Secondary organic aerosols (SOA) refer to that organic components of particulate matter that transfer to the aerosol phase from the gas phase as products of gas-phase oxidation of parent organic species. SOA contribute to many important atmospheric processes, including visibility reduction, cloud formation by serving as cloud condensation nuclei, direct and indirect radiative forcing, and also to a variety of adverse health effects. We reported recently a bibliometric analysis of research trends on SOA based on data from 1711 papers by 4038 authors retrieved from SCI-EXPANDED journal publications using “secondary organic aerosol” and “secondary organic carbon” as the keywords to search titles, abstracts and keywords until end of 2011 . (The first mentioning of SOA there refers to an ACS Meeting Abstract in 1976 ). The papers were cited totally 42 482 times with an h-index of 90. Here we update these results with an analysis of data till end of year 2013 (retrieved on January 12th, 2014).
All publications referring to SOA from 1976 to 2013 were assessed according to: characteristics of publication outputs, distribution of outputs in journals, times cited per publication and analysis of paper titles, author keywords, KeyWords Plus, and abstracts. The words in titles and abstracts were separated, and then conjunctions and prepositions were discarded, as they were meaningless for further analysis. Keywords were defined as comma-separated items of one or more words. All keywords, both those reported by authors and those attributed by Thomson Reuters as well as the search words in titles and abstracts, were identified for 1990 to 2013 and separated into four 6-year periods, and then their ranks and frequencies were calculated in order to thoroughly and precisely analyze the variations of trends. Different words with identical meaning and misspelled keywords were grouped and considered as a single keyword. A word cluster analysis combination of the words in titles, author keywords, KeyWords Plus, and words in abstracts was used in the analysis. Besides, h-index was used to evaluate the total publications and leading journals in the SOA research. It is defined as the number of papers with citation number ≥ h and primarily used to characterize the scientific output of a researcher .
Results and discussion
Characteristics of publication outputs
From this study, 5461 authors published 2396 publications, most of which are articles (2179, 90.9 %). The h-index increased steadily after 2000 with a linear rate of 7.4 per year (R2 = 0.99) and reached 106 in 2013. SOA research developed from one publication in 1976 especially rapidly after 1998, to 323 publications in 2013 (Fig. 1), with secondary organic aerosol and secondary organic carbon as the search keywords in topics. It should be noticed that the number of publications was lower in 2013 than that in 2012 because their update has some delay. Usually it takes at least half a year to get all data. Actually for 2013 by end of March 2014 it rose by 43 to a total of 2439 publications. The two peaks in 1992 and 1995 show the impact of the developed Lagrangian trajectory model  and of the primary aerosol model with elemental carbon (EC) tracer method  on the field. Both enabled to quantify the primary and secondary organic aerosol concentrations to simulate the formation, transport and deposition of SOA that showed the major uncertainties in predicting SOA concentrations were the reactive organic gas emissions, aerosol yields and partitioning of condensable gases between the phases.
Publications appeared in 188 journals. Of the top five Atmospheric Chemistry and Physics published the most (501, 21 %), with Atmospheric Environment (432, 18 %), Environmental Science and Technology (296, 12 %), Journal of Geophysical Research-Atmospheres (202, 8.4 %) and Aerosol Science and Technology (75, 3.1 %) as next. The h-index of these five journals are 53, 58, 61, 49 and 23.
As in many other research fields, USA showed the highest counts (1464, 61 %), almost five times the number of China (320, 13 %). Germany ranked third (257, 11 %), followed by UK (176, 7.3 %), France (159, 6.6 %), and Switzerland (136, 5.7 %). The h-index is 98 for USA, 36 for China, 48 for Germany, 37 for UK, 31 for France, 32 for Switzerland, respectively.
Several atmospheric research field campaigns were dedicated to evaluating the roles of ozone, precursors, reaction products, meteorology, geography, etc. in the development of air pollution and its hazardous effects in megacities and their surroundings, such as the Mexico City Metropolitan Area Field Campaign (MCMA) in 2003 and 2006, Pearl River Delta (PRIDE-PRD in 2006, 3C-STAR in 2008 and so on), Megacity Initiative: Local and Global Research Observation (MILAGRO) in 2009, and European Integrated Project on Aerosol-Cloud-Climate and Air Quality Interactions (EUCAARI) from 2007 to 2010. Through the end of 2013, a total of 8 and 35 publications based on the EUCAARI and MILAGRO campaigns, 25 and 33 publications related to MCMA and Pearl River Delta (PRD) are indexed in SCI-EXPANDED. The number of publications in PRD almost tripled during the last 2 years.
Modeling, chamber, AMS (aerodyne aerosol mass spectrometer) and source apportionment are the most popular methods in SOA research (Fig. 2). Modeling is the most predominant. Modeling is a powerful tool to assess our knowledge of atmospheric processes related to SOA, including emissions, chemical formation, transport, deposition, and gas/particle partitioning. Various models were developed to simulate SOA, such as Lagrangian trajectory model, gas/particle partitioning absorption model, Secondary Organic Aerosol Model (SORGAM), PMCAMx, CMAQ, and WRF/Chem. Using WRF/Chem fully coupled with SORGAM, Jian et al.  found that in China anthropogenic sources contributed 35 % of total SOA and annual SOA production reached 3.05 Tg/yr in 2006, accounting for about 4–25 % of global SOA formation. Chamber is developed to quantify the aerosol formation potential (aerosol yield) of SOA precursors and to elucidate fundamental aspects such as chemical mechanisms and gas/particle partitioning of a given mixture of products in an experiment and then extend that knowledge to the atmosphere. Evidence from controlled smog chamber experiments and field measurements  showed that in the presence of high levels of nitrogen oxides (NOx = NO + NO2) typical of urban atmospheres, 2-methyloxirane-2-carboxylic acid (methacrylic acid epoxide, MAE) is a precursor to known isoprene-derived SOA tracers, and ultimately to SOA. How current models could improve the way that SOA form has been attributed to isoprene based on ambient tracer measurements, and lead to SOA parameterizations that better capture the dependency of yield on NOx. AMS allows for online measurements of organic aerosols with a high time resolution. While the AMS reports bulk organic matter, the identification of specific chemical markers is required for more comprehensive source apportionment. Methods for source apportionment include EC tracer method and its variant CO tracer method, positive matrix factorization (PMF) and chemical mass balance (CMB). AMS is always combined with PMF.
Biogenic hydrocarbons (1222 publications), anthropogenic hydrocarbons (475), and organic compounds generated from biomass burning (286) are three main precursors of SOA, among which terpenes (monoterpenes and pinene), isoprene, dicarbonyls (glyoxal and methylglyoxal), toluene, and xylenes have attracted the most attention (Fig. 3). Although the research on the contribution of dicarbonyls to SOA started late in 1999, the number of relative publications increased fast, even surpassed that of toluene in 2005 and during 2009–2013. Two recent SOA tracer-based measurement studies in Hong Kong and Guangzhou in PRD in southern China show a rather considerable difference in the relative SOA contributions from biogenic VOCs (isoprene and monoterpenes) and anthropogenic VOCs (toluene + xylenes). In Hong Kong, monoterpenes are the major contributor (up to 70 %), followed by isoprene (14–48 %) and toluene + xylenes (15–43 %). In Guangzhou, toluene + xylenes contribute more to SOA than isoprene and monoterpenes (up to 76 % from toluene + xylenes vs. 13–44 % from isoprene and 10–45 % from monoterpenes) due to the significantly larger toluene + xylene emissions in Guangzhou .
The most popular mechanisms included chemical processes (oxidation, gas-phase, heterogeneous, and aqueous-phase reactions) and physical processes (gas/particle partition and nucleation) (Fig. 4). Obviously, the number of publications related to gas-phase reactions decreased dramatically in 2013, even less than that in 2011. On the contrary, the number of publications on aqueous phase reactions continued to increase in 2013. The most recent research revealed that aqueous chemistry of dicarbonyls (glyoxal and methylglyoxal) accounted for 53 % of the total surface SOA in the PRD in fall .
The most hot research topic is emission, followed by chemical composition, yield, kinetics, volatility, size distribution and health (Fig. 5). There are large uncertainties associated with both anthropogenic and natural emission inventories on regional and global scales. The Po Valley (Italy) model inter-comparison exercise (POMI) showed that all models including CAMx, RCG, TCAM, CHIMERE, AURORA, EMEP underestimated the carbonaceous fractions (EC and OM) both in winter and in summer, which was related to missing information on local biogenic/biomass burning emissions of primary particulate matters in winter and lack of knowledge on sources of SOA from biogenic emissions . Better understanding of emission inventory will contribute to regulate emissions efficiently and reduce SOA concentrations. The number of publications related to chemical composition increased the fastest during 2011–2013. Due to the complexity of SOA, its precursors and intermediates, identification of chemical composition remains a challenge in SOA research. Multiple instruments were developed, but a perfect field instrument with full completeness, high chemical resolution, and high time/size resolution does not exist. Volatility is a hot topic because it influences the extent of gas/particle partitioning and aerosol yield, and chemical processes (oxidation reactions in the gas phase and reactions in the particle phase) can change the volatility of organic compounds. The volatility changes arising from these chemical reactions must be parameterized and included in models in order to gain a quantitative and predictive understanding of SOA formation. The increased number publications with health in their titles, keywords and abstracts indicated people’s concern on the health effects of particulate matter, however, there are only a limited number of studies on the health effects of SOA. This is mainly due to the lack of suitable particle exposure techniques for studies of in vitro toxicity effects of SOA . Polar organic and oligomeric compounds in SOA are likely more water-soluble and to some extent with unknown potential toxicity; thus dose may be enhanced by their solubility in the aqueous lung lining fluid relative to the reduced forms of these compounds . With the particle deposition chamber, Baltensperger et al.  proved that SOA with oligomers, which accounts for up to 23% of the total organic carbon, may induce distinct effects in lung cells. Modeling by CMAQ indicated that the estimated public health impacts from exposure to PM2.5 originating from aromatic hydrocarbons in gasoline, which are important precursors of SOA, lead to a central estimate of approximately 3800 predicted premature mortalities in the USA .
The ranking of the most cited publications changed a lot during 2010–2013. First, the most frequently cited publication was “Organic aerosol and global climate modeling: a review”  published in Atmospheric Chemistry and Physics and cited 920 times through 2013 (Table 1). Second, the publication “The formation, properties and impact of secondary organic aerosol: current and emerging issues” by Hallquist et al.  ranked first during 2010–2013 and was cited as high as 120 times per year on average. Third, the number of citations to the publications by Rogge et al.  and Kalberer et al.  presented a rapidly declining trend after 2006. Analysis of the most cited publications demonstrated the researchers’ attention to the modeling, formation, properties and impact of organic aerosols.
In this study, 2396 publications on SOA research were obtained from SCI-EXPANDED database during 1976–2013. New findings since the last bibliometric studies on SOA research during 1976–2011 include that Environmental Science and Technology ranked first with an h-index of 61 among the 188 journals; USA was the most prominent country in both the total number of publications (1464, 61%) and the h-index (98); the number of publications related to PRD almost tripled during the last 2 years. By analyzing the distribution and changes of author keywords, KeyWords Plus, words in publication titles and abstracts, the development of SOA research during last decade was described, and the future orientation of SOA research was predicted. It can be concluded that modeling will continue to be the predominant research method, the foci of SOA research will be the precursors of terpenes, isoprene and dicarbonyls, the mechanisms of oxidation and aqueous phase reaction, emission inventories and chemical composition in the 21st century.
This study was supported by the Chemistry and the Environment Division of the International Union of Pure and Applied Chemistry (No. 2012-028-2-600). Discussions with Dr. Weiwei Hu, Dr. Song Guo, Dr. Yunpeng Li and Prof. Yuh-shan Ho are gratefully acknowledged.
J. Li, Y. Zhang, M. Veber, P. H. Wine, L. Klasinc. Pure Appl. Chem.85, 1241 (2013).Google Scholar
D. Grosjean. Abstracts Papers of the American Chemical Society. 1, 2 (1976).Google Scholar
B. J. Turpin, J. J. Huntzicker. Atmos. Environ.29, 3527 (1995).Google Scholar
F. Jiang, Q. Liu, X. Huang, T. Wang, B. Zhuang, M. Xie. J. Aerosol Sc.43, 57 (2012).Google Scholar
Y. H. Lin, H. F. Zhang, H. O. T. Pye, Z. F. Zhang, W. J. Marth, S. Park, M. Arashiro, T. Q. Cui, H. Budisulistiorini, K. G. Sexton, W. Vizuete, Y. Xie, D. J. Luecken, I. R. Piletic, E. O. Edney, L. J. Bartolotti, A. Gold, J. D. Surratt. P. Natl. Acad. Sci. USA110, 6718 (2013).Google Scholar
N. Li, T. M. Fu, J. Cao, S. Lee, X. F. Huang, L. Y. He, K. F. Ho, J. S. Fu, Y. F. Lam. Atmos. Environ.76, 200 (2013).Google Scholar
D. Pernigotti, P. Thunis, C. Cuvelier, E. Georgieva, A. Gsella, A. De Meij, G. Pirovano, A. Balzarini, G. M. Riva, C. Carnevale, E. Pisoni, M. Volta, B. Bessagnet, A. Kerschbaumer, P. Viaene, K. De Ridder, A. Nyiri, P. Wind. Air Qual. Atmos. Hlth.6, 701 (2013).Google Scholar
M. Hallquist, J. C. Wenger, U. Baltensperger, Y. Rudich, D. Simpson, M. Claeys, J. Dommen, N. M. Donahue, C. George, A. H. Goldstein, J. F. Hamilton, H. Herrmann, T. Hoffmann, Y. Iinuma, M. Jang, M. E. Jenkin, J. L. Jimenez, A. Kiendler-Scharr, W. Maenhaut, G. McFiggans, T. F. Mentel, A. Monod, A. S. H. Prevot, J. H. Seinfeld, J. D. Surratt, R. Szmigielski, J. Wildt. Atmos. Chem. Phys.9, 5155 (2009).Google Scholar
P. A. Solomon, P. K. Hopke, J. Froines, R. Scheffe. J. Air Waste Manag Assoc.58, S3 (2008).Google Scholar
U. Baltensperger, J. Dommen, M. R. Alfarra, J. Duplissy, K. Gaeggeler, A. Metzger, M. C. Facchini, S. Decesari, E. Finessi, C. Reinnig, M. Schott, J. Warnke, T. Hoffmann, B. Klatzer, H. Puxbaum, M. Geiser, M. Savi, D. Lang, M. Kalberer, T. Geiser. J. Aerosol Med. Pulm. D. 21, 145 (2008).CrossrefGoogle Scholar
K. von Stackelberg, J. Buonocore, P. V. Bhave, J. A. Schwartz. Environ. Health, 12, 19 (2013).Google Scholar
M. Kanakidou, J. H. Seinfeld, S. N. Pandis, I. Barnes, F. J. Dentener, M. C. Facchini, R. Van Dingenen, B. Ervens, A. Nenes, C. J. Nielsen, E. Swietlicki, J. P. Putaud, Y. Balkanski, S. Fuzzi, J. Horth, G. K. Moortgat, R. Winterhalter, C. E. L. Myhre, K. Tsigaridis, E. Vignati, E. G. Stephanou. J. Wilson. Atmos. Chem. Phys.5, 1053 (2005).CrossrefGoogle Scholar
W. F. Rogge, M. A. Mazurek, L. M. Hildemann, G. R. Cass, B. R. T. Simoneit. Atmos. Environ. a-Gen.27, 1309 (1993).Google Scholar
M. Kalberer, D. Paulsen, M. Sax, M. Steinbacher, J. Dommen, A. S. H. Prevot, R. Fisseha, E. Weingartner, V. Frankevich, R. Zenobi, U. Baltensperger. Science303, 1659 (2004).Google Scholar