Structural effects due to the incorporation of Ar atoms in the lattice of ZrO2 thin films prepared by ion beam assisted deposition

S uc u al effec s due o he inco po a ion o A a oms in he la ice o Z O2 hin films p epa ed by ion beam assis ed deposi ion J.P. Holgado a , R. Escoba Galindo b , A. an Veen b , H. Schu b , J.Th.M. de Hosson c , A.R. Gonz aalez-Elipe a,* a Ins i u o de Ciencia de Ma e iales de Se illa (CSIC/Uni e sidad de Se illa) and Dp o. de Qu ıımica, Ino g aanica Uni . Se illa., A . Ame ico Vespucio S/N, E-41092 Se illa, Spain b In e acul y Reac o Ins i u e, De ec s in Ma e ials, Del Uni e si y o Technology, Mekelweg 15, 2629 JB Del , The Ne he lands c Ma e ials Science Cen e , Uni e si y o G oningen, Nijenbo gh 4, 9747 AG G oningen, The Ne he lands Recei ed 4 Oc obe 2001; ecei ed in e ised o m 27 Janua y 2002 Abs ac Two se s o Z O2 hin films ha e been p epa ed a oom empe a u e by ion beam induced chemical apou de- posi ion and subsequen ly annealed up o 1323 K. The wo se s o samples ha e been p epa ed by using ei he Oþ 2o mix u es o (Oþ 2þA þ) ions o he decomposi ion o a ola ile me allo ganic p ecu so o zi conium. The s uc u e and mic os uc u e o hese wo se s o samples ha e been de e mined by means o X- ay diff ac ion, Fou ie ans o m in a ed spec oscopy and posi on beam analysis (PBA). The samples we e e y compac and dense and had a e y low-su ace oughness. A e annealing in ai a TP573 K bo h se s o films we e anspa en and showed simila e ac ion indexes. Fo he (Oþ 2þA þ)-Z O2 hin films i is shown by X- ay pho oelec on spec oscopy and Ru he o d back sca e ing ha a ce ain amoun o inco po a ed A (5–6 a .%) emains inco po a ed wi hin he oxide la ice. No changes we e de ec ed in he amoun o inco po a ed A e en a e annealing a T¼773 K. Fo highe annealing empe a u es (T>1073 K), he amoun o A s a s o dec ease, and a T¼1223 K only esidual amoun s o A (<0.4%) emain wi hin he la ice. I has been ound ha as a as A a oms emain inco po a ed wi hin he Z O2ne wo k, he (Oþ 2 –A þ)-Z O2films p esen a cubic/ e agonal phase. When he amoun o ‘‘embedded’’ A dec eases, he c ys alline phase e e s o monoclinic, he majo i y phase obse ed o he (Oþ 2)-Z O2films a e any annealing ea men s. The mic os uc u e o he films a e diffe en annealing ea men s has been in es iga ed by PBA. The p esence o A ions and he ini ial amo phous s a e o he laye s we e de ec ed by his echnique. An inc ease o he open olume was obse ed a e annealing up o 773 K in bo h se s o samples. Fo highe annealing empe a u es he samples showed a p og essi e c ys allisa ion esul ing in a dec ease o he open olume. Du ing his sin e ing he samples wi hou embedded A p esen a highe concen a ion o open olume de ec s. A e he elease o A occu s (TP1223 K) bo h samples app oach o a simila de ec ee s a e. Nuclea Ins umen s and Me hods in Physics Resea ch B 194 (2002) 333–345 www.else ie .com/loca e/nimb * Co esponding au ho . Fax: +34-954-460-665. E-mail add ess: [email p o ec ed] (A.R. Gonz aalez-Elipe). 0168-583X/02/$ - see on ma e Ó2002 Else ie Science B.V. All igh s ese ed. PII: S0168-583X(02)00695-X The inco po a ion o A wi hin he Z O2 hin film s uc u e, is p oposed as he main ac o con ibu ing o he s abilisa ion o he cubic/ e agonal phase o Z O2a oom empe a u e. Ó2002 Else ie Science B.V. All igh s ese ed. Keywo ds: Chemical apou deposi ion; Ion bomba dmen ; Thin film; Posi on annihila ion; Phase ans o ma ion; Ion assis ed deposi ion; Z O2 1. In oduc ion Among many o he applica ions, such as op i- cal coa ings [1], buffe laye s o g owing supe - conduc o s [2], hea esis an coa ings [3], oxygen senso s [5], e c. ... , Z O2 hin films ha e a ac ed g ea a en ion because o hei p ope ies as ion conduc o s [4], o be used in he so-called solid oxide uel cells (SOFCs) [5]. Fo all hese appli- ca ions, and specially in he case o SOFCs, i is gene ally desi able o ob ain he e agonal o cubic phase o zi conia. I is known ha al hough he e agonal phase o Z O2is uns able a oom empe a u e, i can be s abilised by doping wi h ca ions such as Y3þ,Ca 2þ, e c. [6], o by dec easing he pa icle size o he c ys alline domains [7]. The syn hesis o Z O2o Z O2doped hin films (i.e. Z O2-M) has been a emp ed by many diffe en me hods including sp ay deposi ion [8], chemical apou deposi ion (CVD) [9], elec on e apo a ion [10], and o he me hods [11]. Ve y compac Z O2ðMÞ hin films can be p e- pa ed by ion beam assis ed deposi ion p ocedu es (IBAD) [12] ha in ol e he bomba dmen o he g owing film wi h low-ene gy ion beams (i.e. gene ally EP1000 eV), usually Oþ 2,A þo mix- u es o bo h. By hese me hods, i is e y common o obse e he appea ance o diffe en c ys allo- g aphic effec s, such as amo phisa ion o he p e - e en ial g ow h o some c ys allog aphic planes [13,14] ha a e a ibu ed o he influence o he bomba dmen wi h low-ene gy ions [15]. Ma in [16] has shown ha monoclinic o e agonal phase ans o ma ions can be induced in Z O2 hin films by bomba dmen wi h A þions o ela i ely low ene gy. Fo bulk Z O2 he effec o high-ene gy A þbomba dmen in inducing c ys allog aphic ans o ma ions and he s abilisa ion o he cubic/ e agonal phase is well known [17]. In mos cases an implici explana ion o his ype o monoclinic o cubic/ e agonal ans o ma ion elies on he assump ion ha ballis ic in e ac ions be ween he ene ge ic ions and he a ge a oms a e esponsi- ble o he phase change. The esul s p esen ed in he p esen pape do no exac ly suppo such an explana ion based on ene ge ic effec s, bu a- he emphasise he ole o A embedded wi hin he Z O2 hin film in p oducing he s abilisa ion o he cubic/ e agonal phase. We compa e he s uc u al e olu ion o Z O2 hin films p epa ed by ion beam induced chemical apou deposi ion (IBICVD) by using ei he Oþ 2o mixed Oþ 2þA þ ion beams. In he la e case he cubic/ e agonal phase is s abilised upon annealing, as long as A a oms emain inco po a ed wi hin he hin films. In a p e ious pape [18] we ha e s udied in de ail how he ex u e, mo phology and op ical p ope - ies o he esul ing films a e affec ed by he use o A þions. The fi s mo i a ion o he p esen pa- pe is hen o highligh how he p esence o in- co po a ed A may s abilise he cubic/ e agonal s uc u e o zi conia ac ing in a simila way han ‘‘doping’’ ca ions such as Y3þ,Ca 2þ, e c. and o p o ide an analysis o he e olu ion o s uc u al de ec s in he samples upon annealing, as de e - mined by posi on beam analysis (PBA). 2. Expe imen al Thin films ha e been p epa ed by IBICVD [19]. This me hod consis s o he decomposi ion o a ola ile me allic p ecu so by means o an ion beam. Fu he de ails abou he me hod as applied o he syn hesis o oxides and ni ides can be ound in p e ious publica ions [19–21]. Zi conium e a e -bu oxide, Z (Obu )4has been used as p ecu so . Typical pa ial p essu e 334 J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 du ing deposi ion was 4 105mba . Oþ 2o mix- u es o Oþ 2þA þions, accele a ed o 400 eV, we e supplied by an ‘‘IQP’’ b oad ion sou ce. O2þA mix u es wi h a mola a io 10:1 we e used o p epa a ion o A -inco po a ed samples. A beam cu en o a ound 40 lAcm 2was mea- su ed a he sample holde du ing deposi ion. The o al p essu e in he chambe du ing deposi ion was 5 104mba . Unde hese condi ions a g ow h a e o app oxima ely 2.0 nm min1was measu ed wi h a qua z c ys al moni o a he sample posi ion. Silicon (1 0 0) wa e s and used qua z ha e been used as subs a es o p epa a ion o he samples. X- ay diff ac ion (XRD) spec a we e eco ded by using he Cu Ka adia ion wi h a Siemens D5000 diff ac ome e . An incidence angle o 0.5° was chosen o inc ease he sensi i i y. IR spec a we e ob ained wi h a Nicole spec- ome e a 8 cm1 esolu ion o samples p epa ed on Si wa e s. The silicon subs a e was pa ially anspa en o in a ed adia ion showing a small ea u e a 609 cm1. Annealing o he samples in ai was ca ied ou in a con en ional a mosphe ic u nace, using a linea hea ing amp o 10 K min1. A he end o he amp he selec ed final empe a u e was kep o 3 h ollowed by cooling down slowly. Ru he o d back sca e ing (RBS) spec a we e eco ded wi h a andem ion accele a o (Pelle on 9SH-2 om NEC) a he Cen o Nacional de Acele ado es (CNA) in Se ille. Alpha pa icles o 2.021 MeV and usual se ings o sca e ing angle and de ec o geome y we e used o eco ding o spec a. The da a acquisi ion was done wi h a o al sample cha ge o 5 lC. Simula ions o composi- ion p ofiles ha e been done wi h he RUMP p og am code [22]. X- ay pho oelec on spec oscopy (XPS) was ca ied ou using an ESCALAB 210 spec ome e wo king in pass ene gy cons an mode wi h a alue o 50 eV. The su ace concen a ion o he ele- men s was de e mined by measu ing he peak a eas and co ec ing by he app op ia e sensi i i y alues. The PBA was pe o med using he Del a i- able ene gy posi on beam (VEP) [23]. The posi- ons we e injec ed in he samples wi h ene gies be ween 100 eV and 25 keV. Annihila ion o pos- i ons wi h elec ons esul s, in he cen e -o -mass sys em, in he emision o wo collinea pho ons o 511 keV each. In he labo a o y ame he non- ze o momen um o he posi on-elec on pai causes an ene gy shi o he pho ons. This esul s in a b oade ene gy dis ibu ion, called Dopple b oadening o he annihila ion adia ion (DBAR) a ound he 511 keV peak. This b oadened ene gy spec um is cha ac e ised by wo pa ame e s: S and W. The S(shape) pa ame e is defined as he a io be ween he cen al a ea and he o al a ea o he annihila ion pho opeak. This pa ame e e- flec s he annihila ion o posi ons wi h alence elec ons (low-momen um elec ons). In gene al, a high alue o Sindica es posi on annihila ion in open olume de ec s. I he ma e ial allows he o ma ion o a Posi onium s a e (Ps) his will con ibu e o he DBAR wi h a na ow componen because o he low-in insic momen o he pa a- posi onium. A second use ul pa ame e o he analysis o DBAR is he W(wing) pa ame e , which eflec s he posi on annihila ion wi h high- momen um elec ons (co e elec ons). The W- pa ame e is sensi i e o changes in he chemical en i onmen . Bo h pa ame e s can be combined in S–W maps wi h a hi d a iable (e.g. he implan- a ion ene gy, empe a u e o s ain) as a unning pa ame e . These maps a e use ul o ace posi on apping in, e.g., laye s as in he samples desc ibed in his wo k. The da a was analysed wi h he a iable ene gy posi on fi (VEPFIT) p og am [24]. The samples s udied he e a e modelled by a numbe o s acked slabs. Each o hese slabs is cha ac e ised by a hickness (di), Siand Wipa- ame e s, and a posi on diffusion leng h (Li). The measu ed SðEÞand WðEÞas a unc ion o he implan a ion ene gy E, a e a sum o he cha ac- e is ic Siand Wi alues o he apping laye s (clus e poin s) weigh ed by he ac ion o posi- ons apped in each laye iðEÞ. The spec a we e eco ded wi h a single Ge solid-s a e de ec o . All expe imen s we e ca ied ou a oom empe a u e unde a acuum o abou 106Pa. The alues o S and Wpa ame e s a e no malised wi h espec o he Sand W alues o a c ys al Si subs a e (SSi ¼0:58, WSi ¼0:031). J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 335 3. Resul s 3.1. Chemical composi ion The de e mina ion o he chemical composi ion o he zi conia films was done by XPS and RBS. The pho oelec on spec a o bo h se o samples, p epa ed by using Oþ 2o (Oþ 2þA þ) ions, we e ypical o a Z O2s oichiome y. Besides a ca bon peak, which is easily emo ed by bomba dmen wi h A þo Oþ 2(and hus p o ing ha i is due o con amina ion by exposu e o ai ), no o he con- aminan species we e de ec ed a he su ace o samples. In he case o (Oþ 2þA þ)-Z O2samples, A 2p and A 2s peaks could be de ec ed in he pho oelec on spec a o he films. Thei in ensi ies we e equi alen o an a omic pe cen age o ap- p oxima ely 5–6%, a alue ha did no change a e cleaning by spu e ing wi h Oþ 2ions o 3 keV ene gy. The samples we e also analysed by RBS o ob ain a mo e p ecise quan i a i e de e mina ion o he amoun o A p esen in he samples, as well as an e alua ion o he chemical composi ion p ofile o he whole film hickness. Fig. 1 shows expe imen al and heo e ically calcula ed spec a o he as p epa ed (Oþ 2þA þ)-Z O2and Oþ 2- Z O2samples. The igh o his figu e shows, in an enla ged scale, he A egion o some spec a co esponding o (Oþ 2þA þ)-Z O2samples an- nealed a inc easing empe a u es up o 1423 K. In he (Oþ 2)-Z O2and (Oþ 2 –A þ)-Z O2samples he O/ Z a io was close o 2.0, al hough a sligh ly smalle a io (i.e. ffi1.95–2.00) was ob ained o he as p epa ed (Oþ 2 –A þ)-Z O2 hin film. Fo hese samples, A (5–6 a .%) is ound o be dis ibu ed homogeneously o e he whole film hickness e en a e hei annealing a empe a u es as high as 873 K. A e annealing o he samples a T>773 K he concen a ion o A wi hin he film dec eases. I is ound ha a 1073 K only 3% o A is s ill e ained by he hin film, and ha a smalle con- cen a ion is obse ed a i s su ace. A e an- nealing a 1323 K, only a 0.5 a .% o A is p esen , wi h an e en lowe concen a ion a he su ace o he hin film. These esul s indica e ha annealing ea men s abo e 1073 K p o oke a p og essi e loss o he A inco po a ed wi hin he s uc u e o zi conia films. This loss is mo e p onounced om he opmos su ace laye s o he Z O2 hin films. Fig. 1. Expe imen al and simula ed RBS spec a o he (Oþ 2þA þ)-Z O2sample. The inse shows, in an enla ged scale, he egion co esponding o embedded A a e annealing ea men s a indica ed empe a u es. 336 J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 3.2. C ys allog aphic s uc u e Fig. 2 shows XRD pa e ns o (Oþ 2)-Z O2 hin films p epa ed a 298 K and a e annealing a inc easingly highe empe a u es up o 1323 K. Fo compa ison, he e e ence pa e ns o he mos in ense peaks o he diffe en phases o zi - conia a e included in he op pa o he figu e. The spec um o he as p epa ed (Oþ 2)-Z O2sample e eal he exis ence o a leas wo phases: mono- clinic and cubic and/o e agonal. I should be ema ked he e ha he diff ac og ams co e- sponding o hese wo phases p esen a se ies o peaks wi h e y simila in ensi y and wi h e y close posi ions. Due o he la ge wid h o he peaks ob ained in he diag am o he hin film and he possible displacemen o he posi ions o he peaks espec o hose o he pu e phases due o s ess and s ain effec s, e en a ca e ul analysis o he posi ions o he peaks do no pe mi o disc imi- na e be ween he cubic and/o he e agonal phase o zi conia in he hin film. Thus, in he ollow- ing discussion a men ion o cubic/ e agonal phase should always be aken as a possible exis- ence o any o hose phases o a mix u e o bo h. Annealing a T<773 K leads o a ela i e in- c ease o he monoclinic phase, al hough only a small change in he wid h o he peaks is obse ed in his ange o empe a u es. By con as , when he samples a e annealed a highe empe a u es (T>873 K), a d as ic dec ease is obse ed in he in ensi y o he peaks associa ed wi h he cubic/ e agonal phase. These peaks p ac ically disap- pea om he spec um when he film was an- nealed a TP1223 K. This annealing ea men induces an inc ease in he heigh and a dec ease in he wid h o he peaks co esponding o he monoclinic phase (Table 1). Fo he whole em- pe a u e ange o annealing, he ela i e in ensi y o he peaks o he monoclinic phase was a om he a io obse ed in a single c ys al, hus indi- ca ing he p e e en ial g ow h in he films o some c ys allog aphic planes (c . Table 1). XRD pa e ns o (Oþ 2þA þ)-Z O2films p e- pa ed a 298K and subsequen ly annealed a in- c easingly highe empe a u es up o 1323 K a e p esen ed in Fig. 3. The XRD pa e n o he ‘‘as deposi ed’’ (Oþ 2þA þ)-Z O2films depic s only wo b oad ea u es a diff ac ion angles which sugges he exis ence o a badly o de ed cubic o e agonal s uc u e o Z O2. Again, he wid h o he peaks makes i e y difficul o disc imina e be ween hese wo phases, and in he ollowing discussion i should always be kep in mind he possible exis- ence o ei he cubic/ e agonal phases o e en a mix u e o bo h. Al hough he posi ion o he di - ac ion lines is consis en wi h hose o he cubic/ e agonal phases, he obse ed in ensi ies a e a Fig. 2. XRD pa e ns o Oþ 2-Z O2 hin film o he as p epa ed samples and a e annealing ea men s a indica ed empe a- u es. The pa e ns o he mos in ense peaks o monoclinic (bo om), cubic ( op, ull line) and e agonal ( op, dashed line) phases o zi conia a e included o compa ison. J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 337 om e e ence pa e ns. In pa icula , he high in- ensi y o he peak a 2hffi59:4°, indica es a p e - e en ial g ow h o he film acco ding o he co esponding (3 1 1) planes o he cubic phase and/ o he (2 1 1) plane o he e agonal phase. The p e e en ial g ow h o planes wi h high Mille in- dices has been ecen ly epo ed by us o Fe2O3 hin films p epa ed also by IBICVD [25]. Such a p e e en ial g ow h was ela ed wi h he p esence o A embedded in he s uc u e. By annealing hese zi conia samples a TP573 K he diff ac ion pa e n changed. The maximum o he peak a 2h¼59:4°becomes sligh ly b oade wi h a small asymme y a highe diff ac ion an- gles. Simul aneously, a b oad ea u e appea s a 2h¼33–35°. The compa ison o hese s uc u es wi h he lines co esponding o he e e ence pa - e n o he bulk ma e ial, sugges s he exis ence o a badly o de ed cubic/ e agonal phase o zi conia in he as p epa ed samples and hei p og essi e o de ing o es uc u ing wi h he annealing ea men s. A significan change in he diff ac ion diag am o he (O2þA þ)-Z O2samples is obse ed a e annealing a T¼1073 K. In his case he peak co esponding o planes (3 1 1) is s ill he mos in ense, bu becomes na owe (c . Table 2). In his diag am peaks a 33–35°a e now clea ly isible, p obably co esponding o some diff ac ion planes o he cubic/ e agonal phase o zi conia and/o (2 0 0) o he monoclinic phase. The peaks o he e agonal/cubic phases o he zi conia a e no longe isible a e annealing a T¼1223 K, being subs i u ed by a clea pa e n co esponding o he monoclinic phase. The diff ac ion peaks o his phase a e also much na owe now (c . Table 2), ha ing a ull wid h a hal maximum (FWHM) simila o ha o he peaks o he (Oþ 2)-Z O2 hin films epo ed in Fig. 2 (c . Table 1). A e an- nealing a 1323 K, a diff ac ion diag am simila o ha depic ed by he Oþ 2-Z O2sample annealed a he same empe a u e was ob ained. Fou ie ans o m in a ed spec oscopy (FTIR) spec oscopy can be used as a complemen o XRD o a mo e accu a e desc ip ion o he s uc u e o Z O2 hin films. Al hough FTIR is no mally used on a ‘‘finge p in ’’ basis, a sys em- a ic analysis o IR spec a may also p o ide insigh in o he Z O2 hin film s uc u e [25–29]. Fig. 4 shows FTIR spec a o he (Oþ 2)-Z O2 sample as p epa ed and a e diffe en annealing ea men s up o 1323 K. Ini ially, he FTIR spec a o he as p epa ed Oþ 2-Z O2film is cha - ac e ised by a s uc u e whe e some b oad bands a e defined a 340, 407 and 490 cm1. These bands can be a ibu ed o some o he 15 ib a- ional modes o he monoclinic phase o Z O2[30]. By annealing a T6773 K, he bands become sligh ly sha pe and inc ease in in ensi y. This endency is e en clea e a e annealing TP1223 K, while some new bands cen ed a 258, 460 and 570 cm1de elop in he spec um. These bands can be also assigned o some o he 15 ib a ional modes o he monoclinic phase o Z O2[30]. The inc ease in in ensi y a TP1073K, he de elop- men o new bands and he smalle wid h o he obse ed bands abo e his empe a u e a e ac s which a e consis en wi h he obse ed inc ease in he c ys allini y o he monoclinic phase. This endency can be also deduced om he dec ease o Table 1 No malised in ensi y, e e ed o mos in ense peak o each phase and FWHM (in b acke s) o he mos significan peaks appea ing in XRD diag ams o Fig. 2 o (Oþ 2)-Z O2samples a e indica ed annealing ea men s Tempe a u e Monoclinic Cubic/ e agonal (1 1 0) (1 1 1) (1 1 1) (1 1 1) (2 2 0) (3 1 1) 300 K 18 (0.98) 18 (0.90) 100 (0.82) 100 (0.92) – – 573 K 19 (0.85) 45 (0.83) 100 (0.82) 100 (0.88) – – 773 K 21 (0.75) 58 (0.75) 100 (0.62) 100 (0.81) – – 1073 K 14 (0.72) 58 (0.57) 100 (0.39) 100 (0.40) – – 1223 K 18 (0.32) 21 (0.32) 100 (0.26) – – – 1323 K 25 (0.25) 63 (0.29) 100 (0.23) – – – Single c ys al 25 100 70 100 50 20 338 J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 he FWHM o he peaks o he XRD diag ams p esen ed in Fig. 2 (c . Table 1). Fig. 5 shows he FTIR spec a co esponding o a(O þ 2þA þ)-Z O2sample subjec ed o diffe en annealing ea men s. The as p epa ed sample p esen s only a wide s uc u e wi h a maximum a 399 cm1. A e annealing a T6773 K, his s uc u e becomes sligh ly sha pe and wo max- ima can be obse ed a 343 and 434 cm1. To ou knowledge ypical equencies o he ans e se ib a ional modes o he cubic o e agonal pha- ses o pu e Z O2a e no epo ed in li e a u e. The e o e, compa ison is made wi h he ib a- ional equencies o he cubic and e agonal phases o Y ia-s abilised Z O2[31] epo ed in Table 3. Assuming ha he p esence o y ia migh p oduce some shi s in he posi ion o bands, he spec a in Fig. 5 a e annealing a TP573 K could be consis en wi h he o ma ion o he e agonal phase. By con as , he b oad band a 399 cm1 ound in he as p epa ed sample, could be a ibu ed o a badly o de ed cubic and/o e agonal phase. The b oad shapes o all hese spec a make i difficul o disca d some mino i y con ibu ion o a cubic phase a e an- nealing a 573 o 773 K. The FTIR spec a o he (Oþ 2þA þ)-Z O2samples annealed a TP1073 K unde go d as ic changes. The shape o spec a is be e defined, and shows bands cen ed a 330, 405 and 490 cm1. The posi ion o hese new bands is simila o hose ha we e asc ibed o he monoclinic phase o zi conia obse ed in he (Oþ 2)- Z O2samples. Howe e , he supe posi ion o he b oad IR s uc u e o he e agonal phase canno be disca ded in hese spec a. I is in e es ing ha no clea e idence o he p esence o he monoclinic phase can be deduced om he XRD pa e n only o he (Oþ 2þA þ)-Z O2sample annealed a his empe a u e (c . Fig. 2). The diffe en long ange o de sensi i i y o XRD and IR is likely he ea- son o he diffe en in o ma ion p o ided by hese wo echniques and indica e he ad an age o using bo h me hods o a p ope e alua ion o he s uc u e o some hin films. A e annealing a T¼1323 K, he bands inc ease in in ensi y and become be e defined, while some o he bands a 258, 340, 407, 460, 490 and 570 cm1become also isible. A his final empe a u e, he spec um is almos iden ical o ha ob ained o he (Oþ 2)- Z O2sample annealed a he same empe a u e. 3.3. Posi on annihila ion expe imen s The S-pa ame e e sus implan a ion ene gy in he as p epa ed samples o (Oþ 2þA þ)-Z O2and (Oþ 2)-Z O2se ies is plo ed in Fig. 6(a). Bo h samples p esen almos iden ical a ia ion o his Fig. 3. XRD pa e ns o (Oþ 2þA þ)-Z O2 hin film o he as p epa ed samples and a e annealing ea men s a indica ed empe a u es. The pa e ns o he mos in ense peaks o monoclinic (bo om), cubic ( op, ull line) and e agonal ( op, dashed line) phases o zi conia a e included o compa ison. J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 339 pa ame e , indica ing ha he ac ion o open olume in he samples is e y simila . By con as , he a ia ion o he W-pa ame e , p esen ed in Fig. 6(b), shows some diffe ences be ween he as p epa ed samples o (Oþ 2þA þ)-Z O2and (Oþ 2)- Z O2. In p inciple, his change can be asc ibed o he posi on annihila ion in he icini y o a gon ions, as hey a e hea ie han oxygen ions and, acco dingly, posi on annihila ion wi h i s co e elec ons would be enhanced. Fig. 4. FTIR spec a o he (Oþ 2)-Z O2 hin film o he as p epa ed sample and hose ob ained a e annealing ea men s a indica ed empe a u es. Fig. 5. FTIR spec a o he (A þþOþ 2)-Z O2 hin film o he as p epa ed sample and hose ob ained a e annealing ea - men s a indica ed empe a u es. Table 2 No malised in ensi y, e e ed o mos in ense peak o each phase and FWHM (in b acke s) o he mos significan peaks appea ing in XRD diag ams o Fig. 3 o (A þþOþ 2)-Z O2samples a e indica ed annealing ea men s Tempe a u e Monoclinic Cubic/ e agonal (1 1 0) (1 1 1) (1 1 1) (1 1 1) (2 2 0) (3 1 1) 300K–––––100(2.56) 573K–––––100(2.56) 773K–––––100(2.10) 1073 K –––––100(1.22) 1223 K 18 (0.29) 21 (0.32) 100 (0.26) – – – 1323 K 25 (0.23) 64 (0.29) 100 (0.23) – – – Single c ys al 25 100 70 100 50 20 340 J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 Al hough he a ia ion o Sand Wpa ame e s eflec s he p ope ies o he samples, a simul a- neous ep esen a ion o hese wo pa ame e s in S–W plo is usually p e e ed o a clea e desc ip- ion o he e olu ion o he hin film p ope ies. Fig. 7 p esen s he S–W map o he as p epa ed (Oþ 2þA þ)-Z O2and (Oþ 2)-Z O2samples. In his kind o plo s, in o ma ion is ob ained o he di - e en laye s o which he sample is composed. In he cu en s udy a h ee-laye sys em (su ace/ zi conia laye /silicon subs a e) was applied. As expec ed, he S–W clus e poin s o he su ace and he silicon subs a e a e iden ical in bo h samples bu diffe en in hose o he zi conia laye . Thus, changes in he clus e poin s o he zi conia laye will eflec changes in he s uc u e o he film. The a ia ion o he fi ed S–W poin s co esponding o he zi conia laye o samples o (Oþ 2þA þ)- Z O2and (Oþ 2)-Z O2a e annealing ea men s a diffe en empe a u es is shown in Fig. 8. In Fig. 9 he fi ed Sand Wpa ame e s o he zi conia laye a e plo ed independen ly e sus he annealing empe a u e. In bo h figu es o he diffe en se s o samples wo annealing s ages can be conside ed: annealing om oom empe a u e o 773 K and annealing abo e ha empe a u e. In he fi s case, he S–W cha ac e is ic poin s mo e o highe S alues while he Wpa ame e dec eases. Changes in he posi on diffusion leng h we e also obse ed. In ac , o he (Oþ 2)-Z O2samples, he posi on diffusion leng h inc eases om 7.5 nm in he as p epa ed sample, o 50 nm a e annealing a 773 K. Fo he as p epa ed (Oþ 2þA þ)-Z O2sample a sho e posi on diffusion leng h was measu ed (3 nm) and i emains sho e a e annealing a Table 3 Some o he IR ib a ional modes o e agonal and Y ia s abilised cubic Z O2( aken om [30]) Phase mT(cm1) Te agonal 467 339 Y ia-s abilised cubic 358 Fig. 6. Va ia ion e sus ene gy o implan a ion in he as p e- pa ed sample o (Oþ 2þA þ)-Z O2and (Oþ 2)-Z O2 hin films (a) o he S-pa ame e (b) o he W-pa ame e . The lines indica e he fi ed alues ob ained om VEPFIT. Fig. 7. S–W plo o he as p epa ed (Oþ 2þA þ)-Z O2and (Oþ 2)-Z O2as p epa ed, indica ing he h ee-laye sys em (su - ace/zi conia laye /silicon subs a e). The a ows indica e he posi on implan a ion dep h. J.P. Holgado e al. / Nucl. Ins . and Me h. in Phys. Res. B 194 (2002) 333–345 341