Sublimed C60 for efficient and repeatable perovskite-based solar cells

Said, Ahmed A.; Aydin, Erkan; Ugur, Esma; Xu, Zhaojian; Deger, Caner; Vishal, Badri; Vlk, Aleš; Dally, Pia; Yildirim, Bumin K.; Azmi, Randi; Liu, Jiang; Jackson, Edward A.; Johnson, Holly M.; Gui, Manting; Richter, Henning; Pininti, Anil R.; Bristow, Helen; Babics, Maxime; Razzaq, Arsalan; Allen, Thomas G.; Ledinský, Martin; Yavuz, Ilhan; Rand, Barry P.; De Wolf, Stefaan

doi:10.1038/s41467-024-44974-0

Download PDF

Article
Open access
Published: 24 January 2024

Sublimed C₆₀ for efficient and repeatable perovskite-based solar cells

Nature Communications volume 15, Article number: 708 (2024) Cite this article

8073 Accesses
1 Citations
17 Altmetric
Metrics details

Subjects

Abstract

Thermally evaporated C₆₀ is a near-ubiquitous electron transport layer in state-of-the-art p–i–n perovskite-based solar cells. As perovskite photovoltaic technologies are moving toward industrialization, batch-to-batch reproducibility of device performances becomes crucial. Here, we show that commercial as-received (99.75% pure) C₆₀ source materials may coalesce during repeated thermal evaporation processes, jeopardizing such reproducibility. We find that the coalescence is due to oxygen present in the initial source powder and leads to the formation of deep states within the perovskite bandgap, resulting in a systematic decrease in solar cell performance. However, further purification (through sublimation) of the C₆₀ to 99.95% before evaporation is found to hinder coalescence, with the associated solar cell performances being fully reproducible after repeated processing. We verify the universality of this behavior on perovskite/silicon tandem solar cells by demonstrating their open-circuit voltages and fill factors to remain at 1950 mV and 81% respectively, over eight repeated processes using the same sublimed C₆₀ source material. Notably, one of these cells achieved a certified power conversion efficiency of 30.9%. These findings provide insights crucial for the advancement of perovskite photovoltaic technologies towards scaled production with high process yield.

Uncovering the predictive pathways of lithium and sodium interchange in layered oxides

Article 16 April 2024

High carrier mobility along the [111] orientation in Cu2O photoelectrodes

Article Open access 24 April 2024

Flexible quasi-2D perovskite solar cells with high specific power and improved stability for energy-autonomous drones

Article 17 April 2024

Introduction

Perovskite-based solar cells (PSCs) are emerging high-efficiency photovoltaic (PV) technologies on the verge of commercialization^1,2. In their single-junction (1-J) implementation, initial PSCs were fabricated in the so-called n-i-p architecture, i.e. by first depositing the electron transport layer (ETL, n-layer), followed by the perovskite absorber (i-layer), and hole transport layer (HTL, p-layer). So far, majority of the certified champion 1-J cells have been reported in this configuration^3,4. However, in recent years, p-i-n devices have drawn more attention due to their compatibility with low-temperature processing, enabling applications such as tandem solar cells^5,6,7,8, and more promising material choices for the HTL in terms of device stability and materials cost⁹. In this configuration, C₆₀-based fullerenes are usually chosen as the ETL^{8,10,11,12,13,14,15,16,17,18,19}, although several alternative materials have been explored as well^{20,21,22,23,24,25,26,27,28}.

The preference of C₆₀ as the ETL for p-i-n PSCs largely relates to their small conduction band offset and large valence band offset with respect to the perovskite, resulting in favorable electron-extraction and hole-blocking properties, respectively, essential for a good electron-selective contact. Other advantages are their high electron mobility, hydrophobic nature, and capability to passivate perovskite surface antisite defects^10,11,12,13. Their poor solubility is usually resolved by synthesizing C₆₀ to feature specific functional groups²⁹. This enables processing with solution-based methods such as spin-coating, spray, or blade coating^30,31. Thermal evaporation of C₆₀ is possible too and is arguably the industrially-preferred method for perovskite device fabrication as this technique enables good conformality and thickness control on rough and topologically complex surfaces. This is of specific importance for multijunction perovskite/silicon applications in the common p-i-n architecture, where accurate thickness and conformality control are required to minimize parasitic optical absorption caused by C₆₀ while maintaining its essential electronic role at the ETL in the device³².

In this work, we investigate the evolution of the electronic quality of C₆₀ thin films subjected to repeated evaporation processes and gauge the impact of such a procedure on device performance. Our findings reveal that, as a result of the heating and cooling cycles of the source material during multiple evaporation cycles, C₆₀ undergoes a conversion into higher molecular weight structures through the fusion of C₆₀ molecules. This transformation leads to modifications in the electronic properties of the fullerene, detrimentally affecting device performance. However, we show that further purification of as-received C₆₀ (through a sublimation process, before the evaporation) can help to avoid these issues, and device performance remains unaffected even after repeated deposition cycles. The insights obtained from this research are crucial for the advancement towards practical, scaled applications of perovskite-based p-i-n structured single-junction and tandem PV technologies that employ evaporated C₆₀ as the ETL.

Results

Device performance with repeated processes of as-received C₆₀

We first assessed the performance of commercial as-received C₆₀ source material (henceforth referred to as “as-received”) with repeated deposition cycles directly on 1-J PSCs. For this, we used a device stack in the ITO/NiO_x/MeO-2PACz/perovskite/C₆₀/BCP/Ag configuration, as sketched in Fig. 1a, where ITO is indium tin oxide, NiO_x is nickel oxide, MeO-2PACz is (2-(3,6-dimethoxy-9H-carbazol-9-yl)ethyl)phosphonic acid and BCP is bathocuproine. One thermal cycle involves evacuating the evaporation chamber (to a base pressure of 10^-7Torr), followed by thermal evaporation of C₆₀ layers (crucible-temperature, T_C, and time-temperature profiles are given in Supplementary Fig. 1a), and finished by bringing the half-finished device (NiO_x/MeO-2PACz/perovskite/C₆₀) to the load lock pressure (around 10^-3Torr). We observed that the maximum T_C during the deposition needed to sustain a constant deposition rate increased with every new cycle. We plotted the rise of the set temperature to achieve the 0.1 Å/s (and then 0.25 Å/s) deposition rate in Supplementary Fig. 1a, which is the standard baseline C₆₀ evaporation process in our lab. After the 8^th cycle, we also found that the color of the powder in the crucible changed from black to brown as shown in Supplementary Fig. 1b. From the current density-voltage (J–V) analysis of the devices, we found an average 5 mV drop in V_oc for the corresponding devices after each thermal cycle, as well as a gradual decrease in FF. The V_oc of the device in the 1^st thermal cycle was 1142 mV (average is 1130 mV), which dropped to 1118 mV (average is 1109 mV) after the 8^th thermal cycle, as shown in Fig. 1b and c. We confirmed this trend of declining V_oc and FF with increasing C₆₀ deposition cycles for other types of PSCs, for instance, those based on one-step solution processed Cs_0.05FA_0.8MA_0.15Pb(I_0.745Br_0.255)₃ (1.68 eV) and Cs_0.03(FA_0.90MA_0.10)_0.97PbI₃ (1.55 eV) perovskite absorbers (Supplementary Fig. 2a and 2b, respectively)^33,34. Then, we repeated the same experiment with 1.68 eV bandgap hybrid processed perovskite as commonly used in perovskite/silicon tandem solar cells (see Methods for the details), and found the same trend for V_oc and FF performance drops as shown in Supplementary Fig. 2c.

**Fig. 1: Perovskite solar cells fabricated with commercial as-received C_60.**

We also verified this behavior for larger device areas (0.1 cm²–1 cm²) on perovskite/silicon tandem solar cells with the structure given in Fig. 1d. Again, we observed V_oc and FF losses after repeated processing, as depicted in the statistical analysis of the device parameters in Supplementary Fig. 3. The characteristic J-V curves of the tandem cells and their external quantum efficiency (EQE) spectra are shown in Fig. 1e and Supplementary Fig. 4, respectively. The tandems show a similar performance drop to the 1-J perovskite devices after repeated thermal cycles of the C₆₀ ETL. The quasi-Fermi level splitting (QFLS) from the half-finished device stacks (NiO_x/MeO-2PACz/perovskite/C₆₀) after the 1^st, 6^th, 7^th, and 8^th deposition cycles confirmed the voltage losses on devices stacks, as we again observed ~5 meV loss after each cycle (Fig. 2a).

**Fig. 2: Optoelectronic properties of commercial as-received C₆₀ films.**

The origin of the observed voltage losses after thermal cycling is likely due to a change in the structural and electronic properties of the C₆₀ layer itself and the interface it forms with the perovskite. To gain further insight, we first characterized quartz/perovskite/C₆₀ stacks via photothermal deflection spectroscopy (PDS) to probe whether any deep states were formed or induced in the perovskite band-gap upon contact with the fullerene. We found that the perovskite band edge is almost identical for fresh and thermally cycled samples, indicating that the C₆₀ films do not alter the bulk properties of the perovskite, as expected. The absorption level within the bandgap (from 1 to 1.4 eV) as shown in Fig. 2b is also very similar. We then analyzed the phase changes between the pump beam intensity and the detected signal in PDS measurements, which is a measure of the time shift between the light excitation and the measured PDS signal. As depicted in Fig. 2c, we set the time shift to 0° at the high absorption region (>2.2 eV), where all the incoming light is absorbed by both samples near the perovskite surface. At the band edge area, the phase of fresh C₆₀ samples decreases, which suggests that the light absorption depth increases and therefore excited phonons need more time to travel to the sample surface. This behavior is standard for most PDS measurements. However, in the case of the thermally cycled sample, the phase decrease is minimal and is followed by a strong increase in the phase shift. This means that the thermal signal generation is concentrated at the very top of the sample, close to the surface. This may be caused either by direct absorption of the light at the surface states or by recombination of electron-hole pairs generated within the perovskite film at the surface states. From here, we can infer that fullerene contacts deposited with thermal cycles lead to a higher concentration of electronically active defect states at the perovskite/fullerene interface.

Next, we measured the electron mobility of the C₆₀ layers deposited during both the first and last thermal cycles via top-gate bottom-contact field effect transistors (FETs). The architecture of the fabricated FET is shown in Supplementary Fig. 5. The electron mobilities (μ_e) of the 1st deposition (0.083 cm²/V s) are higher than the 8th deposition (0.034 cm²/V s), on average, as shown in Fig. 2d. This implies that the structure of the C₆₀ layers might have changed during repeated thermal cycles. The work function (WF) and the highest occupied molecular orbital (HOMO) of the distinct films are also other parameters to be considered, which we investigated via ultraviolet photoelectron spectroscopy (UPS) measurements. We obtained a WF value of ~4.4 eV for both the fresh and the thermally cycled C₆₀ films. The valence band maximum, determined with a Gaussian fitting method, also remains unchanged (6.4 eV), as shown in (Supplementary Fig. 6a) and Fig. 2e, respectively. However, we observed a change in the density of states (DOS) on the thermally cycled samples as shown in Fig. 2e. From these characterizations, we conclude that the properties of the C₆₀ powder, and therefore the deposited thin films, have changed with thermal cycling. This is mainly manifested through additional recombination states, in line with the mobility results of the FETs.

Subsequently, we collected the brownish powder from the crucible after the 8^th cycle for further analysis. We first transferred those powders onto quartz glass substrates, sandwiched them with another glass sheet, and verified the PL emission via a hyperspectral imaging system. We observed that the thermally cycled powders showed an emission peak at 835 nm in addition to the main peak typical for pure C₆₀ located at 742 nm, as shown in Supplementary Fig. 6b^35,36. The increased oxygen content for the thermally cycled samples, obtained from X-ray photoelectron spectroscopy (XPS) analysis, reveals the possibility of a chemical reaction occurring during the repeated evaporation processes (Supplementary Fig. 6c, d).

Further analysis indicates that the thermally cycled C₆₀ powder has a wider full width at half maximum (FWHM) of its main peak in the X-ray diffraction (XRD) spectrum. Moreover, a new feature at 11.4 degrees is present, as shown in Supplementary Fig. 7a, b, which is related to the formation of fullerene dimers and different fullerene derivatives at a certain fraction in the powder³⁷. We also utilized the matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectroscopy technique to investigate the change in molecular weights of the powders. Both fresh and thermally cycled C₆₀ powders showed a strong peak at 720 mass/charge number (m/z) which is assigned to the presence of a high fraction of C₆₀ in the sample bed. In addition, a weak peak for C₆₀ with C¹³ at 721 m/z is observed in Supplementary Fig. 8. Interestingly, the thermally cycled C₆₀ powder showed additional peaks at 1444, 1465, 1489, 1513, and 1537 m/z. These peaks show the presence of additional fullerene derivatives such as C₁₂₀, C₁₂₂, C_124, C_126, and C₁₂₈, respectively with different numbers of the C¹³ isotope^38,39, (Supplementary Fig. 8). The peak at 1444 m/z might be ascribed to fullerene with C₁₁₉O. After conducting MALDI-TOF analysis on the thin films, it became evident that films produced from the 1st deposition cycle of the C₆₀ powder lacked C₁₂₀ fragments. In contrast, the MALDI-TOF analysis of the 8^th deposition cycle thin film showed a peak at 1441 m/z which is ascribed to C₁₂₀. We attribute this disparity to the alteration of evaporation temperatures. Furthermore, 1322, 1345, 1369, 1392, and 1417 m/z are found to be present in both C₆₀ thin films with different peak intensities and are assigned to C₁₁₀, C₁₁₂, C₁₁₄, C_116, and C₁₁₈, respectively. Importantly, the peaks of the thermally cycled thin film are stronger than that of fresh thin film, which confirms that the concentration of the higher molecular weight fullerenes in the thermally cycled thin film is higher than that of fresh thin film as shown in Fig. 2f. These findings explain the lower electron mobility of the thermally cycled C₆₀ thin film compared to that of the fresh thin film. The surface morphology of fresh and thermally cycled C₆₀ films was investigated by using atomic force microscopy (AFM). There is no significant change in the morphology of fresh and thermally cycled C₆₀ films as shown in Supplementary Fig. 9a, b, respectively.

We performed further analysis of the thermally cycled powders with high-resolution transmission electron microscopy (HRTEM) as shown in Fig. 3a, Supplementary Fig. 10a and 10b at a low acceleration voltage (80 kV) as the C₆₀ molecules are electron-beam sensitive under conventional TEM conditions. HRTEM and Fast Fourier transform (FFT) spots matching with C₆₀ d-spacing ([111], [222], [602], [522], [511], [624], [800]) confirm that the C₆₀ is molecularly highly crystalline with a long-range spatial ordering (Fig. 3a and Supplementary Fig. 10c). From the HRTEM images given in Fig. 3b, we identified C₆₀, C_120, and C₆₀-dimers via the average diagonal diameter of the resolved molecules with the approach of Goel et al.⁴⁰.

**Fig. 3: Structural and optoelectronic analysis of as-received C₆₀ powders and their films.**

Following this methodology, molecularly resolved HRTEM of C₆₀ and C₁₂₀ appeared with an average diameter of ~0.686 and 1.014 nm, respectively, with an accuracy of ± 0.038 nm. Also, a possible C₆₀-dimer with an intermolecular distance of 0.477 nm was observed as shown in Fig. 3c. The formation of high molecular weight fullerene derivatives in the presence of molecular oxygen has been explained by molecular dynamics simulations previously⁴¹. It is argued that molecular oxygen can react with the carbon-carbon bond in the pentagonal ring of low molecular weight fullerene, resulting in the formation of a carbonyl group bonded to fullerene. The carbonyl group then interconverts to an epoxide radical which attacks other fullerenes, forming higher molecular weight fullerene derivatives.

Next, we studied the correlation between the electronic properties and these structural changes via density functional theory (DFT) calculations. Here, for each case, we considered the thermodynamically most favorable high-symmetry fullerene configurations (Supplementary Fig. 11), which are adopted from the structures we observed in the MALDI-TOF analysis. Stable structures of different fullerene derivatives are shown in Supplementary Fig. 12a, and among all these cases, C₁₂₄ and C₁₂₆ are found to be the most stable, based on DFT-calculated formation energy differences (Supplementary Fig. 12b). Our MALDI-TOF results reveal that C₁₂₄ and C₁₂₆ have the highest mass fraction in the thermally cycled powder after C₆₀, in line with the DFT results. The fraction of C₁₂₄ and C₁₂₆ in the thermally cycled powder, calculated from Supplementary Fig. 8, is estimated to be 5.6% and 2.6%, respectively, while 90% is still C₆₀. We note that the same configuration of fullerene at a different m/z value is related to the number of isotopes of C¹³ found in the structure. Also, DFT calculations suggest that increasing the number of C¹³ isotopes leads to an increased formation energy, and therefore to more stable structures (Supplementary Fig. 12c).

We further elucidated the origin of the DOS changes detected via UPS measurements by modeling the perovskite/C₆₀ (representative of fresh powder) and perovskite/C₁₂₄ (representative of thermally cycled powder) interfaces. For the perovskite/C₆₀ interface, we observed the formation of charge traps near the perovskite band–edge which is attributed to the overlapping of quantum-mechanical wave functions of the perovskite and C₆₀, as reported in our earlier work³⁴. At that time, we overcame this issue by displacing the fullerene contact with ultrathin (<2 nm) MgF₂ interlayers. However, this strategy did not help to reduce the performance losses for fullerene layers deposited via our repeated deposition cycles (Supplementary Fig. 13). This is because interfacing perovskite with C₁₂₄ results in deep trap levels within the band gap as shown in Fig. 3d, which is different than C₆₀. By performing the same simulation for different fullerenes (C₁₁₈, C₁₂₀, C₁₂₂, C₁₂₆, and C₁₂₈), we found a similar behavior as C₁₂₄ (Supplementary Fig. 14a). The observed charge traps might be related to formation of multiple bonds between C-I and C-Pb (Supplementary Fig. 14b), in the absence of continuous contact displacer layers (e.g., CaF₂, MgF₂). Also, in the case of the perovskite/C₆₀ model, the delocalization of electrons over the entire C₆₀ molecule implies effective vertical electron transport from the perovskite layer to the electrode via C₆₀. On the other hand, the charges are localized near the perovskite/C₁₂₄ interface, which can hinder the charge transfer across the perovskite/C₁₂₄ interface as shown in Fig. 3e. Therefore, in the case of thermally cycled C₆₀, contact displacement does not mitigate performance losses since the C₆₀ itself exhibits a defective character.

To understand the possibility of sample storage before the cyclic processing causing performance losses, we conducted an experiment where we compared the performance of half-devices stored in an N₂ glove box and vacuum environment (for a duration equivalent to 8 deposition cycles) with that of a fresh half-device that was not stored in the glove box. The completed devices maintained their V_oc and FF values without any significant change before and after storage, as illustrated in Supplementary Fig. 15a and b. Consequently, we can conclude that the observed variations are attributed to changes in the fullerene itself.

We further investigated the influence of the C₆₀ evaporation heating and cooling cycles on the device performances by performing a single, long deposition cycle, to mimic possible industrial processing. During this process, the substrate shutter was kept closed for a time equivalent to 8 depositions. Note that the number of deposition cycles might be even higher in actual industry processes as additional concerns there will be effective material utilization and throughput. After this period, the shutter was opened for a given time to achieve an identical thickness as in the previous cell fabrication processes. In our study, we observed that the PV parameters of both the long and single deposition batches closely resembled those of the initial short process cycle (Supplementary Fig. 16a). These findings were verified using two separate batches, which yielded consistent results for both the fresh deposition and the long thermal cycle of C₆₀. Interestingly, the rising crucible temperature during the extended cycle (Supplementary Fig. 16b) suggests that temperature increase is not the cause of the performance losses. Instead, we hypothesize that coalescence occurring in or on the fullerene powders during source cooling in half-device exchanges might be responsible for these outcomes.

Repeated deposition cycles with sublimed C₆₀

In search of a solution to overcome C₆₀ coalescence, we further hypothesized that impurities in the as-received powders might be responsible for the dimer formation³⁸. To verify this, we performed high-performance liquid chromatography (HPLC) analysis and we found that the as-received C₆₀ has a purity of ~99.75-99.8% which can contain impurities or by-products of C₆₀ oxide and C₇₀, in addition to the possibility of the presence of C₆₀-dimer as shown in Supplementary Fig. 17. We further purified the as-received C₆₀ powders via vacuum thermal gradient sublimation (see Methods), which resulted in enhancing the purity of C₆₀ to 99.95% as verified by HPLC analysis in Supplementary Fig. 18 (henceforth referred to as “sublimed”). After purification, we performed repeated deposition cycles similar to the as-received series of experiments. Interestingly, the color of thermally cycled sublimed C₆₀ remained visually black even after 8 deposition cycles as shown in Supplementary Fig. 19a. Furthermore, in contrast to the as-received material, the T_c profile remained unchanged when the number of deposition cycles increased. Also, the maximum T_c remained lower than that of the as-received C₆₀ (see temperature profile in Supplementary Fig. 19b), which was confirmed by thermogravimetric analysis (TGA) under nitrogen (Supplementary Fig. 20).

We studied the sublimed C₆₀ layers with the same set of characterization techniques to observe the influence on the film quality. QFLS analysis of half-device stacks (NiO_x/MeO-2PACz/perovskite/sublimed C₆₀) still exhibited losses after 8 deposition cycles. However, compared with that of as-received C₆₀ as shown in Supplementary Fig. 21a, the loss after each cycle is only ~2 meV. To understand the possible occurrence of defect states in the perovskite band gap after repeated deposition cycles of sublimed C₆₀, half stacks of quartz/perovskite/sublimed C₆₀ were characterized by PDS. Here, the measured PDS results were almost identical for fresh and thermally cycled sublimed C₆₀, as shown in Supplementary Fig. 21b, c. Also, WFs and HOMOs of thermally cycled sublimed C₆₀ remained almost unchanged (Supplementary Fig. 21d, e). Importantly, we detected from XPS analysis the same carbon content but an absence of oxygen in the fresh and thermally cycled sublimed C₆₀ samples (Supplementary Fig. 22). MALDI-TOF analysis of fresh and thermally cycled sublimed C₆₀ powders exhibited a very low oxygen percentage, which is around 0.6% as shown in Supplementary Fig. 23a, by showing peaks at 1416, 1417, and 1418 m/z of C₁₁₈. The PL analysis of powders also shows formation of a shoulder at 835 nm, which is less pronounced with sublimed C₆₀ (Supplementary Fig. 23b). This suggests the presence of fewer impurities. This peak is associated with phonon replicas in literature⁴².

Our characterization set revealed that further purification of as-received C₆₀ through a sublimation process is a promising solution to retain the optoelectronic properties of C₆₀ after repeated deposition cycles. So, we studied sublimed C₆₀ on solar cells with the identical stacks given in Fig. 1a. By analyzing the J-V curves of the fabricated devices, we found the average of V_oc and FF of fabricated devices leveled off as shown in Supplementary Fig. 24. Even fully cooling down this powder to room temperature and using it again did not affect the average device performance (Fig. 4a, b). The corresponding J-V curves are shown in Fig. 4c. Also, the performances of the perovskite/silicon tandem solar cells were unaffected by multiple thermal cycles if sublimed C₆₀ was used (Fig. 4d, e). Corresponding J-V curves and EQE graphs of tandem devices are shown in Fig. 4f and Supplementary Fig. 25, respectively. One of our unencapsulated perovskite/silicon tandem devices using sublimed C₆₀ showed a certified PCE of 30.90% by Fraunhofer ISE CalLab as shown in Supplementary Fig. 26. We note that using sublimed C₆₀ provides a quite marginal V_oc enhancement (<10 mV), however, it gives reproducible results in repeated processes, which will be critical to enable the industrial adoption of this technology.

**Fig. 4: Photovoltaic performances with contacts using sublimed C_60.**

Discussion

The objective of this work was to investigate the feasibility of industrial processing of C₆₀ contact layers in perovskite-based solar cells through repeated thermal evaporation depositions. To accomplish this, we employed an evaporation system with a rotating substrate and automated load lock for half-device exchange, without changing the source powders until the end of the batch. After multiple repeated processes, we assessed film quality and solar cell performances and found that if 99.75% pure as-received C₆₀ powders (as-received) are used, the powders should be renewed for each deposition cycle (in case heating and cooling are involved) to avoid a decrease in device performance.

However, in high throughput lines, this demands frequent source replenishment to maintain device performance and may lead to suboptimal materials utilization and added costs. We identified this as a potential problem for industrial processing of perovskite-based solar cells, whether in their single-junction or multijunction (e.g., perovskite/silicon, perovskite/perovskite, or three-junction) implementations. The decrease in device performance is thought to be caused by the coalescence of C₆₀ to larger fullerene derivatives, initiated by the presence of oxygen. Notably, no decrease in device performance was observed when performing a single, long deposition cycle equivalent to the duration of 8 deposition cycles involving heating and cooling.

Our work demonstrates that using 99.95% pure C₆₀ powders, purified via vacuum thermal gradient sublimation, can address this issue and enables continuous processing of C₆₀ contact layers without systematic performance losses. We hypothesized that some oxygen was already enclosed in the crystal structure of 99.75% pure as-received C₆₀. Formation of nearly perfect face-centered-cubic (fcc) crystals in the sublimation process not only eliminates existing oxygen content from voids in the regular crystal lattice but also eliminates defects to prevent subsequent oxygen uptake except for surface adsorption which is easily reversible during degassing prior to the start of the vapor deposition process. Another possibility, beyond sublimation, is the development of processes to produce oxygen-free, and/or oxygen uptake-resistant C₆₀ without the need for sublimation.

Towards the industrialization of C₆₀ contacts, weak mechanical adhesion⁴³, instability under field operation⁴⁴, and non-ideal contact behavior have been reported previously³⁴. Solving these pressing issues is urgent, as commercial PV modules with C₆₀ contacts are already in sight. In addition to the potential of using sublimed C₆₀, this work provides valuable guidance for the preparation of fullerenes that are more suitable, particularly oxygen-free, for commercial perovskite-based solar cell processing with high yield.

Methods

Vacuum thermal gradient sublimation

Commercially purchased C₆₀ (Nano-C) was sublimed via a three-zone tube furnace (Lindberg Blue M, Model number STF55346COMC-1), with quartz collection tubes in it. Prior to use, all quartz pieces were sequentially cleaned by acetone, IPA, and piranha solution (3:1 sulfuric acid and hydrogen peroxide), followed by a quartz bake-out process at 500 °C in the furnace. According to the materials’ vapor pressure, the temperatures of the three zones were set to 570 °C, 395 °C, and 340 °C, respectively, to sublime the C₆₀ powders and create the proper temperature gradient for pure C₆₀ crystals and impurities to be separated. During the purification, the tube environment was kept at 10 Torr by a vacuum pump (Oerlikon Leybold vacuum 501591A1000 Turbolab 80). A constant nitrogen gas flow (20 SCCM flow rate, controlled by Mass Flow Controller SFC5400 LA6UAV) was applied to assist material transport along the tube. The final yield was in the range of 70–80%.

Single-junction solar cell fabrication

17 nm of NiO_x was sputtered on patterned ITO (Xin Yan Technologies, 15 ohm/sq) at room temperature using a NiO target (Plasmaterials, 99.9%) in argon atmosphere. To passivate defects on the surface of NiO_x, potassium chloride, KCl (3.5 mg/mL in water) was spin coated on the NiO_x at 4000 rpm and then annealed at 120 °C for 10 min. Then 1 mg/ml of MeO-2PACz (99 %, TCI) was spin-coated on NiO_x at 5000 rpm followed by annealing at 100 °C for 10 min. MeO-2PACz was dynamic washed by DMF to obtain a monolayer and annealed at 100 °C. NiO_x/MeO-2PACz is considered as an optimal HTL for hybrid-processed perovskite layer. The inorganic scaffold of 240 nm PbI₂-CsBr was evaporated on MeO-2PACz/ NiO_x with rate of 1 and 0.1 A/s for PbI₂ and CsBr, respectively. The final perovskite layer was obtained after conversion of the inorganic scaffold by dynamic dripping 0.65 M ethanol solution of FAI/FABr/MACl with a molar ratio of 6.2/2.07/1.0, respectively at 4000 rpm for 40 s. The crystallization of the perovskite layer was obtained after annealing at 150 °C for 30 min in ambient air. The half devices with the stack of ITO/NiO_x/MeO-2PACz/perovskite were split into 8 batches. 25 nm of C₆₀ was evaporated on each batch from the same C₆₀ without changing or adding additional material. Before the deposition, 310 mg of C₆₀ powder was contained in the crucible. Between the processes, the material in the crucible did not cool down below 240 °C. The devices are finalized with 6 nm BCP and 120 nm Ag contact again with thermal evaporation techniques.

Perovskite/silicon tandem solar cells fabrication

c-Si bottom cells were prepared as in our previous work³⁴. The bottom cells were subjected to UV-Ozone treatment for 10 min. After that, 2PACz was applied via 1 mg/mL solution in ethanol by spin-coating on ITO-coated substrates at 5000 rpm for 30 s, followed by drying at 100 °C for 10 min. We utilized 2PACz as a hole extraction layer in solution-processed perovskite layer following our tandem solar cell baseline procedure⁵. 1.7 M Cs_0.05FA_0.8MA_0.15Pb(I_0.755Br_0.255)₃ perovskite precursor solution was prepared by dissolving a mixture of FAI, MABr, CsI, PbI₂, and PbBr₂ in a solvent mixture of DMF and DMSO with a volume ratio of 4:1, and spin-coated at two steps (2000 rpm for 45 s and 7000 rpm for 10 s) chlorobenzene of 200 μL was dropped in the center of the substrates 12 s before the end of the spin-coating process. The substrates were annealed on a hotplate at 100 °C for 15 min. After perovskite deposition, ~1 nm metal fluorides (MgF₂), and 15 nm C₆₀ were subsequently deposited by thermal evaporation. 10 nm SnO₂ was then deposited by atomic layer deposition (ALD) using a Picosun system at 100 °C. Tetrakis(dimethylamino)tin(IV), TDMASn precursor source was at 80 °C and H₂O source was at 25 °C. The pulse and purge time for TDMASn was 1.6 and 5.0 s with a 90 sccm carrier gas of nitrogen, for H₂O is 1.0 and 5.0 s with 90 sccm N₂, 85 cycles were used. 40 nm IZO was sputtered from a 3-inch IZO ceramic target on top of the SnO₂ through a shadow mask. Ag finger with a thickness of 500 nm was thermally evaporated using a shadow mask. Finally, 100 nm MgF₂ was thermally evaporated as an anti-reflection layer. The evaporation rate and thickness of each experiment were monitored by quartz crystal microbalance sensors.

Solar cell characterizations

Current density-voltage (J–V) curves were measured by Keithley 2400 source meter under calibrated AM1.5 illumination using a solar simulator (Abet Technologies Sun 3000 Solar Simulator) inside a nitrogen-purged glovebox. The range of the forward scan rate and reverse scan rate is −0.1 V to 1.2 V and 1.2 V to −0.1 V, respectively for single junction cell with active area 0.1 cm². For tandem cell the range of the forward scan rate and reverse scan rate was −0.1 V to 2.0 V and 2.0 V to −0.1 V, respectively (area of in house measurement is 1.04 cm²). Both scans were carried out with scan step 50 mV/s. For the tandem solar cells, Wavelabs Sinus 220 LED-based solar simulator was utilized.

Mobility measurements via field effect transistor

Glass substrate with pre-patterned source and drain Au electrodes with a channel length of 40 µm and channel width of 1000 µm were used. C₆₀ layers were thermally evaporated with a thermal evaporator through a shadow mask. Cytop dielectric was spin-coated at 2000 rpm for 60 s on top and annealed at 60 °C for 90 min. The aluminum top metal gate was evaporated using a shadow mask to complete the device architecture. The measurements were performed via Agilent B2912A source/measure unit (SMU) in a glove box.

Transmission electron microscopy

TEM sample was prepared by dispersing thermally cycled C₆₀ in chlorobenzene and drop casting on a holey carbon support film TEM grid. TEM was performed with Cs spherical aberration image corrected Thermofisher Titan 60-300 Cubed TEM, TEM microscope operated at low acceleration voltage (80 kV). Gatan Digital Micrograph was used to process the data.

Hyperspectral PL Imaging and QFLS

To mimic p-i-n device structure, perovskite films were deposited on HTL/ITO films using the same method as for device fabrication. Then, fresh and thermally cycled C₆₀ films were evaporated on top of the perovskite films, respectively. Absolute PL of encapsulated half-finished devices were collected using a hyperspectral imaging system coupled to a microscope with 2 nm spectral resolution (Photon etc. IMA) and stacks were excited with a 405 nm laser at around 1 sun illumination⁴. The collected data were analyzed by home-built MatLab code using modified Würfel’s generalized Plank law to acquire QFLS, ∆μ^45,46.

XPS and UPS

All C₆₀ films were deposited on ITO substrates. The electron spectroscopy measurements of ultraviolet photoelectron spectroscopy (UPS) and X-ray photoelectron spectroscopy (XPS) were carried out with a UHV ScientaOmicron system at 5E-10 mbar. UPS measurements were performed on the samples using a vacuum UV source (focus) operating at 200 mA and 2E-2 mbar He pressure to increase the fraction of the He(II) line (40.8 eV). The kinetic energy range scanned was increased to 30–50 eV and the spectra were adjusted accordingly. The photoelectrons were collected at an angle of 80° between the sample and analyzer, with a normal electron takeoff angle. The constant analyzer pass energy (CAE) was 5 eV for the valence band region and for the secondary electron cutoff SECO. Spectra were adjusted by a scaling factor to provide a suitable comparison. XPS was carried out in the same spectrometer, equipped with a monochromatic Al Ka X-ray Omicron XM1000 X-ray source (hv = 1486.6 eV) operating at a power of 390 W. The survey and high-resolution spectra were collected at a CAE of 50 and 15 eV, respectively. The spectra were analyzed with Casa XPS software. The individual peak envelopes were fitted by a Gaussian–Lorentzian (GL30) using a Tougaard-based background function.

DFT

DFT is used to carry out quantum mechanical calculations in the gas phase, as implemented in Gaussian 09⁴⁷. wB97-XD/6-31 G(d) functional and basis set is used to optimize all gas phase geometries. Possible high symmetry fullerene configurations within n = 118–128 were initially generated using the CaGe code⁴⁸ and each structure if DFT optimized. Final electronic energies are used to collect the most stable geometry for each fullerene type. The formation energy of each fullerene (relative to that of ${C}_{60}$) is then predicted from

$$\Delta E={E}_{n}/n-{E}_{{C}_{60}}/60$$

(1)

where ${E}_{{C}_{60}}$ and ${E}_{n}$ is the gas phase energy of ${C}_{60}$ and ${C}_{n}$. For the bulk phases, we performed first-principles calculations based on DFT by a plane-wave basis set and the projected augmented wave method, as implemented in the VASP package^49,50. Perdew−Burke−Ernzerhof (PBE) functional with a generalized-gradient approximation is employed for the exchange-correlation functional in all geometry optimization and self-consistent field calculations. Fresh⁵¹ A 4 × 4 × 1 gamma-centered k-mesh and a plane wave basis with a 400 eV cutoff energy is used for the geometry optimizations. We first allowed relaxing the atomic positions and cell volumes using a conjugate gradient algorithm, until all residual forces are <0.02 eV/Å. Continuation single-point energy calculations on the optimized geometries were performed to create the charge density difference graphs. We insert a vacuum slab with thickness of 10 − 15 Å between the periodic slab-molecule structure along z direction.

PDS

The photothermal deflection spectroscopy (PDS) measurements were performed using a home-built PDS setup. Samples were immersed in a chemically inert liquid perfluorohexane C₆F₁₄ (FC72). The light from the 150 W Xe lamp was sent through a monochromator equipped with gratings blazed at 300, 750, and 1250 nm, and modulated by a mechanical chopper operating at a frequency of 5 Hz. A probe beam passing parallel to the sample surface is deviated by a change of the refraction index of the FC72 and detected by a four-quadrant detector.

XRD

XRD data was collected by using Bruker D2 Phaser diffractometer (Cu-K_α1 radiation, λ = 1.5406 Å) at room temperature from 5° to 50° (2θ) with a scan speed of 3°/min.

HPLC

High-pressure liquid chromatography (HPLC) was carried out with an Agilent 1100 instrument equipped with a variable wavelength detector (VWD). An analytical Cosmosil Buckprep column (Nacalai, 250 mm × 4.6 mm I.D.) was used with 1 mL/min of toluene for elution and detection at 330 nm. C₆₀ samples were dissolved in toluene at a concentration of 0.25 mg/mL and 20 µm are injected into the instrument.

TGA

Thermogravimetric analysis (TGA) was done in a nitrogen atmosphere using a TA Discovery TGA 5500 instrument after careful purging. The heat rate was 10 K/min from room temperature to 1000 °C.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

All generated data in this study is available in Supplementary Information.

References

Aydin, E. et al. Pathways toward commercial perovskite/silicon tandem photovoltaics. Science 383, eadh3849 (2024).
Article CAS PubMed Google Scholar
Zhu, P. et al. Towards The Commercialization of Perovskite Solar Modules. Adv. Mater. 2307357 https://doi.org/10.1002/adma.202307357 (2024).
Min, H. et al. Perovskite solar cells with atomically coherent interlayers on SnO₂ electrodes. Nature 598, 444–450 (2021).
Article ADS CAS PubMed Google Scholar
Aydin, E. et al. Ligand-bridged charge extraction and enhanced quantum efficiency enable efficient n–i–p perovskite/silicon tandem solar cells. Energy Environ. Sci. 14, 4377–4390 (2021).
Article CAS Google Scholar
Aydin, E. et al. Enhanced optoelectronic coupling for perovskite-silicon tandem solar cells. Nature 623, 732–738 (2023).
Article ADS CAS PubMed Google Scholar
De Wolf, S. & Aydin, E. Tandems have the power. Science 381, 30–31 (2023).
Article ADS PubMed Google Scholar
Chin, X. Y. et al. Interface passivation for 31.25%-efficient perovskite/silicon tandem solar cells. Science 381, 59–63 (2023).
Article ADS CAS PubMed Google Scholar
Mariotti, S. et al. Interface engineering for high-performance, triple-halide perovskite–silicon tandem solar cells. Science 381, 63–69 (2023).
Article ADS CAS PubMed Google Scholar
Yao, Y. et al. Organic hole-transport layers for efficient, stable, and scalable inverted perovskite solar cells. Adv. Mater. 34, 2203794 (2022).
Article CAS Google Scholar
Fang, Y., Bi, C., Wang, D. & Huang, J. The functions of fullerenes in hybrid perovskite solar cells. ACS Energy Lett. 2, 782–794 (2017).
Article CAS Google Scholar
Jeng, J. Y. et al. CH₃NH₃PbI₃ perovskite/fullerene planar-heterojunction hybrid solar cells. Adv. Mater. 25, 3727–3732 (2013).
Article CAS PubMed Google Scholar
Wang, Y.-C. et al. Efficient and hysteresis-free perovskite solar cells based on a solution processable polar fullerene electron transport layer. Adv. Energy Mater. https://doi.org/10.1002/aenm.201701144 (2017).
Castro, E., Murillo, J., Fernandez-Delgado, O. & Echegoyen, L. Progress in fullerene-based hybrid perovskite solar cells. J. Mater. Chem. C. 6, 2635–2651 (2018).
Article CAS Google Scholar
Liu, D. et al. Impact of ultrathin C60 on perovskite photovoltaic devices. ACS Nano 12, 876–883 (2018).
Article CAS PubMed Google Scholar
Isikgor, F. H. et al. Concurrent cationic and anionic perovskite defect passivation enables 27.4% perovskite/silicon tandems with suppression of halide segregation. Joule 5, 1566–1586 (2021).
Article CAS Google Scholar
Zhumagali, S. et al. Linked Nickel Oxide/Perovskite interface passivation for high-performance textured monolithic tandem solar cells. Adv. Energy Mater. 11, 2101662 (2021).
Article CAS Google Scholar
Xu, Z. et al. Reverse-bias resilience of monolithic perovskite/silicon tandem solar cells. Joule 7, 1992–2002 (2023).
Article Google Scholar
Toniolo, F. et al. Efficient and reliable encapsulation for perovskite/silicon tandem solar modules. Nanoscale 15, 16984–16991 (2023).
Article CAS PubMed Google Scholar
Tockhorn, P. et al. Nano-optical designs for high-efficiency monolithic perovskite–silicon tandem solar cells. Nat. Nanotechnol. 17, 1214–1221 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Said, A. A. et al. NDI-based small molecules as electron transporting layers in solution-processed planar perovskite solar cells. J. Solid State Chem. 270, 51–57 (2019).
Article ADS CAS Google Scholar
Said, A. A. et al. Efficient inverted perovskite solar cells by employing N-Type (D-A(1) -D-A(2)) polymers as electron transporting layer. Small 15, e1803339 (2019).
Article PubMed Google Scholar
Said, A. A., Xie, J. & Zhang, Q. Recent progress in organic electron transport materials in inverted perovskite solar cells. Small 15, e1900854 (2019).
Article PubMed Google Scholar
Chen, W. et al. Sulfur position in Pyrene-based PTTIs plays a key role to determine the performance of perovskite solar cells when PTTIs were employed as electron transport layers. ACS Appl. Energy Mater. 2, 5716–5723 (2019).
Article CAS Google Scholar
Shaikh, D. B. et al. Influences of structural modification of naphthalenediimides with benzothiazole on organic field-effect transistor and non-fullerene perovskite solar cell characteristics. ACS Appl Mater. Interfaces 11, 44487–44500 (2019).
Article CAS PubMed Google Scholar
Wang, D., Ye, T. & Zhang, Y. Recent advances of non-fullerene organic electron transport materials in perovskite solar cells. J. Mater. Chem. A 8, 20819–20848 (2020).
Article CAS Google Scholar
Jameel, M. A. et al. Naphthalene diimide-based electron transport materials for perovskite solar cells. J. Mater. Chem. A 9, 27170–27192 (2021).
Article CAS Google Scholar
Steparuk, A. S. et al. Thieno[3,2-b]indole–benzo[b]thieno[2,3-d]thiophen-3(2H)-one-based D–π–A molecules as electron transport materials for perovskite solar cells. N. J. Chem. 46, 16612–16617 (2022).
Article CAS Google Scholar
Shi, Y. et al. Imide-functionalized acceptor–acceptor copolymers as efficient electron transport layers for high-performance perovskite solar cells. J. Mater. Chem. A 8, 13754–13762 (2020).
Article CAS Google Scholar
Cao, T. et al. Interfacial engineering via inserting functionalized water-soluble fullerene derivative interlayers for enhancing the performance of perovskite solar cells. J. Mater. Chem. A 6, 3435–3443 (2018).
Article CAS Google Scholar
Uddin, M. A. et al. Blading of conformal electron-transport layers in p-i-n perovskite solar cells. Adv. Mater. 34, e2202954 (2022).
Article PubMed Google Scholar
Zheng, Y. et al. Spray coating of the PCBM electron transport layer significantly improves the efficiency of p-i-n planar perovskite solar cells. Nanoscale 10, 11342–11348 (2018).
Article CAS PubMed Google Scholar
Sittinger, V., Schulze, P. S. C., Messmer, C., Pflug, A. & Goldschmidt, J. C. Complex refractive indices of Spiro-TTB and C60 for optical analysis of perovskite silicon tandem solar cells. Opt. Express 30, 37957–37970 (2022).
Article ADS CAS PubMed Google Scholar
Azmi, R. et al. Damp heat-stable perovskite solar cells with tailored-dimensionality 2D/3D heterojunctions. Science 376, 73–77 (2022).
Article ADS CAS PubMed Google Scholar
Liu, J. et al. Efficient and stable perovskite-silicon tandem solar cells through contact displacement by MgF_x. Science 377, 302–306 (2022).
Article ADS CAS PubMed Google Scholar
Chow, P. C., Albert-Seifried, S., Gelinas, S. & Friend, R. H. Nanosecond intersystem crossing times in fullerene acceptors: implications for organic photovoltaic diodes. Adv. Mater. 26, 4851–4854 (2014).
Article CAS PubMed PubMed Central Google Scholar
Elistratova, M. A., Zakharova, I. B. & Romanov, N. M. Obtaining and investigation of C60 <A2B6> semiconductor compounds with a view to create effective solar cells. J. Phys.: Conf. Ser. 661, 012030 (2015).
Google Scholar
Davydov, V. A. et al. Particularities of C60 Transformations at 1.5 GPa. J. Phys. Chem. B 103, 1800–1804 (1999).
Article CAS Google Scholar
Wang, G.-W., Komatsu, K., Murata, Y. & Shiro, M. Synthesis and X-ray structure of dumb-bell-shaped C₁₂₀. Nature 387, 583–586 (1997).
Article ADS CAS Google Scholar
Burlingame, Q. et al. Photochemical origins of burn-in degradation in small molecular weight organic photovoltaic cells. Energy Environ. Sci. 8, 1005–1010 (2015).
Article CAS Google Scholar
Goel, A., Howard, J. B. & Vander Sande, J. B. Size analysis of single fullerene molecules by electron microscopy. Carbon 42, 1907–1915 (2004).
Article CAS Google Scholar
Martin, J. W. et al. Giant fullerene formation through thermal treatment of fullerene soot. Carbon 125, 132–138 (2017).
Article CAS Google Scholar
Capozzi, V. et al. Optical spectra and photoluminescence of C60 thin films. Solid State Commun. 98, 853–858 (1996).
Article ADS CAS Google Scholar
De Bastiani, M. et al. Mechanical reliability of fullerene/tin oxide interfaces in monolithic perovskite/silicon tandem cells. ACS Energy Lett. 7, 827–833 (2022).
Article Google Scholar
De Bastiani, M. et al. Toward stable monolithic perovskite/silicon tandem photovoltaics: a six-month outdoor performance study in a hot and humid climate. ACS Energy Lett. 6, 2944–2951 (2021).
Article Google Scholar
El-Hajje, G. et al. Quantification of spatial inhomogeneity in perovskite solar cells by hyperspectral luminescence imaging. Energy Environ. Sci. 9, 2286–2294 (2016).
Article CAS Google Scholar
Wurfel, P. The chemical potential of radiation. J. Phys. C: Solid State Phys. 15, 3967 (1982).
Article ADS Google Scholar
Gaussian 16 Rev. C.01 (Wallingford, C. T., 2016).
Brinkmann, G., Lisken, D. F. O., Peeters, S., Van Cleemput, A. & CaGe, N.A virtual environment for studying some special classes of plane graphs—an update. MATCH Commun. Math. Comput. Chem. 63, 533–552 (2010).
MathSciNet CAS Google Scholar
Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B 47, 558–561 (1993).
Article ADS CAS Google Scholar
Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 6, 15–50 (1996).
Article CAS Google Scholar
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Article ADS CAS PubMed Google Scholar
Aydin, E. et al. Interplay between temperature and bandgap energies on the outdoor performance of perovskite/silicon tandem solar cells. Nat. Energy 5, 851–859 (2020).
Article ADS CAS Google Scholar

Download references

Acknowledgements

This work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No: OSR-CARF URF/1/3079-33-01, KAUST OSR-CRG 400 RF/1/3383, KAUST OSR-CRG2018-3737. M.L. and A.V. acknowledge Czech Ministry of Education, Youth and Sports grant no. LUASK 22202 and the use of the CzechNanoLab research infrastructure (LM2018110). Computing resources used in this work were provided by the National Center for High-Performance Computing of Turkey (UHeM) under the grant number of 1015902023. The authors thank Shruti Sarwade for her contribution to the c-Si bottom cell preparation and Dr. Safiye Sag Erdem for fruitful discussions on DFT analysis.

Author information

These authors contributed equally: Ahmed A. Said, Erkan Aydin.

Authors and Affiliations

King Abdullah University of Science and Technology (KAUST), KAUST Solar Center (KSC), Physical Science and Engineering Division (PSE), Thuwal, 23955-6900, Kingdom of Saudi Arabia
Ahmed A. Said, Erkan Aydin, Esma Ugur, Badri Vishal, Pia Dally, Bumin K. Yildirim, Randi Azmi, Jiang Liu, Anil R. Pininti, Helen Bristow, Maxime Babics, Arsalan Razzaq, Thomas G. Allen & Stefaan De Wolf
Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ, 08544, USA
Zhaojian Xu, Holly M. Johnson, Manting Gui & Barry P. Rand
Department of Physics, Marmara University, Istanbul, Türkiye
Caner Deger & Ilhan Yavuz
Laboratory of Nanostructures and Nanomaterials, Institute of Physics, Academy of Sciences of the Czech Republic, v. v. i., Cukrovarnická 10, Prague, 162 00, Czech Republic
Aleš Vlk & Martin Ledinský
Nano-C, Inc., 33 Southwest Park, Westwood, MA, 02090, USA
Edward A. Jackson & Henning Richter

Authors

Ahmed A. Said
View author publications
You can also search for this author in PubMed Google Scholar
Erkan Aydin
View author publications
You can also search for this author in PubMed Google Scholar
Esma Ugur
View author publications
You can also search for this author in PubMed Google Scholar
Zhaojian Xu
View author publications
You can also search for this author in PubMed Google Scholar
Caner Deger
View author publications
You can also search for this author in PubMed Google Scholar
Badri Vishal
View author publications
You can also search for this author in PubMed Google Scholar
Aleš Vlk
View author publications
You can also search for this author in PubMed Google Scholar
Pia Dally
View author publications
You can also search for this author in PubMed Google Scholar
Bumin K. Yildirim
View author publications
You can also search for this author in PubMed Google Scholar
Randi Azmi
View author publications
You can also search for this author in PubMed Google Scholar
Jiang Liu
View author publications
You can also search for this author in PubMed Google Scholar
Edward A. Jackson
View author publications
You can also search for this author in PubMed Google Scholar
Holly M. Johnson
View author publications
You can also search for this author in PubMed Google Scholar
Manting Gui
View author publications
You can also search for this author in PubMed Google Scholar
Henning Richter
View author publications
You can also search for this author in PubMed Google Scholar
Anil R. Pininti
View author publications
You can also search for this author in PubMed Google Scholar
Helen Bristow
View author publications
You can also search for this author in PubMed Google Scholar
Maxime Babics
View author publications
You can also search for this author in PubMed Google Scholar
Arsalan Razzaq
View author publications
You can also search for this author in PubMed Google Scholar
Thomas G. Allen
View author publications
You can also search for this author in PubMed Google Scholar
Martin Ledinský
View author publications
You can also search for this author in PubMed Google Scholar
Ilhan Yavuz
View author publications
You can also search for this author in PubMed Google Scholar
Barry P. Rand
View author publications
You can also search for this author in PubMed Google Scholar
Stefaan De Wolf
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

A.A.S. and E.A. conceived the idea and wrote the manuscript. A.A.S., B.K.Y., R.A., and J.L. fabricated and characterized the single-junction perovskite solar cells. E.U. and E.A. fabricated and characterized the perovskite/silicon tandem solar cells. B.V. performed TEM measurements and related data analysis. M.B., T.G.A., and A.R. fabricated silicon bottom cells. C.D. and I.Y. performed the DFT calculations and related data analysis. A.A.S. and H.B. contributed to the MALDI-TOF data analysis. A.V. and M.L. performed the PDS measurements and related data analysis. A.P. performed the mobility measurements. P.D. performed the XPS and UPS measurements and related data analysis. Z.X., H.M.J., M.G. and B.P.R. performed the sublimation process of as-received C₆₀. H.R. and E.J. performed TGA and HPLC analyses. All authors contributed to the writing. E.A. and S.D.W. supervised the project and S.D.W. secured the funding.

Corresponding authors

Correspondence to Ahmed A. Said, Erkan Aydin or Stefaan De Wolf.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Mohammad Khaja Nazeeruddin, Beom Soo Kim and Yongzhen Wu for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Reporting Summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Said, A.A., Aydin, E., Ugur, E. et al. Sublimed C₆₀ for efficient and repeatable perovskite-based solar cells. Nat Commun 15, 708 (2024). https://doi.org/10.1038/s41467-024-44974-0

Download citation

Received: 15 July 2023
Accepted: 08 January 2024
Published: 24 January 2024
DOI: https://doi.org/10.1038/s41467-024-44974-0

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.