Elsevier

Solar Energy

Volume 197, February 2020, Pages 78-83
Solar Energy

High efficiency perovskite solar cells using nitrogen-doped graphene/ZnO nanorod composite as an electron transport layer

https://doi.org/10.1016/j.solener.2019.12.062Get rights and content

Highlights

  • Nitrogen-doped graphene/ZnO nanorod composite (NG-ZnO NR) was used as electron transport layer (ETL) to replace TiO2 for perovskite solar cells.

  • NG-ZnO NR composite based perovskite solar cells achieved a highest efficiency (16.82%) as compared to pristine ZnO nanorods (12.87%).

  • NG-ZnO NR composite promotes formation of uniform perovskite thin films with a high surface coverage area in comparison to pristine ZnO nanorods.

  • NG-ZnO NR composite based perovskite solar cells exhibited improved photocurrent and photon to electron conversion efficiency due to enhancement in light absorption.

Abstract

We demonstrate for the first time, the photovoltaic performance of perovskite solar cells (PSCs) by employing nitrogen-doped graphene/ZnO nanorod nanocomposites (NG-ZnO NR NCs) as an electron transport layer (ETL). A series of PSCs are fabricated by varying the NG concentration from 0 to 1 wt% with an increment of 0.2 wt% in ZnO NRs having a device structure of FTO/NG-ZnO NR NCs/CH3NH3PbI3/Spiro-MeOTAD/Ag. The perovskite films grown on NG-ZnO NR NCs exhibited a uniform film with high surface coverage area with an improved crystallinity. By employing photoluminescence measurements, it was found that NG-ZnO NR NCs reduced the interfacial resistance by improving charge carrier extraction, which is in correlation with an improvement in photocurrent and fill factor in the corresponding devices. At an optimized concentration of NG (0.8 wt%) in ZnO NRs, PSCs obtained an improvement in photocurrent from 17.38 mA/cm2 to 21.98 mA/cm2 and the power conversion efficiency from 12.87% to 16.82%, respectively.

Introduction

Since the first report of Kojima et al. in 2009, perovskite solar cells (PSCs) have captured a tremendous attention and became a game changer in photovoltaic community due to an excellent optoelectronic properties of hybrid organic-inorganic perovskite halides (CH3NH3PbX3, X = Cl, Br, and I). It possesses high light absorption coefficient (1.5 × 104 cm−1), direct bandgap, longer electron/hole diffusion lengths (~100 nm and ~1 µm for CH3NH3PbI3 and CH3NH3PbI3-xClx, respectively), high carrier mobility, and an ease of fabrication (Kojima et al., 2009, Stranks et al., 2013, Dong et al., 2015, Xing et al., 2013). Indeed, PSCs have skyrocketed in their power conversion efficiency (PCE) from 3.8% to above 20% and recently reported a highest PCE of 25.2% by NREL (National Renewable Energy Laboratory in the USA) with continuous research thrust in device engineering, solvent engineering, optimizing processing conditions, and by employing various device fabrication strategies (Burschka et al., 2013, Zhou et al., 2014, Collavini et al., 2015, Saliba et al., 2016, nrel, xxxx, Sun et al., 2018, Wang et al., 2018, Di et al., 2018).

PSCs broadly fabricated in two different configurations; one is based on mesoporous scaffold structure and the other one is based on planar structure (p-i-n and n-i-p). Here, our focus is on mesoporous configuration, mesoscopic TiO2 nanostructure has been widely used as an electron transport layer (ETL). However, using TiO2 as an ETL in PSCs may be a hurdle towards commercialization of PSCs because of high temperature annealing process at ~450 °C, and lower electron mobility in TiO2 causes an unbalanced charge transportation and degradation of devices photovoltaic performance (Cheng et al., 2015, Ponseca et al., 2014). Most importantly, high temperature processing makes unfavorable for the fabrication of PSCs on flexible substrates and increases the production cost. Therefore, an alternate ETL such as n-type ZnO having direct band gap and superior electron mobility, has been reported previously as a potential candidate to replace TiO2 (Chen et al., 2012, Kim et al., 2013, Bi et al., 2013). ZnO has been successfully demonstrated as an efficient ETL in both mesoporous and planar PSCs (Aleksandra et al., 2017, Liu and Kelly, 2014, Tseng et al., 2015). PSCs using ZnO as ETL require a lower annealing temperature (80–100 °C) than the devices using TiO2 as ETL. However, the PCE of PSCs using ZnO nanoparticles (NPs) as ETL is low due to charge recombination at ZnO NPs/perovskite interface. Previous reports showed that NPs have low electron mobility due to presence of grain boundaries, which causes charge carrier recombination (Kim et al., 2014, Kim et al., 2013). In addition, the perovskite crystal structure undergo decomposition on the surface of ZnO thin films due to presence of hydroxide groups, which limits the PCE of PSCs (Cheng et al., 2015).

Therefore, one dimensional (1-D) ZnO nanostructures have been employed in lieu of ZnO NPs to improve charge transportation and infiltration of perovskite material into the mesoporous nanostructures to enhance the PCE of PSCs. For instance, Kumar et al. reported a PCE of 8.90% by employing chemical bath deposited ZnO nanorods in PSCs (Kumar et al., 2013). Son et al. demonstrated a PCE of 11.13% using hydrothermally grown ZnO nanorods by optimizing their length and diameter. The photovoltaic response of ZnO nanorods as ETL shows efficient charge collection in PSCs when compared to PSCs fabricated using TiO2 nanorods ETL (Son et al., 2014). Nevertheless, the PCE of PSCs using ZnO nanorods is still not high as that of a standard TiO2 based device architecture. The low performance is attributed to nonoptimized dimensions of ZnO nanorods (length and diameter) having improper perovskite crystal growth because of low annealing temperature and time (Son et al., 2014). It is noted that fabrication of high quality perovskite films on nanorods surface is difficult because of unique structure and morphology of ZnO nanorods (Zhang et al., 2018). To further improve the PCE of PSCs, recently, graphene oxide (GO) nanocomposites and graphene derivatives also have been introduced into PSCs to enhance the PCE by improving charge transportaion and quality of perovskite films. Wu et al. reported a PCE of around 12% by employing GO as hole conductor in planar heterojunction PSCs which helped to improve crystallization, surface coverage area and hole extraction from perovskite to GO (Wu et al., 2014). Ahmed et al. demonstrated GO in ZnO matrix as ETL having low resistivity and fabricated HTL free PSCs with maximum PCE of 4.52% (Ahmed et al., 2016). Chandrasekhar et al. demonstrated the effect of graphene concentration on the photovoltaic performance of PSCs by employing spray deposited G/ZnO nanocomposites as ETL (Chandrasekhar and Komarala, 2017). Tavakoli et al. achieved a PCE of 15.2% by employing quasi core-shell structure of ZnO/rGO quantum dots as ETL and found to be an improvement in charge carrier extraction quickly from perovskite layer to reduce the carrier recombination (Tavakoli et al., 2016). Wang et al. reported a PCE of 15.6% by employing low temperature processed graphene/TiO2 nanocomposites as ETL (Wang et al., 2014). Thus, an improved performance by using graphene nanoflakes and its derivatives in nanocomposites suggested that graphene/metal oxide nanocomposites have potential to contribute towards fabrication of efficient and low-cost PSCs.

In this work, we systematically investigated the photovoltaic performance of PSCs by employing nitrogen-doped graphene (NG) and ZnO NR nanocomposites (NG-ZnO NR NCs) as an ETL. The effect of NG on the device performance is explored in relation with morphology and optical properties of perovskite films by varying different concentrations of NG in the range of 0 to 1 wt% with an interval of 0.2 wt%. At an optimized concentration (0.8 wt%) of NG, we observed a significant improvement in photocurrent by ~21% from 17.38 mA/cm2 to 21.98 mA/cm2 and PCE by ~ 23% from 12.87% to 16.82%, respectively. Thus, graphene derivative-ZnO NR composites appear to be a promising electron transport layer for obtaining high efficiency PSCs.

Section snippets

Experimental section

NG was synthesized by a previously reported hydrothermal method (Chen et al., 2014). Initially, graphene (1 mg/ml) was dispersed in deionized water (25 ml) using an ultrasonication bath and well mixed with the urea solution (200 mg/ml). The resulting solution was transferred to a 100 ml Teflon vessel and subjected to hydrothermal treatment at 180 °C for 5 h. The obtained black product was washed with deionized water and ethanol for several times. Finally, NG was obtained by vacuum drying at

Results and discussion

Fig. 2a shows the XRD patterns of vertically grown ZnO NRs on FTO glass substrate. The diffraction peaks at (1 0 0), (0 0 2), (1 0 1), and (1 0 2) planes are corresponding to hexagonal wurtzite crystal structure of ZnO as per the JCPDS file No: 36-1451. An intense (0 0 2) diffraction peak reflects the c-axis orientation of elongated ZnO nanorods on the FTO glass substrate (Vayssieres, 2003). The other diffraction peaks denoted by star (*) symbol corresponds to FTO glass substrate. The XRD

Conclusions

In summary, the incorporation of NG into ZnO NRs as an ETL has improved the photovoltaic performance of PSCs. Results indicated that NG facilitates better infiltration of perovskite precursor in ZnO NRs resulting in uniform perovskite film with high surface coverage and improved hole transport properties. With increase in NG concentration, a significant enhancement in Jsc, FF, and PCE are achieved due to improved light absorption and charge transport. At an optimized concentration (0.8 wt%) of

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

P. S. Chandrasekhar would like to thank Department of Science and Technology (DST), Govt. of India for providing fellowship (code no. IF120755) under DST INSPIRE program, and gracefully acknowledging Indo-US science and technology forum (IUSSTF) for providing an internship through Bhaskara Advanced Solar Energy Fellowship (BASE) program at South Dakota State University (SDSU), USA. We would like to thank central research facility (CRF) at IIT Delhi for their kind help in SEM analysis. This work

References (40)

  • Y. Di et al.

    Sol. Energy Mater. Sol. Cells

    (2018)
  • Q. Sun et al.

    Sol. Energy Mater. Sol. Cells

    (2018)
  • M. Wang et al.

    Sol. Energy Mater. Sol. Cells

    (2018)
  • M.I. Ahmed et al.

    AIP Adv.

    (2016)
  • D. Aleksandra et al.
  • D. Bi et al.

    Nanoscale

    (2013)
  • J. Burschka et al.

    Nature

    (2013)
  • P.S. Chandrasekhar et al.

    RSC Adv.

    (2017)
  • S. Chen et al.

    J. Mater. Chem.

    (2012)
  • Y. Chen et al.

    Nanoscale Res. Lett.

    (2014)
  • Y. Cheng et al.

    ACS Appl. Mater. Interfaces

    (2015)
  • S. Collavini et al.

    Angew. Chem. Int. Ed.

    (2015)
  • Q. Dong et al.

    Science

    (2015)
  • L.E. Greene et al.

    Angew. Chem. Int. Ed.

    (2003)
  • G.S. Han et al.

    ACS Appl. Mater. Interfaces

    (2015)
  • J.-H. Im et al.

    Nanoscale

    (2011)
  • N.J. Jeon et al.

    Nat. Mater.

    (2014)
  • H.-S. Kim et al.

    Nano Lett.

    (2013)
  • H.-S. Kim et al.

    J. Phys. Chem. C

    (2014)
  • K.-D. Kim et al.

    ACS Appl. Mater. Interfaces

    (2013)
  • Cited by (0)

    View full text