Subscriber access provided by READING UNIV
Article
Kinetic isotope effects provide experimental evidence for proton tunneling in methylammonium lead triiodide perovskites Yan-Fang Chen, Yu-Tang Tsai, Lionel Hirsch, and Dario M. Bassani J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b09526 • Publication Date (Web): 25 Oct 2017 Downloaded from http://pubs.acs.org on October 25, 2017
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.
Journal of the American Chemical Society 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 23
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
Journal of the American Chemical Society
Kinetic isotope effects provide experimental evidence for proton tunneling in methylammonium lead triiodide perovskites Yan-Fang Chen, Yu-Tang Tsai, Lionel Hirsch,* Dario M. Bassani* Univ. Bordeaux, CNRS, UMR 5218, IMS, F-33400 Talence, France Univ. Bordeaux CNRS UMR 5255, ISM, F-33405 Talence, France. Hybrid perovskites, proton migration, tunneling, isotope effect
ABSTRACT:
The occurrence of proton tunneling in MAPbI3 hybrid organic inorganic
perovskites is demonstrated through the effect of isotopic labeling of the methylammonium (MA) component on the dielectric permittivity response. Deuteration of the ammonium group results in the acceleration of proton migration (inverse primary isotope effect), whereas deuteration of the methyl group induces a normal secondary isotope effect. The activation energies for proton migration are calculated to be 50 and 27 meV for the tetragonal and orthorhombic phases, respectively, which decrease upon deuteration of the ammonium group. The low activation barrier, and the deviation from unity of the ratio of the pre-exponential factors (AH/AD = 0.3 – 0.4) are consistent with a tunneling mechanism for proton migration. Deuteration
ACS Paragon Plus Environment
1
Journal of the American Chemical Society
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 23
of the PEDOT:PSS hole transport layer results in a behavior that is intermediate between that of the deuterated and undeuterated perovskite, due to extrinsic ion migration between the two materials.
INTRODUCTION Due to their rapid increase in performance, there is considerable interest in the use of hybrid organic-inorganic perovskites (HOIPs, Figure 1), in particular methylammonium lead halide perovskites (MAPbX3), as solution-processable materials for solar energy conversion devices. Intense research effort has rapidly brought the power conversion efficiency from an initial 3.8%1 to over 20%2 in mixed halide systems. While the toxicity of lead and long-term device stability are still major obstacles to be overcome for their widespread deployment in solar cells and electroluminescent devices, HOIPs represent a class of materials that combine promising electronic properties with solution processing technology.3 Perovskites in general are known to possess ionic conductivity4 and ion movement has been postulated to be directly or indirectly responsible for their lack of long-term stability and unusual behavior, including hysteresis.5 Therefore, understanding ion migration in perovskites is of major importance in improving their performance.6 Ion movement in HOIPs may occur extrinsically, i.e. from one of the contacts,7 or take the form of ion migration, where an ion is displaced from an occupied to a vacant site. Several theoretical investigations concur in proposing that iodine is the most mobile ion with respect to migration, with calculated activation energies that are systematically lower than those for Pb2+ or methylammonium (MA) ions.8 Egger et al.9 have calculated that hydrogen, due to its small size, can migrate between interstitial sites resulting in a net migration of defects but,
ACS Paragon Plus Environment
2
Page 3 of 23
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
Journal of the American Chemical Society
although other perovskite materials are well known proton conductors,10 there is as yet no experimental evidence for proton migration in HOIPs. Not surprisingly, direct experimental evidence for ion movement in thin film HOIP devices has proven difficult to obtain.11 Numerous studies confirmed the rotational movement of the MA ions within the inorganic cage formed by the PbI2 octahedra,12 which has been associated to the temperature and frequency dependence of the complex permittivity.12e,13 The plausible agreement between MA rotation rates and the high frequency response of the dielectric constant relaxation supports this hypothesis and dielectric studies of HOIP materials are often interpreted based on the assumed dynamic orientational disorder of the MA dipoles.14 However, although the low frequency response (< 102 Hz) has been assigned to ion conduction involving iodide,11a there is an intermediate frequency response generally observed at ca. 1 – 3 kHz (Figure 1) that has so far eluded assignment.11a,15
It has been loosely attributed by different groups to chemical
capacitance in the device,16 to the existence of microscopic dipolar units,17 or to the migration of defects.8b To test the relationship between MA rotation and the intermediate frequency dielectric response in HOIP materials, we conducted a series of experiments in which the hydrogen atoms in MAI were selectively substituted with deuterium. Isotopic labeling does not directly affect the electronic properties of the material and is typically used in chemistry and in biology to evidence molecular sites that participate in the transition state of a reaction. Generally, replacement of a hydrogen atom by deuterium results in a slower reaction rate due to the smaller zero point energy (ZPE) of bonds incorporating D vs. H (normal primary isotope effect). Secondary isotope effects are instead much smaller and are observed when isotopic labeling is conducted at a site that is adjacent to the reaction center. Our results unambiguously demonstrate that MA migration or
ACS Paragon Plus Environment
3
Journal of the American Chemical Society
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 23
rotation is not directly responsible for the frequency dependence of the dielectric constant at intermediate frequencies at 1 – 3 kHz. Instead, we observe that deuteration of the ammonium group leads to a large inverse kinetic isotope effect (KIE), and show that proton tunneling is responsible for the mid-range frequency-dependence of the dielectric constant in HOIP materials. Deuteration of the methyl group in MA induces a normal secondary KIE that is consistent with these findings.
Figure 1. General structure of HOIPs (top) and temperature dependence of the dielectric constant (ε', black curve). The presence of an intermediate relaxation phenomenon at 3 kHz is clearly visible. RESULTS AND DISCUSSION
ACS Paragon Plus Environment
4
Page 5 of 23
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
Journal of the American Chemical Society
The activation energy of the relaxation phenomena (Ea) at mid-range frequency was determined from its temperature dependence according to the Arrhenius equation. In general, the values of Ea we find are similar to those determined by neutron scattering (27 to 50 meV) which were assigned to MA rotation.18 We proceeded to change the mass of the MA ion by replacement of the hydrogen atom by deuterium. The relative difference between the mass of the deuterated and the non-deuterated forms of MA is significant (up to 19%), and this is expected to affect the rotational (or translational) dynamics of the MA ion according to Table 1. The perdeuterated molecular ion CD3ND3+ (D-D) possesses the largest rotational momentum and would therefore show the slowest rotational dynamics. Also, H-bonds involving a deuterium atom are stronger than those based on 1H. We may therefore expect that if H-bonding is present in the material, it will contribute to raising the activation barrier for rotation for those MA where deuteration is located on the amine nitrogen. In contrast, the two partially deuterated species (H-D and D-H) possess similar rotational momenta, but only the H-D species will show a primary isotope effect on H-bonding. An analogous argument can be made for the rotation of the CD3 vs CH3 or ND3 vs. NH3 fragments, with similar conclusions. Deuterated MAI samples were prepared in a straightforward manner by H / D exchange in D2O from commercially available CH3NH3I and CD3NH3Cl (see SI) and characterized by 1H, 2
H, and 13C NMR, all of which are consistent with the expected deuteration pattern. In agreement
with detailed analysis of the crystalline phase transitions of isotopic MAPbI3 (CH3NH3PbI3, CH3ND3PbI3, and CD3ND3PbI3),19 we find that all four samples possess identical crystalline packing as evidenced by X-ray diffraction experiments. The samples are also indistinguishable from UV–vis absorption and electroluminescence, showing that they do not differ electronically. Details of the characterization can be found in SI. As a test, the compounds were also used to
ACS Paragon Plus Environment
5
Journal of the American Chemical Society
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
prepare
planar
perovskite
solar
cells
with
a
straightforward
Page 6 of 23
ITO/PEDOT-
PSS/MAPbI3/PCBM/Ca/Al architecture. As expected, we find that the deuteration of the MA does not affect the performance of the solar cells (avg. efficiency of 12%, see SI). We do note some variability in the device parameters which we attribute to statistical variations, but these were not investigated further as a simplified device architecture was instead used for the temperature dependent admittance spectroscopy to reduce as the internal electric field under short circuit conditions. To this end, a top gold electrode was thermally deposited on the perovskite layer to obtain an ITO/PEDOT-PSS/MAPbI3/Au device architecture. In previous work, we showed that the gold electrode deposition induces only shallow traps at this interface.20
Table 1. Deuterated MA ions and their translational and inertia moments. Translational moment (a.m.u.)
Moment of inertia (a.m.u. Å2)
H-H
32
36.3
H-D
35
39.1
D-H
35
39.7
D-D
38
43.2
Sample
Formula
Rotor
The dielectric measurements were carried out as function of frequency from 10 Hz to 1 MHz and as function of temperature from 270 K to 90 K. Over this temperature and frequency region, the contribution of the low frequency ion migration is minimal, allowing clear identification of
ACS Paragon Plus Environment
6
Page 7 of 23
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
Journal of the American Chemical Society
the intermediate relaxation phenomena. At lower temperatures, the relaxation overlaps with the larger contributions from ion migration and cannot be readily isolated. The characteristic relaxation frequency represents the frequency at which the rate of exchange of the system between two states is equal, and this can be determined from the plot of ε'(f) from either the intersection of 2 tangential lines, or from the lowest point on the plot of the second derivative of the permittivity vs. frequency. Both methods yield the same values (see SI), and the effect of tetragonal-to-orthorhombic phase transition occurring at 162 K for MAPbI3 can be observed as a jump in the characteristic frequency. In both the tetragonal and orthorhombic phases, the relaxation frequency exhibits an Arrhenius behavior (Figure 2). The activation energies (Ea) and pre-exponential factors (A) determined for each sample in either crystalline phase are reported in Table 2. These activation energies are low compared to those associated with ion migration, which are typically on the order of hundreds of meV.4,8b,21 On the other hand, the activation energies for the rotation of organic cations in MAPbI3 are similar to these values.12a,18,22 While the exact value varies (from a few meV up to 100 meV), all reports agree that MA rotation is restricted in the low temperature orthorhombic phase due to steric hindrance. The same behavior is observed for the intermediate relaxation process for the four deuterated and non-deuterated MAPbI3.
ACS Paragon Plus Environment
7
Journal of the American Chemical Society
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 23
Figure 2. Comparison of the Arrhenius plots of the intermediate relaxation frequencies of deuterated and non-deuterated MAPbI3 samples: H-H (filled black circles), H-D (filled red circles), D-H (filled green circles), and D-D (filled blue circles). Shaded area shows temperature region of the tetragonal to orthorhombic phase transition.
Table 2. Summary of the pre-exponential factors (A) and activation energies (Ea) of the intermediate relaxation of ITO/PEDOT:PSS/MAPbI3/Au devices with deuterated/non-deuterated MAPbI3.a Tetragonal phase
Orthorhombic phase
Sample
A /kHz
Ea /meV
A /kHz
Ea /meV
H-H
7 (1)
50 (3)
5.1 (0.2)
27 (1)
H-D
17 (2)
43 (2)
4 (1)
14 (3)
D-H
6 (1)
57 (4)
2.1 (0.3)
29 (2)
D-D
20 (3)
54 (3)
4.9 (0.8)
21 (2)
ACS Paragon Plus Environment
8
Page 9 of 23
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
Journal of the American Chemical Society
a
The error bar of the values from the standard deviations of linear fitting is given in parenthesis. To our surprise, deuteration of the ammonium group (H-D) actually results in increase in the characteristic relaxation frequency compared to the non-deuterated H-H sample.
This
acceleration of the process is contrary to what is expected based on the moment of inertia of compounds H-D vs. H-H (Table 1) and is indicative of an inverse isotope effect. The difference in reactivity of D vs. H was first proposed by Cremer and Polanyi23 and by Eyring24 based on the difference in ZPE between the two isotopes. According to transition state theory, the lower ZPE of D contributes to increase the activation energy which results in a lower reaction rate. The magnitude of the KIE is then defined as the ratio of kH / kD which is largest and greater than unity when isotopic labeling directly involves the reaction center. In such cases, values of KIE are typically of 2 – 6, and can exceed 7 or more for some reactions.25 A value of KIE less than unity is termed an inverse isotope effect and reflects the observation that the heavy isotope reacts faster than the lighter one. In comparing the characteristic relaxation frequency for H-H and HD at 270K, we can determine KIE = 0.30 (270 K), which corresponds to a large, inverse, isotope effect. This is reflected in both the pre-exponential factor (A), which is significantly larger for HD vs. H-H (17 vs. 7 kHz, respectively), and the activation energy (Ea), which is lower for H-D vs. H-H (43 vs. 50 meV, respectively). The observation of an inverse isotope effect generally implies the occurrence of a preequilibrium mechanism that scales the observed KIE by an equilibrium isotope effect.26 Other situations leading to an inverse KIE include a narrower potential energy surface for the transition state (Figure 3a) or the occurrence of an ionic H-bond formed between A•••H+•••A (where A is an electronegative atom) due to the lower difference in ZPE between the initial state and transition state (Figure 3b).27 The ionic nature of perovskites and the presence of charged MA ions support
ACS Paragon Plus Environment
9
Journal of the American Chemical Society
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 23
the latter mechanism, thereby assigning the origin of the intermediate relaxation phenomena to the motion of H+ in the HOIP. This is consistent with the observation of a primary KIE for the breaking of a H2N+–H bond. With respect to the mechanism of proton migration (classical vs. tunneling), we note that the process is active even at temperatures as low as 90 K, and that it is characterized by a low probability as reflected by the small value of A (logA = 3.8 and 4.2 for HH and H-D, respectively). This behavior is associated with quantum tunneling, and further evidence for tunneling stems from the ratio of pre-exponential factors, which tends towards unity for non-tunneling regimes. The ratio of the pre-exponential factors for H-H and H-D is 0.41, in agreement with a moderate tunneling regime as it lies beyond the lower limit of 0.5 admitted for semiclassical isotope effects.28 Deuteration of the CH3 fragment in MA (D-H, Table 1) should provide the same change in the moment of inertia as for the H-D sample discussed above, but without affecting proton migration as the hydrogens in CH3 are not labile. Instead, we find that KIE = 1.6 and that Ea is increased upon deuteration (from 50 to 57 meV). The low value of the KIE is typical for a normal (KIE > 1) secondary isotope effect, where isotopic substitution occurs at a site whose vibrational motions are not directly associated with the reaction coordinate. This further supports the hypothesis that migration of the NH3 protons is involved in the intermediate relaxation phenomena. We may also compare the behavior of the fully deuterated MA (D-D) with D-H to again focus on the effect of deuteration on the nitrogen site. The observed KIE = 0.26 is in good agreement with the result from the H-D vs. H-H system evidencing yet again an inverse isotope effect associated with the putative occurrence of an ionic H-bond formed during proton migration.
The mechanism again possesses a tunneling contribution as suggested by the
deviation from unity of the ratio of the pre-exponential factors (AD-H / AD-D = 0.30).
ACS Paragon Plus Environment
10
Page 11 of 23
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
Journal of the American Chemical Society
Figure 1. Illustration of inverse isotope effect caused by (a) curvature at the saddle point or by (b) the occurrence of an ionic H-bond between two electronegative atoms A (adapted with permission from reference 27). Both of these phenomena lead to a reduction of Ea for the heavier vs. lighter isotope. The situation for the orthorhombic phase is analogous to that of the tetragonal phase described above. In short, an inverse primary KIE is observed on going from H-H to H-D or from D-H to D-D, whereas a normal secondary KIE is present in going from H-H to D-H or from H-D to DD. Both the Ea and A are lower in the orthorhombic phase than in the tetragonal phase, suggesting that the decreased freedom of the MA directly affects the proton migration governing the intermediate relaxation phenomena. Interestingly, the ratio of the pre-exponential factors remains below 0.5 for AD-H / AD-D = 0.43, indicative of tunneling, but tends towards unity for AHH
/ AH-D = 1.1. This does not necessarily indicate a lower contribution from tunneling in the later
case, however, as recently discussed by Sharma and Klingman.29 Furthermore, we may remark that the values of the secondary KIEs for both the tetragonal and the orthorhombic phases are
ACS Paragon Plus Environment
11
Journal of the American Chemical Society
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 23
temperature-dependent, as generally observed for proton tunneling in enzymatic reactions.30 This difference in behavior between the inverse primary and normal secondary KIE can be seen in Figure 4.
Figure 4. Temperature-dependence of the inverse primary (H-H vs. H-D and D-H vs. D-D, blue and red filled circles, respectively) and normal secondary (H-H vs. D-H and H-D vs. D-D, black and purple filled triangles, respectively) kinetic isotope effect (KIE).
Direct experimental investigation of proton migration HOIPs has largely gone unreported.8a In contrast, the effect of proton migration in inorganic perovskite oxides has been investigated using hydrogen-31 or moisture-containing environments.32 Simulations suggest that the mechanism involves protons hopping between neighboring oxygen atoms while forming transitional hydrogen bonds, thereby causing a reorientation of the OH dipoles.33 In HOIPs, the migration of protons has been considered theoretically by Egger et al.8c,9 using minimum energy path calculation. However, the direct observation of the migration is challenging,8a possibly due to the instability of HOIPs under water or hydrogen containing atmosphere. The simulation
ACS Paragon Plus Environment
12
Page 13 of 23
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
Journal of the American Chemical Society
studies by Egger et al. show that protons migrate in MAPbI3 by forming transient hydrogen bonds with iodides, with a barrier as low as 170 meV.9 The authors propose that this barrier would be further reduced if nuclear quantum effects such as tunneling are taken into consideration. Indeed, the quantum effect on proton conduction is a usual occurrence in the case of perovskite oxides,34 or metals.35 In light of the theoretical simulations reported by Egger et al., which considered deprotonation of the MA ion as a possible source of migrating hydrogen, it is reasonable to assume that proton migration is coupled to the MA ion rotation.9
This
hypothesis is consistent with the observation that the proton transition in the tetragonal phase is faster than in the orthorhombic phase where MA movement is hindered even though the Ea in the orthorhombic phase is smaller. We may further surmise that the proton movement occurring under the experimental conditions of an external oscillating electric field is a reversible shuttling between two sites, one of which is necessarily located on MA to account for the observation of a secondary KIE upon deuteration of the CH3 fragment. The second site involved in the exchange may be a vicinal iodide as it constitutes a potential energy well for protons according to calculations. Quantitative analysis of the admittance graphs provides information regarding the total number of charges that respond to the applied oscillating electric field. By investigating the effect of device thickness on the response of the intermediate relaxation phenomena, we can determine if the observed proton movement is a bulk or interfacial phenomena, and whether it is intrinsic to the material or due to defects. To this end, the intermediate relaxation was measured on samples with thickness varying between 185 – 300 nm (Figure S8). Both the characteristic relaxation frequency and the magnitude of the intermediate relaxation phenomena are found to be independent of the film thickness, suggesting that the proton movement is principally located
ACS Paragon Plus Environment
13
Journal of the American Chemical Society
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 23
at an interface. The quantity of mobile protons can then be estimated from the capacitance of the intermediate relaxation (13.5 nF for an 8-mm2 device), which is scaled by the device area to afford a surface density of mobile protons ≈ 2 x 1010 cm–2. The surface density of MA ions can be obtained from the crystal packing to be ca. 1013 cm–2, suggesting that proton movement occurs at interfacial defect sites affecting about 0.1% of the MA ions at the contact / HOIP interface. To identify which interface is responsible for the formation of defects leading to mobile protons, we proceeded to modify the sample structure and in order to cross-reference the interfaces. To do so, a thin insulating aluminum oxide layer was used to selectively passivate the contacts of MAPbI3 layer. In this experiment, a 45-nm thick Al2O3 layer was e-beam evaporated under high vacuum (~ 10–7 mbar). Control experiments (see SI) were performed to eliminate the possibility of a response by the added oxide layer in the frequency range of interest. The capacitance spectra of ITO/Al2O3/MAPbI3/Au, ITO/PEDOT:PSS/MAPbI3/Al2O3/Al, and ITO/Al2O3/PEDOT:PSS/MAPbI3/Al2O3/Al devices are shown in Figure S10. The intermediate relaxation phenomenon is absent in the case of the ITO/Al2O3/MAPbI3/Au device but present in the other 2 samples, though reduced in magnitude due to the series capacitor effect. From this, we conclude that the defects responsible for the mobile protons accumulate near the MAPbI3/PEDOT:PSS interface. The accumulation of defects at MA sites near the PEDOT:PSS layer is not altogether unreasonable as the polymeric hole transport layer is highly acidic and is known to be an ion conductor36 in which the protons from the dissociation of the PSS acid may diffuse into the top layers.37 To verify whether this is occurring in HOIPs, we investigated the effect of deuteration of the PEDOT:PSS layer on the proton mobility as evidenced by the intermediate relaxation
ACS Paragon Plus Environment
14
Page 15 of 23
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
Journal of the American Chemical Society
phenomena. Deuterated PEDOT:PSS (PEDOT:d-PSS) was obtained through deuterium exchange in D2O (see SI), and the temperature dependence of the relaxation frequencies of the PEDOT:d-PSS experiments are shown in Figure 5. The behavior of the PEDOT:dPSS/CH3NH3PbI3 and PEDOT:d-PSS/CH3ND3PbI3 samples mirror that of H-D in that they exhibit an inverse KIE. The magnitude of the effect for the PEDOT:d-PSS/CH3NH3PbI3 sample is intermediate that of H-H and H-D both for the tetragonal and the orthorhombic phases, confirming the exchange of protons between the MAPbI3 material and the PEDOT:PSS layer. This observation confirms the occurrence of extrinsic ion migration from the PEDOT:PSS layer to the HOIP. In contrast, the response of the PEDOT:d-PSS/CH3ND3PbI3 sample is near that observed for H-D, suggesting that the bulk of the dielectric response originates from the perovskite material and that it is only marginally affected by exchange with the underlying PEDOT:PSS layer.
ACS Paragon Plus Environment
15
Journal of the American Chemical Society
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 23
Figure 5. Arrhenius plots of the relaxation frequencies of MAPbI3 on PEDOT:dPSS/CH3NH3PbI3 sample (filled purple diamonds) show that its behavior is intermediate between that of the H-H sample (filled blue circles) and that of the H-D sample (filled red circles). The response of PEDOT:d-PSS/CH3ND3PbI3 sample (filled black diamonds) is similar to that of the H-D sample. Shaded area shows temperature region of the tetragonal to orthorhombic phase transition. CONCLUSIONS Deuteration of methylammonium iodide in HOIP provides direct evidence for the occurrence of proton migration in MAPbI3 perovskites used in solar cells. Primary inverse KIEs are observed upon deuteration of the protons of the ammonium fragment, whereas a secondary normal isotope effect is observed upon deuteration of the methyl fragment. This is consistent with reversible shuttling of the labile N–H protons between the amine nitrogen atom and a nearby site. Although isotopic labeling is frequently used in biology to determine enzymatic reaction
ACS Paragon Plus Environment
16
Page 17 of 23
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
Journal of the American Chemical Society
pathways involving proton movement and tunneling, this is the first time, to the best of our knowledge, that it is applied to the investigation of proton movement in HOIP materials. This has allowed us to assign the mid-frequency dielectric response in HOIPs to proton motion through a tunneling mechanism. In agreement with this assignment, examination of the frequency dependence of the dielectric response of CsPbBr3, a perovskite that does not contain labile protons, shows that the mid-frequency dielectric response is absent.38 We determine the Ea associated with proton migration to be 50 and 27 meV for the tetragonal and orthorhombic phase, respectively, which decreases substantially upon deuteration of the ammonium group. Proton migration is principally localized at the interfacial region with the PEDOT:PSS layer which is shown through isotopic labeling to participate in proton exchange with the MAPbI3 material (extrinsic ion migration).
The small Ea, the ratio of the pre-
exponential parameters, and the temperature dependence of the KIE, all support the intervention of quantum tunneling in the mechanism for proton transport in HOIPs. These results thus provide the first experimental evidence for proton movement in HOIPs, which we believe will be of value in understanding the complex behavior of this fascinating class of materials. More importantly, we show that chemistry, through isotopic labeling, provides a facile and very important methodology for investigating proton migration in HOIPs which can be readily applied to the elucidation of interfacial exchange processes.
ASSOCIATED CONTENT Supporting Information. The following files are available free of charge.
ACS Paragon Plus Environment
17
Journal of the American Chemical Society
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 23
Synthesis and characterization of labeled MAI, HOIP material (X-ray diffraction, UV-Vis and electroluminescence), and further device characterization. (PDF) AUTHOR INFORMATION Corresponding Authors *
[email protected] *
[email protected] ACKNOWLEDGMENT We are grateful to financial support from the LabEx AMADEus (ANR-10-LABX-0042AMADEUS through grant ANR-10-IDEX-0003-02). We thank Dr. I. Pianet for assistance with the 13C and 2H NMR spectroscopy. REFERENCES (1)
Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am. Chem. Soc. 2009, 131, 6050-
6051. (2)
Saliba, M.; Matsui, T.; Seo, J.-Y.; Domanski, K.; Correa-Baena, J.-P.; Nazeeruddin, M.
K.; Zakeeruddin, S. M.; Tress, W.; Abate, A.; Hagfeldt, A.; Gratzel, M. Energy Environ. Sci. 2016, 9, 1989-1997. (3)
(a) Nie, W.; Tsai, H.; Asadpour, R.; Blancon, J.-C.; Neukirch, A. J.; Gupta, G.; Crochet,
J. J.; Chhowalla, M.; Tretiak, S.; Alam, M. A.; Wang, H.-L.; Mohite, A. D. Science 2015, 347, 522-525; (b) Pedesseau, L.; Sapori, D.; Traore, B.; Robles, R.; Fang, H.-H.; Loi, M. A.; Tsai, H.; Nie, W.; Blancon, J.-C.; Neukirch, A.; Tretiak, S.; Mohite, A. D.; Katan, C.; Even, J.; Kepenekian, M. ACS Nano 2016, 10, 9776-9786; (c) Berry, J.; Buonassisi, T.; Egger, D. A.;
ACS Paragon Plus Environment
18
Page 19 of 23
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
Journal of the American Chemical Society
Hodes, G.; Kronik, L.; Loo, Y.-L.; Lubomirsky, I.; Marder, S. R.; Mastai, Y.; Miller, J. S.; Mitzi, D. B.; Paz, Y.; Rappe, A. M.; Riess, I.; Rybtchinski, B.; Stafsudd, O.; Stevanovic, V.; Toney, M. F.; Zitoun, D.; Kahn, A.; Ginley, D.; Cahen, D. Adv. Mat. 2015, 27, 5102-5112. (4)
Eames, C.; Frost, J. M.; Barnes, P. R.; O’regan, B. C.; Walsh, A.; Islam, M. S. Nat.
Commun. 2015, 6, 7497. (5)
(a) Snaith, H. J.; Abate, A.; Ball, J. M.; Eperon, G. E.; Leijtens, T.; Noel, N. K.; Stranks,
S. D.; Wang, J. T. W.; Wojciechowski, K.; Zhang, W. J. Phys. Chem. Lett. 2014, 5, 1511; (b) Calado, P.; Telford, A. M.; Bryant, D.; Li, X.; Nelson, J.; O'Regan, B. C.; Barnes, P. R. F. Nat. Commun. 2016, 7, 13831; (c) Wang, Y.; Zhang, Y.; Pang, T.; Xu, J.; Hu, Z.; Zhu, Y.; Tang, X.; Luan, S.; Jia, R. Phys. Chem. Chem. Phys. 2017, 19, 13002-13009; (d) Seol, D.; Jeong, A.; Han, M. H.; Seo, S.; Yoo, T. S.; Choi, W. S.; Jung, H. S.; Shin, H.; Kim, Y. Adv. Funct. Mater. 2017, 27, 1701924. (6)
(a) Chen, M.; Shan, X.; Geske, T.; Li, J.; Yu, Z. ACS Nano 2017, 11, 6312-6318; (b)
Zhang, T.; Long, M.; Yan, K.; Qin, M.; Lu, X.; Zeng, X.; Cheng, C. M.; Wong, K. S.; Liu, P.; Xie, W.; Xu, J. Adv. Energy Mater. 2017, 1701235. (7)
Li, Z.; Xiao, C.; Yang, Y.; Harvey, S. P.; Kim, D. H.; Christians, J. A.; Yang, M.; Schulz,
P.; Nanayakkara, S. U.; Jiang, C.-S.; Luther, J. M.; Berry, J. J.; Beard, M. C.; Al-Jassim, M. M.; Zhu, K. Energy Environ. Sci. 2017, 10, 1234-1242. (8)
(a) Yuan, Y.; Huang, J. Acc. Chem. Res. 2016, 49, 286-293; (b) Azpiroz, J. M.; Mosconi,
E.; Bisquert, J.; De Angelis, F. Energy Environ. Sci. 2015, 8, 2118; (c) Egger, D. A.; Rappe, A. M.; Kronik, L. Acc. Chem. Res. 2016, 49, 573-581; (d) deQuilettes, D. W.; Zhang, W.; Burlakov, V. M.; Graham, D. J.; Leijtens, T.; Osherov, A.; Bulović, V.; Snaith, H. J.; Ginger, D. S.; Stranks, S. D. Nat. Commun. 2016, 7, 11683.
ACS Paragon Plus Environment
19
Journal of the American Chemical Society
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
(9)
Egger, D. A.; Kronik, L.; Rappe, A. M. Angew. Chem. Int. Ed. 2015, 54, 12437-12441.
(10)
(a) Schober, T.; Condon, J. B. Ionics 1996, 2, 132-136; (b) Matsushita, E.; Tanase, A.
Page 20 of 23
Solid State Ionics 1997, 97, 45-50; (c) Matsushita, E.; Sasaki, T. Solid State Ionics 1999, 125, 31-37. (11)
(a) Yang, T. Y.; Gregori, G.; Pellet, N.; Grätzel, M.; Maier, J. Angew. Chem. Int. Ed.
2015, 54, 7905; (b) Yuan, Y.; Chae, J.; Shao, Y.; Wang, Q.; Xiao, Z.; Centrone, A.; Huang, J. Adv. Energy Mater. 2015, 5, 1500615. (12)
(a) Leguy, A. M. A.; Frost, J. M.; McMahon, A. P.; Sakai, V. G.; Kockelmann, W.; Law,
C.; Li, X.; Foglia, F.; Walsh, A.; O/'Regan, B. C.; Nelson, J.; Cabral, J. T.; Barnes, P. R. F. Nat Commun 2015, 6, 8124; (b) Bakulin, A. A.; Selig, O.; Bakker, H. J.; Rezus, Y. L. A.; Müller, C.; Glaser, T.; Lovrincic, R.; Sun, Z.; Chen, Z.; Walsh, A.; Frost, J. M.; Jansen, T. L. C. Acc. Chem. Res. 2015, 6, 3663-3669; (c) Mosconi, E.; Quarti, C.; Ivanovska, T.; Ruani, G.; De Angelis, F. Phys. Chem. Chem. Phys. 2014, 16, 16137-16144; (d) Wasylishen, R. E.; Knop, O.; Macdonald, J. B. Solid State Commun. 1985, 56, 581-582; (e) Poglitsch, A.; Weber, D. J. Chem. Phys. 1987, 87, 6373-6378. (13)
(a) Frost, J. M.; Walsh, A. Acc. Chem. Res. 2016, 49, 528-535; (b) Brivio, F.; Walker, A.
B.; Walsh, A. Appl. Mater. 2013, 1, 042111. (14)
Chen, Y.-F.; Tsai, Y.-T.; Bassani, D. M.; Hirsch, L. Appl. Phys. Lett. 2016, 109, 213504.
(15)
(a) Dualeh, A.; Moehl, T.; Tétreault, N.; Teuscher, J.; Gao, P.; Nazeeruddin, M. K.;
Grätzel, M. ACS Nano 2014, 8, 362-373; (b) Sanchez, R. S.; Gonzalez-Pedro, V.; Lee, J.-W.; Park, N.-G.; Kang, Y. S.; Mora-Sero, I.; Bisquert, J. J. Phys. Chem. Lett. 2014, 5, 2357-2363.
ACS Paragon Plus Environment
20
Page 21 of 23
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
Journal of the American Chemical Society
(16)
(a) Kim, H.-S.; Ivan, M.-S.; Gonzalez-Pedro, V.; Fabregat-Santiago, F.; Juarez-Perez, E.
J.; Park, N.-G.; Bisquert, J. Nat. Commun. 2013, 4, 2242; (b) Bisquert, J. Phys. Chem. Chem. Phys. 2003, 5, 5360-5364. (17)
Almora, O.; Zarazua, I.; Mas-Marza, E.; Mora-Sero, I.; Bisquert, J.; Garcia-Belmonte, G.
J. Phys. Chem. Lett. 2015, 6, 1645-1652. (18)
Chen, T.; Foley, B. J.; Ipek, B.; Tyagi, M.; Copley, J. R. D.; Brown, C. M.; Choi, J. J.;
Lee, S.-H. Phys. Chem. Chem. Phys. 2015, 17, 31278-31286. (19)
Whitfield, P. S.; Herron, N.; Guise, W. E.; Page, K.; Cheng, Y. Q.; Milas, I.; Crawford,
M. K. Sci. Rep. 2016, 6, 35685. (20)
Chen, Y. F.; Tsai, Y. T.; Bassani, D. M.; Clerc, R.; Forgacs, D.; Bolink, H. J.; Wussler,
M.; Jaegermann, W.; Wantz, G.; Hirsch, L. J. Mater. Chem. A 2016, 4, 17529-17536. (21)
Haruyama, J.; Sodeyama, K.; Han, L.; Tateyama, Y. J. Am. Chem. Soc. 2015, 137,
10048. (22)
Xu, Q.; Eguchi, T.; Nakayama, H.; Nakamura, N.; Kishita, M. Z. Naturforsch., A: Phys.
Sci. 1991, 46, 240-246. (23)
Cremer, E.; Polanyi, M. Z. Physik. Chem. 1932, B19, 443-450.
(24)
(a) Eyring, H. Proc. Natl. Acad. Sci. U. S. A. 1933, 19, 78-81; (b) Eyring, H.; Sherman,
A. J. Chem. Phys. 1933, 1, 345-349. (25)
Westheimer, F. H. Chem. Rev. 1961, 61, 265-273.
(26)
Parkin, G. Acc. Chem. Res. 2009, 42, 315-325.
(27)
Scheiner, S. BBA-Bioenergetics 2000, 1458, 28-42.
(28)
Stojkovic, V.; Kohen, A. Isr. J. Chem. 2009, 49, 163-173.
(29)
Sharma, S. C.; Klinman, J. P. J. Am. Chem. Soc. 2008, 130, 17632-17633.
ACS Paragon Plus Environment
21
Journal of the American Chemical Society
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
(30)
Sen, A.; Kohen, A. J. Phys. Org. Chem. 2010, 23, 613-619.
(31)
(a) Ishihara, T. Perovskite Oxide for Solid Oxide Fuel Cells Springer: New York, 2009;
Page 22 of 23
(b) Iwahara, H.; Esaka, T.; Uchida, H.; Maeda, N. Solid State Ionics 1981, 3-4, 359-363. (32)
(a) Haile, S. M.; West, D. L.; Campbell, J. J. Mater. Res. 1998, 13, 1576-1595; (b)
Kreuer, K. D. Annu. Rev. Mater. Res. 2003, 33, 333-359. (33)
(a) Islam, M. S.; Davies, R. A.; Gale, J. D. Chem. Mater. 2001, 13, 2049-2055; (b)
Merinov, B.; Goddard, W. A., III J. Chem. Phys. 2009, 130, 194707; (c) Yamazaki, Y.; Blanc, F.; Okuyama, Y.; Buannic, L.; Lucio-Vega, J. C.; Grey, C. P.; Haile, S. M. Nat. Mater. 2013, 12, 647-651. (34)
(a) Matsushita, E. Solid State Ionics 2001, 145, 445-450; (b) Spahr, E. J.; Wen, L.;
Stavola, M.; Boatner, L. A.; Feldman, L. C.; Tolk, N. H.; Lupke, G. Phys. Rev. Lett. 2009, 102, 075506. (35)
(a) Stoneham, A. M. J. Nucl. Mater. 1978, 69-70, 109-116; (b) Stoneham, A. M.; Flynn,
C. P. J. Phys. F: Metal Phys. 1973, 3, 505-512. (36)
Rivnay, J.; Inal, S.; Collins, B. A.; Sessolo, M.; Stavrinidou, E.; Strakosas, X.; Tassone,
C.; Delongchamp, D. M.; Malliaras, G. G. Nat. Commun. 2016, 7, 11287. (37)
Lee, J.-K.; Chatzichristidi, M.; Zakhidov, A. A.; Hwang, H. S.; Schwartz, E. L.; Sha, J.;
Taylor, P. G.; Fong, H. H.; De Franco, J. A.; Murotani, E.; Wong, W. W. H.; Malliaras, G. G.; Ober, C. K. J. Mater. Chem. 2009, 19, 2986-2992. (38)
Fabregat-Santiago, F.; Kulbak, M.; Zohar, A.; Vallés-Pelarda, M.; Hodes, G.; Cahen, D.;
Mora-Seró, I. ACS Energy Lett. 2017, 2, 2007-2013.
ACS Paragon Plus Environment
22
Page 23 of 23
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
Journal of the American Chemical Society
TOC image for graphical abstract
ACS Paragon Plus Environment
23