Abstract

For some time, the chalcopyrite semiconductor CuInSe2 and its alloy with Ga and/or S [Cu(InGa)Se2 or Cu(InGa)(Se,S)2], commonly referred as CIGS, have been leading thin-film material candidates for incorporation in high-efficiency photovoltaic devices. CuInSe2-based solar cells have shown long-term stability and the highest conversion efficiencies among all thin-film solar cells, reaching 20%. A variety of methods have been reported to prepare CIGS thin film. Efficiency of solar cells depends upon the various deposition methods as they control optoelectronic properties of the layers and interfaces. CIGS thin film grown on glass or flexible (metal foil, polyimide) substrates require p-type absorber layers of optimum optoelectronic properties and n-type wideband gap partner layers to form the p-n junction. Transparent conducting oxide and specific metal layers are used for front and back contacts. Progress made in the field of CIGS solar cell in recent years has been reviewed.

1. Introduction

Current trends suggest solar energy will play an important role in future energy production [1]. Silicon has been and remains the traditional solar cell material of choice. Although silicon is a highly abundant material, it requires an energy intensive process to purify and crystallize. Furthermore, installations of silicon cells require heavy glass protection plates, which reduce residential applications [2].

Recently, commercial interest is beginning to shift towards thin-film cells [3]. Material, manufacturing time, and weight savings are driving the increase in thin-film cells. Cu(In,Ga)Se2 (CIGS) is one of the most promising semiconductors for the absorber-layer of thin-film solar cells [4]. The conversion efficiency of such cells on glass substrates is approaching 20% [5]. Chalcopyrite-based solar modules are uniquely combining advantages of thin-film technology with the efficiency and stability of conventional crystalline silicon cells. Copper indium gallium selenide (CIGS) solar cells have the highest production among thin film technologies. Advances in preparation and efficiency have allowed these cells to be produced rapidly and are approaching market values for carbon-based energy production [6].

The first report on chalcopyrite-based solar cell was published in 1974 [7]. The cell was prepared from a p-type CuInSe2 (CISe) single crystal onto which a CdS film was evaporated in vacuum. This combination of a p-type chalcopyrite absorber and a wide-gap n-type window layer still is the basic concept upon which current cell designs are based. The typical design, first described in 1985 [8] is shown in Figure 1 and a typical cross-section CIGS device structure is shown in Figure 2. The CuInSe2 crystal was replaced by a polycrystalline thin film of the more general composition Cu(In,Ga)(S,Se)2.

Many groups across the world have developed CIGS solar cells with efficiencies in the range of 15–19%, depending on different growth procedures. Glass is the most commonly used substrate, but now efforts are being made to develop flexible solar cells on polyimide [917] and metal foils [2, 1829]. Highest efficiencies of 14.1% and 17.6% have been reported for CIGS cells on polyimide [30] and metal foils [31], respectively. Recently a slight increase in efficiency of 14.7% and 17.7% has been reported for CIGS cells on polyimide and metal foils [32], respectively.

CIGS solar cells also attract considerable interest for space applications due to their two main advantages. It offers specific power up to 919 W/Kg, the highest for any solar cell [23]. CIGS cells are also superior to GaAs cells in radiation hardness [33]. Moreover, the flexibility of these cells allows for novel storage and deployment options [23].

There are several reviews available dealing with different aspects of CIGS solar cells [3539]. The emphasis of the present paper is placed on the progress made in different aspects of CIGS solar cells in the recent times.

2. CIGS Cell Configuration

The Cu-chalcopyrites exhibit the highest efficiencies among the thin-film solar cells with a record labscale efficiency reaching nearly 20% [5]. Most commonly CIGS solar cells are grown in substrate configuration (see Figure 3). This configuration gives the highest efficiency owing to favourable process conditions and material compatibility. Cell preparation starts with the deposition of back contact, usually Mo, on glass, followed by the p-type CIGS absorber, CdS or other weakly n-type buffer layer, undoped ZnO, n-type transparent conductor (usually doped ZnO or In2O3), metal grid, and antireflection coating. It requires an additional encapsulation layer and/or glass to protect the cells surface.

The structure of a CIGS solar cell is quite complex because it contains several compounds as stacked films that may react with each other. Fortunately, all detrimental interface reactions are either thermodynamically or kinetically inhibit at ambient temperature [40]. The cover glass used in substrate configuration is not required for the cells grown in the superstrate configuration (see Figure 3). CIS-based superstrate solar cells were investigated by Duchemin et al. [41] using spray pyrolysis deposition, but efficiencies did not exceed 5%. The main reason for this low efficiency in CdS/CIGS superstrate cells is the undesirable interdiffusion of Cd into CIS (or CIGS) during the elevated temperatures required for absorber deposition on CdS buffer layers [42]. To overcome this problem of interdiffusion more stable buffer materials and low-temperature deposition processes such as electrodeposition (ED), low-substrate temperature coevaporation and screen printing were investigated. Nakada and Mise [43] achieved a breakthrough by replacing CdS with undoped ZnO and coevaporating Se during CIGS deposition. With the additional introduction of composition grading in absorber layer, 12.8% efficiency cells were developed [43].

This coevaporation of Se for incorporation of sodium in CIGS is essential for high-efficiency cells, as the ZnO front contact acts as diffusion barrier for Na from the glass substrate and leads to a low net carrier density in CIGS and cells with low open-circuit voltage ( ) and fill factor (FF) [44]. (The influence of Na on CIGS optoelectronic properties is discussed in Section 6) Investigations of the interface between the ZnO buffer layer and CIGS revealed the presence of a thin layer of Ga2O3 which acts as barrier against photocurrent transport [43, 45, 46]. However, Na-free superstrate solar cells with efficiencies up to 11.2% was obtained, but a strong light-soaking treatment was necessary [47].

Preparing a blocking contact in superstrate structure has been difficult. Only small-area cells have been demonstrated so far and even those show limited performance (see Table 1). It is interesting to note that approaches not using buffer layers have resulted in higher efficiency than those using CdS buffers prepared by various methods.

Another interesting application for superstrate solar cells are tandem solar cells. These solar cells employ two separate solar cell structures for a more efficient conversion of the illumination. Superstrate solar cells are then required as top cell for the short wavelength part of the solar irradiation. Tandem solar cell will not be the part of present discussion. The details can be referred to in the following articles [5356].

3. Back Contact

Molybdenum (Mo) is the most common metal used as a back contact for CIGS solar cells. Several metals, Pt, Au, Ag, Cu, and Mo, have been investigated for using as an electrical contact of CIS- and CIGS-based solar cells [5759]. Mo emerged as the dominant choice for back contact due to its relative stability at the processing temperature, resistance to alloying with Cu and In, and its low contact resistance to CIGS. The typical value of resistivity of Mo is nearly  Ω cm or less. The preferred contact resistivity value is 0.3 cm. Results have been reported in several papers [57, 60, 61] concerning the influence of the mechanical and electrical properties of the Mo films on the performance of the photovoltaic devices. Molybdenum is typically deposited by e-gun evaporation [61, 62] or sputtering [6365] on soda-lime glass which ideally provides an inexpensive, inert, and mechanical durable substrate at temperatures below C. Intrinsic stresses in molybdenum films depend on the experimental deposition parameters [6166], inducing significant changes in the structural and electrical properties. Films with compressive stresses have near bulk like values of the electrical resistivity and a dense microstructure, but films under tensile stresses exhibit altered physical properties and a more open porous structure [6164]. Gross stress may be determined by visual inspection in that highly compressed films tend to buckle up, frequently in zigzag patterns, whereas films under extreme tensile stress develop a system of stress lines that look scratches. Orgassa et al. [67] fabricated CIGS solar cells with different back-contact materials, emphasizing the role of the back contact as an optical reflector. Early results by Russell et al. [68] and Jaegaermann et al. [69] suggested that Mo back contacts for CIS form a Schottky-type barrier with a barrier height of 0.8 eV for the intimate p-doped CIS/Mo contact. The work of Shafarman et al. [70], who analyzed the Mo/CIS interface separately from the cell, shows the contact to be ohmic.

The influence of MoSe2 on the ohmic contact behaviour at the CIGS/Mo interface makes MoSe2 formation an important issue. Fundamental work by Raud and Nicolet [71] on Mo/Se, Mo/In, and Mo/Cu diffusion couples showed Se to react with Mo, forming MoSe2 in very small amounts after annealing at C. Kohara et al. [72] have also shown the formation of nearly ideal ohmic contact between Mo and CIGS that occurs via an intermediate MoSe2 layer. Jones et al. [73] investigated the interface properties of d.c.-sputtered Mo on CIS layers, deposited by coevaporation, and concluded that MoSe2 does not form below C and it might be an artifact of the sputtering process. Similar results have been obtained by Schmid et al. [74] they detected Mo–O and Mo–O–Se compounds, while selenizing the Mo-coated substrate prior to the CIS deposition at C. They concluded that there should be a Schottky-type barrier at the CIS-Mo/MoO2 interface. Wada et al. [75] have also suggested that CIGS/Mo heterocontact including a MoSe2 layer is not Schottky type, but a favorable ohmic type contact. Nishiwaki et al. [76, 77] have also studied the formation of MoSe2 layer at the CIGS/Mo interface during “3-stage” process.

Wada et al. [78] reported the formation of a MoSe2 layer at the Mo/CIGS interface during the second stage of the three-stage process, yet only under (In,Ga)-rich growth and for substrate temperatures higher than C. They found Na to enhance the formation of MoSe2. Assmann et al. [79] have also shown the presence of MoSe2 at the Mo/CIGS interface; they conclude that mechanical stable MoSe2 at the interface gives good adhesion. Recently, Shimizu et al. [80] have studied the variation of Mo thickness from 0.2  m to 0.07  m on the properties of CIGS solar cells. They conclude that there is a tradeoff between the decreased sodium diffusion for thicker Mo layers and decreased fill factor for thin layers. The optimum Mo thickness suggested was 0.2  m. They have also found that water vapour introduced during CIGS growth improve the overall photovoltaic properties.

MoSe2 layers were confirmed also in CuGaSe2-based solar cells by Würz et al. [81]. Contrary to the above results, Ballif et al. [82] could not detect any intermediate compound within the Mo/CIGS interface. Mo back contact for flexible polyimide is also been investigated by Zhang et al. [83].

The properties of molybdenum thin films evaporated onto large area ( ) soda-lime glass substrates at different deposition rates have been investigated by Guillén and Herrero [84]. During the formation of films, Na ion diffuse from the soda lime glass substrate through the Mo back contact into the absorber layer. The diffusion of Na into absorber film depends on the deposition conditions of the Mo back contact [8587]. Nowadays, Mo growth by sputtering or e-beam evaporation is the most commonly used back contact for CIGS solar cells.

Kim et al. [88] have tried Na-doped Mo/Mo bilayer on Alumina substrate and have shown improvement in photovoltaic properties. Nakada [89] has tried transparent conducting oxide as back contact. The TCO back contact deteriorated at high absorber deposition temperature. The formation of Ga2O3 was also reported at the CIGS/ITO and CIGS/ZnO:Al interfaces.

4. CIGS Absorber Layer—Deposition Methods

I–III–VI2 semiconductors, such as CIS or CIGS are often simply referred to as chalcopyrites because of their crystal structure. These materials are easily prepared in a wide range of compositions and the corresponding phase diagrams are well investigated [9092]. For the preparation of solar cells, only slightly Cu-deficient compositions of p-type conductivity are suited [93, 94]. The details of material properties will not be discussed here.

A wide variety of thin-film deposition methods has been used to deposit Cu(In,Ga)Se2 thin films. To determine the most promising technique for the commercial manufacture of modules, the overriding criteria are that the deposition can be completed at low cost while maintaining high deposition or processing rate with high yield and reproducibility. Compositional uniformity over large areas is critical for high yield. Device considerations dictate that the Cu(In,Ga)Se2 layer should be at least 1  m thick and that the relative compositions of the constituents are kept within the bounds determined by the phase diagram.

The most promising deposition methods for the commercial manufacture of modules can be divided into two general approaches that have both been used to demonstrate high device efficiencies and in pilot scale manufacturing. The first approach is vacuum coevaporation in which all the constituents, Cu, In, Ga, and Se, can be simultaneously delivered to a substrate heated at 40 C to C and the Cu(In,Ga)Se2 film is formed in a single growth process. The second approach is a two-step process that separates the delivery of the metals from the reaction to form device-quality films. Typically the Cu, Ga, and In are deposited using low-cost and low-temperature methods that facilitate uniform composition. Then, the films are annealed in a Se atmosphere, also at 40 C to C. The reaction and anneal step often takes longer time than formation of films by coevaporation due to diffusion kinetics, but is amenable to batch processing.

4.1. Coevaporation

The most successful technique for deposition of CIGS absorber layers for highest-efficiency cells is the simultaneous evaporation [95] of the constituent elements from multiple sources in single processes where Se is offered in excess during the whole deposition process. The process uses line-of-sight delivery of the Cu, In, Ga, and Se from Knudsen-type effusion cells or open boat sources to the heated substrate. While a variation of the In-to-Ga ratio during the deposition process leads to only minor changes in the growth kinetics, variation of the Cu content strongly affects the film growth.

The sticking coefficients of Cu, In, and Ga are very high, so the film composition and growth rate are determined simply by the flux distribution and effusion rate from each source. Different deposition variations, using elemental fluxes deliberately varied over time, have been explored using coevaporation. Four different sequences that have been used to fabricate devices with efficiencies greater than 16% are shown in Figure 4.

The first process (Figure 4(a)) is the simplest stationary process in which all fluxes as well as substrate temperature is constant throughout the deposition process [96]. Advanced preparation sequences include a Cu-rich stage during the growth process and end up with an In-rich overall composition in order to combine the large grains of the Cu-rich stage with the otherwise more favourable electronic properties of In-rich composition. The use of this kind of procedure is called “Boeing or bilayer process” (Figure 4(b)) originates from the work of Mickelsen and Chen [97, 98].

This bilayer process yields larger grain sizes compared to the constant rate (single stage) process. This is attributed to the formation of a Se phase during the Cu-rich first stage, which improves the mobility of group III atoms during growth [99101]. Another possibility is the inverted process where first (In,Ga)2Se3 is deposited at a lower temperature (typically ~3 C). Then Cu and Se are evaporated at an elevated temperature until an overall composition close to stoichiometry is reached [102104]. This process leads to smoother film morphology than bilayer process. The so-called “three-stage process” introduced by Gabor et al. [103] from NREL is shown in Figure 4(c).

This method leads, up to now, to the most efficient solar cells. The smoother surface obtained with three-stage process reduces the junction area and thereby is expected to reduce the number of defects at the junction and also facilitates the uniform conformal deposition of a thin buffer layer and prevents ion damage in CIGS during sputter deposition of ZnO/ZnO:Al. Variations of the Ga/In ratio during deposition (Figure 4(d)) allow the design of graded band-gap structures [105].

In one of the other process (shown in Figure 5) is an in-line process in which the flux distribution results from the substrate moving sequentially over the Cu, Ga, and In sources. This was first simulated in a stationary evaporation system [106] and demonstrated by Hanket et al. [107] and has subsequently been implemented by several groups in pilot manufacturing systems.

4.2. Sequential Approach—Selenization of Precursor Material

The interest in sequential processes is sparked by its suitability for large-area film deposition with good control of the composition and film thickness. Such processes consist of the deposition of a precursor material, followed by thermal annealing in controlled reactive or inert atmosphere for optimum compound formation via the chalcogenization reaction. This is commonly referred as selenization of stacked metal or alloy layers. The metals and alloys can be deposited by variety of methods which involve vacuum or no vacuum. The most common of vacuum process is sputtering [108114] and thermal evaporation [113, 115131]. The two step process has many variations in both the precursor deposition and the Se reaction step.

4.2.1. Vacuum-Based Approach

This general approach was first demonstrated by Grindle et al. [132] who sputtered Cu/In layers and reacted them in hydrogen sulfide to form CuInS2. This was first adapted to CuInSe2 by Chu et al. [121]. The highest-efficiency Cu(InGa)Se2 cell reported using the reaction in H2Se is 16.2%, on the basis of the active area [133], but there have been less effort at optimizing laboratory-scale cell efficiencies than with coevaporated Cu(InGa)Se2. Using the two-step selenization/sulfurization approach, some groups have reported CIGSS-solar cells with and efficiency above 600 mV [134] and 14%, respectively, [135]. Lately, an efficiency of 13% [136] has been achieved on  cm large modules. Recently 14.3% has been reported on cm large modules [137] and 14.7% on  cm minimodules [138]. In both cases, sulphur as well as selenium is used for absorber preparation.

The precursor films are typically reacted in either H2Se or Se vapor at 40 C to C for 30 to 60 min to form the best device quality material. Poor adhesion [139] and formation of a MoSe2 layer [140] at the Mo/CuInSe2 interface may limit the reaction time and temperature. Reaction in H2Se has the advantage that it can be done at atmospheric pressure and can be precisely controlled, but the gas is highly toxic and requires special precautions for its use. The precursor films can also be reacted in a Se vapor, which might be obtained by thermal evaporation, to form the CuInSe2 film [141]. A third reaction approach is rapid thermal processing (RTP) of either elemental layers, including Se [142, 143] or amorphous evaporated Cu-In-Se layers [144]. Recently, Chen et al. had tried one step sputtering using Cu-In-Ga alloy target followed by selenization [145].

4.2.2. Nonvacuum-Based Approach

Vacuum-evaporated, polycrystalline copper indium gallium diselenide (CIGS) thin films are used as the absorber layers in the highest efficiency thin film photovoltaic (PV) cells reported to date [5, 146]. However, the high cost and low material utilisation of the equipment used to produce these layers may be a barrier to their commercialization and will increase the cost of the electricity generated by CIGS based systems [147]. Nonvacuum techniques for CIGS deposition offer potential reductions in capital cost and many such techniques have been investigated [148]. These techniques generally split CIGS formation into two stages, one in which the precursor is deposited and one in which the precursor is converted into CIGS.

Nonvacuum approaches to CIGS deposition can be divided into the following categories depending on the deposition method and the scale of mixing of the precursor materials:(1)electrochemical process, (2)particulate process,(3)solution based process.

A detailed review on nonvacuum process dealing with above process is recently published [149]. In view of this, the present section will only update the recent work done in this area, so as not to duplicate the recently published work. In the recent development, Kang et al. [150] have prepared CIGS absorber by selenizing electrodeposited precursor with rich Se and poor Se content. The Se-poor electrodeposited precursor had better crystallinity and increased Ga content. The best cell obtained has efficiency of 1.63% only. In another study, Lai et al. [151] investigated the electrodeposition of CIGS using cyclic voltammetry in a DMF-aqueous solution containing citrate as complexing agent. They performed the cyclic voltammetry study on a ternary Cu-In-Se system, a quaternary system Cu-In-Ga-Se and binary Cu-Se, In-Se, and Ga-Se systems.

Nanoparticle-based approach was carried out by Yoon et al. [152] for the formation of CIS solar cells. They concluded that the Se loss can be minimized by using high heating rate and core-shell structure with a binary compound. The highest efficiency reported was 1.11%. Park et al. [153] have synthesized CIGS absorber using a paste of a Cu, In, Ga, and Se with an aim to develop a simpler and lower cost method of fabricating the absorber layer. Kaigawa et al. [154] have also reported the absorber formation using spray and sintering the film using spot welding machine. Recently a nonvacuum process for preparing nanocrystalline CIGS materials involving an open-air solvothermal route has been demonstrated [155]. In continuation of their earlier work [156, 157], that is, hydrazine-based processor approach for the depositing CIGS and related chalcogenide-based absorber layer, Liu et al. have recently reported 12% efficient CuIn(SeS) solar cell [158]. Hou et al. [159] have also reported the formation of hydrazine-based CuIn(SeS) thin film solar cell.

5. Alternative CIS or CIGS Growth Process

The CIS or CIGS compound has been reported using other alternative techniques. In one of the studies, Ahmed et al. [160] have studied the thermal annealing of flash evaporated CIGS thin films. In another approach [161], efficiency as high as 15.4% was achieved using additional deposition of In, Ga, and Se at high temperatures. In spray pyrolysis, metal salts with a chalcogen reactant are sprayed on heated substrate to form CIS layer. However, a subsequent heat treatment in a reducing atmosphere is still required to improve crystallinity and purity [162164]. Different other approaches have also been successfully adopted for the fabrication of CIGS absorber layers. Flash evaporation has also been used to prepare CIGS film [165]. MOCVD has also been investigated [166] for the deposition of CGS layers as part of a tandem structure, but the growth rate and cell efficiency is rather low. CuInGaSe2 thin films have been prepared by a low pressure metalorganic chemical vapor deposition technique using three precursors without additional Se [167]. A plasma-enhanced CVD has also been reported for fabricating stoichiometric CIS film [168]. CIS thin film has also been deposited by atmospheric pressure metal organic chemical vapor deposition (AP- MOCVD) [169, 170]. Brien et al. [171173] have deposited CuInSe2, CuInS2 and CuGaSe2 thin film by low pressure MOCVD. Recently deposition of CuInSe2 thin film on CuGaSe2 thin film and vice versa has been achieved by a low pressure metal organic chemical vapor deposition technique with three precursors without additional Se [174]. No devices have been attempted by the authors.

Few different techniques have also been used to deposit and characterize CIGS thin film. The technique used are closed-spaced vapor transport [175177] electrodeposition of CIS using ethylene glycol at C [178] CIGSS films by sol-gel route [179] electron beam evaporated CIGS film [180185] CuInSe2 thin films prepared using sequential vacuum evaporation of In, Se, and Cu at moderately low substrate temperatures, avoiding any treatment using toxic H2Se gas [186], CIS using hot wall vacuum evaporation [187], MBE grown CIGS, CIS [188, 189] CIGS using ion-beam plasma evaporation in vacuum [190] and CIGS using a two-stage hybrid sputtering/evaporation method [191].

6. Influence of Sodium

The most important effect of the soda lime glass substrate on Cu(InGa)Se2 film growth is that it supplies sodium to the growing chalcopyrite material. It has been clearly shown that this effect is distinct from the thermal expansion match of soda lime glass [192]. The sodium diffuses through the Mo back contact, which also means that it is important to control the properties of the Mo [193]. Na, incorporated into CIGS absorber layers are widely known to have significantly beneficial effects that lead to enhanced CIGS-related photovoltaic cell efficiencies.

The effect of Na include an improvement in p-type conductivity due to an increase in the effective hole carrier density and improved open circuit voltage ( ) and fill factor for solar cells fabricated from Na doped CIGS [194, 195]. In addition to this, the effect of Na on the growth orientation of CIGS films results in an enhancement of (112) texture [194, 196, 197]. Among the various Na effects, variations in the electrical properties have been well discussed. The observed improvement in has been proposed to originate from an increase in the effective acceptor density [198]. Na in polycrystalline CIGS films is considered to act on the grain boundaries rather than in the bulk [199, 200]. Na substituting on a Cu site results in the formation of a stable compound NaInSe2, which has a larger band gap energy and in turn leads to a larger has also been proposed [201].

The correlation between the CIGS grain size and the presence of Na has been occasionally discussed. While some groups [192, 196, 202] have reported an increase of the grain size in CIGS films containing Na, others did not support these observations [203206]. A decreasing grain size was observed for several Na incorporation methods in a direct comparison [200]. This may be due to the fact that the nature of the effects of Na on grain size depends on the CIGS growth method and the Na-doping process.

The CIS compound formation in rapid-thermal-processed layers was found to be delayed in the presence of Na, resulting in CIS growth at a higher mean temperature, which serves as an explanation for the observed increase in grain size [207]. In any case, the grain size of Na-doped CIGS films seems to have no critical role in solar cell performance [208, 209]. Higher doses of Na are shown to lead to smaller grain sizes, porous films and detrimental to the cell performance [204, 205]. The most obvious electronic effect of Na incorporation into CIGS films is a decrease in resistivity by up to two orders of magnitude [210212]. In one of the recent studies [213] NaF was deposited prior to CIGS absorber and it reported that NaF precursors modify the CIGS growth kinetics: a reduction of the grain size and a slightly enhanced Ga-gradient through the absorber layer were observed. and FF (fill factor) increase when the Na content increases at , during the absorber deposition.

In other studies, Erslev et al. [214] have studied the role of sodium in CIGS solar cells using junction capacitance methods. The increase in solar cell efficiency with sodium was attributed to passivation of a defect state near the CdS/CIGS junction. Recently, Ishizuka et al. [215] have studied the variations in the structural, optical, and electrical properties of polycrystalline Cu(In,Ga)Se2 thin films with Na doping level. Na incorporation into CIGS absorber was controlled using alkali-silicate glass thin layers. They found that the Ga composition gradient in CIGS films became larger and the grain size decreased with increasing Na concentration.

7. CdS Buffer Layers

Semiconductor compounds with n-type conductivity and band gaps between 2.0 and 3.6 eV have been applied as buffer for CIGS solar cells. However, CdS remains the most widely investigated buffer layer, as it has continuously yielded high-efficiency cells. CdS for high-efficiency CIGS cells is generally grown by a chemical bath deposition (CBD), which is a low-cost, large-area process. However, incompatibility with in-line vacuum-based production methods is a matter of concern. Physical vapor deposition- (PVD-) grown CdS layers yield lower efficiency cells, as thin layers grown by PVD do not show uniform coverage of CIGS and are ineffective in chemically engineering the interface properties. For a comprehensive review on CBD-deposited CdS see Ortega-Borges and Lincot [216] and Hodes [217].

The recent trend in buffer layers is to substitute CdS with “Cd-free” wide-bandgap semiconductors and to replace the CBD technique with in-line-compatible processes. The first approach has been to omit CdS and form a direct junction between CIGS and ZnO, but the plasma (ions) during ZnO deposition by RF sputtering can damage the CIGS surface and enhance interface recombination. Possible solutions include ZnO deposited by metal organic chemical vapor deposition (MOCVD), atomic layer deposition (ALD) or a novel technique, called ion layer gas reaction (ILGAR) [218220]. CIS and CIGS solar cells have yielded highest efficiencies with CdS buffer layers deposited by chemical bath deposition (CBD). Omitting the buffer layer always resulted in lower efficiencies [221]. Also, as-grown CIGS cells with CdS buffers deposited by physical vapor deposition (PVD) have always shown significantly lower efficiencies than cells with CBD-CdS buffers [222, 223].

The role of the CdS buffer layer is twofold: it affects both the electrical properties of the junction and protects the junction against chemical reactions and mechanical damage. From the electric point of view the CdS layer optimizes the band alignment of the device [224, 225] and builds a sufficiently wide depletion layer that minimizes tunneling and establishes a higher contact potential that allows higher open circuit voltage value [225]. The buffer layer also play a very important role as a “mechanical buffer” because it protects the junction electrically and mechanically against the damage that may otherwise be caused by the oxide deposition (especially by sputtering). Moreover, in large-area devices the electric quality of the CIGS film is not necessarily the same over the entire area, and recombination may be enhanced at grain boundaries or by local shunts. Together with the undoped ZnO layer, CdS enables self-limitation of electric losses by preventing defective parts of the CIGS film from dominating the open circuit voltage of the entire device [226].

The thickness as well as the deposition method of the CdS layer has a large impact on device performance. During the early days, the device structure consisted of a CuInSe2/CdS junction with a thick (about 1–3  m) CdS layer [227229]. The CdS layer of these devices were most often prepared by evaporation at substrate temperatures between RT and about C or in some cases by sputtering [229]. Also CdS film was often doped with either In [229] or Ga [106] and in some cases a CdS bilayer was used [230, 231] consisting of a thinner high-resistive layer, prepared either by evaporation [143] or chemical bath deposition [194, 230, 231] and a thicker low-resistivity layer, doped with 2% In [231] or Ga [194]. Alternatively, evaporated CdS has been used also in combination with the transparent conducting oxide layer [232234]. Nowadays chemical bath deposition (CBD) is used almost exclusively [235, 236].

In contrast to evaporated films [237], CBD films contain high amounts of oxygen-related impurities that originate from the deposition solution; the amount of oxygen in the films can be as high as at 10–15 at % [237, 238]. Most of the oxygen is present as OH– and H2O [237, 238]. Thus, the composition of the CBD-CdS films is more accurately stated as Cd(S,O,OH) [237]. Additional impurities such as C and N containing compounds result from the side reactions of thiourea [238]. The amount and identity of the impurities, and consequently the performance of the solar cell, depend strongly on the CdS deposition conditions [225, 239241]. Negami et al. [242] for instance, reported an increase of conversion efficiency from 17.6 to 18.5% when the CBD-CdS process was improved.

In addition to the CdS film deposition, the chemical bath also modifies the absorber surface region [235, 243]. Thus, the interface between CIGS and CBD-CdS is not abrupt, but the layers are intermixing to some extent [238, 244]. Both Cu and Cd diffusion play a role, and the intermixing is further enhanced during the post deposition air-annealing [226]. According to Nakada and Kunioka [238], Cu was substituted by Cd at the surface region of CIGS. The diffusion depth of Cd atoms was about 10 nm, which may be related to the thickness of the Cu-deficient surface layer (CuIn3Se5) of CIGS [238]. On the other hand, Heske et al. [244] observe diffusion of Se and In from CIGS into CdS and the diffusion of S from CdS into CIGS. The extent of interdiffusion depends on the structure of the absorber: (224/208) oriented CIGS films have been found to allow more Cd atoms to diffuse into the CIGS film [245].

One advantage of the CBD method as compared to evaporation is that a complete, conformal coverage of the CIGS surface can be obtained at very low thicknesses: already 10 nm has been reported to be sufficient [246]. The coverage depends on deposition conditions, particularly on the concentration ratio of the S and Cd precursors, being better with higher S/Cd precursor ratios [240].

Some studies have been conducted on the fundamental properties of CdS films deposited by an ammonia-free CBD process [247] and very few studies have used an ammonia-free buffer layer for the fabrication of a CIGS solar cell [248]. In a recent work, Mann et al. [249] has also deposited CdS by CBD and used optical reflectance-based measurement of the growing film to determine in situ film thickness.

8. Alternative Buffer Layer

As an alternative to CdS, various materials show promising results. The different buffer layer used and the deposition method for the same is tabulated in Table 2.

The Zn-based compounds tend to form a blocking barrier due to the band alignment with CIGS [253]. Using layers of less than 50 nm thickness, the barrier can be crossed by tunneling of charge carriers, but this poses high requirements on the quality of the deposition process and the CIGS surface to obtain a uniform coverage. The band offset can be reduced as well, if impurities such as hydroxides that can be present in a CBD are incorporated in the CIGS/buffer layer interface [281].

9. Front Contact

There are two main requirements for the electric front contact of a CIGS solar cell device: sufficient transparency in order to let enough light through to the underlying parts of the device, and sufficient conductivity to be able to transport the photo-generated current to the external circuit without too much resistance losses. Transparent conducting metal oxides (TCO) are used almost exclusively as the top contacts. Narrow lined metal grids (Ni–Al) are usually deposited on top of the TCO in order to reduce the series resistance. The quality of the front contact is thus a function of the sheet resistance, absorption and reflection of the TCO as well as the spacing of the metal grids [282].

During the early days of CIS and CIGS substrate cell development, a bilayer of undoped and doped CdS served as a buffer and front contact, respectively [283, 284]. High conductivity in doped CdS was achieved either by controlling the density of donor type defects or by extrinsic doping with Al or In [283, 284]. Spectral absorption loss in the conducting CdS layer was reduced by increasing the bandgap, alloying with ZnS or later replacing it with TCOs with bandgaps of above 3 eV [283]. Transmission spectra of various TCOs are shown in Figure 6.

Today, CIGS solar cells employ either tin doped In2O3 (In2O3: Sn, ITO) [285287] or, more frequently, RF-sputtered Al-doped ZnO. A combination of an intrinsic and a doped ZnO layer is commonly used, although this double layer yields consistently higher efficiencies, the beneficial effect of intrinsic ZnO is still under discussion [226]. It has been shown that device performance increases due to the increase in by 20–40 mV [226]. It has been discussed that resistive oxide layer provides, together with buffer, a series resistance that protects the device from local electrical loses that may originate from inhomogeneties of the absorber [226]. Doping of the conducting ZnO layer is achieved by group III elements, particularly with aluminum [194, 288298]. However, investigations show boron to be a feasible alternative, as it yields a high mobility of charge carriers [290, 299304] and a higher transmission in the long-wavelength spectral region, giving rise to higher currents [305]. For high-efficiency cells the TCO deposition temperature should be lower than C in order to avoid the detrimental interdiffusion across CdS/CIGS interface.

There had been some recent studies of i-ZnO and doped ZnO. Yu et al. [306] have studied the Ni and Al codoped ZnO grown by dc magnetron cosputtering. A comparative study of i-ZnO and ZnO:Al using rf magnetron sputtering and electrodeposition done by Wellings et al. [307] to be used for CIGS solar cells. Pawar et al. [308] have studied the Boron doped ZnO using spray pyrolysis. Few [309, 310] have reported the deposition of ZnO:Al on polymide substrate. Recently, Calnan and Tiwari have discussed in detail regarding High Mobility Transparent Conductor Oxide (HMTCO) [311].

10. Conclusion

Remarkable progress has been made in the development of high efficiency CIGS solar cells. CIGS PV modules have the potential to reach cost-effective PV-generated electricity. Transition from lab to manufacturing has been much more difficult than anticipated.

Each component of the solar cell structure and its manufacturing requires further investigation to simplify the processing and to have more efficient solar cell with lower cost. Few of the key issues related to development of CIGS solar cells are: higher module efficiency, columnar CIGS structures deposited by alternative process for high efficiency cells and modules, thinner absorber layer ( 1 ), and CIGS absorber film stoichiometry and uniformity over large areas.

There is also a need to develop a robust and low temperature (~400 C) deposition process for CIGS for the flexible substrate (polyimide) to facilitate roll to roll manufacturing and to extend the application for space market.

Acknowledgments

The authors would like to thank Mr. Abdul Khader and Ms. Krishna Dalai, for their support during the work. This work has been supported by the Ministry of New & Renewable Energy (31/12/2009/PVSE), New Delhi, Department of Science & Technology (DST), New Delhi (SR/S2/CMP-30/2003), and AICTE, New Delhi (8023/BOR/RID/RPS-78/2007-08).