Researchers generate new knowledge. In science, knowledge is formulated and distributed in the form of academic publications, usually, peer-reviewed journal articles.
One of the questions constantly asked about publications and citations in the professional community is whether publications are a sign of accomplishment. Some colleagues suspect that many researchers produce publications just to publish something to inflate their list of publications.
One CEAS colleague said: “But citations and impact factors have to be handled with a pinch of salt… . Very often, the highly theoretical work involving mathematical derivations typically has lower citations because of the simple fact that only a small percentage of the scientists do/understand that kind of work… Some selected theoretical works do have cult followings and people will cite them (without fully understanding them) just to show that they are doing the right science!”
Another CEAS colleague said: “Citations can have a negative angle, i.e. you can be criticized for incorrect work or thoughts. It is more of a popularity index. The best and strongest work is typically way ahead of time. Citations are just all journal games, with no value in the long term… Many highly cited articles are mistaken.”
All that is correct.
In reality, however, the list of publication is only the first step when judging somebody’s work and contribution to science. One should read these articles and figure out what new ideas, results, or findings presented there. What is the place of these results in the context of the scientific area? Are they important and interesting? Have they influenced the field, i.e., made an impact? Often it is difficult to judge others’ work when it is not in your area. For that reason, confidential letters from external experts are sought when dealing with important decisions, such as tenure and promotion. Some people rely on the citation, which is in a sense “the index of popularity” of a paper and a measure of its impact.
In some cases, you have a great idea, choose a promising topic, obtain interesting results, and many people cite your paper. In other cases, it looks like you are doing the same, but nobody pays attention. It is very difficult to predict, whether a particular study would make an impact or not.
To illustrate that, I decided to pick five of my papers with interesting ideas (as I thought) which received reasonably high citations, and five papers with interesting and promising ideas (again, as I thought) which have not. As well as five intermediate cases. It is always good to analyze success and failure.
* * *
Five papers with relatively high (more than a hundred) citations – success stories:
• B Bhushan, M Nosonovsky 2010 The rose petal effect and the modes of superhydrophobicity, Phil Trans Royal Soc A 368 (1929) 4713-4728 (636 citations) http://doi.org/10.1098/rsta.2010.0203
This is my top-cited paper in the area of superhydrophobicity. At some point around 2010, I realized that two standard models – the Wenzel and Cassie-Baxter models – are insufficient to describe all the diversity of effects related to wetting. Therefore, I introduced several additional modes of superhydrophobicity and found evidence of them in the existing literature. I invited my former boss to be a co-author since he was very energetic in organizing research. I used pictures from his earlier paper to illustrate my ideas. Since the article addresses an issue, which is of interest to many researchers, if was actively cited.
• M Nosonovsky 2007, Multiscale roughness and stability of superhydrophobic biomimetic interfaces, Langmuir 23 (6), 3157-3161 (618 citations) https://doi.org/10.1021/la062301d
I wrote that paper in 2007, and my goal was to answer the question, which was often discussed in the literature: why do Lotus-effect surfaces have multi-scale roughness? I suggested one possible answer: due to the stabilization of the water-air interface (pinning of the triple line) by submicron roughness details. No experimental results reported here. The idea is simple, and it was illustrated by a simple schematic and a simple calculation of a conditional extremum. However, it addresses an important question, and somebody had to formulate it.
• M Nosonovsky 2007 On the range of applicability of the Wenzel and Cassie equations Langmuir 23 (19), 9919-9920 (270 citations) https://doi.org/10.1021/la701324m
This is my opinion in the argument about the Wenzel and Cassie equations. Do they deal with energy or with forces? I suggested the generalized Wenzel/Cassie equation. This paper is on the same issue of two (or more?) wetting states. See also “Generalized Wenzel and Cassie-Baxter equations”
• M Nosonovsky, V Hejazi, 2012, Why superhydrophobic surfaces are not always icephobic, ACS Nano 6 (10), 8488-8491 (444 citation) https://doi.org/10.1021/nn302138r
At some point in 2011, I realized that there is a lot of interest in the “icephobicity,” i.e., the potential of superhydrophobic surfaces to repel ice. Any explanation of how these two are related will have the potential to be well-cited. I was thinking about it and came up with my explanation. It took me about a year to prepare this paper with my PhD student, Vahid Hejazi. We simply compare how the surface energy affect the solid phase (from the point of view of Fracture Mechanics) and how it affects wetting (surface tension). This difference is what makes the icephobicity different from the hydrophobicity. And voila, the paper is well-cited! Not because of any manipulation, but because it provides and answer for an important question. Again, no news experimental results here.
• MS Hasan, A Kordijazi, PK Rohatgi, M Nosonovsky, 2021, Triboinformatic modeling of dry friction and wear of aluminum base alloys using machine learning algorithms, Tribology International 161, 107065 (126 citations) https://doi.org/10.1016/j.triboint.2021.107065
The field of Triboinformatics is a combination of Tribology with machine learning. There is a lot of interest in this new field, this is a “hot” new area. The idea was to combine tribology, which is a data-centric area, with Machine Learning and Artificial Intelligence. We, together with my student and colleagues from Materials Science, were among the first who published on Triboinformatics. The 2021 got 126 citations in four years. (See more details in “Triboinformatics or Intelligent Tribology”)
* * *
Five papers with good ideas and relatively moderate citations (50-150). Success, but more can be done
• AA Fedorets, M Frenkel, E Bormashenko, M Nosonovsky, 2017, Small levitating ordered droplet clusters: stability, symmetry, and Voronoi entropy, The Journal of Physical Chemistry Letters 8 (22), 5599-5602 (60 citations) https://pubs.acs.org/doi/full/10.1021/acs.jpclett.7b02657
This paper is in a series on levitating droplet clusters, in collaboration with Tyumen University in Russia, where the experiments are performed. The arrangement of droplets in cluster are different from colloid crystals. We report here unusual arrangements, with 5-fold and 7-fold symmetry. My idea was to relate them to very “mystical mathematical objects, the Dynkin’s diagrams / Arnold’s ADE-classification / “Mathematical Trinities.” Absolutely amazing and incredible! It could be cited more than just 60 times, by mathematicians and by physicists! (See more details in “Can small droplet clusters be related to the ADE-classification?”)
• SH Yang, M Nosonovsky, H Zhang, KH Chung 2008, Nanoscale water capillary bridges under deeply negative pressure, Chemical Physics Letters 451 (1-3), 88-92 (88 citations) https://doi.org/10.1016/j.cplett.2007.11.068
This was done when I was an NRC Postdoc at NIST. We were four postdocs, and we wanted to understand what capillary bridges (in atomic force microscopy) tell us about properties of water. Which maximum negative pressure water can sustain. (Yes, water can sustain tensile stress, like solids). This is also called the spinodal limit on the water phase diagram. We used the NIST AFMs for experiments and obtained the world record. Although some physicists argued against us, we refuted the criticism. We hoped for Nature / Science / PNAS paper, but it got to the CPL only, as some NAS members opposed it. This paper is important, and it could be cited more than just 88 times!
• R Ramachandran, M Kozhukhova, K Sobolev, M Nosonovsky, 2016, Anti-icing superhydrophobic surfaces: Controlling entropic molecular interactions to design novel icephobic concrete, Entropy 18 (4), 132 (113 citation) https://doi.org/10.3390/e18040132
There are entropic force and enthalpic forces. The hydrophobic interaction is an example of mysterious “entropic” forces. But what about “icephobic interactions”? Do they exist? Are they entropic? This is a conceptually important paper for me which deals with an entropic nature of “icephobic” and hydrophobic interactions. It was written in collaboration with researchers from Civil Engineering department who work on icephobic coatings on concrete, and their results were used to illustrate my concept.
• V Hejazi, AE Nyong, PK Rohatgi, M Nosonovsky, 2012, Wetting transitions in underwater oleophobic surface of brass, Advanced Materials 24 (44), 5963 (76 citations) DOI: 10.1002/adma.201202516
There we observed and reported a new phenomenon: a transition between two wetting states – Wenzel and Cassie – in a metal-water-oil system. It was published in a prestigious journal, Advanced Materials. The paper reports a discovery of a new phenomenon, so could be noticed by the community.
• A Breki, M Nosonovsky 2018, Einstein’s viscosity equation and nanolubricated friction, Langmuir 34 (43), 12968-12973 (59 citations) https://doi.org/10.1021/acs.langmuir.8b02861
What was a mistake in Albert Einstein’s 1905 Doctoral thesis, which he corrected only in 1911? That was in the formula for the viscosity of a dispersion/suspension of small solid particles. Can you deduce the same result from simple dimensional considerations? Yes, but if you do not take into account the rotation of the particles, you would get the same mistake that Einstein made in 1905. The viscosity of a suspension formula is very important for nanolubrication (i.e., when lubricating oil is enhanced with nanoparticle additives). We investigate that using experimental results (for oils with WS2 nanoparticle additives) of my Russian collaborator and colleague.
See also “Albert Einstein and nanofriction”
* * *
Five papers with great ideas but low citations. More should be done to spread the word about amazing ideas:
• M Nosonovsky, B Bhushan 2008, Do hierarchical mechanisms of superhydrophobicity lead to self-organized criticality?, Scripta Materialia 59 (9), 941-944 (29 citations) https://doi.org/10.1016/j.scriptamat.2008.06.013
In this opinion paper, I suggested that wetting and superhydrophobicity involve avalanche-like processes and thus may lead to Self-Organized Criticality (SOC), a particular type of self-organization. While the paper has not received much attention in the wetting community, it was noted by an expert in protein folding, who paid attention to the fact that folding is driven by hydrophobic forces and may lead to SOC. Later, I extended my work on the entropic nature of folding.
• Z Chen, M Nosonovsky 2017 Revisiting lowest possible surface energy of a solid, Surface Topography: Metrology and Properties 5 (4), 045001 (19 citations) https://iopscience.iop.org/article/10.1088/2051-672X/aa84c9/meta
What is the maxumum possible contact angle on a smooth surface? In a celebrated and often cited (more than 1500 citations) paper by Nishino et al. (1999) “The Lowest Surface Free Energy Based on -CF3 Alignment” Langmuir 15:4321-4323, it has been reported that the maximum possible water contact angle with a smooth solid material is 119 degrees. We revisit this statement and find some issues with it. The work was originally a course project in my class on functional surfaces, which I taught during my sabbatical at The Technion. See “Revisiting lowest possible surface energy of a solid”, and “The lowest surface energy is not 6.7 mJ/m2, as it was reported for n-Perfluoroeicosane”
• M Nosonovsky, P Roy 2020 Allometric scaling law and ergodicity breaking in the vascular system Microfluidics and Nanofluidics 24 (7), 53 (11 citations) https://link.springer.com/article/10.1007/s10404-020-02359-x
This is a conceptually important paper in which we study the allometric law of scaling in living organisms. Allometry is a type of scaling in biology, which explains how various parameters (such as the metabolism rate, speed, and even the lifespan of an animal) depend on the average body mass. Ergodicity is a concept in the theory of dynamical systems, which implies that the time average as a parameter can be substituted by the ensemble (or phase-space) average. The flow of biological liquids is often non-ergodic. We discuss the famous West-Brown-Enquist (WBE) model of fractal branching in a vascular network. The WBE model explains, for example, why the average lifespan of an animal is proportional to power four root of its average body mass. We suggest an enhancement of the model and also suggest a possibility that fractal branching can lead to ergodicity breaking.
For more details see “Allometry and Ergodicity breaking in a fractal capillary network”
• MS Hasan, M Nosonovsky 2021 Topological data analysis for friction modeling, Europhysics Letters 135 (5), 56001 (10 citations) https://doi.org/10.1209/0295-5075/ac2655
That was our first paper, with my student, in Triboinformatics on applying the fashionable Topological Data Analysis (TDA) method from “Big Data” to surface roughness. The idea seemed very original and exciting. It was published in a physics journal; however, it only got 10 citations so far in four years. Later, we published more papers on TDA, with more emphasis on experimental data, mostly in materials science journals, and they were cited better.
• AS Blumenthal, M Nosonovsky, 2020, Friction and dynamics of verge and foliot: How the invention of the pendulum made clocks much more accurate, Applied Mechanics 1 (2), 111-122 (5 citations) https://doi.org/10.3390/applmech1020008
In this paper we (together with my student) explained why the accuracy of clocks improved by 30 times after the pendulum was invented in the 1630s. Scaling arguments suggest that the clocks’ accuracy improved by a factor of the order of π/μ, or by about 30 times, which is consistent with actual historical data. (For more details see “The invention of the pendulum clock in the 17th century”, also “A Nobel Prize Winner attended Dr. Nosonovsky’s lecture”)
* * *
I discussed 15 of my many papers with various ideas. There are three categories: (i) “Success” (ii) “Success, but more could be done,” and (iii) “Definitely, these papers could be cited by more people.”
If you have many original ideas and they are mature enough to be transformed into peer reviewed journal articles (i.e., supported by evidence, experimental or computational results), it is not bad to publish them. Some of them will be cited, some of them will not. Papers should be reader-friendly. Even if your paper involves complicated mathematical derivations, which are difficult to follow, you can explain and present the main finding in “layman terms” to make it easier for a reader to understand the underlying idea.
On the other hand (and this is in response to those colleagues who keep saying that there is no need to rush publishing 10 articles every year), if you do not have many original ideas mature enough to be transformed into peer reviewed journal articles, you absolutely do not have to publish many papers. You can publish one paper per year or one paper in five years. Every individual publication will be judged by its own merits, and quantity does not substitute for quality. What matters is new ideas and results that you introduced to the literature, and whether these results influenced others and still remembered after decades. However, this discussion and analysis do not make sense without specific examples, and this is why I brought 15 of my papers as examples.
What is my conclusion? You never know which of your ideas will be picked up by the community. You should try to produce high-quality publications (the best you can) and talk to people about your ideas. Will they like your ideas and your papers? Will they pick up your work to develop it further? You cannot know that in advance. This may depend on whether your work provides a needed answer to a question that everyone is looking for, but also on many other reasons, such as in which journal it was published, and who were your coauthors. However, with experience and scientific intuition, you can sometimes guess which work has a good chance of influencing your colleagues in the scientific community. You also should not be shy in explaining your ideas to people.