Search

Summary The charge transferability of the electron transfer layers (ETLs) is important for achieving high performance in mesoscopic (mp) perovskite solar cells (PSCs). However, because of the low electron extraction efficiency of TiO2, lithium doping is essentially required. In this work, we compared the electrical properties of mp-TiO2 doped with Li-salts with different anions—LiTFSI, Li2CO3, LiCl, and LiF. Interestingly, we found that the anions of the Li-salt dopants affect the electrical properties of the ETLs and the solar cell performance. The Li2CO3 doping of mp-TiO2 led to conduction bands deeper than those of pristine mp-TiO2 or other doped mp-TiO2. The maximum efficiency of 25.28% and certified efficiency of 24.68% (Newport) was obtained by using Li2CO3 dopant. This optimization of the ETLs properties is expected to greatly contribute to the progress of PSCs by leading to further increases in their efficiency.

Control of the constituent phase and stoichiometry of iron pyrite (FeS2) is a prerequisite for high-performance photovoltaic devices based on this material. If the pyrite contains sulfur-deficiency-related secondary phases which have a metallic character and a high possibility of coexistence in pyrite films, then significant carrier recombination is expected. In this work, the beneficial role of Na in suppressing the formation of nanoscale or amorphous sulfur-deficient secondary phases is reported with experimental evidence, leading to a higher phase purity for solution-processed pyrite films. The potential reduction of charge recombination via these metallic secondary phases results in significant improvements in both the photopotential and photocurrent intensity of Na-modified pyrite films compared with reference samples.

Cu(In,Ga)Se2 (CIGSe) solar cells with absorber thicknesses of

Thinning CIGSe absorber layer to less than 500 nm is desirable for reducing the cost per unit watt of photovoltaic-generated electricity, and also, the semitransparent solar cell based on such a thin absorber can be used in bifacial and superstrate configurations if the back electrode is transparent. In this study, a WO x layer is inserted between Cu(In,Ga)Se2 (CIGSe) absorber and tin-doped indium oxide back-contact to enhance the hole collection at the back electrode. A WO x interlayer with a thickness of 6 nm is found to be optimum because it causes a ∼38% relative increase in the fill factor of a ∼450 nm thick CIGSe-based device compared to the reference device without a WO x interlayer. While fixing the thickness of CIGSe, increasing the WO x interlayer thickness to ≥6 nm results in decreases of solar cell parameters primarily because of the emergence of a GaO x interfacial layer at the CIGSe/WO x junction.

Tin monosulfide (SnS) thin film is one of the most promising absorbers layer for high efficiency solar cells using low cost earth-abundant materials. To fabricate high quality SnS thin films, a simple nanoparticle based approach was employed in this work for phase pure material preparation using non-toxic solvents. Uniform SnS thin films were fabricated via a spin-coating process using phase pure SnS-ORT nanoparticle precursor solutions based on a methanol solvent, followed by post-annealing for grain growth. SnS thin film solar cells were then fabricated and their photovoltaic performance was characterized. The best SnS solar cell fabricated with the as-prepared SnS films with a band gap energy of ~1.25 eV showed a conversion efficiency of 0.12%.

We demonstrate production of high-quality Cu(InGa)(SSe) 2 (CIGSSe) films by high-rate supersonic spray deposition. This technique is unique in creating particle-based films without introducing impurities, because no additives or binders are used. The thin film deposition process was investigated computationally, to understand the pulverization of the incoming particles. These simulations were consistent with experimental observations. Grain growth was improved by adding a 300-nm copper layer atop the CIGSSe film; selenization of the resulting bilayer produced a CuSe liquid flux that assisted the sintering process. The final CIGSSe film-based solar cell had a conversion efficiency of 5.49% with J sc = 18.73 mA/cm 2 , V oc = 0.488 V, and FF = 59.99% in an active area of 0.44 cm 2 .

A solar cell based on Cu(In 1−x ,Ga x )Se 2 (CIGS) with bandgap (E g ) of 1.5 eV is superior because of the low coefficient of power loss and resistive losses. CIGS thin-films with the E g = 1.5 eV and Cu/(Ga + In) (CGI) ratios of 0.92, 0.84, and 0.78 were deposited in this study using single-stage process. Photovoltaic (PV) parameters of the solar cell ameliorated on raising CGI values from 0.78 to 0.84, but abruptly deteriorated on further increasing the CGI value to 0.92. The PV properties of CIGS solar cell with CGI = 0.92 were poor due to the high defect density and low shunt resistance. The optimal CGI range for making efficient wide gap CIGS solar cells through the single-stage process was found to be from ∼0.80 to 0.84. The best CIGS solar cell with CGI value of 0.84 exhibited the conversion efficiency of 11.00%.

The temperature variation of solar cells due to white bias light irradiation during the spectral responsivity measurement and its effect on the spectral responsivity measurement result were investigated for various types of solar cells, such as crystalline silicon (c-Si), Cu(In,Ga)Se 2 (CIGS), and dye-sensitized solar cells (DSSCs). For the investigation, a spectral responsivity measurement system, which can employ the well-known sample temperature control methods (such as the “temperature controlled sample stage” method and the “forced air cooling” method) has been used. Hence, the availability of these sample temperature control methods has also been tested. Through the investigation, it was found that the actual temperature of the solar cells located under the AM1.5G-approximated white bias light can be increased significantly during the spectral responsivity measurement, depending on the sample temperature control methods applied. In addition, it was also found that the increase of sample temperature can lead to a significant error in the measured spectral responsivity, depending on the types of solar cells being measured. In addition, a simple analytic model based on the classical heat transfer theory was developed to understand the temperature variation of the solar cells under the spectral responsivity measurement environment.

In this study, we characterized FeS2 thin films grown by the non-vacuum spin-coating method. We annealed the samples under a sulfur atmosphere at different sulfurization temperatures. The phase transformation from marcasite-containing pyrite to pure pyrite occurs between 350 and 400 °C. The structural phase formation on the films depends on the sulfurization temperature. It is known that pure pyrite thin films are more suitable for solar cells. FeS2 thin films with and without the marcasite phase were investigated in terms of their local electrical properties and carrier transport by conductive atomic force microscopy and Kelvin probe force microscopy. Interestingly, the pure pyrite thin film shows less conducting behavior than the mixed phase sample because the mixed phase thin film has other residues on the surface. Pure pyrite shows two major work functions at 4.64 and 4.70 eV, and the mixed phase sample has multiple surface potential peaks below 4.63 eV, which is the main pyrite FeS2 work function. Pure-phase pyrite thin films are promising for earth-abundant solar cell applications.

Different from conventional photovoltaics, building-integrated photovoltaics needs not only high performance but also a high degree of transparency. Nevertheless, the Cu(In,Ga)Se2 (CIGS) solar cell has advantages in terms of the highest conversion efficiency and stability among all thin-film-based solar cells. The semitransparent (ST) CIGS solar cell using an ultrathin CIGS absorber on a transparent conducting oxide (TCO) experiences loss in fill factor and open circuit voltage due to the poor grain morphology and CIGS/TCO interface. Thus, these issues must be addressed to realize ST CIGS solar cells. Wide-bandgap (1.5 eV) submicron CIGS-based solar cells were prepared in this study unlike conventional CIGS with a bandgap of 1.2 eV, in order to enhance see-throughness. But such cells demonstrated low conversion efficiency due to the poor grain morphology and absence of back grading. Therefore, for the sake of improving grain morphology and to create back grading, a sulfurized-AgGa (AGS) layer was deposited between CIGS and the indium tin oxide (ITO) back contact. Ag from the AGS layer diffused throughout the absorber and thus ameliorated the grain morphology. However, Ga and S in the AGS layer remained confined near the back contact, therefore resulting in the creation of back grading. Consequently, a solar cell based on 230 nm thick CIGS modified with a 45 nm thick AGS layer exhibited efficiency of 5.94% with averaged visible transmittance over 25%. This is the highest reported efficiency for a ST CIGS solar cell with over 20% visible transparency. The CIGS solar cell based on this novel approach can be a competent candidate for building-integrated semitransparent photovoltaics applications.

An investigation into the effects of Se flux on absorber thin film growth at each step of a 3-stage co-evaporation process was conducted to further optimize the performance of CIGS solar cells. In ‘step I’ forming an In-Ga-Se precursor thin film during the 3-stage process, Se flux affected the preferred orientation of the CIGS crystal structure, but not the film morphology. In ‘step II’, no correlation was found between Se flux and the crystal structure, although excessively high Se flux employed throughout the 3-stage process degraded the solar cell performance. A CIGS thin film, with a (220/204) crystal orientation, minor physical surface defects and ∼20 nm thick MoSe2 at CIGS/Mo interface, was obtained by fine control of Se flux conditions (high Se flux at ‘step I’ and low Se flux at ‘step II’) at optimum substrate temperatures. The solar cell fabricated using the aforementioned CIGS thin film showed the highest conversion efficiency of 20.02 %.

Carbon-free CuInSe 2 (CIS) thin film with a dense microstructure has been prepared using a novel non-vacuum based fabrication route. Cu x S y and In 2 Se 3 binary nanoparticles, approximately 10 nm in size, were synthesized by a low temperature colloidal process. The precursor film was deposited using the coating ink formulated with the binary nanoparticles and pyridine, and then annealed in the rapid thermal annealing (RTA) chamber at 540 °C for 15 min under selenium (Se) atmosphere. Scanning electron micrographs, X-ray diffraction patterns and Raman spectra showed a phase pure carbon-free and dense CIS thin film was prepared in this method. A solar cell device fabricated using this CIS thin film showed the following photovoltaic characteristics: V OC = 350 mV, J SC = 24.72 mA cm −2 , FF = 38.73% and η = 3.36% under standard AM 1.5 condition.

A primary reference solar cell calibration technique recently established at the Korea Institute of Energy Research in Korea is introduced. This calibration technique is an indoor method that uses a highly collimated continuous-type solar simulator and absolute cavity radiometer traceable to the World Radiometric Reference. The results obtained using this calibration technique are shown with a precise uncertainty analysis, and the system configuration and calibration procedures are introduced. The calibration technique avoids overestimating the short-circuit current of a reference solar cell due to multiple reflections of incident simulator light using a novel method. In addition, the uncertainty analysis indicates that the calibration technique has an expanded uncertainty of approximately 0.7% with a coverage factor of k = 2 for a c-Si reference cell calibration. In addition, the developed primary reference solar cell calibration technique was compared with other techniques established in the World Photovoltaic Scale (WPVS) qualified calibration laboratories to verify its validity and reliability.

To maximize the conversion efficiency of solar cells by improving the trapping of incident light, surface texturing of multicrystalline silicon wafers is a more permanent and effective solution to decrease reflections compared with antireflection coatings. In general, many techniques of wet chemical surface texturing depend on crystal orientation. Also, these processes place a limitation on the growth of solar production capacity due to consumption of the large amount of material, de-ionized water and chemicals. Reactive ion etching (RIE) is therefore useful for mc-Si solar cells with the random grain orientation. In this study, we have fabricated mc-Si solar cells with a large-area (156 mm × 156 mm) by a random RIE texturing method using SF 6 /O 2 plasma chemistry. A large amount of silicon loss, which is due to saw damage removal and texturing process of mc-Si wafers, have dramatically decreased by applying RIE texturing process. As a result of the optimum RIE process, the best mc-Si solar cell showed conversion efficiency, fill factor, short circuit current density, and open circuit voltages as high as 17.4%, 80%, 35 mA/cm 2 and 620 mV, respectively. This study demonstrates that it is possible to apply the thin mc-Si Solar cells fabrication for low cost and high efficiency in conventional industrial production line.

This study examined the effect of UV-spectrum light soaking on solar cell degradation tests. An indoor light soaking test was evaluated over three different spectral ranges: “UV only”, “UV blocked” and “Full spectrum”. a-Si:H and poly-crystalline silicon solar cell technologies were studied by light soaking tests with the same optical filter configurations. The I–V measurement results demonstrated that “UV only” irradiated solar cells exhibited the smallest output power degradation, which was only half of a percent variation compared with the full spectrum light soaking case. Using a filter that excluded the effect of the UV spectral range on light soaking, the “UV blocked” case also exhibited a significant output power degradation of the solar cells. A comparative analysis of the solar cell response, based on the I–V characteristics and the diode ideality factor under these different light soaking spectra, demonstrated the importance of the full spectrum light soaking test in the evaluation of the long-term performance of solar cells.

The influence of the conventional depositing processes on the optical properties of Cu(In,Ga)Se 2 (CIGS) thin films in solar cell structures was investigated by measuring the photoluminescence (PL) and Raman spectra of the CIGS layer at each stage of the solar cell deposition process. The intensities of the PL and the Raman A 1 mode increase after the CdS buffer layer is deposited, suggesting that the CdS layer either improves the optical quality of the CIGS film or protects it from degradation due to environmental factors. The temperature and excitation power dependences of the PL for the bare CIGS sample are very different from those for the samples with the CdS layer, reflecting different characters of the luminescence centers near the surface of the CIGS layer. On the other hand, the lateral homogeneity, as seen in the micro-PL and micro-Raman images, does not seem to improve. After the ZnO window layer is deposited, the overall PL and Raman intensities do not change much, although the intensity distribution becomes more inhomogeneous.

DC and pulse-reverse electrodeposition mode were employed for the deposition of polycrystalline Cu(In,Ga)Se 2 thin films. In comparison with DC electrodeposition mode, films obtained by pulse-reverse electrodeposition were smoother, denser and more uniform with good adhesion. The Ga content in final composition of CIGS thin film was improved in pulse-reverse electrodeposition mode. In addition, pulse-reverse electrodeposited CIGS thin films were more crystalline with chalcopyrite structure. The compact morphology without pores in the deposit was achieved in the pulse-reverse electrodeposited CIGS thin films by varying duty cycle.

CuInSe2 (CIS) thin films were formed on a Mo-coated soda lime glass with a newly formulated hybrid ink composed of binary Cu2−xSe nanoparticles and an In precursor solution. The thin films were fabricated by the spin coating method and a subsequent selenization process, and the compositional and structural properties of the CIS thin films were characterized. Deposition of a hybrid ink provides the unique advantage of producing centers composed of pure material-based nanoparticles that can promote stress-relief and crack-deflection. Additionally, hybrid inks offer efficient binding with nanoparticles by using precursor solutions without other organic binders. Binary Cu2−xSe nanoparticles were synthesized by a low-temperature colloidal process, and an In precursor solution was prepared by using a non-toxic chelating agent for dispersing the In component stably. The band gap of the CIS thin film was measured to be 1.14 eV by the external quantum efficiency (EQE) technique, and the best conversion efficiency of the fabricated device was determined to be 4.19%.

A common feature of the inorganic thin films including Cu(In,Ga)(S,Se)2 fabricated by nonvacuum solution-based approaches is the doubled-layered structure, with a top dense inorganic film and a bottom carbon-containing residual layer. Although the latter has been considered to be the main efficiency limiting factor, (as a source of high series resistance), the exact influence of this layer is still not clear, and contradictory views are present. In this study, using a CISe as a model system, we report experimental evidence indicating that the carbon residual layer itself is electrically benign to the device performance. Conversely, carbon was found to play a significant role in determining the depth elemental distribution of final film, in which carbon selectively hinders the diffusion of Cu during selenization, resulting in significantly Cu-deficient top CISe layer while improving the film morphology. This carbon-affected compositional and morphological impact on the top CISe films is a determining factor for the device efficiency, which was supported by the finding that CISe solar cells processed from the precursor film containing intermediate amount of carbon demonstrated high efficiencies of up to 9.15% whereas the performances of the devices prepared from the precursor films with very high and very low carbon were notably poor.

The compositional distribution of elements is known to be significant for the high conversion efficiency of CZTSe solar cells. As detailed understanding of the Cu/(Zn+Sn) ratio in the light absorption layer is important, Cu2ZnSnSe4 (CZTSe) films grown via the co-evaporation process under different copper fluxes were characterized. It is difficult to evaluate the Cu content effect on the properties of CZTSe films grown using a co-evaporation process with Cu, Zn, Sn, and Se elemental effusion sources because the Cu flux variation during the process also induces other element ratio changes. Furthermore, the Zn/Sn ratio shows significant correlation to the Cu/(Zn+Sn) ratio variation in CZTSe thin films. Replacing the zinc metal effusion source with the ZnSe compound source resulted in less fluctuation in the Zn/Sn variation according to Cu flux change during the CZTSe co-evaporation. This can be useful in evaluating the effect of the different Cu ratios on the CZTSe solar cell characteristics.

This study examined the effects of Ga content in the CIGS absorber layer on the properties of the corresponding thin films and solar cells fabricated using a co-evaporation technique. The grain size of CIGS films decreased with increasing Ga content presumably because Ga diffusion during the 2nd stage of the co-evaporation process is more difficult than In diffusion. The main XRD peaks showed a noticeable shift to higher diffraction angles with increasing Ga content, which was attributed to Ga atoms substituting for In atoms in the chalcopyrite structure. Band gap energy and the net carrier concentration of CIGS films increased with Ga/(In + Ga) ratios. Regarding the solar cell parameters, the short circuit current density (JSC) decreased linearly with Ga/(In + Ga) ratios due to the lack of absorption in the long-wavelength portion of the spectrum, while the open circuit voltage (VOC) increase with those. However, VOC values at high Ga/(In + Ga) regions (>0.35) was far below than those extrapolated from the low Ga contents regions, finally resulting in an optimum Ga/(In + Ga) ratio of 0.28 where the solar cell showed the highest efficiency of 15.56% with VOC, JSC and FF of 0.625 V, 35.03 mA cm−2 and 0.71, respectively.

A non-vacuum process for Cu(In,Ga)Se 2 (CIGS) thin film solar cells from nanoparticle precursors was described in this work. CIGS nanoparticle precursors was prepared by a low temperature colloidal route by reacting the starting materials (CuI, InI 3 , GaI 3 and Na 2 Se) in organic solvents, by which fine CIGS nanoparticles of about 15 nm in diameter were obtained. The nanoparticle precursors were then deposited onto Mo/glass substrate by the doctor blade technique. After heat treating the CIGS/Mo/glass layers in Se gas atmosphere, a complete solar cell structure was fabricated by depositing the other layers including CdS buffer layer, ZnO window layer and Al electrodes by conventional methods. The resultant solar cell showed a conversion efficiency of 0.5%.

A non-vacuum process for fabrication of Cu(In, Ga)Se2 (CIGS) absorber layer from the corresponding nanoparticle precursors was described. CIGS nanoparticle precursors was prepared by a low temperature colloidal route by reacting the starting materials (CuI, InI3, GaI3 and Na2Se) in organic solvents, by which fine CIGS nanoparticles of about 20 nm in diameter with chemical composition of Cu0.9In0.64Ga0.23Se2.00 were obtained. The nanoparticle precursors were mixed with organic binder material for the rheology of the mixture to be adjusted for the doctor blade method. After depositing the mixture of CIGS with binder on Mo/glass substrate, the samples were preheated on the hot plate in air to evaporate remaining solvents and to burn the organic binder material. Subsequently, the resultant (porous) CIGS/Mo/glass sample was selenized in a two-zone Rapid Thermal Process (RTP) furnace in order to get a solar cell applicable dense CIGS absorber layer. The selenization at 550 °C for 15 min with Se gas evaporated at 400 °C resulted in the growth of CIGS particles up to a few hundred nanometers in grain size in the upper parts of the layer with remaining unreacted small particles in the lower parts.

Effects of heat treatment in nitrogen or Se atmosphere on the properties of Cu(In,Ga)Se 2 (CIGS) nanoparticles were investigated to extract optimum sintering conditions for fabrication of solar cell applicable CIGS absorber films. In nitrogen atmosphere, as the temperature increases from 100 to 400 °C the intensity of X-ray diffraction (XRD) peaks corresponding to the (1 1 2), (2 2 0) and (3 1 2) planes of the chalcopyrite CIGS increases, and the peak positions shift to lower angle regions without any particle growth in scanning electron microscopy (SEM) analysis, which is in consistent with the significant In and Ga loss in the EDS data. When the temperature further goes up to 500 °C, parts of CIGS are decomposed and Cu and CuSe 2 phases are observed. From these results, the heat treatment in nitrogen atmosphere is found to have no beneficial effect on the sintering of the particles and only induces loss of In and Ga. On the other hand, heat treatment in Se atmosphere at a substrate temperature of 550 °C with Se vapor evaporated at 250 and 450 °C provided much enhanced growth of the particles, specially up to 500 nm at 450 °C, and increased crystallinity without In or Ga loss, reflecting that Se supply played a critical role in the growth of the CIGS nanoparticles.

To make a dense CIGS absorber layer, spray deposited CIGS films were annealed in the two-zone RTP furnace in Se atmosphere. More Se supply by increasing Se evaporation temperature or by increasing the flow rate of carrier gas resulted in the larger CIGS grains. However, a thick MoSe2 layer was formed between CIGS and Mo, as the Se supply increased, results in partial detachment of CIGS/MoSe2/Mo layers from the glass substrate. From the result, it was found that the short heat- treatment with high Se vapor pressure is better than the long heat-treatment with low Se vapor pressure. The large CIGS grains without peeling off, can be obtained from the following conditions; Se evaporation temperature of 450oC, substrate temperature of 550oC, annealing time of 5 min, and flow rate of carrier gas of 30 sccm.

The photovoltaic properties of CIGS cells on an alumina substrate were improved through the use of Na-doped Mo as the bottom layer of a Mo back contact. Na was supplied to the CIGS bulk region from an alumina/Na-doped Mo/Mo structure, similar to the Na diffusion from soda-lime glass. The diffusion of Na from the Na-doped Mo was controlled effectively compared to that from Soda-lime glass (SLG). The present results indicate that Na-doped Mo acts as a Na source material and that the Na amount can be controlled by adjustment of thickness of Na-doped Mo layer, without the use of an alkali barrier layer. The highest conversion efficiency of 13.34% ( J sc = 34.62 mA/cm 2 , V oc = 0.58 V and FF = 66%) for an active area of 0.45 cm 2 on an alumina substrate was obtained for 100 nm Na-doped Mo/1000 nm Mo.

We prepared CZTSe thin films by the co-evaporation method. In order to investigate the presence of various kinds of secondary phases, we also prepared reference ZnSe, SnSe, CuSnSe, and CuSe thin films in the same way. The depth dependence of the secondary phases was investigated by preparing samples that are sputtered to various depths. We used Raman spectroscopy measurement with several lasers to detect secondary phases. CZTSe films were compared with the reference films, and only ZnSe and MoSe2 phases were detected. The relative amount of ZnSe varied with the depth from the sample surface. The ZnSe Raman peak seems to originate from the top and the bottom of the sample since the signal decreased initially and then increased as deeper parts of the film were probed. Intensity of CZTSe peak did not change much until the bottom part of the sample was reached and the MoSe2 peak appeared.

Multiphase CuInSe2 (CISe) nanoparticles including the secondary CuSe phase are synthesized by a polyol-based, microwave-assisted solvothermal method. The critical role of CuSe phase in generating the pore-free, dense CISe absorber layer for a high performance thin-film photovoltaic device is investigated through the comparative study on CISe absorber layers derived from both multiphase and single phase CISe nanoparticles. The high performance, solution-processed CISe thin-film solar cell, with a conversion efficiency of 8.2%, is obtained based on multiphase CISe nanoparticles that undergo the dramatic microstructural evolution, phase transformation, and composition adjustment during annealing under Se atmosphere.

CuInSe(2) (CIS) absorber layers for thin film solar cells were formed via a nonvacuum route using nanoparticle precursors. A low-temperature colloidal process was used to prepare nanoparticles by which amorphous Cu-In-Se nanoparticles were formed within 1 min of reaction without any external heating. Raman spectra of the particles revealed that they were presumably mixtures of amorphous Cu-Se and In-Se binaries. Selenization of the precursor film prepared by doctor blade coating of the Cu-In-Se nanoparticles resulted in a facile growth of the particles up to micrometer scale. However, it also left large voids in the final film, which acted as short circuiting paths in completed solar cells. To solve this problem, we applied a solution-filling treatment in which a solution containing Cu and In ions was additionally coated onto the precoated nanoparticles, resulting in a complete infiltration of the filler solution into the pores in the nanoparticles based film. By this approach, short circuiting of the device was significantly mitigated and a conversion efficiency of up to 1.98% was obtained.

The wide band-gap CuIn1-xAxlSe2(CIAS) films were obtained by selenization process of metallic precursors. The metallic precursors were deposited sequentially by using sputtering system. As the ratio of Al/(Al+In) in the precursors increased, the chalcopyrite (112) peak shifted to high value and the band-gap of CIAS layer increased to 1.38 eV. However the bi-layer morphology with well crystallized large grain on the surface and small grain thin bottom layer was observed. Although the sequences of precursors were changed in order to get uniform layer, the distinguishable difference was not observed. Moreover the films were peeled off completely during the deposition of buffer CdS. The other process conditions such as selenizing at high temperatures, Se containing precursors and post annealing will be tried to remove the bi layer and to get homogeneous films.

Spray deposited nanoparticle derived Cu(In,Ga)Se2 (CIGS) layers were selenized with a two-zone RTP furnace in order to get a solar cell applicable dense CIGS absorber layer. Effects of various selenization parameters including Se evaporation temperature, flow rate of carrier gas and substrate temperature on the growth of CIGS nanoparticles were investigated. The experimental results revealed that higher amount of Se vapor generated and transported to CIGS nanoparticles, i.e., higher temperature of Se evaporation and/or higher flow rate of carrier gas, resulted in the larger CIGS grains. High substrate temperature also enhanced the growth of individual nanoparticles. However, direct contact of Se vapor with bare Mo surface through the large numbers of pores in the layer gave rise to a formation of very thick MoSe2 layer which at times caused the CIGS/MoSe 2/Mo layers to peel off the glass substrate, presumably due to the significant molar volume difference between Mo and MoSe2

Multiphase CuInSe2 (CISe) nanoparticles including the CuSe phase are synthesized by the microwave-assisted solvothermal method. Without additional processing, multiphase CISe nanoparticles facilitate the solution-processed CISe absorber layer with a dense microstructure, large grains, high crystallinity, and composition controllability, which are essential for acceptable thin-film solar cell performance. The high performance, solution-processed CISe solar cell, with a conversion efficiency of 8.2%, is obtained through phase transformation, microstructural evolution, and composition adjustment by selenization (annealing under a Se atmosphere) at 530 °C.

We report a new approach to fabricating Cu(In,Ga)Se2 (CIGSe) light absorbing layers for thin film solar cells without Ga segregation using a sputtering and single-step selenization process. To mitigate Ga segregation at the CIGSe/back-contact region, which has frequently been observed in the selenization of metal/alloy precursor layers, we used Se-containing precursor layers (Cu-In-Ga-Se) to capture Ga in covalently bonded structures and investigated the effects of Se content in the precursor layers on the properties of the selenized CIGSe films and the devices. As the Se content in the precursor layer increased, Ga segregation was significantly mitigated, resulting in a completely homogenized Ga distribution when the Se/metal ratio of the precursor films is over 0.8. Finally, a thin CIGSe film (∼670 nm) with a uniform Ga distribution was processed to fabricate a solar cell, and the device exhibited a conversion efficiency of 11.7% with an open circuit potential of 0.6 V. An increase of the CIGSe film thickness to 1.55 μm resulted in a device efficiency of up to 13.16%.