**4. Factory panel production**

342 Solar Cells – Thin-Film Technologies

The front and back TCO layers are at the same time electrodes that collect photogenerated carriers. As a semiconductor, the optical transparency and the electrical conductivity are closely related to the band gap structure of the TCO. The short-wavelength cutoff of the transmission spectrum corresponds to the oxide band gap, whereas the long-wavelength transmission edge corresponds to the free carrier plasma resonance frequency. On the other hand, electron conduction in TCO is achieved by degenerate doping that increases the free carrier density and moves the Fermi level into the conduction band. High carrier density and carrier mobility are thus required for TCO layers. There is, however, a tradeoff between high optical transmittance and low electrical resistance. Increasing electron carrier density decreases resistivity but also increases the plasma oscillation frequency of free carriers, thus shifting the IR absorption edge towards the visible. The transmission window is thus

TCO type ITO ZnO:Al SnO2:F

Resistivity (Ω·cm) 1-5×10-4 3-8×10-4 6-10×10-4

Surface roughness Flat Excellent Excellent Plasma stability Low Excellent Good Relative cost High Middle Low

low-pressure chemical vapor deposition. APCVD, atmosphere-pressure CVD.

Table 1. Different TCOs employed in Si thin film solar cells. RF, radio frequency. LPCVD,

The most-widely used TCOs in Si thin film solar cells are doped SnO2 (i.e., SnO2:F) and ZnO (i.e., ZnO:Al) due to their temperature and chemical stabilities. Compared to the more conductive alternative indium-tin-oxide (ITO), they offer a much lower cost by avoiding the use of the costly In. At the same time, surface roughness induced by the crystalline texture of SnO2 and ZnO is widely applied for increasing the optical absorption. These three typical

The reflector layer on top of the back TCO can be Ag, Al, or white paint in a "superstrate" cell, and is the metal foil itself or another Ag/Al coating on the foil in a 'substrate' type cell. Ag is typically used in laboratory research work, while Al is more often used in mass production modules due to its lower cost and better properties in removing module shunts. Back contacts of Oerlikon's thin film panels, on the other hand, use proprietary white paint as the reflector (Meier et al. 2005). The white paint can be rolled on or screen printed directly onto the TCO. At the same time, it offers the following advantages (Berger et al. 2007): 1) high optical reflectance over a broad wavelength band, 2) optimal light scattering pattern which is generally beneficial for solar cells because this maximizes the fraction of photons that are trapped inside the solar cell due to total internal reflection at both cell surfaces, 3) pigmented materials have the potential of low cost. In certain instances, the white paint is a

LPCVD

APCVD, spray pyrolysis

(350-1000 nm) 95% 90-95% 90%

Work function (eV) 4.7 4.5 4.8 Band gap (eV) ~3.7 ~3.4 4.1-4.3

Deposition methods RF sputtering RF sputtering,

narrowed as a result of improved conductivity.

Optical transmission

TCOs are compared in Table 1.

better surface reflector than Al, or TCO/Al reflector.

As previously discussed (c.f., Section 2), manufacturing of the solar panels directly determines the cost of modules, which takes 40-66% of the overall PV system cost. This section focuses on the factory panel production, and addresses various methods of panel efficiency improvement and cost reduction.

Fig. 5. Flow diagram of typical thin film solar panel production, comprising of both the front end of line (FEOL) and back end of line (BEOL) technologies and processes (Bhan et al. 2010).

Development in the a-Si and µc-Si thin film process technology combined with the booming PV market resulted in the fast expansion of a-Si/µc-Si based solar panel manufacturing after 2007. This industry largely benefits from the lab demonstration of thin film solar cells on small size substrates, as well as the large-area thin film deposition techniques developed for thin-film transistor liquid crystal display (TFT-LCD) industry. The growth of high-quality Si thin films for PV applications shares many of the skill sets required for growing Si TFT films, and using similar large-area thin film deposition chambers (Yang et al. 2007). In fact, both thin film solar "turnkey" equipment providers, Oerlikon and Applied Materials, have been manufacturing large-scale TFT-LCD deposition systems for years before becoming thin film solar equipment providers.

2. Secondary gas phase reaction. The reaction between molecules, ions and radicals (product of the previous step) generates reactive species and eventually large Si-H clusters, which are also called dust or power particles. The neutral species diffuse to the substrate, while the positive ions bombard the growing film and the negative ions are

3. Film deposition. The radicals diffusing to the substrate interact with the substrate surface in various ways, like radical diffusion, chemical bonding, hydrogen sticking to

4. Formation of a-Si film. The actively growing film then releases hydrogen and relaxes

RF power supply

Slit valve Observation

To pump

Fig. 7. Challenges met with high deposition rate (RD) for PECVD grown µc-Si films. (a) Efficiencies of the optimal µc-Si single junction solar cell at different very-high frequency radio-frequency power (PVHF). (b) Corresponding deposition rates RD and SiH4 dilution

Gas feed

Susceptor

constrained within the plasma.

into the Si network.

the surface or desorption from the surface.

window

Fig. 6. Schematics of a PECVD process chamber.

concentration (SC) (Mai et al. 2005).

Glass substrate

Diffuser

The general process flow of thin film panel fabrication is shown in the diagram of Fig. 5. This is a typical Applied Materials Sunfab process configuration (Bhan et al. 2010), though that of Oerlikon (Sun et al. 2009) and other 'superstrate' type panel makers are similar. The entire process is divided into the FEOL steps where the front glass substrate is deposited with active layers, and BEOL steps where the module is encapsulated. The FEOL mainly involves several film deposition and patterning steps, including the growth of Si p-i-n junctions by CVD and the growth of TCOs and metal layers by PVD. Laser scribing steps are used in between film depositions to form monolithically interconnected cells across the entire substrate. In the BEOL steps, the front glass with blanket film deposited during FEOL is cut, shaped and encapsulated to make solar panels. Optical and electrical inspections are taken upon finishing of FEOL for quality control purposes. Process of the 'substrate' type module shares many of the FEOL and BEOL steps, and the differences are discussed later in Section 4.3.

#### **4.1 Front end of line (FEOL) process**

The first step of module deposition involves loading of substrates in FEOL. Either float glass or TCO-coated glass is used, though an extra TCO growth step is required for the former case (Kroll et al. 2007; Sun et al. 2009). To prevent the glass from chipping and cracking during the following thermal cycle steps, as the film deposition steps require substrate temperatures ranging from 150 to 250 C, the glass edge is seamed and reinforced. Then the glass goes through washing and drying steps to thoroughly remove debris, particles and organic contaminants. Laser-scribing step P1 follows to form isolated TCO contact strips. After a second washing step, the TCO glass is loaded into the film deposition chambers for the growth of PV active layers.

#### **4.1.1 Growth of a-Si/µc-Si layers**

The PV active Si p-i-n film stack is normally grown by CVD from gaseous precursors. Several CVD deposition techniques have been developed for the deposition of a-Si/µc-Si layers, including plasma-enhanced CVD (PECVD) (Schropp and Zeman 1998), remote plasma enhanced CVD (RPECVD) (Kessels et al. 2001), and hotwire CVD (Schroeder 2003). Though efficient lab-size solar cells are made with various CVD techniques, PECVD is prevailingly used for current industrial, high-throughput thin film PV module fabrication, as it possesses advantages like high deposition rate, in-situ chamber cleaning, good control over film quality, and requires lowest substrate temperature.

A typical PECVD chamber is structured like the schematics in Fig. 6. The substrate is supported by a susceptor, directly facing a gas diffuser. Process gases (SiH4, H2 and dopant gases) are fed into the chamber and dispersed by the diffuser. The diffuser and the susceptor are charged at opposite radio frequency (RF) voltages, thus exciting plasma within the chamber. Commonly used RF plasma is excited at 13.56 MHZ or 40 MHz, while higher RF frequencies are also used (Nishimiya et al. 2008). Higher RF frequencies are reported to deposit Si film faster due to higher plasmornic excitation energy. On the other hand, higher frequency means shorter RF wavelength, potentially forming standing wave inside the chamber that can cause non-uniform plasma distribution and reduce the a-Si/µc-Si film uniformity.

Using silane (SiH4) as the precursor gas, the deposition of a-Si/µc-Si films can be described as a four-step process (Schropp and Zeman 1998):

1. Primary gas phase SiH4 decomposition. The plasma excites and decomposes the SiH4 molecules into neutral radicals and molecules, positively and negatively charged ions, and electrons.

The general process flow of thin film panel fabrication is shown in the diagram of Fig. 5. This is a typical Applied Materials Sunfab process configuration (Bhan et al. 2010), though that of Oerlikon (Sun et al. 2009) and other 'superstrate' type panel makers are similar. The entire process is divided into the FEOL steps where the front glass substrate is deposited with active layers, and BEOL steps where the module is encapsulated. The FEOL mainly involves several film deposition and patterning steps, including the growth of Si p-i-n junctions by CVD and the growth of TCOs and metal layers by PVD. Laser scribing steps are used in between film depositions to form monolithically interconnected cells across the entire substrate. In the BEOL steps, the front glass with blanket film deposited during FEOL is cut, shaped and encapsulated to make solar panels. Optical and electrical inspections are taken upon finishing of FEOL for quality control purposes. Process of the 'substrate' type module shares many of the FEOL and

The first step of module deposition involves loading of substrates in FEOL. Either float glass or TCO-coated glass is used, though an extra TCO growth step is required for the former case (Kroll et al. 2007; Sun et al. 2009). To prevent the glass from chipping and cracking during the following thermal cycle steps, as the film deposition steps require substrate temperatures ranging from 150 to 250 C, the glass edge is seamed and reinforced. Then the glass goes through washing and drying steps to thoroughly remove debris, particles and organic contaminants. Laser-scribing step P1 follows to form isolated TCO contact strips. After a second washing step, the TCO glass is loaded into the film deposition chambers for

The PV active Si p-i-n film stack is normally grown by CVD from gaseous precursors. Several CVD deposition techniques have been developed for the deposition of a-Si/µc-Si layers, including plasma-enhanced CVD (PECVD) (Schropp and Zeman 1998), remote plasma enhanced CVD (RPECVD) (Kessels et al. 2001), and hotwire CVD (Schroeder 2003). Though efficient lab-size solar cells are made with various CVD techniques, PECVD is prevailingly used for current industrial, high-throughput thin film PV module fabrication, as it possesses advantages like high deposition rate, in-situ chamber cleaning, good control

A typical PECVD chamber is structured like the schematics in Fig. 6. The substrate is supported by a susceptor, directly facing a gas diffuser. Process gases (SiH4, H2 and dopant gases) are fed into the chamber and dispersed by the diffuser. The diffuser and the susceptor are charged at opposite radio frequency (RF) voltages, thus exciting plasma within the chamber. Commonly used RF plasma is excited at 13.56 MHZ or 40 MHz, while higher RF frequencies are also used (Nishimiya et al. 2008). Higher RF frequencies are reported to deposit Si film faster due to higher plasmornic excitation energy. On the other hand, higher frequency means shorter RF wavelength, potentially forming standing wave inside the chamber that can

Using silane (SiH4) as the precursor gas, the deposition of a-Si/µc-Si films can be described

1. Primary gas phase SiH4 decomposition. The plasma excites and decomposes the SiH4 molecules into neutral radicals and molecules, positively and negatively charged ions,

cause non-uniform plasma distribution and reduce the a-Si/µc-Si film uniformity.

BEOL steps, and the differences are discussed later in Section 4.3.

over film quality, and requires lowest substrate temperature.

as a four-step process (Schropp and Zeman 1998):

and electrons.

**4.1 Front end of line (FEOL) process** 

the growth of PV active layers.

**4.1.1 Growth of a-Si/µc-Si layers** 


Fig. 6. Schematics of a PECVD process chamber.

Fig. 7. Challenges met with high deposition rate (RD) for PECVD grown µc-Si films. (a) Efficiencies of the optimal µc-Si single junction solar cell at different very-high frequency radio-frequency power (PVHF). (b) Corresponding deposition rates RD and SiH4 dilution concentration (SC) (Mai et al. 2005).

area. Since various PECVD parameters directly affect the growth rate of a-Si / µc-Si film, the non-uniform distribution of these growth parameters induces local film thickness variation. For typical p-i-n type a-Si or µc-Si cells, the open circuit voltage (*V*OC) and fill factor (*FF)* decrease upon increasing i-layer thickness, while *J*SC increases as a result of the higher absorption in the thicker cells (Klein et al. 2002). Other than thickness, the RF power distribution affects the crystallinity of the as-grown µc-Si, which in turn changes the solar cell performance. For example, slight deviation of RF intensity resulted in unbalanced µc-Si crystallization in a Gen 8.5 PECVD chamber, as shown by the smaller fraction of crystallinity (FC) on the left side of chamber before adjusting the RF power supply feed (Fig. 8a), though such deviation could be too small to affect the a-Si and µc-Si film thicknesses (Yang et al. 2009). Small size a-Si/µc-Si tandem junction solar cells cut from the solar panels at corresponding locations, had non-uniform performance distribution. Sample cells from the low RF intensity side (left) had smaller short circuit current density (*J*SC) (Fig. 8b) and higher *V*OC (Fig. 8c) than those from the other side, while *FF* had a uniform distribution despite the RF influence (Fig. 8d). After modifying the RF feed of the PECVD chamber, balanced FC distribution was obtained, and the sample cells showed uniform distribution of *J*SC, *V*OC and *FF*. It is thus important to keep the uniform distribution of all process parameters in large scale process, and special attention must be paid to the RF power distribution across the whole chamber.

Fig. 8. µc-Si film uniformity and its impact to solar cell performance (a) Crystalline fraction (fc) change along substrate diagonal with original and modified RF feed. (b) JSC, (c) VOC, and (d) FF profiles along substrate diagonals. The two substrates were grown before and after

modifying the RF feed location.

The overall film deposition is a complex process of gas and surface reactions, and is controlled by many deposition parameters, including gas composition, flow rate, chamber pressure, RF power density, RF frequency, substrate temperature, and the chamber size and geometry. Extensive studies are carried out to study the influence of those controlling parameters on the a-Si/µc-Si film properties, and are summarized in review papers (Luft and Tsuo 1993; Bruno et al. 1995). Perusing high solar energy conversion efficiency requires high quality, PECVD grown a-Si/µc-Si films, like high optical absorption, low dangling bond density, wide band gap for optical transmission of the p- and n- window layers (Schropp 2006).

Transferring the lab-developed process to solar panel production, however, has its own challenges. The PECVD steps take a significant portion of the cost in energy consumption and equipment depreciation (c.f., Fig. 3), thus substantial changes must be made to the labdeveloped PECVD processes in panel manufacturing to fit the goal of lowering the panel cost. While maintaining the optimal panel performance, the widely adapted strategies are high rate of deposition (*R*D) and large substrate size for high-throughput panel growth.

As previously discussed in Section 2, fast film deposition can lower the energy consumption, facilitating the throughput and process efficiency, thus effectively lower the cost of solar modules. The major obstacle to throughput increase is the µc-Si deposition, which has to be thick (> 1.5 µm) as limited by the finite absorption coefficient, thus requiring appreciably long deposition time. In fact, deposition of the µc-Si bottom cell in a-Si/µc-Si tandem junction solar panels takes the longest process time in the Oerlikon and Applied Materials process lines. To shorten the µc-Si deposition time and improve the overall throughput, research has been focused on increasing the deposition rate of µc-Si films.

Generally, increasing the density of SiH radicals promotes the growth of a-Si/µc-Si, thus increasing SiH4 flow rate, applying higher RF power density, or using a higher RF excitation frequency all lead to higher *R*D. For the simpler case of a-Si, higher RF power and higher gas flow rate result in faster film growth. As for the case of µc-Si, changing these deposition parameters at the same time affects the film crystallization in addition to increasing *R*D. Since a good performed µc-Si solar cell needs to keep at the transition from microcrystalline to amorphous growth, increasing the film deposition rate should not be compromised by the film crystallinity. It is observed that increase of RF power requires higher SiH4 concentration to keep the same crystallinity, which at the same time leads to higher deposition rate. In one example, Fig. 7 compared the high *R*D achieved with increasing very high frequency (VHF) 94.7 MHz RF power (Mai et al. 2005). To maintain the maximum PV efficiency (Fig. 7a) at different VHF power (*P*VHF), the silence concentration (SC) in the mixed gas has to increase with *P*VHF (Fig. 7b). At small *P*VHF, *R*D increases linearly with *P*VHF, and saturates around 1.2 nm/s when *P*VHF is > 60 W. The solar cell efficiency, stays constant for the same region since the µc-Si crystallinity remains unchanged. Further increase of *P*VHF doesn't lead to higher RD, and the optimal cell efficiency drops at this region. It is important to note that higher deposition rate greatly improves the throughput of the panel manufacturing, and lowers the panel cost.

Since the merit of solar cell, \$/Wp, is inversely proportional to the total solar module production (Hegedus and Luque 2003), growing solar cells over large-area substrates is one of the most efficient ways of lowering solar cell cost. For constant direct materials and labor cost, growing films over larger area substrates effectively lowers the module cost per unit area. The major challenge when scaling up the substrate is to maintain uniform film growth over large

The overall film deposition is a complex process of gas and surface reactions, and is controlled by many deposition parameters, including gas composition, flow rate, chamber pressure, RF power density, RF frequency, substrate temperature, and the chamber size and geometry. Extensive studies are carried out to study the influence of those controlling parameters on the a-Si/µc-Si film properties, and are summarized in review papers (Luft and Tsuo 1993; Bruno et al. 1995). Perusing high solar energy conversion efficiency requires high quality, PECVD grown a-Si/µc-Si films, like high optical absorption, low dangling bond density, wide band gap for optical transmission of the p- and n- window layers

Transferring the lab-developed process to solar panel production, however, has its own challenges. The PECVD steps take a significant portion of the cost in energy consumption and equipment depreciation (c.f., Fig. 3), thus substantial changes must be made to the labdeveloped PECVD processes in panel manufacturing to fit the goal of lowering the panel cost. While maintaining the optimal panel performance, the widely adapted strategies are high rate of deposition (*R*D) and large substrate size for high-throughput panel growth. As previously discussed in Section 2, fast film deposition can lower the energy consumption, facilitating the throughput and process efficiency, thus effectively lower the cost of solar modules. The major obstacle to throughput increase is the µc-Si deposition, which has to be thick (> 1.5 µm) as limited by the finite absorption coefficient, thus requiring appreciably long deposition time. In fact, deposition of the µc-Si bottom cell in a-Si/µc-Si tandem junction solar panels takes the longest process time in the Oerlikon and Applied Materials process lines. To shorten the µc-Si deposition time and improve the overall throughput, research has been focused on increasing the deposition rate of µc-Si films. Generally, increasing the density of SiH radicals promotes the growth of a-Si/µc-Si, thus increasing SiH4 flow rate, applying higher RF power density, or using a higher RF excitation frequency all lead to higher *R*D. For the simpler case of a-Si, higher RF power and higher gas flow rate result in faster film growth. As for the case of µc-Si, changing these deposition parameters at the same time affects the film crystallization in addition to increasing *R*D. Since a good performed µc-Si solar cell needs to keep at the transition from microcrystalline to amorphous growth, increasing the film deposition rate should not be compromised by the film crystallinity. It is observed that increase of RF power requires higher SiH4 concentration to keep the same crystallinity, which at the same time leads to higher deposition rate. In one example, Fig. 7 compared the high *R*D achieved with increasing very high frequency (VHF) 94.7 MHz RF power (Mai et al. 2005). To maintain the maximum PV efficiency (Fig. 7a) at different VHF power (*P*VHF), the silence concentration (SC) in the mixed gas has to increase with *P*VHF (Fig. 7b). At small *P*VHF, *R*D increases linearly with *P*VHF,

and saturates around 1.2 nm/s when *P*VHF is > 60 W. The solar cell efficiency,

the panel manufacturing, and lowers the panel cost.

constant for the same region since the µc-Si crystallinity remains unchanged. Further increase of *P*VHF doesn't lead to higher RD, and the optimal cell efficiency drops at this region. It is important to note that higher deposition rate greatly improves the throughput of

Since the merit of solar cell, \$/Wp, is inversely proportional to the total solar module production (Hegedus and Luque 2003), growing solar cells over large-area substrates is one of the most efficient ways of lowering solar cell cost. For constant direct materials and labor cost, growing films over larger area substrates effectively lowers the module cost per unit area. The major challenge when scaling up the substrate is to maintain uniform film growth over large

stays

(Schropp 2006).

area. Since various PECVD parameters directly affect the growth rate of a-Si / µc-Si film, the non-uniform distribution of these growth parameters induces local film thickness variation. For typical p-i-n type a-Si or µc-Si cells, the open circuit voltage (*V*OC) and fill factor (*FF)* decrease upon increasing i-layer thickness, while *J*SC increases as a result of the higher absorption in the thicker cells (Klein et al. 2002). Other than thickness, the RF power distribution affects the crystallinity of the as-grown µc-Si, which in turn changes the solar cell performance. For example, slight deviation of RF intensity resulted in unbalanced µc-Si crystallization in a Gen 8.5 PECVD chamber, as shown by the smaller fraction of crystallinity (FC) on the left side of chamber before adjusting the RF power supply feed (Fig. 8a), though such deviation could be too small to affect the a-Si and µc-Si film thicknesses (Yang et al. 2009). Small size a-Si/µc-Si tandem junction solar cells cut from the solar panels at corresponding locations, had non-uniform performance distribution. Sample cells from the low RF intensity side (left) had smaller short circuit current density (*J*SC) (Fig. 8b) and higher *V*OC (Fig. 8c) than those from the other side, while *FF* had a uniform distribution despite the RF influence (Fig. 8d). After modifying the RF feed of the PECVD chamber, balanced FC distribution was obtained, and the sample cells showed uniform distribution of *J*SC, *V*OC and *FF*. It is thus important to keep the uniform distribution of all process parameters in large scale process, and special attention must be paid to the RF power distribution across the whole chamber.

Fig. 8. µc-Si film uniformity and its impact to solar cell performance (a) Crystalline fraction (fc) change along substrate diagonal with original and modified RF feed. (b) JSC, (c) VOC, and (d) FF profiles along substrate diagonals. The two substrates were grown before and after modifying the RF feed location.

2. Decomposition of H from a-Si:H. The absorbed heat induces the decomposition of a-Si:H, and releases hydrogen at a temperature of > 600 C (Fig. 9b). In fact, the local

3. Destruction of the PV layers and back contact. The gaseous H2 quickly expends its volume and pressure under the high temperature. The pressure of the H2 gas can amount to >1×107 Pa, inducing enormous shear stress on the layers above the heated zone. In one estimation, applying a 532 nm, 12 kHz and 9.5 × 106 W/cm2 laser beam on a-Si single junction module created shear stress of 3.9×108 Pa, enough to break the layers on top of the heating zone, among which the most ductile Ag layer has a shear

4. Formation of heat affected zone (HAZ). Along with the H2 volume expansion, the film cracks quickly followed by blasting off, effectively removing the a-Si/µc-Si layers and the back contact layers above the local, heated zone. The laser heating also damages the film around the removed region, creating a HAZ with high density of defects and poor electrical properties (Fig. 9d). By using high-frequency pulsed laser, the HAZ is limited

It is important to note that the laser scribing removal is not a true thermal process but the mechanical blasting off of the film. By applying different wavelengths of lasers, the laser energy is absorbed by different layers, thus selectively removes those layers without

> Heat affected zone

Fig. 9. Laser-scribing mechanism. (a) absorption of laser beam incidents through the glass, (b) decomposition of H from a-Si:H, (c) destruction of the photovoltaic layer and back electrode, (d) film blasted off and formation of heat affected zone (Shinohara et al. 2006).

H2

AZO Back reflector

a-Si TCO Glass

Laser beam

(a) (b)

Cracks

(c) (d)

temperature in the film can be heated up to 700 C by the laser.

building up in the film.

strength of 107 – 108 Pa (Fig. 9c).

to less than a few tens of nm wide.

affecting other, underlying layers.

at the glass-side of the film (Fig. 9a). Typical laser energy is >1×106 W/cm2, and calculation shows more than 80% of that energy is absorbed and converted to heat
