Thermal Behaviors of Methylammonium Lead Trihalide Perovskites

Jul 6, 2016 - †CAS Key Laboratory of Materials for Energy Conversion, Department of Materials Science and Engineering, ‡Hefei National Laboratory ...
0 downloads 0 Views 1MB Size
Subscriber access provided by La Trobe University Library

Article

Thermal Behaviors of Methylammonium LeadTrihalide Perovskites With or Without Chlorine Doping Fangfang Wang, Xiaoning Li, Xiaofeng Yin, Zhengping Fu, and Yalin Lu J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.6b03147 • Publication Date (Web): 06 Jul 2016 Downloaded from http://pubs.acs.org on July 11, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Thermal Behaviors of Methylammonium Lead-Trihalide Perovskites with or without Chlorine Doping Fangfang Wanga, Xiaoning Lia, Xiaofeng Yina, Zhengping Fua,b,c*, Yalin Lua,b,c,d,e,f*

a

CAS Key Laboratory of Materials for Energy Conversion, Department of Materials Science and Engineering, University of Science and Technology of China, Hefei 230026, P. R. China

b

Hefei National Laboratory for Physical Sciences at Microscale, University of Science and Technology of China, Hefei 230026, P. R. China

c

Synergetic Innovation Center of Quantum Information & Quantum Physics, University of Science and Technology of China, Hefei 230026, P. R. China

d

National Synchrotron Radiation Laboratory, University of Science and Technology of China, Hefei 230026, Anhui, China

e

Hefei Physical Sciences and Technology Center, CAS Hefei Institutes of Physical Sciences, Hefei 230031, Anhui, China

f

Laser Optics Research Center, US Air Force Academy, Colorado 80840, USA

*Corresponding Authors *E-mail: [email protected]; [email protected] Address: Hefei National Laboratory for Physical Sciences at Microscale, University of Science and Technology of China, Hefei 230026, P. R. China

Tel: +86-551-63603194

1

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

Abstract Methylammonium lead-trihalide perovskites have become the harbinger in solar cells lately. However, many fundamental mechanisms, especially those related to the used materials, still need a deeper understanding, in order to further improve the cells’ stability and efficiency. This work discussed the influence of chlorine doping on the materials’ thermal stability and the energy loss by either releasing heat or emitting light in CH3NH3PbI3. The results indicated that CH3NH3PbI3-xClx had no phase transition in the range of 25 - 100 oC, in contrast to CH3NH3PbI3 with a phase transition at ~50 oC. It was discovered that the heat generated in absorbing photon processes in CH3NH3PbI3-xClx could be reduced, about 20% lower than that in CH3NH3PbI3. More heat released in CH3NH3PbI3 may cause the occurrence of the phase transition, and then further decline the cell stability. Photoluminescence results also indicated that adding chlorine into CH3NH3PbI3 could further suppress the energy loss in the means emitting light.

2

ACS Paragon Plus Environment

Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1 Introduction Methylammonium lead-trihalide perovskites have been rapidly used in the solution-processed photovoltaic technology due to their long electron-hole diffusion lengths, high light absorption coefficient, excellent carrier mobility and direct band gap1. In past several years, power conversion efficiency (PCE) has been greatly improved from initial 3.8%2 to 22.1%3 for the perovskite-based solar cells. Unfortunately, a full commercialization of such cells has been less successful, mainly due to the questionable stability of the used perovskite materials and due to the fact that the high efficiency was still not overwhelming. Enormous past efforts have been on either improving the perovskite materials’ syntheses or on finding new hybrid perovskites, and surprisingly, much less on understanding their thermal behaviors when with or without chlorine dopants. On one hand, a solar cell has several main nonequilibrium dynamic processes for those excitons generated by incident photons4, which will release the absorbed energy, as shown in Figure 1: a separation of electrons and holes toward opposite electrodes to generate electricity power, radiative recombination of partial electrons and holes to re-emit photons, nonradiative recombination of partial electrons and holes to give off the heat, as examples. The last two recombination processes are associated with unwanted energy losses, and they should be suppressed in order to promote the efficiency. On the other hand, a solar cell’s stability can be affected by many reasons, including the used materials themselves, over-heating, the used structures, etc. For example, CH3NH3PbI3 will experience a phase transition at about 327K from tetragonal I phase to cubic phase, and the phase transition involves a volume change, which may decreases stability, if the cell is over-heated over the transition temperature5. Chlorine-doping has been played a key role in the efficiency enhancement, as was demonstrated in many previous researches. Adding of chlorine into CH3NH3PbI3 (CH3NH3PbI3-xClx) have displayed superior carrier mobility and a much improved lifetime, compared to those chlorine-free absorber layers. Hui Yu et al revealed that the used chlorine source affects the film formation. The abounding of CH3NH3+ was 3

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 22

crucial to slack the formation process of perovskite films and accordingly mended the crystalline domains during the annealing6. Thus, the added chlorine had an effect to permit the removal of surplus CH3NH3+ at a comparatively lower annealing temperature. Development of a preferred orientation in CH3NH3PbI3-xClx films in contrast to CH3NH3PbI3 films again also indicated that the chlorine played a role in the film formation7, 8. Stranks et al showed that CH3NH3PbI3-xClx had significantly longer diffusion lengths compared to CH3NH3PbI39. In addition while carrier mobility in both CH3NH3PbI3-xClx and CH3NH3PbI3 had a slight difference, the bimolecular recombination rates of the CH3NH3PbI3-xClx films were about one order lower, implying that chlorine can play a role in decreasing the spatial overlap of electrons and holes by adjusted the electronic structure10. Colella et al stated that chlorine within CH3NH3PbI3 can act as a dopant, improving the transporting properties11. Apparently, the exact role of doping chlorine on both efficiency and stability was still unclear.

Figure 1 Schematic illustrating of the energy flowing way.

In this work, we studied both stability and energy losses in both CH3NH3PbI3 and CH3NH3PbI3-xClx by using differential scanning calorimetry (DSC) technique. We have gotten that CH3NH3PbI3-xClx had no phase transition in the range from 25 to 100 oC, in contrast to CH3NH3PbI3 with a clear phase transition at about 50 oC. This implied that CH3NH3PbI3-xClx can be more stable than CH3NH3PbI3, when the device reached to such temperature range. Furthermore, the photo-thermal study showed that the heat generated in CH3NH3PbI3-xClx was about 20% less than that in CH3NH3PbI3. 4

ACS Paragon Plus Environment

Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

In other words, the energy loss by giving off heat may be higher in CH3NH3PbI3 solar cells, which may initialize the phase transition and then further decline the cell efficiency and the stability. Finally, photoluminescence results also indicated that the addition of chlorine into CH3NH3PbI3 can efficiently suppress the alternative energy loss through emitting photons. Detailed results in this work can be very instructive to future perovskite-based solar cells. 2 Experimental Sections 2.1 Sample Preparation. All materials were purchased from Alfa. The precursor solution was prepared by dissolving the organic (CH3NH3I) and the inorganic (PbI2) components in DMF at 1:1 molar ratio.CH3NH3PbI3-xClx precursor solution was obtained through dissolving lead chloride and CH3NH3I in DMF with a molar ratio of 1:3. To prepare the perovskite powder, the prepared precursor solution was spin-coated on the fluorine-doped tin oxide substrates and annealed at 100 °C approximately 10 min for CH3NH3PbI3 and 60 min for CH3NH3PbI3-xClx. CH3NH3PbI3 and CH3NH3PbI3-xClx powders were scratched off the substrate and were used directly for further thermal analysis. All the powders were prepared in ambient air with humidity smaller than 40 % and they were stored in a dry condition. 2.2 Characterization. Morphologies of CH3NH3PbI3 and CH3NH3PbI3-xClx thin films were observed by the field emitted scanning electron microscopy (FESEM, JSM-6700F). The as-deposited CH3NH3PbI3 and CH3NH3PbI3-xClx thin films were described using a TTR-III X-ray diffractometer and the testing condition was: Cu Kα radiation (λ=1.5406 Å), 2θ range of 10°-70° and a step size of 0.02°. Photoluminescence (PL) spectra were surveyed on a steady/transient fluorescence spectrometer (FLES920). 2.3 Thermal Analysis. DSC characterizations were executed by employing a Perkin-Elmer Diamond DSC (Pyris software 6.5) (USA) and an Illuminant (OmniCure, Series 2000). Nitrogen was existed during the experiment process. In isothermal tests, the samples was sealed in an aluminum pan with a content about 2~3

mg, in order to remove the thermal history, the system were firstly heated to 180 °C and held for 5 min, then decreased to the desired temperature: 50 °C, 52 °C, 56 °C, 5

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 22

58 °C at a cooling rate of 80 °C min-1, respectively. In non-isothermal study, the system was cooled at the rate of 10, 20, 30, 40 and 50 oC/min, respectively. In photo-thermal measurements, the samples were first holding at 25 oC for 2 min., then the light was introduced and the samples were illuminated for 2 min. The empty aluminum pan was measured before each test in order to deduct the baseline. The used light bands were 365 nm, 365-500 nm, 400-500 nm, and the white light were used in the experiments with the relative light intensity of 10%, 20%, 30%, 40% and 50%, respectively.

Figure 2 The DSC thermograms at different cooling rate for (a) CH3NH3PbI3, (c) CH3NH3PbI3-xClx and (b) Plots of Tonset, Tp, Tf as a function of cooling rate.

3. Results and discussion 3.1 Crystallization Process The DSC curves of CH3NH3PbI3 and CH3NH3PbI3-xClx at different cooling rates were presented in Figure 2(a, c). For the CH3NH3PbI3-xClx powder, no exothermic peaks were detected under the experimental condition, which was in consistence with 6

ACS Paragon Plus Environment

Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

the previous work of Sang II Seok and Michael Gratzel12. This observation implied that a substantial difference of both crystalline phases and the associated transitions existed between CH3NH3PbI3 and CH3NH3PbI3-xClx, and the crystal phase of the CH3NH3PbI3 and CH3NH3PbI3-xClx was tetragonal which will be stated in the later discussion. The reason may be due to the presence of remaining excess CH3NH3Cl which suppressed the phase transition in some unclear ways12. It was well known that a phase transition should contain two processes: nucleation and growth. Thus, it was urgent to further analyze the crystallization process during the phase transition. For this reason, the phase transitions of CH3NH3PbI3 during isothermal and non-isothermal crystallization process were further analyzed by DSC in much more details, which should provide a further understanding of the CH3NH3PbI3 based materials and the effect on both stability and efficiency. The Non-Isothermal Crystallization Process In Fig. 2(a), the onset temperature (Tonset), the end temperature (Tf) and the crystallization peak temperature (Tp) all decreased when the cooling rate increased, which indicated that it is much easier to crystallize at lower temperatures. Meantime, the crystallization exothermic peak also gradually changed from narrow to wide, which meant that the crystallization temperature range become larger. The reason is due to that the increase of the cooling rate shortened the time to crystallization, so that it would be difficult to completely crystallize in a short period of time and the degree of the crystallization turns to unperfected. Plots of Tonset, Tp and Tf vs the cooling rate were shown in Fig. 2(b). In Fig. 2(b), Tonset is obviously different at diverse cooling rates. The Tp and Tf decreased with increasing the cooling rate. The results show that both Tonset and Tf depended on the cooling rate during non-isothermal crystallization process. Namely, the greater cooling rates contributed to a wider crystallization temperature range. From the crystallization driving force point of view, the relative degree of the undercooling was smaller with the cooling rate slowing down, which means that the period at high temperature is long enough for the nucleation. Therefore, the initial crystallization occurs at a higher temperature and the crystallization process was completed at a 7

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

higher temperature range. In contrast, relative degree of undercooling tunes larger when the cooling rate becomes faster. Then the time for the nucleation is not long enough, and the nucleation mostly took place in the low temperature range. Therefore, the initial crystallization was postponed to the low temperature direction. Nevertheless, the increase of the crystallization driving force raises the crystallization rate, so the crystallization can be done in a relatively short time and in a lower temperature range.

Figure 3 X(t) as a function of (a) temperature T and (b) time t, kinetics parameters from (c) Jeziorny and (d) Mo model for non-isothermal crystallization of CH3NH3PbI3.

Relative crystallinity X(t) is an important index to calculate crystallization parameters. It can be calculated by the proportion of the crystallization heat gives off at any time with the crystallization heat released when the sample was complete crystals. The relation curve of X(t) with time of CH3NH3PbI3 can be gained from the DSC examination. Figure 3 (a) presents X(T) as a function of T of CH3NH3PbI3 under 8

ACS Paragon Plus Environment

Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

different cooling rates, obtained from the area of non-isothermal crystallization exotherms. The temperature can be converted into a time gauge, using the equation13:

 = ( − )/∅

(1)

 values were where ϕ is the cooling rate, which shown in Fig. 3(b). The  /

calculated and listed in Table 1, which represent the time to achieve 50% relative crystallinity. It can also indicate the crystallization rate, i.e., crystallization is fast if   / is small. As showed in the Fig. 3(b), the time for completely crystallization

shorten when the cooling rate is fast, indicating the faster crystallization rate. Many models can be used to investigate the non-isothermal crystallization kinetics, such as Avrami, Ozawa, Mo and Ziabicki methods14. In this paper we choose a modified Avrami and Mo model to analyze the non-isothermal crystallization processes in CH3NH3PbI3. As a general rule, the Avrami equation is applicable only in the isothermal crystallization. By assuming that the cooling rate and the crystallization temperature do not change, Jeziorny adjusted the Avrami equation and made it suitable to describe the non-isothermal crystallization process of the primary crystallization phase. The revised equation is shown as follows15:

ln− ln1 − () = ln + ln

(2)

In this equation, Zc is the crystallization rate constant which is gotten by modified Avrami isothermal crystallization kinetics parameter k through considering the cooling rate, and t is the crystallization time, while n is the Avrami exponent (representing the growth and nucleation behavior). The plotting of ln[-ln[1-X(t)]] versus ln t according to Eq. (2) is shown in Fig. 3(c), n and Zc were obtained from the intercepts and slopes of the plots, and they are listed in Table 1. One can see from Table 1 that n gradually reduces with the raise of cooling rate, implying that the heterogeneous nucleation become more and more popular. In addition, the decreased value of n indicates the more unperfected crystal. At the same time, with the raise of cooling rate, the Zc raises and t reduces, indicating the short crystallization time. The reason may be that the temperature decreased fast when the cooling rate increased, then a large number of nucleation appeared in a short time period, but there was no 9

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 22

enough time to grow up. As a result, the samples could crystallize at a relatively short time but with a low degree of full crystallization. Table 1 Kinetics parameters for non-isothermal crystallization calculated with Mo and Jeziorny model of CH3NH3PbI3. Mo model

/

Jeziorny model

X(t) %

F(T)

a

  /

ϕ (oC/min)

Zc (min-1)

n

20

6.8

0.95

0.96

10

121.5

6.0

40

8.7

0.96

0.54

20

164.0

5.7

60

10.8

0.99

0.34

30

298.9

5.1

80

15.3

0.99

0.23

40

403.4

4.8

/

/

/

0.21

50

411.6

4.4

Mo et al.16derived another model by combining the Avrami and Ozawa model:

ln∅ = ln (T) − ln

(3)

where F(T) is the cooling rate chosen at a unit crystallization time at which the system has a certain degree of crystallinity. Normally, F(T) refers to the difficulty of crystallization, in other words, the larger of F(T), the slower of the crystallinity rate would be. And a = n/m is the ratio of the Avrami exponent n and Ozawa exponent m. Figure 3(d) are plots of ln ϕ versus ln t using Eq. (3) of CH3NH3PbI3. The F(T) and a can be obtained from the slope and the intercept, and listed in Table 1. The curves are almost a set of straight lines, which indicates that the Mo model can be used to express the non-isothermal crystallization process of CH3NH3PbI3. In Table 1, the value of F(T) grows with the raising of the X(t), which shows that under the unit crystallization time, the cooling rate must speed up for the purpose of a higher degree of crystallinity. At the same time, index a increased slowly with the increase of relative crystallinity. The Isothermal Crystallization Process Figure 4(b) displays the X(t) vs. t relation for CH3NH3PbI3 at each isothermal temperature (Tc). One can see that all curves are s-shaped, i.e., the relative 10

ACS Paragon Plus Environment

Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

crystallinity increased rapidly at first, and then increased to a maximum at a slow speed with the change of time. Besides, with the increase of crystallization temperature, the sample need short time to reach full crystallization, which indicates  that crystallization speed become fast. The half-time of crystallization ( / ) gained

from Fig. 4(b) is illustrated in Table 2. It shows that the crystallization

rate

of

CH3NH3PbI3 become fast with the increasing of Tc.

Figure 4 Isothermal crystallization graph of CH3NH3PbI3: (a) Schematic illustration of evolutions of crystalline morphologies; (b) X(t) as a function of t; (c) Plots of ln{-ln[1-X(t)]} vs. ln t; (d) Plots of ln k vs. 1/Tc.

Johnson-Mehl-Avrami (JMA) model was adopted to studied the isothermal crystallization kinetics of CH3NH3PbI3:17 X() = 1 − exp$−% & '

(4)

or ln− ln1 − X() = ln% + ln

(5) 11

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

where k is the crystallization rate constant. Figure 4(c) are Plots of ln{-ln[1-X(t)]} vs. ln t using Eq. (5) of CH3NH3PbI3. The slope and the intersection of the plots are taken to derive the values of k and n, which are also listed in Table 2. As is known, crystallization can take place in the manner of one-dimension, two-dimensions and three-dimensions, corresponding to form line, circle, and spheres, respectively. And the Avrami indices for a heterogeneous nucleation are 1, 2, or 3, while they are 2, 3, or 4 for homogeneous nucleation18. The values of the n for CH3NH3PbI3 are from 1.91 to 1.99 which is close to 2, so the crystallization process of CH3NH3PbI3 is two-dimensional heterogeneous nucleation crystallization. And an increase in the n means that the mechanism for nucleation and growth in CH3NH3PbI3 crystallites are affected by crystallization temperature. In the studied temperature range, the acquired value of n are not as an integer, this may be due to that the crystallization process are complexity, so the nucleation can't completely process in a way, and the crystal form may not necessarily grow in a uniform form. Besides, in the process of crystal nucleus formation and crystal growth, crystal nucleus or crystals may be superimposed with each other19. Table 2 Kinetic parameters for isothermal crystallization of CH3NH3PbI3. Tc (oC)

  / (min)

t'1/2(min)

n

k

50

1.14

0.97

1.91

0.73

52

1.05

0.93

1.92

0.79

56

1.04

0.92

1.93

0.80

58

0.84

0.84

1.99

0.98

△E (kJ/mol)

27.5

 *t / obtained from figure 2, t'1/2 obtained from calculation.

From the above analyses, the possible crystallization process schematic illustration of CH3NH3PbI3 is shown in Fig. 4(a). Generally, the crystallization contains two processes: nucleation and growth. In the first stage, a lot of crystal nucleus appeared at the same time, and then the nucleation and growth carried out simultaneously. Crystal nucleus continues growth with a two-dimensional direction until a collision with other grain, yet crystallization process is complete. The 12

ACS Paragon Plus Environment

Page 13 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

crystallization rate for whole crystallization process depends on the co-effect of these two processes. Moreover, the k are strongly depended on Tc, which is consisted with the trend of   / values for CH3NH3PbI3, the crystallization rate became fast with the rising of Tc.

Substituting X(t)=50% into formula (1), the half crystallization time t'1/2 is gotten: ′  / = (ln2/%) /&

(6)

the values of t'1/2 calculated from Eq. (6) are in table 2. Obviously, t1/2 obtained from Fig. 4(b) are consistent with the values from formula calculation, which indicated that JMA model is suitable for simulate the isothermal crystallization process of CH3NH3PbI3. The activation energy is important for the crystallization behavior of CH3NH3PbI3. On the basis that the crystallization rate constant k is thermally activated, the Arrhenius relation was used to describe the parameter k as follows20: % (T) = % exp +

,-. /0

1

(7)

where k0, Ea, R, Tc is the temperature-independent pre-exponential factor, activation energy, gas constant, and the crystallization temperature, respectively. Therefore, the activation energy Ea can be calculated from the slope of the plots of ln k vs. 1/Tc, as shown in Fig. 4(d), which is 27.5 kJ/mol for CH3NH3PbI3. 3.2 X-ray Diffraction Measurements and SEM images The X-ray patterns of CH3NH3PbI3 and CH3NH3PbI3-xClx films prepared on FTO substrates are shown in Fig. 5(a). The diffraction peaks of both CH3NH3PbI3-xClx and CH3NH3PbI3 films are consistent with the tetragonal phase of CH3NH3PbI321. And no peak shift or broadening is observed in CH3NH3PbI3-xClx film, which revealed that almost paltry chlorine incorporation inside the CH3NH3PbI3 crystal structure22. In addition, the peak intensities of CH3NH3PbI3-xClx were stronger than CH3NH3PbI3 obviously. The SEM image (Fig. 5(b)) of CH3NH3PbI3 film showed the formation of a fibrous structure. However, more uniform and smoother polycrystalline films with less pin holes will be gotten with CH3NH3PbI3-xClx solutions (Fig. 5(c)), which indicated that the crystallization was improved by the addition of chlorine. This is in 13

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 22

accordance with XRD measurements.

Figure 5 (a) XRD pattern of CH3NH3PbI3, CH3NH3PbI3-xClx and FTO, (b) SEM image of CH3NH3PbI3 and (c) SEM image of CH3NH3PbI3-xClx.

3.3 Photo-thermal behaviors Figure 6 (a, b) show the photo-thermal DSC curves of CH3NH3PbI3 and CH3NH3PbI3-xClx at different light intensity. For CH3NH3PbI3 and CH3NH3PbI3-xClx, the exothermic peak reached the maximum immediately as the open of the light, and then remains the same. When the light was turn off, DSC curves dropped to zero at once. Moreover, the exothermic values rise accordingly with the increasing of the light intensity. Comparing their enthalpy change (∆H) shown in Fig. 6(c), CH3NH3PbI3 released more heat than CH3NH3PbI3-xClx under the same illumination conditions, and the difference rose with the increase of light intensity, which is weird since the incorporation of chlorine will not markedly change the crystal structure and the optic absorbance22. This phenomenon is further manifested in Fig. 6(d), which is the enthalpy changes (∆H) of CH3NH3PbI3 and CH3NH3PbI3-xClx illuminated by light with different wavelength, and shows that CH3NH3PbI3 discharged more heat than CH3NH3PbI3-xClx under all wavelength ranges. The CH3NH3PbI3 released 20% more heat than CH3NH3PbI3-xClx, that is CH3NH3PbI3 have 20% more energy translate to heat than CH3NH3PbI3-xClx when absorb the identical illumination condition. Recently, Yang Yet al.23 pointed out that there exists an efficient hot-phonon bottleneck in lead 14

ACS Paragon Plus Environment

Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

iodide perovskites, which slows down the cooling of hot carriers. As a result, the relaxation time is increased to about three orders of magnitude longer compared with GaAs, indicating that the mechanism of photogenerated carrier’s relaxation is unusual. Inspired by this result, we tentatively assume that the carrier relaxation in CH3NH3PbI3-xClx is slower than CH3NH3PbI3, therefore less heat is produced by the photo-induced carrier in CH3NH3PbI3-xClx. Moreover, the more heat released in CH3NH3PbI3 will cause the occurrence of phase transition and so as the decline of the cell efficiency

further,

this may be the

different of CH3NH3PbI3 and

CH3NH3PbI3-xClx.

Figure 6 DSC curves of (a) CH3NH3PbI3, (b) CH3NH3PbI3-xClx at different light intensity and enthalpy of CH3NH3PbI3 and CH3NH3PbI3-xClx at different (c) light intensity and (d) wavelength band.

The photoluminescence of the samples were also measured. Figure 7(a, b) show the PL spectra of CH3NH3PbI3 and CH3NH3PbI3-xClx, excited under light source with different wavelengths. Contrasting Fig. 7(a) to Fig. 7(b), the PL peak have a relatively 15

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

blue-shift in CH3NH3PbI3-xClx, which is due to variation in the band-edge emission24. This suggested that the CH3NH3PbI3 film has a smaller band gap compared to the CH3NH3PbI3-xClx film. Figure 7(c) show the integral of PL peaks, the luminous intensity weakened obviously with the doping of chlorine, though XRD measurements indicated better crystalline of CH3NH3PbI3-xClx film. We have also measured the PL spectra of CH3NH3PbI3 and CH3NH3PbI3-xClx which were deposited on glass substrates (not shown), and the same conclusion was obtained with that on FTO substrate. Therefore, the incorporation of chlorine may promote the separation of electron-hole, thus inhibit the radiation recombination. The photo-thermal and photoluminescence data demonstrate that the addition of chlorine into CH3NH3PbI3 can efficiently suppress the energy loss by emitting heat or by emitting light. These results may explain from a new sight why the addition of chlorine can improve the battery efficiency. However, more transient optical measurements are needed to disclose the mechanisms in detail.

Figure 7 PL spectra of (a) CH3NH3PbI3, (b) CH3NH3PbI3-xClx films at different excitation

wavelength

and

(c)

integral

area

ratio

of

CH3NH3PbI3 and 16

ACS Paragon Plus Environment

Page 17 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

CH3NH3PbI3-xClx.

4 Conclusions The DSC results showed that CH3NH3PbI3-xClx have no phase transition in the range from 25 to 100 oC, indicating that it is more thermal-stable than CH3NH3PbI3 of which there exist phase transition at about 50

o

C. On the other hand, the

photo-thermal study discovered that the heat generated by photon absorbing in CH3NH3PbI3-xClx is 20% less than that in CH3NH3PbI3, which may be due to that the carrier relaxation in CH3NH3PbI3-xClx is slower than CH3NH3PbI3. The more heat released in CH3NH3PbI3 may cause the occurrence of the phase transition and then decline the cell stability. The photoluminescence results also disclosed that the addition of chlorine into CH3NH3PbI3 can efficiently suppress the energy loss through emitting light.

Acknowledgments This work was financially supported by the Chinese Academy of Sciences (211134KYSB20130017),

National

Basic

Research

Program

of

China

(2012CB922000), External Cooperation Program of BIC, Key Research Program of Chinese Academy of Sciences (KGZD-EW-T06), and the Provincial Natural Science Research Project of Anhui Colleges (KJ2014ZD40). Y. L. Lu also feel grateful sustain from the U.S. Air Force Office of Scientific Research (OSR) and DTRA (HDTRA12221). And thanks Professor Qing Yang’s help.

17

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

References (1) Kim, H. S.; Im, S. H.; Park, N.-G. Organolead Halide Perovskite: New Horizons in solar cell research. J. Phys. Chem. C 2014, 118, 5615−5625. (2) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal halide perovskites as visible-light sensitizers for photovoltaic cells. J. Am. Chem. Soc. 2009, 131, 6050-6051. (3) NREL, Best research-cell efficiencies, NREL, 2016. (4) Yamada, Y.; Yamaji, Y.; Imada, M. Exciton lifetime paradoxically enhanced by dissipation and decoherence: toward efficient energy conversion of a solar cell. Phys. Revi. Lett. 2015, 115, 197701 (6). (5) Sum, T. C.; Mathews, N. Advancements in perovskite solar cells: photophysics behind the photovoltaics. Energy & Environmental Science. 2014, 7(8): 2518-2534. (6) Yu, H.; Wang, F.; Xie, F. et al. The role of chlorine in the formation process of “CH3NH3PbI3-xClx” perovskite. Adv. Func. Mater. 2014, 24(45): 7102-7108. (7) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Efficient hybrid solar cells based on meso-superstructured organometal halide perovskites. Science. 2012, 338, 643–647. (8) Burschka, J.; Pellet, N.; Moon, S. J. et al. Sequential deposition as a route to high-performance perovskite-sensitized solar cells. Nature. 2013, 499(7458): 316-319. (9) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J. Electron-hole diffusion lengths exceeding 1 micrometer in an organometal trihalide perovskite absorber. Science. 2013, 342, 341–344. (10) Wehrenfennig, C.; Eperon, G. E.; Johnston, M. B.; Snaith, H. J.; Herz, L. M. High charge carrier mobilities and lifetimes in organolead trihalide perovskites. Adv. Mater. 2013, 26, 1584–1589. (11) Colella, S.; Mosconi, E.; Fedeli, P.; Listorti, A.; Gazza, F.; Orlandi, F.; Ferro, P.; Besgni, T.; Rizzo, A.; Calestani, G.; Gigli, G.; Angelis, F. De; Mosca, R. 18

ACS Paragon Plus Environment

Page 19 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

MAPbI3-xClx mixed halide perovskite for hybrid solar cells: the role of chloride as dopant on the transport and structural properties. Chem. Mater. 2013, 25, 4613–4618. (12) Dualeh, A.; Gao, P.; Seok, S. I. et al. Thermal behavior of methylammonium lead-trihalide perovskite photovoltaic light harvesters. Chem. Mater. 2014, 26(21): 6160-6164. (13) Hay, J.N.; Fitzgerald, P.A.; Wiles, M. Use of differential scanning calorimetry to study polymer crystallization kinetics. Polymer. 1976, 17, 1015-1018. (14) Joo, Y. L.; Sun, J.; Smith, M, D. et al. Two-dimensional numerical analysis of non-isothermal melt spinning with and without phase transition. J. Non-Newt. Fluid Mechanics. 2002, 102 (1):37-70. (15) Jeziorny, A. Parameters characterizing the kinetics of the non-isothermal crystallization of poly (ethylene terephthalate) determined by DSC. Polymer. 1978, 19(10): 1142-1144. (16) Liu, T.; Mo, Z.; Wang, S.; Zhang, H. Nonisothermal melt and cold crystallization kinetics of poly(aryletheretherketoneketone). Polym. Eng. Sci. 1997, 37, 568-575. (17) Avrami, M. Granulation, phase change, and microstructure kinetics of phase change. III. J. Chem. Phys. 1941, 9(2): 177-184. (18) Wang, F. F.; Zheng, K.; Yao, X. Y.; Chen, L.; Tian, X. Y. Synthesis of poly(butylene

terephthalate)/

attapulgite

nanocomposites

by

in-situ

polymerization and its crystallization behavior. J. Macromol. Sci., Part B: Phys. 2015, 54:7, 862-878. (19) Gao, H.; Bao, C.; Li, F. et al. Nucleation and crystal growth of organic–inorganic lead halide perovskites under different relative humidity. ACS applied materials & interfaces. 2015, 7(17), 9110-9117. (20) Cebe, P.; Hong, S. D. Crystallization behaviour of poly (ether-ether-ketone). Polymer. 1986, 27(8), 1183-1192. (21) Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Semiconducting tin and lead iodide perovskites with organic cations: phase transitions, high mobilities, 19

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

and near-infrared photoluminescent properties. Inorg. Chem. 2013, 52(15), 9019-9038. (22) Natalia, Y.; Fang, Y.; Chen, S.; Herlina, A. D.; Pablo, P. B.; Subodh, G. M.; Nripan, M. Unravelling the effects of Cl addition in single step CH3NH3PbI3 perovskite solar cells. Chem. Mater. 2015, 27, 2309−2314. (23) Yang, Y.; Ostrowski, D. P.; France, R. M. et al. Observation of a hot-phonon bottleneck in lead-iodide perovskites. Nature Photon. 2015, 10, 53-59. (24) Luo, D.; Yu, L.; Wang, H. et al. Cubic structure of the mixed halide perovskite CH3NH3PbI3-xClx via thermal annealing. RSC Adv. 2015, 5(104), 85480-85485.

20

ACS Paragon Plus Environment

Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Table of Contents: Table 1 Kinetics parameters for non-isothermal crystallization calculated with Mo and Jeziorny model of CH3NH3PbI3. Table 2 Kinetic parameters for isothermal crystallization of CH3NH3PbI3. Figure 1 Schematic illustrating of the energy flowing way. Figure 2 The DSC thermograms at different cooling rate for (a) CH3NH3PbI3, (c) CH3NH3PbI3-xClx and (b) Plots of Tonset, Tp, Tf as a function of cooling rate. Figure 3 X(t) as a function of (a) temperature T and (b) time t, kinetics parameters from (c) Jeziorny and (d) Mo model for non-isothermal crystallization of CH3NH3PbI3. Figure 4 Isothermal crystallization graph of CH3NH3PbI3: (a) Schematic illustration of evolutions of crystalline morphologies; (b) X(t) as a function of t; (c) Plots of ln{-ln[1-X(t)]} vs. ln t; (d) Plots of ln k vs. 1/Tc. Figure 5 (a) XRD pattern of CH3NH3PbI3, CH3NH3PbI3-xClx and FTO, (b) SEM image of CH3NH3PbI3 and (c) SEM image of CH3NH3PbI3-xClx. Figure 6 DSC curves of (a) CH3NH3PbI3, (b) CH3NH3PbI3-xClx at different light intensity and enthalpy of CH3NH3PbI3 and CH3NH3PbI3-xClx at different (c) light intensity and (d) wavelength band. Figure 7 PL spectra of (a) CH3NH3PbI3, (b) CH3NH3PbI3-xClx films at different excitation

wavelength

and

(c)

integral

area

ratio

of

CH3NH3PbI3 and

CH3NH3PbI3-xClx.

21

ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 22

TOC graphic

(a) Jeziorny and (b) Mo model for non-isothermal crystallization of CH3NH3PbI3, (c) JMA model for isothermal crystallization graph of CH3NH3PbI3, (d) enthalpy of CH3NH3PbI3 and CH3NH3PbI3-xClx at different light intensity.

22

ACS Paragon Plus Environment