Examining the dynamics of an emerging research network using the case of triboelectric nanogenerators

This is a e sion o a publica ion in Please ci e he publica ion as ollows: DOI: Copy igh o he o iginal publica ion: This is a pa allel published e sion o an o iginal publica ion. This e sion can di e om he o iginal published a icle. published by Examining he dynamics o an eme ging esea ch ne wo k using he case o iboelec ic nanogene a o s Suominen A ho, Peng Haoshu, Ranaei Sami a Suominen, A., Peng, H., Ranaei, S. (2018). Examining he dynamics o an eme ging esea ch ne wo k using he case o iboelec ic nanogene a o s. Technological Fo ecas ing and Social Change. DOI: 10.1016/j. ech o e.2018.10.008 Final d a Else ie Technological Fo ecas ing and Social Change 10.1016/j. ech o e.2018.10.008 © 2018 Else ie Inc. Examining he dynamics o an eme ging esea ch ne wo k using he case o iboelec ic nanogene a o s A ho Suominen , Haoshu Peng, Sami a Ranaei A B S T R A C T The analysis o a scien is 's decision o conduc esea ch in a speci fi c scien i fi c fi eld is an in e es ing way o ace he eme gence o a new echnology. The g ow h o a esea ch communi y in size and pe sis ence is an impo an indica o o a new scien i fi c fi eld's i ali y. Using a case s udy on iboelec ic nanogene a o (TENG) echnology, his s udy iden i fi es how esea ch pa icipa ion and communi y dynamics e ol e du ing he eme gence phase o a echnology, and u he wha a e he key condi ions and de e minan s o he eme gen au ho ne wo k. The s udy uses scien i fi c publica ion da a om 2012 h ough 2017 ex ac ed om he Web o Science da abase. Resul s show communi ies eme ging h ough ac o s' close p oximi y a he han om hei sha ed hema ic o ien a ion. Fo indi idual esea che s, he bounda y be ween p io esea ch and TENG esea ch was negligible pa ly ques ioning he exis ence Kuhnian pa adigm shi s. 1. In oduc ion Eme gence is wha a “ sel -o ganizing p ocess p oduces ” (Co ning, 2002). Sel -o ganiza ion equi es ac o s, o ganiza ions and indi iduals ha will ake pa in he p ocess o eme gence. In he con ex o ech- nological eme gence, he dynamics o ac o s aking a ole in he dis- co e y p ocess ha e been b oadly analyzed. Resea che s ha e s udied he eme gence o esea ch ne wo ks h ough co-au ho ship (Suominen, 2014), co-ci a ion (Boyack and Kla ans, 2010), and bibliog aphical coupling (Ja ne ing, 2007). Resea che s ha e used o he s s udies o examine whe he au ho s sha e e minology and c ea e pe sis en new esea ch opics ha migh be eme ging (Guo e al., 2011; Small e al., 2014; Suominen and Toi anen, 2015). In p ac ice, an ac o 's ole has been ope a ionalized h ough p oxies such as he a e age numbe o au ho s pe pape , he numbe o con ibu ing o ganiza ions, and he numbe o coun ies o ci ies in which he au ho s conduc esea ch. In 1969, Ay es (1969) pu o wa d a amewo k o he sel -o ga- nizing dynamic p ocess o ac o s. This p ocess was based on he numbe o ac o s being a unc ion o an al eady known and in e es ing idea le wi hin a fi eld. Ay es ollowed a Humbold ian no ion ha he p og es- sion o echnology and selec ion o esea ch opics a e he unc ion o he a ailabili y o no el ideas. Ay es d ew om Hol on (1962), who s a ed ha only a fi ni e lode o in e es ing ideas exis s wi hin a scien- i fi c fi eld. Once a scien is opens a new lode ia a scien i fi c disco e y, mo e in es iga o s mig a e o he new fi eld. This phenomenon is called a ‘ gold ush ’ as schola s “ de ec om hei old fi eld, in sea ch o g eene pas u e ” (Ay es, 1969). As he mine emp ies, making new disco e ies mo e challenging and sca ce, esea che s a e o ced o mi- g a e ye again o new oppo uni ies (Ay es, 1969). I is clea ha he issue o esea che s pu suing a speci fi c esea ch a ea is much mo e complex han he pu e Humbold ian endea o o a esea che (e.g., Laudel and Gläse , 2014). A esea che , pa icula ly so called “ no mal scien is ” ansi ion easily o agendas ha a e well- unded. (B aun, 2012) Resea che s also look o di e si y in o de o di ff e en ia e his o he wo k om o he scien is s and mi iga e isk associa ed wi h a na ow ocus. The decision o esea che s o endea o in a fi eld is in e es ing when assessing he e olu ion o a echnology. Suominen (2013) analyzed he numbe o en an s in and he cohe- si eness o a fi eld by measu ing he in oduc ion o new e ms. The use o he en an measu e was exempli fi ed in he e alua ion o echnolo- gical p og ession in wo fi elds: di ec me hanol uel cells and dye sen- si ized sola cells. Howe e , he s udy was unable o alida e u he i Hol on ’ s (1962) analogy o a fi ni e space holds ue. The discussion on communi ies and ac o s is no inconsequen ial o he b oade opic o echnological eme gence. Eme ging echnology is de fi ned as a echnology ha will yield signi fi can bene fi s o a wide ange o economic o socie al sec o s (Ma in, 1995). The cha ac e - is ics o eme gence a e no el y, pe sis ence, g ow h, and communi y o ma ion (Suominen and Newman, 2017). These cha ac e is ics a e o en ansla ed o scien ome ic indica o s enabling he ope - a ionaliza ion o eme gence. Temple on and Fleischmann (2013) de- sc ibed eme gence as no iceable h ough he inc ease in ac o s o e ime. Howe e , exis ing s udies mos ly ep esen he dynamics h ough ne wo ks, such as co-au ho ship (Pe ianes-Rod íguez e al., 2010; Glänzel and Schube , 2005), co-ci a ion (Small e al., 2014), and bib- liog aphic coupling (Kuusi and Meye , 2007), seldom conside ing he g ow h o schola s' pa icipa ion and i s unde lying dynamics. E en hough he dynamic o schola s' pa icipa ion is cen al o Ay es's and Hol on's wo k (Ay es, 1969; Hol on, 1962), he e a e limi ed s udies ha quan i a i ely examine he pa icipa ion dynamics o ack he eme gence phenomenon. This s udy elies on bibliome ic da a wi h quali a i e in o ma ion acqui ed om a su ey o explo e one dimen- sion o he eme gence phase o a echnology — how esea ch pa ici- pa ion and communi y dynamics e ol e du ing he eme gence o a new echnological pa hway. The s uc u e o his pape is as ollows: he fi s sec ion desc ibes he s udy's backg ound, ocusing on he dynamics o eme ging scien i fi c communi ies and hei link o he li e a u e on echnological eme - gence. The second pa o he backg ound desc ibes iboelec ic na- nogene a o (TENG) echnology as a case s udy and wha he measu es a e expec ed o unco e . The hi d sec ion e iews he da a collec ion p ocess and he me hodology, ollowed by esul s and discussion in he ou h and fi h sec ions, espec i ely. 2. Backg ound 2.1. Eme gence in communi ies o p ac ice T acing and concep ualizing he eme gence o new echnical in- no a ions has always been o in e es o schola s, as he inno a ions a e closely linked wi h economic p ospe i y (Dosi, 1982). In he pas dec- ades, schola s ha e used di ff e en e ms and axonomies o de fi ne he phenomena and he o igins o eme ging echnologies. Schumpe e (1961) p o ided he seminal explana ion o eme ging echnologies. Schumpe e depic ed echnological de elopmen as a ci cula fl ow dis up ed by spon aneous changes (p ima ily om inno a i e en- ep eneu s) o he p e iously exis ing equilib ium s a e. Eme ging echnologies can be he esul o ei he echnological de elopmen o scien i fi c p og ess. The idea o he ci cula fl ow in echnological change is somewha analogous wi h Kuhn's scien i fi c pa adigm (Kuhn, 1970). Kuhn in oduced he concep o he pa adigm shi in he con ex o scien i fi c disco e ies, an ac ha aligned wi h he Schumpe e ian no ion ha any p og ess in science o echnology is he esul o adical change. Kuhn's iew con as ed wi h he es ablished knowledge o his ime – he la e being ha he d i ing o ce behind scien i fi c ad ances was a s eady accumula ion o knowledge and ideas. Kuhn a gued ins ead ha he p og ess o science occu s du ing a e olu iona y explosion o new knowledge, claiming ha scien i fi c e olu ion has a cyclical pa adigm. The cycle begins in a s able pe iod o no mal science, when esea ch is conduc ed acco ding o a se o accep ed heo ies among scien i fi c communi ies. Resea ch endea o s hen ex end he scope and p ecision o he es ablished knowledge in he n o mal science phase, o puzzle- sol ing phase, which usually has p ede e mined solu ions, p e- cedes a ise in anomalies ha iola e he “ pa adigm-induced expec a ions ha go e n no mal science ” (Kuhn, 1970). These anomalies begin o accumula e a ound ce ain pa adigms, o cing science o explo e al e na i es, o ee alua e cu en heo ies, and fi nally o shi o a new pa adigm. This is simila o Hol on ’ s (1962) image o opening a lode. Pa adigm shi s and echnological eme gence mani es in changes in a gi en fi eld's communi ies o p ac ice o dynamism. Dynamism in e- sea ch communi ies is mos ly analyzed h ough esea ch collabo a ion. The mo i es o in es iga e collabo a i ely s em om six ac o s (Ka z and Shapi o, 1994): (1) inc eased esea ch cos s, (2) educed commu- nica ion cos s and a el cos s, (3) ad ances in science ha depend on in e ac ions among scien is s, (4) inc eased awa eness o he need o in e disciplina y wo k, (5) poli ical d i e s such as unding, and (6) inc eased scien i fi c specializa ion. Collabo a ion is he co e o com- muni y c ea ion, as ac o s sha e and lea n om each o he . Scien o- me ic s udies ha e ex ensi ely examined collabo a ion in a numbe o a eas, such as s em cell esea ch (Li e al., 2009), g aphene esea ch (L e al., 2011), uel cells (Suominen, 2014), ola ile o ganic compounds (Zhang e al., 2010), and global posi ioning sys em esea ch (Wang e al., 2013). S udies o co-au ho ship o en do no conside he g ow h o com- muni ies, a he explaining di ff e ences in exis ing communi ies. Ay es desc ibed scien i fi c esea ch as compa able o ocean explo a ion based on he assump ion o an ocean comp ises a fi ni e pool o ideas. As new esea ch opens new pools o explo e, awa eness o he disco e y sp eads, en icing a numbe o in es iga o s o join in he explo a ion. As esea ch is conduc ed, u u e disco e ies become much ha de o achie e as mos o he fi ni e space has been explo ed. This esul s in fi eld sa u a ion, leading o new pools disco e ed among new s eams o science. This p ocess o e olu ion is illus a ed in Fig. 1. The long dashed line in Fig. 1 desc ibes basic esea ch pa icipa ion. Fig. 1. Ay es's and Hol on's model o esea ch e olu ion (Ay es, 1969). In Hol on (1962), he olume o basic esea ch pa icipa ion g ows apidly a e he opening o a new s eam o esea ch. The inc ease in human labo wi hin he fi eld apidly inc eases he amoun o al eady known and applied ideas. The e ff o needed o fi nd one uni o dis- co e y is highe when less is known abou he subjec . When almos all he in e es ing ideas ha e been disco e ed, he equi ed e ff o o p o- duce new known and applied ideas diminishes, esul ing in a s eepe cu e. Howe e , he lack o emaining ideas educes he olume o esea ch pa icipa ion. Hol on's heo e ical amewo k has emained empi ically unexplo ed. Howe e , unde s anding how esea ch com- muni ies g ow and whe e sub-communi ies eme ge would gi e much insigh o he p ocess o echnological eme gence. 2.2. T iboelec ic nanogene a o To unde s and how en an dynamism e eals he eme gence o a new echnology, he esea ch communi y g ow h o TENG echnology se es as an example. In en ed in 2011 by Z.L. Wang a he Geo gia Ins i u e o Technology and fi s published in 2012 he TENG is a new echnology ha can e ff ec i ely ha es ambien mechanical ene gy om a ious mo ions eadily a ailable bu in a sense was ed in ou daily li es, such as human mo ion, ib a ions, mechanical igge ing, o a ing i es, wind, and fl owing wa e . A nanogene a o comp ises wo s acked shee s made o ma e ials ha ing dis inc ly di ff e en ibo- elec ic cha ac e is ics, wi h me al fi lms deposi ed on he op and bo om o he assembled s uc u e. Resea ch has shown TENG echno- logy's p omising applica ions, such as po able elec onics and sel - powe ed senso ne wo ks (Fan e al., 2012). TENG echnology has g ea comme cializa ion po en ial mainly due o i s capaci y o ha es ene gy om he en i onmen . E ff o s ha e been made o explo e applica ions, such as he po en ial o ealize a sel -sus aining in eg a ed sel -powe ed mic osys em (Zhang e al., 2015), and i s low-cos ab ica ion p ocess (Qiu e al., 2015). Resea ch has sugges ed ha TENG echnology can be used as senso s (Allu i e al., 2015), hyb id ene gy cells (Zheng e al., 2014), po able o wea able elec onics (Zhu e al., 2013), o la ge-scale ene gy (wind o ocean wa e) collec ion de ices (Wen e al., 2014). Compa ed o o he echnologies, TENGs ha e shown ad an ages such as high ou pu , high ene gy-con e sion e ffi ciency, as well as abundan choices o ma e ials, scalabili y, and fl exibili y. The a ea powe densi y eaches 599 W /m 2 , he olume powe densi y eaches 15 MW /m 3 , and he ene gy con e sion e ffi ciency eaches up o 85%. Speci fi cally, he compa ison o TENGs wi h he pe o mances o an- o he mechanical ene gy ha es e , elec omagne ic gene a o s (EMGs), demons a es ha he ou pu pe o mance o EMGs is p opo ional o he squa e o he equency, while ha o TENGs is app oxima ely in p opo ion o he equency. The e o e, TENGs ha e supe io pe o - mance when compa ed o EMGs a low equency ( ypically 0.1 – 3 Hz). Mo eo e , he ex emely small ou pu ol age o EMGs a a low e- quency makes hem almos inapplicable o d i e any elec onic uni ha equi es a ce ain h eshold ol age ( ≈ 0.2 – 4 V). Thus, mos o he ha es ed ene gy is was ed. In con as , TENGs ha e an ou pu ol age ha is usually high enough (> 10 – 100 V) o such an applica ion and is independen o equency so ha mos o he gene a ed powe can be e ff ec i ely used o powe di ff e en de ices (Wang, 2017; Zi e al., 2016). Al hough he es ima ion o wha ep esen s he me apho ical opening o a lode o he occu ence o a pa adigm shi emains highly subjec i e, and in he case o TENG echnology only he u u e migh yield a consensus on i s impac , s ong e idence exis s o suppo TENGs' pa adigm-shi ing na u e. The in en o , Z.L. Wang, cu en ly anks fi s in ci a ions in he fi eld o nano echnology and nanoscience 1 . He and his esea ch g oup ha e ecei ed mul iple awa ds, such as he 1 h p://www.webome ics.in o/en/node/198. En e nazionale id oca bu i S.p.A. (ENI) awa d o he ene gy on ie . In he ENI p ess elease 2 he commi ee highligh ed TENGs as a com- ple ely new g oup o de ices showing signi fi can po en ial in ene gy e ie al and gene a ion. TENG echnology is a aluable case s udy o unde s and esea ch pa icipa ion and communi y dynamics du ing he eme gence o a new echnological pa hway. TENG esea ch would seem o o ff e a me a- pho ical opening o a lode (Hol on, 1962) o a Kuhnian pa adigm shi (Kuhn, 1970) used as a s a ing poin o his analysis. The echnology me ges di ff e en aspec s o na u al sciences om ma e ials science (19% o publica ions), physics (14.1% o publica ions), chemis y (17.5 o publica ions). This c oss-disciplina i y inc eases he applicabili y o he esul s, bu is should be no ed ha ou da a does no ex end he na u al sciences. 3. Da a and me hod 3.1. Da a collec ion Two da ase s we e used o e ie e in o ma ion o his s udy: a scien i fi c publica ion da abase and a ques ionnai e. Publica ion da a was used as a p oxy o unde s and he beha io o esea ch commu- ni ies in TENG echnology. The sea ch que y used o ob ain he da a was o mula ed ia keywo ds collec ed and e iewed by TENG expe s a he Geo gia Ins i u e o Technology. The sea ch was execu ed in May 2018 and e ie ed 1229 eco ds o TENG publica ions om he Web o Science (WoS) Co e Collec ion da abase. The sea ch que y was limi ed o he ime pe iod om Janua y 2012 h ough Decembe 2017. The ques ionnai e was designed o collec in o ma ion on why e- sea che s selec ed o pa icipa e in TENG esea ch and om wha o i- gins. This allowed o unde s and he cen al mo i a ion, which is key in Hol on ’ s (1962) amewo k. Ay es (1969) explained Hol on's ame- wo k as ollows: ha an ac o en e ing esea ch would be mo i a ed by he ease o making new disco e ies. The amewo k also sugges s ha when no mo e easy disco e ies emain, many esea che s fi nd new opics elsewhe e. The ques ionnai e esponden s we e asked, using open-ended ques ions, wha hei mo i a ions o s a TENG esea ch we e and, i hey had conside ed d opping he esea ch, could hey explain why. In addi ion, he opical dis ances be ween esea che s who joined oge he was a ocal poin o he esea ch. Kuhn's no ion ha esea che s making pa adigm shi s a e new o he fi eld was also es ed h ough an open-ended ques ion abou wha esea che we e ac i e p io o hei TENG esea ch. Fu he mo e, he ques ionnai e con ained an open-ended ques ion o iden i y i he esponden s could iden i y communi ies ha had eme ged a ound TENG esea ch. This was used o be e unde s and i ehicles exis ed o suppo communi y c ea ion. Finally, he ques ionnai e inqui ed abou he esea che s' backg ound (e.g., yea s in esea ch). The ques ionnai e ecipien s we e 615 au ho s o TENG publica ions e ie ed om he WoS da abase. Au ho s wi h missing email in o ma ion we e excluded om he su ey. The ques- ionnai e is Appendix A. 3.2. Resea ch pa icipa ion and up ake measu es The publica ion da a was analyzed using a Py hon sc ip , eading he downloaded da a om he WoS. The p ocess ead he abula o - delimi ed fi les and ex ac ed he au ho fi eld (AF) o u he analysis. The names lis ed as au ho s o each publica ion we e sepa a ed in o single en i ies: au ho s. A da a s uc u e was o med o gi e each au ho a unique ID and o ganiza ion, a lis o co-au ho s, and a lis o used emails. O ganiza ions we e connec ed o au ho s in wo ways: fi s , au ho 2 h ps://www.eni.com/en_IT/media/2018/07/winne s-o - he-2018-eni- awa ds-announced. a ffi lia ions ha we e nes ed in a C1 fi eld enabled each au ho o be connec ed wi h a speci fi c o ganiza ion; second, eco ds ha did no ha e a clea de e mina ion o au ho o ganiza ions (i.e., no links om he C1 fi eld) mean ha only he ep in au ho was a ffi lia ed wi h an o ganiza ion. The AF was also used o link co-au ho s. Fo each pape au ho , he sc ip s o ed a lis o co-au ho s. I he au ho s' email add esses we e a ailable, each au ho was also linked o hei co-au ho s' email ad- d esses. The sc ip checked he a ailabili y o email add esses and hen linked email add esses wi h he associa ed ep in au ho s. I mul iple emails we e p o ided, he sc ip examined i he numbe o emails co esponded wi h he numbe o au ho s. I he numbe o emails and au ho s ma ched, he emails we e linked and we e expec ed o appea in he same o de as he au ho s' names appea ed. Finally, each au ho 's eco d was linked o he eco d's publica ion yea s, i le, unding o igin, and scien i fi c subjec ca ego y. The Py hon sc ip ope a ionalized esea ch communi y pa icipa- ion wi h ou a iables: he numbe o au ho s en e ing yea ly, he numbe o au ho s exi ing yea ly, and he yea ly coun o ac i e au- ho s. The fi s measu e was de fi ned by he numbe o au ho s who fi s published in a gi en yea . The second measu e was de fi ned by he numbe o au ho s published in yea who had no subsequen ly pub- lished in he fi eld ( + n). This analysis excluded he las wo yea s in he da ase , since he eliable es ima ions o exi ing could no be made so nea he end o he ime se ies. The hi d measu e was calcula ed by coun ing each au ho ac i e in he yea s hey en e ed and exi ed he fi eld. I an au ho did no exis on he ac i e au ho lis be o e he las wo yea s o he ime se ies, hen ha au ho was calcula ed as ac i e om he ime o fi s publica ion o he end o he ime se ies. The ou h measu e was he numbe o unique au ho s in a gi en yea . This did no ake in o accoun any o he alues han he amoun o unique au ho iden i fi e s. The di ff e ence be ween he ac i e au ho s and he au ho s' yea ly coun is ha he o me did no equi e au ho s o publish in each yea be ween hei fi s and las publica ion o be e- ga ded as an ac i e esea che in he fi eld. 3.3. Communi ies o esea che s To be e unde s and he g ow h o communi ies, indi idual ac o s we e no he only conside a ion in his s udy; co-au ho ship a bo h indi idual and o ganiza ional le els was conside ed. The WoS da a was sliced based on yea s and uploaded o VOS iewe so wa e ( an Eck e al., 2010). An analysis based on yea s enabled an in es iga ion o changes in communi y s uc u e h ough communi y o ma ion. Pa a- me e s used o analyze da a in he VOS iewe we e ull coun ing, in- cluding all publica ions, no expec a ion o a minimum numbe o ci- a ions o publica ions, and calcula ions o all au ho s. In he las s age, au ho s wi h no connec ions o he o he schola s o o ganiza ions wi hin he da ase we e excluded. The esul s om he VOS iewe we e impo ed o Gephi o ne wo k analysis. Fo each yea , basic ne wo k s a is ics we e calcula ed, which allowed o a deepe unde s anding o ne wo k g ow h. Communi ies we e also analyzed a he na ional and o ganiza ional le els. Publica ion eco ds we e connec ed o he communi ies' asso- cia ed coun ies using ull coun ing. Simila ly, he yea ly o ganiza- ional-le el ac i i ies we e calcula ed using he ull coun ing o iden- i fi ed o ganiza ion names. To unde s and whe he TENG esea ch communi ies we e g owing o becoming mo e dispe sed, he He fi ndahl – Hi schman Index (HHI) was calcula ed on a yea ly basis o bo h na ional and o ganiza ional le els. Finally, he de elopmen o TENG communi ies h oughou he ime se ies was analyzed a an o - ganiza ional le el. The modula i y algo i hm (Blondel e al., 2008a) embedded in Gephi was used o unco e TENG esea ch communi ies om he ull da a. Majo communi ies we e u he examined o de e mine he geo- g aphical and hema ic bounda ies o TENG esea ch communi ies. The geog aphical bounda ies o communi ies we e analyzed using Google's geocoding API. Each o ganiza ion was geocoded o acqui e hei la i- udinal and longi udinal in o ma ion. Then, he dis ances be ween he co-au ho ing au ho s o ganiza ions we e calcula ed. TENG esea ch communi ies we e sc u inized using opical concen a ion and physical dis ance measu emen s. Topical concen a ion was calcula ed based on dis ibu ion o au ho -assigned keywo ds using HHI wi hin each majo esea ch communi y. Physical dis ance was e alua ed as he a e age physical dis ance be ween communi ies. Finally, opical changes wi hin he whole TENG communi y we e e alua ed. Topical change was calcula ed by ex ac ing e ms om abs ac s on a yea ly basis. P io o ex ac ing e ms, common scien i fi c publica ion s opwo ds we e emo ed and n-g ams in he abs ac s we e me ged. Fo each e m ex ac ed, a del a alue was calcula ed as he di ff e ence o he e m appea ing a yea and + 1. This opical change alue was used o unde s and he hema ic changes wi hin he esea ch communi y. The impo an e ms om all majo communi ies we e quali a i ely compa ed o he o e all hema ic changes. 4. Resul s The absolu e olume o TENG esea ch publica ions has been g owing, and we can iden i y se e al eme gen ac o s (see Fig. 2). TENG has a clea in en ion da e and fi s publica ion da e in 2012, which pinpoin s he eme gence imewise. The analysis o he e ie ed WoS da a showed a s ong inc ease in publica ions. Publica ion olume had inc eased om app oxima ely 50 publica ions in 2012, he yea o fi s publica ions, o a high o 402 in 2017. This inc ease o 704 % in publica ion numbe s is he p oduc o esea ch up ake and is much highe han he o e all g ow h o scien i fi c publishing, which is ap- p oxima ely 5% pe yea (La sen and Ins, 2010). I also sugges s a clea pe sis ence, as he echnology has al eady been a ound o se e al yea s. 4.1. Quali a i e insigh s om he ques ionnai e Communi y c ea ion was con fi med ia a ques ionnai e sen o TENG esea che s in he beginning o June 2018. Du ing almos h ee weeks, 41 o 615 esea che s esponded o he ques ionnai e. The e- sul s de i ed om su ey analysis a e p esen ed in Table 1, and he con en o he ques ionnai e is p esen ed in Appendix A. Abou hal o he esponden s iden i fi ed hemsel es as senio scien is s (48.78%), Fig. 2. The yea ly dis ibu ion o TENG scien i fi c publica ions om 2012 h ough 2017. Table 1 The esul o ques ionnai e. Resea ch ime and du a ion Time pe iod (Yea ) Numbe o Numbe o esponden s No es esponden s (%) In ol ed in esea ch ac i i ies Min (1 – 5 yea s) 10 24% A e (5 – 15 yea s) 17 41% Max (mo e han 15) 14 34% In en ion o lea e TENG esea ch Yes 2 5 % The easons a e no epo ed. No 38 95% In ol ed in p epa ing TENG p ojec o Yes 34 82% u u e No 7 17% Responden 's cu en fi eld o esea ch Clus e name Numbe o Numbe o esponden s No es (Clus e numbe ) esponden s (%) 1 Ene gy ha es ing ma e ial 14 26% 2 Senso s, sel -powe ed senso s, senso 9 17% ne wo k analysis 3 T iboelec ic nanogene a o (TENG) 9 17% 3 O he fi elds 6 11% e.g. Vascula biology, p in ed de ice, ib a ion, synch o on adia ion echniques, Li-on ba e y, fl exible elec onics, au oma ed d i ing and ac i e sa e y sys em. 4 Nanogene a o s 4 7% 5 Mic o/nano elec omechanical sys ems 4 7% (MEMS/NEMS) 6 Piezoelec ic elec onics 3 6% 7 Physics 2 4% 9 G aphene 1 2% 10 Mechanical Enginee ing 1 2% 11 Ma e ial science 1 2% Mo i a ion ac o s o engage wi h TENG Clus e name Numbe o Numbe o esponden s No e esponden s (%) 1 Po en ial applica ion in u u e 26 60% e.g. Powe sou ce o LED ligh , elec onic de ices, mic o-senso s, wi eless senso ne wo ks, wea able display, a i fi cial elec onic skin, applica ion In e ne o hings (IoT) 2 No el y 6 14% 3 Pe sonal esea ch in e es 4 9% 4 P omising de elopmen end and cu en 4 9% pe o mance 5 Collabo a ion pu poses 2 5% e.g. Collabo a ion wi h speci fi c companies and colleagues wi hin he esea ch communi ies 6 O he easons 1 2% e.g. Engaged because o being in a fi eld o esea ch hema ically close o TENG. 5 which means hey ha e an independen esea ch-and-de elopmen posi ion in academia wi h signi fi can con ol o e esea ch opics. The emaining esponden s we e a mid-senio - (26.83%) o junio -le el (14.63%) posi ions wi h pa ial o no con ol o e hei esea ch opics. Fo indus y posi ion he esponden s did no ha e senio -le el e- sponden s, bu included 4.88% mid-le el and 2.44% junio -le el e- sponden s. Finally, he esponden s included 2.44% holding a eme i us posi ion. Almos all esponden s had a ffi lia ed hemsel es wi h one o mo e TENG- ela ed con e ences, annual summi s, o jou nals a icles. Based on he esponden s' answe s, he majo scien i fi c enues o he TENG esea ch communi y we e iden i fi ed as he Nanoene gy and Piezo onics In e na ional Con e ence, he Ma e ials Resea ch Socie y Con e ence, and he Nanoene gy and Nanosys ems In e na ional Con e ence. Al hough TENG echnology was in oduced in 2012, 41% o e- sponden s epo ed ha hey had esea ch ca ee s be ween 5 and 15 yea s in leng h, and 34% epo ed a esea ch ca ee o mo e han 15 yea s. Responden s had been engaged wi h esea ch o 12 yea s on a e age. Acco ding o Table 1, 95% o esponden s would con inue hei esea ch on TENG and s ay in he communi y. In addi ion, 82% o esponden s we e hen cu en ly wo king on o planning o p opose esea ch p ojec s wi h a ocus on TENGs. Responden s ac i e in TENG esea ch had di ff e en esea ch back- g ounds. Open-ended esponses we e labeled as 11 ca ego ies (see Table 1). I should be no ed ha each esponden could ha e been a - fi lia ed wi h mo e han one clus e . The majo i y o esponden s (26%) we e ac i e in he opics o ene gy ha es ing ma e ials. The second and hi d clus e s had a simila a e o a ffi lia ed esponden s: 17% each. The ollowing esea ch clus e s con aining less han 10% o esponses: na- nogene a o s, mic o/nano elec omechanical sys ems, piezoelec ic elec onics, physics, g aphene, mechanical enginee ing, and ma e ial science in gene al. Rega ding scien is s' main mo i a ions o join he TENG esea ch communi y, he answe s we e clus e ed in o six main ca ego ies (see he las sec ion o Table 1). “ The po en ial applica ions o TENG ech- nology in he u u e ” and/o “ TENG is a mul i-pu pose eme ging echnology ” a ac ed almos 60% o esponden s o conduc TENG esea ch. “ No el y cha ac e is ic o TENG echnology ” was he second mos impo an eason why esponden s (14%) decided o join he TENG esea ch communi y. Abou 9% o esponden s epo ed ha pe sonal esea ch in e es s mo i a ed hem o engage wi h TENG e- sea ch. Ano he 9% o esponden s iden i fi ed “ he apid de elopmen ” and “ he cu en high pe o mance ” o TENGs as he mo i a ion o pu suing TENG esea ch. “ Building esea ch ne wo k ” and “ Collabo a- ion wi h indus y ” we e he easons o only 5% o esponden . O e all, esponden s seemed o associa e hei TENG esea ch wi h esea ch hey had been conduc ing o a longe pe iod. This is appa en om he ac ha he majo i y o esponden s a ffi lia ed hemsel es wi h TENG esea ch o a pe iod longe han he echnology's in en ion da e. Resea che s we e engaging wi h TENGs mos ly due o he in insic mo i a ion (Lam, 2013) o applying TENGs as a mul ipu pose ech- nology. 4.2. The esea ch communi y analysis esul s Cen al o he no ion o Hol on (1962) was ha a new p omising fi eld would a ac esea che s o join ha fi eld. The idea u he de- eloped by Ay es (1969) claimed ha esea che s a e p one o exi a fi eld i i does no yield esul s. Fo TENGs, he esul s sugges a s ong upwa d end in he size o he esea ch communi y, as seen in Fig. 3. By he end o 2017, a esea ch communi y ha began wi h 200 au ho s in 2012 has g own o app oxima ely 1500 membe s by he end o 2017. The g ow h a e o new esea che s joining he fi eld is also signi fi can . While in he fi s h ee yea s, new publishing esea che s emained unde 300 membe s, by he end o 2015, nea ly 600 new membe s we e doing TENG esea ch. Au ho dynamics we e calcula ed h ough ou measu es: unique, ac i e, new, and lea ing au ho s (Fig. 3). Du ing he fi s yea o pub- lica ion, he fi eld al eady had 261 au ho s. This is signi fi can i we conside ha TENG was in en ed la e 2011 and fi s published in ea ly 2012. I sugges s a apid mig a ion o esea che s om o he fi elds ha we e hema ically close o TENG. This is also suppo ed by ou ques- ionnai e esul s. A he end o 2012, 191 au ho s le he esea ch a ea. By “ lea ing ” we e e o he au ho s who did no publish new esea ch h oughou he es o he ime se ies. This esul ed in 73% o au ho s who did no publish esea ch in any subsequen yea s. Since hen, he numbe o au ho s lea ing he a ea ha e emained ela i ely s able and much lowe han ha o new esea che s joining he fi eld. In addi ion o pa icipa ion g ow h in he fi eld, eme gence equi es some cohe ence. TENG esea ch is a highly coope a i e esea ch a ea. Co-au ho ship o TENG publica ions desc ibes p og ess o TENG com- muni y de elopmen . Fig. 4 shows he co-au ho ship changes h oughou he s udy's pe iod. As seen in he fi gu e, wo dis inc clus- e s o esea che s a e iden i fi able, bo h connec ed by a ew cen al au ho s bu sepa a ed by a numbe o esea che s who do no co-au ho b oadly. In he fi gu e, we can also clea ly iden i y he cen al ole o he in en o , see as he la ges o ange node. Complemen ing Fig. 4, he analysis o he yea ly ne wo k o ma ion o TENGs enables unde s anding o he a ea's g ow h. Ne wo k mea- su es a e shown in Table 2. The a e age deg ee, he a e age o all au ho connec ions wi h o he au ho s, has emained ela i ely s able. An au ho has, on a e age, ou o fi e co-au ho s in a gi en yea . Co- au ho ship is o en s udied on a pape le el, whe eas esul s he e ocus on he communi y a ound a esea che pe yea . The li e a u e shows ha pape -le el co-au ho ship is on a e age app oxima ely ou au ho s (Glänzel and Schube , 2004). In his con ex , TENG esea ch does no di ff e om o he scien i fi c endea o s. When he au ho coun inc eases he diame e o he ne wo k, he longes pa h in a ne wo k g ows as new esea che s join a he ends o he ne wo k, wi h limi ed coope a ion wi hin he communi y. The a e age pa h leng h also inc eases, which means no only one o wo esea che s a e a he ends o he ne wo k, bu he o e all communi y is becoming mo e spa se. Ne wo k diame e , he a io be ween au ho connec ions o all possible connec ions, also dec eases o suppo he no ion o a mo e spa se communi y. One cha ac e is ic o eme gence is global p esence (Ro olo e al., 2015). Al hough he esul s o his s udy indica e ha he communi y has g own in e ms o indi idual ac o s, hey ell li le o he commu- ni y's global g ow h. In Fig. 5, he global sp ead o TENG esea ch is e iden . The fi gu e shows ha while a communi y is g owing by he numbe o ac o s, i eally is only cen e ed on h ee coun ies: he USA, China, and Sou h Ko ea; all o he coun ies show only modes pub- lica ion coun s. The numbe o coun ies wi h a leas one TENG pub- lica ion has g own om se en in 2012 o 32 in 2017. This de elopmen is simila o he fi ndings in he eme gence o uel cell echnology (Suominen, 2014), whe e he numbe o coun ies g ew linea ly. In- e es ingly, i a h eshold o coun ies wi h a leas fi e publica ions is used, as in Suominen (2014), only wo coun ies me ha limi in 2012, g owing o 10 in 2017. A simila pa e n was seen in uel cell ech- nology. The connec ion be ween au ho s and coun ies iden i fi ed he spa se con ibu ions om all excep he co e coun ies. The majo i y o au- ho s in he da ase we e a ffi lia ed wi h an o ganiza ion based ei he in China, he USA, o Sou h Ko ea. In 2012, he numbe o publica ions om China inc eased om 11 in 2012 o 226 in 2017 and he USA inc eased om se en in 2012 o 120 in 2017. I should also be no ed ha some au ho s can ha e se e al a ffi lia ions, which we e whole coun ed o acc edi each men ioned coun y. No able inc eases in he numbe o publica ions ha e aken place in Sou h Ko ea. While Sou h Ko ea had jus h ee publica ions in 2013, in 2017 i s publica ion numbe had g own o 84. All o he coun ies emained a an ex emely Fig. 3. Size o he TENG esea ch communi y. The fi gu e shows all ac i e au ho s, unique au ho s yea ly, new au ho s, and au ho s lea ing. The ime se ies o au ho s lea ing concludes a he end o 2015, as he calcula ions we e based on an au ho con inuously publishing. low publica ion g ow h a e. Coun ies such as he Uni ed Kingdom and Ge many, which accoun o a signi fi can amoun o global scien i fi c p oduc ion, had less han 20 publica ions each. The HHI highligh ed he concen a ion o he scien i fi c communi y. On a na ional le el, TENG esea ch was signi fi can ly concen a ed. HHI alues had g own om 29% in 2012 o a high o 37% in 2013, and hen o 25% in 2017. Fo compa ison, he o e all concen a ion o scien i fi c esea ch is app oxima ely 10% (Veugele s, 2010). Ul ima ely, al hough he TENG esea ch communi y appea s o be global, i has ac ually been concen a ed in a small numbe o coun ies. A an o ganiza ional le el, he Chinese Academy o Sciences and he Geo gia Ins i u e o Technology a e he co e o ganiza ions in he fi eld. F om 2012 h ough 2017, hese wo o ganiza ions accoun ed o nea ly 30% o publica ions, o en wi h esea che s sha ing a ffi lia ions. Compa ing he wo la ges o ganiza ions wi h he es , i is no ewo hy ha he 34 nex -la ges o ganiza ions p oduced oughly he same amoun o publica ions as he wo la ges . Table 3 highligh s o gani- za ions wi h o e 20 publica ions, 2012 – 2017. Focusing on he eme gence cha ac e is ic o global p esence, he numbe o o ganiza ions had g own mo e d ama ically han has he numbe o coun ies wi h a signi fi can ole. F om he s a o 2012 o he end o 2017, he numbe o o ganiza ions had g own om 27 o 274, as seen in Table 4. Using he HHI o o ganiza ional au ho ship, TENG esea ch has no been a pa icula ly concen a ed esea ch communi y, especially when compa ing on a na ional le el. Table 4 shows ha he fi eld con inued o become mo e concen a ed om he beginning o 2012 o he end o 2015, when i began o diminish in concen a ion o he end o 2017. I is no ewo hy ha e en hough he wo la ges o ganiza ions ha e played a signi fi can ole, he inc ease in he numbe o o ganiza ions keeps he HHI alues small. The communi y o ma ion is isualized by co-au ho ship ne wo k on an o ganiza ional le el, as seen in Fig. 6. In he fi gu e, s ong links a e e iden be ween he Chinese Academy o Sciences and he Geo gia Ins i u e o Technology seen as he la ges g een nodes. Howe e , i is wo h men ioning he dual posi ion o Z.L. Wang as he cen al au ho in Fig. 4; Wang has led he TENG esea ch in bo h leading o ganiza ions. This connec ion migh o e emphasize he link be ween he o ganiza ions. The co-au ho ship ne wo k om 2012 h ough 2017 was used o e alua e he ypes o communi ies o med (as seen in Fig. 6). The communi ies we e clus e ed using he modula i y algo i hm (Blondel e al., 2008b). The analysis esul ed in 87 communi ies, among which only ou had o e 5% o he au ho s. The la ges o ganiza ional clus e (17.32 %) was cen e ed in Sou h Ko ea. The second la ges (16.23%) was cen e ed in he wo la ges o ganiza ions, complemen ed wi h a numbe o geog aphically spa sed Chinese o ganiza ions. The hi d la ges communi y (11.4 %) was a sp ead o cen al o ganiza ions, wi h Soochow Uni e si y con ibu ing a signi fi can po ion o he publica- ions. The ou h la ges clus e (7.68%) was a mix o No h Ame ican and Chinese o ganiza ions, such as Huazhong Uni e si y o Science & Technology and Uni e si y o To on o. In addi ion o he la ge com- muni ies eme ging, i is signi fi can o no e ha Fig. 6 shows a numbe o o ganiza ions no connec ed o he o e all communi y o TENG e- sea ch (in g ay). These o ganiza ions emained isola ed om 2012 h ough 2017. To be e unde s and he communi ies embedded in F h i g e. 6, physical dis ance and hema ic concen a ion o each o he ou la ges communi ies was calcula ed. As can be seen in Table 5, he hema ic concen a ion and physical dis ance had a modes ly nega i e co ela- ion ( = − 0.44, p < 0.05). The ela i ely low co ela ion did no allow o s ong conclusions, bu he able does clea ly demons a e ha in addi ion o he clus e o au ho s, new communi ies g ew om e- gionally bound spaces, such as a communi y ha has a high con- cen a ion o Sou h Ko ean o ganiza ions. Table 6 desc ibes he hema ic changes in TENG esea ch o e all. The mos impo an e ms a e cen e ed on he co e echnology ele- men s. Te ms such as “ TENG, ” “ iboelec ic, ” and “ de ice ” emain among he mos eme gen . The only signi fi can ly eme gen applica ion on he able is he eme gence o senso s and wea able applica ions. Conce ning di ff e en communi ies, he second communi y, on which mos TENG esea ch is cen e ed including he in en o o he echnology, he mos equen ly used e ms we e “ TENG ” o “ ene gy ha es ing ” . The e m occu ence sugges s ha his communi y has been ocused on he co e echnology. O he communi ies a ound he echnology ha e had di ff e en hema ic o ien a ions. The fi s com- muni y was hema ically concen a ed on impo an e ms such as “ sel - powe ed senso a ays ” and “ silk fi b oins. ” These e ms a e highligh ed as hey a e no p esen ed in he o he communi ies. The hi d com- muni y appea s o ha e been specialized h ough e ms such as “ sel - healing ” and “ TENGs ” . The ou h communi y was connec ed h ough e ms such as “ in i o ene gy ha es ing ” and “ a e ial pulse mon- i o ing, ” which did no appea in o he communi ies. In e es ingly, hese di ff e ences a e no isible in Table 6; hey a e much sub le . The selec ed e ms a e highligh ed as hey appea in a pa icula communi y Fig. 4. Co-au ho ship in TENGs, 2012 – 2017. Colo ep esen s clus e esul ing om an analysis done by VOS iewe . The ne wo k g aph is a ailable online a h p:// a hosuominen. fi /TENG/au ho / and he ela ed da a fi le a h p://a hosuominen. fi /TENG/au ho /au ho _TENG.gex . (Fo in e p e a ion o he e e ences o colo in his fi gu e legend, he eade is e e ed o he web e sion o his a icle.) Table 2 Ne wo k measu es o each yea o TENG co-au ho ship ne wo ks. Ne wo k measu e 2012 2013 2014 2015 2016 2017 Nodes 91 128 233 568 763 1094 Edges 365 579 1107 2792 3863 5680 A e age deg ee 4.011 4.523 4.751 4.915 5.063 5.192 Ne wo k diame e 5 4 5 7 9 9 G aph densi y 0.089 0.071 0.041 0.017 0.019 0.01 A g. pa h leng h 2.216 2.121 2.709 3.105 3.439 3.682 bu a e no isible in any o he majo communi y. 5. Discussion and conclusion In his pape , we s udied he au ho ship dynamics o a newly eme ging esea ch fi eld — TENG echnology. The aim was o fi nd he cha ac e is ics o esea ch communi y de elopmen . This is impo an because s udies analyzing echnological eme gence usually use e ms as a measu emen , while he heo e ical backg ound on eme gence would sugges a b oade an age poin (e.g. Ay es, 1969). While au ho s such as Kuhn (1970) ocused on he pa adigm shi , and mo e con empo a y s udies on echnological eme gence ha e ocused on he cha ac e is ics o a echnical en i y (e.g. Ro olo e al., 2015), a esea che 's decision o join an eme gen fi eld is cen al o i s eme gence and de elopmen . The e ha e ce ainly been s udies on esea che mo i a ions (Lam, 2013), bu he li e a u e on au ho s' decisions o join a new esea ch fi eld does no eally exis . In his s udy, we ound ha a no el disco e y quickly engaged e- sea che s o join ha disco e y's fi eld. Sp eading h ough he cen al ac o s, new schola s joined he esea ch on he pe iphe y o he au ho ne wo k. Wi hin six yea s, a s ong o ganiza ional ne wo k had shi, q ay, eh wang, dhaka , l wu, h lee, c zhang, q wang, h yang, z sun, m han, md li, y sun, l liu, h wu, n shi, m xia, y chen, meng, b yu, h jeon, sb zhang, hx peng, h zhang, x yu, b zhou, j wang, w chen, hzhang, j uddin, asmi kim, wg zhang, xs sun, x huang, l zhong, q hu, b zhang, h su, z zhong, j cheng, x song, y liu, w chung, gs han, jw miao, l seol, ml pa k, sj kim, b han, m kim, d li, w cho, iw choi, yk wang, b zhou, chen, x zheng, y wang, d cui, n lee, ps shao, j pa ida, k yang, y qin, y liu, j yu, y hu, y li, j cheng, l bai, s wang, s jiang, y yi, gu, l yang, w xiong, j liao, q zi, y wu, c du, z wei, x zou, h jing, q jin, l cheng, g wu, w bai, p ma, j wu, y kim, y chand asekha , a gu, gq zhang, b ma, m liu, z chen, l hu, w jiang, zhou, y li, x zhang, lm yang, b wu, z he, c zhang, z en, l zhu, y chen, y wang, x ang, w allu i, n deng, w yang, zw zhai, j jin, h pa k, j zhu, m zhang, k willande , m liu, y lee, cj he, x yang, pk yang, c yang, j luo, j guo, w li, p he, j chen, m p adel, kc peng, m huang, y li, wu, jm zhang, l pang, y yin, y wen, x du, x zhu, l wang, y wang, zl li, s niu, s zhou, ys zhang, sl han, c sa a anakuma , b li, q yu, js wang, z yeh, mh zheng, l xu, l deng, j li, hy shi, b jie, y xia, x xu, y jiang, q li, h wang, l zhou, han, y lu, c su, l xu, c kim, sj lin, y liang, q zhang, g yu, a liu, m liu, li, c gao, s liu, x liao, x wang, m xie, y liu, g chen, j you, z yan, x ding, w choi, d shi, h zheng, q an, x wang, c liu, c lin, zh meng, xs lee, sh ko, yh zhang, y zhong, x sun, j wang, n lu, s leng, q zhang, c yang, x xu, w liu, l zhang, w zhao, y xu, z chung, j xue, c kim, ds zhu, h pu, x zhu, j jin, y xi, y wang, j wang, q wen, z li, xh su, j lin, z saada nia, z li, z yu, zu, j lee, s zhou, x guo, h zhang, n xue, mu, x hu, c ian, h cao, x chen, s zhu, g dai, y han, cb zhao, z xue, x ouyang, h li, l su, y yao, y zheng, x kim, dy pan, c hassan, i lee, y ahmed, a an, kuang, sy lin, l kim, hsjung, jh kim, dh kim, s pa k, jy pa k, s pa k, c kim, h lee, kj lee, j jeong, ck kim, j kim, jb ha, j lee, mh kim, sw shin, hj yu, h kim, y kim, knseung, w kim, y nah, j jung, jy ye, bu baik, jm kwon, yh shin, sh chun, j gup a, mk kim, jw hinche , lee, jh kim, jh lee, jw lee, ky jeong, j hong, d