scieee Open visual document viewer

Quantification of the H Content in Diamondlike Carbon and Polymeric Thin Films by Reflection Electron Energy Loss Spectroscopy

Yubero, Francisco; Rico, Víctor; Espinós Manzano, Juan Pedro; Cotrino Bautista, José; Rodríguez González-Elipe, Agustín

Abstract

A nondestructive method to determine the hydrogen content at the surface of diamondlike carbon and polymeric thin films is proposed. The method relies on the analysis of the elastic peak produced by backscattering of electrons from the hydrogen atoms present at the sample surface. Quantitative analysis of the H content at the surface is achieved through use of a phenomenological sensitivity factor for elastic electron backscattering by H atoms with respect to other atoms present at the surface of reference polymeric samples. The validity of the method is checked with elastic recoil detection measurements and infrared spectroscopy analysis of the same samples. The accuracy of the method in the determination of H content at the sample surface is estimated to be ±10%.

Full text

Quan i ica ion o he H con en in diamondlike ca bon and polyme ic hin ilms by e lec ion elec on ene gy loss spec oscopy F. Yube o,a兲V. J. Rico, J. P. Espinós, J. Co ino, and A. R. González-Elipe ICMSE (CSIC-USE) Amé ico Vespucio s/n, E-41092 Se illa, Spain 共Recei ed 24 Ma ch 2005; accep ed 28 June 2005; published online 15 Augus 2005兲 A nondes uc i e me hod o de e mine he hyd ogen con en a he su ace o diamondlike ca bon and polyme ic hin ilms is p oposed. The me hod elies on he analysis o he elas ic peak p oduced by backsca e ing o elec ons om he hyd ogen a oms p esen a he sample su ace. Quan i a i e analysis o he H con en a he su ace is achie ed h ough use o a phenomenological sensi i i y ac o o elas ic elec on backsca e ing by H a oms wi h espec o o he a oms p esen a he su ace o e e ence polyme ic samples. The alidi y o he me hod is checked wi h elas ic ecoil de ec ion measu emen s and in a ed spec oscopy analysis o he same samples. The accu acy o he me hod in he de e mina ion o H con en a he sample su ace is es ima ed o be ±10%. © 2005 Ame ican Ins i u e o Physics.关DOI: 10.1063/1.2011786兴 Quan i ica ion o he hyd ogen con en in diamondlike ca bon 共DLC兲and polyme ic hin ilms is o he mos impo - ance because hei p ope ies a e igh ly linked o his con en .1Typically he hyd ogen con en in his ype o ma- e ials has been de e mined wi h echniques such as elas ic ecoil de ec ion analysis 共ERDA兲, nuclea eac ion analysis 共NRA兲using he 1H共15N, ␣ 兲12C nuclea eac ion, nuclea magne ic esonance 共NMR兲, he mal hyd ogen e olu ion, o in a ed spec oscopy 共IR兲. Conscious o he g ea demand o al e na i e and simple me hods,1–3 we ha e de eloped a p o- cedu e o quan i y he hyd ogen con en a he su ace o DLC and polyme ic hin ilms. As i will be shown in his pape , he H con en wi hin he su ace egion o hese ype o ma e ials can be de e mined by looking a he elas ically backsca e ed elec ons wi h kine ic ene gies in he o de o 1 o 2 keV as in s anda d e lec ion elec on ene gy-loss spec- oscopy 共REELS兲analysis. The basis o he quan i ica ion is he same as ha o Ru he o d backsca e ing spec oscopy 共RBS兲, wi h he di e ence ha in RBS he p obe pa icles a e usually alpha pa icles and in ou me hod elec ons 共no e he di e ence o mass be ween hese wo p ojec iles兲. In his espec , i mus be men ioned ha e y ecen ly O osz e al.4 ha e epo ed ha H a oms in polye hylene can be de ec ed by his echnique. I is also wo h men ioning ha REELS was used in he pas o de ec H induced chemical shi s in he plasmon exci a ions.5 The p esence o H can be easily de ec ed by s anda d REELS measu emen s wi h 1 o 2 keV elec on beams. The physical basis o he de ec ion o H a oms a he su ace o ma e ials con aining ligh elemen s is he expec ed di e ence in he ecoil ene gy ⌬E 6in he in e ac ion o a as elec on wi h a s a ic nucleus, ei he hyd ogen o ano he ype o a om, which can be es ima ed by he exp ession ⌬E =4me MH E0sin2 ␪ 2 冉 1− 1 A 冊 ,共1兲 whe e me,MH, and Aa e he elec on mass, he a omic weigh o he H a om and he o he nucleus conside ed, e- spec i ely, ␪ is he sca e ing angle, and E0is he p ima y ene gy o he incoming elec ons. This exp ession p edic s ha , o a ypical sca e ing angle o ␪ =120° and a p ima y ene gy o 1500 eV, he di e ence in ecoil ene gies be ween H and o he a oms hea ie han C is ⬃2.1 eV, so ha we migh easily dis inguish be ween H and hese o he a oms p esen in he sample wi h he s anda d ene gy esolu ion o REELS measu emen s 共i.e., ew en hs o eV兲. Figu e 1 shows he REELS spec a, aken wi h a p ima y elec on beam o 1500 eV, o se e al e e ence samples 关polyca bona e 共PC, -C16H14O3-兲, polys y ene 共PS, -C8H8-兲, polyme hyl-me hac yla e 共PMMA, -C5H8O2-兲, and poly e - a luo e hylene 共PTFE, -CF2-兲and highly o ien ed py oli ic g aphi e 共HOPG, C兲兴. The REELS measu emen s we e pe - o med wi h special ca e o minimize elec on damage a he su ace o he samples. Thus, ypically he measu ing ime was o ⬃1 min. wi h elec on beam cu en s below 1 ␮ A/cm2.7The angle o incidence o he elec on beam wi h espec o he su ace no mal was 60° 共i.e., ␪ =120°兲and he elec on de ec ion was se no mal o he sample su ace. In he elas ic egion o hese spec a 共be ween ⬃1497 and 1502 eV兲 he H signal can be clea ly iden i ied as a weak peak a 2.1 eV lowe kine ic ene gy han he mo e in ense ea u e in he spec a a 1500 eV. This la e peak is due o he elec ons backsca e ed by he a oms o he han H p esen a he su ace o he samples. No e ha he small peak a 2.1 eV is only p esen in he H con aining polyme s 共i.e., PS, PMMA, and PC兲and is no de ec ed in he case o PTFE o HOPG, as expec ed. This peak shi s o 3.0 eV lowe kine ic ene gy han he main peak when he exci a ion ene gy is changed o 2000 eV 共no shown兲. This esul con i ms i s assignmen as elec ons backsca e ed by he H a oms in he samples. A his poin i is wo h men ioning ha using p i- ma y beam ene gies la ge han ⬃1800 eV has he e ec o inc easing he backg ound be ween he wo con ibu ions o he elas ic peak p obably due o mul iple elas ic sca e ing and damage. On he o he hand, he wo con ibu ions o he elas ic peak do no sepa a e enough o acili a e quan i ica- ion when p ima y elec on ene gies smalle han ⬃1200 eV a e conside ed. Besides he small ea u e a ⬃2.1 eV om he main elas ic peak o E0=1500 eV, hese spec a a e cha ac e ized by he ypical ␲ plasmon ea u es a ⬃6eV ene gy loss, co esponding o he benzoic ings in PC and a兲Au ho o whom co espondence should be add essed; elec onic mail: [email p o ec ed] APPLIED PHYSICS LETTERS 87, 084101 共2005兲 0003-6951/2005/87共8兲/084101/3/$22.50 © 2005 Ame ican Ins i u e o Physics87, 084101-1 12 June 2025 14:52:16 PS, and o he sp2cha ac e o g aphi e. Besides, he co e- sponding ␴ + ␲ bulk plasmons a e indica ed o each ma e- ial. All hese e e ence samples we e also analyzed by XPS o con i m he le el o su ace con amina ion. I was ound ha a 5%–10% o O con amina ion was p esen a he su - ace o polyme s ha do no con ain oxygen in hei s uc- u es. A simila de ia ion om he heo e ical oxygen con en was ound in hose ma e ials such as PC o PMMA ha con ain his elemen in hei s uc u es. F om he ela i e in ensi y o he H peak wi h espec o he main ea u e o he spec a, i is possible o es ima e a phenomenological sensi i i y ac o ␴ exp ha , in he o m o a calib a ion cu e, se es o co ec o he di e en c oss sec ions o elas ic elec on backsca e ing by H o o he a - oms 共C and, o a mino ex en , O a oms兲p esen a he su - ace o he polyme s. The inse in Fig. 1 shows he co ela- ion be ween he in ensi y a io IH/Ino_H, and he nominal a io be ween H and o he ype o a oms 共C and o O兲p esen in he polyme s conside ed in his s udy. He e IHis he a ea o he H signal in he back e lec ed elec ons and Ino_His he a ea o he elas ic peak coming om a oms o he han H. I can be obse ed ha he e exis a linea co ela ion wi hin he expe imen al unce ain ies be ween he nominal a omic a io and he expe imen al in ensi y a io. In his analysis, we did no a emp quan i ica ion o high-H con en polyme s such as polyp opylene 共-C3H6-兲o polye hylene 共-CH2-兲be- cause hey a e known o deg ade easily wi h elease o hy- d ogen when hey a e bomba ded wi h low-ene gy elec ons.7By pe o ming a linea eg ession passing h ough he o igin o coo dina es wi h he o he poin s in he g aph, he slope o he i ed line is 37. This alue can be conside ed as a phenomenological ela i e c oss sec ion ␴ exp be ween he wo con ibu ions o he elas ic peak signal. No e ha he alue ob ained o ␴ exp is in easonable ag eemen wi h he p edic ions o he NIST da abase o elas ic sca e ing c oss sec ions8共neglec ing he p esence o oxygen a oms a he su ace and mul iple elas ic sca e ing, ha may be signi i- can in his case9兲. A ough heo e ical es ima ion o ␴ exp is 42, ob ained as he a io o he di e en ial elas ic sca e ing c oss sec ion wi h espec o he solid angle be ween C and H o 120° sca e ing angle and 1500 eV elec on ene gy.8In he ollowing we will used ou phenomenological ␴ exp=37 o quan i y he H con en in DLC samples. Thus, he H quan i- ica ion a he su ace o unknown samples by REELS can be pe o med acco ding o he exp ession %H= 100 IH ␴ exp IH ␴ exp +Ino_H ,共2兲 whe e he meaning o he di e en magni udes has been in- oduced p e iously. To check he alidi y o his p ocedu e, a i s compa a- i e s udy was ca ied ou by ERDA/RBS and REELS o a se ies o DLC samples p epa ed wi h ixed bias ol age and di e en con en o H in he plasma gas. The DLC hin ilms we e p epa ed by plasma Enhanced Chemical Vapo Deposi- ion in a plasma eac o wi h a pla e con igu a ion. Mix- u es o C2H2, A , and H2we e used as he plasma gas, while a sel -induced bias ol age was applied o he pla e elec ode con aining he subs a e o he samples. I was ound ha he pe cen age o H a oms in he DLC ilms measu ed by bo h echniques 共no shown兲was he same wi hin 10%, so ha , wi h he unce ain y limi s o hese echniques, hey yield simila alues o he H con en o he ilm. This p o es ha REELS can be an al e na i e echnique o he de e mina ion o H con en a he su ace in DLC ilms. To check u he he alidi y o he me hod, he hyd o- gen con en has been also de e mined o a se ies o DLC hin ilms p epa ed wi h a s anda d plasma composi ion by a ying he bias ol age applied o he subs a e pla e. Bias ol age is ecognized as a c i ical pa ame e o he con ol o he ha dness and o he p ope ies o DLC hin ilms.10 Fo his se ies o samples, he pe cen age o hyd ogen in he ilms has been plo ed in Fig. 2 agains he bias ol age. Fo compa ison, his igu e also includes he e olu ion o he in ensi y o he 2750–3050 cm−1 IR band, co esponding o he s e ching C-H ib a ions, no malized o he ilm hick- ness as de e mined by x- ay luo escence spec oscopy in a scanning elec on mic oscope. I is in e es ing ha , wi hin he e o ba s in bo h p epa a ion and quan i ica ion p o o- FIG. 1. REELS spec a measu ed wi h 1500 eV kine ic ene gy o PTFE, g aphi e 共G兲, PC, PS, and PMMA. Inse : Co ela ion be ween he heo e ical H con en in he e e ence polyme samples and he a io IH/Ino_Has de e - mined by REELS measu emen s. FIG. 2. Co ela ion be ween he no malized in ensi y o he in a ed C-H s e ching modes, accele a ion ol age du ing p epa a ion and he a omic pe cen age o H in he ilm as de e mined by quan i ica ion o he elas ic signal in REELS da a. Full ci cles: In a ed quan i ica ion; open squa es: REELS quan i ica ion. 084101-2 Yube o e al. Appl. Phys. Le . 87, 084101 共2005兲 12 June 2025 14:52:16 cols, hese wo magni udes ollow simila endencies. This e olu ion is cha ac e ized by a p og essi e dec ease in he hyd ogen con en as he bias ol age inc eases. This co ela- ion shows ha , despi e he su ace cha ac e o REELS, i s quan i ica ion da a o hyd ogen can be aken as oughly ep- esen a i e o bulk composi ion. To conclude, he p esen esul s ha e shown ha REELS may p o ide a way o quan i ying he amoun o hyd ogen p esen a he su ace o DLC o polyme ic ma e ials whe e C and H a e he majo i y cons i uen elemen s. The su ace cha ac e o REELS is con olled by he mean ee pa h o elec ons ha a el h ough he analyzed ma e ials ha , o he usual elec on ene gies u ilized in his echnique, is ⬃3 nm. We wan o emphasize ha his analysis p ocedu e measu es di ec ly he amoun o hyd ogen by a su ace sen- si i e echnique. I could hus be use ul o combine REELS wi h XPS o su ace analysis since XPS does no di ec ly de ec H. We hank he Spanish Minis y o Science and Educa ion 共MAT2004-01558兲 o inancial suppo . 1J. Robe son, Ma e . Sci. Eng., R. 37, 129 共2002兲. 2W. Jacob and M. Unge , Appl. Phys. Le . 68, 475 共1996兲. 3J. Ris ein, R. T. S ie , L. Ley, W. Beye , J. Appl. Phys. 84, 3836 共1998兲. 4G. T. O osz, G. Ge gely, M. Menyha d, J. To h, D. Va ga, B. Lesiak, A. Jablonski, Su . Sci. 566-568, 544 共2004兲. 5G. Ge gely, P og. Su . Sci. 71,31共2002兲. 6A. J. Be olo, J. Vac. Sci. Technol. A 3, 1312 共1985兲,and e e ences he e in. 7D. R. Bae , D. J. Gaspa , M. H. Engelha d, A. S. Lea, Beam E ec s Du ing AES and XPS Analysis in Su ace Analysis by Auge and X- ay Pho oelec on Spec oscopy, edi ed by D. B iggs and J. T. G an 共IM Pub- lica ions and Su ace Spec a Limi ed, UK, 2003兲,Chap.9. 8A. Jablonski, F. Sal a , C. J. Powell, NIST Elec on Sca e ing C oss- sec ion da abase- Ve . 3.0. Na ional Ins i u e o S anda ds and Technology, Gai he sbu g, MD 共2002兲. 9A. Jablonski and C. J. Powell, Su . Sci. 551, 106 共2004兲. 10A. R. González-Elipe, F. Yube o, and J. M. Sanz, Low Ene gy Ion Assis ed Film G ow h 共Impe ial College P ess, UK, 2003兲,Chap.5. 084101-3 Yube o e al. Appl. Phys. Le . 87, 084101 共2005兲 12 June 2025 14:52:16