Topic-I

Solar Energy Conversion and Si Solar cell

1.1 Introduction

Sunlight or solar spectrum is a portion of electromagnetic radiation, in particular, infra

red, visible and ultraviolet light are emitted by sun. The solar spectrum falling on earth is
regulated by attenuation of atmosphere. Average solar energy falling on the earth's surface is
known as air mass 1.5 global (AM 1.5G) irradiation. The power density corresponding to AM
1.5 radiation is approximately 1000 Wm-2. If the spectrum does not pass through any air mass
it is known as air mass 0 (AM 0) radiation. It has integrated power density of 1366 Wm -2.
Another spectrum known as air mass 1.5 direct (AM 1.5D), which is defined for solar
concentrator work has an integrated power density of 900 Wm-2 [1.1].

Figure 1.1 The approximate shape and power density of solar spectrum under AM 1.5G, AM
1.5D and AM0 condition

Two important technologies have been developed by harnessing solar energy, namely, (i) Solar
thermal technology and (ii) Solar photovoltaic technology. Solar thermal technology involves
harnessing solar energy (particularly IR radiation) for thermal energy. Various devices like (i)
Solar water heater, (ii) Solar thermal power, (iii) Solar cooker etc. based on this technology are
of widespread use. Photovoltaic technology involves conversion of UV and visible light into
electricity. As the energy carried by photons of these portion of electromagnetic spectrum gets
converted to voltage, the technology is known as photovoltaics (PV). First three chapter of this
book are based on various PV cells and next three chapters focus on other energy materials and
devices.

1.1.1 Steps for PV conversion

There are four fundamental steps [1.2] to convert light energy to electrical energy in
photovoltaic cell or solar cell:

(i) Selection of a suitable material for efficient absorption of photons

(ii) Generation of charge carriers (electron hole pairs) due to absorbed energy in the material
(iii) Separation of charge carriers before they recombine
(iv) Transport of charge carriers from the point of generation to outer circuit through
appropriate electrical contacts.

There are two important aspects involved in this physical process, namely efficient absorption
of light and creation of charge carriers by selecting materials with appropriate physical
properties and making device configuration to transport charges to the outer circuit. Efficiency
of a cell depends upon both these processes and hence the main challenges of solar cell science
and technology depend on selection of right materials and appropriate device configuration. In
this book we will discuss how researchers have explored new materials with appropriate
properties and how these materials have been used in different PV device configuration like,
homojunction solar cells, heterojunction solar cells, sensitized solar cells and tandem solar
cells.

1.2 Semiconducting materials for PV

Semiconductors are a class of materials which due to their intermediate values of band
gaps received maximum attention to materials scientists as PV material. Elementary
semiconductors like silicon (Si) germanium (Ge) or binary compounds like gallium arsenide
(GaAs), Cadmium sulphide (CdS), cadmium telluride (CdTe) have intermediate conductivities
between those of metals and insulators, high mobility with intrinsic charge carrier and
temperature activated conductivity make them very promising materials in PV conversion.
These materials doped with n or p type dopants can be used for fabrication of p-n junction and
act as a device for high performance solar cells. Other essential properties of semiconductors
to be used as junction solar cell are: (i) high optical absorption, (ii) less recombination centres,
(iii) good mechanical strength and stability and (iv) other physical properties appropriate to
make a solar cell circuit configuration for best transport. Another important aspect is nature of
band gap of semiconductors. Direct band gap semiconductors (GaAs, CdS, CdTe) are better in
terms of absorption of light as compared to indirect band gap semiconductors (Si). This
demands much higher thickness (few hundred microns) of indirect band gap semiconductors
as compared to direct band gap (few microns only). Although reduced thickness makes the
direct band gap semiconductor based solar cell more cost effective, but issues related to surface
and interface states# make material more complex.

----------------------------------------------------------------------------------------------------------------
# Surface and interface

An interface is defined as the boundary between two spatial regions occupied by different
materials. The surface is basically an interface between material and vacuum or (air). Surface
and interface states are electronic states formed at the surface and interface, which could be
totally different from bulk electronic states. A detailed understanding of the surface and
interface chemical structure of thin film-based PV device and their correlation with electronic
structure (finally, cell performance parameters) is extremely important [Weinhardt. et al. AdV.
Mat. https://doi.org/10.1002/adma.201806660].
----------------------------------------------------------------------------------------------------------------

1.3 Semiconductor junction

The simplest and effective configuration of semiconductor based solar cell is p-n
junction. Let us first discuss briefly about p-n junction [Fig. 1.2] followed by solar cell circuit
configuration and necessary configuration. When a junction of p-type and n-type
semiconductor is made, electrons from n-type semiconductor diffuse to p-type due to
concentration gradient, and a layer of fixed positive charge in the n-type semiconductor is left
behind. Similarly, holes from p-type semiconductors flow towards n-side, and a layer of
negative charge in the p-type semiconductor is created. Further diffusion of charge carriers
across the junction is prohibited as the space charge sets up an electrostatic force. This opposes
the motion of charges eventually establishing an equilibrium condition. The junction region
does not have any mobile charge carriers and acts like a barrier that opposes the flow of
electrons from n-side and holes from p-side. This region is known as depletion region. The
width of the depletion region can be varied by applying bias voltage to the junction.
Figure 1.2 Transfer of charge carriers across p-n junction and formation of depletion region

For forward bias (n type connected to the negative polarity of battery), the depletion width
reduces, whereas for reverse bias (n type connected to positive polarity of battery), the width
increases. For junction solar cell device, p-n junction is the main active region, and associated
electrical contacts for extracting charge carriers. Therefore, it is a device with two interfaces
namely, (i) semiconductor homo (n and p are same materials) or heterojunction (n and p are
different materials) and (ii) semiconductor-metal junction. A built-in potential$ is eventually
developed across the junction.

Let us see what happens when photons strike this junction configuration. A photon with energy
(E = h), when falls on the junction (i.e., depletion region), gets absorbed, if E is greater than
the band gap energy (Eg) of the semiconductor. The energy so absorbed creates electron-hole
(e-h) pair (Fig. 1.3) and electric field across the junction separates these photo generated e-h
pair towards the electrical contacts on both sides of the junction. There will be no photocurrent
in the outer circuit if the built-in potential is not strong enough to separate photo generated
charge carriers. Charge carriers will recombine due to their opposite polarity.
Figure 1.3 Generation of e-h pair due to interaction with photon with semiconductor junction

----------------------------------------------------------------------------------------------------------------
$ Built-in potential

Built-in potential barrier or built-in voltage is basically the potential difference across the
depletion region.

For acceptor concentration na in the p side and for donor concentration nd in the n side, built in
potential vb can be expressed as

vb = vT ln [nand / ni2]
where vT = kT/e
k = Boltzmann constant
T = absolute temperature
e = magnitude of the electronic charge
vT : thermal voltage is ~ 0.025 V at temp T = 300 K.
As mentioned before, the importance of this built in potential is that it opposes both the flow
of holes and electrons across the junction and hence it is termed as potential barrier.
----------------------------------------------------------------------------------------------------------------

1.4 Working principle of a Solar cell

In order to understand the working principle and electrical behavior of a junction solar
cell, it is important to consider an equivalent circuit and analyze all circuit parameters. As the
heart of the solar cell is semiconductor n-p junction, the ideal equivalent circuit can be
considered as diode with a current source parallel to it. The current source is basically added
due to photogeneration of charge carriers in the junction under illumination. The intensity of
photocurrent increases with increase in incident light intensity. The current is divided in two
pathways i.e. through the diode and the load respectively [1.2] as shown in Fig. 1.4.

Figure 1.4 The equivalent circuit of an ideal junction solar cell

The amount of current flowing through both the sections depends on the intensity of light
falling on the diode and the resistance value of the load. If the load resistance is higher, the
current flowing through the load is smaller as compared to that flowing through the diode. A
potential difference is thus created between the cell terminals with smaller current through the
load, and hence the diode provides a photovoltage. When the voltage across the cell is zero,
the current flowing through the cell is known as photocurrent. Such condition can be satisfied
if external circuit is short circuit (load resistance is zero), and hence this photocurrent is also
known as short circuit current (Isc). This current depends on incident light intensity, band gap,
absorption coefficient of the semiconductor, charge collection efficiency and active area of the
cell. Open circuit voltage (VOC) is defined as the maximum voltage obtained from the solar
cell, when the current through the cell is zero.

1.5 I-V characteristics of a Solar cell

As we have discussed before junction solar cell consists of two important interfaces,
namely, n-p junction between semiconductors and electrical contacts between semiconductors
and metals to collect charge carriers. Let us analyze various circuit parameters based on a metal
semiconductor rectifying Schottky diode and conversion efficiency of the cell. For a Schottky
diode under dark condition, the current-voltage relation can be written as [1.2]:

ID = I0 [exp (eV/nkT) - 1] ---- (1.1)

where,
ID = Electric current in dark condition
I0 = Saturation current
e = Electronic charge
k = Boltzmann constant
n = Ideality factor of the diode

In order to determine some of the physical quantities, we can consider externally applied
voltage, V across the diode ~ 75 mV or higher, which gives a condition,
exp (eV/nkT) >> 1
Therefore, equation 1.1 can be written as [1.2]

ID = I0 exp (eV/nkT) ---- (1.2)

Writing it in logarithmic form,

log10 (ID) = (e / 2.303nkT).V + log10 (I0) ----- (1.3)

One can determine the ideality factor (n) of the diode, the gradient of the straight-line portion
of the plot between log10(ID) versus V as given in equation (1.3). Ideality factor is basically is
a factor indicating recombination process in semiconductor diode. If the recombination occurs
in the neutral region of the diode the value of n = 1. In this case current transport basically takes
place by thermionic process. For various recombination and generation (R & G) centres in the
depletion region, the current transport is dominated by R & G centres and the value of n
becomes ~ 2. Generally, the value of n lies between 1 and 2 in a practical device as both the
processes occur simultaneously. The intercept of the straight line gives the value of saturation
current (I0). Saturation current depends on various parameters including the value of potential
barrier (b) between semiconductor-metal junction.* From a liner plot of I-V characteristics,
two other parameters of a solar cell namely series resistance (Rs) and shunt resistance (Rsh) can
be evaluated. The forward and reverse characteristics can be used to estimate the values of R s
and Rsh respectively. Accordingly, the equivalent circuit of a solar cell including both these
resistances are depicted in Figure 1.5

Figure 1.5 Equivalent circuit of solar cell including series and shunt resistances
----------------------------------------------------------------------------------------------------------------
* Metal-semiconductor junction

Metal and semiconductor having different work function form a heterojunction. Depending
upon work function values they are of two types: (i) when work function of metal is higher
than that of semiconductor (m > s), Schottky junction and (ii) when work function of metal
is less than semiconductor (m < s), Ohmic junction. The bending of energy bands in these
junctions depends upon work function values. The energy band diagrams for metal/n-doped
semiconductor and metal/p-doped semiconductor ae shown below:
In the first case (a) the electron affinity of semiconductor is lower than the work function of
the metal, thus a barrier for electron is induced. In the second case (b) it is opposite where a
barrier for holes is induced. Silicon solar cell is generally analyzed based on Schottky diode
formulation.

----------------------------------------------------------------------------------------------------------------

Let us now see what happens when a solar cell with external load resistance is illuminated.
There is a voltage across the external load which reduces the potential barrier for carrier
injection as it opposes the built-in potential. This is similar to p-n junction with forward current,
which opposes the photocurrent generated due to the solar radiation as in case of
photogenerated current, electrons move towards n-side and holes towards p-side. Therefore,
the current through the diode under illumination is [1.2]:

IL = ID -ISC = I0 exp (eV/nkT) - ISC --------- (1.4)

The forward current which basically arises due to minority carrier movement constitutes dark
current of the cell and the I-V characteristics is typically a forward bias characteristic of a
diode. The I-V curve shifts to negative current axis as photocurrent opposes the forward bias
dark current. The I-V characteristics of solar cell under dark and illuminated condition are
shown in Figure 1.6.
Figure 1.6 I-V characteristics of solar cell under dark and illumination condition. Maximum
power is indicated by shaded area

The short circuit current and open circuit voltage are indicated in Figure 1.6. The short circuit
current density (mA cm-2) is defined as
JSC = ISC / area of the contact ------------ (1.5)
Open circuit voltage (VOC) which is measured when the circuit is in open condition depends
on various factors. Under this condition from equation 1.4, we can write

0 = I0 exp (eVOC/nkT) - ISC

I0 exp (eVOC/nkT) = ISC

VOC = n [b + kT/e ln (JSC/A*T2)] --- -------- (1.6)

From equation 1.6, it can be stated that open circuit voltage depends on ideality factor,
temperature, short circuit current and saturation current. Optimization of VOC depends on
interplay between n and ISC. Further, saturation current which depends on barrier height, can
be enhanced by incorporating an insulating layer within metal-semiconductor junction [1.2].
The value of JSC can be maximized by generating a higher number of charge carriers, which is
possible by absorption of photons from different regions of solar spectrum. The output power
delivered by a solar cell as a function of voltage is shown in Fig. 1.7 [1.3].

Figure 1.7 Power of an ideal solar cell as a function of voltage

1.5.1 Efficiency of a cell

The solar cell delivers power in the operating voltage range i.e. from 0 to V OC. The power
delivered from the solar cell is defined as

P = I  V ---- ------- (1.7)

As shown in Fig. 1.7, the maximum is at the cell's operating point (i.e. maximum power point
in the I-V curve) and the maximum power is defined as,

Pm = Im  Vm ----- ----- (1.8)

The fill factor (FF) of a solar cell is defined by the relation:

FF = Vm  Im / JSC  VOC ---- ------ (1.9)

The efficiency () of the cell is defined as

 = Jm  Vm / Pi --- ------- (1.10)

 = JSC  VOC  FF / Pi --- ------ (1.11)

where Pi is the incident light power. If it is considered for /unit area, then P i is the incident
power density. Therefore, optimization of efficiency depends on four important parameters,
i.e. JSC, VOC, FF and . All these quantities are defined for a particular illumination condition.
Generally, an incident power density of 1000 Wm-2, air mass 1.5 spectrum and operating
temperature 25 0C are considered as standard test conditions.

1.5.2 Parasitic resistances

There are two parasitic resistances series (RS) and shunt (Rsh) resistance, which are
electronically equivalent to contact resistance, leakage resistance etc. As Rs depends on various
contact resistances, interfaces within the device and materials, ISC as well as FF decreases with
increase in RS. Ideally the total series resistance should be zero. On the other hand, R sh, arises
due to leakage resistance, can be enhanced by minimizing recombination and generation
process of charge carriers. In ideal case Rsh should be infinity. Adding these parasitic
resistances, the equation for current density can be written as [1.2]:

J = J0 exp [e(V+JRs/nkT)] - (V+JRs) / Rsh - ISC -------- (1.12)

1.6 Recombination in solar cell

Recombination, an opposite process of charge carrier generation, where an electron

recombines with a hole and produces light or heat. There are three types of recombination, (i)
radiative or band to band, (ii) defect mediated and (iii) Auger recombination.
In radiative recombination an electron from the conduction band directly combines with a hole
and releases a photon dominates in direct band gap semiconductors.
The first recombination process which is basically mediated by defects is known as Schokley-
Read-Hall (SRH) process. SRH occurs when an energy state in the forbidden region traps an
electron (hole). These states can either be deliberately added to the material or unintentionally
introduced. If a hole (or an electron) moves up to the same energy state before the electron is
thermally re-emitted into the conduction band, then it recombines. When an electron and a hole
recombine, but rather than emitting the energy as heat or as a photon, the energy is given to a
third carrier, an electron in the conduction band, Auger recombination process takes place.
One important issue in solar cell is the rate at which recombination occurs. Recombination rate
again depends on number of excess minority carrier. If minority carrier concentration is small,
recombination rate is low. Minority carrier lifetime and carrier diffusion length are two
important parameters which decide recombination rate in solar cell. If n is the number of
excess minority carrier and R the recombination rate, the minority carrier lifetime is defined
as:

 = n / R ------ (1.13)

Considering all three-recombination process as discussed above, the bulk recombination rate
(bulk) can be written as:

1/bulk = 1/band + 1/SRH + 1/Auger ------ (1.14)

It is to be noted that solar cells having higher minority carrier lifetime has more efficiency. The
average length that a carrier moves between generation and recombination is known as
diffusion length. Semiconductor materials with higher dopant concentration have greater
recombination rates and have shorter diffusion length. Diffusion length (dl) is related to the
carrier lifetime through the following equation:

dl = (D)1/2 ----------- (1.15)

where D is the diffusivity in m2sec-1 and is a measure of how quickly a group of particles fill a
space.

1.7 Basic design of a Si Solar cell

Although Si has relatively lower light absorption coefficient due to its indirect band
gap, is still most widely used material in commercial solar cell due to its large abundance and
well-known process of manufacturing. A typical thickness of 200 to 400 m Si wafer is used
as a base material of Si solar cell. Generally, n-type silicon is placed in the front of the cell
where most of the light is absorbed. Due to absorption of light, a large fraction of charge
carriers is generated within the diffusion length of p-n junction of the cell. A lower resistivity
metal grid is used to conduct away the current. However, as metal grid shades the cell from the
incoming light, there is a compromise between light collection and resistance of the metal grid.
Due to high reflection of bare Si, the light trapping in the cell is less. This problem can be
tackled by texturing the surface or coating the surface with anti-reflection coating (ARC). ARC
is basically a thin layer of dielectric material whose thickness is chosen in such a manner so as
to make reflected light energy as minimum as possible. Further reduction in reflection can be
achieved through double layer or multiple layer anti reflection coating. Rear contact is not as
important as front contact as it is not close to junction and need not be transparent. A schematic
of basic structure of solar cell is given in Figure 1.8.

Figure 1.8 Schematic of basic design of a solar cell

1.8 Si solar cells and a solar module

There are three types of Si solar cell in commercial use, (i) monocrystalline Si solar
cell, (ii) polycrystalline Si solar cell and (iii) amorphous Si silicon cell. Monocrystalline Si
solar cells are prepared from a high purity single crystal ingot [1.2]. Standard Czochralski
method is adopted to grow these ingots. A precisely oriented rod-mounted seed crystal is
dipped into the molten silicon. The rod is then slowly pulled upwards and rotated
simultaneously, allowing the pulled material to solidify into a monocrystalline cylindrical ingot
of few meters in length and weighing several hundred kilograms. In order to make cells the
ingots of size (typically 4 and 5 inches) are cut into pieces or sliced into thin wafers.
Monocrystalline Si solar cell has reported efficiency of ~ 27% is one among the highest in
various categories of solar cells, namely Poly -Si, a-Si, thin film solar cells, dye sensitized solar
cell, perovskite solar cell etc. Relatively lower number of recombination centres and better
absorption of light are the primary reasons for higher efficiency. However, mono-Si solar cells
are expensive and generally used in satellites powered by solar energy and other spacecrafts.
Efficiency can be further improved by making multi-layer solar cells and will be discussed
later. Polycrystalline silicon, or multicrystalline silicon a high purity polycrystalline form of
Si, used as a raw material by the solar photovoltaics and electronic industry [1.2]. Polysilicon
is produced from metallurgical grade silicon by a chemical purification process, called the
Siemens process. For making solar modules, polysilicon is first melted at high temperature and
ingots are formed. These are then sliced into wafers and processed for cell and module.
Polycrystalline solar cell is relatively less efficient as compared to monocrystalline, however,
they are much cheaper. Polycrystalline cells do not function efficiently at high temperatures as
they have less heat tolerance as compared to monocrystalline Si and hence disadvantageous to
use in the places with hot climatic condition.
A PV module/single panel is generally an assembly of connected solar cells that absorb sunlight
as a source of energy to develop electricity. A group of PV modules (also called PV panels) is
wired into an array (known as PV) array to gain a required current and voltage. A
monocrystalline solar panel is shown in Fig. 1.9 below.

Figure 1.9 Schematic of a monocrystalline silicon solar cell

The number of cells connected and their size generally decide the peak output power of a solar
module. Based on standard Standard Test Conditions (STC), i.e. irradiance of 1,000 W/m²,
solar spectrum of AM 1.5 and module temperature at 25°C, module performance is rated in
peak watts [Wp]. Thus, a 50 Wp module can be expected to supply 50 W of power under optimal
conditions [1.3].
In Fig. 1.10 illustrates various stages of Polysilicon solar panel starting from poly silicon.

Figure 1.10 Steps from Polysilicon to PV module stage

1.8.1 Amorphous silicon solar cell

The third category of Si solar cell is amorphous silicon solar cell, which is of thin film form.
Among various allotropes, amorphous silicon (a-Si) is the non-crystalline allotropic form of
Si. In a-Si long range order does not exist, which results in large number of dangling bonds, i.
e. unsaturated valances of immobilized Si atom of the order of 1019/cm3 [1.4]. These dangling
bonds create two serious issues, (i) generation of recombination centers which trap charge
carrier and (ii) pin the Fermi level and hence having difficulty to dope into either n-type or p-
type semiconductor. Therefore, a good solar cell cannot be made out of a-Si. However, one can
overcome this problem by incorporating H in a-Si. Chittik et al. [1.5] have demonstrated that
dissociation of silane gas (SiH4) on heated substrate can incorporate 10% hydrogen in a-Si and
defect density can be drastically reduced. Hydrogen atoms combined with the dangling bonds
make an amorphous hydrogenated silicon (a-Si:H), which can be doped for making junction
devices. A comparison between photovoltaic properties of c-Si and a-Si solar cells are given in
table 1.1. In amorphous structure bond lengths are not as well defined as crystalline structure
and hence internal scattering of light takes place through structural inhomogeneities, resulting
in higher absorption in the former.
Table 1.1 Various photovoltaic properties of c-Si and a-Si solar cells [1.6]

----------------------------------------------------------------------------------------------------------------

Photovoltaic properties c- Si solar cell a-Si solar cell

----------------------------------------------------------------------------------------------------------------

Band gap 1.1eV ~1.8eV

Spectral range (75%-85% QE) 440 nm- 650 nm 550 nm-750 nm

Optical to electrical conversion efficiency 22.3% - 26.1% 14.0%

Sufficient thickness 100 µm 1-2 µm

Area required/kW (m2) 7 -8 15

----------------------------------------------------------------------------------------------------------------

Therefore, concerning absorption of light, a-Si exhibits higher advantage as compared to c-Si
and its absorption coefficient is order of magnitude larger as compared to c-Si. Therefore, the
thickness required is much smaller in a-Si as compared to c-Si as shown in Table 1.1 for
absorption of same amount of light. Apart from that the temperature stability of a-Si solar ell
is slightly better than c-Si solar cells. Moreover, they can be deposited on various substrates,
not only on glass but also plastic etc. at relatively lower temperature. However, as far as
conversion efficiency is concerned, c-Si solar cell outcompetes a-Si solar cell as seen in table
1.1. This sharp difference occurs due to large number of dangling bonds in amorphous silicon
(even after incorporation of H) solar cell which act as recombination center and reduce carrier
lifetime and hampers transport of charges. Moreover, hydrogenation deteriorates under
illumination. It has been seen in the literature that within the first 100 hours under exposure to
light, a-Si solar cells exhibit a steady drop in efficiency measured by a decrease of
photoconductivity and dark conductivity. This arises basically due to relatively higher diffusion
of hydrogen and change in local bonding coordination [1.7].

Instead of its lower efficiency, amorphous silicon cells are appropriate to use in: (i) artificial
(indoor) light condition due to its capability of higher optical absorption, (ii) micro power
applications like solar powered calculators, watches etc., (iii) applications with design
aesthetics due to its flexibility, (iv) high temperature environment as it is less sensitive to
temperature as compared to c-Si and (v) environment friendly PV applications.
1.8.2 A-Si/C-Si heterostructure solar cells

In order to reduce the surface defect density and to increase the efficiency of the cell,
passivation layers are used. The first idea is to use dielectric surface passivation layers, e.g.,
SiO2, on the front side of the solar cell [1.8], which reduces surface recombination. However,
metal contacts are directly made on the silicon wafer, which forms a metal semiconductor
interface, and hence recombination at this interface remains a problem. One of the most
successful options to reduce recombination is to make a p-n heterojunction comprising of
doped a-Si:H thin film on a doped silicon crystalline wafer [1.9]. However, as amorphous
silicon is highly doped there are lot of interface state density which can enhance recombination
centres. To overcome this issue, a heterojunction with intrinsic thin layer structure is formed
by growing a thin buffer layer of a-Si:H film between emitter and the wafer. This type of
heterostructure has another advantage that open circuit voltage increases as amorphous silicon
has higher band gap. This type of solar cell also shows a better performance when temperature
increases. It can give higher efficiency at lower fabrication cost. A typical configuration with
a-Si:H on both sides is shown in Fig. 1.11 as suggested by Wolf et al. [1.8].

Figure 1.11 Schematic diagram of A-Si/C-Si hetero structure solar cell

In the above configuration, before sunlight hits the crystalline layer, it crosses through a thin
film at the top and it also grabs some sunlight that reflects off the layers below. A heterojunction
solar panel made out of sandwich of three different photovoltaic layers, can reach efficiencies
of 21% or higher. Researchers have developed several others heterostructure silicon solar cells
like TMO/c-Si solar cell, PEDOT:PSS/c-Si solar cell etc. Discussion on these solar cells is not
included in this book.

1.9 Plasmon enhanced light trapping in Si solar cell

As discussed above the typical thickness of c-Si solar cell is few hundred micrometers.
In order to reduce the production cost of Si, reduction of wafer thickness is important, i.e.,
materials cost must be reduced. However, this is not possible due to lower light absorption
capability of Si. Other approach to improve performance is to geometrically trap light by
texturing of Si surface. In order to trap more light for conventional thick silicon solar cell,
pyramid structures have been made on the surface. These structures scatter light into the solar
cell thereby increasing optical path length of the light. In order to achieve further light trapping,
a combination of surface texture and a reflector layer is demonstrated by many researchers
[1.10]. The mechanism involved in this process is simple. First light gets scattered due to the
textured surface and then it gets reflected to Si layer by the back reflector (e.g. Al). These two
processes (Fig. 1.12) eventually enhance the optical path in Si and enhance overall absorption
of light.

Figure 1.12 Schematic representation of light capture in Si by surface texture and back
reflector

When light harvesting layers are thicker than the wavelength of sunlight in the visible and near
infra-red region such surfaces are designed. However, such surface is basically designed for
the. For thin Si wafers large geometric texturing is a challenge and hence researchers adopted
a methodology by using the surface plasmon resonance (SPR) of metal nanoparticles to
enhance light trapping in the active layer. Surface plasmons (SP) are coherent electron
oscillations that propagate between the interface of a metal and a dielectric [1.11].
Electromagnetic field can be strongly confined within metal/dielectric or semiconductor
interface. The enhanced electromagnetic field along with increased scattering cross section can
be obtained through excitation of surface plasmons. Overall absorption of light in the thin
semiconducting layer enhances when both enhanced field (i.e. stronger absorption) and
scattering cross section enhance. Instead of thin metal layer if metal nanoparticles (NPs) are
grown on the surface or interface, SPs can be localized, which is known as localized surface
plasmon (LSPs), which play a major role in enhanced light trapping. In this subsection, firstly
some fundamental concepts of SP and LSP will be discussed, which will be followed by brief
survey on plasmon enhanced silicon solar cell and their performances.

1.9.1 Surface plasmon of metal nanostructures

Due to strong interaction of the free electron on the metal surface and incident light, enhanced
capabilities of light trapping is possible. A collective oscillation known as localized surface
plasmon resonance (LSPR) sets in when the frequency of the incident photons matches well
with the plasmon frequency of the free electrons of the metal nanostructures [1.12]. LSPR is
schematically represented in Fig. 1.13. Unlike surface plasmon polaritons (SPP)# , which
propagates along the interface, LSP is totally bound and cannot propagate.

Figure 1.13 Schematic representation of LSPR

----------------------------------------------------------------------------------------------------------------
-#Surface plasmon polariton:

Electrons and positive ion cores in a metal is considered as plasma and is used to explain some
natural phenomena like color of metal and temperature dependent conductivity. ‘Plasmon’ is
defined as quantum of plasma oscillation, with an eigen frequency [ Ref: Charles Kittel,
Introduction to Solid State Physics]

p = (ne2 / m0) 1/2

where n is the electron density. They can be excited by free electron beams and by UV light
source. Coherent electron oscillation at the interface between any two materials, results in
surface plasmon and can be excited conveniently and efficiently by polarized light source.
Surface plasmon can also be treated as a collective set of surfaces plasmon polaritons (SPPs),
which are basically electromagnetic waves that travel along a metal dielectric or metal air
interface. [Ref: Huang et al. Plasmonic thin film solar cell, doi: 10.5772/65388 ]

There would be no SPPs, if the interface supporting SPPs shrinks to a scale of nanometer and
takes shape like sphere or ellipsoid. However, considering electrons in this tiny metal ball as a
neutral plasma, an intrinsic resonance similar to volume plasma resonance is expected. This is
known as ‘localized surface plasmon’(LSP) in literature.

----------------------------------------------------------------------------------------------------------------

The optical properties of the metal NPs depend on their size, shape, surface coverage on the
substrates, surrounding medium etc. [1.13, 1.14]. The basic function of LSP suggests that light
can be either scattered or absorbed by metal NPs [1.15] and hence the scattering should be
maximized and absorption should be minimized across the wavelengths of interest in the solar
cells. Mathematically, the scattering and absorption cross section can be represented by
following equations [1.16]

scatt = (1/6) (2/)4 ||2 ------- (1.16)

abs = (2/) Im () ------- (1.17)

Where,  (the polarizability of the particle) is given by

 = 3V [ (p/m - 1) /( p/m + 2) ] ------ (1.18)

V is the particle volume, p and m are dielectric functions of particle and the embedding
medium respectively. Here the dielectric medium surrounding metal NPs could be air, water.
Organic species, inorganic coating etc. Sum of scattering and absorption known as extinction
decides the ability to concentrate light at the metal dielectric interface. A strong correlation
with composition and morphologies of metal nanostructures as well as the dielectric medium
with LSPR has been shown by various researchers. Optimum combination of all these physical
quantities gives an opportunity to tune wavelength of light for maximum extinction. For
example, the  values for maximum extinction for Ag, Au and Cu nanospheres of some specific
dimensions are 420 nm, 520 nm and 610 nm respectively, which fall in the near ultraviolet and
visible region of the spectrum. For one dimensional structure (Au nanorods), two different
extinction peaks can be observed. These are attributed to longitudinal and transverse plasmon
resonance. Aspect ratio of the nanorods basically controls the position of longitudinal
resonance peak. It has been shown that longitudinal LSPR can be red shifted from 700 nm to
1200 nm with change in aspect ratio, keeping transverse mode unchanged [1.17]. Further
resonant frequency can be tuned by changing refractive index of the embedding medium. It has
been shown that the wavelength of maximum extinction of Au nanodisk increases with increase
in dielectric constant of the embedding medium [1.18].

1.9.2 Plasmonic solar cell configuration

Plasmonic nanostructures can be integrated with PV cell in three different geometries: (i)
metallic nanostructures at the top of the solar cell, where light gets trapped by scattering from
metal nanoparticles, (ii) metal nanostructures embedded in the semiconductor, where light gets
trapped primarily by excitation of LSPR, (iii) metal nanostructures at the bottom of the
semiconductor, where light gets trapped by excitation of SPPs.

For scheme 1 (Fig. 1.14), i.e., when metallic nanostructures are placed on the top of the
semiconductor, part of the light will be scattered in to the semiconductor as it has higher
permittivity.
Fig. 1.14 Plasmonic nanostructures on the surface of the semiconductor

The incident light scattered into the semiconductor will be scattered again from the back side
of the metal nanostructure. Derkacs et al. [1.19] have reported 8.1% increase in short circuit
current density and 8.3 % enhancement in efficiency after depositing Au NPs on amorphous Si
of thickness ~ 240 nm. Pillai et al. [1.20] have shown significant absorption of light over a
broad range in 1.25 µm Si-on-insulator solar cell after growing Ag NPs on top of it. Literature
also shows a strong dependence of particle size, shape and surrounding medium on solar cell
performance. Catchpole et al. [1.21] have shown that hemisphereical and cylindrical
nanoparticles coating result in much stronger light absorption as compared to spherical NPs.
This is due to the fact that they have more efficient coupling of the scattered light into the
semiconductor substrate due to less average distance. Uhrenfeldt et al. [1.22] have
experimentally demonstrated a broadband photocurrent enhancement ranging from UV to IR
due to deposition of Al NP array on top of a Si solar cell. Combination of AR coatings and
metal nanostructures is another approach to enhance performance of solar cell has been
demonstrated by researchers. As shown by Munday et al. [1.23], a combination of AR coating
and plasmonic grating structure on ultrathin Si layer exhibits much stronger absorption as
compared to individual structures, i.e., only AR coating or plasmonic structure.

The second scheme i.e., metal nanoparticle embedded in semiconductors can efficiently
concentrate light in the form of surface plasmons as they work like antenna to intensely absorb
incident light and localize the sunlight energy on NP surface. This particular scheme of light
trapping has also been demonstrated as an effective way to enhance performance of Si solar
cell, polymer solar cell and other system.

In the third configuration a periodic corrugated metal nanostructure is made which helps to
generate surface plasmon polariton (SPP), a transverse wave propagation along the metal-
semiconductor interface. SPP facilitates absorption of light along lateral direction, increases
the light propagation length, eventually enhancing performance of solar cell device. Paetzold
et al. [1.24] designed a periodic nanostructured plasmonic back contact for an a-Si:H solar cell,
which showed 26% enhancement in short circuit current density. Let us now make a brief
comparison between these three different configurations: It is usually seen for relatively thicker
layers with dielectric spacer between metallic particles, the scattering cross section for the front
located particles is larger than that for rear located particles [1.25]. However, for front loaded
particles, light absorption below plasmon resonance wavelength is reduced due to Fano effect
resulting from the interreference effects between the scattered and the incident light. Silver
(Ag) nanocones as the plasmonic particles have been demonstrated as plasmonic particles in c-
Si/Al2O3/Ag nanocone solar cell structure by Yan et al. [1.26]. Particles is treated as front
located when light is incident from air and they are treated as rear ones when light is illuminated
from Si side.

Literature reveals that [1.27] performance of plasmonic solar cell can be further enhanced when
metal nanostructures are further integrated with AR coatings. Metal nanostructures can also be
integrated with AR coatings to enhance performance of plasmonic solar cell [1.27]. Aluminum
(Al) NPs coated on 75 nm silicon nitride antireflection coating in 1 µm silicon film provides a
strong broadband light absorption enhancement [1.27]. As discussed before, plasmonic
resonance and hence absorption is also strongly influenced by refractive index of the
surrounding medium. Existence of SiO2 layer between Si and Ag particles enhances light
absorption by of Ag coated Si (rear side) [1.28]. In case of thin film solar cells, combination of
front and rear nanostructures or dual interface nanostructures where plasmonic NPs (rear) and
front dielectric grating have been adopted to enhance absorption efficiency [1.29]. Shi et al.
[1.30] have shown that a combination of silicon front surface grating and the rear located
bilayer Ag nano-hemispheres can be combined to optimize short-circuit current density.

1.10 Nanostructured Si solar cell (Black silicon) solar cell

Engineering materials down to nanoscale or nanostructuring is another potential

process to modify the properties of a material and improve the performance when used in a
device. Silicon solar cells with best possible fabrication process has achieved efficiency more
than 25%. For further enhancement of efficiency nanostructuring has been adopted for better
anti reflection and light trapping. This also reduces cost related to usage of bulk material.
Various nanostructures like nanopyramids, nanopillar, nanocones, nanodisks etc. have been
considered as materials which can substantially improve light trapping in silicon solar cells and
a new cost-effective technology is emerging [1.31]. Literature indicates that near zero
reflection over a broad range of wavelength is possible [1.32], which results in so called “black
silicon”. Moreover, over a wide range of incident angles, these nanostructures can also
effectively suppress the surface reflection. This is extremely beneficial for advanced solar
power applications. Wu et al. [1.33] have shown that there is an enhanced light trapping due to
surface morphology of black silicon and higher absorption. Black silicon textures have been
achieved by various techniques, which include (i) laser chemical, (ii) electrochemical and (iii)
reactive ion etching. M. Halbwax et al. [1.34] have prepared micro and nanostructured silicon
for PV cell by fs laser. They have demonstrated an enhancement of photocurrent of ~30% in
the laser treated areas, which were like nanostructured forest created by laser impact followed
by boron doping and rapid thermal annealing (RTA). T. Sarnet et al. [1.35] have demonstrated
high absorption and appreciable enhancement in photocurrent by a nanosurface structure
created by laser treatment followed by traditional doping method.

For more detailed discussion of nanostructured silicon solar cell interested readers may go
through recent literature on this topic.

Bibliography -1

[1.1] https://www.pveducation.org/pvcdrom/appendices/standard-solar-spectra.

[1.2] I. M. Dharmadasa, Advances in Thin Film Solar Cells, Pan Stanford Publishing Pvt. Ltd.
2013.

[1.3] Tom Markvart and Luis Castañer. Principles of solar cell operation. Practical Handbook
of Photovoltaics: Fundamentals and Applications, page 7, 2003.

[1.4] K. L. Chopra, P. D. Paulson, and V. Dutta. "Thin‐film solar cells: an overview." Progress
in Photovoltaics: Research and applications 12.2‐3 (2004): 69-92.

[1.5] R. C. Chittick, J. H. Alexander, and H. F. Sterling Journal of the Electrochemical Society

116.1 (1969): 77.

[1.6] Paul R. Hart, "Crystalline vs. Amorphous Silicon - a Comparison of their Respective
Properties and their Significance in Photovoltaic Applications." Seventh EC Photovoltaic Solar
Energy Conference. Springer, Dordrecht, 1987.

[1.7] Fritzsche, Hellmut. "Development in understanding and controlling the Staebler- Wronski
effect in a-Si: H." Annual Review of Materials Research, 31.1 (2001): 47-79.

[1.8] Stefaan De Wolf, Antoine Descoeudres, Zachary C Holman, and Christophe Ballif. High-
efficiency silicon heterojunction solar cells: A review. Green, 2(1):7–24, 2012.

[1.9] K.Niemann W.Fuhs and J.Stuke. AIP Conf. Proc., 20(345), 1974.
[1.10] Modou Beye, Djicknoum Diouf, Babacar Dieng, Amadou S.Maiga, Chapter
6 - Nanostructured silicon for antireflection and light trapping in crystalline silicon solar cells,
Micro and Nano Technologies, 2020, Pages 149-173.

[1.11] C. H. Fosli, A. Thøgersen, S. Karazhanov, E.S. Marstein, Energy Procedia, 2011, 10,
287-291.

[1.12] F.J. Beck, A. Polman, K.R. Catchpole, Journal of Applied Physics, 105 (2009),
p. 114210.

[1.13] H.A. Atwater, A. Polman, Nature Materials, 9 (2010), p. 205213.

[1.14] Thøgersen A, Bonsak J, Fosli CH, Muntingh G., Journal of Applied Physics 110, 044306
(2011).

[1.15] Pillai S. Surface plasmons for enhanced thin-film silicon solar cells and light emitting
diodes. Ph.D. thesis. School of Photovoltaic & Renewable Energy Engineering, UNSW 2007.

[1.16] Bohren, C. F.; Huffman, D. R. Absorption and Scattering of Light by Small Particles;
Wiley: Weinheim, 2004.

[1.17] X. Ye, C. Zheng, J. Chen, Y. Gao and C. B. Murray, Nano Lett., 2013, 13, 765–771.

[1.18] M. M. Miller and A. A. Lazarides, J. Phys. Chem. B, 2005, 109, 21556–21565.

[1.19] Derkacs D., Lim S. H., Matheu P., Mar W., Yu E. T. Appl. Phys. Lett. 2006; 89:093103–
3.
[1.20] Pillai S., Catchpole K. R., Trupke T., Green M. A., J. Appl. Phys. 2007; 101:093105–8.

[1.21] Catchpole K. R., Polman A., Appl. Phys. Lett. 2008; 93:191113.

[1.22] Uhrenfeldt C., Villesen T. F., Tetu A., Johansen B., and Larsen A. N. Opt. Express.
2015; 23:A525.

[1.23] Munday J. N., and Atwater H. A. Nano Lett. 2011; 11:2195.

[1.24] Ulrich W. Paetzold, Etienne Moulin, Bart E. Pieters, Reinhard Carius, and Uwe Rau,
Optics Express, 19,S6, A1219-A1230, (2011).

[1.25] Beck F., Mollapati S., Polman A., Catchpole K. Appl. Phys. Lett. 2010; 96:033113.
[1.26] Yan W., Stokes N., Jia B., and Gu M. Opt. Lett. 2013; 38:395.

[1.27] Zhang Y., Chen X., Ouyang Z., Lu H., Jia B., Shi Z., et al. Opt. Express. 2013 ;3:489.

[1.28] Park J., Park N., Varlamov S. , Appl. Phys. Lett. 2014; 104: 033903.

[1.29] Abass A., Le K., Alu A., Burgelman M., Maes B. Phys. Rev. B. 2012; 85:115449.

[1.30] Shi Y., Wang X., Liu W., Yang T., Ma J., Yang F. Appl. Phys. Express. 2014; 7:062301.

[1.31] Jian Lv, Ting Zhang, Peng Zhang, Yingchun Zhao and Shibin Li, Nanoscale Research
Letters (2018) 13:110.
[1.32] Serpengüzel A, Kurt A, Inanç I, Cary JE, Mazur E (2008) J of Nanophotonics,
2(1):021770.

[1.33] Wu C, Crouch CH, Zhao L, Carey JE, Younkin R, Levinson JA, Mazur E (2001), Appl.
Phys. Lett. 78(13):1850–1852.

[1.34] Halbwax M, Sarnet T, Delaporte P, Sentis M, Etienne H, Torregrosa F, Vervisch V,

Perichaud I, Martinuzzi S (2008), Thin Solid Films 516(20): 6791–6795.

[1.35] Sarnet T, Halbwax M, Torres R, Delaporte P, Sentis M, Martinuzzi S, Vervisch V,

Bastide S (2008) , In Proc SPIE 6881:688119.