[Image above] Replication studies have demonstrated how slight changes in methodology can significantly affect results. Identifying such seemingly minor factors can help researchers conduct more accurate replications. Credit: Pixabay
I remember reading many news articles in the early- to mid-2010s on the “replication crisis” in science. This “crisis” was sparked by the publication of some high-profile papers that found a large percentage of results in published research articles across different disciplines (especially the social sciences) could not be reproduced.
While some news organizations were quick to claim these findings as evidence that science was failing, researchers swiftly stepped up to defend the scientific process and point out that in many cases these replication studies simply revealed challenges already known to scientists when it comes to conducting experiments.
For example, researchers have shown that studies deemed contextually sensitive are harder to reproduce because the original study’s exact circumstances can be difficult or impossible to replicate. As New York University associate professor of psychology Jay Van Bavel writes in The New York Times, “Imagine a study that examined whether an advertisement for a ‘colorblind work environment’ was reassuring or threatening to African-Americans. We assumed it would make a difference if the study was conducted in, say, Birmingham, Ala., in the 1960s or Atlanta in the 2000s.”
In addition, slight changes in methodology can significantly affect the results. I once read an anecdote by two research groups attempting to perform the same experiment. It took them months to identify the reason for their differing results—one group stirred the solution using a magnetic stir bar, while the other group swirled the flask by hand.
For many researchers, then, this “replication crisis” is not so much a crisis as it is an indicator of how to further improve research practices.
Providing more detailed methodologies in a paper is one practice that can help improve reproducibility. As the stirring anecdote above illustrates, the smallest change in experimental conditions sometimes has outsized impacts on results. Thus, providing detailed notes on the experimental setup—and understanding how seemingly minor methodological changes can impact experiments—allows researchers to conduct accurate replications.
In a recent study published in Materials Letters, researchers from the National Polytechnic Institute of Mexico and Tecnológico de Monterrey investigate the effect that one seemingly minor methodological change can have on experiments—the radius of the indenter tip used in scratch testing.
Scratch testing is a common method used to test the adhesion strength of thin films and hard coatings. The test involves applying a progressively increasing indenting load over the coated sample, which moves at constant speed until a failure mechanism is identified along the scratch groove. The load at which a well-defined failure mechanism is observed is called the critical load.
Researchers have evaluated the effect of indenter tip radius on the critical loads of various coating and substrate combinations, for example, here and here. However, despite the variety of these studies, “there is no information focusing on adhesion strength analysis of a coating/substrate system developed by a thermochemical diffusion process such as nitriding, cementing or more precisely, boriding,” the researchers write.
So, they investigated the effect that indenter tip radius has on critical loads of borided materials, specifically double phase boride layers (FeB/Fe2B) on AISI H13 steel, which is a tool steel grade standardized for hot working.
The researchers used a diamond Rockwell-C indenter tip for the scratch adhesion tests. Initial loads of 1, 2.5, 5, and 10 N and final loads of 10, 25, 50, and 100 N were selected for indenter tip radiuses of 20, 50, 100, and 200 μm, respectively.
Scanning electron microscopy analysis revealed that the predominant failure mechanisms in borided steels were lateral cracking, chipping, and gross chipping. While an indenter tip radius of 20 μm developed more severe damage on the borided surface, the more severe substrate plastic deformation occurred with an indenter tip radius of 200 μm. Thus, overall, “the smaller indenter tip radius generated more severe damage,” the researchers write.
The paper, published in Materials Letters, is “Scratch test in boride layers: influence of indenter tip radius on failure mechanisms” (DOI: 10.1016/j.matlet.2022.131918).
Author
Lisa McDonald
CTT Categories
- Basic Science