experimental novel design

Literary Theory and Criticism

Home › Experimental Novels › Experimental Novels and Novelists

Experimental Novels and Novelists

By NASRULLAH MAMBROL on March 6, 2019 • ( 2 )

Literature is forever transforming. A new literary age is new precisely because its important writers do things differently from their predecessors. Thus, it could be said that almost all significant literature is in some sense innovative or experimental at its inception but inevitably becomes, over time, conventional. Regarding long fiction, however, the situation is a bit more complex.

It is apparent that, four centuries after Miguel de Cervantes wrote what is generally recognized the first important novel, Don Quixote de la Mancha (1605, 1615), readers have come to accept a certain type of long fiction as most conventional and to regard significant departures from this type as experimental. This most conventional variety is the novel of realism as practiced by nineteenth century giants such as Gustave Flaubert, Leo Tolstoy, Charles Dickens, and George Eliot .

The first task in surveying contemporary experiments in long fiction, therefore, is to determine what “conventional” means in reference to the novel of realism. Most nineteenth century novelists considered fiction to be an imitative form; that is, it presents in words a representation of reality. The underlying assumption of these writers and their readers was that there is a shared single reality, perceived by all—unless they are mad, ill, or hallucinatory—in a similar way. This reality is largely external and objectively verifiable. Time is orderly and moves forward. The novel that reflects this view of reality is equally orderly and accountable. The point of view is frequently, though not always, omniscient (all-knowing): The narrators understand all and tell their readers all they need to know to understand a given situation. The virtues of this variety of fiction are clarity of description and comprehensiveness of analysis.

After reading Flaubert’s Madame Bovary (1857; English translation, 1886), one can be confident that he or she knows something about Emma Bovary’s home, village, and manner of dress; knows her history, her motivations, and the way she thought; and knows what others thought of her. Not knowing would be a gap in the record; not knowing would mean, according to standards against which readers have judged “conventional” novels, a failure of the author.

Modernism and its Followers

Early in the twentieth century, a disparate group of novelists now generally referred to as modernists— James Joyce , Virginia Woolf , William Faulkner , Franz Kafka , Marcel Proust , and others—experimented with or even abandoned many of the most hallowed conventions of the novel of realism. These experiments were motivated by an altered perception of reality. Whereas the nineteenth century assumption was that reality is external, objective, and measurable, the modernists believed reality to be also internal, subjective, and dependent upon context. Reflecting these changing assumptions about reality, point of view in the modernist novel becomes more often limited, shifting, and even unreliable rather than omniscient.

This subjectivity reached its apogee in one of the great innovations of modernism, the point-of-view technique dubbed “stream of consciousness,” which plunges the reader into a chaos of thoughts arrayed on the most tenuous of organizing principles—or so it must have seemed to the early twentieth century audience accustomed to the orderly fictional worlds presented by the nineteenth century masters.

Once reality is acknowledged to be inner and subjective, all rules about structure in the novel are abandoned. The most consistent structuring principle of premodernist novelists—the orderly progression of time—was rejected by many modernists. Modern novels do not “progress” through time in the conventional sense; instead, they follow the inner, subjective, shifting logic of a character’s thoughts. Indeed, the two great innovations of modernist fiction—stream of consciousness and nonchronological structure—are inseparable in the modern novel.

Among the most famous and earliest practitioners of these techniques were Joyce (especially Ulysses , 1922, and Finnegans Wake , 1939), Woolf (especially Jacob’s Room , 1922; Mrs. Dalloway , 1925; and To the Lighthouse , 1927), and Faulkner (especially The Sound and the Fury , 1929; As I Lay Dying , 1930). Many of the experimental works of post-World War II long fiction extended these techniques, offering intensely subjective narrative voices and often extreme forms of stream of consciousness, including disruptions of orderly time sequences.

In La traición de Rita Hayworth (1968; Betrayed by Rita Hayworth , 1971), Manuel Puig employs (among other techniques) the words of several sets of characters in different rooms of a house without identifying the speakers or providing transitions to indicate a change in speaker. The effect may be experienced by the reader as a strange solipsistic cacophony, or something like a disjointed choral voice; in fact, the technique is a variation on stream of consciousness and not so very different from the tangle of interior monologues in Faulkner ’s novels.

Tim O’Brien’s novel of the Vietnam War, Going After Cacciato (1978, revised 1989), is another example of a work that makes fresh use of a modernist strategy. Here, reality at first seems more external and hence clearer than in Betrayed by Rita Hayworth . The bulk of the action concerns a rifle squad that follows a deserter, Cacciato, out of Vietnam and across Asia and Europe until he is finally surrounded in Paris, where he once again escapes. The chapters that make up this plot, however, are interspersed with generally shorter chapters recounting the experiences of the point-of-view character, Paul Berlin, at home and in Vietnam. In another set of short chapters, Berlin waits out a six-hour guard shift in an observation post by the sea. The most orderly part of the novel is the pursuit of Cacciato, which moves logically through time and place. The perceptive reader eventually realizes, however, that the pursuit of Cacciato is a fantasy conjured up by Berlin, whose “real” reality is the six hours in the observation post, where his thoughts skip randomly from the present to the past (in memories) to a fantasy world. As in the best modernist tradition, then, the structure of Going After Cacciato reflects an inner, subjective reality.

The post-World War II writer who most famously and provocatively continued the modernist agenda in long fiction was Samuel Beckett, especially in his trilogy: Molloy (1951; English translation, 1955), Malone meurt (1951; Malone Dies , 1956), and L’Innommable (1953; The Unnamable , 1958). In each successive novel, external reality recedes as the narrative voice becomes more inward-looking. In Molloy , for example, the title character searches for his mother, but he is lost from the beginning. He can find neither her (if she truly exists) nor his way back home, wherever that is—nor can he be sure even of the objective reality of recent experience. In one passage Molloy notes that he stayed in several rooms with several windows, but then he immediately conjectures that perhaps the several windows were really only one, or perhaps they were all in his head. The novel is filled with “perhapses” and “I don’t knows,” undermining the reader’s confidence in Beckett’s words.

The subjectivity and uncertainty are intensified in Malone Dies . At least in Molloy, the protagonist was out in the world, lost in a countryside that appears to be realistic, even if it is more a mindscape than a convincing geographic location. In Malone Dies , the protagonist spends most of his days immobile in what he thinks is a hospital, but beyond this nothing—certainly not space or time—is clear. As uncertain of their surroundings as Molloy and Malone are, they are fairly certain of their own reality; the unnamed protagonist of the final volume of the trilogy, The Unnamable , does not know his reality. His entire labyrinthine interior monologue is an attempt to find an identity for himself and a definition of his world, the one depending upon the other. In those attempts, however, he fails, and at no time does the reader have a confident sense of time and place in reference to the protagonist and his world.

The New Novel

With The Unnamable , long fiction may seem to have come a great distance from the modernist novel, but in fact Beckett was continuing the modernist practice of locating reality inside a limited and increasingly unreliable consciousness. Eventually, voices cried out against the entire modernist enterprise. Among the earliest and most vocal of those calling for a new fiction—for le nouveau roman, or a new novel—was a group of French avantgarde writers who became known as the New Novelists. However, as startlingly innovative as their fiction may at first appear, they often were following in the footsteps of the very modernists they rejected.

Among the New Novelists (sometimes to their dismay) were Michel Butor, Nathalie Sarraute, and Claude Simon. Even though Simon won the Nobel Prize in Literature, probably the most famous (or infamous) of the New Novelists was Alain Robbe-Grillet.

Alain Robbe-Grillet

Robbe-Grillet decried what he regarded as outmoded realism and set forth the program for a new fiction in his Pour un nouveau roman (1963; For a New Novel: Essays on Fiction , 1965). His own career might offer the best demonstration of the movement from old to new. His first published novel, Les Gommes (1953; The Erasers, 1964), while hardly Dickensian, was not radically innovative. With Le Voyeur (1955; The Voyeur , 1958), however, his work took a marked turn toward the experimental, and with La Jalousie (1957; Jealousy, 1959) and Dans le labyrinthe (1959; In the Labyrinth, 1960), the New Novel came to full flower.

The most famous technical innovation of the New Novelists was the protracted and obsessive descriptions of objects—a tomato slice or box on a table or a picture on a wall—often apparently unrelated to theme or plot. The use of this device led some critics to speak of the “objective” nature of the New Novel, as if the technique offered the reader a sort of photographic clarity. On the contrary, in the New Novel, little is clear in a conventional sense. Robbe-Grillet fills his descriptions with “perhapses” and “apparentlys” along with other qualifiers, and the objects become altered or metamorphosed over time. After Robbe-Grillet’s early novels, time is rarely of the conventional earlier-to-later variety but instead jumps and loops and returns.

One example of the transforming nature of objects occurs in The Voyeur , when a man on a boat peers obsessively at the figure-eight scar left by an iron ring flapping against a seawall. Over the course of the novel the figure-eight pattern becomes a cord in a salesman’s suitcase, two knotholes side by side on a door, a bicycle, a highway sign, two stacks of plates, and so on—in more than a dozen incarnations. Moreover, Robbe-Grillet’s objects are not always as “solid.” A painting on a wall (In the Labyrinth) or a photograph in a newspaper lying in the gutter ( La Maison de rendez-vous , 1965; English translation, 1966) may become “animated” as the narrative eye enters it, and the action will transpire in what was, a paragraph before, only ink on paper or paint on canvas.

Such techniques indeed seemed radically new, far afield of the novels of Joyce and Faulkner. However, it is generally the case with the New Novelists, especially with Robbe-Grillet, that this obsessive looking, these distortions and uncertainties and transformations imposed on what might otherwise be real surfaces, have their origins in a narrative consciousness that warps reality according to its idiosyncratic way of seeing. The point-of-view character of In the Labyrinth is a soldier who is likely feverish and dying; in The Voyeur , a psychotic murderer; in Jealousy, a jealous husband who quite possibly has committed an act of violence or contemplates doing so. In all cases the reader has even more trouble arriving at definitive conclusions than is the case with the presumably very difficult novels of Joyce and Faulkner.

Ultimately, the New Novelists’ program differs in degree more than in kind from the modernist assumption that reality is subjective and that fictional structures should reflect that subjectivity. As famous and frequently discussed as the New Novelists have been, their fiction has had relatively little influence beyond France, and when literary theorists define “postmodernism” (that is, the literary expression that has emerged after, and is truly different from, modernism), they rarely claim the New Novel as postmodern.

Metafiction

A far more significant departure from modernism occurred when writers began to reject the notion that had been dominant among novelists since Miguel de Cervantes: that it is the chief duty of the novelist to be realistic, and the more realistic the fiction the worthier it is. This breakthrough realization—that realism of whatever variety is no more than a preference for a certain set of conventions—manifests itself in different ways in fiction. In metafiction, also known as self-mimesis or selfreferential fiction, the author (or his or her persona), deliberately reminds the reader that the book is a written entity; in the traditional novel, however, the reader is asked to suspend his or her disbelief.

Often the metafictive impulse appears as little more than an intensification of the first-person-omniscient narrator, the “intrusive author” disparaged by Henry James but favored by many nineteenth century writers. Rather than employing an “I” without an identity, as in William Makepeace Thackeray’s Vanity Fair: A Novel Without a Hero (1847-1848), metafiction makes clear that the “I” is the novel’s author. Examples of this technique include the novels of José Donoso in Casa de campo (1978; A House in the Country, 1984) and of Luisa Valenzuela in Cola de lagartija (1983; The Lizard’s Tail, 1983).

In other novels, the metafictive impulse is more radical and transforming. When Donald Barthelme stops the action halfway through Snow White (1967), for example, and requires the reader to answer a fifteen-question quiz on the foregoing, the readers’ ability to “lose themselves” in the novel’s virtual world is hopelessly and hilariously destroyed. Another witty but vastly different metafictive novel is Italo Calvino ’s Se una notte d’inverno un viaggiatore (1979; If on a Winter’s Night a Traveler , 1981), in which the central character, Cavedagna, purchases a novel called If on a Winter’s Night a Traveler by an author named “Italo Calvino.” Cavedagna finds that his copy is defective: The first thirty-two pages are repeated again and again, and the text is not even that of Calvino; it is instead the opening of a Polish spy novel. The remainder of the book concerns Cavedagna’s attempts to find the rest of the spy novel, his blossoming romance with a woman on the same mission, and a rambling intrigue Calvino would surely love to parody had he not invented it. Furthermore, the novel is constructed around a number of openings of other novels that never, for a variety of reasons, progress past the first few pages.

Metafiction is used to represent the impossibility of understanding the global world, particularly the complexity of politics and economics. Critically acclaimed metafictive novels include Australian Peter Carey’s Illywhacker (1985), American Mark Z. Danielewski’s House of Leaves (2000), Canadian Yann Martel’s Life of Pi (2001), and South-African Nobel laureate J. M. Coetzee’s Slow Man (2005). Junot Díaz won the Pulitzer Prize and the National Book Critics Circle Award for his first novel, The Brief Wondrous Life of Oscar Wao (2007). The novel is ostensibly a love story about a Dominican American man in New Jersey, but through a series of extended footnotes and asides the narrator relates and comments on the history of the brutal dictatorship of Rafael Trujillo in the Dominican Republic. The novel has different narrators and settings, as well as occasional messages from the author, and is filled with wordplay and lively slang in English and Spanish.

Fiction as Artifice

One might well ask if metafiction is not too narrow an endeavor to define an age (for example, postmodernism). The answer would be yes, even If on a Winter’s Night a Traveler , for example, might more properly be described as a novel whose subject is reading a novel rather than writing one. Metafiction is best considered one variation of a broader, more pervasive impulse in post-World War II long fiction: fiction-as-artifice. Rather than narrowly focusing on the process of writing fiction (metafiction), in fiction-as-artifice the author directly attacks the conventions of realism or acknowledges that all writing is a verbal construct bearing the most tenuous relationship to actuality.

One of the earliest examples of fiction-as-artifice in the post-World War II canon is Raymond Queneau’s Exercices de style (1947; Exercises in Style , 1958). The title states where Queneau’s interests lie: in technique and in the manipulation of language, rather than in creating an illusion of reality. The book comprises ninety-nine variations on a brief scene between two strangers on a Parisian bus. In each retelling of the incident, Queneau uses a different dialect or style (“Notation” and “Litotes”). The almost endless replication of the single scene forces the reader to see that scene as a verbal construct rather than an approximation of reality. Such “pure” manifestations of fiction-as-artifice as Exercises in Style are relatively rare. More often, fiction-as-artifice is a gesture employed intermittently, side by side with realist techniques. The interplay of the two opposing strategies create a delightful aesthetic friction.

One of the most famous and provocative examples of fiction-as-artifice is Vladimir Nabokov ’s Pale Fire (1962). The structure of the work belies all traditional conventions of the novel. Pale Fire opens with an “editor’s” introduction, followed by a long poem, hundreds of pages of annotations, and an index. The reader discovers, however, that this apparatus tells a hilarious and moving story of political intrigue, murder, and madness. Does Pale Fire , ultimately, underscore the artifices of fiction or, instead, demonstrate how resilient is the writer’s need to tell a story and the reader’s need to read one? Either way, Pale Fire is one of the most inventive and fascinating novels of any genre.

The same questions could be asked of Julio Cortázar’s Rayuela (1963; Hopscotch, 1966), a long novel comprising scores of brief, numbered sections, which, the reader is advised in the introductory “Instructions,” can be read in a number of ways: in the order presented, in a different numbered sequence suggested by the author, or perhaps, if the reader prefers, by “hopscotching” through the novel.

A similar strategy is employed in Milorad Pavic’s Hazarski renik: Roman leksikon u 100,000 reci (1984; Dictionary of the Khazars: A Lexicon Novel in 100,000 Words , 1988). The work is constructed as a dictionary with many brief sections, alphabetized by heading and richly cross-referenced. The reader may read the work from beginning to end, alphabetically, or may follow the cross-references. An added inventive complication is the Dictionary of the Khazars’s two volumes, one “male” and the other “female.” The volumes are identical except for one brief passage, which likely alters the reader’s interpretation of the whole.

Although fiction-as-artifice is European in origin— indeed, it can be traced back to Laurence Sterne’s Tristram Shandy (1759-1767)—its most inventive and varied practitioner is the American writer John Barth . In work after work, Barth employs, parodies, and lays bare for the reader’s contemplation the artifices of fiction.

In his unified collection Lost in the Funhouse (1968), for example, Barth narrates the history—from conception through maturity and decline—of a man, of humankind, and of fiction itself. The story’s telling, however, highlights the artificiality of writing. The first selection (it cannot be called a “story”) of the novel, “Frame Tale,” is a single, incomplete sentence—“Once upon a time there was a story that began”—which is designed to be cut out and pasted together to form a Möbius strip. When assembled, the strip leads to the complete yet infinite and never-ending sentence “Once upon a time there was a story that began Once upon a time. . . .” The novel’s title story, “ Lost in the Funhouse, ” contains graphs illustrating the story’s structure. “Glossolalia” is formed from six brief sections all written in the rhythms of the Lord’s Prayer . In “ Menelaid ,” the dialogue is presented in a dizzying succession of quotation marks within quotation marks within quotation marks. Barth’s experiments in Lost in the Funhouse are continued and intensified in later novels, especially Chimera (1972) and Letters (1979).

In The Broom of the System (1987) by David Foster Wallace, the protagonist, Lenore Stonecipher Beadsman, feels sometimes that she is just a character in a novel. Wallace’s Infinite Jest (1996) is a massive work about a future North America where people become so engrossed in watching a film called Infinite Jest that they lose all interest in other activities. Wallace’s fiction moves back and forth in time without warning and combines wordplay, long sentences, footnotes and endnotes, transcripts, and acronyms to create a disjointed postmodern language.

Mark Dunn’s Ella Minnow Pea: A Progressively Lipogrammatic Epistolary Fable ( 2001) is an epistolary novel about a fictional island off the coast of South Carolina, where the government forbids the use of one letter of the alphabet, at a given time; in effect, the government is parsing the alphabet. The novel is a collection of letters and notes from inhabitants, often less than a page in length, written with a diminishing set of alphabet letters.

Fiction or Nonfiction?

Even at his most experimental, however, Barth never abandons his delight in storytelling. Indeed, virtually all the long fiction addressed thus far show innovations in certain technical strategies but do not substantially challenge the reader’s concept of what is “fictional.” A number of other writers, however, while not always seeming so boldly experimental in technique, have blurred the distinctions between fiction and nonfiction and thus perhaps represent a more fundamental departure from the conventional novel.

The new journalists—such as Truman Capote ( In Cold Blood , 1966), Norman Mailer ( The Armies of the Night: History as a Novel, the Novel as History , 1968), Tom Wolfe ( The Electric Kool-Aid Acid Test , 1968), and Hunter S. Thompson with his Fear and Loathing series (beginning in 1972)—blur the lines between fiction and nonfiction by using novelistic techniques to report facts. However, the subtitle of Mailer’s work notwithstanding, the reader rarely is uncertain what side of the fictionnonfiction line these authors occupy. The same cannot be said for Don DeLillo (Libra, 1988). For his interpretation of the Kennedy assassination, DeLillo spent countless hours researching the voluminous reports of the Warren Commission and other historical documents. With this factual material as the basis for the novel and with assassin Lee Harvey Oswald as the central character, the degree to which Libra can be considered fictional as opposed to nonfictional remains a challenging question.

The question is even more problematic in reference to Maxine Hong Kingston’s The Woman Warrior: Memoirs of a Girlhood Among Ghosts (1976). Kingston conducted her research for her memoir not in library stacks but by plumbing her own memory, especially of stories told her by her mother. At times, Kingston not only is imaginatively enhancing reconstructed scenes but also is inventing details. Is this a work of autobiography or a kind of fiction?

Publishers had trouble classifying Nicholson Baker ’s The Mezzanine (1988) and W. G. Sebald’s Die Ausgewanderten (1992; The Emigrants , 1996). The reader is fairly certain that the point-of-view character in The Mezzanine is fictional, but in what sense is his experience fictional? The work, made up of essaylike contemplations of whatever the persona’s eye falls on as he goes about his business on a mezzanine, recalls in some ways the intensely detailed descriptions of the New Novelists but with even less of an apparent conflict or movement toward climax one expects in fiction.

Sebald’s work is in some ways even odder. His short biographies of a selection of dislocated Europeans have a documentary feel—complete with photographs. The photographs, however, have a vagueness about them that makes them seem almost irrelevant to their subjects, and the reader has the uneasy impression that the book may well be a fabrication.

The distinction between fiction and nonfiction was brought into new relief in 2003, when James Frey published A Million Little Pieces , his memoir of escaping drug addiction. The memoir was aggressively gritty in its detail of the author’s struggles, and Frey was widely praised for his courage and honesty in revealing his own mistakes and weaknesses. In 2005 the book was named to Oprah’s Book Club , and soon after topped nonfiction best-seller lists. In 2006, however, much of the material in the “memoir” was found to have been fabricated. The resulting clamor from talk-show host Oprah Winfrey, Frey’s publishers, and readers led to a lively and interesting public debate over art and truth. Frey claimed that the work presented the “essential truth” of his life. Many readers continued to value the book as an honest accounting of what life is like for some addicts, but readers who felt that they had been defrauded were offered a refund. The Brooklyn Public Library, for example, moved the book to its fiction section.

Just as Baker and Sebald call into question what earlier generations would have thought too obvious to debate—the difference between fiction and nonfiction— one consistent impulse among experimenters in long fiction has been the question, What is necessary in fiction and what is merely conventional? Their efforts to test this question have brought readers some of the most provocative and entertaining works of fiction in the late twentieth and early twenty-first centuries.

BS Johnson with pages from his 1968 novel-in-a-box The Unfortunates , which could be shuffled and read in any order.

Bibliography Currie, Mark, ed. Metafiction. London: Longman, 1995. Collection of articles on experimental themes and techniques. Friedman, Ellen G., and Miriam Fuchs, eds. Breaking the Sequence: Women’s Experimental Fiction. Princeton, N.J.: Princeton University Press, 1989. Levitt, Morton. The Rhetoric of Modernist Fiction from a New Point of View. Hanover, N.H.: University Press of New England, 2006. Robbe-Grillet, Alain. For a New Novel: Essays on Fiction. 1965. New ed. Translated by Richard Howard. Evanston, Ill.: Northwestern University Press, 1996. Seltzer, Alvin J. Chaos in the Novel: The Novel in Chaos. New York: Schocken Books, 1974. Shiach, Morag, ed. The Cambridge Companion to the Modernist Novel. New York: Cambridge University Press, 2007. Waugh, Patricia. Metafiction: The Theory and Practice of Self-Conscious Fiction. New York: Routledge, 2005. Rollyson, Carl. Critical Survey Of Long Fiction. 4th ed. New Jersey: Salem Press, 2010.

Share this:

Categories: Experimental Novels , Literature , Novel Analysis

Tags: 000 Words , A House in the Country , A Million Little Pieces , Alain Robbe-Grillet , American Literature , Analysis of Robbe-Grillet's Novels , Analysis of Truman Capote’s Novels , As I Lay Dying , aymond Queneau , Betrayed by Rita Hayworth , BS Johnson , Claude Simon , David Foster Wallace , Dictionary of the Khazars: A Lexicon Novel in 100 , Die Ausgewanderten , Don DeLillo , Donald Barthelme , Ella Minnow Pea: A Progressively Lipogrammatic Epistolary Fable ( , Exercises in Style , Experimental Long Fiction , Experimental Novels and Novelists , Fear and Loathing series , Fiction as Artifice , Finnegans Wake , For a New Novel: Essays on Fiction , Fragmentation in Novels , Franz Kafka , French avantgarde writers , French New Novelists , Going After Cacciato , Hunter S. Thompson , If on a Winter’s Night a Traveler , Illywhacker , In Cold Blood , In the Labyrinth , Infinite Jest , Italo Calvino , J. M. Coetzee , Jacob’s Room , James Frey , James Joyce , John Barth , Julio Cortázar , Junot Díaz , Laurence Sterne , Lee Harvey Oswald , Lenore Stonecipher Beadsman , Literary Criticism , Literary Theory , Lost in the Funhouse , Malone Dies , Manuel Puig , Marcel Proust , Mark Dunn , Mark Z. Danielewski , Maxine Hong Kingston , Metafiction , metafictive novels , Michel Butor , Milorad Pavic , Mrs. Dalloway , Nathalie Sarraute , new journalists novels , New Novelists , Nicholson Baker , nonchronological structure of novels , Norman Mailer , Novel Analysis , Oprah’s Book Club , Pale Fire , pastiche , Pastiche in Novels , Peter Carey , post-World War II fiction , Rayuela , Samuel Beckett , stream of consciousness , The Brief Wondrous Life of Oscar Wao , The Broom of the System , The Electric Kool-Aid Acid Test , The Emigrants , The Erasers , The Lizard’s Tail , The Mezzanine , The Novels of Robbe-Grillet , The Sound and the Fury , The Unfortunates , The Unnamable , The Woman Warrior: Memoirs of a Girlhood Among Ghosts , Tim O’Brien , To the Lighthouse , Tom Wolfe , Tristram Shandy , Truman Capote , Ulysses , Virginia Woolf , Vladimir Nabokov , W. G. Sebald , William Faulkner , Yann Martel

Related Articles

• Epistolary Novels and Novelists | Literary Theory and Criticism
• Postmodern Novels and Novelists | Literary Theory and Criticism

Leave a Reply Cancel reply

You must be logged in to post a comment.

Home For Fiction – Blog

for thinking people

March 29, 2020

Experimental Fiction: Examples and Dynamics

art , experimental , fiction , literature , writing

Experimental fiction examples aren’t easy to find. The thing is, experimental fiction is the kind of writing that rarely produces a commercial hit – though it does so every now and then.

Imagine literary fiction on LSD, and you have something like experimental fiction. But no, not all experimental fiction is literary – and certainly most of literary fiction isn’t experimental.

A definition of experimental fiction isn’t easy, either. In some sense, experimental fiction is the kind of fiction that displays significant deviation from literary norms – mostly in structural, formatting, or methodological aspects. We’ll see more about all these in a moment, through the examples I’ve picked.

At the same time, we must also recognize an interesting paradox: Whether a novel is experimental or not must be seen in connection with its context. In other words, what is groundbreaking, radical, and unorthodox today, might be the norm tomorrow.

For some, experimental fiction is “weird”. For others, it’s “awesome”. As for me, it’s an interesting opportunity to learn a thing or two about how fiction, writing, and even literary criticism operate.

Examples of Experimental Fiction

I’m a great fan of examples – they make it easy to immediately picture something. This is especially the case with difficult or vague concepts. And, I assure you, experimental fiction is such a concept.

Therefore, I have put together a list of three novels that exemplify this kind of writing. I’ve read all three of them. One I liked a lot, one I hated, and one I enjoyed more about its how than about its what – which is a major sign of experimental fiction, as we’ll see in more detail below.

The three novels in question are The Raw Shark Texts , Beautiful Losers , and House of Leaves . As I said, every now and then you do get some commercial hits, too. Let’s take a look at these novels one by one, to see how experimental fiction looks like. Afterward, we’ll talk about the how’s and why’s of the genre Is it a genre or a mode? That is a difficult question to which I don’t have a clear answer. I would favor "mode", but I’m open to discussions on the matter. For more on genre vs mode, also look at my post on Gothic vs science fiction and horror . .

The Raw Shark Texts : Text, Concept, Writer, Reader

The plot of The Raw Shark Texts is fairly simple, on the surface. Eric Sanderson, suffering from what appears to be a rare form of amnesia, discovers a note from a previous instance of himself, addressed to himself. He begins to follow the clues and enters a world where words and text can take the form of a conceptual shark and kill you.

If that sounds crazy alright, wait till you see the book.

Yes, that’s a shark made of words there. And guess what, it’s also moving . You read that right; a physical book, made of printed text, can show a shark – made of printed text – move toward you. Here’s how:

The whole point about The Raw Shark Texts and its experimental nature is to draw attention to the power of words . Particularly, the novel makes a compelling case about how we approach (and formulate) reality in a way that, though subjective, has very objective consequences.

Beautiful Losers : Hey, Where Are the Paragraphs?

Leonard Cohen… His music, I like; his writing, not so much. If I recall correctly, Beautiful Losers was written under the Greek summer sun on the lovely island of Hydra . It was also written under… the influence, which clearly shows, as you can see in these excerpts:

It’s not so much the lack of paragraphs (not all the book is like that, but some of it is), but just the fact that there’s so little to grab onto. The Raw Shark Texts had a plot and some sort of evolution; Beautiful Losers , far less so.

Come to think of it, Beautiful Losers reminds me of some results I got using my JavaScript random text generator . Though, actually, the random text generator was probably more coherent!

House of Leaves ; or, Mercilessly Mocking the Academia

This last example of experimental fiction is my favorite, though not because of the book itself. I did enjoy House of Leaves – though not as much as The Raw Shark Texts , which was far more philosophical and speculative in a purely artistic way This is a subtle issue, and difficult to properly explain without you having read the books. Very simply, I mean that The Raw Shark Texts engaged the reader in a philosophical discussion without doing it directly or overtly; merely as a consequence of its form. Conversely, House of Leaves , by paying excessive attention to concepts such as academic criticism , journalism , and generally our record-obsessed culture , leaves far less room for ambiguity, artistry , and reader-based interpretation. .

House of Leaves I enjoyed because it mercilessly mocks the academia . It taunts criticism This includes the present post, in a way! It takes a brilliant mind to actually anticipate criticism and preemptively absorb it into the literary product. I know, because I’ve done it myself with one of my novels (besides brilliant I’m also very modest, what can I say…) . It exposes the clinical, hopelessly boring deconstruction of art that has become a self-sustaining process, having become so engorged with the blood of artists that it’s about to burst in a supernova of amusing gore.

In plain English, House of Leaves shows how hopelessly out of touch with reality modern intelligentsia can be.

It basically takes a horror story – where the house of a family isn’t what it seems – and through a very complex series of (entirely fictitious) interviews, pieces of research, books, videos, and what not, it creates an echo that’s supposed to assign authenticity to the whole thing, only it does the exact opposite. It’s the “ Blair Witch Project genre”, basically.

Moreover, and more crucially, it precisely exposes academia for what it is, by showing all these (fictitious) researchers approaching with excessively dry and clinical methods events that are inherently emotional .

To do that, House of Leaves deploys the most bizarre formatting you’ve ever seen.

I particularly liked the instance where there is a footnote, which itself contains a footnote, which itself contains yet a third footnote.

The book is 600 pages long. If we only kept the “basic story”, formatted traditionally, the book would probably be 60-70 pages long.

What Is the Purpose of Experimental Fiction?

In its most basic form, experimental fiction is about drawing attention. If the author is skillful, this isn’t merely about drawing attention in general , but about drawing attention to specific processes – mostly in regard to reading, writing, literature, and the dissemination of information.

Experimental fiction is an auto-referential process .

If the author is really skillful, then the final product is also perfectly readable and enjoyable – I’d characterize The Raw Shark Texts as such an example; House of Leaves less so, and Beautiful Losers not at all.

Another option is when a book becomes partially – that is, to an extent – experimental by… accident. This is usually a result of the novel escaping the author’s conscious control .

Issues of Historical Context

Bram Stoker’s Dracula would be a fantastic such example. To us today it might not seem all that experimental, but for the late-1800s Victorian audience, the merging of Gothic elements with modernity and the “upgrading” of the epistolary forms to include phonographs, photos, and other media, was surely unorthodox.

Another example would be Mary Shelley’s Frankenstein . As I mentioned in my post on 3 Gothic characters with a secret :

[I]n the general complexity of the plot, with stories within stories and horrifying events easily attracting all the attention, few readers realize the importance of [Margaret Saville] for the narrative: “[A]s the designated recipient of her brother’s writings, she will ultimately hold Robert’s letters which contain Victor’s story which contains the monster’s story which includes the story of the DeLacy’s [sic]” (Dickerson, Vanessa D. “The Ghost of a Self: Female Identity in Mary Shelley’s Frankenstein ”. Journal of Popular Culture . 27.3. 1993. p.84). It is a remarkable realization, that the entire narrative is dependent on the hands of a woman. Furthermore, the lack of an omniscient narrator in  Frankenstein  leaves open all possibilities, including the one that the text of  Frankenstein  is, after all, edited by none other than Margaret Saville herself. Extending the skeptical hypothesis to its opposite limits, there is nothing to prove Margaret’s existence in the first place. She could as well be a creation of Robert’s imagination, just like the entire narrative. The metatextual twist is, of course, that the entire narrative  is  a product of the imagination, namely Mary Shelley’s – who shares the same initials as Walton’s sister, MS.

Again, Frankenstein doesn’t seem all that experimental to us today. Perhaps not even at its own time. And yet, this auto-referential essence that we saw in more recent (and more… traditionally experimental) works is obvious.

Experimental Fiction Is a Discourse of Doubt

You might like it or hate it – I did hate Beautiful Losers . But one thing’s for certain: Experimental fiction will make you feel something.

This is already a big deal.

In a literary world full of pointlessness and regurgitation, authenticity is always welcome. Sometimes it doesn’t work – to some people at least – but the effort is always worth it.

On the other hand, the examples of experimental fiction I showed you in this post are the exception rather than the rule, when it comes to commercial viability. If you care about making money from writing books ( I certainly don’t ), I doubt experimental fiction is the way to go.

I’ve written some experimental fiction myself – the Self Versus Self project. Not only does it come in two volumes – one narrative poetry, one a novel – that somewhat narrate the same plot, but even the novel itself is not quite traditional. Here’s what I said about it in the post linked above:

[ Self Versus Self ] contains various forms of texts, from Instagram hashtags to twitter comments and from news reports to journal entries. It also contains stories within stories, and some experienced readers will surely notice many other details, all related to  the writing process itself . Is the work self-referential? You bet.

It’s all about challenging your preconceptions.

What is art?

What is criticism?

Why should you care?

And, in the words of one of the characters of Self Versus Self , there are no answers; but there are always questions.

Novlr is now writer-owned! Join us and shape the future of creative writing.

24 February 2023

6 Books That Break the Mould and Use Experimental Storytelling Techniques

Reading and writing go hand in hand. To be a truly great writer, you must immerse yourself in the written word. It’s a deeply personal craft, and no two writers will express themselves in exactly the same way. In fact, some storytellers push boundaries and experiment with style and expression to challenge both themselves, and their readers.

Experimental and unconventional writing techniques take many forms. Non-linear narratives, epistolary styles, second-person voice , and grammatical experimentation are just some of the fascinating ways that authors push boundaries in service of their stories.

Whether you’re a reader or a writer, here are some amazing works of fiction to explore that use experimental or unusual writing techniques to build their narratives in unique and fascinating ways.

I’m Thinking of Ending Things by Iain Reid

I’m Thinking of Ending Things is a bold and completely unique psychological horror thriller that keeps its secrets until the very end. It’s a simple story on the surface; an un-named woman agrees to go on a road trip to visit her boyfriend’s family, despite having doubts about their relationship. But underneath that simplicity lies as deeply disturbing, complicated narrative that is impossible to forget.

Iain Reid masterfully weaves a compelling story that is completely turned on its head as the reader reaches the climax. It’s a book you can’t just read once. There are things hinted at below the surface that require a second read-through to really get to grips with. Every element of the plot, every tiny revelation, is carefully included with nothing left to chance, but as you reach the end, instead of things clicking into place, you’re left wondering how those elements feed into the resolution.

I’m Thinking of Ending Things is a book with immense staying power. It will live in your head rent free, and nip at your thoughts until you dive back in to unravel the mystery. The narrative is scary and unsettling, but you’re never really sure why you feel that way. And it’s all achieved through an interesting mix of narrative points of view and a non-linear arc, interspersed with hints at a deeper reality, without revealing too much too soon.

Stories of Your Life by Ted Chiang

I will never stop sharing my love of Ted Chiang. In my view, he is one of the most innovative writers out there, with the deft skill to make big, complicated concepts understandable. He has published two short story collections, but his first collection, Stories of Your Life , contains the titular novella, Story of Your Life, which changed the way I look at grammatical structure completely. It’s one of the most fascinating works of short fiction that really exemplifies how experimental writing can convey new and unique meaning.

Story of Your Life is about a linguist, Dr. Louise Banks, who narrates the story. When aliens appear on earth, she is tasked with opening a line of communication with them, and to do that she must learn how they communicate.

The aliens do not use language the same way that humans do, so Chiang explores the concept of linguistic relativity — the idea that the way we use language influences the way we experience and conceive of reality. As Banks learns the alien language, she experiences reality as they do, which Ted Chiang represents by experimenting with grammatical structure and his use of tense. He seamlessly blends past, present, and future through his exacting and careful manipulation of grammar. It should be hard to understand, and yet, it isn’t. An absolute must-read if you want a perfect example of having to know the rule to break them.

House of Leaves by Mark Z. Danielewski

House of Leaves is probably the most famous example of experimental structure to convey a story. its unconventional layout and style is designed to mirror the events and the mental instability of the book’s protagonists, and it does so to wonderful effect.

The novel itself is difficult to classify. It is primarily marketed as a horror story, and yet easily falls into classification as romance and satire as well. It’s a postmodern, contemporary work about a young family who discover that their new home is bigger on the inside than it is on the outside and become increasingly unsettled.

Multiple narrators tell different stories in different ways. The book is filled with marginalia has multiple narrators, colours, footnotes, and typographical tricks that help to convey the disorienting and chaotic nature of the story.

Besides exploring themes of perception, reality, and madness, House of Leaves is also a meditation on the nature of storytelling itself, and a fascinating parody of academia. It’s a challenging read that demands close attention and active engagement, but it’s a wonderful example of what can be achieved by a writer who thinks outside the box.

Sleeping Giants by Sylvain Neuvel

The first book in Sylvain Neuvel’s science-fiction series, the Themis File s, Sleeping Giants , is a wonderful example of epistolary fiction at work. It tells the story of a mysterious artifact discovered buried deep beneath the Earth’s surface, and the efforts to unlock its secrets, but with a unique narrative structure.

The novel is written as a series of documents, including interview transcripts, journal entries, official reports. and newspaper articles. What this achieves, is to give the reader a real sense of immersion in a narrative that is incredibly expansive, and yet feels both immediate and personal.

One of the hallmarks of science fiction is a grandiose extrapolation and explanation of complicated scientific concepts. By choosing an epistolary style for his series, Neuvel foregoes those elements, as they become irrelevant. We only need to know what we can piece together from the documents with which we are presented. The interviews especially, limit what information readers are furnished with, giving a sense of being part of something greater, while not being overwhelmed by the “infinity” of it all.

Sleeping Giants incorporates elements of mythology and science fiction, weaving together a complex and thought-provoking narrative that explores themes of power, control, and the consequences of human ambition. Through its unique narrative structure and storytelling, it offers readers a fresh and exciting take on the traditions of science fiction.

Horrorstör by Grady Hendrix

The unique design and layout of Grady Hendrix’s Horrorstör is the first thing that jumps out at any prospective reader. The IKEA-esque catalogue style is immediately recognisable, and it’s hard not to be drawn in.

The novel tells the story of a group of employees who work at Orsk, a fictional Scandinavian furniture store that bears a striking resemblance to IKEA, and who decide to spend the night in the store to find out what is happening overnight that is resulting in a lot of smashed products whenever they open up for the day. They patrol the deserted showroom, trying to solve the mystery, with chilling results.

The thing that makes Horrorstör so unique is its use of design elements to enhance the reading experience. The novel is designed to resemble a furniture catalog, with product descriptions, diagrams, and even fake advertisements interspersed throughout the text. This design choice helps to create a sense of immersion for the reader and adds to the overall creepy and unsettling atmosphere of the story; especially since it’s a design that is easily recognisable and the product descriptions become bleaker as the story progresses.

Horrorstör uses its unique design to incorporate elements of satire and social commentary into its horror elements, skewering the consumerist culture of modern-day America and the soulless corporate environments that can often come with it. This blend of horror and humour, along with its unconventional design elements, makes Horrorstör a book that’s well worth picking up for its concept alone.

If on a Winter’s Night a Traveller by Italo Calvino

If on a Winter’s Night a Traveller is a highly experimental work of fiction. It is very much a book lover’s book, and firmly post-modern.

The novel follows the story of “you” as “you” set out to read a new novel by Italo Calvino titled If On A Winter’s Night a Traveller . However, after a few pages, “you” discover that there is a printing error, and the novel has been replaced with a different book entirely. As “you” attempt to track down the correct novel, “you” are thrust into a series of different stories, each with its own unique style and genre.

Using alternative chapters of second-person narration, Calvino puts “you,” the reader, at the very centre of his tale. It’s a love story, both in the truest sense, as well as a love story to the act of narrative storytelling. It’s disjointed due to its alternating narrative and varying opening chapters of other works, but there is still a substantive arc that runs through the second-person narration; a story of an international conspiracy involving fraudulent books, a cunning translator, a secluded author, a failing publishing company, and a series of oppressive governments.

Every element of Calvino’s work feeds in to the next. The narration and how it is used is interesting on its own, but the thematic content of each of the fictional works relates to narration that follows in the chapter after it. And the titles of each of the stories, when put together, create the beginnings of yet another tale. It’s highly experimental, deliciously complex, and beautifully unique — a masterclass in the art of experimental storytelling.

Whether you plan to write experimental works of your own, or you prefer to stick to the tried-and-true methods of storytelling, these novels are worth taking the time to read. They are fascinating and dynamic narratives that challenge the traditional modes of storytelling, exploring the limitless potential of the written word.

Note: All purchase links in this post are affiliate links through BookShop.org, and Novlr may earn a small commission – every purchase supports independent bookstores.

• Craft and Criticism
• Fiction and Poetry
• News and Culture
• Lit Hub Radio
• Reading Lists

• Literary Criticism
• Craft and Advice
• In Conversation
• On Translation
• Short Story
• From the Novel
• Bookstores and Libraries
• Film and TV
• Art and Photography
• Freeman’s
• The Virtual Book Channel
• Behind the Mic
• Beyond the Page
• The Cosmic Library
• The Critic and Her Publics
• Emergence Magazine
• Fiction/Non/Fiction
• First Draft: A Dialogue on Writing
• The History of Literature
• I’m a Writer But
• Lit Century
• Tor Presents: Voyage Into Genre
• Windham-Campbell Prizes Podcast
• Write-minded
• The Best of the Decade
• Best Reviewed Books
• BookMarks Daily Giveaway
• The Daily Thrill
• CrimeReads Daily Giveaway

Renaissance of the Weird: Experimental Fiction as the New American Normal

John domini on amber sparks, percival everett, laura van den berg, lance olsen, paul beatty, karen russell, and more.

So you pick up a New Yorker short story, hoping to find something fresh. Here’s one that seems to have gotten a lot of attention, “The Ghost Birds,” by Karen Russell. In no time you find yourself spellbound, swept up in a world where no one would want to live, a near-future biosphere so toxic it’s killed off all the birds. Gloomy stuff indeed, and yet you turn pages or swipe screens fascinated, compelled in large part by the sheer strangeness.

Russell will kick off a section with an arresting line like “To be a kid requires difficult detective work,” thereby opening an alternative point of view for which the only correlative might be the children’s barracks at Auschwitz. She’ll interrupt the narrative with lists of the bird species lost, or with flashbacks so compelling, they could be whole novels in thumbnail: “The fires spread to every continent. The air turned a peppery orange, making each unfiltered breath a harrowing event.” At the story’s climax, things turn supernatural, decidedly ambiguous, and to confirm the outcome, you need to reread a fleeting earlier reference or two.

A magnificent piece of work, “The Ghost Birds” depends—for its impact—on stretching, not to say manhandling, the fictional form. It bears little resemblance to what’s generally considered “a New Yorker story,” the strained domesticity of contributors from John O’Hara to Ann Beattie. At the same time, despite its future tech and apocalyptic apparatus, Russell’s piece doesn’t feel right for, say, Fantasy & Science Fiction . It achieves both emotional sting and political savvy (a harsh critique of capitalism) beyond what’s generally considered SF turf, the materials of Asimov and Bradbury.

But then again, who cares what’s “generally considered?” Don’t perceptions like that always wind up off-base, whether the subject’s the New Yorker or F&SF ? These days, Hugo and Nebula winners claim the proud heritage of George Orwell and Mary Shelley, and many prefer to call their genre “spec-fic,” if not “cli-fi.” That last category, fiction about the climate crisis, seems the best fit for Russell’s splendid work⎯insofar as it needs a fit.

“Ghost Birds” succeeds by defying any such demands, disrupting the norms of dramaturgy, and so makes an excellent introduction to my larger point. I’d argue that nothing so animates contemporary American novels and short stories as the spirit of experiment. Experimental fiction is flourishing, as we near the century’s quarter-mark, in a way this country has rarely if ever seen.

The evidence lies scattered far and wide, so easy to spot that I’m baffled by how little critics have noticed. Granted, the term “experimental” still turns up, most often applied to esteemed elders like Don DeLillo. Yet the same energy crackles, unmistakably, in a good many talents now in mid-career, if not just hitting their stride.

The outstanding case in point would be Colson Whitehead; the novel that may rank as his most celebrated, The Underground Railroad, also presents his wildest Rube Goldberg contraption. And Whitehead hardly stands alone. Even setting aside the so-called Afrofuturists, like N.K. Jemisin and her alternative worlds, many of the recent knockout fictions from Black Americans display a glittering eccentric streak. You spot it even in texts very different from Whitehead’s, like Paul Beatty’s The Sellout (2015) . Then there’s the prolific Percival Everett, lower profile but always in some way iconoclastic.

Naturally, the phenomenon is more common in the smaller independent houses, like Fiction Collective 2, Two Dollar Radio, or Dzanc. Among these, Lance Olsen would be the exemplar, now into his third decade of narrative chimera. But a number are with commercial houses, among them some of the country’s most exciting women authors. Alongside Russell, I’d put Amber Sparks and Laura van den Berg.

In short, there’s a lot of activity. Naturally, it hasn’t spread everywhere or changed everything. Telling your story strangely, in a literary culture of such variety, has by no means become the only way for an American to work. Still, an awful lot of writers are embracing the strange, impatient with established standards and practices. I daresay such titles dominate the New Fiction shelves these days, and certainly the work has won notice, including more than a few awards. Still, no one has stood up to raise a shout, a salute, and this leaves it up to me, I suppose⎯a toast to our Renaissance of the Weird.

More than celebration, to be sure, my essay intends illumination. I’ll start abroad, in Europe particularly, considering how it treats narrative rowdiness, and then I’ll draw out the contrast to the situation Stateside, our long-troubled relationship with such work. Once I’ve established that difference, that history, I’ll return to the contemporary, a range of unfettered homegrown talents. That range is remarkable, as I say⎯vineyards of every terroir have produced stunning varietals⎯and demands exploration of its roots, its reasons. That’s where I’ll end, with the question of Why now?

Now, over on the Continent, this wouldn’t be news. Paris and Berlin reclaimed their place on the cutting edge following the last World War, and European fiction has sprouted all sorts of wild hairs, whether by Alain Robbe-Grillet in the 1950s or Jenny Erpenbeck this past decade. Other pertinent names would fill several pages in Oulipo format, with paste-ins and marginalia, and the emotional range would run from the sourpuss Thomas Bernhard to the upwardly striving Bernardine Evaristo.

Naturally, more straightforward narrative has seen its champions as well. The latest is Elena Ferrante, with her Marxist economics and family sorrows. If any Italian since Dante can match that woman’s stature, however, it’s Italo Calvino, so radical an imagination that his Invisible Cities (1972) invented a fresh form for the novel. Reframing narrative, as Cities did, may provide the best handle on the recent European contribution to the artform. Other primary shape-shifters would include Beckett and Sebald.

The effort to construct a nouveau roman also energizes a good deal of the fiction out of the former European colonies. Again, the list could go on and on, but consider that India gave us the protean Salman Rushdie, Central Africa Alain Mabanckou, with his shaggy supernatural tales, and Oman (formerly a British “protectorate”) the Booker-winner Jokha Alharthi. Her novels are garnished with original poetry and hopscotch across a century of women’s lives.

For an American, it would seem only natural to venture something similar: to risk, at least once in a while, getting lost in the funhouse. I’m citing John Barth, of course, his watershed innovation from 1967, and among more recent US novelists, at least one went out of his way to pay that story homage. David Foster Wallace, rather a turn-of-the-millennium rock star, made free use of Barth’s Funhouse , the book as well as its title piece, in his ’89 novella “Westward the Course of Empire Takes its Way.” DFW’s story bogs down badly, it must be said, but the set in which it appears, Girl With Curious Hair, includes one or two of his most dazzling, and these reference previous experiments other than Barth’s. Throughout his too-brief career, this author honored his elders, the writers generally known as “the Postmoderns.”

Yet even as he brought off his own provocative spec-fic, Wallace took care to set it apart from the work of these same predecessors. In interviews he worried that novels and stories like Barth’s could be “enervating,” especially in their tendency to self-commentary, or meta-fiction. Misgivings like that come with the territory, part of any artist’s response to another, but what Wallace had to say was angrily amplified by his colleague Jonathan Franzen. The author of The Corrections has always subscribed to a more accessible model for fiction, and he defended it at length in a 2002 New Yorker essay, “Mr. Difficult”⎯otherwise an unforgiving takedown of William Gaddis, perhaps the greatest of the Postmoderns.

Franzen’s complaints were far from the first. Attacks on the freaks of US fiction go back to their freaky heyday, what you could call the Postmodern moment. This lasted about five years, roughly the first half of the 1970s, and the rock star of the group was Donald Barthelme. If New Yorker norms were sabotaged, Barthelme was the culprit, and he drew major media attention while sharing a girlfriend with Miles Davis (see the biography Hiding Man ). During those same years, too, Barth, Pynchon, and Gaddis each took home a National Book Award. But as they enjoyed the limelight, others sat grumbling in the dark.

An early reprisal came from Gore Vidal, who in “American Plastic” (1976) wielded his usual viper’s tongue. As for an argument, that was largely absent, but then John Gardner’s On Moral Fiction (1980) had even less to say, while serving up more vitriol. This book got some traction⎯Gardner, like Vidal, could write powerfully⎯but it’s little more than an executioner’s roll call: the Righteous and the Damned. The latter included even Gardner’s friend William Gass, sent to the block for seeking fictional alternatives.

To be sure, Gass and the rest of the condemned had their defenders; these days, Barthelme’s in the Library of America. Nevertheless, the beat-down went on for decades; first Tom Wolfe grabbed a cudgel (“The Billion-Footed Beast,” ’89), then Franzen. The attacks came with such regularity, and sounded so similar, it’s hard not to think of Puritanism, its lingering chill. The ripple effect was withering; the lascivious biplay of Robert Coover’s “The Babysitter” (1968), for instance, would get bumped off the syllabus or out of the anthology, while there was always space for the bleak monosyllables of Raymond Carver’s “What We Talk About When We Talk about Love” (‘81).

Surely an argument can be made for both sorts of stories, but in the US this was largely lacking until 2008, when Zadie Smith published her splendid and clarifying essay, “Two Paths for the Novel.” Still, even now, the more crooked and crazy path may be marked CLOSED. In recent weeks, I’ve heard a talented and well-published writer claim that, in the offices of Manhattan publishing, “anything experimental” will get “combed out” of a manuscript. I’ve read the sensitive critic Ron Charles, in his review of Jennifer Egan’s The Candy House , disdainful about her playfulness: “A second-person narrator? You shouldn’t have.” Recently I’ve witnessed the recent Twitterstorm, the thunder rumbling on both sides of the issue, after journalist Ben Judah called for, well, moral fiction: for “the great society novels” of the past, “the Zolas, Balzacs, Austens, Tolstoys….”

Whether prompted by our grim-faced forefathers or the smirking Gore Vidal, the US critical establishment developed a discomfort with the label “Postmodern.” These days, a text like Gaddis’ JR (1975) might instead be called a “systems novel.” The systems in question are the larger controlling forces of our lives, and certainly JR makes a good example; it turns us all to hamsters on the wheel of Big Finance. Still, this handle too proves slippery. I first encountered the expression in Tom LeClair’s work on DeLillo ( In the Loop , 1988), and in Libra (’88) or Underworld (’97), doesn’t “the system” carry guns? CIA, FBI, Mafia?

More recently, the NYTBR review of Whitehead’s Harlem Shuffle (2022) called it a systems novel, and in Whitehead’s imagination, it’s race that holds the reins. The Prime Mover, the source of our troubles, provides a central subject of any good drama, after all. That includes social realism like Ferrante’s. What’s more, to insist that a single purpose unifies all the new American efforts to bend and fold the fictional form⎯doesn’t that violate the very project?

DeLillo’s case tells us something, since he wears the dual descriptors of Postmodern and systems novelist. Critical consensus holds that the man’s peak came a while back, with the brilliant run from Libra to Underworld , while the 21st-Century work is regarded, by and large, as a letdown. Michiko Kakutani of the Times gave Falling Man (2007) her Imperial thumbs-down, declaring it “spindly” and “inadequate.” Even DeLillo himself, when he relents to an interview, will admit he’s no longer writing the way he used to. Beyond that, to be sure, the man remains cagey, and yet any objective comparison between his latest half-dozen and their honored predecessors at once reveals the salient difference.

The work since Body Artist (‘01) is more experimental . Their materials tend to the surreal; Zero K (’16) even tosses in a voice from the dead⎯or is it the undead? Their central action tends to totter off-center, and as for “systems,” most of the protagonists drift at a level of comfort that leaves the question moot. The author has left the ballpark, the sweat and grit so essential to Mao II (1991) and “Pafko at the Wall.” Instead, he’s going for shadow, suggestion, myth. Whether that makes for great fiction is another question, to be sure, but if you ask me, whichever muse inspired Cosmopolis, (‘03) she’s a magical seductress. I’d call that one a fable of renunciation: the Emperor strips off his own clothes.

Insofar as DeLillo belongs in any New Wave, though, he’s its gnomic Elder. Among the younger and more approachable are writers like Laura van den Berg and Amber Sparks. Again, these two are but a small sample; forerunners include Aimee Bender, who’s spent a quarter-century reinventing the fairy tale; newer on the scene would be Missouri Williams. Her debut The Doloriad (a title that recalls Barth) presents a post-apocalyptic society to make your hair stand on end⎯as do some of the imagined futures in Sparks and van den Berg. Find Me (2015) , van den Berg’s first novel, takes place in a US devastated by a memory-destroying plague. “The Men and Women Like Him,” from Sparks’ 2016 collection The Unfinished World , depicts the unexpected agonies of a civilization that’s figured out time-travel.

Of the two, Sparks is the more ostensibly out there. Her three collections are full of specters, witches, and old gods reborn. She makes mischief with form and language, too; stories sprawl “like some strange, bloody, chaotic plant,” and several wouldn’t look out of place in an Escher exhibit. Yet if Sparks is playing tricks, the joke’s on us. Her fictions are animated by a prickly social consciousness, one you sense in the very title of her splendid latest, And I Do Not Forgive You (2020). One of that book’s best, “Everyone’s a Winner at Meadow Park,” may be a ghost story, but its howls are those of the downtrodden.

As for van den Berg, her materials are more down to earth, but nonetheless spooky. In her 2018 novel The Third Hotel , a recent widow winds up seducing her husband’s ghost⎯in Cuba, where she’s gone on dubious pretenses. In “Karolina,” from her 2020 collection I Hold a Wolf by the Ears , the ghost confronting another solo voyager is estranged family, half-crazy yet something of a Cassandra. Van den Berg sets all her women eerily adrift; even a dreary apartment complex can seem “a kind of purgatory where we docked until our souls were called elsewhere.”

And as in Sparks, these elements feel feminist: a demonstration of how quickly and callously women can be stripped of care and support. The resulting riddle of identity, van den Berg’s abiding conundrum, finds vital expression in wild verb coinages. A woman doesn’t just sigh over how close she and her friends used to be; “We wept secrets,” she mourns. “We eavesdropped nightmares.”

The displacement that haunts these young women feels very 21st-century. It’s the uneasiness of the refugee, really, a defining trauma for our time, and widespread in this country. On America’s margins, hardscrabble cultures express their turmoil in all sorts of novels and stories, and some of the best-known are the most unconventional. Viet Thanh Nguyen fled Vietnam as a child, and his violent, comedic refashioning of his bicultural experience, in The Sympathizer (2015), won the Pulitzer. The Cherokee writer Brandon Hobson has clearly studied Louise Erdrich’s crazy quilts of Lakota culture; his The Removed (2021) includes whole chapters of dissociative nightmare.

The dread has seeped into mainstream cultures as well. The fictions of Sparks or van den Berg look like cases in point, but a more striking one would be Lance Olsen, plainly of Northern European extraction and raised, as he puts it in his memoir [[there]] (2014), in a “bland foliaged suburb” of New Jersey. Nevertheless, his fiction (as well as his mélange of a memoir) has focused more and more on men, women, and the occasional monster who’ve lost their bearings. Some are bushwhacked by fate, flailing but most likely done for, while a few others rush headlong towards self-destruction. In every case, though, the text in which they turn up looks outrageous.

“Experimental writing,” as most people conceive the term, suits Olsen’s work better than that of all the writers I’m discussing. Percival Everett might object, but even he would admit Olsen has birthed some mooncalves, his typography all over the place and even the books themselves, in a case like the double-sided Theories of Forgetting (2014), oddly configured.

Those two authors also share a breathtaking prolixity. I’ll be getting to Everett, and Olsen’s latest, Skin Elegies (2021), brings his bibliography to thirty titles, mostly novels. Lately, these have asserted a fresh power. The author’s recent saturation in European culture, in particular his sojourns in Berlin, has deepened his sensitivity to human precariousness. Nothing so gnaws at his people, in these teeming later fictions, as the awareness “that every meeting is / the origin of a leaving.” The quote is from a Fukushima survivor, in Skin Elegies , texting about her harrowing escape. The entire novel teeters on the verge of death, in Fukushima, on the doomed shuttle Challenger , in a Swiss euthanasia clinic, and elsewhere. Each setting has its own layout and style, too, from lines of text to stream of consciousness.

Collage also provides the form in Olsen’s previous, My Red Heaven (2020), though the elements of this composition are very different, in keeping with a very different narrative surrogate. Red Heaven takes place over a single day in Berlin, 1927, and visits with everyone from Goebbels and Hitler to Kafka’s late-life wife Dora (a Jew who escaped the Holocaust, the novel reminds us) and the obscure and struggling Walter Benjamin (a Jew who didn’t get out; the novel flashes forward to his suicide). These and many others are all caught in the evanescence, the superlunary glow just before an eclipse; all suffer the giddiness Benjamin jots down in 1927, sounding as if it’s already 1940 and he’s taken his fatal dose of pills: “ I’m falling in love with my lostness.”

Olsen may have more powerful novels⎯ Calendar of Regrets (2010), possibly. Nevertheless, My Red Heaven and Skin Elegies stand like twin peaks worthy of the David Lynch reference: story-substitutes no one else could’ve brought off, and yet in their shock, life-giving.

Colson Whitehead has delivered plenty of shocks as well, though his apparatus doesn’t look so abnormal. Regardless of how he ranks as a rulebreaker, though, Whitehead’s into a stretch as stunning as DeLillo’s in the ‘80s and ‘90s. Nickel Boys (2019) may have picked up more awards than Underground Railroad, I can’t keep track, and Harlem Shuffle has so far done splendidly. More pertinent for this essay is how the most recent novels tame the author’s wild streak.

Both books present the America we know all too well, where anyone born the wrong color contends with Sisyphean economics and due process, and both keep their chronology straightforward. Granted, the three linked sequences of Shuffle conceal a trompe l’œil or two, and Nickel Boys has a metafictional aspect, with its constant notetaking and rumor-sharing. Even shackles tell a story: “The iron is still there. Rusty. Deep…. Testifying.” Still, when I call Whitehead a champion of the New Eccentrics, I’m thinking primarily of the work that began with his debut The Intuitionist (’99)⎯no doubt the world’s only novel of mystic elevator repair⎯and culminated in The Underground Railroad.

The narrative of Whitehead’s masterpiece depends on an imaginative leap now known to any Goodreads user, but previously unseen outside of steampunk: an actual refugee railway deep in the earth. Whitehead’s invention does largely without nuts and bolts, too, it’s dreamy around the edges, so that the lone correlative I think of comes in Song of Solomon, with its climactic discovery that “the people could fly.”

Certainly, Toni Morrison presides like fertile Demeter over the contemporary efflorescence of Black literature, but Whitehead takes risks all his own. He’ll apply a refined rhetorical balance to inhuman abuse: “the travesties so routine and familiar that they were like weather, and the ones so imaginative in their monstrousness that the mind refused to accommodate them.” He’ll adopt perspectives as antiphonal as an escaped slave and a “slave catcher,” also putting each viewpoint though acute reversals. Stories so far apart yet so entwined come to embody a core insight: “truth was a changing display in a shop window, manipulated… when you weren’t looking.” If there’s anything that can resist that manipulation, it’s the supernatural journey underground: “the miracle beneath. The miracle you made with your own sweat and blood.”

I could go on pulling citations, but Whitehead’s accomplishment is best appreciated not in its parts but as a whole⎯a novel. This one had forerunners (one thinks also of Ellison’s Invisible Man ), but Underground Railroad offers such a ride, bruising and mind-blowing, it creates a fresh model for imagining the tormented history of race.

Among the texts that share the model, a signal case would be Sing, Unburied, Sing (2017). Jessamyn Ward’s unhappy family is unhappy in its own way, to be sure, and her story’s set largely in the present. I doubt any of the novel’s rave reviews termed it “experimental.” Nevertheless, in this case too, the Deep South is a place of ghosts, with many a wrinkle in time.

Yet while the same shadows fall across a number of recent texts, Whitehead’s historical revisionism is by no means the rule, in Black fiction these days. No neighborhood in America is monolithic, anymore, and this applies on the aesthetic fringes as well. Both Paul Beatty’s The Sellout (2015) and Rivers Solomon’s Sorrowland (2021) rival Underground Railroad for praise, but they upend expectation in very different ways. Both honor a different ancestor, the acerbic Ishmael Reed, whipping up horror and satire very much of the moment. When Beatty imagines a Black man calling for the reinstatement of slavery, or Solomon a bisexual teen mother who defends herself by turning into a monster, these wild developments bristle with insight. And for still greater diversity within this writing community, check out the lengthening shelf of titles from Percival Everett.

Novels make up the majority of his work, and altogether, they display a carnivalesque flexibility. Even when Everett observes Aristotelian unities, he knocks them out of whack. A few titles, however, rival the Postmoderns at their most radical, and one of those ranks among his most celebrated: Percival Everett by Virgil Russell (2013).

The novel’s all stories within stories, with new hybrids sprouting just as you’d got the hang of the last. Most are sprinkled with nonsense rhymes or literary allusions, and these yield, in turn, yet more stories. The sheer imagination compels the reading, that and the way each whirling prose dervish puts a fresh spin on our mortality. The dedication is to the author’s late father, and the novel opens on a father in assisted living, in halting conversation with his son. Even the more surreal images convey a chill: “We can fall asleep in a room full of the snoring dead.”

Near the end, a mathematical formula appears, and while I’ve no idea what it means, I do see that it features the infinity symbol, and that it brackets, at beginning and end, a brief, berserk chapter. This begins: “The real question was whether there was some real value to which all of this, all of our naming, thinking, speaking, breathing, wanting, loving, lusting….” The three pages that follow (and close with the math widget) are nothing but present participles: the linguistic formula which signifies life.

An alternative fiction without parallel, if you ask me, Percival Everett also prompted the actual Everett to yet another surprise. His followup, So Much Blue (2017), works with rhetoric and sequencing you might call commonplace. The recollections of an aging painter, its identities remain stable and its prose solid Strunk & White. While interactions can get extreme, they’re never impossible; people even speak between quotation marks. Yet the text juggles its narratives oddly, keeping secrets that any ordinary drama would have to let out of the bag, and the resolution has an unsettling ambivalence. Besides that, So Much Blue shares with its predecessor something crucial about the protagonists. The men are Black (the father too, in the earlier novel), but this doesn’t much matter. The character’s color is never germane to any of the plotlines.

Now, this author has by no means ignored the subject, over his career. In 2009, I Am Not Sidney Poitier had a nasty laugh at what white folks expect from Black literature, and his latest, The Trees , exacts bloody new revenge for the murder of Emmett Till. Overall, though, Everett’s fiction isn’t yoked to the torments of racism. Rather, his project keeps raising further questions, wanting, loving, lusting… experimenting.

Which leaves him in excellent company, nowadays. Busy company, at that: with each succeeding book season, Americans of all backgrounds and orientations are finding new ways to warp Freytag’s Triangle. Naturally, their gathering momentum by no means steamrolls over the sturdy old structures of character and catharsis, rising action and climax. Nor is the structures’ toolkit, with items like scrupulous observation and psychological acuity, in danger of rusting. So long as novels matter, there will always be a place for Edward P. Jones or Mary Gaitskill. By the same token, if the alternatives are enjoying a Renaissance, sooner or later it’ll confront its Savonarola, building bonfires of the vanities. Some of the writers I’ve praised might even take as bad a pasting as the Postmoderns.

That earlier backlash, from which the dust still hasn’t settled, does seem one of the prompts for the current eruption. Smart young writers aren’t oblivious to what their culture approves and forbids, and inevitably, the outcasts start to look intriguing. Then too, MFA workshops have grown notorious for how they rein in the high-kickers, insisting grimly on ”show don’t tell” and “less is more.”

Even Lan Samantha Chang, director at Iowa, has complained about the constraints. She’s mounted her own small rebellion, too, this year in The Family Chao . But all that’s mere literature, how it’s taught, read, and critiqued. Art remains a response to the whole world, not just its texts, and for any Stateside talent born since the Baby Boom, our world suffers a terrible need for new perceptions and configurations. A hundred years after “The Waste Land,” hasn’t the devastation grown worse? Isn’t the air full of ghost birds? Aren’t streets crowded with the displaced? Then why not⎯to cite other milestones from the same earlier ‘22⎯ dream up a new Ulysses , or “The Hunger Artist,” or À la recherche du temps perdu ?

To think in such terms, to set whole centuries in balance, takes you to the issue of ultimate value. You wonder where, on the scales of human storytelling, the needle might land for an Amber Sparks, a Paul Beatty. But that’s another question, though a perfectly good one. I mean, it’s still America. But this is still American fiction⎯that’s my point. This is our own witching hour, full of strange cries and peculiar apparitions, and that seems to me worth a drink. It’s not nothing to once more show the world that this artform just can’t be contained.

John Domini

Previous article, next article.

• RSS - Posts

Literary Hub

Created by Grove Atlantic and Electric Literature

Sign Up For Our Newsletters

How to Pitch Lit Hub

Advertisers: Contact Us

Privacy Policy

Support Lit Hub - Become A Member

Become a Lit Hub Supporting Member : Because Books Matter

For the past decade, Literary Hub has brought you the best of the book world for free—no paywall. But our future relies on you. In return for a donation, you’ll get an ad-free reading experience , exclusive editors’ picks, book giveaways, and our coveted Joan Didion Lit Hub tote bag . Most importantly, you’ll keep independent book coverage alive and thriving on the internet.

Become a member for as low as $5/month

• NONFICTION BOOKS
• BEST NONFICTION 2023
• BEST NONFICTION 2024
• Historical Biographies
• The Best Memoirs and Autobiographies
• Philosophical Biographies
• World War 2
• World History
• American History
• British History
• Chinese History
• Russian History
• Ancient History (up to 500)
• Medieval History (500-1400)
• Military History
• Art History
• Travel Books
• Ancient Philosophy
• Contemporary Philosophy
• Ethics & Moral Philosophy
• Great Philosophers
• Social & Political Philosophy
• Classical Studies
• New Science Books
• Maths & Statistics
• Popular Science
• Physics Books
• Climate Change Books
• How to Write
• English Grammar & Usage
• Books for Learning Languages
• Linguistics
• Political Ideologies
• Foreign Policy & International Relations
• American Politics
• British Politics
• Religious History Books
• Mental Health
• Neuroscience
• Child Psychology
• Film & Cinema
• Opera & Classical Music
• Behavioural Economics
• Development Economics
• Economic History
• Financial Crisis
• World Economies
• Investing Books
• Artificial Intelligence/AI Books
• Data Science Books
• Sex & Sexuality
• Death & Dying
• Food & Cooking
• Sports, Games & Hobbies
• FICTION BOOKS
• BEST NOVELS 2024
• BEST FICTION 2023
• New Literary Fiction
• World Literature
• Literary Criticism
• Literary Figures
• Classic English Literature
• American Literature
• Comics & Graphic Novels
• Fairy Tales & Mythology
• Historical Fiction
• Crime Novels
• Science Fiction
• Short Stories
• South Africa
• United States
• Arctic & Antarctica
• Afghanistan
• Myanmar (Formerly Burma)
• Netherlands
• Kids Recommend Books for Kids
• High School Teachers Recommendations
• Prizewinning Kids' Books
• Popular Series Books for Kids
• BEST BOOKS FOR KIDS (ALL AGES)
• Ages Baby-2
• Books for Teens and Young Adults
• THE BEST SCIENCE BOOKS FOR KIDS
• BEST KIDS' BOOKS OF 2023
• BEST BOOKS FOR TEENS OF 2023
• Best Audiobooks for Kids
• Environment
• Best Books for Teens of 2023
• Best Kids' Books of 2023
• Political Novels
• New History Books
• New Historical Fiction
• New Biography
• New Memoirs
• New World Literature
• New Economics Books
• New Climate Books
• New Math Books
• New Philosophy Books
• New Psychology Books
• New Physics Books
• THE BEST AUDIOBOOKS
• Actors Read Great Books
• Books Narrated by Their Authors
• Best Audiobook Thrillers
• Best History Audiobooks
• Nobel Literature Prize
• Booker Prize (fiction)
• Baillie Gifford Prize (nonfiction)
• Financial Times (nonfiction)
• Wolfson Prize (history)
• Royal Society (science)
• Pushkin House Prize (Russia)
• Walter Scott Prize (historical fiction)
• Arthur C Clarke Prize (sci fi)
• The Hugos (sci fi & fantasy)
• Audie Awards (audiobooks)

Make Your Own List

The Best Fiction Books » Contemporary Fiction

The best experimental fiction, recommended by rebecca watson.

little scratch by Rebecca Watson

Experimental fiction often uses unusual forms of syntax, style, or form—perhaps taking the form of fragments, footnotes or parallel narratives. Here Rebecca Watson , author of the critically acclaimed experimental novel little scratch , recommends five of the best experimental novels and explains why a writer might choose to bend the rules—and to what effect.

Interview by Cal Flyn , Deputy Editor

When I Hit You: Or, A Portrait of the Writer as a Young Wife by Meena Kandasamy

The Lesser Bohemians by Eimear McBride

Between the Acts by Virginia Woolf

Diary of a Bad Year by J M Coetzee

Dept. of Speculation by Jenny Offill

1 When I Hit You: Or, A Portrait of the Writer as a Young Wife by Meena Kandasamy

2 the lesser bohemians by eimear mcbride, 3 between the acts by virginia woolf, 4 diary of a bad year by j m coetzee, 5 dept. of speculation by jenny offill.

T hanks for joining us on Five Books to discuss five of the best examples of experimental fiction. Could you tell us about your own novel little scratch , and its formal invention?

I guess the book is about the way that consciousness works; more specifically, present tense immediacy. That kind of consciousness. I think that it began with the challenge of feeling that prose does not really represent the bombardment and overwhelming simultaneity of everyday live experience. There’s so much going on. But when we write prose, we have this very neat, linear way in which we inhabit a moment. So my challenge was to represent the opposite of that on the page. Trauma can hyper-sensitise the ordinary, so it gave the experiment an extra charge.

Get the weekly Five Books newsletter

I split the page into channels. It breaks into two or three columns, goes between prose and half-prose. As you go down the page, you pass through time, and you have the internal, the external—sensory information, what you hear, what you smell—and how they, essentially, conflict with but also inform each other. It was kind of a game of association.

One of the things I found so interesting about it, and accurate—at least, according to my own experience of moving through the world—is how the interior monologue can be the defining element of a day. A theatre of drama, or conflict, when outwardly nothing is really happening. But as well as this issue of trauma it also struck me that your book was also fun , in how imaginative it was and how it told little jokes to itself. Do you think that’s important in experimental fiction, that sense of play?

Although perhaps ‘play’ is not the right word for the first book we’re going to discuss? This is When I Hit You by Meena Kandasamy. It’s about quite extreme domestic abuse; a woman marries an Indian intellectual who uses his ideas “as a cover for his own sadism.”

Um, actually I think you can use the word play. It’s uncomfortable, but the beauty of what she does is that she takes an incredibly disturbing centrepiece—this abusive marriage—and turns it into a creative challenge, and a performance.

It’s fiction, with an unnamed first-person narrator, and the narrator is telling the story retrospectively, about five years after these few months of marriage. It’s about how to tell the story: how do you go back and look at something so invasive, so encompassing, while reclaiming, or asserting, your sense of self.

“Experimental writing needs an openness and willingness from a reader, to go beyond what you might be used to”

What is the centre of the book is a playfulness about what it is to be a writer, and what it is to live a life—how we write narratives. And she has this very playful voice that shifts and flirts and diverts. It moves in lots of different registers.

I don’t think it looks experimental on the page. It’s the shifting register that makes it feel really, really new. Kandasamy studies the different vehicles of how to tell a story. The novel moves between prose poetry, to bits where it’s like a Q&A. The narrator is simultaneously the actor who breaks the fourth wall, and the writer dictating the stage directions. So even though we’re looking at this abusive marriage, most of the time this person is almost testing you to see if she can make you laugh, think, shift your expectations of a ‘victim’.

That’s interesting. I’ve actually come to Kandasamy backwards, I think. The first book of hers I read was her novella-memoir hybrid, Exquisite Cadavers . Which was written, I think, in response to how the earlier, highly garlanded novel was received. If When I Hit You does not appear overtly experimental on the page, Exquisite Cadavers certainly is—it’s split into two columns, one strand of which is fictional and the second a sort of metafictional commentary that reflects on her life and the writing process. One expects quite a lot of the reader, when writing in such a form. Or maybe you disagree?

I think we act like there’s more we need from the reader than we actually do. That’s the barrier for people getting into this kind of writing. I think often experimental fiction is used as a warning term rather than as a way of elaborating what the writing is.

Absolutely. The second book you’ve chosen is The Lesser Bohemians by Eimear McBride.

I had to look at this again yesterday, because although I have such a vivid recollection of the feeling of reading it, I couldn’t remember anything about it on a line level! I opened it and thought, ‘wow, so this is what it looks like.’ I guess because it’s so voice-led, you completely enter that head. It’s true consciousness. McBride’s really good at, like, skipping to the image of the association—the kind of narrative preamble, or narrative signposting, that writers often give she will skip, not because she’s trying to be cryptic, but because in real consciousness none of those things exist. You’ve instantly leapt to the next thought or sensation.

It’s a love story between an 18-year-old girl who just moved from Ireland to London to attend drama school, and the older actor she meets in a pub. They slowly unveil their stories to each other, but both are hiding parts of themselves.

Partly it’s about the unknowability of the other, but also the ability to learn so much through love. The lyricism of it is just something else. There’s such a music, such a lilt to it. There’s rhythm and movement to how you read it, any reader would get that. And that’s something incredible to establish.

A lot of people have remarked upon the sexually explicit nature of this book. But one wonders if sex is not perhaps the perfect experience to be rendered as stream of consciousness . I guess I’m also thinking of certain scenes in Ulysses , which McBride herself has talked about as a prime literary influence. But does McBride bring a new approach to this form?

Experimental fiction both points forward and back. We often describe things as new, when they are speaking as part of a tradition. But there is a noticeable newness to this book, and it’s in the deftness of language and the immersion—you feel so entirely in this person’s head whilst also being so clearly told a story. The way she gets inside people’s bodies, and writes sex in such a brilliant, honest and felt way. I think that is genuinely new. And if sex is something that is more instinctive, relies more on sensation and the unifying of desire and the body, then yes, surely experimental fiction serves that best: like sex, it defies a linear, straightforward recounting.

It was a highly anticipated follow up to her very acclaimed debut A Girl is a Half-Formed Thing . How similar are they—as books, as reading experiences?

Next up on our experimental fiction reading list is a Virginia Woolf book I admit I wasn’t familiar with: Between the Acts , her final novel which was published posthumously.

Yes, there’s a split over whether it was finished or not. She never did her final revise of it, and often she did do quite a lot of work in that act of revision. But it doesn’t read like something unfinished. It’s very, very accomplished, and one of my favourite of Woolf’s books.

It takes place at a village pageant in the summer preceding World War Two. I read it as a teenager and was so obsessed with it. I remember getting into an argument with a literature professor at an interview for a university that I very much got rejected from. It was so weird. We both loved Virginia Woolf, but both had very different visions of what that book was. I was young, so sure! I was the naïve one, more likely to be wrong but, at the same time, I remember this man puffing out—because I disagreed with him.

What I was making the case for—and what I still agree with—is how obsessed that book is with the idea of individualism versus society, the exhausting nature of being one person whilst also being so recognisably within the midst of a group, connected to other people.

Support Five Books

Five Books interviews are expensive to produce. If you're enjoying this interview, please support us by donating a small amount .

I remember him being like, ‘it’s about the war!’ And I was like, I get that it’s about the war, but it’s also about these other things. We just couldn’t connect. I mean, it is about the war,  to be fair—so much of it is about the burden of retrospect. It’s just before the war arrives, but they know the war is coming. Another war’s happened already. And so everything’s laced with this very aggressive, violent imagery. There’s something simmering. It’s all about living in what seems a simple present tense that’s about to combust into historical significance.

Let’s talk about J. M. Coetzee’s Diary of a Bad Year next. This is another one that’s pretty interesting to look at on the page. It’s separated into parallel narratives; could you tell us a little about the plot or rather plots, plural?

It begins with the page split into two, horizontally. We have on the top an extract from a book by an old Australian writer, called Strong Opinions . It’s political essays, stuff like that. That’s the top narrative, and initially the bottom narrative is him, first-person, his life as he’s writing the manuscript. Quickly he meets a woman called Anya, in the laundry room of his apartment building. He’s very struck by her, she’s very attractive, she’s a lot younger… there’s something about her that intrigues him. So he asks her to start typing up his manuscript for the book, the first narrative.

Suddenly she appears as a third column, or rather a horizontal division. So now you have the book, his first-person, her first-person. And each informs the others. So you see two perspectives – how she sees him, and how he sees her. He embroils her in his life, but she also embroils herself, because she and her boyfriend start to plot how to insinuate themselves. He’s got a lot of money—maybe they could get some of his cash.

“It sounds very complicated, but it’s incredibly easy to read, and it’s very profound”

You have these interfering narratives, and you can read across the double spread of the page, or down each page. Depending on which way you do it, you learn different things about each of the segments. That sounds very complicated, but it’s incredibly easy to read, and it’s very profound. This guy’s dying, so it’s about being a writer, how to make meaning, and the difference between the creator and the product—seeing this fallible human being behind the authoritative essay.

It does also have a very propulsive story. A classic narrative, of someone thinking they are intentionally bringing someone into their life, and at the same time they are being undone. Who has the power?

Mmm, yes. And this three-stranded form really represents exactly what you were getting at in the start of our conversation: the clarity of prose, the mess of reality. But one thing I suppose I worry about—and maybe other readers worry about this too—when you mention there are multiple ways of reading this book, I guess that allows for the possibility of reading it ‘wrong.’

I think that is often the reader’s fear. I always want to say to people: ‘Hey, don’t worry. Just trust yourself.’ That’s what I say to myself, sometimes. Readers often fear they don’t have the authority to tackle a text. Or just see it as something they have to ‘tackle.’ But a writer doesn’t write something with an authoritative insistence of how it should be read. They lay out a path of how to read it, and you as a reader will follow that – or make your own. little scratch , particularly, exists to be read in different ways, and the reader is meant to make their own choices. Depending on what choice they make, they get different things. That’s the same for Diary of a Bad Year ; those decisions are important, but they’re meant to be fun. They’re not meant to be a stress. The reader should remember that once the book is in their hands, they’re the authority. It’s their reading experience, they’re in control.

Just before we move on, I wanted to read you a quote from the Guardian : “The ensuing comedy of conflicting perspectives, of high rhetoric and low aims, is an amazingly strange thing for Coetzee to have decided to write.” What do you think about that? Is invention itself the aim?

I don’t know. I didn’t think it was strange. I mean maybe, in as much that it’s kind of crazy that it came out of someone’s head. But who are we to say what a writer will write next? And why would we want that to be predictable? The novel is exciting in its formal invention, but as a story in its own right, it’s interesting. The form is a way of getting you closer to the story. It’s not an indulgent thing.

Right, this brings us to our final work of experimental fiction. This is Jenny Offill’s Dept. of Speculation , one of my own top favourite books. I think I’ve read it, honestly, upwards of 100 times. I carry it around on my phone, on my Kindle app, and just reread it on the bus, or anywhere. Do you feel the same? What do you like about it?

It’s quite a hard book to summarise, because it’s so much about those myriad moments that make up a life, or relationship, or anything. Firstly, it’s really funny. People don’t talk directly enough about how funny Jenny Offill is.

It’s essentially about a test of marriage, a kind of imperfect love story. Right? It’s both very sad and very funny.

Right. And it’s told in fragments, each of which are a sentence, or a paragraph, long.

You have these moments between the fragments to take a break, just to laugh, or absorb it. She takes the ordinariness of life and she makes it mean something. I think she proves why every moment of life can have significance, and how the quiet moments of existence contribute to our sense of self.

Last year, when her new book Weather came out, there was a picture of the fragments of Weather , how she’d rearrange them. Seeing that patchwork was amazing, because you can forget when a product is complete that so much work goes into the order of these things. Even just choosing the right ones… there will be so many rejected fragments. It affirmed the level of precision with which she works in order to create these novels.

She’s said that the gaps are moments for the reader to have an imagination. She doesn’t want to fill in the gaps. Because in doing so you eliminate or pin the story down. By fragmenting you allow the reader to immerse themselves in, be part of the world in a more intense way. The absences give the created world a greater imaginative potential.

I really enjoy fragmented fiction, especially this book. But as a style it’s recently become common enough—perhaps riding the wave of Offill’s success—to have inspired a rather funny satire of the form in Lauren Oyler’s new novel Fake Accounts . And of course, you’ve found mainstream success with little scratch. Do you feel like experimental fiction is becoming more influential, more popular?

I don’t know. I think that the fact that little scratch was published and accepted and treated as a novel speaks to a healthy publishing culture. Certainly there’s more space for it and, I think, more commercial viability, which is the key sign that people are open to it.

There’s still a lot of pushback. There’s still a strange treatment of experimental writing. And, you know, when I see people talking about my book, it’s the first thing they do, right? They say, ‘okay, guys, this looks weird, but don’t worry, you’re going to be able to get into it.’ There’s an apology at the beginning. I find that strange. ‘Experimental’ has taken on this negative association. It’s something that we have to forgive, that you can get something from despite it.

I think that kind of negates the whole purpose of this type of writing, which is to help immerse the reader further in the story. It serves a purpose. It’s there to do something beyond looking funny. It’s meant to open up new possibilities. I find it sad that it is often seen as a boundary. We are still very obsessed with keeping that boundary, putting the signpost up. I think we need to work on that.

Do you feel that is anti-intellectual, somehow?

Maybe? I think some readers fear they are being pushed away, or that it’s coded for a certain level of education, rather than for a general reader—which I don’t think is true.

But I think what you were saying earlier about being scared of not approaching the text ‘right’—those kind of insecurities, it’s not that formal or informal education helps you read better, but that maybe it gives you a level of self-trust. Sometimes we need an ego to trust ourselves as a reader. But it’s too complicated to make generalisations. Some think ‘experimental’ fiction is elitist, like it’s trying to shut out a certain reader. But all I can say for myself is that I’m writing for all readers. And I hope little scratch will show some readers who might feel hesitant over formally inventive writing, that shaking up the page can sometimes be a more natural way to read. It might bring you closer, rather than push you away.

April 12, 2021

Five Books aims to keep its book recommendations and interviews up to date. If you are the interviewee and would like to update your choice of books (or even just what you say about them) please email us at [email protected]

Rebecca Watson

Rebecca Watson is the author of little scratch , which is published by Faber in the UK and Doubleday in the US. She was picked as one of The Observer 's ten best debut novelists of 2021. Her work has been published in The TLS, The Guardian, Granta and elsewhere. In 2018, she was shortlisted for the White Review Short Story Prize. She works part-time as assistant arts editor at the Financial Times and lives in London.

We ask experts to recommend the five best books in their subject and explain their selection in an interview.

This site has an archive of more than one thousand seven hundred interviews, or eight thousand book recommendations. We publish at least two new interviews per week.

Five Books participates in the Amazon Associate program and earns money from qualifying purchases.

© Five Books 2024

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base

Methodology

• Guide to Experimental Design | Overview, Steps, & Examples

Guide to Experimental Design | Overview, 5 steps & Examples

Published on December 3, 2019 by Rebecca Bevans . Revised on June 21, 2023.

Experiments are used to study causal relationships . You manipulate one or more independent variables and measure their effect on one or more dependent variables.

Experimental design create a set of procedures to systematically test a hypothesis . A good experimental design requires a strong understanding of the system you are studying.

There are five key steps in designing an experiment:

• Consider your variables and how they are related
• Write a specific, testable hypothesis
• Design experimental treatments to manipulate your independent variable
• Assign subjects to groups, either between-subjects or within-subjects
• Plan how you will measure your dependent variable

For valid conclusions, you also need to select a representative sample and control any  extraneous variables that might influence your results. If random assignment of participants to control and treatment groups is impossible, unethical, or highly difficult, consider an observational study instead. This minimizes several types of research bias, particularly sampling bias , survivorship bias , and attrition bias as time passes.

Table of contents

Step 1: define your variables, step 2: write your hypothesis, step 3: design your experimental treatments, step 4: assign your subjects to treatment groups, step 5: measure your dependent variable, other interesting articles, frequently asked questions about experiments.

You should begin with a specific research question . We will work with two research question examples, one from health sciences and one from ecology:

To translate your research question into an experimental hypothesis, you need to define the main variables and make predictions about how they are related.

Start by simply listing the independent and dependent variables .

Then you need to think about possible extraneous and confounding variables and consider how you might control  them in your experiment.

Finally, you can put these variables together into a diagram. Use arrows to show the possible relationships between variables and include signs to show the expected direction of the relationships.

Here we predict that increasing temperature will increase soil respiration and decrease soil moisture, while decreasing soil moisture will lead to decreased soil respiration.

Prevent plagiarism. Run a free check.

Now that you have a strong conceptual understanding of the system you are studying, you should be able to write a specific, testable hypothesis that addresses your research question.

The next steps will describe how to design a controlled experiment . In a controlled experiment, you must be able to:

• Systematically and precisely manipulate the independent variable(s).
• Precisely measure the dependent variable(s).
• Control any potential confounding variables.

If your study system doesn’t match these criteria, there are other types of research you can use to answer your research question.

How you manipulate the independent variable can affect the experiment’s external validity – that is, the extent to which the results can be generalized and applied to the broader world.

First, you may need to decide how widely to vary your independent variable.

• just slightly above the natural range for your study region.
• over a wider range of temperatures to mimic future warming.
• over an extreme range that is beyond any possible natural variation.

Second, you may need to choose how finely to vary your independent variable. Sometimes this choice is made for you by your experimental system, but often you will need to decide, and this will affect how much you can infer from your results.

• a categorical variable : either as binary (yes/no) or as levels of a factor (no phone use, low phone use, high phone use).
• a continuous variable (minutes of phone use measured every night).

How you apply your experimental treatments to your test subjects is crucial for obtaining valid and reliable results.

First, you need to consider the study size : how many individuals will be included in the experiment? In general, the more subjects you include, the greater your experiment’s statistical power , which determines how much confidence you can have in your results.

Then you need to randomly assign your subjects to treatment groups . Each group receives a different level of the treatment (e.g. no phone use, low phone use, high phone use).

You should also include a control group , which receives no treatment. The control group tells us what would have happened to your test subjects without any experimental intervention.

When assigning your subjects to groups, there are two main choices you need to make:

• A completely randomized design vs a randomized block design .
• A between-subjects design vs a within-subjects design .

Randomization

An experiment can be completely randomized or randomized within blocks (aka strata):

• In a completely randomized design , every subject is assigned to a treatment group at random.
• In a randomized block design (aka stratified random design), subjects are first grouped according to a characteristic they share, and then randomly assigned to treatments within those groups.

Sometimes randomization isn’t practical or ethical , so researchers create partially-random or even non-random designs. An experimental design where treatments aren’t randomly assigned is called a quasi-experimental design .

Between-subjects vs. within-subjects

In a between-subjects design (also known as an independent measures design or classic ANOVA design), individuals receive only one of the possible levels of an experimental treatment.

In medical or social research, you might also use matched pairs within your between-subjects design to make sure that each treatment group contains the same variety of test subjects in the same proportions.

In a within-subjects design (also known as a repeated measures design), every individual receives each of the experimental treatments consecutively, and their responses to each treatment are measured.

Within-subjects or repeated measures can also refer to an experimental design where an effect emerges over time, and individual responses are measured over time in order to measure this effect as it emerges.

Counterbalancing (randomizing or reversing the order of treatments among subjects) is often used in within-subjects designs to ensure that the order of treatment application doesn’t influence the results of the experiment.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

Finally, you need to decide how you’ll collect data on your dependent variable outcomes. You should aim for reliable and valid measurements that minimize research bias or error.

Some variables, like temperature, can be objectively measured with scientific instruments. Others may need to be operationalized to turn them into measurable observations.

• Ask participants to record what time they go to sleep and get up each day.
• Ask participants to wear a sleep tracker.

How precisely you measure your dependent variable also affects the kinds of statistical analysis you can use on your data.

Experiments are always context-dependent, and a good experimental design will take into account all of the unique considerations of your study system to produce information that is both valid and relevant to your research question.

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Student’s  t -distribution
• Normal distribution
• Null and Alternative Hypotheses
• Chi square tests
• Confidence interval
• Cluster sampling
• Stratified sampling
• Data cleansing
• Reproducibility vs Replicability
• Peer review
• Likert scale

Research bias

• Implicit bias
• Framing effect
• Cognitive bias
• Placebo effect
• Hawthorne effect
• Hindsight bias
• Affect heuristic

Experimental design means planning a set of procedures to investigate a relationship between variables . To design a controlled experiment, you need:

• A testable hypothesis
• At least one independent variable that can be precisely manipulated
• At least one dependent variable that can be precisely measured

When designing the experiment, you decide:

• How you will manipulate the variable(s)
• How you will control for any potential confounding variables
• How many subjects or samples will be included in the study
• How subjects will be assigned to treatment levels

Experimental design is essential to the internal and external validity of your experiment.

The key difference between observational studies and experimental designs is that a well-done observational study does not influence the responses of participants, while experiments do have some sort of treatment condition applied to at least some participants by random assignment .

A confounding variable , also called a confounder or confounding factor, is a third variable in a study examining a potential cause-and-effect relationship.

A confounding variable is related to both the supposed cause and the supposed effect of the study. It can be difficult to separate the true effect of the independent variable from the effect of the confounding variable.

In your research design , it’s important to identify potential confounding variables and plan how you will reduce their impact.

In a between-subjects design , every participant experiences only one condition, and researchers assess group differences between participants in various conditions.

In a within-subjects design , each participant experiences all conditions, and researchers test the same participants repeatedly for differences between conditions.

The word “between” means that you’re comparing different conditions between groups, while the word “within” means you’re comparing different conditions within the same group.

An experimental group, also known as a treatment group, receives the treatment whose effect researchers wish to study, whereas a control group does not. They should be identical in all other ways.

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

Bevans, R. (2023, June 21). Guide to Experimental Design | Overview, 5 steps & Examples. Scribbr. Retrieved August 9, 2024, from https://www.scribbr.com/methodology/experimental-design/

Is this article helpful?

Rebecca Bevans

Other students also liked, random assignment in experiments | introduction & examples, quasi-experimental design | definition, types & examples, how to write a lab report, get unlimited documents corrected.

✔ Free APA citation check included ✔ Unlimited document corrections ✔ Specialized in correcting academic texts

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, automatically generate references for free.

• Knowledge Base
• Methodology
• A Quick Guide to Experimental Design | 5 Steps & Examples

A Quick Guide to Experimental Design | 5 Steps & Examples

Published on 11 April 2022 by Rebecca Bevans . Revised on 5 December 2022.

Experiments are used to study causal relationships . You manipulate one or more independent variables and measure their effect on one or more dependent variables.

Experimental design means creating a set of procedures to systematically test a hypothesis . A good experimental design requires a strong understanding of the system you are studying. 

There are five key steps in designing an experiment:

• Consider your variables and how they are related
• Write a specific, testable hypothesis
• Design experimental treatments to manipulate your independent variable
• Assign subjects to groups, either between-subjects or within-subjects
• Plan how you will measure your dependent variable

For valid conclusions, you also need to select a representative sample and control any  extraneous variables that might influence your results. If if random assignment of participants to control and treatment groups is impossible, unethical, or highly difficult, consider an observational study instead.

Table of contents

Step 1: define your variables, step 2: write your hypothesis, step 3: design your experimental treatments, step 4: assign your subjects to treatment groups, step 5: measure your dependent variable, frequently asked questions about experimental design.

You should begin with a specific research question . We will work with two research question examples, one from health sciences and one from ecology:

To translate your research question into an experimental hypothesis, you need to define the main variables and make predictions about how they are related.

Start by simply listing the independent and dependent variables .

Then you need to think about possible extraneous and confounding variables and consider how you might control  them in your experiment.

Finally, you can put these variables together into a diagram. Use arrows to show the possible relationships between variables and include signs to show the expected direction of the relationships.

Here we predict that increasing temperature will increase soil respiration and decrease soil moisture, while decreasing soil moisture will lead to decreased soil respiration.

Prevent plagiarism, run a free check.

Now that you have a strong conceptual understanding of the system you are studying, you should be able to write a specific, testable hypothesis that addresses your research question.

The next steps will describe how to design a controlled experiment . In a controlled experiment, you must be able to:

• Systematically and precisely manipulate the independent variable(s).
• Precisely measure the dependent variable(s).
• Control any potential confounding variables.

If your study system doesn’t match these criteria, there are other types of research you can use to answer your research question.

How you manipulate the independent variable can affect the experiment’s external validity – that is, the extent to which the results can be generalised and applied to the broader world.

First, you may need to decide how widely to vary your independent variable.

• just slightly above the natural range for your study region.
• over a wider range of temperatures to mimic future warming.
• over an extreme range that is beyond any possible natural variation.

Second, you may need to choose how finely to vary your independent variable. Sometimes this choice is made for you by your experimental system, but often you will need to decide, and this will affect how much you can infer from your results.

• a categorical variable : either as binary (yes/no) or as levels of a factor (no phone use, low phone use, high phone use).
• a continuous variable (minutes of phone use measured every night).

How you apply your experimental treatments to your test subjects is crucial for obtaining valid and reliable results.

First, you need to consider the study size : how many individuals will be included in the experiment? In general, the more subjects you include, the greater your experiment’s statistical power , which determines how much confidence you can have in your results.

Then you need to randomly assign your subjects to treatment groups . Each group receives a different level of the treatment (e.g. no phone use, low phone use, high phone use).

You should also include a control group , which receives no treatment. The control group tells us what would have happened to your test subjects without any experimental intervention.

When assigning your subjects to groups, there are two main choices you need to make:

• A completely randomised design vs a randomised block design .
• A between-subjects design vs a within-subjects design .

Randomisation

An experiment can be completely randomised or randomised within blocks (aka strata):

• In a completely randomised design , every subject is assigned to a treatment group at random.
• In a randomised block design (aka stratified random design), subjects are first grouped according to a characteristic they share, and then randomly assigned to treatments within those groups.

Sometimes randomisation isn’t practical or ethical , so researchers create partially-random or even non-random designs. An experimental design where treatments aren’t randomly assigned is called a quasi-experimental design .

Between-subjects vs within-subjects

In a between-subjects design (also known as an independent measures design or classic ANOVA design), individuals receive only one of the possible levels of an experimental treatment.

In medical or social research, you might also use matched pairs within your between-subjects design to make sure that each treatment group contains the same variety of test subjects in the same proportions.

In a within-subjects design (also known as a repeated measures design), every individual receives each of the experimental treatments consecutively, and their responses to each treatment are measured.

Within-subjects or repeated measures can also refer to an experimental design where an effect emerges over time, and individual responses are measured over time in order to measure this effect as it emerges.

Counterbalancing (randomising or reversing the order of treatments among subjects) is often used in within-subjects designs to ensure that the order of treatment application doesn’t influence the results of the experiment.

Finally, you need to decide how you’ll collect data on your dependent variable outcomes. You should aim for reliable and valid measurements that minimise bias or error.

Some variables, like temperature, can be objectively measured with scientific instruments. Others may need to be operationalised to turn them into measurable observations.

• Ask participants to record what time they go to sleep and get up each day.
• Ask participants to wear a sleep tracker.

How precisely you measure your dependent variable also affects the kinds of statistical analysis you can use on your data.

Experiments are always context-dependent, and a good experimental design will take into account all of the unique considerations of your study system to produce information that is both valid and relevant to your research question.

Experimental designs are a set of procedures that you plan in order to examine the relationship between variables that interest you.

To design a successful experiment, first identify:

• A testable hypothesis
• One or more independent variables that you will manipulate
• One or more dependent variables that you will measure

When designing the experiment, first decide:

• How your variable(s) will be manipulated
• How you will control for any potential confounding or lurking variables
• How many subjects you will include
• How you will assign treatments to your subjects

The key difference between observational studies and experiments is that, done correctly, an observational study will never influence the responses or behaviours of participants. Experimental designs will have a treatment condition applied to at least a portion of participants.

A confounding variable , also called a confounder or confounding factor, is a third variable in a study examining a potential cause-and-effect relationship.

A confounding variable is related to both the supposed cause and the supposed effect of the study. It can be difficult to separate the true effect of the independent variable from the effect of the confounding variable.

In your research design , it’s important to identify potential confounding variables and plan how you will reduce their impact.

In a between-subjects design , every participant experiences only one condition, and researchers assess group differences between participants in various conditions.

In a within-subjects design , each participant experiences all conditions, and researchers test the same participants repeatedly for differences between conditions.

The word ‘between’ means that you’re comparing different conditions between groups, while the word ‘within’ means you’re comparing different conditions within the same group.

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the ‘Cite this Scribbr article’ button to automatically add the citation to our free Reference Generator.

Bevans, R. (2022, December 05). A Quick Guide to Experimental Design | 5 Steps & Examples. Scribbr. Retrieved 5 August 2024, from https://www.scribbr.co.uk/research-methods/guide-to-experimental-design/

Is this article helpful?

Rebecca Bevans

6 Great Books That Experiment With Text and Genre

Natalie Layne Baker

Natalie Layne Baker's writing has appeared at Audible, Hachette, Book Riot, Submittable, Entropy, Memoir Mixtapes, Howl Round, and Bone & Ink Lit Zine. She currently resides in Philadelphia.

View All posts by Natalie Layne Baker

Like any other medium, books have their tried and true conventions. Generally speaking, when one sits down with a novel or an essay collection, they can expect blocks of text broken up into paragraphs, sections, chapters, and so on. There is nothing wrong with having expectations of how a reading experience will go, otherwise reading would be a chaotic and inaccessible endeavor. But like any series of conventions, when consumed in rapid succession with no variation, curious readers may find themselves wanting a little more spice. For those occasions, there are a plethora of experimental books that experiment with text, genre, and structure to choose from.

City on Fire by Garth Risk Hallberg

The sale of Hallberg’s—a relatively unknown fiction writer—debut novel for seven figures was a huge event in publishing a few years back. It is not the kind of deal one encounters commonly, where conventional wisdom says debut literary novels should be a tight 80,000–100,000 words at most. However, all the buzz about the book’s contract tended to overshadow just how engaging and unexpected the journey through City on Fire is.

Following the intertwining plots of several New Yorkers in the days leading up to the 1977 blackout, City on Fire contains many passages of textual experimentation. Within its 900 pages, you will find tear-stained, handwritten letters, a lengthy fictional investigative report on people who design fireworks displays, and a fully laid out punk rock zine. Despite the novel’s length, Hallberg’s prose is neither dense nor overly obfuscated. Thus, City on Fire is a fantastic and accessible novel for readers looking to add a twist to their to-be-read pile of experimental books.

Some of Us Are Very Hungry Now by Andre Perry

Multiple choice questions may elicit feelings of dread, dragging up memories of school testing that those of us who are no longer students may wish to leave behind us. They’re certainly not something that most might think to use as literary devices, yet that’s exactly what Andre Perry does, among other textual experimentations, in Some of Us Are Very Hungry Now .

Forthcoming from  Two Dollar Radio  in Columbus, Ohio, Perry’s essay collection contains personal, deeply felt ruminations on identity, racism, and belonging. Other genre-bending flourishes include passages formatted as screenplays and transcripts of imaginary talk-show interviews. The boldness of these forays from the expectations of an essay collection make the passages in which Perry writes in straightforward prose all the more stark and arresting.

Oreo by Fran Ross

Following the journey of a mixed race girl searching for self-actualization, Oreo is a picaresque novel that experiments with text and structure. Set mostly in New York City, it is modeled after the myth of Theseus, as the title character searches for her Jewish father.

With its episodic structure, Oreo ‘s narrative is purposefully fragmented. Its chapters are divided into subsections, much like a textbook, and in its pages appear diagrams, restaurant menus, advertisements, and tests for the reader. Published in 1974, Oreo is a prime example of postmodern literature made famous by writers like Vonnegut and Pynchon. However, it did not achieve similar widespread attention and went out of print. Thankfully, it was rediscovered in the 2000s and has since become recognized as a postmodern masterpiece.

Monster by Walter Dean Myers

Lest we believe nontraditional text can only be found in adult literature, Monster is a young adult novel formatted almost entirely as a screenplay, with interludes from a handwritten diary. The story follows Steve, a 16-year-old boy who is on trial for murder after a robbery gone wrong, to which it is uncertain if he has any plausible connection. Steve’s interest in filmmaking motivates the novel’s structural choice, and the interplay between the stylized descriptions of shots and voiceovers with his diary entries builds toward a stunning conclusion. I received a copy of Monster from a book fair in middle school, and it was the first of many books that experiment with text I encountered.

Special Topics in Calamity Physics by Marisha Pessl

Literary fiction is a diverse genre when it comes to authorial voice and novel structure. Across novels, as with any genre, one can spy certain common themes. Domestic anxiety, infidelity, financial stress, untimely death. Marisha Pessl’s Special Topics in Calamity Physics explores this familiar territory, but turns it on its head by formatting the novel as an English syllabus.

Each chapter of Special Topics is linked to a source material, like Othello and Wuthering Heights . Throughout, the text is filled with references in footnotes, many of which lead to fictional sources. Reading this novel is an exercise in shifting your expectations.

Dictee by Theresa Hak Kyung Cha

Sometimes, structural and textual experiments can lead to dense and challenging literature. There is a fine line between challenging and willfully obscure, the latter of which can lead to frustrating reading experiences in which the reader might toss the book aside, believing the author is only interested in being weird for weird’s sake.

A shining example of the former—a book whose density and difficulty makes it rewarding rather than infuriating—is Theresa Hak Kyung Cha’s Dictee . Unlike many literary experiments, Dictee is a slim book, which nevertheless contains a wide range of textual and visual experiments. Photographs and letters appear with little or no context. There are long passages in which a character’s inability to form a coherent sentence is described. Sentences appear in English, French, Korean, and Chinese, displaying the author’s multilingualism and creating a visual experience as much as a linguistic one.

Dictee is not light reading, and its context is heartbreaking— the author was murdered shortly after its publication. However, it is an essential read for anyone seeking out experimental books.

These six are just a small sample of experimental books that experiment with text, genre, structure, and form. Want more strange and unusual books? We got you .

You Might Also Like

• International edition
• Australia edition
• Europe edition

Experimental fiction: is it making a comeback?

A couple of years ago, Zadie Smith wrote an essay in the New York Review of Books comparing Tom McCarthy's Remainder (see below) with Joseph O'Neill's acclaimed novel about cricket in post-9/11 New York, Netherland . As the essay's title – " Two Paths for the Novel " – suggested, Smith saw the two books as exemplifying competing strands within western literature: Netherland was a "lyrical realist" novel in the mould of Balzac and Flaubert, while Remainder was heir to the works of 20th-century experimentalists ranging from Joyce and Kafka to Donald Barthelme and William Gaddis.

In healthy times, Smith said, these two traditions – the realist and the avant garde – would comfortably co-exist. But "these aren't particularly healthy times", and one reason for this is that the experimentalist tradition has been "relegated to a safe corner of literary history", dismissed as a "fascinating failure". As Smith put it: "A breed of lyrical realism has had the freedom of the highway for some time now, with most other exits blocked." In order for our literary culture to rebalance itself, she suggested, more writers need to follow McCarthy in attempting novels that set out to challenge the dominant realist mode.

Whether or not one agrees with her assessment of Netherland and Remainder , it's hard to quibble with Smith's contention that avant-garde fiction, at least in Britain and America, isn't flourishing. For many, the death of David Foster Wallace in 2008 represented the end point of a project that had become synonymous with obscurantism, pretentiousness and boredom. In Britain, which has its own lively tradition of literary experimentation, ranging from Virginia Woolf's excursions into consciousness to BS Johnson's eccentric games with form (including his notorious 1969 "shuffle" novel, The Unfortunates ) there is little sense of mourning for the passing of the avant garde. As Smith noted, a kind of fatalism has entered our literary culture, a sense that all other routes have been tested and found wanting. The well-made realist novel, inherited from the 19th century, is what we are stuck with now, and even if we aren't excessively fond of it, it seems to be pretty much all we have.

But is this the case? The assumption that genuine experimentation is no longer possible is in many ways a parochial quirk of the anglophone world. Things are very different, for example, in Latin America, where anti-realist techniques have long been part of the mainstream, and where the recent success of writers such as Roberto Bolaño and César Aira (see right) shows that novelists can still be lauded for striking out in new ways. France may not be the hotbed of literary radicalism that it once was, but the avant-garde tradition represented by the likes of Georges Perec – famous, among much else, for writing a novel without the letter "e" – continues to be venerated.

Closer to home, there are a notable few who remain committed to experimentation. David Mitchell, the author of Cloud Atlas , is hardly an avant-garde figure – his novels are too mainstream for that – but with his twisting, time-bending narratives, he is a high-profile example of someone doing something different. And this month there's a double reminder that experimentation is still possible, with the publication of Tom McCarthy's new novel, C , and the Collected Stories of American author and translator Lydia Davis. McCarthy and Davis are in many ways antithetical figures: while McCarthy is something of a literary showman, a disseminator of maverick manifestos, Davis keeps herself out of public view, and offers few explanations for what she is doing. But in their different ways, both writers help us see that, where fiction is concerned, it is a mistake ever to assume that there should be limits on what is possible. Even if, as Zadie Smith says, lyrical realism has the run of the highway now, there are still a few slip roads down which others might go.

Stewart Home: Cut and paste

Stewart Home is a kind of subcultural chameleon, capable of playing many roles: artist, pamphleteer, film-maker, activist, hoaxer and writer. Born in London in 1962 , he has spent his adult life immersed in leftist counterculture. "One day in the spring of 1982 I woke up and decided I would be an artist," he said. He formed a punk band, created a "one- man movement" known as "Generation Positive" and founded the parodic art fanzine SMILE , before becoming involved with the underground "anti-art" Neoism movement. Home also loves playing pranks: after claiming in a magazine that he had witnessed an arms dump involving musician Jimmy Cauty, the unfortunate guitarist was arrested while the police raided his house.

But Home is best known as a novelist. Radical in form and content, his books brashly flout conventional ideas of the "literary", mixing filth and the highfalutin and plundering a diverse range of sources. Home's deployment of collage techniques makes him an experimentalist in the tradition of William Burroughs, while his preoccupation with moral subversion and explicit sex draws parallels with Henry Miller and Jean Genet. Slow Death (1996) follows the progress of a sexually voracious skinhead who attempts to take on the art establishment, while 69 Things to Do with a Dead Princess (2002) charts a bizarre erotic relationship through a cut-and-paste collage of pornography, political theory and occult conspiracy. His most recent novel, Blood Rites of the Bourgeoisie , merges penis enlargement junk emails with philosophical pontification to strangely comic effect. Anna Winter

Cesar Aira: Forward motion

One of Argentina's leading contemporary writers, César Aira specialises in short, roughly 100-page novels that he churns out at a rate of up to four a year. His productivity is partly explained by his credo of el continuo , or forward motion, which involves making up his plots as he goes along and never revising his work. His fiction is at the more playful end of the experimental spectrum, leading one critic to brand him "the Duchamp of Latin American literature". In one of his books, a character drowns in a vat of ice cream; in another, a mad scientist dreams up a plot to clone Mexican novelist Carlos Fuentes. His most recent work to be translated, Ghosts , concerns the apparitions who inhabit an almost completed luxury apartment block in Buenos Aires. Like Roberto Bolaño, Aira sees himself as an adversary of the likes of Márquez and Fuentes. Interest in him, particularly in the US, is growing and it seems likely we'll be hearing lots more about him. William Skidelsky

Ben Brooks: Emotional montage

It was picking up a £1 secondhand novel that set British schoolboy Ben Brooks writing. The book was Noah Cicero's The Human War , a savage tirade set two hours before the start of the Iraq conflict. Up until then, Brooks explains, he was reading, "just classics mainly, like Dostoevsky and Tolstoy and stuff". That and "pretty awful set texts" – because, despite having three published novellas to his name and another on the way, 18-year-old Brooks only finished school a month ago.

At 16, he began sending his writing to James Chapman, who runs Fugue State, a small New York press devoted to experimental novels (including The Human War ). Brooks's first, Fences , was published by them last year and its "emotional montage" style sent ripples across the Atlantic. He's also become known for what he calls "the font thing" – using text size as a sort of punctuation, so that some words whisper and others loom large. The unbound quality of these visual crescendos and diminuendos is echoed in the style of his imagery: all torrents of poppies and floods and liquid gold.

Youth has its downsides: "None of the books I've written do I particularly like any more," he says equably. But, he adds, "it's good for picking up girls and stuff." Hermione Hoby

• The Observer
• Tom McCarthy

Comments (…)

Most viewed.

KathySteinemann.com: Free Resources for Writers

Word lists, cheat sheets, and sometimes irreverent reviews of writing rules. kathy steinemann is the author of the writer's lexicon series..

Be Brave: Experimental Writing Ahead: But There Are Rules

Experimental but Not Chaotic

Nolan Liebert provides a few basic rules for experimental fiction, including links to pieces written by authors Ani King and C. J. Harrington.

Nolan’s Advice

My name is Nolan, and I write speculative and experimental fiction and poetry. The majority of my published work falls in the experimental category, in one way or another. I edit Pidgeonholes , a weekly online journal of experimental and international literature, and volunteer as a first reader for freeze frame fiction . So, while I’m not an MFA-holding, rhetoric-wielding expert, I think I’ve got some practical insight into what makes experimental writing successful. So, without further ado, let’s deconstruct this mystery together.

Before you begin with experimental fiction, you need to know one thing: writing something incomprehensible and calling it “experimental” defeats the point of literature. There are rules. Yes. Rules.

The main rule, the most important rule, the only rule you really need to remember is to start with knowing, truly understanding the form or forms you are working with. Some examples are:

• computer code
• false non-fiction
• metafiction
• dialogue -only
• invented dialects
• stream-of-consciousness
• asemic writing

Of course, there are many more, and many authors will invent their own.

By Choosing a Form, You Will Draw Attention to the Form

This is because most readers aren’t accustomed to reading fictional lists, quizzes that tell stories, or unfiltered thoughts. The basic story elements of character, plot, and theme are all filtered through the form you choose, as if the form were a camera and the story is seen through the viewfinder. If the camera shifts, so does the subject. Any form chosen is a base, and it is up to you, as an author, to find a new way to approach it.

Take, for example, “ Your Elegant Noose ” by Ani King, published in freeze frame fiction . This piece is clearly a numbered list. In it, the described character progresses through different medications and mental states until ultimately committing suicide. The premise of the story is simple enough that it could be told in a straightforward, traditional narrative. However, this would leave us without questions and without the impact of the ending.

Over the course of the piece, King evolves her method slightly, only slightly, as the piece progresses, which is what leaves us with the satisfying punch as readers. The other thing that really works is that she has treated the entire form as an evolution of the list. It’s a list, but it’s not. It’s a progression, a terror-ride down the drain, a lexical meandering to the unlucky #13. The form here, for me at least, is like the noose it indicates — long in the diction, smooth in feel, dangerous in the execution. The piece is the form, but it has taken the base and evolved it to something new, unfamiliar, unsettling, and wholly interesting .

Another Piece of Experimental Fiction

Consider “ What the Ocean Does ” by C. J. Harrington, published in Pidgeonholes . First, note that the narrative is written in second-person, featuring “you” as a character, which is not a common perspective. In this way, it is akin to an epistle, but is not in simple letter format. The separated sections form a sort of list. These are not defined by numbers, but rather events, separated by time, that make up a larger experience that the reader must piece together. These deviations from the epistle and the list are not gradual, they are immediate and pervasive. However, because they subvert reader expectations, they remain interesting for the length of the piece. Top this off with language that borders on poetry, and you have an experimental piece that is vivid, engrossing, and compelling.

Experimental but Comprehensible

These two are what I would consider “soft” examples of experimental writing. They’re experimental but still easily comprehensible. Works like these are great jumping-off-points for those looking to get their feet wet trying new writing styles.

As you experiment, you may run into problems, especially when it comes to submitting work for potential publication. Tabs do not easily translate for Internet publication, tables are often hard to incorporate, and things like concrete poetry that forms an image with words are a formatting nightmare. I have received numerous submissions like this and only have one currently being prepared for publication in a special National Poetry Month volume. The formatting is so strange that I am publishing it as an image on the Pidgeonholes website instead of text. Of course, each publisher will approach these types of problems in different ways — some will specifically state in their guidelines to query about pieces that may have formatting problems, some will figure out a way to make them work as long as they love them, and others will turn them down flat. So before you submit, do some reading and research of your target market.

Ultimately, there are some truly strange and beautiful works available on the Internet, both in subject matter and form. I frequently read the works available on sites like PANK , Gone Lawn , DIAGRAM , ExFic , and decomP . I would encourage anyone hoping to get into experimental fiction to check out the work not only over at Pidgeonholes , but these and other magazines, as well.

© Nolan Liebert

Nolan Liebert edits Pidgeonholes , a weekly webzine of experimental and international writing. He also volunteers as a reader for freeze frame fiction . He writes short fiction and poetry that can be found littering the Internet. Interacting with authors, both new and established, is important to him, so feel free to harass him on X @pidgeonholes . You can read more about him and his work at https://nolanliebert.wordpress.com/ .

See also The Master List of So-called “Rules” for Writers .

Discover more from KathySteinemann.com: Free Resources for Writers

Subscribe to get the latest posts sent to your email.

Type your email…

Please don't be shy. Leave a reply. Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

Notify me of new posts by email.

4 thoughts on “ Be Brave: Experimental Writing Ahead: But There Are Rules ”

Very informative. Kathy would you be interested in reviewing my book, “Emojis Vs. Punctuation Marks:Battle Of The Keyboard”?

Hi, Mansu. Thanks for dropping by. I haven’t been able to do any book reviews lately. If you return for next week’s blog post, you’ll see why.

Keep writing!

Thank you very much for this informative article.

Thanks, Sam!

If you have a chance, please visit Nolan’s Pidgeonholes magazine. You’ll find excellent experimental fiction that pushes the boundaries of creativity.

An Introduction to Experimental Design Research

• First Online: 18 May 2016

Cite this chapter

• Philip Cash 4 ,
• Tino Stanković 5 &
• Mario Štorga 6  

6049 Accesses

4 Citations

Design research brings together influences from the whole gamut of social, psychological, and more technical sciences to create a tradition of empirical study stretching back over 50 years (Horvath 2004 ; Cross 2007 ). A growing part of this empirical tradition is experimental, which has gained in importance as the field has matured. As in other evolving disciplines, e.g. behavioural psychology, this maturation brings with it ever-greater scientific and methodological demands (Reiser 1939 ; Dorst 2008 ). In particular, the experimental paradigm holds distinct and significant challenges for the modern design researcher. Thus, this book brings together leading researchers from across design research in order to provide the reader with a foundation in experimental design research; an appreciation of possible experimental perspectives; and insight into how experiments can be used to build robust and significant scientific knowledge. This chapter sets the stage for these discussions by introducing experimental design research, outlining the various types of experimental approach, and explaining the role of this book in the wider methodological context.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Experimental Design

Literature on Experimental Design and Discussion of Some Topics Not Covered in the Text

Research Design: Toward a Realistic Role for Causal Analysis

Keyword: experiment in abstract , title or keywords from 1990 to 2015.

American Psychological Association (2010) Ethical principles of psychologists and code of conduct. Am Psychol 57:1060–1073

Google Scholar  

Ball LJ, Ormerod TC (2000a) Putting ethnography to work: the case for a cognitive ethnography of design. Int J Hum Comput Stud 53:147–168

Ball LJ, Ormerod TC (2000b) Applying ethnography in the analysis and support of expertise in engineering design. Des Stud 21:403–421

Berdichevsky D, Neuenschwander E (1999) Toward an ethics of persuasive technology. Commun ACM 42:51–58. doi: 10.1145/301353.301410

Article   Google Scholar  

Blessing LTM, Chakrabarti A (2009) DRM, a Design Research Methodology. Springer, New York

Book   Google Scholar  

Brandt E, Binder T (2007) Experimental design research: genealogy, intervention, argument. In: IASDR international association of societies of design research, pp 1–18

Carroll JM, Swatman PA (2000) Structured-case: a methodological framework for building theory in information systems research. Eur J Inf Syst 9:235–242

Cash P, Culley S (2014) The role of experimental studies in design research. In: Rodgers P, Yee J (eds) The Routledge companion to design research. Routledge, New York, pp 175–189

Cash P, Elias EWA, Dekoninck E, Culley SJ (2012) Methodological insights from a rigorous small scale design experiment. Des Stud 33:208–235

Cash P, Piirainen KA (2015) Building a cohesive body of design knowledge: developments from a design science research perspective. In ICED 15 international conference on engineering design. Milan, Italy (in press)

Cross N (2007) Forty years of design research. Des Stud 28:1–4

Dong A, Lovallo D, Mounarath R (2015) The effect of abductive reasoning on concept selection decisions. Des Stud 37:37–58. doi: 10.1016/j.destud.2014.12.004

Dorst K (2008) Design research: a revolution-waiting-to-happen. Des Stud 29:4–11

Dyba T, Dingsoyr T (2008) Empirical studies of agile software development: a systematic review. Inf Softw Technol 50:833–859

Ernst Eder W (2011) Engineering design science and theory of technical systems: legacy of Vladimir Hubka. J Eng Des 25:361–385

Eisenhardt KM (1989) Building theories from case study research. Acad Manag Rev 14:532–550

Eisenhardt KM, Graebner ME (2007) Theory building from cases: opportunities and challenges. Acad Manag J 50:25–32

Glasgow RE, Emmons KM (2007) How can we increase translation of research into practice? Types of evidence needed. Annu Rev Public Health 28:413–433

Gorard S, Cook TD (2007) Where does good evidence come from? Int J Res Method Educ 30:307–323

Guala F (2005) The methodology of experimental economics. Cambridge University Press, Cambridge

Horvath I (2004) A treatise on order in engineering design research. Res Eng Design 15:155–181

Hubka V (1984) Theory of technical systems: fundamentals of scientific Konstruktionslehre. Springer, Berlin

Kirk RE (2009) Experimental design. Sage Publications, London, UK

Kitchenham BA, Pfleeger SL, Pickard LM, Jones PW, Hoaglin DC, El-Emam K, Rosenberg J (2002) Preliminary guidelines for empirical research in software engineering. IEEE Trans Softw Eng 28:721–734

Levin JR, O’Donnell AM (1999) What to do about educational research’s credibility gaps? Issues Educ 5:177–229

Lilley D, Wilson GT (2013) Integrating ethics into design for sustainable behaviour. J Des Res 11:278–299

National Academy of Sciences (2009) On being a scientist: a guide to responsible conduct in research

Nesselroade JR, Cattell RB (2013) Handbook of multivariate experimental psychology, vol 11. Springer Science & Business Media

Radder H (2003) The philosophy of scientific experimentation. University of Pittsburgh Press, Pittsburgh

Reiser OL (1939) Aristotelian, Galilean and non-Aristotelian modes of thinking. Psychol Rev 46:151–162

Robson C (2002) Real world research, vol 2nd. Wiley, Chichester

Saunders MNK, Lewis P, Thornhill A (2009) Research methods for business students, vol 3rd. Pearson, Essex

ScienceDirect (2015) Science Direct: paper repository (Online). www.sciencedirect.com

Shadish WR, Cook TD, Campbell DT (2002) Experimental and quasi-experimental designs for generalized causal inference. Mifflin and Company, Boston

Simon HA (1978) The science of the artificial. Harvard University Press

Snow CC, Thomas JB (2007) Field research methods in strategic management: contributions to theory building and testing. J Manage Stud 31:457–480

Tiles M, Oberdiek H (1995) Living in a technological culture: human tools and human values. Routledge

Wacker JG (1998) A definition of theory: research guidelines for different theory-building research methods in operations management. J Oper Manage 16:361–385

Weisberg RW (2006) Creativity: understanding innovation in problem solving, science, invention, and the arts. John Wiley & Sons

Winston AS, Blais DJ (1996) What counts as an experiment? A transdisciplinary analysis of textbooks, 1930–1970. Am J Psychol 109:599–616

Download references

Author information

Authors and affiliations.

Department of Management Engineering, Technical University of Denmark, Diplomvej, 2800, Lyngby, Denmark

Philip Cash

Engineering Design and Computing Laboratory, Department of Mechanical and Process Engineering, Swiss Federal Institute of Technology Zurich, Zurich, Switzerland

Tino Stanković

Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, Zagreb, Croatia

Mario Štorga

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Philip Cash .

Editor information

Editors and affiliations.

Technical University of Denmark, Lyngby, Denmark

Swiss Federal Institute of Tech. Zürich, Zurich, Switzerland

University of Zagreb, Zagreb, Croatia

Rights and permissions

Reprints and permissions

Copyright information

© 2016 Springer International Publishing Switzerland

About this chapter

Cash, P., Stanković, T., Štorga, M. (2016). An Introduction to Experimental Design Research. In: Cash, P., Stanković, T., Štorga, M. (eds) Experimental Design Research. Springer, Cham. https://doi.org/10.1007/978-3-319-33781-4_1

Download citation

DOI : https://doi.org/10.1007/978-3-319-33781-4_1

Published : 18 May 2016

Publisher Name : Springer, Cham

Print ISBN : 978-3-319-33779-1

Online ISBN : 978-3-319-33781-4

eBook Packages : Engineering Engineering (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Experimental Design: Novel Methodologies and Experimental Designs

Recent development of novel methodologies and experimental designs hold much promise for accelerating the pace of testing the efficacy of digital health interventions.

To promote, educate, and support the use of these innovative scientific tools and approaches, the CTBH has compiled a list of resources (e.g., websites with instructional materials and published manuscripts that review or outline novel experimental and analytic approaches) that can be accessed through the links below. In addition, a number of educational and instructional resources for open source platforms for app development and usability testing are listed.

Methodology Resources

Just-in-time Adaptive Interventions

Just-in-time Adaptive Interventions, by Design lab at the Institute for Social Research at the University of Michigan ( https://d3c.isr.umich.edu/intervention-designs/just-in-time-adaptive-interventions/ )

• Nahum-Shani, I., Hekler, E. B., & Spruijt-Metz, D. (2015). Building health behavior models to guide the development of  just-in-time adaptive interventions: A pragmatic framework. Health Psychology, 34(S), 1209.

Micro Randomized Trials

• Klasnja, P., Hekler, E. B., Shiffman, S., Boruvka, A., Almirall, D., Tewari, A., & Murphy, S. A. (2015). Microrandomized trials: An experimental design for developing just-in-time adaptive interventions. Health Psychology, 34(S), 1220.

Sequential Multiple Assignment Randomized Trials (SMART) Advancing Intervention Data Science, by Design lab at the Institute for Social Research at the University of Michigan ( http://d3lab-isr.com/resources/ ) – includes information on SMART designs (a design that can be used to test adaptive interventions) with resources, training modules and tech reports on adaptive interventions and SMARTs (including R and SAS code for analyzing data from SMARTs)

• Collins, L. M., Murphy, S. A., & Bierman, K. L. (2004). A conceptual framework for adaptive preventive interventions. Prevention science, 5(3), 185-196.
• Nahum-Shani, I., Almirall, Yap, J.R., D. McKay, J., Lynch, K., Freiheit, E., & Dziak, J.J. (in press). SMART Longitudinal Analysis: A Tutorial for Using Repeated Outcome Measures from SMART Studies to Compare Adaptive Interventions. Psychological Methods.
• Nahum-Shani, I., Qian, M., Almirall, D., Pelham, W., Gnagy, B., Fabiano, G., … Murphy, S. A. (2012). Experimental design and primary data analysis methods for comparing adaptive interventions. Psychological Methods, 17, 457-77.

Multiphase Optimization Strategy (MOST) – Framework for intervention development

Optimizing Behavioral and Biobehavioral Interventions, by Intervention Optimization Initiative at NYU School of Global Public Health ( https://publichealth.nyu.edu/w/ioi/optimizing-behavioral-biobehavioral-interventions/overview)

• Collins, L. M. (2018) Optimization of Behavioral, Biobehavioral, and Biomedical Interventions: The Multiphase Optimization Strategy (MOST). Springer.

Factorial Experiments for optimizing multi-component interventions – Research method related to MOST

• Collins, L. M., Dziak, J. J., Kugler, K. C., & Trail, J. B. (2014). Factorial experiments: Efficient tools for evaluation of intervention components. American Journal of Preventive Medicine, 47, 498-504.

Digital Intervention Development and Human Coaching Northwestern’s Center for Behavioral Intervention Technologies – includes a focus on the role of human coaching in supporting digital interventions – describes the Center’s research intervention development and coaching models and provides references: http://cbits.northwestern.edu/research-models/

• Schueller SM, Tomasino KN, Mohr DC. Integrating Human Support into Behavioral Intervention Technologies: The Efficiency Model of Support. Clinical Psychology: Science and Practice. 2016(24):27-45.

Center for Digital Health Interventions in St. Gallen Switzerland Mobile Coach website – provides an open source behavioral intervention platform: https://www.mobile-coach.eu/

Single-Case Experimental Design and Analysis The International Collaborative Network (ICN) for N-of-1 Clinical Trials and Single-Case Experimental Designs (SCEDs) provides resources on design and analysis, including links to online courses, effect size and analysis calculators, and other resources. https://nof1andsced.wixsite.com/home

• Dallery, J., Cassidy, R., Raiff, B. R. (2013). Single-case experimental designs to evaluate novel technology-based health interventions. Journal of Medical Internet Research. 15:e22. https://www.jmir.org/2013/2/e22/

Other examples of available open source platforms and instructional methods for mHealth app development and testing which may help accelerate mHealth research:

http://www.openmhealth.org/features/features-overview/

https://www.usability.gov/what-and-why/user-research.html

https://developer.apple.com/healthkit/

https://smarthealthit.org/an-app-platform-for-healthcare/about/

• Search Menu
• Sign in through your institution
• Advance articles
• Editor's Choice
• Special Collections
• Author Guidelines
• Submission Site
• Open Access
• Reasons to submit
• About BioScience
• Journals Career Network
• Editorial Board
• Advertising and Corporate Services
• Self-Archiving Policy
• Potentially Offensive Content
• Terms and Conditions
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Development of in-class paper-and-pencil activities focused on experimental design, implementation of the activities, assessment of experimental design ability, identification of students’ accurate and inaccurate conceptions regarding experimental design, comparing introductory students with advanced students to determine whether inaccurate conceptions persist throughout the curriculum, statistical analysis, conclusions, references cited.

• < Previous

How Students Think about Experimental Design: Novel Conceptions Revealed by in-Class Activities

• Article contents
• Figures & tables
• Supplementary Data

Sara E. Brownell, Mary Pat Wenderoth, Roddy Theobald, Nnadozie Okoroafor, Mikhail Koval, Scott Freeman, Cristina L. Walcher-Chevillet, Alison J. Crowe, How Students Think about Experimental Design: Novel Conceptions Revealed by in-Class Activities, BioScience , Volume 64, Issue 2, February 2014, Pages 125–137, https://doi.org/10.1093/biosci/bit016

• Permissions Icon Permissions

Experimental design is a fundamental skill for scientists, but it is often not explicitly taught in large introductory biology classes. We have designed two pencil-and-paper in-class activities to increase student understanding of experimental design: an analyze activity, in which students are asked to evaluate data, and a design activity, in which students are asked to propose a novel experiment. We found that students who completed the design activity but not the analyze activity performed significantly better on the Expanded Experimental Design Ability Tool (E-EDAT) than did students who attended a didactic lecture about experimental design. By using grounded theory on student responses on the in-class activities, we have identified a novel set of accurate and inaccurate conceptions focused on two aspects of experimental design: sample size and the repetition of experiments. These findings can be used to help guide science majors through mastering the fundamental skill of designing rigorous experiments.

Experimental design is a fundamental skill, essential for achieving success in science (Coil et al. 2010 ) and gaining fluency in scientific literacy and critical thinking in general (Brewer and Smith 2011 ). However, explicit instruction and practice in experimental design is often lacking in introductory biology lecture courses because of perceived time pressures, large class sizes, and the need to emphasize content rather than skills (Dirks and Cunningham 2006 ).

Efforts have been made to develop activities to test and improve students’ skills and knowledge of experimental design (Hoefnagels 2002 , Hiebert 2007 , White et al. 2009 , Pollack 2010 , Sirum and Humburg 2011 , D’Costa and Schlueter 2013 , Kloser et al. 2013 ). In addition, students’ understanding of the significance of controls has been addressed in many studies (Shadmi 1981 , Chen and Klahr 1999 , Lin and Lehman 1999 , Boudreaux et al. 2008 , Shi et al. 2011 ). In biology, the development of a validated assessment to measure student understanding of experimental design (Sirum and Humburg 2011 ) illustrated how challenging it is for introductory-level college students to design well-controlled experiments. However, despite this body of research, there are few studies in which students’ accurate and inaccurate conceptions about experimental design have been identified (Kanari and Millar 2004 , White et al. 2009 , Colon-Berlingeri and Burrows 2011 ). Therefore, there is a need for additional studies to investigate how best to teach this topic to introductory biology students and what aspects of experimental design are particularly difficult for students to grasp.

In this Education article, we describe two pencil-and-paper in-class group activities designed to test alternative hypotheses about how best to teach experimental design in a large introductory biology classroom. We describe the relative effectiveness of these activities in the improvement of students’ experimental design ability and discuss specific accurate and inaccurate conceptions that we identified from student responses to the in-class activities. To determine whether inaccurate conceptions persist as students progress through the undergraduate biology curriculum, we have also assessed students enrolled in upper-level biology courses on their understanding of two key elements of experimental design: sample size and repeating an experiment.

Numerous studies have demonstrated that active learning approaches more effectively increase student learning than do traditional lectures (Hake 1998 , Beichner et al. 2007 , Freeman et al. 2007). However, there have been few studies on the relative effectiveness of different types of active learning approaches for helping students learn specific concepts (Eddy et al. 2013 ). Our first goal was to develop in-class activities to test alternative hypotheses for how best to teach experimental design in a large introductory biology lecture hall. On the basis of previous studies (Boudreaux et al. 2008 , Crowe et al. 2008 ), we reasoned that student understanding of experimental design could be improved by working in a group (a) to develop a hypothesis and design an experiment to test that hypothesis (the design activity; see supplemental appendices A1 and A2) or (b) to analyze and draw appropriate conclusions from mock experimental data (the analyze activity; see supplemental appendices B1 and B2). Both tasks require higher-order thinking and are considered high level on Bloom's taxonomy of cognitive domains (Bloom and Krathwohl 1956 ); however, the first activity requires synthesis-level skills, whereas the second relies on analysis and evaluation skills.

The in-class activities were based on a nontechnical scenario approachable for students taking their first biology course. Both activities prompted the students to consider the basic elements of a well-designed experiment and were structured as a series of guided prompts (Lin and Lehman 1999 ). We designed both activities with the goal of improving students’ ability to explain the importance of the elements included in an experimental design and to recognize the iterative nature of science and the tentative nature of results (Giere 2004 ). The activities were pilot tested twice with students in a large introductory biology course and revised before the final versions were administered. The data reported in this article are from the student responses to the final version of the activities.

Students enrolled in Biology 180, the first course of a three-quarter introductory biology series at a large public research university, completed the activities on the second day of class in autumn 2011. This introductory biology course is a required gateway course for all students interested in majoring in biological sciences and is focused on the topics of ecology and evolution. Typical enrollment is approximately 600–800 students, primarily sophomores. All of the students attend the same lecture period and are asked to sit with students from their lab sections. For the in-class activity, the ­students self-aggregated into groups of two or three on the basis of where they were sitting in the large traditional lecture hall, and the student groups were randomly assigned one of the two activities to complete in the lecture hall. While they were working on the activity, the students could request help from other groups, the instructor, or teaching assistants (TAs). One instructor and 16 TAs (approximately 40 students per TA) circulated around the room to answer questions while the students were working on the activities. Because these activities were administered on the second day of class, the TAs had only minimal experience facilitating group discussions, so they primarily responded to student-generated questions. Participation points were given for completing the activity, independent of performance. Approval for this study was obtained from the University of Washington Institutional Review Board (application no. 36743).

To measure the impact of the in-class activity on students’ ability to design an experiment, we adapted the Experimental Design Ability Tool (EDAT; Sirum and Humburg 2011 ) to create the Expanded EDAT (E-EDAT). The EDAT is an open-ended response instrument administered as a pre- and posttest in which students are asked to design an investigative strategy to address a company's claim regarding one of its products (Sirum and Humburg 2011 ). The instrument is content independent; it works particularly well for introductory biology students and nonmajors, because it does not require any technical expertise. However, in our initial administration of the EDAT, we found that several of the grading criteria did not discriminate among our students. To enhance the discrimination ability of the test, we created the E-EDAT by adding prompts to the EDAT that direct students to (a) design an experiment that would test a company's claim, (b) provide justification for each element of their research design, and (c) state whether the conclusions drawn from their proposed study could prove the company's claim (supplemental appendix C1).

To score the students’ responses on the E-EDAT, we developed an expanded scoring rubric (supplemental appendix C2) that awarded the students points for recognizing that they needed to include an experimental design element (e.g., a large sample size) and for giving an appropriate explanation of why that element was needed (e.g., to account for natural variability in the population). We were particularly interested in their reasoning, because it had previously been shown that students often understand what components are important for experimental design but do not necessarily know why they should use them (Boudreaux et al. 2008 ). We did not introduce any novel elements to the E-EDAT, but through an iterative process based on student responses on the EDAT, we altered the scoring to reflect both an inclusion of essential elements and appropriate reasoning for including those elements. Whether a response warranted partial credit or full credit for a criterion on the E-EDAT was determined through an iterative process with four different raters, who scored the E-EDAT responses independently and then held norming sessions to come to agreement on what score to give a particular response. Content validity of the final rubric was affirmed by asking three experts in biology to confirm that the scoring rubric was scientifically accurate and relevant to the understanding of experimental design. Unlike the binary system employed in the original EDAT rubric, this expanded scoring system allowed us to identify students with an intermediate understanding of a concept. The students could receive 17 points using the E-EDAT scoring rubric, as opposed to the 10 points possible on the original EDAT (Sirum and Humburg 2011 ).

The students completed the E-EDAT online, outside of class; the pretest was completed the evening before the in-class activity, and the posttest was completed the evening after the in-class activity. Only those students who participated in the in-class activity, completed the pre- and posttests, and consented to have their data analyzed were included in the study. From a consenting population of 357 students who completed the design activity and 276 students who completed the analyze activity, we selected a random subset of pre- and posttests to score with the expanded rubric and then included only the responses from the consenting students in the final analysis ( n = 87 in the design activity group; n = 95 in the analyze activity group). All of our future references to the effectiveness of the in-class activities are based on the data from this randomly selected subset of the students’ pre- and posttests. The E-EDATs were scored blindly by two independent graders. To assess the level of agreement between the graders, we calculated the interrater reliability using Cohen's kappa coefficient for a subset of blinded responses (90 responses each on the pre- and posttests) graded independently by each grader. The interrater reliability on individual questions ranged from .54 to .89, with a reliability of .76 across all of the questions. This indicates moderately strong agreement between the two graders and illustrates that the expanded scoring rubric provides sufficient guidance to achieve reliability between independent graders.

We compared the experimental design ability of students who completed the in-class activities with that of students who learned experimental design through a traditional lecture by measuring learning gains on the E-EDAT. The students in all three comparison groups (i.e., design activity, analyze activity, lecture) were enrolled in Biology 180, but the students who participated in the didactic lecture took the course in a different term and with a different instructor. The students in the traditional lecture course were assigned the same readings and received instruction of equivalent length to the in-class activities on the scientific method and experimental design. The lecture included examples of experimental data that illustrated inherent variation in a population and variable outcomes from repeating an experiment with a different population. The lecture also included explicit statements that a hypothesis can never be definitively proven (one of the elements scored on the assessment). One hundred student pre- and posttest responses were randomly selected from the lecture group and scored; we compared the lecture students’ learning gains with the gains attained by the students who completed the activities.

Many terms have been used to describe student conceptions, including naive conceptions (Strike and Posner 1992 ), alternative conceptions (Mak et al. 1999 , Poehnl and Bogner 2013 ), preconceptions (Clement et al. 1989 , Ryan and Aikenhead 1992 ), misconceptions (Coley and Tanner 2012 , Yates and Marek 2013 ), and inaccurate conceptions (Zuckerman 1994 , Edens and Potter 2003 ). In this article, we will use a model of describing student conceptions as either accurate or inaccurate, defining accurate as being in accordance with what is known to be scientifically true and confirmed by a group of expert scientists. The students’ 
handwritten responses to questions posed on the in-class activity worksheets were transcribed and coded as accurate, inaccurate, a mixed model that was a combination of accurate and inaccurate, or too vague to determine accuracy. A response was classified as vague if it was incomplete, did not answer the question, or was so general that we could not determine whether the student held an inaccurate or accurate conception. We took a conservative approach in our analysis by removing vague answers from the data set so that we did not incorrectly infer what the students were thinking (Gormally et al. 2012 ).

We chose to focus on two aspects of experimental design that we determined were challenging for the students on the basis of their low E-EDAT scores: sample size and repeating an experiment (supplemental table S1). The majority of the students’ answers on the E-EDAT did not include any mention of sample size or repetition, which may be because of the open-ended nature of the E-EDAT, which does not contain specific prompts for students to address each of these aspects of experimental design. Alternatively, it could be because the students did not think that sample size and repeating an experiment are important elements of experimental design. The commonality between these two elements is that both sample size and repeating an experiment are relevant to one's confidence in a conclusion based on a given set of data and require an understanding of the inherent variation that exists in biological populations; they both help students understand the iterative, tentative nature of scientific results.

In order to assess the quality of the students’ conceptions, we analyzed their responses on the in-class worksheets, because the students were explicitly asked to consider sample size and repeating an experiment as they completed those worksheets (table 1 ). Grounded theory was used to identify specific conceptions—both accurate and inaccurate conceptions—that the students held about sample size and repeating an experiment from the in-class activities. Grounded theory is a process by which researchers do not hold previous ideas or hypotheses about the data; rather, the themes emerge from the data itself (Glaser and Strauss 1967 ). We decided to use this method as a way to examine the students’ responses without prior bias in order to uncover potentially novel conceptions. Two raters then scored the students’ written answers on the in-class activities for the presence of these conceptions. The raters’ agreement averaged 70%, and disagreement in coding was discussed to achieve consensus. To achieve expert validation, we asked a group of five expert biologists ( expert was defined as holding a PhD in a biology-related field) in our research group to review the list of identified accurate and inaccurate conceptions (see the tables), and they agreed with the raters’ designations. In addition, we asked a group of three outside expert biologists to confirm the designations.

Prompts on each of the activities for sample size and repeating an experiment.

We also surveyed undergraduate biology majors enrolled in 400-level (senior-level) courses ( n = 122) to assess their understanding of the importance of sample size and repeating an experiment and to investigate whether they maintained the inaccurate conceptions held by the introductory students. Using an online survey, the advanced students were asked a subset of the questions from the analyze activity, because the question prompts were more direct and elicited fewer vague conceptions than did the question prompts of the design activity. These students received participation points for completing the questions regardless of the accuracy of their responses. The same two independent raters who coded the introductory student responses coded the advanced student responses, using the same set of categories described for the introductory students. Rater agreement was established to be over 70%, and disagreements in coding were discussed to achieve consensus.

As a preliminary analysis, Student's t -tests were used to compare the students’ gain scores on the E-EDAT among the three groups: the students who completed the design activity, the students who completed the analyze activity, and the students who were in the lecture course. The gains were calculated as the posttest score minus the pretest score. However, there are some differences in the characteristics of the students in the three groups (see table 2 ) that may be correlated with the test score gains. To control for these differences, we used a multiple linear regression model in which each student's test score gain was the response variable, and observable student characteristics, including gender, ethnicity, socioeconomic status, grade point average, and verbal SAT score, serve as predictor variables. This regression model produces estimates of two treatment effects: the treatment effect of the design activity relative to the lecture and the treatment effect of the analyze activity relative to the lecture. In each case, the treatment effect is the average difference in test score gains between the two groups, holding observable characteristics of students in the two groups constant.

Demographic information for the subset of students who completed the activities or experienced a didactic lecture and whose scores were analyzed.

a Low socioeconomic status was measured by admission into the Equal Opportunity Program.

* p < .05.

In order to determine significant differences in the student conceptions derived from grounded theory, a chi-squared analysis was used to compare the inaccurate, mixed model, and accurate conceptions, and t -tests were used for comparing the specific student conceptions ( α = .05).

Finding 1: An active approach leads to greater understanding of experimental design than passive lecturing does.

The introductory biology students showed gains in their ability to design an experiment after the in-class pencil-and-paper activities, independent of the type of activity, and the students who completed the design activity had significantly higher gains than did the students who attended a didactic lecture.

We found that the students who completed the design and the students who completed the analyze activity had similar learning gains when their pre- and posttests were compared ( p = .21). However, only the students who completed the design activity demonstrated significantly higher gains on the E-EDAT than the group of students who learned about experimental design through the didactic lecture ( t- test, p < .05; figure 1 ). The average score on the posttest for all of the groups was 7.6 out of 17, with no individual group scoring higher than 8.1, which indicated no ceiling effect (table S1). The learning gains could be seen in multiple criteria (table S1) and could not be attributed to one particular aspect of the E-EDAT.

Students who completed the design activity scored higher on the Expanded Experimental Design Ability Tool (E-EDAT) than did students who experienced only a lecture on experimental design (design activity, n = 87; analyze activity, n = 95; lecture, n = 100; design–analyze comparison, p = .210; design–lecture comparison, p = .018; analyze–lecture comparison, p = .275). The gain was calculated as the posttest score minus the pretest score. The error bars represent the standard error.

The results of the linear regression model used to control for differences in the observable characteristics of the students in the three groups (table 2 ) indicate that, holding individual student characteristics constant, the students who completed the design experiment gained almost half a point more out of 17 possible points, on average, than did the students in the lecture class, so the differences in the E-EDAT scores are likely due to the differences in activities, not to differences in the student population ( p = .022; table 3 ). The observation that the students in the lecture group did not show gains from the pretest to the posttest indicates that the process of taking the E-EDAT itself did not lead to learning.

Multiple linear regression of gain scores against individual student characteristics.

Finding 2: Introductory students do not have a strong understanding of the importance of sample size and repeating an experiment.

The analysis of the introductory students’ written responses on the in-class activity worksheets revealed that the students harbored three distinct levels of understanding about sample size and repeating an experiment: accurate, inaccurate, and a combination of accurate and inaccurate conceptions (figure 2a , 2b ). If we combine the student responses that were completely inaccurate with those that contained mixed conceptions, we see that the majority of the student responses on the design activity and the analyze activity contained some inaccurate conceptions about both sample size ( design , 75.2%; analyze , 69.8%) and repeating an experiment ( design , 63.4%; analyze , 81.1%). The students’ responses revealed more inaccurate conceptions surrounding the purpose of repeating an experiment than regarding the importance of sample size for both the design (45.5%) and the analyze (52.2%) activities ( t -test, p < .01). Interestingly, the design activity responses contained significantly fewer inaccurate conceptions about sample size than did the analyze activity responses (10.2% and 39.6%, respectively; t -test, p < .01).

Introductory students’ conceptions of (a) sample size and (b) repeating an experiment. For responses regarding the rationale for sample size, the data are shown as percentages of the total number of student group responses ( n = 69 for the design group, n = 96 for the analyze group). The student group responses for the design activity were completely inaccurate (10.2%), completely accurate (24.6%,) or a mix of accurate and inaccurate (65.2%) conceptions; the differences between these groups are statistically significant (chi-squared analysis, p < .001). The student group responses for the analyze activity were completely inaccurate (39.6%), completely accurate (30.2%), or a mix of accurate and inaccurate (30.2%). Chi-squared analysis indicates that these differences are not statistically significant. The student responses that were too vague to code have been removed. There was a statistically significant difference between the design and analyze student responses for inaccurate and mixed conceptions (Student's t-test, p < .01). For responses regarding the rationale for repeating an experiment, the data are shown as percentages of the total number of student responses ( n = 112 for the design group and n = 90 for the analyze group). The student responses for the design activity were completely inaccurate (45.5%), completely accurate (36.6%), or a mix of accurate and inaccurate (17.9%). The student responses for the analyze activity were completely inaccurate (52.2%), completely accurate (18.9%), or a mix of accurate and inaccurate (28.9%). Chi-squared analysis for both the design and the analyze groups indicates that these are statistically significant differences ( p < .001).

Notably, there were significantly more vague answers to the question about sample size in the design activity (47.3%) than in the analyze activity (12.7%) (data not shown). However, when students were prompted to provide reasoning for repeating an experiment, their responses contained similar percentages of vague responses in both activities ( design , 11.8%; analyze , 17.4%). Vague responses may reflect confused thinking, a misinterpretation of the question, or a low level of motivation to answer the question.

Finding 3: Novel accurate conceptions and inaccurate conceptions were identified from introductory student responses for sample size and repeating an experiment on the in-class activities.

In order to further explore what conceptions the students held about sample size and repeating an experiment, we used grounded theory to identify three distinct accurate conceptions (table 4a ) and three distinct inaccurate conceptions (table 4b ) about sample size. We also identified three distinct accurate conceptions (table 5a ) and seven distinct inaccurate conceptions (table 5b ) about repeating an experiment. Several students who completed the design activity (8.9%) stated that it was not necessary to repeat an experiment, particularly if the sample size was large enough. Because the analyze activity did not allow the students this option, we cannot conclude whether this idea is a general inaccurate conception held by introductory students or whether the nature of the design exercise led the students to this conception.

Introductory students’ accurate conceptions about sample size.

Note: The data are shown as percentages of the number of student responses on each in-class activity ( n = 69 for the design group, n = 96 for the analyze group). Student responses that were too vague to code have been removed.

* p < .05 (Student's t -test).

Introductory students’ inaccurate conceptions about sample size.

Introductory students’ accurate conceptions about repeating an experiment.

Note: The data are shown as percentages of the number of student responses on each in-class activity ( n = 112 for the design group, n = 90 for the analyze group). Student responses that were too vague to code have been removed.

Introductory students’ inaccurate conceptions about repeating an experiment.

Finding 4: Some inaccurate conceptions are “sticky.”

The advanced students held fewer inaccurate and more accurate conceptions than did the introductory students who completed the analyze activity, but over a third of the advanced students continued to harbor inaccurate conceptions.

We surveyed the advanced biology majors’ understanding of experimental design by asking them the same questions about sample size and repeating an experiment that were included in the analyze activity. We found that the advanced students held significantly more accurate conceptions (advanced, 57.3%; introductory, 30.2%) and fewer inaccurate conceptions about sample size (advanced, 14.5%; introductory 39.6%) than did the introductory students who completed the analyze activity ( p s < .05; figure 3 ). Similarly, in describing their reasoning for repeating an experiment, the advanced students held significantly more accurate conceptions (advanced, 56.1%; introductory, 18.9%) and fewer inaccurate conceptions (advanced, 9.6%; introductory, 52.2%) than did the introductory students ( p s < .05).

Advanced students’ conceptions of (a) sample size and (b) repeating an experiment. For sample size, the data are shown as percentages of the total number of student responses ( n = 110). The advanced student responses were completely inaccurate (14.5%), completely accurate (57.3%), or a mix of accurate and inaccurate (28.2%). Compared with 
the introductory students, who completed the analyze activity, the advanced students had significantly more accurate conceptions and fewer inaccurate conceptions (chi-squared analysis, p < .01). Student responses that were too vague to code have been removed. For repeating an experiment, the data are shown as percentages of the total number of student responses ( n = 114). The advanced student responses were completely inaccurate (9.6%), completely accurate (56.1%), or a mix of accurate and inaccurate (34.2%). The advanced students had significantly more accurate conceptions than inaccurate conceptions (chi-squared analysis, p < .01). The introductory student group is the same as the analyze group (see figure 2 ).

When we combine the student responses that were completely inaccurate with those containing mixed conceptions, we see that the majority of responses from the advanced students contained accurate conceptions about both sample size (57.3%) and repeating an experiment (56.1%). However, we still identified a notable percentage of inaccurate conceptions in the advanced students’ responses (42.7% for sample size, 43.8% for repeating an experiment), which indicate that these conceptions about experimental design persist even among students who have almost completed the undergraduate biology curriculum.

We found that the advanced students held more accurate conceptions (table 6a ) and fewer inaccurate conceptions (table 6b ) about sample size than did their introductory-level counterparts. Specifically, significantly more advanced student responses (77.3%) than introductory student responses (53.1%) exhibited the accurate conception that a larger sample size is good ( p < .05; table 6a ). However, there was no significant difference in the advanced and introductory students’ recognition that too big of a sample size is not cost effective (advanced, 3.6%; introductory, 2.1%). Importantly, we did not observe a statistically significant difference between the advanced and introductory students’ recognition 
that sample size is important because of inherent natural biological diversity (advanced, 20%; introductory, 10.4%), the primary reason that most biologists would give for including a large sample size.

Advanced students’ accurate conceptions about sample size.

Note: The data are shown as percentages of the total number of advanced student responses ( n = 110). Student responses that were too vague to code have been removed. The advanced student responses were compared (using Student's t -tests) with the introductory student group responses on the analyze activity (see table 3a ).

Advanced students’ inaccurate conceptions about sample size.

Note: The data are shown as percentages of the total number of advanced student responses ( n = 110; see table 3a ). Student responses that were too vague to code have been removed. The advanced student responses were compared (using Student's t -tests) with the introductory student group responses on the analyze activity (see table 3a ).

In general, the advanced students held fewer inaccurate conceptions about sample size than did the introductory students (table 6b ); however, the inaccurate idea that a larger sample size gives more accurate data was more common in the advanced group (advanced, 31.8%; introductory, 15.6%).

The advanced students held more accurate conceptions and fewer inaccurate conceptions than did the introductory students (table 7a , 7b ) regarding the importance of repeating an experiment. A significantly higher percentage of the advanced students correctly stated that it is important to repeat an experiment to reduce the likelihood that an uncontrolled variable has affected the results (advanced, 21.9%; introductory, 3.3%) or to reduce the impact of chance or randomness on the interpretation of the results (advanced, 36.8%; introductory, 6.7%; p s < .05). However, the advanced students and the introductory students were equally likely to recognize that repeating an experiment and getting a reproducible result would increase confidence in the data (advanced, 52.6%; introductory, 43.3%; table 7a ). Significantly fewer of the advanced students held the inaccurate conception that you repeat an experiment to eliminate outliers or to account for chance (advanced, 0.9%; introductory, 22.2%), to make the data more accurate (advanced, 14.9%; introductory, 28.9%), or to make certain that the findings are correct (advanced, 13.2%; introductory, 28.9%; p s < .05; table 7b ).

Advanced students’ accurate conceptions about repeating an experiment.

Note: The data are shown as percentages of the total number of advanced student responses ( n = 114; see table 4a ). Student responses that were too vague to code have been removed. The advanced student responses were compared (using Student's t -tests) with the introductory student group responses on the analyze activity (table 4a ).

Advanced students’ inaccurate conceptions about repeating an experiment.

Note: The data are shown as percentages of the number of advanced student responses ( n = 114; see table 4b ). Student responses that were too vague to code have been removed. The advanced student responses were compared (using Student's t -tests) with the introductory student group responses on the analyze activity (table 4a ).

We originally set out to identify which activity would lead to higher learning gains, but the finding that both in-class activities were beneficial for improving students’ experimental design ability may be even more interesting. It suggests that there may not be one “right” way to teach a skill as complex as experimental design. Because the students completed these activities in groups, we cannot disaggregate their responses by demographic characteristics for this study. However, it would be an interesting area for future research to see whether certain students learn better with an analysis or evaluation task than with a synthesis-level task.

Although we found no significant differences in E-EDAT score gains between the two activities, we did observe a difference of 0.5 point between the design activity and the passive lecture. Although this number is small, it is statistically significant, and we believe that it demonstrates a meaningful improvement. We were not expecting a large difference between the pre- and posttest scores for an intervention of only 30 minutes, and, given the time spent on the task, this gain is similar to what has previously been reported (Sirum and Humburg 2011 ).

There are at least two possible explanations for why only the design activity was significantly more effective than the lecture: There may be closer alignment of the design task with the assessment instrument, or the students’ ability to apply concepts to new situations may increase after completing a synthesis-level activity. The development of additional validated tools to assess student understanding of experimental design would allow us to differentiate these possibilities. The students completed either the design activity or the analyze activity in this study, but a possible area for future exploration would be to determine whether there could be a synergistic effect resulting from students completing both activities. In particular, it may be interesting to investigate whether there is an order effect, such that we see higher gains in students who complete the analyze activity before the design activity or vice versa.

We also uncovered several novel accurate and inaccurate student conceptions about experimental design. The structure of the two in-class activities probably affected which accurate and inaccurate conceptions were identified. The increased number of vague responses on the design activity, for example, may be a result of the more open-ended nature of the questions on that activity. Interestingly, the students in the design activity group were much more likely to consider the advantages of a large sample size, as well as the cost and logistical challenges associated with using a large sample size; the design activity was better than the analyze activity at eliciting these accurate conceptions concerning sample size. Furthermore, the design activity responses contained significantly fewer inaccurate conceptions about sample size than did the analyze activity responses. This suggests that the process of interpreting data may reveal more inaccuracies in students’ ways of thinking than does the act of designing an experiment, which indicates that the analyze activity may be more effective at eliciting inaccurate student ideas. Alternatively, designing an experiment may help move students toward more accurate conceptions. Questions that prompted students to consider the purpose of repeating an experiment elicited significantly more inaccurate conceptions than did those focused on sample size. This may be because of the difficulty of the topic, the wording of the prompt, or because students are not often asked to consider why they should repeat an experiment (e.g., in “cookbook” lab courses).

Much of the research focused on gaining insight into student understanding of biological concepts has generally relied heavily on identifying student misconceptions (Nelson 2008 ), which are defined as scientifically inaccurate ideas. Although the identification of misconceptions has been valuable for the biology education community, recent findings indicate that students’ misunderstandings of concepts cannot simply be described as misconceptions; rather, there is a continuum of student understanding known as a learning progression (Alonzo and Gotwals 2012 ). Learning progressions are research-based models of how core ideas are formed over time, often focused on students’ ways of thinking (Songer et al. 2009 , Duncan and Rivet 2013 ). Learners develop their understanding of complex biological concepts in stages that build on each other. It has been shown that scientifically inaccurate answers may be useful for students at early stages of learning, allowing them to partially understand a topic (Duncan and Rivet 2013 ). Although there are a few studies in which inaccurate but possibly productive student conceptions related to experimental design have been identified (Kanari and Millar 2004 , White et al. 2009 , Colon-Berlingeri and Burrows 2011 ), we currently lack a learning progression for undergraduate biology majors’ understanding of experimental design, and we hope that findings from our study can help move the field toward this goal.

Specifically, our investigation of the differences between introductory and advanced students could be useful for developing a learning progression. Overall, the advanced students held significantly more accurate conceptions and fewer inaccurate conceptions for sample size and repeating an experiment than did the introductory students. There are a few surprising observations about the differences between the advanced and introductory students’ ideas concerning the significance of sample size and repeating an experiment. First, the advanced students were more likely to hold the inaccurate conception that repeating an experiment leads to more accurate data. A possible reason for this is that the advanced students were actually thinking correctly about how repeating an experiment could lead to a more accurate interpretation of the results but simply used the phrase “more accurate data” to convey the idea that repeating an experiment will lead to increased confidence in one's interpretation of the data. This difference in language, although it is subtle, is important but may not be clear to these students. Intriguingly, this inaccurate idea of “more accurate data” could be viewed as a productive misunderstanding in a student learning progression of experimental design (Duncan and Rivet 2013 ). As students move from not thinking about how sample size affects the quality of the data to thinking that a larger sample size often leads to a more accurate interpretation of those data, perhaps thinking incorrectly about accurate data is an indication that the students are on the path to building a deeper understanding.

Although the advanced students were more likely to provide accurate justifications for experimental design elements, we think that it may be primarily because of an increased proficiency with statistics and may not be reflective of improved understanding of how biological variation influences experimental design. The vast majority of the introductory and advanced students who mentioned anything related to variation discussed “unique individuals” or “mutants,” as opposed to the variation that exists on a natural continuum (data not shown). Although we do not have an explanation for why these biology students’ did not think about inherent variation in a population, our study suggests that this concept may need to be more explicitly taught in the biology classroom.

Although it does not provide a complete picture, this study is an important first step toward revealing the types of conceptions that students hold about experimental design. The underlying reasons for which students hold these inaccurate conceptions remain to be explored. Is it a result of how we teach experimental design in lab courses—with a very small sample size and typically never repeating experiments? Do students not understand inherent variability in the population because we often present the data as averages, and undergraduate students rarely see raw data? Do students understand the underlying biological principles but have linguistic difficulties describing precise aspects of experimental design (e.g., the difference between “decreasing the effect of outliers” and “decreasing outliers”) that make their responses inaccurate? Future directions for this research include developing research tools that target these specific inaccurate conceptions and using think-alouds and interviews to more deeply probe student understanding.

It is possible that the present results underrepresent the extent to which students hold inaccurate conceptions about experimental design. We relied on the students to come up with an inaccurate conception, as opposed to asking a specific question about the inaccurate conception, so it is possible that some of the students may have had inaccurate conceptions that they did not write down. The students also worked in groups, so they had the opportunity to discuss their responses with each other. They were able to ask the TAs and the instructor questions about the activities during the class session, so, perhaps, some of their inaccurate conceptions were clarified in class. Although our work is an important first exploration into possible conceptions that students may harbor, more work needs to be done to determine how prevalent these conceptions are for undergraduates.

There are a few limitations to this study that necessitate caution in generalizing the results. First, different instructors taught the lecture course and the course in which the activities were administered, so an instructor effect could have affected the difference we saw between the design activity and lecture students on the E-EDAT. Next, we collected data from the introductory students through in-class, handwritten worksheets, whereas we collected data from the advanced students though online questions. The questions were identical, but the method of delivery was different, which could have influenced the results. Finally, although we anticipate that we would obtain similar results with a different population of students, especially because we controlled for student ability in our regression model, collecting only one set of data is a limitation but is not uncommon for educational research. Determining the impact of these activities in different student populations is an interest of ours, and we encourage others to use our activities and the E-EDAT to see whether they obtain similar gains in different contexts.

In this article, we have presented two in-class activities that instructors can use to teach experimental design, a modified EDAT (E-EDAT), and a rubric to assess students’ ability to design an experiment and justify their reasoning. This study also provides novel insight into how students think about specific elements of experimental design, which could be the basis for building a learning progression of undergraduate thinking about experimental design. Much work still needs to be done before we can begin to model what the learning progression may be, but we believe that this study is an important step in our own learning progression of understanding student thinking about experimental design.

This work was supported in part by National Science Foundation grant no. DUE-0942215, awarded to AJC and MPW, and a Washington Research Foundation-Hall fellowship awarded to CLW-C. We would like to thank the University of Washington (UW) students who participated in the study and the UW faculty members who were supportive of these efforts. Special thanks go to John Parks for his invaluable support administering the activities, the UW Biology Education Research Group for helpful discussions, and Sarah Eddy for statistical advice.

Google Scholar

Google Preview

Supplementary data

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1525-3244
• Copyright © 2024 American Institute of Biological Sciences
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

• Privacy Policy

Home » Experimental Design – Types, Methods, Guide

Experimental Design – Types, Methods, Guide

Table of Contents

Experimental Design

Experimental design is a process of planning and conducting scientific experiments to investigate a hypothesis or research question. It involves carefully designing an experiment that can test the hypothesis, and controlling for other variables that may influence the results.

Experimental design typically includes identifying the variables that will be manipulated or measured, defining the sample or population to be studied, selecting an appropriate method of sampling, choosing a method for data collection and analysis, and determining the appropriate statistical tests to use.

Types of Experimental Design

Here are the different types of experimental design:

Completely Randomized Design

In this design, participants are randomly assigned to one of two or more groups, and each group is exposed to a different treatment or condition.

Randomized Block Design

This design involves dividing participants into blocks based on a specific characteristic, such as age or gender, and then randomly assigning participants within each block to one of two or more treatment groups.

Factorial Design

In a factorial design, participants are randomly assigned to one of several groups, each of which receives a different combination of two or more independent variables.

Repeated Measures Design

In this design, each participant is exposed to all of the different treatments or conditions, either in a random order or in a predetermined order.

Crossover Design

This design involves randomly assigning participants to one of two or more treatment groups, with each group receiving one treatment during the first phase of the study and then switching to a different treatment during the second phase.

Split-plot Design

In this design, the researcher manipulates one or more variables at different levels and uses a randomized block design to control for other variables.

Nested Design

This design involves grouping participants within larger units, such as schools or households, and then randomly assigning these units to different treatment groups.

Laboratory Experiment

Laboratory experiments are conducted under controlled conditions, which allows for greater precision and accuracy. However, because laboratory conditions are not always representative of real-world conditions, the results of these experiments may not be generalizable to the population at large.

Field Experiment

Field experiments are conducted in naturalistic settings and allow for more realistic observations. However, because field experiments are not as controlled as laboratory experiments, they may be subject to more sources of error.

Experimental Design Methods

Experimental design methods refer to the techniques and procedures used to design and conduct experiments in scientific research. Here are some common experimental design methods:

Randomization

This involves randomly assigning participants to different groups or treatments to ensure that any observed differences between groups are due to the treatment and not to other factors.

Control Group

The use of a control group is an important experimental design method that involves having a group of participants that do not receive the treatment or intervention being studied. The control group is used as a baseline to compare the effects of the treatment group.

Blinding involves keeping participants, researchers, or both unaware of which treatment group participants are in, in order to reduce the risk of bias in the results.

Counterbalancing

This involves systematically varying the order in which participants receive treatments or interventions in order to control for order effects.

Replication

Replication involves conducting the same experiment with different samples or under different conditions to increase the reliability and validity of the results.

This experimental design method involves manipulating multiple independent variables simultaneously to investigate their combined effects on the dependent variable.

This involves dividing participants into subgroups or blocks based on specific characteristics, such as age or gender, in order to reduce the risk of confounding variables.

Data Collection Method

Experimental design data collection methods are techniques and procedures used to collect data in experimental research. Here are some common experimental design data collection methods:

Direct Observation

This method involves observing and recording the behavior or phenomenon of interest in real time. It may involve the use of structured or unstructured observation, and may be conducted in a laboratory or naturalistic setting.

Self-report Measures

Self-report measures involve asking participants to report their thoughts, feelings, or behaviors using questionnaires, surveys, or interviews. These measures may be administered in person or online.

Behavioral Measures

Behavioral measures involve measuring participants’ behavior directly, such as through reaction time tasks or performance tests. These measures may be administered using specialized equipment or software.

Physiological Measures

Physiological measures involve measuring participants’ physiological responses, such as heart rate, blood pressure, or brain activity, using specialized equipment. These measures may be invasive or non-invasive, and may be administered in a laboratory or clinical setting.

Archival Data

Archival data involves using existing records or data, such as medical records, administrative records, or historical documents, as a source of information. These data may be collected from public or private sources.

Computerized Measures

Computerized measures involve using software or computer programs to collect data on participants’ behavior or responses. These measures may include reaction time tasks, cognitive tests, or other types of computer-based assessments.

Video Recording

Video recording involves recording participants’ behavior or interactions using cameras or other recording equipment. This method can be used to capture detailed information about participants’ behavior or to analyze social interactions.

Data Analysis Method

Experimental design data analysis methods refer to the statistical techniques and procedures used to analyze data collected in experimental research. Here are some common experimental design data analysis methods:

Descriptive Statistics

Descriptive statistics are used to summarize and describe the data collected in the study. This includes measures such as mean, median, mode, range, and standard deviation.

Inferential Statistics

Inferential statistics are used to make inferences or generalizations about a larger population based on the data collected in the study. This includes hypothesis testing and estimation.

Analysis of Variance (ANOVA)

ANOVA is a statistical technique used to compare means across two or more groups in order to determine whether there are significant differences between the groups. There are several types of ANOVA, including one-way ANOVA, two-way ANOVA, and repeated measures ANOVA.

Regression Analysis

Regression analysis is used to model the relationship between two or more variables in order to determine the strength and direction of the relationship. There are several types of regression analysis, including linear regression, logistic regression, and multiple regression.

Factor Analysis

Factor analysis is used to identify underlying factors or dimensions in a set of variables. This can be used to reduce the complexity of the data and identify patterns in the data.

Structural Equation Modeling (SEM)

SEM is a statistical technique used to model complex relationships between variables. It can be used to test complex theories and models of causality.

Cluster Analysis

Cluster analysis is used to group similar cases or observations together based on similarities or differences in their characteristics.

Time Series Analysis

Time series analysis is used to analyze data collected over time in order to identify trends, patterns, or changes in the data.

Multilevel Modeling

Multilevel modeling is used to analyze data that is nested within multiple levels, such as students nested within schools or employees nested within companies.

Applications of Experimental Design 

Experimental design is a versatile research methodology that can be applied in many fields. Here are some applications of experimental design:

• Medical Research: Experimental design is commonly used to test new treatments or medications for various medical conditions. This includes clinical trials to evaluate the safety and effectiveness of new drugs or medical devices.
• Agriculture : Experimental design is used to test new crop varieties, fertilizers, and other agricultural practices. This includes randomized field trials to evaluate the effects of different treatments on crop yield, quality, and pest resistance.
• Environmental science: Experimental design is used to study the effects of environmental factors, such as pollution or climate change, on ecosystems and wildlife. This includes controlled experiments to study the effects of pollutants on plant growth or animal behavior.
• Psychology : Experimental design is used to study human behavior and cognitive processes. This includes experiments to test the effects of different interventions, such as therapy or medication, on mental health outcomes.
• Engineering : Experimental design is used to test new materials, designs, and manufacturing processes in engineering applications. This includes laboratory experiments to test the strength and durability of new materials, or field experiments to test the performance of new technologies.
• Education : Experimental design is used to evaluate the effectiveness of teaching methods, educational interventions, and programs. This includes randomized controlled trials to compare different teaching methods or evaluate the impact of educational programs on student outcomes.
• Marketing : Experimental design is used to test the effectiveness of marketing campaigns, pricing strategies, and product designs. This includes experiments to test the impact of different marketing messages or pricing schemes on consumer behavior.

Examples of Experimental Design 

Here are some examples of experimental design in different fields:

• Example in Medical research : A study that investigates the effectiveness of a new drug treatment for a particular condition. Patients are randomly assigned to either a treatment group or a control group, with the treatment group receiving the new drug and the control group receiving a placebo. The outcomes, such as improvement in symptoms or side effects, are measured and compared between the two groups.
• Example in Education research: A study that examines the impact of a new teaching method on student learning outcomes. Students are randomly assigned to either a group that receives the new teaching method or a group that receives the traditional teaching method. Student achievement is measured before and after the intervention, and the results are compared between the two groups.
• Example in Environmental science: A study that tests the effectiveness of a new method for reducing pollution in a river. Two sections of the river are selected, with one section treated with the new method and the other section left untreated. The water quality is measured before and after the intervention, and the results are compared between the two sections.
• Example in Marketing research: A study that investigates the impact of a new advertising campaign on consumer behavior. Participants are randomly assigned to either a group that is exposed to the new campaign or a group that is not. Their behavior, such as purchasing or product awareness, is measured and compared between the two groups.
• Example in Social psychology: A study that examines the effect of a new social intervention on reducing prejudice towards a marginalized group. Participants are randomly assigned to either a group that receives the intervention or a control group that does not. Their attitudes and behavior towards the marginalized group are measured before and after the intervention, and the results are compared between the two groups.

When to use Experimental Research Design 

Experimental research design should be used when a researcher wants to establish a cause-and-effect relationship between variables. It is particularly useful when studying the impact of an intervention or treatment on a particular outcome.

Here are some situations where experimental research design may be appropriate:

• When studying the effects of a new drug or medical treatment: Experimental research design is commonly used in medical research to test the effectiveness and safety of new drugs or medical treatments. By randomly assigning patients to treatment and control groups, researchers can determine whether the treatment is effective in improving health outcomes.
• When evaluating the effectiveness of an educational intervention: An experimental research design can be used to evaluate the impact of a new teaching method or educational program on student learning outcomes. By randomly assigning students to treatment and control groups, researchers can determine whether the intervention is effective in improving academic performance.
• When testing the effectiveness of a marketing campaign: An experimental research design can be used to test the effectiveness of different marketing messages or strategies. By randomly assigning participants to treatment and control groups, researchers can determine whether the marketing campaign is effective in changing consumer behavior.
• When studying the effects of an environmental intervention: Experimental research design can be used to study the impact of environmental interventions, such as pollution reduction programs or conservation efforts. By randomly assigning locations or areas to treatment and control groups, researchers can determine whether the intervention is effective in improving environmental outcomes.
• When testing the effects of a new technology: An experimental research design can be used to test the effectiveness and safety of new technologies or engineering designs. By randomly assigning participants or locations to treatment and control groups, researchers can determine whether the new technology is effective in achieving its intended purpose.

How to Conduct Experimental Research

Here are the steps to conduct Experimental Research:

• Identify a Research Question : Start by identifying a research question that you want to answer through the experiment. The question should be clear, specific, and testable.
• Develop a Hypothesis: Based on your research question, develop a hypothesis that predicts the relationship between the independent and dependent variables. The hypothesis should be clear and testable.
• Design the Experiment : Determine the type of experimental design you will use, such as a between-subjects design or a within-subjects design. Also, decide on the experimental conditions, such as the number of independent variables, the levels of the independent variable, and the dependent variable to be measured.
• Select Participants: Select the participants who will take part in the experiment. They should be representative of the population you are interested in studying.
• Randomly Assign Participants to Groups: If you are using a between-subjects design, randomly assign participants to groups to control for individual differences.
• Conduct the Experiment : Conduct the experiment by manipulating the independent variable(s) and measuring the dependent variable(s) across the different conditions.
• Analyze the Data: Analyze the data using appropriate statistical methods to determine if there is a significant effect of the independent variable(s) on the dependent variable(s).
• Draw Conclusions: Based on the data analysis, draw conclusions about the relationship between the independent and dependent variables. If the results support the hypothesis, then it is accepted. If the results do not support the hypothesis, then it is rejected.
• Communicate the Results: Finally, communicate the results of the experiment through a research report or presentation. Include the purpose of the study, the methods used, the results obtained, and the conclusions drawn.

Purpose of Experimental Design 

The purpose of experimental design is to control and manipulate one or more independent variables to determine their effect on a dependent variable. Experimental design allows researchers to systematically investigate causal relationships between variables, and to establish cause-and-effect relationships between the independent and dependent variables. Through experimental design, researchers can test hypotheses and make inferences about the population from which the sample was drawn.

Experimental design provides a structured approach to designing and conducting experiments, ensuring that the results are reliable and valid. By carefully controlling for extraneous variables that may affect the outcome of the study, experimental design allows researchers to isolate the effect of the independent variable(s) on the dependent variable(s), and to minimize the influence of other factors that may confound the results.

Experimental design also allows researchers to generalize their findings to the larger population from which the sample was drawn. By randomly selecting participants and using statistical techniques to analyze the data, researchers can make inferences about the larger population with a high degree of confidence.

Overall, the purpose of experimental design is to provide a rigorous, systematic, and scientific method for testing hypotheses and establishing cause-and-effect relationships between variables. Experimental design is a powerful tool for advancing scientific knowledge and informing evidence-based practice in various fields, including psychology, biology, medicine, engineering, and social sciences.

Advantages of Experimental Design 

Experimental design offers several advantages in research. Here are some of the main advantages:

• Control over extraneous variables: Experimental design allows researchers to control for extraneous variables that may affect the outcome of the study. By manipulating the independent variable and holding all other variables constant, researchers can isolate the effect of the independent variable on the dependent variable.
• Establishing causality: Experimental design allows researchers to establish causality by manipulating the independent variable and observing its effect on the dependent variable. This allows researchers to determine whether changes in the independent variable cause changes in the dependent variable.
• Replication : Experimental design allows researchers to replicate their experiments to ensure that the findings are consistent and reliable. Replication is important for establishing the validity and generalizability of the findings.
• Random assignment: Experimental design often involves randomly assigning participants to conditions. This helps to ensure that individual differences between participants are evenly distributed across conditions, which increases the internal validity of the study.
• Precision : Experimental design allows researchers to measure variables with precision, which can increase the accuracy and reliability of the data.
• Generalizability : If the study is well-designed, experimental design can increase the generalizability of the findings. By controlling for extraneous variables and using random assignment, researchers can increase the likelihood that the findings will apply to other populations and contexts.

Limitations of Experimental Design

Experimental design has some limitations that researchers should be aware of. Here are some of the main limitations:

• Artificiality : Experimental design often involves creating artificial situations that may not reflect real-world situations. This can limit the external validity of the findings, or the extent to which the findings can be generalized to real-world settings.
• Ethical concerns: Some experimental designs may raise ethical concerns, particularly if they involve manipulating variables that could cause harm to participants or if they involve deception.
• Participant bias : Participants in experimental studies may modify their behavior in response to the experiment, which can lead to participant bias.
• Limited generalizability: The conditions of the experiment may not reflect the complexities of real-world situations. As a result, the findings may not be applicable to all populations and contexts.
• Cost and time : Experimental design can be expensive and time-consuming, particularly if the experiment requires specialized equipment or if the sample size is large.
• Researcher bias : Researchers may unintentionally bias the results of the experiment if they have expectations or preferences for certain outcomes.
• Lack of feasibility : Experimental design may not be feasible in some cases, particularly if the research question involves variables that cannot be manipulated or controlled.

About the author

Muhammad Hassan

Researcher, Academic Writer, Web developer

You may also like

Applied Research – Types, Methods and Examples

Descriptive Research Design – Types, Methods and...

Mixed Methods Research – Types & Analysis

Qualitative Research – Methods, Analysis Types...

Textual Analysis – Types, Examples and Guide

Case Study – Methods, Examples and Guide

Experimental Design: Types, Examples & Methods

Saul McLeod, PhD

Editor-in-Chief for Simply Psychology

BSc (Hons) Psychology, MRes, PhD, University of Manchester

Saul McLeod, PhD., is a qualified psychology teacher with over 18 years of experience in further and higher education. He has been published in peer-reviewed journals, including the Journal of Clinical Psychology.

Learn about our Editorial Process

Olivia Guy-Evans, MSc

Associate Editor for Simply Psychology

BSc (Hons) Psychology, MSc Psychology of Education

Olivia Guy-Evans is a writer and associate editor for Simply Psychology. She has previously worked in healthcare and educational sectors.

On This Page:

Experimental design refers to how participants are allocated to different groups in an experiment. Types of design include repeated measures, independent groups, and matched pairs designs.

Probably the most common way to design an experiment in psychology is to divide the participants into two groups, the experimental group and the control group, and then introduce a change to the experimental group, not the control group.

The researcher must decide how he/she will allocate their sample to the different experimental groups.  For example, if there are 10 participants, will all 10 participants participate in both groups (e.g., repeated measures), or will the participants be split in half and take part in only one group each?

Three types of experimental designs are commonly used:

1. Independent Measures

Independent measures design, also known as between-groups , is an experimental design where different participants are used in each condition of the independent variable.  This means that each condition of the experiment includes a different group of participants.

This should be done by random allocation, ensuring that each participant has an equal chance of being assigned to one group.

Independent measures involve using two separate groups of participants, one in each condition. For example:

• Con : More people are needed than with the repeated measures design (i.e., more time-consuming).
• Pro : Avoids order effects (such as practice or fatigue) as people participate in one condition only.  If a person is involved in several conditions, they may become bored, tired, and fed up by the time they come to the second condition or become wise to the requirements of the experiment!
• Con : Differences between participants in the groups may affect results, for example, variations in age, gender, or social background.  These differences are known as participant variables (i.e., a type of extraneous variable ).
• Control : After the participants have been recruited, they should be randomly assigned to their groups. This should ensure the groups are similar, on average (reducing participant variables).

2. Repeated Measures Design

Repeated Measures design is an experimental design where the same participants participate in each independent variable condition.  This means that each experiment condition includes the same group of participants.

Repeated Measures design is also known as within-groups or within-subjects design .

• Pro : As the same participants are used in each condition, participant variables (i.e., individual differences) are reduced.
• Con : There may be order effects. Order effects refer to the order of the conditions affecting the participants’ behavior.  Performance in the second condition may be better because the participants know what to do (i.e., practice effect).  Or their performance might be worse in the second condition because they are tired (i.e., fatigue effect). This limitation can be controlled using counterbalancing.
• Pro : Fewer people are needed as they participate in all conditions (i.e., saves time).
• Control : To combat order effects, the researcher counter-balances the order of the conditions for the participants.  Alternating the order in which participants perform in different conditions of an experiment.

Counterbalancing

Suppose we used a repeated measures design in which all of the participants first learned words in “loud noise” and then learned them in “no noise.”

We expect the participants to learn better in “no noise” because of order effects, such as practice. However, a researcher can control for order effects using counterbalancing.

The sample would be split into two groups: experimental (A) and control (B).  For example, group 1 does ‘A’ then ‘B,’ and group 2 does ‘B’ then ‘A.’ This is to eliminate order effects.

Although order effects occur for each participant, they balance each other out in the results because they occur equally in both groups.

3. Matched Pairs Design

A matched pairs design is an experimental design where pairs of participants are matched in terms of key variables, such as age or socioeconomic status. One member of each pair is then placed into the experimental group and the other member into the control group .

One member of each matched pair must be randomly assigned to the experimental group and the other to the control group.

• Con : If one participant drops out, you lose 2 PPs’ data.
• Pro : Reduces participant variables because the researcher has tried to pair up the participants so that each condition has people with similar abilities and characteristics.
• Con : Very time-consuming trying to find closely matched pairs.
• Pro : It avoids order effects, so counterbalancing is not necessary.
• Con : Impossible to match people exactly unless they are identical twins!
• Control : Members of each pair should be randomly assigned to conditions. However, this does not solve all these problems.

Experimental design refers to how participants are allocated to an experiment’s different conditions (or IV levels). There are three types:

1. Independent measures / between-groups : Different participants are used in each condition of the independent variable.

2. Repeated measures /within groups : The same participants take part in each condition of the independent variable.

3. Matched pairs : Each condition uses different participants, but they are matched in terms of important characteristics, e.g., gender, age, intelligence, etc.

Learning Check

Read about each of the experiments below. For each experiment, identify (1) which experimental design was used; and (2) why the researcher might have used that design.

1 . To compare the effectiveness of two different types of therapy for depression, depressed patients were assigned to receive either cognitive therapy or behavior therapy for a 12-week period.

The researchers attempted to ensure that the patients in the two groups had similar severity of depressed symptoms by administering a standardized test of depression to each participant, then pairing them according to the severity of their symptoms.

2 . To assess the difference in reading comprehension between 7 and 9-year-olds, a researcher recruited each group from a local primary school. They were given the same passage of text to read and then asked a series of questions to assess their understanding.

3 . To assess the effectiveness of two different ways of teaching reading, a group of 5-year-olds was recruited from a primary school. Their level of reading ability was assessed, and then they were taught using scheme one for 20 weeks.

At the end of this period, their reading was reassessed, and a reading improvement score was calculated. They were then taught using scheme two for a further 20 weeks, and another reading improvement score for this period was calculated. The reading improvement scores for each child were then compared.

4 . To assess the effect of the organization on recall, a researcher randomly assigned student volunteers to two conditions.

Condition one attempted to recall a list of words that were organized into meaningful categories; condition two attempted to recall the same words, randomly grouped on the page.

Experiment Terminology

Ecological validity.

The degree to which an investigation represents real-life experiences.

Experimenter effects

These are the ways that the experimenter can accidentally influence the participant through their appearance or behavior.

Demand characteristics

The clues in an experiment lead the participants to think they know what the researcher is looking for (e.g., the experimenter’s body language).

Independent variable (IV)

The variable the experimenter manipulates (i.e., changes) is assumed to have a direct effect on the dependent variable.

Dependent variable (DV)

Variable the experimenter measures. This is the outcome (i.e., the result) of a study.

Extraneous variables (EV)

All variables which are not independent variables but could affect the results (DV) of the experiment. Extraneous variables should be controlled where possible.

Confounding variables

Variable(s) that have affected the results (DV), apart from the IV. A confounding variable could be an extraneous variable that has not been controlled.

Random Allocation

Randomly allocating participants to independent variable conditions means that all participants should have an equal chance of taking part in each condition.

The principle of random allocation is to avoid bias in how the experiment is carried out and limit the effects of participant variables.

Order effects

Changes in participants’ performance due to their repeating the same or similar test more than once. Examples of order effects include:

(i) practice effect: an improvement in performance on a task due to repetition, for example, because of familiarity with the task;

(ii) fatigue effect: a decrease in performance of a task due to repetition, for example, because of boredom or tiredness.

Data Science Texts

Discover what you don't know, and attack your weaknesses, nb: we may earn a commission if you buy something via an affiliate link., experimental design, strongly recommended prerequisites.

• Elementary Statistics

Recommended Prerequisites

Last updated: 8/29/2021.

Experiments are a very important component of data science; they're the only reliable way to to measure the change in one variable caused by another. Since it is often expensive to perform experiments, it is important to conduct them in an optimal way. There is no single design that is optimal for every scenario, so data scientists must be familiar with the specific designs that work well in the kinds of situations they generally encounter.

Recommended Books

Design and analysis of experiments with r, john lawson.

Key Features

• In-text exercises
• Example R code
• Completely Randomized Design
• Crossover and Repeated Measure Designs
• Designs to Study Variances
• Experimental Strategies for Increasing Knowledge
• Factorial Designs
• Fractional Factorial Designs
• Incomplete and Confounded Block Designs
• Linear Models
• Mixture Experiments
• Nested Designs
• Randomization
• Randomized Block Designs
• Replication
• Response Surface Designs
• Robust Parameter Design Experiments
• Split Plot Designs

Description

They say, Don't judge a book by its cover, but this book is almost worth buying for its cover alone. Lawson systematically covers the most common experimental situations and teaches you when each is appropriate. In many cases the basic designs that are introduced early in the text will work well. However, if you find yourself in a situation where your experimental units are highly heterogeneous with multiple factors that are hard to vary, you'll be glad you have this book to teach you the more obscure but more performant design. Lawson provides example code that takes some of the guesswork out of working with arcane R packages.

Statistics for Experimenters: Design, Innovation, and Discovery

George e.p. box, j. stuart hunter, william g. hunter.

• Solutions to some exercises
• Blocking and Randomization
• Data Transformation
• Designing Robust Products
• Elementary Probability and Statistics
• Evolutionary Process Operation
• Latin Squares Design
• Process Control, Forecasting, and Time Series
• Response Surface Methods
• Split Plot Design

This is the OG text on experimental design, and any data scientist who does a lot of experimentation will benefit from reading through it. We feel it's not the best possible experimental design book because it doesn't work as well as a reference as our top pick, and it tries to be too many things: an introductory statistics book, an experimental design book, an operations research book.... That said, it does have a lot of wisdom to offer on those subjects, so if you're interested in them this book will serve you well.

© 2019-2021 Pluripotent Limited

As an Amazon Associate Data Science Texts earns from qualifying purchases.

Accessibility Links

• Skip to content
• Skip to search IOPscience
• Skip to Journals list
• Accessibility help
• Accessibility Help

Click here to close this panel.

The Electrochemical Society (ECS) was founded in 1902 to advance the theory and practice at the forefront of electrochemical and solid state science and technology, and allied subjects.

Find out more about ECS publications

Simplified Thermal Runaway Model for Assisting the Design of a Novel Safe Li-Ion Battery Pack

Paul T. Coman 3,1 , Eric C. Darcy 2 and Ralph E. White 4,1

Published 11 April 2022 • © 2022 The Author(s). Published on behalf of The Electrochemical Society by IOP Publishing Limited Journal of The Electrochemical Society , Volume 169 , Number 4 Citation Paul T. Coman et al 2022 J. Electrochem. Soc. 169 040516 DOI 10.1149/1945-7111/ac62bd

Article metrics

3590 Total downloads

Share this article

Author e-mails.

[email protected]

Author affiliations

1 Department of Chemical Engineering, University of South Carolina, Columbia, South Carolina 29208, United States of America

2 NASA Johnson Space Center, Houston, Texas 77058, United States of America

Author notes

3 Electrochemical Society Member.

4 Electrochemical Society Fellow.

Paul T. Coman https://orcid.org/0000-0003-2876-7441

Eric C. Darcy https://orcid.org/0000-0002-8426-8607

• Received 8 December 2021
• Revised 10 March 2022
• Published 11 April 2022

Buy this article in print

This paper presents a simplified thermal runaway model (FEM) used to guide the design of a novel battery pack designed to resist thermal runaway propagation passively. The model is based on the heat equation for a 2D geometry with a heat generation term based on the maximum amount of energy measured using a custom-made calorimeter. The model was validated against experimental data using a 48-cell subscale of a full-scale battery pack for three different runs with three trigger cells with Internal Short Circuit Devices (ISCD) implanted in the separators. One trigger cell was placed at the edge, one placed in the middle, surrounded by six cells, and one placed in one corner of the subscale pack. It was shown that by simplifying the geometry and looking at the complex thermal runaway propagation mechanism only from a thermal perspective (no electrochemical reactions or fluid flow), the model predicted the experimental data with good precision. Furthermore, such a model was used to validate some experimental observations, which indicated the practicality of such a simplified design tool.

Export citation and abstract BibTeX RIS

This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited.

Li-ion batteries and cells are currently used in almost every application that requires the storage of electrical energy, from simple small devices, such as smartphones, to large aerospace applications. Although they possess superior qualities compared to other chemistries, Li-ion batteries have demonstrated that safety can be seriously altered due to thermal runaway if not operated correctly or electrically, mechanically, or thermally abused. 1 Thermal runaway is a chaotic phenomenon that is not yet fully understood and can yield unexpected catastrophic outcomes if the system is not adequately designed. The effect of a thermal runaway was seen in recent incidents, from the notorious Dreamliner incident to many other accidents and events in portable devices such as Samsung Galaxy Note 7 battery fires, exploding headphones, hoverboards, and even in stationary applications. 2 – 4 Due to such incidents, Li-ion batteries were the subject of various changes in the standards and regulations. 5 For these reasons, it is crucial to design battery packs that can withstand or even continue to function in the event of thermal runaway.

When designing battery packs, experimental testing can be time-consuming and expensive, requiring special safety features for the laboratory instruments; therefore, modeling thermal runaway is crucial for assisting the design of safe battery packs. However, modeling thermal runaway is generally complicated because the inside phenomena depend on many parameters such as cell chemistry, geometry, embedded safety features, electrolyte blends, etc. A significant number of models for predicting thermal runaway in single cells are available in the literature. Most of them are based on the work pioneered by Hatchard et al. 6

It is commonly agreed that the energy released during thermal runaway is due to exothermic reactions caused by the decomposition reactions between the active materials and the electrolyte. This was proven both in Refs. 6 , 7 , but also in other significant works that followed. 8 , 9 In addition to the decomposition reactions between the electrolyte and electrodes, it was also found that the electrochemical energy due to the short circuit has a significant contribution to the energy amount released during thermal runaway. 10 , 11 Later, endothermic reactions due to vaporization and venting of electrolyte were added to the decomposition reactions, proving that for more accurate modeling of thermal runaway, it is important to include both vaporization and venting. 12 , 13 Physics-based electrochemical models (e.g., pseudo-2D-models or P2D) can also be used to provide an understanding of the internal phenomena and to provide tools that can be used to model thermal runaway, but they are scarce due to the high number of parameters required both for the electrochemical reactions and for the thermal runaway decomposition reactions. Some papers used the P2D approach to understand the internal short circuit, for example. 14 , 15 Experimentally, some works done by Donal Finegan et al. 16 – 18 have shown the complexity of the thermal runaway phenomena and helped to gain more insight into the dynamics occurring inside cylindrical cells with different safety features (top vent, bottom vent, etc.) during thermal runaway. Significant work on experimental testing and modeling of thermal runaway was also done by Xuning Feng et. al., 19 – 22 which helped to deconvolute the energy released by different reactions in some sets of cells and provide a better insight into the dynamics inside a cell during thermal runaway.

Understanding and accurately modeling such phenomena requires good knowledge of the interaction between each component inside the cell and include structural or thermodynamic models, as seen in Refs. 23 – 25 . However, even with all the state-of-the-art understanding of the thermal runaway phenomena, there is still a broad distribution of energy released for the same cell type that cannot be predicted, 25 , 26 which makes it challenging to find a modeling pattern. Using calculations of probability and performing sensitivity analysis, Akos et al. 27 also demonstrated significant variability in the severity of thermal runaway, which depends on many parameters. They indicated that the triggering input energy before and after the internal-short circuit is influential and can alter the severity of thermal runaway. They also indicated that reproducibility and repeatability might be achieved for sets of initial parameters.

Modeling battery packs or modules add even more complexity because of the different design elements that can play an essential role in conducting the heat or even the fumes inside the pack or outside. In addition to the thermal design and heat flow, the electrical series-parallel configuration can change the battery pack's temperature dynamics by adding extra heat in a parallel configuration, as shown in Refs. 28 , 29 . Different modeling strategies were developed and applied for investigating thermal runaway propagation, such as using a thermal resistance network and electric circuit analogy, 20 , 21 , 30 – 35 internal resistance methods, or electrochemical-thermal physics-based model with decomposition reactions as the main contributors to the heat generation. 30

The thermal runaway models available in the literature are not one-fits-it-all, especially if the goal is to design a pack that can withstand a worst-case-scenario thermal runaway event. Additionally, all the models need adjustments and fitting to experimental data. For example, modeling a thermal runaway event in a battery pack using a physics-based electrochemical model would make it very complex and would require many parameters such as the activation energy, and the pre-exponential factors just for the thermal runaway event, let alone including the diffusion and electric charge distribution. Additionally, one would still need to have complex experimental data to get those parameters. Including a reduced model such as the single-particle model will reduce the complexity significantly, but still requires many parameters, including the activation energies and also diffusion coefficients, etc. Equivalent circuit models can also be used, but, again, fitting is required and coupling with electrochemical decomposition reaction. Besides, when designing a battery pack that can withstand a severe thermal runaway event, one needs to know a realistic worst-case scenario, based on real cells of different chemistries. This means that a heat generation profile or a function needs to be defined to account for thermal runaway. Whether the heat generation is anchored to some physics-based interpretation of the phenomenon inside the battery cell or defined as a set of decomposition reactions with the enthalpy of each reaction, in the end, the result is a heat generation function that goes into the heat equation. The path to obtaining this function can lead to a variety of interpretations and models, but having a simple heat generation rate that is based on a wide range of experimental data for different cells, can be helpful to assist the design of a safe battery pack, by providing the most severe scenario, when the trigger cell delivers the highest amount of energy for a specific type of cell configuration.

Therefore, for analyzing designs, simplified models might provide the same insights as more complicated ones, but with fewer resources and fewer parameters to fit. In this paper, a simplified semi-empirical thermal runaway model based on experimental-data determined heat generation is presented and validated against experimental data performed on a novel battery design. The novel design includes a combination of highly conductive paths for dissipating the heat during thermal runaway and performance (charge and discharge) but in combination with an insulating ceramic matrix that prevents the propagation of thermal runaway. Despite the abstraction of the modeling system, the model can easily guide the design of battery packs without sacrificing computational resources and with minor differences. An important question guided this study: can an overly simplified thermal model be used to model such a complex phenomenon and validate experimental findings and observations?

Battery Pack Description

The battery pack in this project was designed to provide 2 kWh of energy at high rates (3C) and to resist thermal runaway propagation passively. The full-scale battery pack is shown in Fig. 1 and contains design features chosen after a thorough electrical and thermal investigation.

Figure 1.  Sketch showing (a) the CAD design for the full battery pack, (b) the design components of the cell nests with the ceramic putty matrix and the Superwool compression paper, and (c) the CAD design of a single 16-cell nest.

Download figure:

The full pack consists of two modules of 96 cells per module (Fig. 1 a). The 18650-format cells are nested in pairs of 16, and each cell was glued on a 60° contact angle to the aluminum spine with Masterbond epoxy (Fig. 1 c). The spines serve as heat sinks, conducting the cells' heat to the top and bottom headers. In the final design, the headers are exposed to active cooling through cooling coils embedded in an external casing. To prevent cell-to-cell propagation between different nests and the adjacent cells inside the nest, insulating putty (Thermeez 7020) putty was added between the cells, and Superwool compression paper was added between the nests, as seen in Fig. 1 b. For keeping the cells in place, channeling ejecta, and hosting the nickel strips installed for conducting the electricity to the cells, each nest has capture plates placed at the top and bottom of the cells. Electrically, the cells are connected in a 12S-8P configuration to deliver the 2kWh. However, modeling this entire full-pack system is the subject of future work. The system was split into subscales to demonstrate the importance of some design details and test the margins at a smaller scale, as seen in the next section.

Model Description

Modeling thermal runaway propagation can be done in many creating ways, depending on the level of details required for each design. For understanding the physics of the thermal runaway phenomena, it was shown in a previous publication that the decomposition reactions could and should be added to the model, 30 but this can add too much complexity. Such an approach would slow down the simulations and require extra fitting parameters that are not always easy to obtain. If, for example, the influence of design parameters is the investigation's goal, a simplified model might provide a fast and smart design tool. The model described here was a two-dimensional FEM built in COMSOL v5.4 and was based on the heat equation (Eq. 1 ):

Table I.  Material properties for each computation domain.

a) -assumed parameter, due to lack of material specifications. However, these values do not influence the 0.152 mm layer of Epoxy and are small for the 2 mm Superwool compression paper. b) This parameter was adjusted to 0.188 W m −1 K −1 to fit experimental data.

The heat generation rate, in this case, was taken from empirical observation and calorimetric measurements for an LG 18650-MJ1 cell (Nominal theoretical capacity: 3.5Ah, Capacity at testing: 3.43 Ah, Nominal voltage: 3.64 V, Energy at 100% SoC: 12.7 Wh, Mass: 49.0 g), which is the cell with the highest energy released during thermal runaway. According to Ref. 26 , for an MJ1 cell, the total average thermal runaway energy released is 75.2 kJ, with a standard deviation of 6.6 kJ, and the distribution with the probability density is given in Fig. 2 (and Fig. 5 in Ref. 26 ).

Figure 2.  Predicted TR Energy for an LG 18650 MJ1 cell. 26

This wide range indicates how different the amount of energy coming from the same cell can be. This energy, however, must be decoupled, as parts of it are leaving the cell. For simulating thermal runaway in this design, the maximum amount of energy measured experimentally for such a cell was 82.9 kJ (see the last data point for MJ1 in Fig. 5 in Ref. 26 ), and a percentage of 38.8% of that represents the maximum amount of energy that stays in the cell body, contributing to the heat generation. This percentage represents the percentage of energy released only through the cell body. This amount is very important for a heat transfer analysis, in a system in which the adjacent cells are generally protected from ejecta, as is the case in the pack presented in Fig. 1 . The amount was measured experimentally, and it is the highest amount measured in a batch of 15 cells. However, in Ref. 26 , the median percentage is reported. That is, 21.7%. However, for determining the worst-case scenario for a thermal runaway, the median values are not enough, therefore 38.8% of the maximum amount released by an MJ1 (82.9 kJ) was chosen for this study. A total maximum amount of 32.2 kJ was used as a reference for the energy remaining in the cell. It was experimentally observed that the median TR duration was approximately 1s to reach the peak point in the heat generation and 5 s until the exothermic reactions lapse, which yields the approximated heat rate profile shown in Fig. 3 .

Figure 3.  Approximated heat generation rate for MJ1 cells. Integrated over time, the heat rate yields a total energy of 32.2 kJ.

For each run, the time for the heat rate activation was adjusted so that the starting point corresponds to the thermal runaway onset time, which will be explained in the Results section for each case.

Boundary Conditions

It is known from experimental observations that the pack was exposed to natural convection, except during thermal runaway, when smoke and flames are flowing in between the capture plates and towards the vents. This might influence two of the boundary conditions (left and right) inside the pack. The high-speed flow should, in reality, change the boundary conditions from natural convection to forced convection for a short amount of time. However, for simplicity, only natural convection was applied to all the boundaries with a heat transfer coefficient of h = 20 W m −2 K −1 . This assumption is discussed in detail in the Results section.

Assumptions

A couple of assumptions were made to simplify the model:

• •   No pack cover was included in the model.
• •   No air or gas flow.
• •   No radiation.
• •   The mass loss due to the presence of ejecta was modeled through the change in the density when thermal runaway occurs.
• •   The heat transfer coefficient value was effective and constant for all the exterior boundary conditions.

Experimental set-up

The full-scale battery pack was split into different subscales for modeling purposes. A 48-cell pack was initially built and inserted in a casing with a copper mesh at one side that filters flames and ejecta. This strategy allows the flow of gases only through a Gore-Tex vent, but not the flames. Three runs with three trigger cells were chosen for validating the model in three different positions inside the pack, as seen in Fig. 4 :

• One corner cell (Cell 25) close to the heat sink header, surrounded by only two adjacent cells.
• One middle cell (Cell 46) surrounded by six adjacent cells.
• One edge cell (Cell 5) surrounded by four adjacent cells.

Figure 4.  Pictures showing (a) the sketch of the thermocouple map, (b) the 46-cell design assembled, and (c) the casing which hosts the battery pack.

The middle and the edge trigger cells (C#5 and C#46) are LG MJ1 18650 format, whereas the corner trigger cell (C#25) was an LG M36 18650 format. Each position was unique and emulated both the worst-case scenarios as well as the least aggressive ones. The trigger cells were equipped with Internal Short Circuit Devices (ISCD), 30 which trigger thermal runaway internally when heated to approximately 60 °C. Each trigger cell was heated through electric heaters at the bottom until thermal runaway was triggered, then the heater was turned off. Thermocouples were placed mid-height in the positions shown in Fig. 4 .

Edge cell trigger cell

This trigger cell position's uniqueness consists of the fact that it was surrounded by four adjacent cells, two across the highly conductive aluminum spine, and two across the ceramic putty (Fig. 5 b). The temperatures measured for this run are shown with marked lines in Fig. 5 a. In this run, it took 2487 s (41.45 min) for the trigger cell to reach the temperature of 60 °C.

Figure 5.  Plots showing (a) the temperatures measured on the trigger cell skin (TC5) and the adjacent cell (TC6) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time (2492 s).

Figure 6.  Plot showing the heater power as a function of time for the edge trigger cell.

Figure 7.  Plots showing (a) the temperatures measured on the trigger cell skin (TC5) and the adjacent cell (TC6) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time when the thermal conductivity of the Epoxy changed to air.

Although this difference was not very obvious for this particular cell position, it became valuable for the next run, as explained later. Next, the thermal conductivity of the Superwool layer around the trigger cell was changed from k = 0.06 W m −1 K −1 (Table I ) to k = 1.7 W m −1 K −1 (which corresponds to Carbon), assuming that the area of the Superwool carbonized. By looking at Fig. 8 , it can be seen that the temperature of the trigger cell predicted the experimental data slightly better.

Figure 8.  Plots showing (a) the temperatures measured on the trigger cell skin (TC5) and the adjacent cell (TC6) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time when the thermal conductivity of the Epoxy changed to air, and the thermal conductivity of Superwool changed to Carbon.

Middle pack trigger cell

This particular trigger cell position's uniqueness consists of the fact that the cell was surrounded by six adjacent cells, two across the highly conductive aluminum spine, and four across the ceramic putty. Because the trigger cell was in the middle of the pack, both the adjacent cells and the aluminum spine around the trigger cell removed the heat better than in the previous run. This arrangement caused this run to take longer to trigger thermal runaway. It took 3580 s (∼1 h) for the trigger cell to reach 60 °C. For modeling purposes, the trigger cell's heat generation corresponding to the maximum amount of energy for an MJ1 was reduced by 12%. For this run, space did not permit the installation of a thermocouple that would measure the trigger cell's temperature. Therefore two thermocouples were placed on the adjacent cell at the bottom and the heat sink above, as seen in Fig. 9 b. In this case, the thermocouple's temperature in the heat sink, next to the adjacent top cell, was predicted higher by the model than the temperature measured experimentally. This indicates that there was too much heat removed through the heat sink and that, in reality, there was another phenomenon that should impede the heat transfer. As was the case in the previous run, Epoxy's change was considered in this run as well. It was speculated that large parts of the epoxy layer volatilized, and an air gap was formed, insulating the trigger cell and preventing the heat dissipation to the adjacent cells across the highly conductive heat sink.

Figure 9.  Plots showing (a) the temperatures measured on the two adjacent cells (TC47 and TC45hs) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time, with no changes in the epoxy layer's properties.

The next step was to include the change in the thermal conductivity of the epoxy layer. It can be seen from Fig. 10 a that the modeling predictions were very similar to the experimental data, proving that the observation that the epoxy layer vanishes is plausible. The Superwool layer was placed only at the edges of the pack; therefore, no changes in its properties are required for modeling the trigger cell in the middle of the subscale. By looking at the comparison between the surface plots (Fig. 9 b and Fig. 10 b), it can be visually observed that if the epoxy layer does not change properties, a large thermal gradient appears at the boundary between the trigger cell and the heat sink, indicating a fast thermal path. In contrast, an insulating layer keeps the heat trapped in the trigger cell. The question that arises is, what happens with the adjacent cells across the ceramic putty in this case? The comparison between Fig. 9 a and Fig. 10 a shows that the adjacent cell C#47 was not affected by having an insulating barrier in the case of epoxy vaporization/volatilization. This observation indicates that having an insulating putty between the trigger cells on one side and a conductive material on the other side can prevent thermal runaway propagation, especially if the epoxy layer becomes an insulator at high temperatures. Simultaneously, during operation (charge and discharge), the conductive epoxy layer helps dissipate the heat from the cells to the aluminum heat sink, reducing the overall pack temperature.

Figure 10.  Plots showing (a) the temperatures measured on the two adjacent cells (TC47 and TC45hs) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time when the thermal conductivity of the Epoxy changed to air.

Corner cell trigger cell

This trigger cell position's uniqueness is that the cell was surrounded by two adjacent cells only, one across the highly conductive aluminum spine and one across the ceramic putty. For testing the pack's high margins, the cell used in this run was more energetic than the MJ ones. An LG M36 18650-format cell was used for this run and, for fitting the experimental data, the heat generation was increased by 30%. Figure 11 a shows that the adjacent cell's temperature across the aluminum spine predicted by the model was higher than the experimental one, which leads to the same observation from the other two runs.

Figure 11.  Plots showing (a) the temperatures measured on the two adjacent cells (TC26 and TC25) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time, with no changes in the epoxy layer's properties.

When adding a change in the Masterbond epoxy layer's thermal conductivity, the model predicted the experimental temperatures with modest differences (less than 5K difference at peak temperatures), as seen in Fig. 12 . Even with a cell with higher energy in the body during thermal runaway, the adjacent cell temperatures were under 100 °C, which means no propagation was expected in this novel design. Finally, the Superwool layer's carbonization was also considered since this was a corner cell with such a layer pressed against it by the G10 edge plate. By comparing the temperatures predicted by the model for the no-carbonization case from Fig. 12 with the carbonization case in Fig. 13 , it can be observed that there are minor differences between the two. The difference in TC25 was barely noticeable, but the adjacent cell's temperature over the ceramic putty (TC26) shifts to the left in time, yielding a slightly better experimental fitting. However, the difference was still small, but this shift in the temperature indicates that a change in the material properties for the Superwool layer was still something to account for a better prediction.

Figure 12.  Plots showing (a) the temperatures measured on the two adjacent cells (TC26 and TC25) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time when the thermal conductivity of the Epoxy changed to air.

Figure 13.  Plots showing (a) the temperatures measured on the two adjacent cells (TC26 and TC25) and (b) the surface plot of the area around the trigger cell 5 s after the onset thermal runaway time when the thermal conductivity of the Epoxy changed to air.

Although no gas flow was considered in the model, it is worth mentioning that the thermal runaway dynamics are extremely complicated and, as mentioned previously, chaotic in behavior and random, making it difficult to model or formulate a precise pattern. For this run, for example, the thermal runaway process was so aggressive that the cell walls were breached and ejecta came from both the top, side, and bottom, as seen in Fig. 14 .

Figure 14.  Pictures showing the post-mortem analysis of the corner trigger cell (Cell TC25) and the bottom and side breach holes.

Modeling such an effect is beyond the scope of this project. Still, it is important to observe that by using a simplified thermal model with a percentual change in the heat rate with reference to a maximum measured energy, the model predicted the experimental data with acceptable precision. The model can reveal and validate some of the experimental findings.

Finally, it is important to look at the assumption regarding the heat transfer coefficient. As previously specified, the heat transfer coefficient was assumed effective and equal to 20 W m −2 K −1 for all the boundaries. However, for an ideal system, the heat transfer coefficient can be calculated from the Nusselt number, which depends on the film temperature around the walls exposed to air and the airflow velocity. Considering that a high flux of hot gases is circulating inside the sealed box during and after the thermal runaway, it is difficult to find an ideal approximation for this coefficient for each wall (e.g., the left wall that is closer to the vent will experience higher gas flow than the one to the right). However, for the sake of comparison, a heat transfer coefficient is calculated for natural convection considering an isothermal vertical wall at an ambient temperature of 23 °C (ambient filled with air). By using the Nusselt number, the calculated heat transfer coefficient is approximately 8 W m −2 K −1 for a wall temperature of 60 °C and 18 W m −2 K −1 for a wall temperature of 600 °C, which can be reached right after the thermal runaway. Using the lowest value (8 W m −2 K −1 ), it can be seen from Fig. 15 that the model still shows that the difference between the experimental data and the model is small, especially when looking at the comparison around the peak points, which are of high interest when looking for the margins to thermal runaway propagation. The difference is more visible in the thermocouple placed on the adjacent cell closer to the edge, whereas, in the cell's temperature across the heat sink, the difference is minimal.

Figure 15.  Plot showing the comparison between the experimental data and the model when two different heat transfer coefficients were considered.

Even with a higher heat transfer coefficient (h = 100 W m −2 K −1 ) on the top and bottom boundary conditions, the difference was very small. This means that with the current system, the boundary conditions are not bottlenecks for this design if thermal runaway mitigation is the goal. The most important parameter, so far, was the transition of the Epoxy to air. Overall, it can be concluded that the model, despite its simplification, can be used to investigate the design parameters and validate some experimental observations.

Conclusions

Thermal runaway is a complex phenomenon that can be modeled using different creative methods, depending on the system and the final goals. Regardless of the method chosen, the current models require the fitting of many parameters, which makes the modeling process very slow. If the decomposition reactions during thermal runaway are included in the model, for example, the activation energies for every single reaction need fitting. Additionally, the chaotic behavior of thermal runaway causes each cell to have unique behavior, which makes it challenging to find a pattern to model. The study's goal presented in this paper was to show that such a complex phenomenon can be modeled using an overly simplified thermal model with a heat generation term based on the maximum amount of energy released during thermal runaway. Compared with the experimental data, the model showed a good prediction of experimental data. Using the model enables observing and validating experimental observations, which made it a fast and useful tool. However, it should be noted that the model proved to be suitable for analyzing the design elements that were of high interest in this particular battery pack design. Still, it has limitations if other phenomena such as gas flow, wall breaching, current distribution, etc. are of interest. However, some of these limitations can be reduced or eliminated by deconvoluting the heat generation term even more and by adding the percentage of ejecta that leaves the cell, as it is available experimentally in references such as Ref. 40 .

• INNOVATION FESTIVAL
• Capital One

08-02-2024 DESIGN

10 great ‘beach reads’ for art and design lovers

It’s summer vacation season! Here’s a (hefty) list of the season’s best new design books you should take with you.

BY  Zachary Petit 5 minute read

A “beach read” means different things to different people. There are those who simply shove a couple paperback thrillers or romance novels into their vacation bag and call it a day. And there are those who thrill at the nerd joy of a bit more substance—like, say, a no-holds-barred treatise on the ethics of their chosen profession (*heart flutters*). Or, a book seemingly containing every single piece of famous graphic design ever created. Sure, all that substance is often substantially more substantial—less “beach read” and more “tidepool tome.” But here’s to those who schlep hardbound books with 80-pound uncoated stock to the seaside respite of their choice.

No matter how loadbearing your beach tote is, if you’re looking to escape the mass-market mystery for some sublime design reading this summer, these 10 books will help you do just that.

Graphic Classics

Phaidon selected a remarkably humble title for a remarkably vast volume—one containing 500 of the most iconic pieces of design ever made. From record sleeves to Gutenberg bibles and virtually every designer you’ve ever heard of (per the publisher, 400 of them, from 33 countries and five continents), this book is a powerhouse of inspiration, not to mention an overwhelming collection of the raw materials that have shaped decades of designers. The book isn’t out just yet—but will ship mid-August, in time for your summer’s end sojourn. 

Atlas of Never Built Architecture by Sam Lubell and Greg Goldin

If Graphic Classics contains every piece of brilliant design ever made, this book captures every brilliant (and, well, sometimes just brilliantly strange) piece of architecture never made. From the Domino’s Leaning Tower of Pizza to Frank Lloyd Wright’s (literally) mile-high skyscraper , the authors capture an endlessly fascinating taxonomy of facade futures that were not meant to be. 

What It Means to Be a Designer Today: Reflections, Questions and Ideas from AIGA’s Eye on Design

I’m proud to have published many articles in Eye on Design over the years—because it was one of the best venues around for insightful industry commentary and critique, not to mention delightfully unexpected design odds and ends. AIGA may no longer be publishing it at the moment, but the good news is that this book (edited by Fast Company ’s Liz Stinson and Scratching the Surface ’s Jarrett Fuller) distills all the best parts of the site down into a 240-page volume that strikes at the heart of the book’s title—and, against all odds, nails it.

Cuban Mid-Century Design: A Modernist Regime by Abel González Fernandez, Laura Mott and Andrew Satake Blauvelt

In the wake of the Cuba’s revolution, designers navigated the balance of international embargoes and consumer demand—and ultimately created a wholly Cuban take on Modernism in the process. This book (which accompanies a Cranbook exhibition that’s currently on view) bills itself as the first to document Cuban Midcentury design—and it does so with great discoverability, while revealing much more. Per the publisher, “This volume brings together a prolific cohort of artists, designers and architects that materialized the ideology of their time, and ultimately narrate[d] the country’s arc from revolutionary promise to authoritarian retrenchment.”

American Modern: Architecture; Community; Columbus, Indiana by Matt Shaw, with photographs by Iwan Baan

You may be surprised, as I—a Midwest native—was, that of all places, Indiana was home to a preeminent creative hub for incredible and impactful Modernist architecture. When I last visited Columbus, I cobbled together a tour based on Wikipedia pages—which is why I’m delighted that Shaw and Baan’s guide now exists and fills in the gaps of how this all came to be, covering works by Eero Saarinen, Eliot Noyes, I.M. Pei, “and the generations-long quest to develop the ideal American city through design excellence.” 

Shopkeeping: Stories, Advice and Observations by Peter Miller

Compared to some of the behemoths on this list, this slim hardcover is indeed suitably portable and beach-worthy. In an airy 144 pages, shopkeeper Peter Miller—of his eponymous design bookshop in Seattle, which he opened in 1980—riffs on his trade, from the philosophies of designing a retail space to the nature of time within a shop. But one needn’t be a shopkeeper to appreciate what Miller brings to the page: meditative, humorous New Yorker –style musings that reveal the secret inner life of the highly designed ecosystems we take for granted.

Iris Alba: Art Director

Largely a fascinating visual collection—with the exception of a prologue by Ellen Lupton and a handful of essays— Iris Alba introduces you to Iris Alba, a “missing link when trying to understand Latin American book cover design,” as the book’s co-editor, Leandro Castelao, dubs her.

While Milton Glaser, Seymour Chwast, and the rest of the Pushpin group was working its pop and psychedelic magic in New York City, Alba flexed her own interpretation of it in Argentina, reshaping book cover design in Latin America in the process.

Modern Man in the Making by Otto Neurath

On face value, Neurath’s legendary 1939 book simply documents the era in which he lived via facts and figures. But viewed retrospectively, as Lars Müller Publishers notes, “The book can be seen as one of the most influential predecessors of today’s ever-present infographics. … Modern Man in the Making shows Neurath’s democratic endeavor to make knowledge intelligible and available to all.” This facsimile edition cleanly falls into the realm of a storied design curiosity—and because of that, I can confirm that there’s no better time to explore it than the veritable void of a vacation, when time is less of a premium and you’re free to let your mind frolic and weigh its impact. 

How Directors Dress

Midsommar . Ex Machina . Everything Everywhere All At Once . You probably know A24’s eccentric (and perfect) films, but you might not know that the studio also has a damn good publishing line. Case in point: How Directors Dress , which “uses clothing to tell exciting new stories about directors, their lives, their movies, and the times in which they were made.” Featuring more than 200 photographs and writing by fashion and style journalists, the book grants entirely new insights into familiar films and their creators—and a symbiosis that’s far from surface level.

Footnotes From the Most Fascinating Museums by Bob Eckstein

Sure, this is a list of beach reads. But that doesn’t mean vacations don’t leave some among us yearning for the confines of a museum. Get the best of both worlds with this New Yorker cartoonist’s collection that captures more than 75 institutions (from MoMA and the Guggenheim to the La Brea Tar Pits and Museum) in 155 pieces of art. Each is a charming profile accompanied by an insight or reflection on the museum—and collectively they showcase just why, exactly, some of us might choose to haunt dark air-conditioned corridors when the warm ocean is right there in front of us.

Apply to the Most Innovative Companies Awards and be recognized as an organization driving the world forward through innovation. Early-rate deadline: Friday, August 23.

ABOUT THE AUTHOR

Zachary Petit is a contributing writer for Fast Company and an independent journalist who covers design, the arts and travel. His words have appeared in Smithsonian, National Geographic, Eye on Design, McSweeney’s, Mental_Floss and PRINT, where he served as editor-in-chief of the National Magazine Award–winning publication   More

Explore Topics

• Tech Why Impossible Foods isn’t focusing on the climate impact of its plant-based meat alternatives
• Tech Microsoft says Iran increasing cyber activity to influence U.S. election
• Tech Samsung stove recall hits 1 million electric ranges; hundreds catch fire, causing injuries, property damage, and pet deaths
• News EpiPen gets a new competitor: FDA approves anaphylaxis nasal spray
• News Cash App might owe you money: How to find out and file a settlement claim
• News The Kamala Harris campaign is hiring for almost 60 positions: What they say about her tech-focused, social-first strategy
• Design Vapes for migraines and asthma? Companies seek to speed up relief time, but some have concerns
• Design With new changes at Southwest and Spirit, it’s getting harder to tell the airline brands apart
• Design Michaels says it accidentally sold AI Halloween art
• Work Life MrBeast’s confidential memo to ‘Team Beast’ employees ordered big changes. Here’s what it said
• Work Life A therapist explains how knowing your work archetype could help with stress
• Work Life How to make AI more emotionally intelligent

Grab your spot at the free arXiv Accessibility Forum

Help | Advanced Search

General Relativity and Quantum Cosmology

Title: two novel $f(q)$ models.

Abstract: We propose two novel models in the framework of $f(Q)$ gravity to explain our accelerated universe, namely the exponential $f(Q)_{EXP}$ model and the hyperbolic tangent $f(Q)_{HT}$ model. The current cosmological electromagnetic observations including the cosmic microwave background anisotropies (CMB), the baryon acoustic oscillations(BAO), the type Ia supernovae (SN) and the direct measurements of H(z), combined with the simulated gravitational-wave data are used to constrain the $f(Q)$ models. We find that the Hubble tension can be significantly alleviated to $1.40\sigma$ level in the $f(Q)_{EXP}$ model. The fitting $\chi^2$ of the $f(Q)_{HT}$ model is $9.75\sigma$ poorer than that of the $f(Q)_{EXP}$ model, implying the $f(Q)_{HT}$ model would be excluded by future gravitational-wave observation.

Submission history

Access paper:.

• HTML (experimental)
• Other Formats

References & Citations

• INSPIRE HEP
• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 08 August 2024

Optimization design of hydrocyclone with overflow slit structure based on experimental investigation and numerical simulation analysis

• Shuxin Chen 1 , 3 ,
• Donglai Li 1 ,
• Jianying Li 2 &
• Lin Zhong 1  

Scientific Reports volume  14 , Article number:  18410 ( 2024 ) Cite this article

Metrics details

• Chemical engineering
• Mechanical engineering

This study aims to address the issue of high energy consumption in the hydrocyclone separation process. By introducing a novel slotted overflow pipe structure and utilizing experimental and response surface optimization methods, the optimal parameters were determined. The research results indicate that the number of slots, slot angles, and positioning dimensions significantly influence the performance of the hydrocyclone separator. The optimal combination was found to be three layers of slots, a positioning dimension of 5.3 mm, and a slot angle of 58°. In a Φ100mm hydrocyclone separator, validated through multiple experiments, the separation efficiency increased by 0.26% and the pressure drop reduced by 24.88% under a flow rate of 900 ml/s. CFD simulation verified the reduction in internal flow field velocity and pressure drop due to the slotted structure. Therefore, this study provides an effective reference for designing efficient and low-energy hydrocyclone separators.

Similar content being viewed by others

A novel strategy for comprehensive optimization of structural and operational parameters in a supersonic separator using computational fluid dynamics modeling

Investigation of novel passive methods of generation of swirl flow in supersonic separators by the computational fluid dynamics modeling

Numerical study on the stimulation effect of boundary sealing and hot water injection in marine challenging gas hydrate extraction

Introduction.

Hydrocyclones are commonly used rotary flow separation and classification devices in industrial applications, owing to their simple structure, high separation efficiency, small footprint, and large processing capacity 1 , 2 . However, hydrocyclone separation performance is affected by structural parameters, with the overflow pipe being particularly important and the major factor influencing pressure drop 3 . Overflow pipe design parameters include length, insertion depth, diameter, and dimensions 4 .

Previous research has made significant progress in hydrocyclone structural optimization and numerical simulation. Some studies focused on optimizing the overflow pipe, such as increasing the distance for short-circuiting flow to enter the bottom, which improved internal pressure distribution 5 , 6 . Additionally, computational fluid dynamics (CFD) simulation of a hydrocyclone with conical section and dual tapered inlet showed significantly increased tangential velocity and axial velocity. This enhances centrifugal force on particles and reduces misplaced particles 7 . Adding a conical top to the overflow pipe improved fine particle separation efficiency but did not affect pressure drop 8 .

Despite these advances, inherent fluid flow characteristics lead to imperfect separation and high energy loss regardless of geometry. To further enhance performance, various designs have been explored, including introducing a center body 9 , 10 , inner cone 11 , 12 , double overflow pipe 13 , 14 , 15 , 16 , overflow pipe with conical top 17 , overflow cap 18 , 19 , and slit cone 20 , 21 , 22 . By altering hydrocyclone geometry, these designs improved separation performance. The overflow cap reduced air core diameter, decreasing energy consumption while increasing tangential velocity and centrifugal force, and decreasing axial velocity 18 , prolonging particle separation time and improving efficiency.

Numerical simulation has also been utilized to study multiphase flow in hydrocyclones. Despite different models and methods, these simulations accurately described the complex phenomena, demonstrating the extensive application of numerical techniques in multiphase flow research 23 , 24 , 25 , 26 , 27 .

While previous studies have focused on the overflow pipe, optimization of other structural parameters has been inadequate. Moreover, past research primarily considered specific particulate types and concentrations rather than comprehensive optimization across operating conditions. To address these limitations and further enhance performance, this study aims to design a slit conical overflow pipe hydrocyclone and optimize multiple key structural parameters. The significance of this research is that it will provide new perspectives to improve hydrocyclone performance and application in industrial fields, holding promise for resource conservation and environmental protection.

The research will combine experimental investigation and numerical modeling to obtain separation data under different parameters. Accurate numerical simulation will be utilized to model internal multiphase flow and determine optimal designs. Through improved design and accurate modeling, this study will provide new perspectives to enhance hydrocyclone performance and application, holding promise for resource conservation and environmental protection in industrial fields.

Overflow pipe structural design scheme

Geometry and dimensions of the overflow pipe structure are crucial factors affecting the pressure drop of a hydrocyclone. In order to increase throughput and reduce flow losses, thus lowering pressure drop, enlarging the diameter of the overflow pipe can be adopted. However, it should be noted that excessively large overflow pipe diameter may increase the probability of solid particles entering the overflow pipe region, leading to reduced separation efficiency of the hydrocyclone 28 , 29 . In this study, we improved the overflow pipe structure of a conventional 100 mm hydrocyclone by incorporating a slotted design to ensure separation efficiency remains unaffected while enhancing throughput and reducing energy consumption.

The introduction of the slotted structure significantly reduces the pressure drop and energy consumption of the hydrocyclone. The main mechanism behind this improvement lies in the increased outlet area of the overflow pipe achieved through the slotted design, thereby reducing fluid kinetic energy losses. According to the Bernoulli energy conservation law, higher fluid velocities result in greater kinetic energy losses and lower pressures. Hence, the slotted structure reduces fluid velocity inside the overflow pipe, thereby increasing outlet pressure, effectively lowering the overall pressure drop of the hydrocyclone. Moreover, theoretically, the slotted structure helps reduce short-circuit fluid flow entering the bottom outer vortex of the hydrocyclone, thereby reducing kinetic energy losses in the bottom region and contributing to overall pressure drop reduction within the hydrocyclone. Properly setting the number of slots, angles, and positioning dimensions of the slotted structure decreases turbulence intensity in the internal flow field of the hydrocyclone, mitigating energy losses caused by turbulent states and facilitating pressure drop reduction.

In summary, the improvement of the hydrocyclone through the slotted design of the overflow pipe optimizes internal flow dynamics, reduces energy losses in each component, and significantly lowers the pressure drop and energy consumption. This study provides a strong theoretical basis for designing efficient and low-energy hydrocyclones.

The design involves the uniform distribution of 4 narrow slots along the circumferential direction of each layer, with each slot having a height of 2 mm. The inter-layer spacing is fixed at 6 mm. Concurrently, an optimization design is conducted for the number of slot layers, slot positioning dimensions, and slot angles. Figure  1 illustrates the schematic diagram of the conventional structure of the hydrocyclone, while Fig.  2 presents the schematic diagram of the cone overflow pipe with a slot structure.

Conventional schematic diagram of the hydrocyclone.

Schematic diagram of the cone overflow pipe with seam structure in the hydrocyclone. Note : The total height of the overflow pipe is 120mm, with a slotted design featuring four uniform slots per layer, each slot having a height of 2mm. The bottom inner diameter (φ) of the conical overflow pipe is 20mm, while the top outlet inner diameter (φ) is 28.8mm. The wall thickness of the overflow pipe is 5mm, with a layer spacing of 4mm.

By considering the variation characteristics of the flow field in the hydrocyclone, significant optimization results were achieved. The number of slot layers (n) for the cone overflow pipe was varied at 1 layer, 2 layers, and 3 layers, while the slot angle (θ) was set to 30°, 45°, 60° and 75°. The slot positioning dimension (a) was tested at 3 mm, 4 mm, 5 mm, and 6 mm. These parameters were systematically combined and organized with specific codes to comprehensively investigate the influence of slot structure parameters on the separation performance of the hydrocyclone. In the overflow pipe of a hydrocyclone separator, optimizing the design by increasing the number of slots, adjusting slot angles, and positioning dimensions effectively reduces the pressure drop of the hydrocyclone. Increasing the number of slots enlarges the open area of the overflow pipe, reducing fluid resistance as it passes through the pipeline. Additionally, this optimization helps to decrease local pressure at the bottom inlet of the overflow pipe and reduces dynamic pressure drop as fluid flows through the hydrocyclone. However, increasing the number of slots to 5 or 6 layers, while further increasing the open area of the overflow pipe, also introduces potential issues. Excessive layers may position the slots in the short-circuit flow area within the hydrocyclone, potentially causing coarser overflow and thereby impacting separation efficiency and performance. Therefore, in the design optimization process, it is crucial to balance the number and placement of slots to ensure improved efficiency of the hydrocyclone while mitigating potential adverse effects from excessive layering.

At the outset, distinct levels of the three variables, namely the number of slot layers, slot angle, and slot positioning dimension, were meticulously planned. Subsequently, an orthogonal experiment was carried out to investigate multiple combinations of these variables. For more detailed information, kindly refer to Tables 1 and 2 .

Experimental procedure and analysis

Experimental setup.

The experimental setup for the hydrocyclone mainly consists of a batching system including a stirrer, material tank, and a feed system comprising a centrifugal pump and material pipelines. The separation and testing system consist of various types of hydrocyclones and testing instruments. Under identical experimental conditions, separation experiments are conducted on different types of hydrocyclones. Overflow and underflow samples are collected three times and averaged to reduce experimental errors. Figure  3 illustrates the experimental equipment for the hydrocyclone separator, while Fig.  4 depicts the process flow diagram for the hydrocyclone separation experiments. In this study's evaluation of hydrocyclone performance, precision-engineered differential pressure sensors, specifically the Honeywell STD720-E1HC4AS-1-A-AHB-11S-A-10A0-F1-0000 model, were strategically installed at the hydrocyclone's inlet, overflow, and underflow points for meticulous pressure measurement. This strategic deployment facilitated the real-time surveillance of pressure shifts at pivotal junctures, enabling an accurate determination of the hydrocyclone's pressure differential. Rigorous calibration of each sensor ensured the reliability of the data captured. Employing high-frequency sampling, which exceeded ten instances per second, allowed for the documentation of transient pressure variations. Subsequent data analysis yielded the computation of the average pressure drop. To affirm the experiments' accuracy and reproducibility, each testing scenario was conducted in triplicate, bolstering the confidence in the outcomes and providing a robust dataset for hydrocyclone optimization efforts.In this study, the mixed fluid was extracted from the blending tank and delivered to the hydrocyclone feed inlet via a pump designed for handling flow rates ranging from 600 to 5000 ml per second. The pump's flow rate was precisely measured using an electromagnetic flow meter, ensuring accurate control and monitoring of the fluid dynamics processes within the hydrocyclone.Among them, in Fig.  4 , the 8-Centrifugal pump is used for liquid extraction, with a working flow rate ranging from 500 to 3500 ml/s.

Diagram of experimental apparatus.

The process flowchart of the hydrocyclone separation experiment.

Experimental method

The experiment utilized a mixture of 1% mass concentration of glass bead fine powder and water. The median particle size of the glass beads was measured as 41.52 μm using an Eyetech laser particle size analyzer. The true density of the glass beads was determined to be \(2.6\text{ g}/{\text{cm}}^{3}\) . Figure  5 presents the particle size distribution of the glass bead experimental raw material.

The particle size distribution chart of the glass bead experimental raw material.

To collect samples from the overflow and underflow outlets, the mixture was filtered and weighed. Subsequently, the collected samples were subjected to filtration, extraction, drying, and weighing processes.

During the experimental process, the overflow and underflow flow rates were measured using electromagnetic flowmeters. The inlet and outlet pressures were measured using pressure gauges, and the pressure drop across the hydrocyclone was calculated based on Eq. ( 1 ). The mass of the glass bead samples after drying was weighed, and the separation efficiency of the hydrocyclone was calculated using Eq. ( 2 ).

Pressure Drop Calculation Formula:

In the equation, \({\text{P}}_{\text{in}}\) represents the inlet pressure of the hydrocyclone, and \({\text{P}}_{\text{out}}\) represents the overflow outlet pressure of the hydrocyclone.

The efficiency calculation formula is as follows:

In the equation, \({\text{C}}_{\text{u}}\) represents the concentration before separation (the inlet concentration into the hydrocyclone); \({\text{C}}_{\text{o}}\) represents the concentration of the overflow material (the output of the hydrocyclone); and \({\text{C}}_{\text{f}}\) represents the concentration of the underflow waste material.

Numerical calculation method

Calculation model and grid generation.

Numerical simulations were conducted to study the internal flow of the hydrocyclone, and the computational domain was established. Firstly, three-dimensional models of the three types of hydrocyclones were constructed using SolidWorks software. Subsequently, the constructed three-dimensional models were imported into CFD mesh software for grid generation.

To better represent the fluid motion, a tetrahedral structured grid was used as the fluid domain model for the hydrocyclone. During the grid generation process, refinement was applied to regions such as the tangential inlet of the hydrocyclone to capture the flow characteristics more accurately. Grid independence tests were also performed to reduce the influence of grid quantity on the numerical simulation results. Taking Type A conventional hydrocyclone as an example, since the fluid domain models had the same diameter and length before and after improvement, different grid numbers (approximately 200,000, 400,000, 600,000, and 900,000) were used for numerical simulation. In numerical simulations of fluid flow, maintaining an aspect ratio of the grid within a moderate range is crucial for optimizing the balance between simulation accuracy and computational efficiency. This strategy not only ensures the precision of simulation outcomes and the stability of the computational process but also aids in managing the consumption of computational resources. In this simulation, the grid aspect ratio was set at 2.8. Such a selection allows for the accurate capture of fluid dynamics within the hydrocyclone, including velocity profiles, pressure fields, and the trajectories of solid particles, while avoiding the computational instability and unnecessary cost increases associated with higher aspect ratios.

Moreover, particular attention was devoted to the optimization of near-wall grid refinement in simulations to adjust wall shear stress (Y +) values, a critical aspect for ensuring simulation accuracy. The correct Y + values are imperative for selecting turbulence models and wall treatment strategies, as they accurately depict the flow characteristics within the boundary layer. This approach enables precise identification of flow separation and reattachment points. Through meticulously designed grids and suitable simulation strategies, this measure not only guarantees the quality of simulations but also enhances computational efficiency, providing reliable data support for the design and optimization of hydrocyclones.

Through these numerical simulations, the influence of different grid quantities on the simulation results was evaluated, and an appropriate grid number was determined to obtain accurate and reliable simulation results. This exploration is crucial for further analyzing the performance of the hydrocyclone and the effects of improvements.

Numerical calculation method and boundary conditions

ANSYS Fluent software was used to conduct numerical simulations for different types of hydrocyclones. For the simulation, the Reynolds Stress Model (RSM) was chosen as the turbulence model for the fluid in the hydrocyclone, and standard wall functions were adopted 29 . The Reynolds Stress Model adequately accounts for the stress tensor induced by fluid rotation and is particularly suitable for high-intensity turbulent flow, making it a suitable option in this study.

The Volume of Fluid (VOF) model was employed for multiphase flow simulations. The VOF model can be used to simulate the interface between two or more immiscible fluids and track the movement of the phase interface by solving the continuity equation. The Volume of Fluid (VOF) model is principally utilized for capturing the dynamics between the liquid and air phases within the hydrocyclone, notably including the formation of the air core. The simulated fluid does not include glass particles.This method enables the simulation and thorough analysis of intricate flow phenomena inside the hydrocyclone, such as the efficiency of solid–liquid separation and the pressure drop. In parallel, the experimental component assessed the separation performance of glass particles, with these observations being integrated with numerical simulation outcomes to refine the hydrocyclone's design.

This study meticulously investigates the fluid dynamics within hydrocyclones, focusing primarily on the interaction between water and air, and the pivotal role of air core formation in influencing hydrocyclone performance. Acknowledging the core objective to unravel the intricacies of liquid–gas interactions on hydrocyclone efficiency, and given the minimal concentration of solid particles, it is argued that while particles do exert an influence on separation efficacy, their effect is marginal relative to the principal phenomena of interest—flow dynamics and air core genesis. Consequently, the disturbance effects of particulate matter on fluid flow are considered negligible for the scope of this investigation. This targeted approach allows for a nuanced exploration of the interaction between water and air, facilitating a more refined analysis of their collective impact on the hydrocyclone's internal flow field.

The genesis of the air core is ascribed to the negative pressure generated by the fluid's rotational movement within the hydrocyclone, compelling air to be drawn into the vortex. This fluid dynamic-induced negative pressure zone is identified as the direct catalyst for air core formation, critically influencing the hydrocyclone's separation efficiency and flow characteristics. Through a focused examination of water–air interactions, this research endeavors to enhance the understanding of hydrocyclone operational mechanisms, specifically analyzing the air core's effect on performance.In the simulation of the hydrocyclone, the main phase was set as the mixture liquid, with a constant temperature, density of \({998.2\text{kg}/\text{m}}^{3}\) , and viscosity of \(0.001\text{Pa}\bullet \text{s}\) . The air phase was considered as the second phase, with a density of \({1.293\text{kg}/\text{m}}^{3}\) and viscosity at room temperature. The overflow and underflow outlets were set as pressure outlets, and the air backflow rate was set to 1.

In this study, the initial stage of the calculation used a mixture liquid calculation, and after convergence, it transitioned to two-phase calculation. The implicit transient pressure–velocity coupling method used the SIMPLEC method. To ensure computational stability, the pressure gradient was computed using the Green-Gauss Cell-Based method, the pressure discretization used the PRESTO! method, the momentum discretization used the Second Order Upwind method, and the turbulent kinetic energy and turbulent kinetic energy dissipation rate used the first-order upwind scheme. The convergence criterion was set at a residual tolerance of 1e-5, and the balance of mass flow rates at the inlet and outlet phases was used as the criterion for convergence judgment. In this simulation, the results were subject to temporal averaging to ensure they accurately reflect the mean state of the flow within the hydrocyclone. Three complete flow cycles were selected for the temporal averaging process, guaranteeing the precision and representativeness of the outcomes.

The validation process of CFD simulation credibility

In this investigation, a sequence of meticulous validation procedures was conducted to affirm the robustness and fidelity of the computational fluid dynamics (CFD) simulations. Figure  6 depicts the diagram of different cross-sectional positions of the hydrocyclone. The inaugural phase entailed a grid independence verification (refer to Fig.  7 ), aiming to ascertain the sensitivity of the results to the computational cell size. Through systematic refinement of the mesh and scrutiny of solution convergence, spatial resolution was confirmed as adequate to capture the flow dynamics with precision. The superposition of velocity profile curves across varying mesh densities indicates that additional refinement does not significantly modify the outcomes, thereby asserting grid independence. For computational efficiency, the mesh count was selected in the order of 600,000 cells.

Hydrocyclone cross-sectional position.

Mesh independence verification.

Upon establishing grid independence, a time-step independence verification was executed (refer to Fig.  8 ), ensuring the temporal discretization was sufficiently detailed to capture essential time-dependent characteristics of the fluid flow. The consistency of simulation results across varying time steps, paired with negligible variations in the velocity profiles at a time step of 1e-5, suggests that the simulation has attained a quasi-steady state, exhibiting insensitivity to further reduction in the time step. In this study, the selection of the time step adheres to the Courant-Friedrichs-Lewy (CFL) condition to ensure the numerical stability of Computational Fluid Dynamics (CFD) simulations. The CFL condition, a critical criterion, guarantees that the distance a fluid particle travels within a time step does not exceed the size of a computational cell 30 . Through preliminary simulations, the impact of various time steps on the outcomes was assessed, and the time step was adjusted to maintain the CFL number within a range of less than or equal to 1. This procedure ensures the accuracy and stability of the simulations.

Time-step independence verification for hydrocyclone simulations.

Conclusively, to solidify the accuracy of the simulations, a numerical simulation accuracy test was performed (refer to Fig.  9 ). This entailed juxtaposing simulation outputs with experimental data. The high congruence between simulated axial velocity profiles and experimental observations substantiates the numerical model's precision, especially in predicting peak velocities pivotal to the hydrocyclone's performance.

Model accuracy validation through comparison with experimental data.

To comprehensively elucidate the computational approach adopted in the investigation of hydrocyclone separator performance, Table 3 consolidates the pivotal simulation parameters employed within the study.

Overflow pipe slotted structure optimization

Impact of overflow pipe slotted structure on hydrocyclone separation performance.

In this study, solid–liquid separation experiments were conducted for the hydrocyclone. Firstly, based on the desired feed concentration and separation target, the concentration of the mixture liquid was adjusted to obtain a glass bead fine particle mass concentration of 1%. Subsequently, the mixture liquid was adequately covered by the stirrer, and the motor was adjusted to start the stirrer, initiating the mixing of the material and water.

Simultaneously, the centrifugal pump's rotational speed was controlled to achieve the experimentally preset initial reading of the electromagnetic flowmeter, which was set at an initial flow rate of 680 ml/s. During the experimental stage, after the mixture liquid was fully and uniformly mixed under the action of the stirrer, and the flow rates at the overflow and underflow outlets of the hydrocyclone stabilized, the beakers were quickly placed at the overflow and underflow outlets for sampling.

The collected samples were subjected to drying, and the dried samples were weighed using a precise balance. The mass data of the samples obtained from the experiment were recorded. Specifically, in the experiment, weighing equipment (as shown in Fig.  10 ) was used to ensure the accurate weighing of the samples, ensuring the accuracy and reliability of the data. The experimental protocol followed the established procedure of drying the specimens at 105 degrees Celsius for around 24 h. This method was employed to remove all moisture from the samples, guaranteeing that the weight measurements accurately represent the dry mass of the specimens collected.

The equipment diagram for accurately weighing the experimental samples of hydrocyclone separation efficiency.

The separation performance of the hydrocyclone with a single-layer slotted conical overflow pipe Type B hydrocyclone and the conventional Type A hydrocyclone under equivalent operating conditions is illustrated in Fig.  11 . The graph depicts the influence of different inlet flow rates on the separation efficiency (η) and pressure drop (ΔP) for both types of hydrocyclones. The x-axis represents the hydrocyclone inlet flow rate (Q), the left y-axis represents the separation efficiency (η) of the hydrocyclone, and the right y-axis represents the pressure drop (ΔP) across the hydrocyclone.

Flow rate-efficiency pressure drop relationship chart.

When the inlet flow rate is the same, the improved Type B hydrocyclone shows a slight decrease in separation efficiency compared to the conventional Type A hydrocyclone. However, it also achieves a certain degree of pressure drop reduction, resulting in energy-saving benefits. Under the operating conditions with inlet flow rates ranging from 680 to 920 ml/s, the improved Type B hydrocyclone exhibits a relatively small reduction in pressure drop. However, when the inlet flow rate exceeds 780 ml/s, the pressure drop reduction of the Type B hydrocyclone gradually increases, reaching its maximum at 860 ml/s. Compared to the conventional Type A hydrocyclone, the Type B hydrocyclone shows a pressure drop reduction of 6.8 units. The pressure drop for the conventional Type A hydrocyclone is 42.04 kPa, while it is 39.18 kPa for the Type B hydrocyclone.

Furthermore, after the slotted modification, the separation efficiency of the improved Type B hydrocyclone is slightly lower than that of the conventional hydrocyclone. When the inlet flow rate is greater than 760 ml/s, the separation efficiency of the Type B hydrocyclone approaches that of the conventional Type A hydrocyclone. At an inlet flow rate of 880 ml/s, the separation efficiency of the conventional Type A hydrocyclone is 97.96%, while the Type B hydrocyclone achieves a separation efficiency of 97.62%. Compared to the conventional Type A hydrocyclone, the separation efficiency of the Type B hydrocyclone decreases by 0.35 percentage points. Moreover, with the increase in inlet flow rate, the separation efficiency of the Type B hydrocyclone gradually approaches that of the conventional Type A hydrocyclone, while the pressure drop reduction increases.

Based on the experimental data, it can be observed that compared to the conventional Type A hydrocyclone, the slotted conical overflow pipe structure has a relatively minor impact on separation efficiency as the inlet flow rate increases. However, it has a significant effect on pressure drop reduction. The slots act as fluid passages, increasing the outlet area of the overflow pipe, reducing the axial velocity of the fluid inside the hydrocyclone, and thereby reducing the kinetic energy loss and pressure drop.

Optimization of slotted layer number

In order to further reduce the energy consumption of the Type B hydrocyclone, an optimization design of the slotted layer number was conducted. The slotted layer number was set from 1 to 4, with a layer spacing of 6 mm, slot angle of \(30^\circ \) , and slot position size of 3 mm. These were designated as Type B to Type E, and separation experiments were carried out for each design. The relationship curves between different slotted layer numbers, inlet flow rates, and the hydrocyclone's separation efficiency and pressure drop are shown in Fig.  12 .

Inlet flow rate—separation efficiency and pressure drop curves under different numbers of seams.

The separation efficiency of the five types of hydrocyclones is positively correlated with the inlet flow rate. With an increase in the number of slots, the overall trend of the separation efficiency in Type B to Type E hydrocyclones gradually decreases. Among them, Type B to Type D hydrocyclones (with 1–3 layers of slots) exhibit a slow decline in separation efficiency, with a small reduction. The Type E hydrocyclone (with 4 layers of slots) shows a relatively larger decrease in separation efficiency because the increased number of slots elevates the slot position, causing short-circuit flow in the overflow pipe region, leading to the entrainment of solid particles from the slots into the overflow pipe, thereby increasing the separation efficiency reduction.

Regarding the pressure drop, as the inlet flow rate increases, all five types of hydrocyclones show a gradual upward trend in pressure drop. With an increase in the number of slots, compared to the conventional Type A hydrocyclone, the pressure drop reduction in Type B to Type E hydrocyclones gradually increases. Type B and Type C hydrocyclones (with 1 to 2 layers of slots) experience minor changes in pressure drop reduction, while Type D and Type E hydrocyclones (with 3 to 4 layers of slots) demonstrate a significant increase in pressure drop reduction. The increase in the number of slots results in a larger slot area, which increases the flow rate entering the overflow pipe, reduces the local pressure at the bottom inlet of the overflow pipe, decreases the overall dynamic pressure of the internal swirling flow in the overflow pipe, and increases the outlet static pressure of the overflow pipe. According to fluid dynamics principles, the change in velocity has a significant impact on fluid kinetic energy, which is a key reason for the significant reduction in pressure drop after slot modification. Based on the analysis above, Type D hydrocyclone exhibits a remarkable pressure drop reduction while maintaining almost the same separation efficiency.

During the actual experimental process, at an inlet flow rate of 680 ml/s, the Type D hydrocyclone achieved a separation efficiency of 90.6% with a pressure drop of 36.31 kPa. Compared to the conventional Type A hydrocyclone, the separation efficiency of the Type D hydrocyclone decreased by 3.04%, and the pressure drop decreased by 1.83%.As the inlet flow rate reached the working condition of 900 ml/s, the Type D hydrocyclone showed a turning point in separation efficiency, reaching its maximum value. At this point, the separation efficiency and pressure drop for the conventional Type A hydrocyclone were 97.69% and 43.34 kPa, respectively, while for the Type D hydrocyclone, they were 97.53% and 38.65 kPa, respectively. Compared to the conventional Type A hydrocyclone, the separation efficiency of the Type D hydrocyclone decreased by 0.16%, and the pressure drop decreased significantly by 10.28%. These results indicate that the Type D hydrocyclone is more suitable for separation operations under high inlet flow rate conditions.

Optimization of slot position and angle

The different slot positions in the overflow pipe will have a certain impact on the separation efficiency and pressure drop of the hydrocyclone. An experiment was conducted to explore the effect of slot positions on the Type D hydrocyclone. The slot size "a" was set to 4 mm, 5 mm, and 6 mm, corresponding to Type T, Type Jj, and Type Zz, respectively. Figure  13 shows the flow rate-separation efficiency and flow rate-pressure drop curves for different types of hydrocyclones under inlet flow rates ranging from 680 to 920 ml/s.

Inlet flow rate—pressure drop curves at various seam positions.

At an inlet flow rate of 680 ml/s, the separation efficiency of the Type Jj hydrocyclone is 90.72%, with a pressure drop of 26.0 kPa. Compared to the conventional Type A hydrocyclone, the separation efficiency of the Type Jj hydrocyclone decreases by 1.91%, and the pressure drop decreases by 2.99%.

When the inlet flow rate reaches the working condition of 900 ml/s, the Type Jj hydrocyclone achieves its highest separation efficiency at 97.84%, with a pressure drop of 37.87 kPa. Compared to the conventional Type A hydrocyclone, the separation efficiency of the Type Jj hydrocyclone increases by 0.15%, and the pressure drop decreases by 12.62%.Regarding the other three types of hydrocyclones with different slot positions, the relationship between efficiency, pressure drop, and slot position changes is not very pronounced. However, for the Type Zz hydrocyclone, a relatively significant decrease in separation efficiency is observed. This is because the top slot position is close to the short-circuit flow region, allowing some particles to enter the overflow pipe through the slots along with the fluid motion, resulting in a reduction in the hydrocyclone's separation efficiency. On the other hand, the variation in the slot position below the short-circuit flow has little effect on the hydrocyclone's separation performance.

To achieve continuous analysis of different levels of various factors within the experimental conditions and obtain a more accurate optimal solution, the response surface optimization method was utilized. In this approach, the inlet flow rate (Q) and the slot size (a) were selected as the influencing factors. The ranges of these two factors were determined, and the experimental data corresponding to these two factors' levels were input into the Design-Expert design software. By employing central composite design, specific values for the three levels of each factor were obtained (as shown in Table 4 ). The three levels are lower limit, center point, and upper limit, respectively.

Regarding the experimental data, a response surface optimization design method was employed to conduct multivariate regression analysis. The experimental data was input into the Design-Expert software to establish the quadratic polynomial response surface regression equations for the target functions, separation efficiency ( \({\text{Y}}_{\text{e}}\) ) and pressure drop ( \({\text{Y}}_{\text{p}}\) ), with respect to the variables X1 and X2, as shown in Eqs. ( 3 ) and ( 4 ):

Figure  14 a,b illustrate the interaction effects of inlet flow rate and orifice size on the objective functions \({\text{Y}}_{\text{e}}\) and \({\text{Y}}_{\text{p}}\) . With other parameters kept constant, an increase in the inlet flow rate leads to higher pressure drop and separation efficiency. In this simulation, while maintaining the other dimensions of the hydrocyclone unchanged, increasing the orifice size initially enhances the separation efficiency but then causes a decrease, and the pressure drop shows a decreasing trend followed by an increasing trend. When the orifice size is set to 5.3 mm, a better balance between separation efficiency and pressure drop can be achieved.

The influence of flow rate and positioning dimension on separation performance.

To investigate the influence of orifice angle on the separation efficiency and pressure drop of the hydrocyclone, four different angles, namely \(30^\circ \) , \(45^\circ \) , \(60^\circ \) , and \(75^\circ \) , were designed, corresponding to the models Type Jj, Type Nn, Type Rr, and Type Vv, respectively. These models were compared with the conventional Type A hydrocyclone under the same inlet flow rate condition. The flow rate-separation efficiency and pressure drop curves of the five hydrocyclone models are shown in Fig.  15 .

Inlet flow rate-efficiency pressure drop curves at different seam angles.

Type Jj, Type Nn, and Type Rr hydrocyclones exhibit similar separation efficiencies, while Type Vv hydrocyclone experiences a more significant decrease in separation efficiency.The pressure drop reduction follows the order from the largest to the smallest: Type Vv, Type Rr, Type Nn, and Type Jj hydrocyclones.

As the orifice angle increases, the overflow flow rate gradually increases, leading to a decrease in the kinetic energy loss of the internal fluid. When solid particles are carried into the orifice, they need to change direction to enter the overflow pipe. Part of the particles experiences inertial impact with the pipe wall and undergo secondary separation. With the increase in orifice angle, the fraction of particles being impacted and re-separated decreases gradually, which significantly reduces the separation efficiency of the hydrocyclone. Among them, the Type Rr hydrocyclone experiences a substantial decrease in pressure drop while maintaining the separation efficiency nearly constant.

At an inlet flow rate of 900 ml/s, the Type Rr hydrocyclone achieves the highest separation efficiency of 97.75% and a pressure drop of 31.56 kPa. Compared to the conventional Type A hydrocyclone, the separation efficiency increased by 0.06%, and the pressure drop decreased by 24.85%.

Figure  16 a,b represent the interaction between inlet flow rate and orifice angle on the objective functions \({\text{Y}}_{\text{e}}\) and \({\text{Y}}_{\text{p}}\) , respectively. When other parameters remain constant, an increase in the inlet flow rate leads to a rise in both separation efficiency and pressure drop. In this simulation, with the hydrocyclone's other dimensions unchanged, increasing the orifice angle initially enhances the separation efficiency and subsequently decreases it, while the pressure drop exhibits a gradual decline. An orifice angle of \(58^\circ \) appears to strike a balance between separation efficiency and pressure drop, providing better performance for the hydrocyclone.

Influence of multiple factors on separation performance.

To further investigate the optimization scheme with three orifice layers, a 5.3 mm orifice size, and a \(58^\circ \) orifice angle, experimental research is conducted with an initial inlet flow rate of 800 ml/s. The results are compared with the conventional Type A hydrocyclone, as shown in Fig.  17 , illustrating the contrast in pressure drop and separation efficiency. The efficiency-related data were meticulously compiled and analyzed using SPSS Statistics 22 software, employing a one-way ANOVA to conduct significance tests with Student's t-test at a P < 0.05 significance level. Graphical representation was created using Origin 2021.

The separation efficiency and pressure drop of the hydrocyclone before and after optimization.

Figure  18 illustrates the comparison of particle size efficiency between the optimized and conventional hydrocyclones at an inlet flow rate of 900 ml/s. Based on the results from Fig.  17 and the comparative chart in Fig.  18 , it can be concluded that within the range of inlet flow rates from 900 to 920 ml/s, the optimized hydrocyclone exhibits higher separation efficiency compared to the conventional type. However, as the inlet flow rate increases, the improvement in separation efficiency gradually diminishes, while the pressure drop also increases. At an inlet flow rate of 900 ml/s, the optimized hydrocyclone achieves the highest separation efficiency, reaching 97.77%, representing a 0.26% improvement compared to the conventional hydrocyclone. The corresponding pressure drop is 32.98 kPa, resulting in a reduction of 24.88%.Within the particle size range larger than 30 µm, the optimized hydrocyclone's particle size efficiency remains essentially unchanged compared to the conventional hydrocyclone.

Comparison of particle efficiency before and after optimization in hydrocyclone.

These results indicate that the optimized hydrocyclone can achieve higher separation efficiency and relatively smaller pressure drop within a certain range of inlet flow rates. This is of great significance for improving the hydrocyclone's performance and efficiency.

Numerical simulation analysis

Numerical simulation analysis is conducted on the optimized hydrocyclone, referred to as Type I, with three orifice layers, an orifice size of 5.3 mm, and an orifice angle of \(58^\circ \) . Numerical simulations are performed at an inlet flow rate of 900 ml/s and compared with the conventional Type A hydrocyclone. By comparing the two hydrocyclones in terms of fluid axial velocity, tangential velocity, pressure distribution, and other aspects, this numerical simulation analysis provides deeper insights into the improvement achieved by Type I hydrocyclone, thereby serving as a reference for further research and optimization.

Grid independence and numerical method validation

By examining the average tangential velocity at different sections of the hydrocyclone, it was observed that the average tangential velocity remained relatively constant when the grid size increased to approximately 600,000 cells. To validate the numerical simulation of the Type A hydrocyclone, the tangential velocities at various cross-sections were compared with experimental values. The results from the numerical simulations were found to be in close agreement with the experimental values, indicating that the numerical model used in this study can reasonably predict the solid liquid separation performance of the hydrocyclone. Therefore, the grids for Type A and Type I hydrocyclones were set to similar orders of magnitude, with 643,541 and 674,512 cells, respectively.

Pressure analysis

Based on the pressure distribution analysis, it was observed that as both types of hydrocyclones approached the center radially, the pressure gradually decreased, forming negative pressure regions. Figures  19 and 20 illustrate the pressure distribution at different cross-sectional positions. Compared to the Type A hydrocyclone, the modified hydrocyclone exhibited significantly reduced overall pressure, with an increased diameter of the air column and a noticeable decrease in pressure drop along the column. This indicates that the modified overflow pipe had a significant impact on the pressure distribution along the hydrocyclone column. The improved overflow pipe possessed a larger equivalent diameter, resulting in increased fluid discharge within the overflow pipe, thereby reducing the internal pressure of the hydrocyclone.

Pressure contour maps of hydrocyclones with different cross-sectional designs before and after improvement.

Before and after improvement, axial cross-sectional pressure contour maps of the hydrocyclone.

Based on the pressure distribution curves at different axial cross-sections in the hydrocyclone, as shown in Fig.  21 , it can be observed that the overall pressure trend exhibits an approximate "V" shape, and the negative pressure region at the axis of both hydrocyclones shows similar pressure values. The pressure is positively correlated with the radial position. Compared to the Type A hydrocyclone, the improved Type I hydrocyclone shows a gentler pressure curve in the external region of the overflow pipe, resulting in a significant overall pressure reduction.

Pressure distribution curves in different axial sections of the hydrocyclone before and after improvement.

Furthermore, the pressure of both hydrocyclone types is negatively correlated with the axial position. Specifically, in the axial positions ranging from the Y = − 0.015 m cross-section to the Y = − 0.04 m cross-section, the pressure variation in the Type I hydrocyclone is greater than that in the Type A hydrocyclone. Additionally, the pressure at the column cross-section located at Y = 0.01 m is higher than the pressure at the overflow pipe cross-section. The improved design of the overflow pipe in the Type I hydrocyclone reduces internal frictional resistance, leading to a notably lower pressure at the overflow pipe cross-section compared to the column cross-section. However, the Type I hydrocyclone adopts a tapered slotted design, resulting in a rapid increase in fluid velocity as it enters the overflow pipe, leading to localized turbulence and increased energy loss. As a consequence, the Type I hydrocyclone exhibits a slightly higher pressure drop compared to the TypeA hydrocyclone.

In summary, the optimization of the hydrocyclone's overflow pipe design in the Type I hydrocyclone reduces the overall pressure and improves the pressure distribution compared to the conventional Type A hydrocyclone.However, due to the introduction of the tapered slotted structure, the Type I hydrocyclone experiences a slightly higher pressure drop, indicating a trade-off between pressure reduction and energy loss in the design optimization.

The changes in the internal pressure distribution of the hydrocyclone before and after the optimization of the slotted structure are jointly presented in Figs. 19, 20 and 21.The results demonstrate that the pressure distribution of the optimized hydrocyclone is more reasonable and symmetrical in multiple cross-sections and axial profiles compared to the original hydrocyclone, and the pressure level is noticeably reduced. Specifically, the slotted structure leads to a reduction in pressure in the region near the outlet, a gradual decrease in the axial pressure gradient, and an overall pressure reduction across the hydrocyclone. The combined information from the three figures indicates that the introduction of the slotted structure significantly improves the internal pressure distribution of the hydrocyclone, which explains the observed phenomenon of reduced pressure drop from the perspective of the flow field. Therefore, the regulatory effect of the slotted structure on the internal pressure field is one of the key reasons for achieving the optimization of the hydrocyclone's performance.

Axial velocity analysis

In the analysis of axial velocity, detailed distribution simulations of the axial velocity were conducted at axial cross-section positions (Y = 0.04 m and 0.08 m) for both hydrocyclone types, and the results are presented in Fig.  22 .By observing the axial velocity distribution of the two hydrocyclone types, it can be seen that the velocity gradually increases from the wall to the axis and sharply rises to its maximum value in the central region, presenting a generally symmetrical profile.

Comparison of axial velocity distribution before and after improvement in the hydrocyclone.

The improved symmetry in the pressure and velocity distributions of the optimized hydrocyclone compared to the original hydrocyclone confirms the effectiveness of the slotted structure optimization in achieving a more balanced and stable flow field inside the hydrocyclone. The changes in pressure and velocity distributions provide valuable insights into the flow behavior, contributing to the understanding of the improved separation performance and reduced pressure drop observed in the experimental results.

It is noteworthy that, compared to the Type A hydrocyclone, the Type I hydrocyclone exhibits a slight decrease in its axial velocity. In the Type I hydrocyclone, the reduction in axial velocity is more pronounced in the inner swirling region than in the outer swirling region. The optimized hydrocyclone with overflow slits shows a significant decrease in axial velocity in the inner swirling region near the overflow outlet. Specifically, at the Y = 0.04 m section, the maximum axial velocity of the prototype hydrocyclone is approximately 3.2 m/s, while the optimized version only reaches 2.8 m/s. Similarly, at the Y = 0.08 m section, the maximum axial velocity decreases from 2.9 to 2.6 m/s. This reduction in axial velocity is attributed to the enlargement of the outlet area by the overflow slits, which weakens the intensity of the inner swirling vortex flow, leading to a decrease in the axial velocity of the vortex flow.

The increase in the number of overflow slits will further expand the outlet area and cause a further decrease in the axial velocity of the inner swirling flow. However, excessive slit numbers may lead to a saturation effect. Additionally, the opening angle of the slits affects the outlet flow rate, where too large an angle can result in excessively low axial velocities. On the other hand, the height of the slit controls its range of influence and directly determines the distribution pattern of the axial velocity field.

Figure  23 provides a visual representation of the X-direction velocity (axial velocity component) distribution in the axial section of the two hydrocyclones. From the figure, it is evident that the optimized hydrocyclone with overflow slits exhibits a more uniform and symmetric axial velocity distribution within its interior, especially in the region near the overflow outlet, where the velocity field distribution appears more reasonable. Specifically, after the slit optimization, the maximum axial velocity near the overflow outlet reduces significantly from the original 3.2–2.8 m/s. This indicates that the introduction of the overflow slits weakens the intensity of the vortex flow in the overflow tube region, leading to a notable reduction in the axial velocity component.

Comparison of axial section x velocity distribution before and after improvement in the hydrocyclone.

The Type I hydrocyclone can effectively control the distribution of axial velocity to match the tangential velocity field, thereby achieving the goal of improving the hydrocyclone's separation efficiency. The axial velocity distribution plays a crucial role in optimizing the hydrocyclone's performance.

In addition, after introducing the overflow slit in the hydrocyclone, the axial velocity of the outer swirling region near the hydrocyclone wall shows a slight decrease, although this effect is relatively minor. However, as the radial position moves towards the axis, the axial velocity in the inner swirling region experiences a significant reduction, with the impact of the overflow slit becoming more pronounced. This phenomenon can be explained by the fact that, under the same inlet flow conditions, the overflow slit structure enlarges the equivalent diameter of the overflow outlet. As a result, the rotational speed of the fluid around the central axis decreases, causing the zero-velocity envelope surface to move inward. This process increases the time for medium and large particles in the outer swirling region to participate in the separation, resulting in a more thorough separation effect. Additionally, the overflow slit structure also reduces the likelihood of coarse particles in the outer swirling region re-entering the inner swirling flow. Therefore, the influence of the overflow slit on hydrocyclone performance is mainly manifested in the reduction of axial velocity in the inner swirling region and the enhancement of solid–liquid separation efficiency. The optimized combination of the overflow slit parameters in Type I hydrocyclone satisfies the separation requirements of the axial velocity field, thereby improving the overall separation performance of the hydrocyclone.

Tangential velocity analysis

In this study, the tangential velocity of the fluid in the hydrocyclone with an inlet flow rate of 900 ml/s was analyzed. The comparison of the tangential velocity distribution curves at different cross-sectional positions for both hydrocyclone types is shown in Fig.  24 .Overall, the tangential velocity distribution curve exhibits an "S"-shaped trend. As the distance from the hydrocyclone wall decreases, the tangential velocity increases with decreasing radius. It reaches its maximum value near the hydrocyclone wall and then gradually decreases with further reduction in radius. When approaching the vicinity of the air core, the tangential velocity drops sharply, eventually becoming zero at the central axis.

Velocity distribution curves of different axial sections in the hydrocyclone before and after improvement.

The design of overflow slit in the hydrocyclone reduces the internal fluid velocity, causing small-sized solid particles to lack sufficient centrifugal force to enter the outer swirling region for separation. Instead, they are eventually discharged through the overflow outlet, leading to a decrease in the hydrocyclone's particle size efficiency for small particles. However, large-sized particles, due to their larger volume and mass, can still overcome the reduced centrifugal force and enter the outer swirling region, thus their particle size efficiency remains unaffected. Compared to Type A hydrocyclone, the overall tangential velocity in Type I hydrocyclone slightly decreases, resulting in a reduction of the centrifugal force experienced by solid particles.

Additionally, when observing the tangential velocity above the overflow slit (Y = − 0.04 m) in Fig.  25 , it is evident that the decrease in tangential velocity above the overflow slit is more significant compared to the cylinder and cone sections, with the cone section experiencing a larger reduction than the cylinder section. This phenomenon is attributed to the greater influence of diameter size on the tangential velocity, and the impact of the overflow slit structure becomes more pronounced above the overflow slit level.

Comparison of tangential velocity distribution at the upper section of the overflow pipe.

As a result, the overflow slit design in the hydrocyclone has selective effects on particle size efficiency. It reduces the separation efficiency for small-sized particles due to reduced centrifugal force, while having limited impact on the efficiency of large-sized particles. Moreover, the influence of the overflow slit structure on tangential velocity is more evident above the overflow slit level, especially in the cone section.

Based on the combined analysis of the axial velocity distribution in Fig.  24 and the tangential velocity distribution in Fig.  25 at different axial cross-sections, it is evident that the Type I hydrocyclone, after optimization with the slotted structure, exhibits a more symmetrical and stable tangential velocity distribution compared to the Type A hydrocyclone. Specifically, at multiple cross-sections in Fig.  24 , the tangential velocity near the hydrocyclone wall is reduced by 0.2–0.4 m/s in the optimized hydrocyclone compared to the Type A hydrocyclone, and the negative tangential velocity in the central region is also decreased. In Fig.  25 , the tangential velocity distribution above the slotted structure shows an overall reduction of 0.3–0.5 m/s, with a smaller slope in the curve. This indicates that the introduction of the slotted structure weakens the internal vortex, resulting in a decrease in the tangential velocity component. Moderating the tangential velocity can contribute to achieving a more stable separation performance. Therefore, the regulation of the tangential velocity field through the slotted structure is one of the significant factors in optimizing the hydrocyclone's performance.

Furthermore, the proportion between axial and tangential velocities directly influences the hydrocyclone's separation efficiency. According to the above analysis, the velocity matching between the two components needs to be adjusted according to the particle size of different materials. For fine or low-density particles, increasing the axial velocity is necessary to rapidly remove them from the hydrocyclone wall and prevent excessive fine particles from entering the underflow. At the same time, providing a higher tangential velocity allows light particles to obtain sufficient centrifugal force to enter the overflow outlet. For coarse or high-density particles, reducing the axial velocity appropriately can increase their residence time inside the hydrocyclone for adequate separation. The tangential velocity can also be adjusted accordingly to reduce turbulence losses inside the hydrocyclone. For materials with a wide particle size distribution, a moderate combination of axial and tangential velocities should be chosen to achieve good separation performance for particles of different sizes. The axial velocity should not be too high or too low, and the tangential velocity needs to be controlled within an appropriate range. By adjusting the proportion between these two velocities when the operating conditions change, customized separation of materials can be achieved, thus expanding the hydrocyclone's applicability range.

The comprehensive experimental study with multiple factors reveals that the interaction of overflow slit design parameters, including positioning size, number of slits, and angle, significantly affects the separation performance of the hydrocyclone under identical operating conditions.

The number of overflow slits has a considerable impact on the pressure drop of the hydrocyclone. As the number of slits increases, the pressure drop also gradually increases. However, this is accompanied by a decrease in the hydrocyclone's separation efficiency. After optimizing the number of slits to three layers, a better compromise between separation efficiency and pressure drop is achieved.

Changing the positioning size of the overflow slits has a minor effect on the separation performance of the hydrocyclone. Excessively increasing the positioning size can lead to a sharp decrease in the separation efficiency. The positioning size of 5.3 mm provides a good balance between separation efficiency and pressure drop.

Altering the angle of the overflow slits has a significant impact on the hydrocyclone's separation performance. An excessively large angle causes a drastic reduction in separation efficiency. At an inlet flow rate of 900 ml/s, compared to the conventional hydrocyclone, the hydrocyclone with three layers of slits, a positioning size of 5.3 mm, and an angle of \(58^\circ \) exhibits an increase in separation efficiency of 0.26% and a substantial reduction in pressure drop, reaching 24.88%. This demonstrates that the optimized design of the conical overflow slits enables the hydrocyclone to maintain its separation efficiency under high inlet flow conditions while significantly reducing pressure drop. This results in remarkable energy savings and achieves the goal of optimized design, providing valuable reference for the development of new hydrocyclones.

The findings from this study provide essential insights into the impact of overflow slit design on the performance of hydrocyclones, offering valuable guidance for the development and optimization of hydrocyclone separators.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Ding, J. et al. Research progress on the application of hydrocyclone separators in water treatment. Environ. Eng. 39 (08), 1–6 (2021) ( (in Chinese) ).

Google Scholar  

Chen, T. W. et al. Effect of cyclone split ratio on carbon release performance of excess sludge. Chin. J. Chem. Eng. 72 (11), 5761–5769 (2021) ( (in Chinese) ).

CAS   Google Scholar  

Martínez, L. F., Lavín, A. G., Mahamud, M. M. & Bueno, J. L. Improvements in hydrocyclone design flow lines stabilization. Powder Technol. 176 (1), 1–8 (2007).

Article   Google Scholar  

Song, M. H. et al. Discussion on the deep improvementof separation efficiency of liquid-liquid hydrocyclone. Prog. Chem. Ind. 40 (12), 6590–6603 (2021) ( (in Chinese) ).

Li, F., Liu, P., Yang, X., et al. Purification of granular sediments from wastewater using a novel hydrocyclone. Powder Technol. 393 (2021).

Li, F., Liu, P., Yang, X., et al. Numerical simulation on the effects of different inlet pipe structures on the flow field and seperation performance in a hydrocyclone. Powder Technol . 373 (2020).

Wakizono, Y., Maeda, T., Fukui, K. & Yoshida, H. Effect of ring shape attached on upper outlet pipe on fine particle classification of gas-cyclone. Sep. Purif. Technol. 141 , 84–93 (2015).

Article   CAS   Google Scholar  

Wakizono, Y. et al. Effect of ring shape attached on upper outlet pipe on fine particle classification of gascyclone. Sep. Purific. Technol. 141 , 84–93 (2015).

Huang, L. et al. Numerical analysis of a novel gas-liquid pre-separation cyclone. Sep. Purific. Technol. 194 , 470–479 (2018).

Jiayu, Z. et al. Study on the influence of central cone structure on flow field and separation efficiency of hydrocyclone. Mineral Conserv. Util. 4 (6), 65–69 (2018).

Lixin, Z. et al. Flow field analysis and structural optimization of internal cone-type oil removal hydrocyclone. Chem. Eng. Mach. 38 (02), 202–205 (2011).

Xiao, X. U. et al. Dissolved gas separation using the pressure drop and centrifugal characteristics of an inner cone hydrocyclone. Sep. Purific. Technol. 161 , 121–128 (2016).

Chen, B. et al. Experimental study on separation performance of dual overflow pipe hydrocyclone. Light Metals 475 (05), 9–13 (2018).

Peikun, L. et al. Numerical simulation and experimental study on separation performance of dual overflow pipe hydrocyclone. Coal Mine Machinery 41 (02), 40–43 (2020).

Showalter S, Kosteski Edward G. Three-phase cyclonic fluid separator: US, US007288138B2[P]. 2007-10-30

Zhang, Y. et al. The study on numerical simulation and experiments of four product hydrocyclone with double vortex finders. Minerals 9 (1), 23 (2019).

Article   ADS   Google Scholar  

Hongyan, L. et al. Effect of Hydrocyclone Overflow Pipe Structure on Fine Particle Separation. Journal of Chemical Engineering 68 (05), 1921–1931 (2017).

Hongyan, L. et al. Influence of novel outlet baffle structure on separation performance of hydrocyclone. J. Chem. Eng. 69 (05), 2081–2088 (2018).

Peikun, L. et al. Study on flow field characteristics and separation performance of hydrocyclone with overflow cap structure. Fluid Machinery 49 (01), 1–6 (2021).

Jihai, D. et al. Influence of conical slot on solid-liquid separation performance of hydrocyclone. J. Chem. Eng. 70 (05), 1823–1831 (2019).

Xiulin, L. et al. Experimental study on structural optimization of PV type cyclone separator. China Powder Sci. Technol. 25 (05), 72–77 (2019).

Xiulin, L. et al. Experimental study on structural optimization of cyclone separator. Modern Chem. Industry 39 (12), 205–209 (2019).

Ghodrat, M. et al. Numerical analysis of hydrocyclones with different vortex finder configurations. Minerals Eng. 63 , 125–138 (2014).

Liu, H., Jia, X. & Wang, B. Simulation study on the influence of overflow pipe structural parameters on cyclone separator performance. Fluid Machinery 48 (11), 6–10 (2020).

Jianxiang, Z. & Tianhe, Z. Numerical simulation of optimizing the convergent nozzle radius of cyclone separator exhaust pipe. Fluid Machinery 43 (12), 28–32 (2015).

Huang, Q., Xiao, H., Chen, A., et al. Hydraulic cyclone with conical slot structure. Patent No. CN109225687B, Shandong Province, 19 Mar 2021.

Ren, L. et al. Scheme design of filtration-type hydrocyclone. J. Southwest Pet. Inst. 01 , 82–85 (2005).

Yamei, L. et al. Analysis of the influence of cyclone separator structural parameters on its performance. Chem. Eng. Machinery 48 (05), 678–682 (2021).

Yang, L. & Zhenbo, W. Research progress on factors affecting separation efficiency of hydrocyclone. Fluid Machinery 44 (02), 39–42 (2016).

Zhang, W. et al. Study on flow field characteristics and separation performance of conical overflow pipe slotted hydrocyclone. Fluid Machinery 51 (08), 64–72 (2023).

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China (U2031142) and Heilongjiang Provincial Natural Science Foundation of China (LH2023F050).Technology Innovation Center of Agricultural Multi-Dimensional Sensor Information Perception, Heilongjiang Province (DWCGQKF202107) This work was supported by the Tianjin Research Innovation Project for Postgraduate Students (No. 2021KJ088).

Author information

Authors and affiliations.

School of Mechanical and Electrical Engineering, Qiqihar University, Qiqihar City, 161006, Heilongjiang Province, China

Shuxin Chen, Donglai Li & Lin Zhong

School of Mechanical and Power Engineering, Harbin University of Science and Technology, Harbin City, 150080, Heilongjiang Province, China

Jianying Li

Tianjin Ren’ai College, Tianjin City, 300000, China

Shuxin Chen

You can also search for this author in PubMed   Google Scholar

Contributions

C.S.: Designed and led the research project, responsible for overall project planning, contributed important ideas and theoretical support in paper writing. L.D.: Responsible for data collection and preprocessing. Provided detailed descriptions and analysis of the experimental section for paper writing. Conducted data analysis and statistical processing, offering strong support for interpreting the paper's results. L.J.: Provided significant insights in the discussion section. Supervised and guided the entire research process, offering valuable professional opinions. Made important revisions and additions in the literature review and conclusion sections. L.Z.: Responsible for data collection and graphical representation. All authors collaborated actively, contributing to different stages of the research task, and collectively played essential roles in completing the paper.

Corresponding author

Correspondence to Shuxin Chen .

Ethics declarations

Competing interests.

We, Chen Shuxin, Li Donglai, Li Jianying, and Zhong Lin, hereby declare that during the process of writing and submitting this paper, we have no financial or non-financial competing interests. We do not have any direct or indirect financial relationships or other interests that could potentially lead to conflicts of competing interests with any institutions, organizations, or individuals. The design, data collection, analysis, and interpretation of results for this study have been conducted with objectivity and integrity, unaffected by any competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Chen, S., Li, D., Li, J. et al. Optimization design of hydrocyclone with overflow slit structure based on experimental investigation and numerical simulation analysis. Sci Rep 14 , 18410 (2024). https://doi.org/10.1038/s41598-024-68954-y

Download citation

Received : 06 August 2023

Accepted : 30 July 2024

Published : 08 August 2024

DOI : https://doi.org/10.1038/s41598-024-68954-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.