What does citation factor reflect? And actually, does it really reflect anything?

I remember hearing Prof. Carolyn Bertozzi’s invited talk at San Diego ACS meeting in 2012. She mentioned that the key method applied in bioorthogonal chemistry has been published by Rolf Huisgen in 1960s and forgotten afterwards till early 2000s.

1,3-Dipolar cycloaddition developed by Huisgen has been applied by every scientist working with synthetic biology at least once in his carrier if he/she stepped into the field after 2000. This is after the Huisgen’s reaction was reawaken by the team of Sharpless and then Bertozzi’s group, and shortly after them applied by the entire community of life scientists.

But formally speaking, impact factor of the original Huisgen’s work within first 40 years after publication was rather low. According to nowadays evaluation criteria for research, that means that this was a low impact idea/method that was not relevant to the field. Consequences today would be low chances of getting promotions, funding and other recognition forms from the research community. But you see, this is not the case for the actual impact of the work on science, just taking it on a bit longer run:)

For more detail, I took a look at ISI Web of Science. First of all, I wanted to check the citations for Huisgen’s paper in Angewandte Chemie International Edition 1963, pp 565–598, title: 1,3-Dipolar Cycloadditions. Past and Future. The paper is actually not on ISI! That would be really bad for the authors.. Other papers by Huisgen and co-authors published in 1960-1980 are cited 30-100 citation per each, most after 1990 (!). Cycloadditions topic is cited 3,680 times after 1970. You can see the pattern for citations on the topic in the Figure below – it speaks for itself.

Then I got curious on other organic reactions that had a high impact on synthetic chemistry and biology in the last 30 years. The results of ISI were truly amazing. For example, Andre Mortreux’s work on metathesis published in 1974 "Metathesis of alkynes by a molybdenum hexacarbonyl–resorcinol catalyst". Chemical Communications (19): 786–787. It had almost no impact on the community for 20 years after publication date (see Figure 2). Nevertheless, the topic “metathesis” got its 29,207 citation times, most of them were after 1997.

My final search topic gave shocking results. ISI Web of Science repot on the topic “free radical polymerization” is in Figure 3. Any chemist knows that this is a method of polymerization by which a polymer forms from free radical building blocks. At present worldwide about 45% of the manufactured plastics and 40% of synthetic rubber are produced using free radical polymerization. The first free radically synthesized polymers were produced between 1910 and 1930 by initiation with peroxy compounds (read great story here: http://www.hindawi.com/journals/ijps/2009/893234/). Groundbreaking work on polymerization was done in 1920s by Staudinger. Nevertheless, his papers have not been cited much since industry does not publish their work! One example is the paper “On highly polymeric compounds, 14(th) Announcement - On polystyrol, a model of rubber“ by Staudinger et al. published in 1929 in Berichte, pp 241-263. It has been cited 36 times only.. Overall topic got 50,638 citations after 1935, with 8 of them (!) in 1935-1950.

One can argue that overall number of citations for all papers was initially low and then increased within last 30 years. There are more scientists, and way more papers are being published. Therefore trends observed on Figures 1-3 might seem typical. But this is not my point. I think that the number of citations right after the publication date (in case of metathesis this is first 20 years, for Staudinger's polymerization - ever), still seems to be unreasonable to evaluate if the work is of high importance or not.

ISI Web of Science search performed on the 2nd October 2015.