Practical Efficiency zpv (at a Junction)

Relative to theoretical efficiency zt (see Section 3.4.2), the practical efficiency zpvof a solar cell is further reduced by the following phenomena: [3]



Figure 3.26 Light trapping in a c-Si solar cell: solar cell reflection loss is reduced and solar cell light absorption is improved through the use of suitably designed (i. e. textured) surfaces and reflective back contacts. In this process, incident light is refracted at the solar cell surface in such a way that it traverses the silicon at an angle, is reflected on the back, and is captured by the silicon to the maximum possible extent via total reflection. The multiple light reflections thus achieved in the cell elongate the light path by a factor of upwards of 20 and allow for full absorption of virtually all photons. This in turn means that solar cells can be considerably thinner, i. e. recombination is reduced using the same material quality and material use is also reduced. With these methods, and despite the reduced cell thickness, the light traverses a path that is long enough to allow for full absorption of even low-energy (i. e. red light, near-infrared) photons [Gre95]. This effect can be further enhanced through the use of such methods on the back of the cell, or through integration of an optical lattice. In addition, an anti-reflection coating is applied to the textured area of most solar cells

• Recombination loss: In some materials, silicon being one of them, photon penetration depth is determined by wavelength. Hence not all electron-hole pairs separated by the internal photoelectric effect are generated in or near the space charge zone, where they can be separated by the field instantaneously. In c-Si some energized photons are already absorbed on the n-layer area (see Figure 3.10), whereas some low-energy photons with h ■ n > EG are not absorbed until they are beyond the space charge zone.

Some electrons that are generated by low-energy photons (in red light or near-infrared light) at unfavourable locations (far beyond the space charge zone) recombine with the abundant holes in the p-zone before being diffused in the space charge zone, where the electric field can retain them via pre-recombination transfer to the n-site, thus allowing them to be definitively shunted to the outer circuit. This recombination mechanism is particularly prevalent at lattice imperfections and on the semiconductor surface (e. g. at the back contact).

In today’s solar cells, design engineers try to minimize back-contact recombination loss by creating a back-area field (BSF) via substantial additional doping (p+) on the p-field area (see Figure 3.27), thus increasing Isc, Voc and Pmax.

Hence the extent to which photons with h ■ n > EG are efficiently leveraged in working solar cells is determined by photon wavelength. Figure 3.28 shows the spectral quantum efficiency for a number of solar cell materials, and represents the state of the art in 1986.

• Self-shading attributable to opaque front electrodes: Front electrode optimization through the use of buried contact cells (inserting front contacts in grooves made using lasers; see Figure 3.29) can reduce self-shading loss from opaque front electrodes by a considerable amount (normally by up to several percentage points). For cells with back contacts, contact is established for the n – and p-zones from the back of the cell, thus completely obviating self-shading (see Figure 4.4).

• Ohmic loss in semiconductor materials: RS and RP in the equivalent circuit shown in Figure 3.12 result in solar cell ohmic loss, which in c-Si solar cells can be reduced in the n-zone via relatively high doping (n+).

• Lower efficiency attributable to temperature: As with Pmax (see Figure 3.18) efficiency Zpvdeclines with rising temperature, whereby the characteristic value for the temperature coefficient of c-Si solar cell efficiency is —0.004 to —0.005 per K. In such cases, the somewhat higher open-circuit voltage of












p p+


! |









Cross section of a solar cell with BSF (back surface field)


Figure 3.27 By application of a more heavily doped coating (p +) just in front of the back contact, a back-area field (BSF) and thus a potential barrier are created that deflect electrons away from the solar cell area, where they are particularly prone to recombination. This in turn greatly increases the chances that an electron will be liberated by a low-energy photon near the back contact in such a way that recombination will be averted and the electron will be diffused while still in the space charge zone, where the electron can be definitively separated from the ubiquitous holes in the p-field and can thus contribute to the outer current flow. Thus realization of a BSF increases the likelihoodрт that a photon-generated electron can be successfully separated and thus be made to flow through the outer circuit. This in turn increases solar cell Isc an (by virtue of reduced Is) also VOc [Gre95]

newer c-Si solar cells has a beneficial effect on this coefficient in that the relative change in the presence of higher Vqc is somewhat lower.

Solar cell efficiency is an extremely important parameter for PV system operation. Needless to say, both researchers and vendors in the PV sector would like to be able to indicate the highest possible efficiency


Figure 3.28 Spectral quantum efficiency of the following materials: Cz, monocrystalline Si (sc-Si); TFS:H, amorphous thin-film Si (a-Si:H); CIS, thin-film CuInSe2. State of the art as of 1986. Arco Solar [3.3]/Willi Maag


Figure 3.29 Reduced self-shading via a front electrode buried-contact arrangement. The front contact is realized via narrow laser grooves. A specially designed back contact increases light reflection, while higher p+ doping creates a BSF that reduces recombination. BP Solar makes this type of solar cell. UNSW, Centre for Photovoltaic Engineering

levels for their research outcomes or products (as the case may be); these values for commercially available products are indicated in Section 1.4.1.

Efficiency levels exceeding those of commercially available products by several percentage points have been achieved in laboratory settings. However, a certain amount of fudging has gone on in pursuit of the highest possible solar cell efficiency levels, in that many researchers base efficiency solely on active surfaces (i. e. those not shaded by front contacts) and thus automatically ‘achieve’ higher values. In some cases, the efficiency values indicated are missing key data such as spectrum AM count, cell area, cell temperature and so on, thus often making it impossible to compare the values in the literature head to head. The solar cell efficiency Zpv data in Table 3.2, achieved in laboratory settings, can be regarded as reliable.

Figure 3.30 shows the structure of a PERL cell developed by a research team led by M. A. Green from the University of New South Wales, which has long since held the record (as at 2009) for the highest monocrystalline Si solar cell efficiency. The efficiency zpv of this cell, which integrates virtually all of the efficiency optimization technologies that are realizable today, is very close to the theoretical limit as in Figure 3.25.

Table 3.2 The lab efficiency level zpv that has been attained (and confirmed by impartial observers) with small cells (area 1 to 4 cm2) and only one p-n junction, at STC (AM1.5 spectrum, 1kW/m2, cell temperature 25 °C) [3.2]. The differences between the data shown here and the data in [Hab07] are partially attributable to a minor change that was effected in the standard AM1.5 spectrum in 2008 (new IEC 60904-3:2008) [3.2]. Reproduced by permission of Wiley-Blackwell





GaAs (thin film)


sc-Si (monocrystalline)


mc-Si (multicrystalline)




Cu(In, Ga)Se2 (CIGS)






rear contact oxide

Подпись: finger "inverted" pyramids Figure 3.30 Structure of an sc-Si PERL cell, which is endowed with a special area design featuring inverted pyramids, bilateral area passivation and point contacts with a BSF. Owing to these characteristics, the PERL solar cell currently holds the record for c-Si cell efficiency, amounting to ZPV = 24% [Gre95]. However, as a result of the many process steps entailed by this cell, producing it is still extremely cost intensive. UNSW, Centre for Photovoltaic Engineering
Figure 3.31 shows the characteristic V-I curve for such a solar cell. Noteworthy here is that the open – circuit voltage Voc, short-circuit current density JSC and fill factor FF are considerably higher than with commercially available monocrystalline solar cells.

A key factor for commercially available solar cells is not only high efficiency, but also a reasonable price. It is no easy matter to integrate all available efficiency optimization technologies into an industrial production process at a reasonable cost. However, the efficiency of commercially available solar cells is bound to increase somewhat in the coming years, in view of the above lab values.

Updated: August 3, 2015 — 11:42 pm