14 Optical Absorption-Solar Cell and Photodetectors

Dr. Ayushi Paliwal and Dr. Monika Tomar

Introduction

Optoelectronic devices refer to those electronic devices whose principle of operation is dependent on both light and electrical currents. They come under the category of photonic devices and generally include electrically driven light sources such as laser diodes and light-emitting diodes, components for converting light to an electrical current such as solar and photovoltaic cells and devices that can electronically control the propagation of light. P-n junctions are an integral part of several optoelectronic devices and are widely employed in commercially available devices. Optoelectronic devices such as photodiodes, solar cells, LEDs and laser diodes are specifically designed to optimize the light absorption and emission, resulting in high conversion efficiency. In the subsequent sections, two important Optoelectronic devices will be discussed in detail i.e. LEDs and LASERs.

1.1 Homojunction and Heterojunction devices

At forward bias in a normal homojunction photonic device (i.e. p-n junctions fabricated from a single material with different doping), electrons are injected into the p-type material and holes are injected into the n-type material. The electron-hole recombination then takes place on both sides of the junction, and light is generated in both p-type and n-type regions. But in a heterojunction photonic device i.e. a p-n junction between two materials where say the n-type material has a wider energy gap than does the p-type layer, minority-carrier injection into the n-type material can be eliminated. Such a device is called a single heterostructure (SH) device, and if a structure consists of two barriers, i.e. two large-bandgap semiconductors, then the structure is called a double heterostructure (DH).

In homo-junction devices, carriers (electrons and holes) diffuse on an average over diffusion lengths (Ln and Lp) before recombining. However, in heterojunctions, carriers are confined by the heterojunction barriers.

Advantages of heterostructures

(i) The inclusion of heterostructures allows one to improve the efficiency of devices by confining carriers to the active region, thereby avoiding diffusion of minority carriers over long distances.

(ii) Ability to use wide band gap material.

Disadvantages of heterostructures

(i) If the thickness of the active layer is much larger than barrier then carriers may escape out of the region

(ii) The overflow of carriers from the active region into the confinement regions at high injection current densities.

1.2 Electroluminescence in semiconductors: Radiative and non-radiative transitions

In devices such as photo-detectors and solar cells, light illumination results in an increase in the current due to the flow of photo-excited charge carriers. Opposite to this phenomenon, emission of light can be observed in some semiconductors upon passing electrical current through the sample and this process is termed as electroluminescence.

Generally light emission is caused by injection electroluminescence in which radiative recombination of electrons injected into p-side and holes injected into n-side material of the forward biased p-n junction is observed. This technique is employed in Light emitting diodes, where radiative recombination of charge carriers is primarily required. Direct band gap semiconductors are advantageous over indirect band gap semiconductor for the same reason.

However, not all transitions in direct band gap semiconductors are radiative in nature. The presence of defects and traps in a material alters its properties and might result in non-radiative transitions via dissipation of heat in most cases.

These two recombination path ways can be considered as parallel process occurring across the energy gap of the semiconductor as shown in Figure 2

Under continuous injection of carriers, an excess of carrier density will be created in the semiconductor material. For charge neutrality to exist the excess electrons “ΔN” and excess holes “ΔP” must be equal after injection ceases; the excess carrier density has been found to decrease exponentially with time with a characteristic life time = (− ⁄ ) (1.1) Where ΔNo = initial excess carrier concentration = recombination life time.

The recombination life time consist of both radiative (τrad) and non-radiative (τnon) life time.

Light Emitting Diodes (LEDs)

Light emitting diodes come under the category of photonic devices which convert electrical energy into optical radiation. LEDs are narrow-band light sources based on semiconductor materials, with emitted wavelengths ranging from the infrared to the ultraviolet. The first LEDs were studied and fabricated during the 1950s and 1960s in several laboratories. They emitted light at different wavelengths, from the infrared to the green.

Working:

When a p – n junction is forward biased, a resultant current flows across the junction between the p and n regions with two components: holes are injected from the p-region into the n region and electrons are injected from the n region into the p region. This minority-carrier injection across the junction perturbs the carrier density distribution from its equilibrium condition. The injected minority carriers undergo recombination with majority carriers until thermal equilibrium is re-established [Figure 1]. As long as the current continues to flow, the minority carrier injection continues. On both sides of the junction, a new steady-state carrier distribution is established such that the recombination rate becomes equal to the injection rate. Recombination of minority-carriers is not an instantaneous process and requires fulfilment of conservation laws. Both, energy and momentum conservation have to be satisfied. Energy conservation can be readily established because a photon can take up the energy of the recombining electron-hole pair, but the photon doesn’t contribute anything for conservation of momentum. Thus, an electron can only combine with a hole of practically identical and opposite momentum. The emission wavelength is governed by the band gap of the concerned semiconductor and is given by

Key requirements for fabrication of LED:

Direct Band gap semiconductors are advantageous for the fabrication of electroluminescent devices because radiative process is a first-order transition process with no phonon contribution. Quantum efficiency is therefore much higher in case of materials with direct band gap compared to indirect band gap materials.

For pn-junction LEDs, the wavelength of the emitted light depends on the bandgap of the material. For narrow-bandgap materials, the wavelength is in the infrared region, and for wide-bandgap materials, the wavelength is in the visible and ultraviolet spectral regions. For fabricating visible LEDs, the semiconductor materials to be employed must have an energy band gap ( ) greater than 1.8 eV.

In case of direct band gap semiconductors, such as GaAs and GaAs1-xP (x≤ 0.45), the electrons at the bottom of the conduction band and holes at the top of the valence band have equal momentum [Figure 2]. However, in case of indirect band gap semiconductors like GaAs1-xP with x > 0.45 and GaP, contribution from phonons or other scattering agents is very crucial for radiative transition. The conduction band minima and valence band maxima in such materials do not lie at the same momentum value. Therefore, special recombination centres are incorporated in the host material to facilitate radiative recombination [Figure 3].

In GaAsP (x>0.45), nitrogen (N) impurities are introduced to create recombination centres in the forbidden gap. The outer electronic structure of N is similar to phosphorus (P), and thus replaces it at its lattice sites. Such an incorporation results in the formation of isoelectronic trap centres near the conduction band. Another isoelectronic trap in GaP is formed by ZnO pairs (Zn on a Ga site and O on a P site). The ZnO trap is deeper than the N trap, resulting in longer wavelength emission in the red region of the spectrum [Figure 4].

Figure 4: Formation of excitons (bounded electron-hole pairs) by the addition of iso-electronic dopants such as N and ZnO to an indirect semiconductor. The excitons have a high probability to recombine radiatively.

2.1 Loss mechanisms in LEDs

The photons generated at the junction are emitted in all directions, however, only a fraction of electrons can reach the external environment. There are three phenomenons due to which the quantity of extracted photons decreases i.e.

(a) Absorption within the LED material

Absorption losses are large in case of LEDs with substrates opaque to light (such as GasAs) where maximum percentage of emitted photons is absorbed internally.

(b) Fresnel loss

When photons pass from a medium of high refractive index n2 (semiconductor) to a medium of low refractive index n1 (air), a portion of light is reflected back to the medium interface. This loss incurred refers to Fresnel’s loss and the Reflection coefficient for normal incidence is

(c) Critical angle loss

When the emitted photons strike the surface at an angle greater than the critical angle, they undergo total internal reflection and reducing the number of photons extracted. The critical angle θc is defined by Snell’s law as

2.2 LED efficiencies

(a) Internal efficiency

The active region of an ideal LED emits one photon for every electron injected. Each charge carrier (electron) produces one light quantum-particle (photon). Thus the ideal active region of an LED has a quantum efficiency of unity. The internal quantum efficiency is defined as where, Pint refers to the optical power emitted from the active region and I is the injection current.

(b) Extraction efficiency

Photons emitted by the active region should escape from the LED die. In an ideal LED, all photons emitted by the active region are also emitted into free space. Such an LED has unity extraction efficiency. However, practically, the power emitted from the active region is not fully emitted into the free space. Some photons cannot be extracted out of a semiconductor die. Several loss mechanisms are responsible for such attenuation. For example, light emitted by the active region can be reabsorbed in the substrate of the LED, assuming that the substrate is absorbing at the emission wavelength. Light may be incident on a metallic contact surface and be absorbed by the metal. In addition, the phenomenon of total internal reflection, also referred to as the trapped light phenomenon, reduces the ability of the light to escape from the semiconductor.

The light extraction efficiency is defined as

where P is the optical power emitted into free space. The extraction efficiency can be a severe limitation for high-performance LEDs. It is quite difficult to increase the extraction efficiency beyond 50% without resorting to highly sophisticated and costly device processes.

External Quantum Efficiency

The best measure for the LED efficiency is how much light is generated for the injected current. The external quantum efficiency can be expressed as the ratio of photons out per injected current = ( )/( ) Or in other words, external quantum efficiency can be defined as

For a semiconductor material to be employed for the fabrication of LEDs, it should satisfy the following three criterions:

(a) A pathway for radiative recombination

(b) Appropriate energy band gap

LASERS

The word LASER is an acronym for “light amplification by stimulated emission of radiation.” The distinguishing characteristic of lasers is emission of strong narrow beams of monochromatic light. An important application of semiconductor lasers is the generation of the monochromatic light that carries information through optical-fiber communication systems. Similar to LEDs, the current of a forward-biased P–N junction causes recombination of the excess minority carriers, leading to light emission. The difference is that laser light is monochromatic, having resulted from the process of stimulated emission, as distinct from the spontaneous emission in the case of LEDs.

Stimulated Emission, Inversion Population and Other Fundamental Concepts

To understand the process of stimulated emission at fundamental level, it is useful to know some of the key properties of photons, which distinguish them as particles from the electrons. It is known that that not more than one electron can occupy a single electron state (commonly referred to as Pauli Exclusion Principle). Thus, if there is an electron in a particular state, the probability that another electron will get into that state is 0. As opposed to this behaviour, photons “flock” into a single state. When there are n identical photons, the probability that one more photon will enter the same state is enhanced by the factor (n + 1). The probability that an atom will emit a photon with particular energy hν is increased by the factor (n + 1) if there are already n photons with this energy. According to this property, electrons are classified as Fermi particles (being described by the Fermi–Dirac distribution), whereas the photons are classified as Bose particles and obey a different, Bose–Einstein, distribution.

Let us put mirrors at the two ends of a P–N junction that emits light due to a forward bias current. The emitted light will reflect from the mirrors, so that the intensity of the light in the direction normal to the mirrors becomes dominant. More importantly, the presence of this light with frequency ν will increase the probability of minority excess electrons falling from the conduction band down to the valence band emitting light with the same frequency ν. As the presence of more hν photons causes further increase in the emission of light with frequency ν, a chain reaction is triggered, leading to what is known as stimulated emission. Clearly, the stimulated emission can amplify a small-intensity incoming light beam to produce a large-intensity light beam. This is called optical amplification

The concentration of excess electrons in the conduction band needs to be maintained at a high level; otherwise, as the photons get out, the excess electrons will be spent and the light that was generated will die away (as shown in figure 1). This is achieved by maintaining the forward-bias current above the needed level (threshold current), so that the concentration of the minority electrons injected into the P-type region is sufficiently higher than the equilibrium level. This condition is referred to as population inversion.

In nonequilibrium case, the concentrations of electrons and holes are expressed by two separate quasi-Fermi levels (EFN and EFP). Therefore, the stimulated emission can exceed the absorption only if strong non-equilibrium concentrations of electrons and holes are maintained, such that the difference between the respective quasi-Fermi levels is larger than the energy of the emitted photons i.e. when

EFN − EFP > hν = E2 − E1 (2.1)

3.1 Typical Heterojunction Laser: Confinement of charge carriers for Lasing action

Similar to LEDs, semiconductor lasers are made of direct band gap semiconductors, typically III–V and II–VI compound semiconductors. In addition, it is practically impossible to reach the condition given by eqn. (2.1) by an ordinary P–N junction. Also, there is a need to confine the emitted light beam inside the active laser region. All these requirements can be met by using a structure with layers of different III–V and/or II–VI semiconductors—a structure with heterojunctions. Different materials have different energy gaps, creating energy-band discontinuities (offsets) at the heterojunction, such as the band offsets at the AlGaAs–GaAs N–P junction. These offsets can help to achieve the condition given by eqn. (2.1). Also, different materials have different refractive indices, which can help to achieve total beam reflection at the parallel interfaces, so that the light is confined inside the active region.

Molecular-beam epitaxy is one of the widely employed techniques to deposit one semiconductor layer over another, creating heterojunctions with precise control. The choice of the semiconductor that will emit the light (from the active layer) depends on what light frequency (colour) is required.

Figure 2 (a) shows a laser with GaAs as the active layer, as an illustration of a typical semiconductor laser. To create appropriate energy-band discontinuities, layers with wider energy gaps are needed at each side of the active layer. In this example, N-type and P-type AlGaAs are used for this purpose. Let us concentrate on the N-AlGaAs–P-GaAs heterojunction. If the doping of N-AlGaAs is high enough, the electron quasi-Fermi level (EFN) is close enough to the bottom of the conduction band that the conduction-band offset brings the bottom of the conduction band in the GaAs region below EFN. Analogously, the valence-band offset at the P-AlGaAs–P-GaAs heterojunction places EFP below the top of the valence band. This provides the condition for population inversion when appropriate forward bias (VF) is applied. The emitted light is reflected backward and forward by the mirror surfaces capturing enough photons to trigger the stimulated emission. The photons that get through the partially transparent mirror (the useful laser beam) have to be replaced by new electrons and holes provided by the laser forward current (IF ).

The conduction-band offsets at both heterojunctions (N-AlGaAs–P-GaAs and PGaAs– P-AlGaAs) are important. They create a potential well that traps the minority carriers (electrons) so that they cannot diffuse away from the active region, as would happen in the case of an ordinary P–N junction (shown in figure 8(b)). This carrier confinement is important for maintaining the population inversion and maximizing the stimulated recombination and light emission. On the other hand, light (photons) confinement is achieved by the fact that the refractive index of AlGaAs is smaller than the refractive index of GaAs. The outside GaAs layers, the P-type capping layer and the N-type substrate layer, also create heterojunctions with the adjacent P-type and N-type AlGaAs layers. These also lead to energy-band offsets; since, however, the associated depletion layers are narrow because of the high doping, they do not present a practical problem for the current flow.

In a Double-Heterojunctions (DH) Laser, the light is confined and guided by the dielectric waveguide. The confinement factor T, may be defined as

Confinement (Guided) factor T ≈ 1- exp [-cΔηd] (2.2)

Where c is a constant, Δη is the difference in refractive index, d is the active layer thickness. Thus, T will be large when Δη increases or d increases.

Threshold Current Density (Jth) can be defined as the minimum current density (Jth) required for lasing action to occur. Jth is higher for homo-junction laser and lower for DH Laser. Also Jth increases slowly with temperature for DH laser in comparison to that for homo-junction Laser. Thus, DH laser can operate continuously at room temperature. Both carrier and optical confinement is poor in GaAs (Homo) junction Laser. Both carrier and light are confined in narrow region in DH laser which leads to enhanced stimulated emission and low value of Jth.

Summary:

Concept of optoelectronic devices and electroluminescence in semiconductors (radiative and non radiative transitions)
Homojunction and heterojunctions mentioning their advantages and disadvantages are discussed.
Basic working of Light Emitting Diodes (LED) has been discussed along with the essential requirements for the fabrication of LED. The different types of losses have been discussed with their brief mechanism.
Fundamental principle of LASER working is explained
The structural details of a typical Heterojunction Laser are discussed and the concept of charge carrier confinement is discussed.