Hole Selective Contacts: A Brief Overview

Abstract

Carrier selective solar cell structure has allured curiosity of photovoltaic researchers due to the use of wide band gap transition metal oxide (TMO). Distinctive p/n-type character, broad range of work functions (2 to 7 eV) and risk free fabrication of TMO has evolved new concept of heterojunction intrinsic thin layer (HIT) solar cell employing carrier selective layers such as $MoO_x$, $WO_x$, $V_2O_5$ and $TiO_2$ replacing the doped a-Si layers on either front side or back side. The p/n-doped hydrogenated amorphous silicon (a-Si:H) layers are deposited by Plasma-Enhanced Chemical Vapor Deposition (PECVD), which includes the flammable and toxic boron/phosphorous gas precursors. Due to this, carrier selective TMO is gaining popularity as analternative risk-free material in place of conventional a-Si:H. In this work hole selective materials such as $MoO_x$, $WO_x$ and $V_2O_5$has been investigated. Recently $MoO_x$, $WO_x$ & $V_2O_5$ hetero-structures showed conversion efficiency of 22.5%, 12.6% & 15.7% respectively at temperature below $200^{\circ}C$. In this work a concise review on few important aspects of the hole selective material solar cell such as historical developments, device structure, fabrication, factors effecting cell performance and dependency on temperature has been reported.

Subscript

HIT : Heterojunction intrinsic thin layer solar cell

EHP : Electron Hole Pair

BSF : Back Surface Field

TMO : Transition Metal Oxides

1. Introduction

Amorphous silicon heterojunction has proved to be a cost effective alternative to high temperature crystalline silicon (C-Si) technology. Even though the conventional C-Si solar cells possess appreciative photovoltaic (PV) conversion efficiency, in spite of that, this technology demands excessive fabrication cost, higher thermal budget and availability of admirable Si.(1) HIT attracts researchers and gain popularity due to its characteristics like elevated power conversion efficiency (PCE) and lower thermal budget²⁾. Evolution of HIT cells has been started in the year 1991^1-2). An inauguration of a thin intrinsic a-Si: H layer sandwiching in between c-Si from both front and back end exhibits an improvement in device performance. In hetero-structures, a-Si:H(i) layers has been used for surface passivation, which also acts as window layer for both front and backside of n-type crystalline silicon (c-Si) wafer. A p-type a-Si layer serves as emitter while the n+ a-Si layer acts as a back surface field (BSF). Heterostructures necessarily encourage tunneling across the heterointerfaces for better performance^2-4). HIT does not involve high process temperature or removal of Boron rich layer or segregation of junction for better performance. Current survey reports HIT with efficiency 26.3%^3-5).

Recently, a further modification on heterostructures with the replacement of doped emitter layer with carrier selective layers has been reported⁶⁾. According to the literature this modified structure does not involve any poisonous gas during its fabrication unlike conventional heterostructures at the same time reduce absorption losses due to high band gap^6-7). Different theoretical and experimental study has been implemented on such HIT structures for characterization and superb efficiency.

Transition metal oxides (TMO) has evolved as a revolutionary carrier selective material to replace amorphous silicon layers in HIT structures⁸⁾. TMOs can be categorized into holes as well as electron selective material. MoOx, V2O5, and WOx etc are some examples of hole selective material. These hole selective material replace p-type emitter in case of hole selective cells due to which it has been misidentified as p-type^9-11). Whereas, they show n-type character due to intrinsic oxygen vacancies in the atomic structure. The alteration in stoichiometry ‘x’, varies the property of MoO_x from insulators (MoO₃) to metal like a conductor (MoO₂). The energy gap (Eg) lies in between O₂p- and metal d-bands and thus we can say the d state decides the conductivity of a material.

There is an alternative hole and electron selective technique rather than TMOs i.e. organic thin film photovoltaic devices. It consist of material having privileged hole or electron selective contacts which uses low temperature^11,12). Besides this, it also provide segregation of carriers along with low recombination rate and lower contact resistivity. These Organic semiconductor materials like P3HT and PEDOT: PSS have depicted hole injection and extraction properties for buffer layers in organic photovoltaic devices^12,13). The main issue with this organic material is that it suffers from chemical instability due to hygroscopic profile which lead to device destruction. Therefore TMO is a better alternative. Molybdenum oxide (MoOx, x<3) , WOx (x<3) and V2O5 has emerged as rebellious materials to serve as hole selective stacks¹⁴⁾.In comparison to a-Si: H(p) & μ c-SiOx:H(p) layer, TMO reveals high transmittance, high bandgap, high affinity high work-function and low absorption coefficient¹⁵⁾. Fig. 1 represented the schematics of HIT solar cells with variation in emitter from conventional a-Si: H(p) to different TMO (MoOx and WOx) structures considering n-type C-Si as base wafer.

 

Fig. 1. Schematics of cell with (a) conventional HIT (b) MoOx as hole transport layer (c) WOx as hole transport layer

1.1 Historical Development of HIT solar cells

According to literature, in 1991 HIT cell exhibited 16% PCE which increased to 18% by the following year¹⁵⁾.In 1994, PCE for HIT improved to 20% that persisted till 2000 to 2003 around 21%. In 2011, it was reported that a thinner absorber layer of approximate 98 µm has performed better on HIT cells improving PCE to 23%15). Recent times HIT technology has already achieved an efficiency of 25.6%. Whereas, carrier selective cells has illustrated a record of 22.5% efficiency with MoOx as hole transport layer¹⁶⁾. This structure has high potential for improvement in efficiency as higher as HIT devices.

1.2 Device structure

n type Si wafer: Schematics of HIT structures considering n-type c-Si as base wafer (thickness: 220 μm) is presented in Fig. 1. The Base wafer is sandwiched between hydrogenated intrinsic a-Si layers of thickness 5-7 nm which acts as passivation as well as tunneling layer^17-18). Consequently, carrier selective layer is replaced by a-Si:H(p) layer of approximately 10-15 nm thickness. The BSF is fabricated with a highly doped a-Si:H(n) layer. The outermost layers both front and back is filled with ITO having an electron density of 1×1020 which will be in the order of 120 nanometers. ITO serves as an anti-reflection coating or we can say ARC¹⁸⁾. Ag is used for metal contact both in front and backside

The energy band diagram of MoOx and amorphous cells on n-type wafer is shown in Fig. 2. P-type emitter has a lower Fermi Energy level as compared to n-type BSF. Therefore diffusion has to be carried out from n zone to p zone electrons will diffuse to the zone with lower Fermi Energy level¹⁹⁾. Therefore, bands of n-type C-Si will be bended towards energy bands of MoO_x layer as exhibited in Fig. 2.

Fig. 2. Energy band diagram for silicon solar cells with hole selective (a) Standard n-type a-Si:H emitter, (b) MoOx contact

MoO_X , ITO and a-Si are having wide optical band gap. As Electron hole pair (EHP) generates at C-Si i.e absorber layer ¹⁹⁾. Carriers will move towards the contact region. As MoOx layer acts as hole extracting layer, the conduction band offset does not favor tunneling of electrons to MoOx layer whereas, the valance band offset is reduced by MoOx layer which help holes to tunnel through i–layer to MoOx layer and finally metal contact²⁰⁾ as represented in Fig. 1 and Fig. 2 describes band diagram of (a) MoOx cell (b) a-Si cell

As discussed in literature Light current voltage (LIV) data of different hole selective cells are shown in Table 1¹⁾.

Table 1. Electrical parameters of cells

p-type Si wafer: As described in literature, in case of p-wafer cells i-layer at front surface yields interface passivation. Schematics of (a) c-Si (p) / a-Si:H/ MoOx heterojunction solar cell structure and (b) c-Si(p)/a-Si:H/WOx heterojunction solar cell has been depicted in Fig. 3.

Fig. 3. Schematics of the (a) C-Si (p) / a-Si:H/ MoOx and (b) C-Si(p)/ a-Si:H/WOx heterojunction solar cell structure

Transportation of holes at the back contact, through tunneling via intrinsic and hole extracting layer is significant to reduce barrier effects and better performance21). For n-type wafer, the thicker top a-Si (i) layer results in a lower short circuit current whereas, in case of p-wafer cell front i-layer should be as thick as 50 nm (approx) so that top layer generates high electric field that results in accumulation of carriers originated by short-wavelength light²²⁾.

As reported in literatures hole extracting materials can serve as both p-type replacement and additional contact layer to assist hole extraction²³⁾. The analysis revealed if MoOx and WoO_x act as additional contact layers, the offset between ITO and amorphous silicon is reduced which enhance Voc and FF of cell as compared to conventional HIT structure.

1.3 Fabrication

The fabrication of carrier selective HIT cell involves prior removal of saw damage followed by proper pyramidal texturing of high-quality c-Si (n) wafer having resistivity 1.5 Ω cm and thickness of 200 µm24). Consequently, standard RCA cleaning has been adapted for removal of organic/inorganic contaminants succeeded by HCl/HF dipping, rinsing in DI water and drying with nitrogen gas. The a-Si:H (i) layer of thickness 8 and 6 nm was deposited in both front and rear sides of C-Si respectively via cluster PECVD at 200°C. A n+ phosphorus-doped layer of thickness 12 nm is also deposited in back side of device through PECVD. Likewise, MoOx 10 nm has been deposited on front side of device by thermal evaporation²⁵⁾. ITO is placed on front and back layers as of 100 nm and 80 nm by pulsed DC magnetron sputtering at 100°C. Silver (Ag) paste is screen printed as continuous metal on rear side and grid type on front side²⁶⁾. Lastly samples are cured at 160°C for 30 min in industrial belt furnace.

1.4 Material requirement

Eventually, recent investigations have reported carrier selective materials utilizing low thermal budget, superb separation of carriers and lower recombination velocity and insignificant contact resistivity²⁷⁾. These materials can be implemented as an alternative solution to silicon dopants. It has been reported in literatures that PEDOT:PSS have exhibited marvelous open-circuit voltage (Voc) of 657 mV and conversion efficiency more than 20%^27-28). But it shows chemical instability due to its hygroscopic character which leads to severe device degradation. TMOs are a kind of material that possess excellent carrier selective properties. The use of TMOs as p-type emitters in n-type c-Si has also been investigated for MoO3, WO3 and V2O5 demonstrating a potential conversion efficiency of 22.5%, 12.5% & 15.7% for this novel solar cell concept28).Molybdenum trioxide (MoO3), tungsten trioxide (WO3) and vanadium pentoxide (V2O5) are some examples of hole extracting TMOs which can replace p type silicon dopants in HIT structures. TMOs possess large work function. These TMOs work as hole-selective contacts as they show high work functions (>5 eV) and lie very nearer to the High Occupied Molecular Orbital (HOMO) level of p-type organic semiconductors, which support ohmic contact origination.

2. Factors affecting HIT cell performance

n-type Silicon wafer: The majority of high efficiency HIT cells have been fabricated by considering n-type C-Si wafer as base material. Even though it is possible to fabricate HIT cells on p-wafer, still Voc is inferior as compared to n-wafer cells as depicted in Fig. 4²⁹⁾. Reports say faster recombination rate of p-type wafer would lead to this issue³⁰⁾.

Fig. 4. J-V curve for MoOx contact based on (a) n-wafer (b) p-wafer

Hole extracting layers: As reported in literature, amorphous silicon (a-Si) doped layers suffer from optoelectronic losses, complexities and deposition as it requires PECVD which introduces toxic gases, require explicit control to optimize proper Voc and Jsc. and quite costly deposition methods³⁰⁾

Moreover, the doping technique comprises high cost, high temperature treatments (more than 800°C), tiny contact fractions, removal of boron rich layer and junction removal^30-31).

3. Temperature dependency of carrier selective HIT

The insertion of i-layer in carrier selective cells creates large valance band offset energy³¹⁾. Drift diffusion velocity of i-layer hinder transfer of holes due to large valance band offset energy. This property enhances with increased forward bias voltage and enhances transmission of minority carriers i.e holes from C-Si base to carrier selective layers³¹⁾. This property opposes enhanced dark current that turned into lower dependence of carrier selective cells on temperature.

4. Effect of work function on cell

The work function (ψ) of hole selective material shows remarkable impact on the hetero contact characteristics of device and plays crucial role in determining the charge transport as well as contact resistance behavior31).

Work function decides transportation mechanism of holes. The potential barriers for electrons depend on work function of hole selective material32). For an efficient hole selective cell height of electron barrier should be higher and height of hole barrier should be lower for efficient transportation of holes32). As work-function decreases electron barrier height also decreases which leads to inefficiency of cell32-33).

For work function below 4.5 eV no remarkable band bending occurs and simultaneously Voc decrease. Decreasing work function increases barrier height for holes33). Thus, value of work-function transported nearby interface which leads to assembly of electrons near interface between hole selective and i layer resulting into increased recombination. This strongly effect Voc, FF and ƞ34). Thus work-function of hole extraction material should be high.

5. Effects on HIT cell parameters

VOC is an important cell parameter and can be conveyed as Voc = kT/q ln (Isc/I01). Leakage current is indirectly proportional to Voc. Therefore with increasing leakage current Io1, Voc decreases. The thickness of emitter does not control Voc. until the thickness goes beyond 27 nm^34-35).Increased thickness escorts to lowering of blue response of the QE and Jsc which leads to use of a thinner p emitter. It has also been investigated that proper surface texturing and improved passivation is the two controlling factors of Voc. Jsc depends on the thickness of the emitter layer, absorption by wafer, BSF structure and modified surface passivation³⁶⁾. Moreover, an increase in emitter thickness leads to lower Jsc.

6. Conclusion

This work focuses on HIT employing hole selective material to replace boron doped layer. These devices have shown the potential to challenge the conventional HIT cells. It can be fabricated on both n-type and p-type. While implementing on n-base with hole extracting layer as emitter and n+ as BSF it shows higher efficiency as compared to p-base. HIT cells impose lower thermal budget and faster fabrication process. At the same time carrier selective HIT cells has lower dependency on temperature, evolving as a risk-free materials to be deposited alternative to p / n doped a-Si:H. These devices depict wide band gap with a distinctive p- or n-type character and a broad range of work functions varying from 2 to 7 eV. For hole selective contacts work-function has to be higher i.e in the range of 4 - 7 eV. In order to achieve best cell results proper surface texturing and improved passivation are the two major factors which control Voc. Jsc depends on thickness of emitter layer, absorption by wafer, BSF structure and modified surface passivation.