Tuning optical / electrical properties of 2 D / 3 D perovskite by the inclusion of aromatic cation

The application of bulky aliphatic cations in the manufacture of moisture ́s stable materials has triggered the development and application of 2D/3D perovskites as sensitizers in stable to moisture solar cells. While it is true the moisture ́s stability increases, it is also true that the photovoltaic performance of the 2D/3D PVK material is severely limited owing to quantum and dielectric confinement effects. Accordingly, it is necessary the synthesis and deep optical characterization of materials with an adequate management of dielectric contrast between the layers. Here, we demonstrate the successful dielectric confinement tuning by the inclusion of a conjugated molecule, as bulky cation, in the fabrication of 2D/3D PVK material (C6H5NH3)2(CH3NH3)n1PbnI3n+1 where n=3 and 5. The absence of excitonic states related to n≥1 at room temperature, as well as the very low concentration of excitons after 1 ps for samples where n≥3, are a strong evidence of an excellent ability to dissociate excitons in free charge carriers. As consequence films with low n presenting higher stability than standard 3D perovskites, improve significantly their performance showing one of the highest short circuit current density (Jsc≈13.8) obtained to date for perovskite materials within the 2D limit (n<10).


Introduction
During the last years, hybrid organic-inorganic halide perovskites (PVK), ABX 3 where A=methylammonium, formamidinium; B=Pb, Sn, Ge; X=Br, I are continuously making breakthroughs in optoelectronic devices spanning photovoltaics, light emitting, lasers and photodetectors owing to their unique optical and electrical properties. (1,2) Their performance in perovskite solar cells (PSCs) has been exceptionally remarkable, solution processed devices now deliver power conversion efficiency (PCE) close to 23%.(3) However, challenges such as stability under environmental conditions need to be addressed thinking in commercial application. (4,5) The inclusion of bulky hydrophobic organic cations, in perovskite´s chemical formulation, is one of the most successful strategies followed to overcome the issue related to stability towards moisture. The generated material adopts a layered arrangement named 2D or 2D/3D PVK of general chemical formula R 2 R´n -1 Pb n X 3n+1 (R=aryl-, alkyl-cation, R´=methylammonium, formamidinium, etc; X=Cl, Br I and n=number of inorganic layers). This structural arrangement adopts the form of a natural multi quantum well structure, producing two important phenomena: 1) the excitons (or charge carriers) are confined to inorganic layers (the wells) owing to a very different band gap compared to the organic layers (the barriers); consequently, it is possible to see highly stable excitonic transition, with high binding energy, inclusive at room temperature (quantum confinement effect); and 2) a stronger difference in dielectric properties between the inorganic and organic layers produce a still more intense coulombic interaction of the exciton than systems where a less pronounced dielectric contrast is observed (dielectric effect). (6) The dielectric mismatch, and its effects in the exciton binding energy, has been described previously by Ishihara. (7,8) From that study, it is possible determine that the dielectric confinement can be modulated through the organic cation. Several works have been reported where bulky cations have been successfully applied in the synthesis of layered materials, with applications in solar cells highly moisture´s stable. (9)(10)(11) Despite that stability of the device is improved, the poor photovoltaic performance of the 2D/3D PVK material confirms the disadvantages related to quantum and dielectric confinements effects. A quick glance at the reported results, it becomes evident that the principal reason should be related to the aliphatic nature of all the cations used. The Fig. S1 shows the chemical structure of the principal bulky cations used in the fabrication of solar cells. These organic molecules are represented by butylammonium (BA) (and propyl, not showed) iodide, (11)(12)(13)(14) valeric acid derived ammonium iodide (AVA), (15) 2-iodoethylammonium iodide (EA), (16) phenylethylammonium bromide (17) or iodide (9,18) and anilinium iodide. (19) Recently, it was demonstrated that the inclusion of polarizable guests, within crystal lattice of 2D materials, leads to a significant reduction of the excitons´ confinement, owing to a very important decrease in the exciton binding energy. (20) Analyzing the polarizability as key parameter to improve the photovoltaic performance through decreasing the exciton binding energy, the aliphatic derivatives (BA or propyl, the later not showed), have just σ-electrons located between the non-polar covalent bonds (C-C), such that the polarizability (ε) is very low. Conversely, the inclusion of aromatic rings, with free and polarizable π electrons, lead to a higher ε than the observed in aliphatic chains. (7,19) The case of AVA (15) and EA(16) is very interesting. In spite of their structures are formed by a non-polar carbon backbone, with slightly polarizable electrons clouds (butyl and ethyl chains), the inclusion of the acid (COOH) and iodide functionalities confer them a polarizable fragment.
In a previous work, we demonstrated that photovoltaic devices, based on anilinium (Any) 2D and 2D/3D perovskite,(19) exhibit a higher performance than the ones prepared with BA cation, especially due to the higher photocurrent exhibited by the former. Based on that result and considering that the choice of the bulky cations has direct implications not only in the improved stability but also in the global both electronic and photovoltaic performance of devices, we focused our efforts in a complete optical and morphological characterization of a new type of material, with the principal hypothesis of reduce the stronger effects, as high exciton binding energy, through the modulation of dielectric contrast by the introduction of a bulky cation with free and polarizable π-electrons. Steady state and transient absorption measurements were used to study energy structure and dynamics of photoexcited charge carriers to verify the differences in internal charge carrier processes between our samples and traditional 2D/3D PVK. We have showed that our samples are capable of efficient dissociation of initially generated excitons that is in agreement with expectation of low exciton binding energy. Posteriorly, we focused our efforts on the fabrication of a 2nd generation of Any 2 MA 4 Pb 5 I 16 2D/3D perovskite based solar cells through hot-casting method. It was demonstrated the positive´s effect of the temperature in the materials properties with a direct correlation between temperature with the crystal sizes and PCE.

Results and discussion
As first step, we focused on the fabrication of thin films with formula Any 2 MA n-1 Pb n I 3n+1 , hereafter AnyPb n , with n=3 and 5 and their morphological characterization by X-ray diffraction XRD and scanning electron microscopy SEM. Although the n=1 film represents exactly the 2D material his low absorption across the visible range restricts his application as sensitizer in useful photovoltaic devices. XRD pattern and SEM images of thin films prepared by hot-casting method (11) are shown in the  Regard the optical properties, the absorbance spectra of 2D PVK films (n=1-5) chemically formed just by aliphatic chains, (12,14) and inclusive phenylethylammonium,(9, 18) (see Fig. S1) are characterized by an easily recognized exciton transition at the edge of the absorption band, exhibiting a red-shift progression on going from lower to higher n. ( Fig. 2c. The PL intensity quenching with temperature can be accounted for by an activation energy in the range of 23-53 and 21 meV for n=3 and n=5 samples, respectively (Fig. 2d, e). The temperature dependence of the non-radiative recombination time is also consistent with the PL intensity quenching. TR-PL experiments as function of the temperature shows an exponential dependence where the decay time can be deduced (see Fig. S3). In this way, non-radiative decay time can be extracted from measured decay times and the PL intensity at the different temperature by assuming unit quantum yield at around 110 K, and it is characterized with 8-21 and 23 meV for n=3 and 5 samples, respectively (Fig.   2d,e). These activation energies responsible of the non-radiative recombination mechanism in our samples could be ascribed to the exciton dissociation process as referred for PVK in literature. (25,26) In  to energetically lower states, (26,29) that is in agreement with the presented temperature-dependent measurements, Fig. 2d, e. The Fig. 3b shows the normalized representative fs-transients for n=3 and 5 at 510 nm, and 3D perovskite at 485 nm. These decays were well fitted using a bi-exponential function (Table S1), which has been successfully applied to analogous systems.  Table   S2, Fig.s S4 a For further clarification of the charge carrier processes in our samples, the TA measurements of 2D/3D PVK n=5 sample upon excitation at 500, 600 and 700 nm were done and compared to the one obtained using 400 nm excitation (Fig. S4c-e). The TA spectra at 2 ps pump-probe delay show comparable spectral shapes (minimal signal in visible spectra interval and strong signal in red and IR), suggesting that independently on excess energy the photoexcited charges are transferred to states at 770 nm (Fig. S2c).
These results suggest that the excited charges are transferred to common low energy states, as we observe previously. (19) The TA dynamics observed at 770 nm using all excitations are presented in Fig. S4d and S4e, respectively. As can be seen, the initial rise dynamics, that are related to charge transport from higher energy states, are similar perovskite samples (~ few tens of percentage and from tens to hundreds of ps). (26,29,30,33) This difference can be also interpreted in terms of a higher efficiency of excitons dissociation and charge transfer from initial excitonic states to the low energy ones, related to thicker perovskite layers (high n) in Any 2 MA n-1 Pb n I 3n+1 sample, in contrast to other 2D/3D perovskite films.
In the next step we focus on the characterization and fabrication of AnyPb 5 2D/3D PVK based solar cell devices through modified hot-casting method. (11,19) The choice of n=5 as sensitizer was made not only on the basis that the photovoltaic performance of solar cells based on 2D/3D perovskites increases with n(18) but also considering that the long term stability of the devices decreases with n, as a consequence of the decrease of concentration of hydrophobic cations. Considering the deficient coating by our 2D/3D PVK material previously discussed, and with the aim of improves it, we fabricated thin films through a scanning of temperatures from 110 to 190 °C. (35) The structural properties of the 2D/3D perovskite layers formed at different substrate temperatures have been analyzed by XRD. The XRD patterns are shown in Fig. S5a.
The obtained values matches perfectly with those described in past section for n=5. In  Considering that in 2D/3D perovskites the transport properties are severely reduced, in comparison with 3D counterparts, owing to the presence of bulky organic cation, we have measured local current maps via a conducting AFM under N 2 flow at an applied bias 4 V (Fig. 4b, d and f). Interestingly, we note nearly an order of magnitude increase in local current for perovskite films treated at higher temperature, see Fig. 4g. Systematically, in all the samples, we note a lower current at grain boundaries (dark regions), suggesting that grain boundaries limit the performance and increase the trap density in the films. We also note a minor variation of local current within a single leaf structure (Fig. S8), suggesting a highly polycrystalline surface, probably due to mixed phases with different n and/or crystalline orientations. This is further confirmed from the surface potential maps, recorded via Kelvin Probe Force Microscopy (KPFM) (Fig. S8), After fundamental characterization of the perovskite material, the photovoltaic performance of PSC fabricated with different temperatures of substrate pre-heating has been analyzed. (11,35) The fabrication of our 2 nd generation of devices was carried out, with n-i-p architecture was carried out i.e., FTO, compact and mesoporous layers TiO 2 , 2D/3D perovskite with n=5, spiro-OMeTAD as hole selecting contact and gold. Fig. 6 shows the current-potential (J-V) curves and the photovoltaic parameters, for champion reported values for 3D is probably due to a higher recombination rate as it has been studied by impedance spectroscopy, (37-39) see Fig. S9, however, a complete study will be reported separately. Interestingly average values obtained for devices fabricated at different temperatures do not follow the same trend than champion cells; see Table S3 and Fig. S10. Concerning the average PCE values, an improvement is evident with increase in the temperature; however, when the processing temperature increases to 170 °C and above, the dispersion of the obtained results increases causing that the average PCE gradually decreases. This variability observed in the results we suggest is directly related to the experimental conditions. Pre-heated substrates are quickly transferred from the hot plate to the spin coater, immediately the perovskite precursor solution is added and the spin coating process start. When the samples are in a low temperature range, e.g., 110 to 150 °C, the change of temperatures is less abrupt and the results can be more reproducible compared to the higher temperature range.
Even though the introduction of bulky cation with high value of ε suggest a stronger interaction with water than cations with low value of ε (and therefore lees moisture stability), we carried out a stability test under ambient conditions for the fabricated devices. Their stability was compared with a pure 3D PVK based solar cells, (Fig. S11). The devices were kept under ambient humidity over 40% for a period of 288 h. Interestingly, the devices prepared between 110 and 150 °C, exhibit higher long term stability than 3D perovskite based devices stored at similar conditions. In the other hand, the results showed that the devices fabricated with the highest temperatures (170 and 190 °C) exhibit lower long term stability than 3D perovskites. Very recently, we have shown that the presence of a reduced amount of moisture during the perovskite deposition processes is positive from the stability point of view. (42) In this way, the high temperature utilized in the high temperature regime could avoid the presence of this little humidity, preventing its beneficial role.
In conclusion, we demonstrate that the use of a conjugated anilinium cation, with a cloud of free and polarizable π-electrons, in the preparation of 2D/3D PVK led to obtain a material with optical and electrical properties improved. The determinant factor was the tuning of dielectric contrast between inorganic and organic layers, which has direct influence in the decrease of exciton binding energy. Through steady state and time-resolved absorption experiments we have proved high efficiency of initially generated exciton dissociation even for samples with low n (about 3) and thus high potential of these structures in photovoltaics. The scanning of temperature´s fabrication stated a relation between temperature and size/crystallinity of the material.
Thus it was observed a gradual increment in the size of crystallites on going from low to high fabrication´s temperatures. When the crystallinity of the material was improved through the temperature and a better coverage of the film was achieved, the PCE it got better. Importantly, the higher J sc observed, comparing to the reported values, is a strong indicative of both: 1) a lower exciton binding energy by tuning of dielectric contrast between wells and barriers and 2) an increase in the electronic transport properties of this material. Surprisingly, in spite of the intrinsic nature of the employed cation suggest an important interaction with water, the devices fabricated with 2D/3D perovskite showed a higher stability (near to 70%) after 288 h than 3D perovskite (less than 40%) under the same period and environmental conditions. Further improvement on the performance of anilinium based 2D/3D perovskites can be anticipated if higher control of the nucleation and crystal growth conditions are obtained, work that it is currently in progress.

Conflict of interest
There are no conflicts to declare.