Enhanced Power Conversion Efficiency of Perovskite Solar Cells with an Up- Conversion Material of Er 3+ -Yb 3+ -Li + Tri-doped TiO 2

Transcription

1 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 NANO EXPRESS Enhnced Power Conversion Efficiency of Perovskite Solr Cells with n Up- Conversion Mteril of Er 3+ -Y 3+ -Li + Tri-doped TiO 2 Zhenlong Zhng 1, Jinqing Qin 1, Wenji Shi 1, Ynyn Liu 1, Yn Zhng 1, Yuefeng Liu 1, Huiping Go 1 nd Ynli Mo 1,2* Open Access Astrct In this pper, Er 3+ -Y 3+ -Li + tri-doped TiO 2 (UC-TiO 2 ) ws prepred y n ddition of Li + to Er 3+ -Y 3+ co-doped TiO 2. The UC-TiO 2 presented n enhnced up-conversion emission compred with Er 3+ -Y 3+ co-doped TiO 2. The UC-TiO 2 ws pplied to the perovskite solr cells. The power conversion efficiency (PCE) of the solr cells without UC-TiO 2 ws 14.0%, while the PCE of the solr cells with UC-TiO 2 ws incresed to 16.5%, which presented n increse of 19%. The results suggested tht UC-TiO 2 is n effective up-conversion mteril. And this study provided route to expnd the spectrl sorption of perovskite solr cells from visile light to ner-infrred using up-conversion mterils. Keywords: Enhnced power conversion efficiency, Perovskite solr cells, Up-conversion mteril Bckground Orgnoled hlide perovskite solr cells (PSCs) hve ecome ttrctive in the solr cell field, which is due to their dvntges, such s high efficiency, lost cost, nd simple friction [1 4]. In few yers, the power conversion efficiency (PCE) of PSCs hs een improved to 22.1% [5]. However, perovskite solr cells only sor smll frction of incident light in UV nd visile rnges due to the nrrow energy nd gp of perovskite sensitizer; thus, lrge portion of incident light is lost due to its non-sorption of ner-infrred (NIR) [6, 7]. One promising method to solve the NIR energy loss issue is to pply up-conversion mterils to PSCs, which cn convert NIR to visile light. Some uthors hve reported the pplictions of up-conversion mterils to perovskite solr cells [8 10], in which the up-conversion mterils dopted were minly sed on et-phse sodium yttrium fluoride (β-nyf 4 ). While the β-nyf 4 up- * Correspondence: ylmo1@163.com 1 School of Physics nd Electronics, Henn University, Kifeng , Chin 2 Institute of Micro/Nno Photonic Mterils nd Applictions, Henn University, Kifeng , Chin conversion mterils cn reduce chrge trnsport ility of electron trnsfer lyer [11]. It hs een reported tht Er 3+ -Y 3+ -F tri-doped TiO 2 cn improve the PCE of dye-sensitized solr cells (DSSCs) due to its enhnced up-conversion emission compred with Er 3+ -Y 3+ codoped TiO 2 [12]. In our previous puliction [13], we reported the ppliction of Er 3+ -Y 3+ co-doped TiO 2 nnorods to PSCs. Some reserchers hve proved tht the ddition of Li + into Er 3+ -Y 3+ co-doped TiO 2 could increse the up-conversion emission [14, 15]. And it hs een reported tht the perovskite solr cells sed on Lidoped TiO 2 produce higher performnces compred to the device sed on un-doped TiO 2 [16]. Thus, we wonder if the up-conversion mterils of Er 3+ -Y 3+ -Li + tridoped TiO 2 cn e pplied to PSCs to further improve the performnce. Therefore, in the present study, we prepred Er 3+ -Y 3 + -Li + tri-doped TiO 2 (UC-TiO 2 ) y ddition of Li + into Er 3+ -Y 3+ co-doped TiO 2, which presented n enhnced up-conversion emission compred with Er 3+ -Y 3+ codoped TiO 2. The UC-TiO 2 ws pplied to perovskite solr cells. The PCE of the solr cells with UC-TiO 2 is The Author(s) Open Access This rticle is distriuted under the terms of the Cretive Commons Attriution 4.0 Interntionl License ( which permits unrestricted use, distriution, nd reproduction in ny medium, provided you give pproprite credit to the originl uthor(s) nd the source, provide link to the Cretive Commons license, nd indicte if chnges were mde.

2 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 2 of 8 Fig. 1 Up-conversion emission spectr of Er 3+ -Y 3+ -Li + tri-doped TiO 2 (Er:Y:Li:Ti = 0.5:10:x:100, x = 0, 10, 15, 20, nd 25, molr rtio). Schemtic energy-trnsfer mechnisms of the Er 3+ ions incresed to 16.5 from 14.0% for the solr cells without UC-TiO 2, which presents n increse of 19%. Methods Synthesis of Er 3+ -Y 3+ -Li + Tri-Doped TiO 2 The nnocrystls of Er 3+ -Y 3+ -Li + tri-doped TiO 2 (UC-TiO 2 ) were synthesized y modified method [15]. A titnium (IV) n-utoxide ws prepred y mixing n-utyl titnte (Ti(OBu) 4 ) with cetylcetone (AcAc) t room temperture for 1 h under gittion. Then, the iso-propyl (i-proh) ws put in the titnium (IV) n-utoxide. Next, i-proh, deionized wter, nd concentrted nitric cid (HNO 3 ) ws mixed nd dropped into the solution. A light yellow TiO 2 sol ws otined fter stirring for 6 h. The molr rtios of AcAc, H 2 O, nd HNO 3 to Ti(OBu) 4 were 1:1, 2:1, nd 0.3:1, respectively. Then, Er(NO 3 ) 3 5H 2 O, Y(NO 3 ) 3 5H 2 O, nd LiNO 3 were dded into the TiO 2 sol to mke the molr rtios of Er:Y:Li:Ti = 0. 5:10:x:100 (x = 0, 10, 15, 20, 25). The solvent in the Er 3+ -Y 3+ -Li + tri-doped TiO 2 sol (UC-TiO 2 sol) ws removed y drying for 8 h t 100 C. Then, the UC- TiO 2 ws clcined t 500 C for 30 min. Friction of Perovskite Solr Cells Ptterned FTO glss sustrte ws clened in cetone, 2-propnol, nd ethnol y soniction for 20 min, respectively. Then, UV-O 3 ws used to tret the FTO for 15 min. A compct lyer ws formed y spincoting precursor solution on FTO nd nneled t 500 C for 30 min. The precursor solution is 0.1 M titnium diisopropoxide is (cetylcetonte) (75 wt% in isopropnol, Aldrich) solution in 1-utnol. A mesoporous TiO 2 film ws otined y spin-coting TiO 2 solution on the compct lyer t 4000 rpm for 30 s, followed y nneling t 100 C for 10 min nd 500 C for 30 min, respectively. The TiO 2 solution ws prepred y diluting TiO 2 pste (30NR-D, Dyesol) with ethnol (1:6, weight rtio) or y mixing the UC-TiO 2 sol nd the diluted TiO 2 solution (UC-TiO 2 : TiO 2 = x:100, v/v, x = 10, 20, 30, nd 40). A perovskite lyer ws formed on UC-TiO 2 lyer y spin-coting perovskite precursor solution in two steps t 1000 rpm for 10 s nd 4000 rpm for 30 s, nd 200 μl chloroenzene ws poured on the sustrte duringthesecondstepeforetheendof20s.then, the smple ws heted on the hotplte t 100 C for 1 h. The perovskite precursor solution ws otined y mixing PI 2 (1.1 M), FAI (1 M), PBr 2 (0.22 M), nd MABr (0.2 M) in the mixed solvent of nhydrous DMF/DMSO (4:1 v:v). Then, 1.5 M stock solution of CsI pre-dissolved in DMSO ws dropped into the Fig. 2 XRD ptterns of commercil TiO 2 (30NR-D) nd UC-TiO 2

3 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 3 of 8 c d Fig. 3 XPS spectr of UC-TiO 2. Ti 2p, Er 4d, c Y 4d, nd d Li 1s mixed perovskite precursor solution [17]. A holetrnsfer lyer ws formed y spin-coting solution of spiro-meotad t 4000 rpm 30 s. Finlly, n 80- nm-thick gold lyer ws thermlly evported on the top of the device. Chrcteriztion Up-conversion fluorescence, stedy-stte photoluminescence (PL), nd time-resolved photoluminescence (TRPL) spectr were mesured with FLS 980 E fluorometer (Edinurgh Photonics). X-ry diffrction (XRD) spectr were cquired on diffrctometer (DX- 2700). X-ry photoelectron spectroscopy (XPS THS-103) with Al K s X-ry source ws pplied to mesure the chemicl stte of the smples. Ultrviolet visile ner-infrred (UV-vis-NIR) sorption spectr were collected on Vrin Cry 5000 spectrophotometer. Microstructure nd morphologies were oserved with scnning electron microscope (SEM; JEM-7001F, JEOL). Photocurrent-voltge (J-V) curves were mesured with Keithley 2440 Sourcemeter under AM 1.5 G illumintion. The electrochemicl impednce spectroscopy Fig. 4 UV-vis-NIR sorption spectr of commercil TiO 2 (30NR-D) nd UC-TiO 2. Tuc plots

4 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 4 of 8 Fig. 5 SEM imges of the mesoporous lyers. 30NR-D film without UC-TiO 2. 30NR-D film with UC-TiO 2 (EIS) ws otined with n electrochemicl worksttion (CHI660e, Shnghi CHI Co., Ltd.). Results nd Discussion Up-conversion emissions were mesured with n excittion of 980-nm lser. Figure 1 shows the up-conversion emissions spectr of Er 3+ -Y 3+ -Li + tridoped TiO 2 (Er:Y:Li:Ti = 0.5:10:x:100, x = 0, 10, 15, 20, nd 25, molr rtio). Figure 1 shows the schemtic energy-trnsfer mechnisms of the Er 3+ ions. The green emission nds locted t out 525 nd 545 nm cn e ttriuted to 2 H 11/2 4 I 15/2 nd 4 S 3/ 2 4 I 15/2 trnsitions of Er 3+ ions, respectively. And the red emission nds centered t out 658 nm could correspond to 4 F 9/2 4 I 15/2 trnsitions of Er 3+ ions [15, 16]. With the increse of Li + doping content, the intensity of the spectrum increses firstly, nd then decreses, which is the mximum when the doping content of Li + is x =20. Hereinfter, the up-conversion mteril of Er 3+ -Y 3 + -Li + tri-doped TiO 2 (Er:Y:Li:Ti = 0.5:10:20:100, molr rtio) ws pplied. Figure 2 displys the XRD ptterns of commercil TiO 2 (30NR-D) nd UC-TiO 2. The XRD pttern of UC-TiO 2 is similr to tht of 30NR-D. The peks t 25.3, 37.8, 48.0, nd 53.8 in the XRD ptterns re ssigned to the (101), (004), (200), nd (105) plnes (JCPDS crd no ), respectively, which indictes tht UC-TiO 2 nd 30NR-D elong to the ntse phse of TiO 2. To confirm the doping of Er, Y, nd Li into TiO 2, XPS spectr of UC-TiO 2 were recorded nd shown in Fig. 3. The peks t nd ev in Fig. 3 could elong to Ti 2p 3/2 nd Ti 2p 1/2, respectively, nd the peks t ev in Fig. 3 nd t ev in Fig. 3c could e ttriuted to Er 4d nd Y 4d, respectively [18]. The pek t 55.5 ev in Fig. 3d cn correspond to Li 1s [19]. The survey XPS spectrum of UC-TiO 2 nd O1s pek were lso presented in Additionl file 1: Figure S1. The results demonstrted tht Er, Y, nd Li toms were doped into TiO 2. Figure 4 displys the UV-vis-NIR sorption spectr of TiO 2 (30NR-D) nd UC-TiO 2. Compred with 30NR- D, UC-TiO 2 presents stronger sorption, especilly t the rnge from 900 to 1000 nm. The energy nd gp without UC-TiO2 with UC-TiO2 Fig. 6 PCE dependence of solr cells on the content of UC-TiO 2 (UC-TiO 2 :30NR-D=x:100, v/v) in the mixture. I-V curves of the est performnce devices without UC-TiO 2 nd with UC-TiO 2

5 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 5 of 8 Tle 1 Photovoltic prmeters of the solr cells with UC-TiO 2 (20%) nd without UC-TiO 2 Smples V oc (V) I sc (ma/cm 2 ) FF (%) PCE (%) Control 1.01 ± ± ± ± 0.3 With UC-TiO ± ± ± ± 0.2 (E g ) could e estimted with the Tuc plot [20]. The Tuc plots re shown in Fig. 4, from which the vlues of E g could e otined to e 3.20 nd 3.10 ev for 30NR-D nd UC-TiO 2, respectively. The E g of UC-TiO 2 is smller thn tht of un-doped TiO 2. Figure 5 shows the SEM imge of 30NR-D film formed on the compct lyer. The nnoprticle size is out 30 nm, nd the size distriution is uniform. Figure 5 shows the SEM imge of the film contining 30NR-D nd UC-TiO 2 deposited on the compct lyer y spin-coting method. There is no ovious difference etween the two films, which displys tht the prticle size nd morphology of UC-TiO 2 re similr to those of 30NR-D. In the present work, the perovskite film ws formed with the method previously reported [17]. According to report, the composition of the perovskite film is Cs 5 (MA 0.17 FA 0.83 ) 95 P(I 0.83 Br 0.17 ) 3, nd the role of the CsI is to mke the perovskite solr cells thermlly more stle, with less phse impurities, nd less sensitive to processing conditions [17]. The scheme of the device is presented in Additionl file 1: Figure S2. The perovskite solr cells sed on mesoporous lyer formed with the mixture of UC-TiO 2 sol nd diluted TiO 2 solution (UC-TiO 2 :TiO 2 = x:100, v/v, x = 0, 10, 20, 30, nd 40) were fricted nd their I-V curves were mesured. The photovoltic prmeters were otined from the I-V mesurements. Figure 6 shows the PCE dependence of solr cells on the content of UC-TiO 2 (UC-TiO 2 :TiO 2 = x:100, v/v) inthe mixture. With the increse of UC-TiO 2 content, the power conversion efficiency (PCE) of solr cells increses firstly, nd then decreses, which is the mximum t the content of x =20 for UC-TiO 2. The detiled photovoltic prmeters of solr cells with 20% UC-TiO 2 nd without UC-TiO 2 were listed in Tle 1. Compred with those of the devices without UC-TiO 2, the photovoltic prmeters of the solr cells with UC-TiO 2 present n improvement. The PCE of the solr cells with 20% UC-TiO 2 is incresed to 16.5 from 14.0% for the solr cells without UC-TiO 2, which presents n increse of 19%. Figure 6 displys the I-V curves of the typicl solr cells with UC-TiO 2 nd without UC-TiO 2. To understnd the enhncement, some investigtions were crried out. Stedy-stte photoluminescence (PL) nd time-resolved photoluminescence (TRPL) cn e pplied to investigte the electron extrtion nd trnsport process. The PL of perovskite lyer on the mesoporous lyers formed y 30NR- D nd 30NR-D with UC-TiO 2 were mesured nd shown in Fig. 7. Compred with tht of 30NR-D/ perovskite, the PL intensity of 30NR-D with UC- TiO 2 /perovskite ecomes reduced, which indictes tht the electron extrtion nd trnsport efficiency cross the interfce etween 30NR-D with UC-TiO 2 nd perovskite is etter thn tht etween 30NR-D nd perovskite [21]. Figure 7 shows the TRPL spectr of perovskite lyer on the mesoporous lyers formed y 30NR-D nd 30NR-D with UC-TiO 2.The TRPL spectrum ws fitted to iexponentil function, in which the fst decy (τ 1 ) my e resulted from trnsporttion of free crriers, nd the slow decy (τ 2 ) cn e originted from rditive recomintion of free crriers [22 24]. The otined prmeters re listed in Tle 2. Compred with tht of 30NR-D/perovskite, the fst decy time (2.8 ns) of 30NR-D with UC-TiO 2 /perovskite ecomes Fig. 7 PL nd TRPL of perovskite lyers on the mesoporous lyers formed y 30NR-D without UC-TiO 2 nd 30NR-D with UC-TiO 2

6 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 6 of 8 Tle 2 Prmeters of the TRPL spectr Smples τ 1 /ns % of τ 1 τ 2 /ns % of τ 2 30NR-D/perovskite NR-D with UC-TiO 2 /perovskite Tle 3 Fitting prmeters for EIS dt Solr cells R s /Ω R co /Ω R rec /Ω C co /Ω CPE-T/F CPE μ -P 30NR-D E-7 5.9E NR-D with UC-TiO E-7 5.5E smller, while the frction of fst decy process (98. 2%) ecomes lrger. This implies tht the chrge trnsfer etween perovskite nd 30NR-D with UC- TiO 2 is fster thn tht etween perovskite nd 30NR-D. Eletrochemicl impednce spectroscopy (EIS) is n effective method to get some informtion on crrier trnsfer ehvior. Figure 8 displys the Nyquist plots of the devices sed on mesoporous lyers formed y 30NR-D nd 30NR-D with UC-TiO 2, in which two rcs were oserved. The rc t high-frequency could e resulted from the contct resistnce etween the interfces, nd the rc t low-frequency could come from the recomintion resistnce (R rec ) nd chemicl cpcitnce (C μ )of the device [25, 26]. The EIS ws fitted with n equivlent circuit shown in Fig. 8, nd the otined prmeters re listed in Tle 3. The series resistnce of the devices sed on 30NR-D with UC-TiO 2 ecomes smller thn tht of the devices on sed on 30NR-D, while the recomintion resistnce of the former ecomes lrger thn tht of the lter. This indictes the chrge Fig. 8 Nyquist plots of the devices sed on mesoporous lyers formed y 30NR-D without UC-TiO 2 nd 30NR-D with UC-TiO 2. Equivlent circuit pplied to fit the EIS dt recomintion ws decresed nd the chrge trnsport ws improved for the device sed on 30NR-D with UC- TiO 2. To further prove the effect of UC-TiO 2 on photocurrents of the devices, the I-V curves of the devices sed on the mesoporous lyers without UC-TiO 2 nd with UC-TiO 2 were mesured under the simulted solr rdition in the wvelength rnge of λ 980 nm with NIR filter, which re shown in Additionl file 1: Figure S3. Compred with tht of thedevicewithoutuc-tio 2, the photocurrent of the device with UC-TiO 2 ws oviously enhnced, which demonstrtes tht the incorportion of UC-TiO 2 in the device cn trnsform the NIR light into visile light, which cn e sored y the devices to generte photocurrent. To explin the incresed open circuit voltge (V oc ) of the solr cells, the energy nd rrngements of UC-TiO 2,TiO 2, perovskite, nd Spiro-OMeTAD were showninfig.9sed on the sorption spectr (Fig. 4) nd the litertures [27, 28]. The conduction nd edge of the UC-TiO 2 is lower thn tht of TiO 2 (30NR-D) due to its smller energy nd gp; thus, the conduction nd offset etween UC-TiO 2 nd perovskite is lrger thn tht etween TiO 2 nd perovskite. This could e one of the resons to hve higher open circuit voltge for UC-TiO 2 sed solr cells [29, 30]. In summry, the PCE increse of the solr cells sed on the mesoporous lyer with UC-TiO 2 is due to the enlrged I sc nd incresed V oc. The enlrged I sc could e due to expnsion of spectrl sorption to ner-infrred (NIR) rnge with up-conversion mteril, reduced recomintion, nd fst chrge trnsfer of photo-generted crriers. The incresed V oc my e ttriuted to the enlrged conduction nd offset etween UC-TiO 2 nd perovskite. Conclusions Er 3+ -Y 3+ -Li + tri-doped TiO 2 (UC-TiO 2 ) ws prepred y ddition of Li + into Er 3+ -Y 3+ co-doped TiO 2, which presented n enhnced up-conversion emission. The UC-TiO 2 ws pplied to the perovskite solr cells. The performnce of solr cells with UC-TiO 2 ws improved compred with tht of control device. The I sc, V oc, nd FF of solr cells with UC-TiO 2 were incresed to ma/cm 2, 1.05 V, nd 70.8% from 21.0 ma/cm 2, 1.01 V,

7 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 7 of 8 Fig. 9 Schemtic energy nd rrngements of UC-TiO 2, TiO 2, perovskite, nd Spiro-OMeTAD nd 66.0% for the control devices, respectively. Thus, the PCE with UC-TiO 2 ws incresed to 16.5 from 14. 0% for the solr cells without UC-TiO 2, which presents n increse of 19%. Bsed on some experimentl results, this PCE increse ws explined. Additionl file Additionl file 1: Figure S1. Survey XPS spectrum of UC-TiO2. O 1s pek. Figure S2 Scheme of the perovskite solr cells. Figure S3 I-V curves of the devices sed on the mesoporous lyers without UC-TiO 2 nd with UC-TiO 2 mesured under the simulted solr rdition in the wvelength rnge of λ 980 nm with NIR filter. (DOCX 81 k) Arevitions EIS: Electrochemicl impednce spectroscopy; NIR: Ner-infrred; PCE: Power conversion efficiency; PL: Photoluminescence; PSCs: Perovskite solr cells; TRPL: Time-resolved photoluminescence Funding This work is supported y the NSFC-Henn Province Joint Fund (U ), Science Fund of Henn Province ( ), Science Project of Eduction Deprtment of Henn Province (No. 17A140005), Science nd Technology Development Project of Henn Province (No ), nd Progrm for Science nd Technology Innovtion Tlents in Universities of Henn Province (No. 16HASTIT043). Avilility of Dt nd Mterils All dt re fully ville without restriction. Authors Contriutions ZZ nd Y-L Mo crried out the min prt of the experiment nd drfted the mnuscript. The other uthors provided ssistnce with experimentl mesurements, dt nlysis nd the mnuscript writing. All uthors red nd pproved the finl mnuscript. Competing Interests The uthors declre tht they hve no competing interests. Pulisher s Note Springer Nture remins neutrl with regrd to jurisdictionl clims in pulished mps nd institutionl ffilitions. Received: 31 Jnury 2018 Accepted: 18 April 2018 References 1. Ahn N, Son DY, Jng IH, Kng SM, Choi M, Prk NG (2015) Highly reproducile perovskite solr cells with verge efficiency of 18.3% nd est efficiency of 19.7% fricted vi Lewis se dduct of led(ii) iodide. J Am Chem Soc 137: McMeekin DP, Sdoughi G, Rehmn W, Eperon GE, Sli M, Horntner MT, Hghighird A, Ski N, Korte L, Rech B, Johnston MB, Herz LM, Snith HJ (2016) A mixed-ction led mixed-hlide perovskite sorer for tndem solr cells. Science 351: Tsi H, Nie WY, Blncon JC, Toumpos CC, Asdpour R, Hrutyunyn B, Neukirch AJ, Verduzco R, Crochet JJ, Tretik S, Pedesseu L, Even J, Alm MA, Gupt G, Lou J, Ajyn PM, Bedzyk MJ, Kntzidis MG, Mohite AD (2016) High-efficiency two-dimensionl Ruddlesden-Popper perovskite solr cells. Nture 536: Aror N, Dr MI, Hinderhofer A, Pellet N, Schreier F, Zkeeruddin SM, Grtzel M (2017) Perovskite solr cells with CuSCN hole extrction lyers yield stilized efficiencies greter thn 20%. Science 358: Yng WS, Prk BW, Jung EH, Jeon NJ, Kim YC, Lee DC, Shin SS, Seo J, Kim EK, Noh JH, Seok SI (2017) Iodide mngement in formmidinium led hlide sed perovskite lyers for efficient solr cells. Science 356: Liu MZ, Johnston MB, Snith HJ (2013) Efficient plnr heterojunction perovskite solr cells y vpour deposition. Nture 501: Zhng JB, Xu B, Johnsson MB, Vlchopoulos N, Boschloo G, Sun LC, Johnsson EMJ, Hgfeldt A (2016) Strtegy to oost the efficiency of mixedion perovskite solr cells: chnging geometry of the hole trnsporting mteril. ACS Nno 10: He M, Png XC, Liu XQ, Jing BB, He YJ, Snith H, Lin ZQ (2016) Monodisperse dul-functionl upconversion nnoprticles enled nerinfrred orgnoled hlide perovskite solr cells. Angew Chem Int Ed 55: Roh JM, Yu HJ, Jng J (2016) Hexgonl β-nyf 4 :Y 3+,Er 3+ nnoprism incorported upconverting lyer in perovskite solr cells for ner-infrred sunlight hrvesting. ACS Appl Mter Interfces 8: Que M, Que W, Yin X, Chen P, Yng Y, Hu J, Yu B, Du Y (2016) Enhnced conversion efficiency in perovskite solr cells y effectively utilizing ner infrred light. Nno 8: Yu J, Yng Y, Fn R, Liu DQ, Wei LG, Chen S, Li L, Yng B, Co WW (2014) Enhnced ner-infrred to visile upconversion nnoprticles of Ho 3+ -Y 3 + -F tri-doped TiO 2 nd its ppliction in dye-sensitized solr cells with 37% improvement in power conversion efficiency. Inorg Chem 53: Yu J, Yng YL, Fn RQ, Zhng HJ, Li L, Wei LG, Shi Y, Pn K, Fu HG (2013) Er 3 + nd Y 3+ co-doped TiO 2-x F x up-conversion luminescence powder s

8 Zhng et l. Nnoscle Reserch Letters (2018) 13:147 Pge 8 of 8 light scttering lyer with enhnced performnce in dye sensitized solr cells. J Power Sources 243: Wng XL, Zhng ZL, Qin JQ, Shi WJ, Liu YF, Go HP, Mo YL (2017) Enhnced photovoltic performnce of perovskite solr cells sed on Er-Y co-doped TiO 2 nnorod rrys. Electrochim Act 245: Co BS, He YY, Sun Y, Song M, Dong B (2011) Opticl high temperture sensor sed on enhnced green upconversion emissions in Er 3+ -Y 3+ -Li + codoped TiO 2 powders. J Nnosci Nnotechnol 11: He YY, Wu JL, Wng XH, Feng ZQ, B. Dong B. (2016) Opticl temperture sensing ehvior through strk sulevels trnsitions of green nd red upconversion emissions for Er 3+ -Y 3+ -Li + codoped TiO 2 phosphors. J Nnosci Nnotechnol 16: Giordno F, Ate A, Ben JPC, Sli M, Mtsui T, Im SH, Zkeeruddin SM, Nzeeruddin MK, Hgfeldt A, Gretzel M (2015) Enhnced electronic properties in mesoporous TiO 2 vi lithium doping for high-efficiency perovskite solr cells. Nt Commun 7: Michel S, Tisuke M, Seo JY, Domnski K, Corre-Ben JP, Nzeeruddin MK, Zkeeruddin SM, Tress W, Ate A, Hgfeldtd A, Grtzel M (2016) Cesium-contining triple ction perovskite solr cells: improved stility, reproduciility nd high efficiency. Energy Environ Sci 9: Ji TH, Liu Y, Zho H, Du HY, Sun JY, Ge GL (2010) Preprtion nd upconversion fluorescence of rre erth (Er 3+ or Y 3+ /Er 3+ )-doped TiO 2 nnoelts. J Solid Stte Chem 183: Rvishnkr TN, Ngrju G, Dupont J (2016) Photoctlytic ctivity of Lidoped TiO 2 nnoprticles: synthesis vi ionic liquid-ssisted hydrotherml route. Mter Res Bull 78: Go XX, Ge QQ, Xue DJ, Ding J, M JY, Chen YX, Zhng B, Feng YQ, Wn LJ, Hu JS (2016) Tuning the fermi-level of TiO 2 mesoporous lyer y lnthnum doping towrds efficient perovskite solr cells. Nno 8: Ling C, Wu Z, Li P, Fn JJ, Zhng YQ, Sho GS (2016) Chemicl th deposited rutile TiO 2 compct lyer towrd efficient plnr heterojunction perovskite solr cells. Appl Surf Sci 391: Chen Q, Zhou H, Song TB, Luo S, Hong Z, Dun HS, Dou L, Liu Y, Yng Y (2014) Controllle self-induced pssivtion of hyrid led iodide perovskites towrd high performnce solr cells. Nno Lett 14: Ling PW, Lio CY, Chu CC, Zuo F, Willims ST, Xin XK, Lin J, Jen AKY (2014) Additive enhnced crystlliztion of solution-processed perovskite for highly efficient pnr-heterojunction solr cells. Adv Mter 26: Yeo JS, Kng R, Lee SY, Jeon YJ, Myoung NS, Lee CL, Kim DU, Yun JM, Seo YH, Kim SS, N SI (2015) Highly efficient nd stle plnr perovskite solr cells with reduced grpheme oxide nnosheets s electrode interlyer. Nno Energy 12: Liu DY, Yng JL, Kelly TL (2014) Compct lyer free perovskite solr cells with 13.5% efficiency. J Am Chem Soc 136: Kim HS, Lee JW, Yntr N, Boix PP, Kulkrni SA, Mhislkr S, Grtzel M, Prk NG (2013) High efficiency solid-stte sensitized solr cell-sed on sumicrometer rutile TiO 2 nnorod nd CH 3 NH 3 PI 3 perovskite sensitizer. Nno Lett 13: Rhee JH, Chung CC, Diu WG (2013) A perspective of mesoscopic solr cells sed on metl chlcogenide quntum dots nd orgnometl-hlide perovskites. NPG Asi Mter 5:e Chen H, Yng B, Shung X, Zhng T, Meng XY, Ng WK, Yng YL, Wong KS, Chen HN, Yng SH (2017) Tuning the A-site ction composition of FA perovskites for efficient nd stle NiO-sed p-i-n perovskite solr cells. J Mter Chem A 5: Kim HS, Prk NG (2014) Prmeters ffecting I-V hysteresis of CH 3 NH 3 PI 3 perovskite solr cells: effects of perovskite crystl size nd mesoporous TiO 2 lyer. J Phys Chem Lett 5: Li YL, Sun WH, Yn WB, Ye SY, Peng HT, Liu ZW, Bin ZQ, Hung CH (2015) High-performnce plnr solr cells sed on CH 3 NH 3 PI 3-x Cl x perovskites with determined chlorine mole frction. Adv Funct Mter 25: