**7. Application of TCO in solar cells**

Solar cells exploit the photovoltaic effect that is the direct conversion of incident light into electricity. Electron–hole pairs generated by solar photons are separated at a space charge region of the two materials with different conduction polarities. Solar cells represent a very promising renewable energy technology because they provide clean energy source (beyond manufacturing) which will reduce our dependence on fossil oil. The principles of operation of solar cells have been widely discussed in detail in the literature and as such will not be repeated here. Rather, the various solar cell technologies will be discussed in the context of conduction oxides. Solar cells can be categorized into bulk devices (mainly single-crystal or large-grain polycrystalline Si), thin film single- and multiple-junction devices, and newly emerged technology which include dye-sensitized cells, organic/polymer cells, highefficiency multi-junction cells based on III–V semiconductors among others. Crystalline silicon modules based on bulk wafers have been dubbed as the "first-generation" photovoltaic technology. The cost of energy generated by PV modules based on bulk-Si wafers is currently around \$3–\$4/Wp and cost reduction potential seems limited by the price of Si wafers. This cost of energy is still too high for a significant influence on energy production markets. Much of the industry is focused on the most cost efficient technologies in terms of cost per generated power. The two main strategies to bring down the cost of

TCO-Si Based Heterojunction Photovoltaic Devices 123

GZO/AZO. For reasons mentioned in the text dealing with the discussion of various TCO materials, FTO films have been widely used in solar cells to replace ITO. Alternatively, FTO coated ITO/glass substrate have been proposed to overcome the shortcomings of pure ITO. FTO is the one typically used but cost-effective SnO2-coated glass substrates on large areas (~1 m2) are still not being used as a standard substrate. On the other hand, AZO has emerged as a promising TCO material for solar cells. The AZO/glass combination has better transparency and higher conductivity than those of commercial FTO/glass substrates. Another benefit is that AZO is more resistant to hydrogen-rich plasmas used for chemical vapor deposition of thin film silicon layers as compared to FTO and ITO. The AZO films on glass for thin film silicon solar cells have a sheet resistance of about 3Ω/sq for a film thickness of ~1000 nm, a figure which degrades for thinner films. They also reported a transmittance of ~90% in the visible region of the optical spectrum for a film thickness of ~700 nm, which enhances for thinner films. These thin film silicon solar cells all have high external quantum efficiencies in the blue and green wavelength regions due to the good transmittance of the AZO films and good index matching as well as a rough interface for avoiding reflections. The highest external quantum efficiency is about 85% at a wavelength of 500 nm. However, as mentioned earlier, AZO degrades much faster than ITO and FTO in

CdTe has a direct optical band gap of about 1.5 eV and high absorption coefficient of >105 cm−1 in the visible region of the optical spectrum, which ensures the absorption of over 99% of the incident photons with energies greater than the band gap by a CdTe layer of few micrometers in thickness. CdTe solar cells are usually fabricated in the superstrate configuration, i.e., starting at the front of the cell and proceeding to the back, as described above for the Si solar cells. CdTe is of naturally p-type conductivity due to Cd vacancies. Separation of the photo-generated carriers is performed via a CdTe/CdS p-n heterojunction. CdS is an n-type material because of native defects, and has a band gap Eg~2.4 eV, which causes light absorption in the blue wavelength range which is undesirable. For this reason, the CdS layer is made very thin and is commonly referred to as a "window layer", emphasizing that photons should pass through it to be absorbed in the CdTe "absorber layer". The basic traditional module of CdTe solar cell is composed of a stack of 'Metal/CdTe/CdS/TCO/glass'. The fabrication begins with the deposition of a TCO layer onto the planar soda lime glass sheet followed by the deposition of the CdS window layer and the CdTe light absorber layer, ~ 5 μm in thickness. Efficiencies of up to 16.5% have been achieved with small-area laboratory cells, while the best commercial modules are presently 10%–11% efficient. The thin CdS window layer poses a problem shared by both CdTe and CIS-based thin film modules, which will be discussed in the next section. Since this layer should be very thin (50–80 nm in thickness), pinholes in CdS provide a direct contact between TCO and the CdS absorber layer, creating short circuits and reducing dramatically the efficiency. This problem is especially severe for CdTe cells, because sulfur readily diffuses into the CdTe layer during post-growth annealing further decreasing the CdS layer

To mitigate this issue, thin buffer layers made of highly resistive transparent oxides are incorporated between the TCO contact and the CdS window. SnO2 layers are commonly used as such buffers, although ZnSnOx films also have been proposed. The exact role of the

dampheat environment.

thickness.

**7.2 CdTe thin film solar cells** 

photovoltaic electricity are increasing the efficiency of the cells and decreasing their cost per unit area. Thin film devices (also referred to as second generation of solar cells) consume less material than the bulk-Si cells and, as a result, are less expensive. The market share of the thin film solar cells is continuously growing and has reached some 15% in year 2010, while the other 85% is silicon modules based on bulk wafers. Alternative approaches also focused on reducing energy price are devices based on polymers and dyes as the absorber materials, which include a wide variety of novel concepts. These cells are currently less efficient than the semiconductor-based devices, but are attractive due to simplicity and low cost of fabrication.

TCO are utilized as transparent electrodes in many types of thin film solar cells, such as a-Si thin film solar cells, CdTe thin film solar cells, and CIGS thin film solar cells. It should be mentioned that, for photovoltaic applications, a trade-off between the sheet resistance of a TCO layer and its optical transparency should be made. As mentioned above, to reduce unwanted free carrier absorption in the IR range, the carrier concentration in TCO should be as low as possible, while the carrier mobility should be as high as possible to obtain sufficiently high conductivity. Therefore, achieving TCO films with high carrier mobility is crucial for solar cell applications.

#### **7.1 Si thin film solar cells**

In addition to the well-established Si technology and non-toxic nature and abundance of Si, the advantage of thin film silicon solar cells is that they require lower amount of Si as compared to the devices based on bulk wafers and therefore are less expensive. Several different photovoltaic technologies based on Si thin films have been proposed and implemented: hydrogenated amorphous Si (a-Si:H) with quasi-direct band gap of 1.8 eV, hydrogenated microcrystalline Si (μc-Si:H) with indirect band gap of 1.1 eV, their combination (micromorph Si), and polycrystalline Si on glass (PSG) solar cells. The first three technologies rely on TCOs as front/back electrodes. This thin film p–i–n solar cell is fabricated in a so-called superstrate configuration, in which the light enters the active region through a glass substrate. In this case, the fabrication commences from the front of the cell and proceeds to its back.

First, a TCO front contact layer is deposited on a transparent glass substrate, followed by deposition of amorphous/microcrystalline Si, and a TCO/metal back contact layer. Therefore, the TCO front contact must be sufficiently robust to survive all subsequent deposition steps and post-deposition treatments. To obtain high efficiency increasing the path length of incoming light is crucial, which is achieved by light scattering at the interface between Si and TCO layers with different refractive indices, so that light is "trapped" within the Si absorber layer. The light trapping allows reduction of the thickness of the Si absorber layer which paves the way for increased device stability. Therefore, TCO layers used as transparent electrodes in the Si solar cells have a crucial impact on device performance. In addition to high transparency and high electrical conductivity, a TCO layer used as front electrode should ensure efficient scattering of the incoming light into the absorber layer and be chemically stable in hydrogen-containing plasma used for Si deposition, and act as a good nucleation layer for the growth of microcrystalline Si. The bottom TCO layer between Si and a metal contact works as an efficient back reflector as well as a diffusion barrier.

To increase light scattering, surface texturing of the front and back TCO contact layers is commonly used. As discussed above, the TCOs for practical applications are ITO, FTO and

photovoltaic electricity are increasing the efficiency of the cells and decreasing their cost per unit area. Thin film devices (also referred to as second generation of solar cells) consume less material than the bulk-Si cells and, as a result, are less expensive. The market share of the thin film solar cells is continuously growing and has reached some 15% in year 2010, while the other 85% is silicon modules based on bulk wafers. Alternative approaches also focused on reducing energy price are devices based on polymers and dyes as the absorber materials, which include a wide variety of novel concepts. These cells are currently less efficient than the semiconductor-based devices, but are attractive due to simplicity and low

TCO are utilized as transparent electrodes in many types of thin film solar cells, such as a-Si thin film solar cells, CdTe thin film solar cells, and CIGS thin film solar cells. It should be mentioned that, for photovoltaic applications, a trade-off between the sheet resistance of a TCO layer and its optical transparency should be made. As mentioned above, to reduce unwanted free carrier absorption in the IR range, the carrier concentration in TCO should be as low as possible, while the carrier mobility should be as high as possible to obtain sufficiently high conductivity. Therefore, achieving TCO films with high carrier mobility is

In addition to the well-established Si technology and non-toxic nature and abundance of Si, the advantage of thin film silicon solar cells is that they require lower amount of Si as compared to the devices based on bulk wafers and therefore are less expensive. Several different photovoltaic technologies based on Si thin films have been proposed and implemented: hydrogenated amorphous Si (a-Si:H) with quasi-direct band gap of 1.8 eV, hydrogenated microcrystalline Si (μc-Si:H) with indirect band gap of 1.1 eV, their combination (micromorph Si), and polycrystalline Si on glass (PSG) solar cells. The first three technologies rely on TCOs as front/back electrodes. This thin film p–i–n solar cell is fabricated in a so-called superstrate configuration, in which the light enters the active region through a glass substrate. In this case, the fabrication commences from the front of the cell

First, a TCO front contact layer is deposited on a transparent glass substrate, followed by deposition of amorphous/microcrystalline Si, and a TCO/metal back contact layer. Therefore, the TCO front contact must be sufficiently robust to survive all subsequent deposition steps and post-deposition treatments. To obtain high efficiency increasing the path length of incoming light is crucial, which is achieved by light scattering at the interface between Si and TCO layers with different refractive indices, so that light is "trapped" within the Si absorber layer. The light trapping allows reduction of the thickness of the Si absorber layer which paves the way for increased device stability. Therefore, TCO layers used as transparent electrodes in the Si solar cells have a crucial impact on device performance. In addition to high transparency and high electrical conductivity, a TCO layer used as front electrode should ensure efficient scattering of the incoming light into the absorber layer and be chemically stable in hydrogen-containing plasma used for Si deposition, and act as a good nucleation layer for the growth of microcrystalline Si. The bottom TCO layer between Si and a metal contact works as an efficient back reflector as well as a diffusion barrier. To increase light scattering, surface texturing of the front and back TCO contact layers is commonly used. As discussed above, the TCOs for practical applications are ITO, FTO and

cost of fabrication.

crucial for solar cell applications.

**7.1 Si thin film solar cells** 

and proceeds to its back.

GZO/AZO. For reasons mentioned in the text dealing with the discussion of various TCO materials, FTO films have been widely used in solar cells to replace ITO. Alternatively, FTO coated ITO/glass substrate have been proposed to overcome the shortcomings of pure ITO. FTO is the one typically used but cost-effective SnO2-coated glass substrates on large areas (~1 m2) are still not being used as a standard substrate. On the other hand, AZO has emerged as a promising TCO material for solar cells. The AZO/glass combination has better transparency and higher conductivity than those of commercial FTO/glass substrates. Another benefit is that AZO is more resistant to hydrogen-rich plasmas used for chemical vapor deposition of thin film silicon layers as compared to FTO and ITO. The AZO films on glass for thin film silicon solar cells have a sheet resistance of about 3Ω/sq for a film thickness of ~1000 nm, a figure which degrades for thinner films. They also reported a transmittance of ~90% in the visible region of the optical spectrum for a film thickness of ~700 nm, which enhances for thinner films. These thin film silicon solar cells all have high external quantum efficiencies in the blue and green wavelength regions due to the good transmittance of the AZO films and good index matching as well as a rough interface for avoiding reflections. The highest external quantum efficiency is about 85% at a wavelength of 500 nm. However, as mentioned earlier, AZO degrades much faster than ITO and FTO in dampheat environment.

#### **7.2 CdTe thin film solar cells**

CdTe has a direct optical band gap of about 1.5 eV and high absorption coefficient of >105 cm−1 in the visible region of the optical spectrum, which ensures the absorption of over 99% of the incident photons with energies greater than the band gap by a CdTe layer of few micrometers in thickness. CdTe solar cells are usually fabricated in the superstrate configuration, i.e., starting at the front of the cell and proceeding to the back, as described above for the Si solar cells. CdTe is of naturally p-type conductivity due to Cd vacancies. Separation of the photo-generated carriers is performed via a CdTe/CdS p-n heterojunction. CdS is an n-type material because of native defects, and has a band gap Eg~2.4 eV, which causes light absorption in the blue wavelength range which is undesirable. For this reason, the CdS layer is made very thin and is commonly referred to as a "window layer", emphasizing that photons should pass through it to be absorbed in the CdTe "absorber layer". The basic traditional module of CdTe solar cell is composed of a stack of 'Metal/CdTe/CdS/TCO/glass'. The fabrication begins with the deposition of a TCO layer onto the planar soda lime glass sheet followed by the deposition of the CdS window layer and the CdTe light absorber layer, ~ 5 μm in thickness. Efficiencies of up to 16.5% have been achieved with small-area laboratory cells, while the best commercial modules are presently 10%–11% efficient. The thin CdS window layer poses a problem shared by both CdTe and CIS-based thin film modules, which will be discussed in the next section. Since this layer should be very thin (50–80 nm in thickness), pinholes in CdS provide a direct contact between TCO and the CdS absorber layer, creating short circuits and reducing dramatically the efficiency. This problem is especially severe for CdTe cells, because sulfur readily diffuses into the CdTe layer during post-growth annealing further decreasing the CdS layer thickness.

To mitigate this issue, thin buffer layers made of highly resistive transparent oxides are incorporated between the TCO contact and the CdS window. SnO2 layers are commonly used as such buffers, although ZnSnOx films also have been proposed. The exact role of the

TCO-Si Based Heterojunction Photovoltaic Devices 125

quality of these materials substantially affects the required thickness of the absorber layers in terms of providing the absorption of an optimal amount of irradiation. Depending on the application, devices are fabricated in either a ''substrate'' or a ''superstrate'' configuration. The superstrate configuration is based on TCO-coated transparent glass substrates, and the layers are deposited in a reversed sequence, from the top (front) to the bottom (back). The deposition starts with a contact window layer of a photodiode and ends with a back

In the superstrate configuration, it is important for the TCO as substrate material to be not only electrically conductive and optically transparent, but also be chemically stable during solar-cell material deposition. The superstrate design is particularly suited for building integrated solar cells in which a glass substrate can be used as an architectural element. In the case of the substrate configuration, solar cells are fabricated from the back to the front, and the deposition starts from the back reflector and is finished with a TCO layer. For some specific applications, the use of lightweight, unbreakable substrates, such as stainless steel,

**8. A novel violet and blue enhanced SINP silicon photovoltaic device** 

Violet and blue enhanced semiconductor photovoltaic devices are required for various applications such as optoelectronic devices for communication, solar cell, aerospace, spectroscopic, and radiometric measurements. Silicon photodetector are sensitive from infrared to visible light but have poor responsivity in the short wavelength region. Since the absorption coefficient of crystal Si is very high for shorter wavelengths in the violet region and is small for longer wavelengths. The heavily doped emitter may contain a dead layer near the surface resulting in poor quantum efficiency of the photoelectric device under short

In order to improve the responsivity of silicon photodiode at the 400-600nm, a novel ITO/SiO2/np Si SINP violet and blue enhanced photovoltaic device (SINP is the abbreviation of semiconductor/insulator/np structure) was successfully fabricated using thermal diffusion of phosphorus for shallow junction, a very thin silicon dioxide and ITO film as an antireflection/passivation layer. The schematic and bandgap structure of the novel SINP photovoltaic device are whown here (Fig.1 and Fig.2). The very thin SiO2 film

reflector. Light enters the cell through the glass substrate.

polyimide or PET (polyethylene terephtalate) is advantageous.

Fig. 1. Schematic of the novel SINP photovoltaic device.

**8.1 Introduction** 

wavelength region.

buffer layers is not fully understood, whether it simply prevents short circuits by introducing resistance or also changes the interfacial energetics by introducing additional barriers, and optimization of this interface is a critical need. TCO materials typically used in CdTe solar cells are ITO and FTO. Reports for AZO in CdTe cells are very few. The use of ZnO-based TCOs in CdTe solar sells of superstrate configuration is hampered by its thermal instability and chemical reaction with CdS at high temperatures (550–650°C) typically used for CdTe solar cells fabrication. To resolve this problem, Gupta and Compaan applied low temperature (250°C) deposition by magnetron sputtering to fabricate superstrate configuration CdS/CdTe solar sells with AZO front contacts. These cells yielded efficiency as high as 14.0%. Bifacial CdTe solar cells make it possible to increase the device NIR transmission as the parasitic absorption and reflection losses are minimized. The highest efficiency of 14% was achieved from a CdTe cell with an FTO contact layer. The device performance depends strongly on the interaction between the TCO and CdS films. Later, the same group has noted a substantial In diffusion from ITO to the CdS/CdTe photodiode, which can be prevented by the use of undoped SnO2 or ZnO buffers. Application of TCO as the back contact also allows fabrication of bifacial CdTe cells or tandem cells, which opens a variety of new applications of CdTe solar cells.

#### **7.3 CIGS thin film solar cells**

Copper indium diselenide (CuInSe2 or CIS) is a direct-bandgap semiconductor with a chalcopyrite structure and belongs to a group of miscible ternary I–III–VI2 compounds with direct optical bandgaps ranging from 1 to 3.5 eV. The miscibility of ternary compounds, that is the ability to mix in all proportions, enables quaternary alloys to be deposited with any bandgap in this range. A large light absorption coefficient of >105 cm−1 at photon energies greater than a bandgap allows a relatively thin (few μm in thickness) layer to be used as the light absorber. The alloy systems with optical bandgaps appropriate for solar cells include Cu(InGa)Se2, CuIn(SeS)2, Cu(InAl)Se2, and Cu(InGa)S2. Copper indium–gallium diselenide Cu(InGa)Se2 (or CIGS) has been found to be the most successful absorber layer among chalcopyrite compounds investigated to date. The bandgap is ~1.0 eV for CuInSe2 and increases towards the optimum value for photovoltaic solar energy conversion when gallium is added to produce Cu(In, Ga)Se2. An energy bandgap of 1.25–1.3 eV corresponds to the maximum gap achievable without loss of efficiency. Further increase in the Ga fraction reduces the formation energies of point defects, primary, copper vacancies which makes them more likely to form. Also, a further increase in gallium content makes the absorber layers too highly resistive to be used in solar cells. Therefore, most CIGS devices are produced with an energy bandgap below 1.3 eV, which limits their VOC at ~700 meV. Note that both CIS- and CIGS-based devices are usually dubbed as the CIS technology in the literature. The CIS technology provides the highest performance in the laboratory among all thin-film solar cells, with confirmed power conversion efficiencies of up to 20.1% for small (0.5 cm2) cells fabricated by the Zentrum fuer Sonnenenrgie-und-Wasserstoff–Forschung and measured at the Fraunhofer Institute for Solar Energy Systems, and many companies around the world are developing a variety of manufacturing approaches aimed at low-cost, high-yield, large-area devices which would maintain laboratory-level efficiencies.

Similarly, TCO layers are generally used for the front contact, whereas a reflective contact material (Ag, frequently in combination with a TCO interlayer, is the most popular one) is needed on the back surface to enhance the light trapping in absorber layers. The optical quality of these materials substantially affects the required thickness of the absorber layers in terms of providing the absorption of an optimal amount of irradiation. Depending on the application, devices are fabricated in either a ''substrate'' or a ''superstrate'' configuration. The superstrate configuration is based on TCO-coated transparent glass substrates, and the layers are deposited in a reversed sequence, from the top (front) to the bottom (back). The deposition starts with a contact window layer of a photodiode and ends with a back reflector. Light enters the cell through the glass substrate.

In the superstrate configuration, it is important for the TCO as substrate material to be not only electrically conductive and optically transparent, but also be chemically stable during solar-cell material deposition. The superstrate design is particularly suited for building integrated solar cells in which a glass substrate can be used as an architectural element. In the case of the substrate configuration, solar cells are fabricated from the back to the front, and the deposition starts from the back reflector and is finished with a TCO layer. For some specific applications, the use of lightweight, unbreakable substrates, such as stainless steel, polyimide or PET (polyethylene terephtalate) is advantageous.
