Recently, organic/inorganic hybrid (e.g. CH3NH3PbI3, CH3NH3PbI3-xClx) perovskite based solar cells attract large attention because of the remarkably high power conversion efficiencies. Most of the devices reported up to now have been fabricated using solution processes. One of the critical issues of the solution processed perovskite solar cells is the reproducibility coming from the difficulties in controlling the uniformity, pin-hole formation, sensitivity to moisture and air of the perovskite film. Vacuum evaporation has a potential for high reproducibility and controllability due to solvent free processes combined with controllable growth parameters in a clean and inert environment. Nevertheless, the fully vacuum?processed perovskite solar cells have been rarely reported up to now. The other crucial issue is the charge transport materials, mainly hole transport materials(HTM’s), to enhance the device performance. The widely used HTM, 2,2’,7,7’-Tetrakis(N,N-di-p-methoxyphenylamine)-9,9’-spirobifluorene (spiro-OMeTAD) is not an optimized HTM for the CH3NH3PbI3 based solar cells, since spiro-OMeTAD possesses a few hundred meV higher HOMO level than valance band (VB) edge of CH3NH3PbI3.
This thesis presents high efficiency perovskite solar cells with high open circuit voltage using full vacuum process employing molybdenum oxide (MoO3) and N,N′-Di(1-naphthyl)-N,N′-diphenyl-(1,1′-biphenyl)-4,4′-diamine (NPB) as a new hole extraction layer. All the layers in the solar cells including the perovskite active layer, hole extraction layers and electron extraction layers were deposited in the vacuum process. Uniform crystalline perovskite layers were able to be grown on the NPB layer under optimized conditions. The grain size of the films were about 100 nm with the root mean square roughness of 9.7 nm. The solar cells employing the MoO3/NPB as the hole extraction layer resulted in a high open circuit voltage (VOC) of 1.12 V, which is one of the highest values reported up to now in perovskite solar cells. Due to the effective hole extraction and high VOC, the devices showed a maximum power conversion efficiency (PCE) of 13.7% with JSC of 18.1 mA/cm2, VOC of 1.12 V and FF of 0.68. It turned out missing one of the MoO3 or NPB layers resulted in much poor solar cell performance due to either the failure of the formation of ohmic contact with the ITO electrode (missing the MoO3 layer) or due to the lack of the electron blocking layer or poor crystallinity of the perovskite layer (missing the NPB layer). The vacuum processed perovskite solar cells showed relatively high reproducibility showing the average value of PCE of 11.1%. The JSC and VOC are distributed within 17.2 mA/cm2 ~ 19.5 mA/cm2 and 0.98 V ~ 1.12 V respectively, exhibiting small deviation within approximately 7% from the average values. The FF values show a relatively wide distribution, 0.39~0.68, hence causing the similar distribution of PCE values, 7.1%~13.7%. Still the variation of the performance of the solar cells fabricated using the vacuum process is fairly narrow compared to the normal solution processed perovskite solar cells. Further study is required to find out the origin of the variation of FF, thereby to enhance the better reproducibility. The vacuum processed perovskite solar cells showed hysteresis depending on sweep direction. The device with MoO3/NPB layer showed PCE of 12.1% with JSC of 18.1 mA/cm2, VOC of 1.09 V and FF of 0.62 in the forward scan direction whereas overall solar cell parameter decreased in the reverse scan direction showing PCE of 10.3% with JSC of 17.1 mA/cm2, VOC of 1.07 V and FF of 0.56. The origin of the hysteresis is important issue, however it has not been clearly figured out yet.