New book - 'Building a Better World in Your Backyard' - on Kickstarter (sponsored friend)

InGaN photovoltaics literature review

From Appropedia
Jump to: navigation, search

Sunhusky.png By Michigan Tech's Open Sustainability Technology Lab.

Wanted: Students to make a distributed future with solar-powered open-source 3-D printing.
Contact Dr. Joshua Pearce - Apply here

MOST: Projects & Publications, Methods, Lit. reviews, People, Sponsors, News
Updates: Twitter, Instagram, YouTube

OSL.jpg


This is in a series of literature reviews on InGaN solar cells, which supported the comprehensive review by D.V.P. McLaughlin & J.M. Pearce, "Progress in Indium Gallium Nitride Materials for Solar Photovoltaic Energy Conversion"Metallurgical and Materials Transactions A 44(4) pp. 1947-1954 (2013). open access
Others: InGaN solar cells| InGaN PV| InGaN materials| InGan LEDs| Nanocolumns and nanowires| Optical modeling of thin film microstructure| Misc.

Contents

InGaN photovoltaics.[edit]

Progress in Indium Gallium Nitride Materials for Solar Photovoltaic Energy Conversion [1][edit]

  • Comprehensive 2013 review

Abstract: The world requires inexpensive, reliable, and sustainable energy sources. Solar photovoltaic (PV) technology, which converts sunlight directly into electricity, is an enormously promising solution to our energy challenges. This promise increases as the efficiencies are improved. One straightforward method of increasing PV device efficiency is to utilize multi-junction cells, each of which is responsible for absorbing a different range of wavelengths in the solar spectrum. Indium gallium nitride (In x Ga1−x N) has a variable band gap from 0.7 to 3.4 eV that covers nearly the whole solar spectrum. In addition, In x Ga1−x N can be viewed as an ideal candidate PV material for both this potential band gap engineering and microstructural engineering in nanocolumns that offer optical enhancement. It is clear that In x Ga1−x N is an extremely versatile potential PV material that enables several known photovoltaic device configurations and multi-junctions with theoretic efficiencies over 50 pct. This potential is driving immense scientific interest in the material system. This paper reviews the solar PV technology field and the basic properties of In x Ga1−x N materials and PV devices. The challenges that remain in realizing a high-efficiency In x Ga1−x N PV device are summarized along with paths for future work. Finally, conclusions are drawn about the potential for In x Ga1−x N photovoltaic technology in the future.

Fabrication and characterization of InGaN p-i-n homojunction solar cell[2][edit]

Abstract: InxGa1-xN p-i-n homojunction solar cells with different In content are studied. The measured open circuit voltages (Voc) are 2.24, 1.34, and 0.96 V, for x = 0.02, 0.12, and 0.15, respectively. By comparing the x-ray rocking curves, the I-V characteristics and the external quantum efficiencies, it’s demonstrated that the deterioration of InGaN crystal quality for larger In contents causes the decrease of Voc. The result demonstrates that reduction of defect is a key factor in the fabrication of nitride solar cell.

  • High absorption efficiency of InGaN at the band edge makes it beneficial in absorbing light with minimal thickness of layer at the nano level
  • X-ray diffraction shows that with increment of In content the FWHM of the intensity spectrum broadens. It can be attributed to the increment of defects due to the enhanced lattice mismatch and thermal mismatch between the InGaN and GaN
  • IV curve illustrates that the FF for different In concentration ranges from 64%-69%
  • Short circuit current increases with the increase of In content- can be attributed to the enhanced absorption of the incident light due to band gap reduction of InGaN. On the other hand, open circuit voltage decreases significantly with the increment of In content (similar outcome obtained for the heterojunction structure) - due to degradation of crystalline quality. Indium content is dominating factor in determining Voc
  • With higher Indium content the dark current rises which indicates higher level of defects and dislocations. The larger reversed bias leakage current is the evidence of deterioration at crystalline quality
  • Peak EQE decreases with Indium rise. The EQE curve for different composition of InGaN shows that the absorption edge of incident light is shifted to longer wavelengths with higher In incorporation –characterizes that more incident light can be absorbed by the device and indicates high prospect


Nearly lattice-matched n, i, and p layers for InGaN p-i-n photodiodes in the 365–500 nm spectral range[3][edit]

Abstract: Author(s) report on nearly lattice-matched grown InGaN based p-i-n photodiodes detecting in the 365–500 nm range with tunable peak responsivity tailored by the i-layer properties. The growth of lattice matched i- and n-InGaN layer leads to improvement in the device performance. This approach produced photodiodes with zero-bias responsivities up to 0.037 A/W at 426 nm, corresponding to 15.5% internal quantum efficiency. The peak responsivity wavelength ranged between 416 and 466 nm, the longest reported for III-N photodiodes. The effects of InN content and i-layer thickness on photodiode properties and performance are discussed.

  • A thick i layer increases the probability of absorption of incoming photons in active region; however, increased thickness leads to higher series resistance, surface roughness, and can affect carrier collection efficiency. The p- and n-InGaN layers should be as thin as possible to minimize absorption of photons in these regions. Bandgaps of doped layers were adjusted so that absorption of photons takes place inside or within the diffusion length distance of the active region, i.e., Egi < Eg</sub>n,p.
  • bandgap properties of the i-layer determine the cutoff and peak responsivity wavelengths in a p-i-n photodiode.
  • deterioration of photodiode performance can be attributed to high defect density in the i-layer instigated by higher InN content and thicker films.
  • peak responsivity wavelength ranged between 416 and 466 nm. This is the longest peak responsivity wavelength reported for III-N based photodetectors.
  • Maximum responsivity has been observed for 0.1 um thick i-layer and an InN composition of x=0.07. The relation between photocurrent and incident light intensity/optical power is linear indicates that the responsivity is persistent with changing light intensity
  • Obtained External Quantum Efficiency (EQE) = 11% and Internal Quantum Efficiency (IQE) = 15.5%
  • Wafers had surfaces with metallic appearance after growth, attributable to In metal segregation at low growth temperatures


InGaN quantum dot photodetectors[4][edit]

Abstract: Nanometer-scale InGaN self-assembled quantum dots (QDs) have been prepared by growth stop during the metalorganic chemical vapor deposition growth. With a 12 s growth interruption, author(s) successfully formed InGaN QDs with a typical lateral size of 25 nm and an average height of 4.1 nm. The QDs density was about 2x1010 cm-2. Nitride-based QD metal–semiconductor–metal (MSM) photodetectors were also fabricated. It was found that author(s) could significantly enhance the photocurrent to dark current contrast ratio of the MSM photodetectors by the use of QD structure.

  • The MSM photodetector has an ultra low intrinsic capacitance and its fabrication process is also compatible with field-effect transistor (FET)-based electronics. Thus, one can easily integrate GaN MSM photodetectors with GaN FET based electronics to realize a nitride-based optoelectronic integrated circuit (OEIC).
  • InGaN alloy inhomogeneity plays a key role in the high efficiency of nitride based LEDs grown on sapphire substrates. (conclusion drawn from references)
  • Nanoscale indium composition fluctuation due to InGaN phase separation could result in the formation of indium-rich clusters, which acts as quantum dots (QDs) (conclusion drawn from references)
  • It is a sophisticated quantum capture system and deeply localizes the charge carriers in quantum dots which eventually hinder the carriers migration toward nonradiative defects (dislocations) (conclusion drawn from references)
  • Author(s) reports the interrupted growth method in MOCVD to fabricate nano-scale InGaN self-assembled QDs.
  • Samples used in this study were grown on (0 0 0 1)- oriented sapphire (Al2O3) substrates in a vertical low-pressure MOCVD reactor with a high-speed rotation disk
  • InGaN QDs were grown by periodically growing and interrupting the deposition process for 12 s untill desired growth size was achieved.
  • The diameter of the circular QDs was in the range of 20–38 nm, with an average height of 4 nm and the density of these circular QDs was estimated to be around 2 X 1010 cm-2
  • To increase photocurrent to dark current contrast ratio of the conventional InGaN MSM photodetector metals with larger Schottky barrier height on InGaN should be used and epitaxial lateral overgrowth should be adopted to reduce the dislocation density in the sample
  • Although dark current of InGaN QD MSM photodetector was about the same as that of conventional InGaN MSM photodetector, its photocurrent was much larger. author(s) reported a photocurrent to dark current contrast ratio larger than 220 by using InGaN QD MSM photodetector. Result seems to suggest that InGaN QDs can actually enhance photo response of nitride-based photodetectors.

InGaN/GaN multiple quantum well solar cells with long operating wavelengths[5][edit]

Abstract: Author(s) report the fabrication and photovoltaic characteristics of InGaN solar cells by exploiting InGaN/GaN multiple quantum wells (MQWs) with In contents exceeding 0.3, attempting to alleviate to a certain degree the phase separation issue and demonstrate solar cell operation at wavelengths longer than previous attainments (420 nm). The fabricated solar cells based on In0.3Ga0.7N/GaN MQWs exhibit an open circuit voltage of about 2 V, fill factor of about 60%, and an external efficiency of 40% (10%) at 420 nm (450 nm).

  • Theoretical calculations has indicated that the requirements of an active material system to obtain solar cells having a solar energy conversion efficiency greater than 50% can be fulfilled by InGaN alloys with In content of about 40% (from reference [1])
  • The realization of high crystalline quality InGaN films in the entire composition range is highly challenging. One of the biggest problems is attributed to the large lattice mismatch between InN and GaN, resulting in low solubility and phase separation. [2],[3]
  • By directly depositing on GaN or AlN epitemplates without buffer layers, single phase InGaN epilayers across the entire alloy range could be produced by MOCVD technique. XRD data for several representative InxGa1−xN epilayers shows that InN peak positions doesn't show multiple peaks- implying that the InxGa1−xN epilayers are not phase separated
  • trend of reduced crystalline quality with increasing x makes the realization of solar cells based on InxGa1-xN with x>0.25 highly challenging.
  • The thickness of p-GaN �(n-GaN�) is �~150nm ��(~0.5u�m)�and the device structure was grown on a GaN epilayer(3u�m)/sapphire template
  • For quantum efficiency versus excitation wavelength characterization, monochromatic illumination was obtained by using white light source in conjunction with a monochromator (with a spectral resolution of about 2.5 nm)
  • strain suppress phase separation in InGaN
  • The open-circuit voltages (Voc) for devices with x ~ 0.3 and 0.4 are about 2.0 and 1.8 V, respectively.
  • Performance of In0.4Ga0.6N/GaN MQWs as active region was much poorer than that of In0.3Ga0.7N/GaN MQWs as active region, despite the fact that In0.4Ga0.6N/GaN MQWs are expected to have a much better spectral overlap with the excitation source. Reason is reduced material quality with increasing x, this leads to a higher loss of photogenerated charge carriers.
  • Obtained fill factor is about 60%
  • Device delivers a quantum efficiency of 40% at 420 nm and 10% at 450 nm.
  • The response in the shorter wavelength region (<300 nm) is limited by the use of p-GaN window and can be improved if a larger band gap material such as p-AlGaN or p-InAlGaN is incorporated
  • There are three quantum efficiency limiting factors: 1) Absorption in the semitransparent p-contact layer 2) Thin light absorption layer and 3) Low crystalline quality of InGaN alloys with relatively high In contents

High quantum efficiency InGaN/GaN solar cells with 2.95 eV band gap[6][edit]

Abstract: Author(s) report on III-nitride photovoltaic cells with external quantum efficiency as high as 63%. InxGa1−xN/GaN p-i-n double heterojunction solar cells were grown by metal-organic chemical vapor deposition on (0001) sapphire substrates with xIn = 12%. A reciprocal space map of the epitaxial structure showed that the InGaN was coherently strained to the GaN buffer. The solar cells have a fill factor of 75%, short circuit current density of 4.2 mA/cm2, and open circuit voltage of 1.81 V under concentrated AM0 illumination. It was observed that the external quantum efficiency can be improved by optimizing the top contact grid.

  • InGaN alloys have been shown to have superior high energy radiation resistance for space based PV applications
  • The III-N alloys also tend to exhibit very strong absorption of approximately 105 cm−1 at the band edge
  • The III-N crystals grown on sapphire contain relatively high densities of threading dislocations, which negatively impact GaN p-n junction device performance by increasing leakage current
  • Large lattice mismatch between InN and GaN, makes it difficult to grow good quality material with high In content due to V-pit formation and InN segregation
  • Strain relaxation can result in defect formation that can increase nonradiative recombination18 and in turn degrade solar cell performance. (from reference)
  • Monochromatic illumination for quantum efficiency measurements was supplied by coupling the Xe lamp to an Oriel 260 monochromator with a spectral linewidth of <5 nm full width at half maximum
  • Measured solar cell parameters for typical devices with 25 and 166 um grid spacing are summarized. The devices with 166 um grid spacing demonstrated high peak external quantum efficiency ne of 63% at 392 nm and a flat (±3%) quantum efficiency response from 370 to 410 nm
  • External quantum efficiency of these devices is limited by several factors: reflection at surface, absorption in semitransparent Ni/Au current spreading layer, and incomplete absorption in InGaN layer
  • An estimated peak internal quantum efficiency ni of 94% was calculated
  • By optimizing p-contact grid spacing, peak external quantum efficiency greater than 60% was achieved.
  • Quantum efficiency spectrum showed a flat spectral response from 370 to 410 nm, efficiency of III-N solar cells can be further improved by optimizing p-GaN contact such as optimizing Ni and Au thicknesses and exploring alternate contact schemes such as ITO and ZnO.
  • Short wavelength response could be enhanced by using p-AlGaN as a window layer instead of p-GaN, this could have added benefit of potentially reducing recombination of electrons at surface.
  • Additionally, an antireflection coating on top surface should also increase performance of these devices.

Photoelectric characteristics of metal/InGaN/GaN heterojunction structure[7][edit]

Abstract: A heterojunction structure photodetector was fabricated by evaporating a semitransparent Ni/Au metal film on the InGaN/GaN structure. The photocurrent (PC) spectra show that both the Schottky junction (NiAu/InGaN) and the InGaN/GaN isotype heterojunction contribute to the PC signal which suggests that two junctions are connected in series and result in a broader spectral response of the device. Secondary electron, cathodoluminescence and electron-beam-induced current images measured from the same area of the edge surface clearly reveal the profile of the layer structure and distribution of the built-in electric field around the two junctions. A band diagram of the device is drawn based on the consideration of the polarization effect at the InGaN/GaN interface. The analysis is consistent with the physical mechanism of a tandem structure of two junctions connected in series.

  • A Schottkey Contact is formed between the InGaN and Ni/Au metal determined by the I-V curve measured between Ni/Au and InGaN. Also it is confirmed by Electron Beam induced current measurements. The diode has an apparently rectifying current characteristic
  • The photoresponse of the front-side illuminated PC spectrum consists of two parts apparently. The signal at the wavelength between 250 and 360 nm comes mainly from the GaN layer (and partly from the InGaN layer) and that between 360 and about 422 nm comes from the InGaN layer
  • The slope of the PC curve at about 422 nm (between 420 and 450 nm) is not as steep as at 360 nm, indicating that the alloy composition and then the related band gap are relatively inhomogeneous in the InGaN layer
  • The signal at the wavelength longer than about 475 nm in the front-side illuminated PC spectrum is not caused by the real PC, but comes from the second harmonic effect of monochromator grating and thus will be neglected
  • The photocurrent of the GaN layer (IGaN) is much higher than that of the InGaN layer (IInGaN) in the device, i.e. a higher optical responsivity is contributed by the GaN layer than the InGaN one
  • Two types of junction are formed between layers i.e InGaN/GaN isotype heterojunction and metal/InGaN Schottky junction. Although the current of the two junctions is not well matched in an optimized way, the two junctions are in series and can produce an additive contribution to the photoresponse
  • Although the current of the two junctions is not well matched in an optimized way, the two junctions are in series and can produce an additive contribution to the photoresponse
  • The EQE under front-side illumination presents a peak value of 14% at 360 nm and it is about 0.7% at 420 nm
  • Polarization will induce a fixed sheet charge density of 2DEG or two-dimensional hole gas (2DHG) due to polarization discontinuities at the nitride heterointerfaces
  • As the 2DEG sheet concentration at the InGaN side is rather high, the depletion layer at InGaN is extremely narrow; correspondingly almost no bright region could be resolved in the EBIC image of figure
  • The intensity of EBIC decreases while the CL intensity increases along the direction from the sample surface to the substrate
  • discussed about photocurrent with respect to different excitation wavelength and overall net photocurrent.
  • discussed about polarization at the InGaN/GaN hetero-junction and band diagram of the metal/InGaN/GaN structure.
  • in wurtzite structure, III-nitride semiconductors, there are often large spontaneous and piezoelectric polarization effects. As a consequence, polarization induce a fixed sheet charge density of two-dimensional hole gas (2DHG) due to polarization discontinuities at the nitride heterointerfaces.
  • Author(s) used combined measurements of SE, CL and EBIC micro-imaging on cleaved edge face of the device to check allocation of two junctions and distribution of related built-in electric field.
  • analysis of polarization effect and band diagram of device supports the explanation of physical process of photocurrent generation in the device.
  • It is beneficial to realize the multi-junction solar cell with less tunneling junctions.

Growth and Characterization of InGaN for Photovoltaic Devices[8][edit]

Abstract: In this work author(s) present the growth and characterization of InxGa1-xN-based materials and solar cells with x up to 0.54. Growth of single phase InxGa1-xN was achieved using Plasma Assisted Molecular Beam Epitaxy (PAMBE) with flux modulation for active species. The material was characterized by x-ray diffraction, electrochemical capacitance-voltage, time-resolved photo-luminescence, and contactless electroreflectance. Fabricated devices are then studied for photo-response under simulated AM0 spectral conditions to evaluate solar cell characteristics. The dark and illuminated J-V results indicate the existence of significant shunt and series resistances arising from material defects and non-optimized device design.

  • To combat the miscibility and indium phase segregation issues, author(s) developed InGaN deposition using modulated growth conditions in order to maintain slightly metal rich growth front while avoiding the formation of indium droplets.
  • Electrochemical capacitance-voltage (ECV) measurements were used to evaluate the carrier levels in the layers. This method was chosen over Hall Effect and standard C-V measurements due to the strong band bending that is present at the InGaN surfaces, especially those with indium compositions above 35 - 40%.
  • The carrier concentration for low indium fraction n-InGaN can be calculated through typical Mott-Schottky C-V analysis, but higher fraction n-InGaN and essentially all p-InGaN require non-standard analysis.
  • Contactless electroreflectance (CER) was used to investigate the optical transitions and built-in electric fields. CER spectroscopy is similar to photoreflectance (PR) spectroscopy as it is also insensitive to localized states in the film, and therefore comparison with PL spectra can give insight into carrier localization phenomena.
  • On the Mott-Schottky representation of ECV data taken from various n-InGaN samples at 1 kHz, The lower composition samples have a linear 1/C2 vs V relationship which lends itself to standard depletion capacitance analysis regardless of doping. On the other hand, higher indium fraction layers do not behave in a linear fashion. Through theoretical modeling a strong correlation between the value of (1/C2) peak and the bulk free electron concentration can be determined. Thus the value of (1/C2) peak becomes the metric by which the n-type behavior of InGaN can be calculated.
  • Evaluation of p-InGaN also relies on the magnitude of (1/C2) peak. Due to the large acceptor energy of Mg in GaN and InGaN, low frequency C-V techniques are sensitive to the acceptor concentration as opposed to the free hole density. Therefore, authors have correlated the ECV behavior of our p-InGaN against SIMS measurements of the Mg concentration. From Mott-Schottky representation of ECV data from p-InGaN and p-GaN taken at 300 Hz, the value of (1/C2) peak decreases with increasing Mg concentration for both p-GaN and p-InGaN. The value of (1/C2) peak for a given Mg concentration is also affected by the indium composition.
  • The room temperature bandgap of the InGaN layers has been determined from CER measurements. Because of suspected carrier localization phenomena in the layers, the corresponding PL peaks were observed below the energy gap determined from CER measurements. The Stokes shift at room temperature has been found to increase from 175 to 250 meV with the decrease of the energy gap from 2.5 to 2.2 eV (i.e. an increase in indium content)
  • Room temperature time resolved photoluminescence measurements did confirmed the carrier localization phenomenon in InGaN layers. PL decay times between 5 ns and 25 ns were observed from layers with indium compositions between 31% and 41%. Author(s) observed that the decay time of PL increases from ~ 5 ns to ~ 15 - 25 ns with the increase in emission wavelength. The decay times of the PL and corresponding spectral dispersions are strong evidence for localized emission in this material. Most likely explanation for the localized emission is from small indium content fluctuations. Although the x-ray data from author(s) samples exhibited only a single peak - thus indicating essentially a single phase material - the breadths of the PL emission and the XRD FWHM’s (~600 to 900 arcsec) support the assumption of small compositional variations in the layers.
  • Photoresponse was observed for InxGa1-xN devices with x up to 0.54, corresponding to turn-on energies below 2.0 eV.
  • Photoresponse turn-on becomes more gradual with increasing indium composition. Most likely causes of this behavior are increased localized indium content fluctuations and reduction in the magnitude of the absorption coefficient. Peak internal quantum efficiencies of the order 0.20 were observed from devices spanning an In compositional range from 15 % to nearly 40%.
  • p-n device structures with indium fractions of up to 54% were fabricated and tested. Isc’s of up to 2.2 mA/cm2 were achieved for a In0.39Ga0.61N cell, but Voc was extremely small. In general, the cell's Isc typically increased with indium mole fraction, but conversely the overall device performance tended to degrade. Coupled with the reduction of Voc and IQE with increase of the cell area, the results indicate material quality issues arising from the lattice mismatch as the indium content is increased
  • Phase segregation occurs due to the high surface mobility of the indium atoms compared to the gallium atoms, the large disparity in the vapor pressures of these two metals, and the resulting narrow window of growth conditions in which reasonable properties can be achieved
  • The decay times of the PL and corresponding spectral dispersions are strong evidence for localized emission in this material. The most likely explanation for the localized emission is from small indium content fluctuations
  • Although the x-ray data from the samples exhibited only a single peak - thus indicating essentially a single phase material - the breadths of the PL emission and the XRD FWHM’s (~600 to 900 arcsec) support the assumption of small compositional variations in the layers
  • Observations- InxGa1-xN devices with x up to 0.54, corresponding to turn-on energies below 2.0 eV
  • Peak internal quantum efficiencies of the order 0.20 were observed from devices spanning an In compositional range from 15 % to nearly 40%
  • Isc’s of the tested devices generally tended to increase with indium mole fraction, the Voc and other device characteristics overall tended to degrade with increasing indium content
  • Coupled with the reduction of Voc and IQE with increase of the cell area, the results indicate material quality issues arising from the lattice mismatch as the indium content is increased


High internal and external quantum efficiency InGaN/GaN solar cells[9][edit]

Abstract: High internal and external quantum efficiency GaN/InGaN solar cells are demonstrated. The internal quantum efficiency was assessed through the combination of absorption and external quantum efficiency measurements. The measured internal quantum efficiency, as high as 97%, revealed an efficient conversion of absorbed photons into electrons and holes and an efficient transport of these carriers outside the device. Improved light incoupling into the solar cells was achieved by texturing the surface. A peak external quantum efficiency of 72%, a fill factor of 79%, a short-circuit current density of 1.06 mA/cm2, and an open circuit voltage of 1.89 V were achieved under 1 sun air-mass 1.5 global spectrum illumination conditions.

  • high measured IQE reflects efficient conversion of absorbed photons into e- and h+ and also reveals efficient transport of these carriers outside device.
  • degrading effect of the polarization fields was avoided with a high doping of the n- and p-GaN, which helps to screen the polarization-related charges at the heterointerfaces
  • improved light incoupling into the solar cells achieved by rough surface, induced during the p-GaN growth, reduces reflection of incident light at device surface and increases path length of light inside device active region.
  • surface roughness assist in achieving higher efficiency.
  • In InGaN spectral region (370<λ<410 nm), where spectral behavior of light absorption followed closely EQE curve, both in shape and magnitude. implying that nearly all absorbed photons in this layer were converted into electrons and holes.
  • generated carriers from strong light absorption in GaN region do not contribute to current generation (EQE ∼ 0); instead, they mostly recombined due to the short carrier diffusion in neutral p-GaN region
  • IQE for both smooth and rough surface solar cells determined from ratio between the EQE and light absorption curves, IQE found higher than 90% in both samples for λ from 380 to 410 nm.
  • smoother sample had slightly higher IQE, reaching values up to 97%, the rough sample had an IQE up to 93%.
  • high measured IQE revealed efficient electron-hole generation and carrier transport out of the active region.
  • absorption can be increased by making surface more rough, roughness can boost absorption upto 80 %
  • solar cell with rough surface presented a short-circuit current density of 1.06 mA/cm2, which was 27% higher than smoother sample (0.83 mA/cm2).
  • open circuit voltages were 1.89 and 1.83 V for rough and smoother samples, respectively. These devices presented high fill factors (FF), 78.6% (rough) and 76.6% (smoother), revealing excellent solar cell performances.
  • Better light coupling of solar cell with rough surface to device active region resulted in an enhancement of 35% in maximum output power
  • The lattice mismatch between InN and GaN and low temperature growth of InGaN induce impurity incorporation, point, extended, and morphological defects, which generates nonradiative recombination centers (NRCs)
  • These NRCs reduce carrier lifetimes and reduce the solar cell shortcircuit current
  • An efficient conversion of lower energy photons requires high In content InGaN, which limits the maximum thickness of this layer, reducing its light absorption
  • There are significant polarization-related charges at the InGaN/GaN interfaces which result in large electric fields in the InGaN layer. For Ga-polar p-GaN/i- InGaN/n-GaN structures, the polarization-related electric field is in opposite sense to the depletion field of the diode,12,13 and forward biasing the device further increases the electric field in the InGaN layer
  • The degrading effect of the polarization fields was avoided with a high doping of the n- and p-GaN, which helps to screen the polarization-related charges at the heterointerfaces
  • An improved light incoupling into the solar cells was achieved by rough surface, induced during the p-GaN growth, which reduced the reflection of the incident light at the device surface and increased the path length of the light inside the device active region
  • The active region was formed by a 60 nm thick InGaN grown at 880 °C, with In content ~12%, below a 300 nm thick p-type doped GaN layer. The n- and p-type GaN doping was [Si] ~6X1018 cm−3 and [Mg] ~8X1019 cm−3
  • The p-GaN in the smoother sample was grown at 955 °C (rms roughness of ~7 nm) and at 880 °C in the rougher sample (rms roughness of ~41 nm)
  • The IQE of the solar cells was assessed from the ratio between the EQE and the light absorption, which were determined by independent measurements
  • The light absorption was measured from the combination of the transmission and reflection measurements, performed in the unprocessed wafer, using a Shimadzu UV- 3600 UV-VIS-NIR spectrophotometer
  • In the GaN spectral absorbing region (λ<365 nm), the generated carriers from the strong light absorption do not contribute to current generation (EQE~0); instead, they mostly recombine due to the short carrier diffusion in the neutral p-GaN region
  • The light absorption follows closely the EQE curve, both in shape and magnitude. This means that nearly all absorbed photons in this layer were converted into electrons and holes, and these charges were efficiently separated and transported out of the device
  • The nonzero absorption observed in λ>410 nm region in the smoother sample is an artifact of the measurement. It corresponds to light propagating laterally in this nonabsorbing medium that was converted from the vertically incoming light by scattering at the surface roughness
  • The measurement artifact was verified by changing the size of the integrating sphere aperture, which consequently changes the amount of collected scattered light
  • The IQE is higher than 90% in both samples for λ from 380 to 410 nm. The smoother sample had a slightly higher IQE, reaching values up to 97%, while the rough sample had an IQE up to 93%
  • The reason for the higher IQE observed in the smoother solar cell is not fully understood and is attributed to run to run variations in the MOCVD growth. The high measured IQE revealed an efficient electron-hole generation and carrier transport out of the active region
  • One initial attempt to improve the light absorption is to increase the surface roughness of the solar cells. This reduces the reflection of the incoming light in the top surface and increases the optical path length of the light inside the solar cell due to a change in the incidence angle
  • Light absorption (at 380 nm) increased from 60 to 80 percent due to rougher surface
  • The peak EQE (average EQE curve) increased from 56% to 72%, at 380 nm, even though the IQE at this same wavelength was slightly reduced from 94% to 90%
  • The solar cell with rough surface presented a short-circuit current density of 1.06 mA/cm2, which was 27% higher than the smoother sample (0.83 mA/cm2)
  • The open circuit voltages were 1.89 and 1.83 V for the rough and smoother samples, respectively. These devices presented high fill factors (FF), 78.6% (rough) and 76.6% (smoother), revealing excellent solar cell performances
  • The better light coupling of the solar cell with rough surface to the device active region resulted in an enhancement of 35% in maximum output power
  • An enhanced light coupling scheme is required to improve the InGaN/GaN solar cell performances, which can be achieved by an improved surface texture, by antireflecting and high reflecting coatings in the top and bottom surfaces, respectively, or by coupling the incoming light to guided light using photonic crystals, which can be then fulabsorbed by the active region

Design and characterization of GaN/InGaN solar cells[10][edit]

Abstract: Author(s) experimentally demonstrate the III-V nitrides as a high-performance photovoltaic material with open-circuit voltages up to 2.4 V and internal quantum efficiencies as high as 60%. GaN and high-band gap InGaN solar cells were designed by modifying PC1D software, grown by standard commercial metal-organic chemical vapor deposition, fabricated into devices of variable sizes and contact configurations, and characterized for material quality and performance. The material was primarily characterized by x-ray diffraction and photoluminescence to understand the implications of crystalline imperfections on photovoltaic performance. Two major challenges facing the III-V nitride photovoltaic technology are phase separation within the material and high-contact resistances.

  • polarization tends to substantially influence the performance of III-V nitride devices
  • Schottky barrier at nonoptimal p-GaN–Ni metal contact interface opposes light-generated current so that current asymptotically approaches zero prior to device reaching its VOC; this Schottky effect was also confirmed through PC1D simulation.
  • major loss mechanisms are (in author(s) prototype) absorption of up to 40% of incident light by semitransparent current spreading layer and transmittance of at least 10% of incident light through the solar cell, and it was also confirmed through spectrometry.
  • quantum efficiency can be further enhanced by optimizing grid contacts for low Ohmic resistance and be brought close to unity as confirmed through simulations.
  • calculated theoretically and measured experimentally that the III-V nitrides are highly pyroelectric materials

Detailed balance modeling indicate that in order to achieve practical terrestrial photovoltaic efficiencies of greater than 50%, materials with band gaps greater than 2.4 eV are required

  • In addition to the wide band gap range, the nitrides also demonstrate favorable photovoltaic properties such as low effective mass of carriers, high mobilities, high peak and saturation velocities, high absorption coefficients, and radiation tolerance
  • To maximize absorption in the field bearing i-region of a p-i-n solar cell for maximum collection, the thickness of the top p-region is limited to 100 nm, which is just enough to provide charge for the junction and the top metal contacts
  • Typical rocking curves for the In0.05Ga0.95N/GaN solar cell epilayers measure full widths at half maximum for GaN (peak: 17.28°) at 24.5 arc sec and In0.05Ga0.95N (peak: ~17.19°) at 72.1 arc sec confirming a high crystalline quality
  • Optical characterization using room-temperature (RT) photoluminescence reveals phase separation within the material
  • Each Gaussian curve in the PL spectra indicates a separate InGaN phase with band gap corresponding to the center of that curve
  • A Schottky barrier at the nonoptimal p-GaN–Ni metal contact interface opposes the light generated current so that the current asymptotically approaches zero prior to the device reaching its VOC
  • The solar cells with semitransparent current spreading layers measure internal quantum efficiencies (IQEs) around 60% at the band edge, as shown in Fig. 4 and external quantum efficiencies around 43%. The major loss mechanisms here are absorption of up to 40% of the incident light by the semitransparent current spreading layer and transmittance of at least 10% of the incident light through the solar cell, as confirmed through spectrometry
  • Devices with interdigitated grid contacts measure a lower IQE of around 50% at the band edge due to the Schottky barrier at the p-GaN–Ni nonoptimized contact and high series resistances. Moreover, this Schottky barrier causes an increase in the cutoff voltage by 0.3 V compared to the devices with semitransparent current spreading layers
  • The III-V nitrides are highly pyroelectric materials. The large polarization charge densities present at nitride heterojunction interfaces profoundly influence electric field and mobile carrier distributions, necessitating their incorporation into device design and analysis
  • Short-period superlattice (SPS) structures provide better tunneling contacts to p-type nitrides and increase the lateral conductivity of the device


Modeling of InGaN/Si tandem solar cells[11][edit]

Abstract: Author(s) investigate theoretically the characteristics of monolithic InGaN/Si two-junction series-connected solar cells using the air mass 1.5 global irradiance spectrum. The addition of an InGaN junction is found to produce significant increases in the energy conversion efficiency of the solar cell over that of one-junction Si cells. Even when Si is not of high quality, such two-junction cells could achieve efficiencies high enough to be practically feasible. Author(s) also show that further, though smaller, improvements of the efficiency can be achieved by adding another junction to form an InGaN/InGaN/Si three-junction cell.

  • derived expression for achieving maximum efficiency by optimizing thickness for InGaN/Si tandem solar cells by using electron and hole current expression.
  • matching operating current rather than short-circuit current in multi-junctions solar cell improves the efficiency and fill factor (but operating current is usually within a few percent of short circuit current)
  • most efficient solar cells are designed such that internal reflections can increase the cell’s effective thickness by as much as a factor of 40.
  • the efficiency rapidly declines with temperature, careful heat sinking of such cells is critical.
  • gain achievable with multicrystalline or otherwise low quality Si by adding an InGaN junction on top of Si can result in increases in energy conversion efficiencies of more than 50% compared to Si alone (27% vs. 17%). Such an increase in efficiency could justify the economic cost associated with increased complexity of growing such cells.
  • At an alloy composition of In0.46Ga0.54N, the conduction band of InGaN has the same energy (relative to the vacuum) as the valence band of Si, and so a n-In0.46Ga0.54N/ p-Si interface should form a low resistance Ohmic junction
  • A two junction InGaN/Si tandem solar cell where the InGaN has an alloy fraction close to In0.46Ga0.54N can be grown without heavy doping of the interface between the two materials, as is required in multijunction cells constructed from traditional III-V materials
  • Within the layers, holes travel toward the surface of the cell and electrons travel toward the back contact of the cell
  • Authors calculation of the short circuit current density explicitly include the absorption coefficients of InGaN and Si, the junction thicknesses, and the diffusion lengths
  • To find the short circuit current density, authors first calculated the short circuit current densities in each layer (p or n) of each junction separately. The mathematical expressions and calculation method of hole and electron current density (for each layer) are discussed thoroughly
  • Maximum efficiency is achieved when the thicknesses of the n-InGaN and p-InGaN layers are adjusted so that the electron and hole currents are equal
  • Increasing the thickness increases the number of photons absorbed and therefore the number of carriers generated, but a larger fraction of the carriers are then lost to recombination processes
  • The thickness of the layers were determined mathematically so that the thicker layer is the one in which the minority carrier diffusion length is longer
  • The reverse saturation current density for each junction was calculated assuming uniform doping of the layers
  • Other than crystalline quality, the only difference between high quality and low quality InGaN is the surface recombination velocities
  • The optimal InGaN bandgap, and consequently the InGaN alloy fraction, depends on the thickness of the Si junction. Maximum efficiency is achieved when the operating currents of the InGaN and Si junctions are matched
  • As the Si thickness increases up to 20 um, its short circuit current density increases because it absorbs a larger fraction of the photons with energies larger than 1.1 eV. Therefore, one must decrease the bandgap of the InGaN junction so that it also absorbs more photons
  • As the Si thickness becomes larger than 20 um, the short circuit current density of the Si junction decreases due to the loss of carriers through recombination and thermalization and the optimal InGaN bandgap increases in order to let more photons through the InGaN junction to be absorbed by Si to compensate
  • The effective optical thickness of the Si junction was taken to be four times larger than the physical thickness
  • The fill factor of the tandem cell is roughly 10% better than that of the single junction Si cell
  • The InGaN and Si bandgaps decrease with temperature, increasing the intrinsic carrier concentration (causing J0 to increase further) and decreasing the cell voltage
  • The improvement in efficiency achieved by adding two InGaN junctions to Si as opposed to one (31%– 35%) is smaller than the improvement from adding a single InGaN junction to a bare Si solar cell (25%–31%)


Photovoltaic Effects of InGaN/GaN Double Heterojunctions With p-GaN Nanorod Arrays[12][edit]

Abstract: The p-GaN/In0.06Ga0.94N/n-GaN double heterojunctional solar cells with solely formed nanorod arrays of p-GaN have been fabricated on sapphire (0001). The p-GaN nanorod arrays are demonstrated to significantly reduce the reflectance loss of light incidence. A stress relief of the intrinsic InGaN region is observed from high-resolution X-ray diffraction analyses. The electroluminescence emission peak is blue shifted compared with the conventional solar cells. These results are reflected by the spectral dependences of the external quantum efficiency (EQE) that show a shorter cutoff wavelength response. The maximum EQE value is 55.5%, which is an enhancement of 10% as compared with the conventional devices.

  • Due to the p-type doping difficulty and crystal quality of In-rich InGaN alloys, most studies are being focused on p-i-n heterojunction solar cells using the Ga-rich intrinsic InGaN layer as the absorber
  • Although InGaN materials have a high absorption coefficient of up to 105 cm-1 at the band edge, due to the critical-thickness issue, there are still many challenges for the device design
  • The thin-film InGaN device showed an enhancement factor of 57.6% in the current density under a standard simulator with the one-sun air mass 1.5 (AM1.5) global light source
  • By optimizing the p-contact grid spacing, the short-circuit current density for the devices with a 166-μm grid spacing was 20% higher than that of the 25-μm device under a concentrated AM0 illumination
  • Group-III nitrides have a reasonably high refractive index (~2.5), which leads to a reflectance of 18% according to Fresnel’s law from reference.
  • till date, due to problems such as material selection and thermal expansion, the devices (InGaN) with antireflective coatings are not reported
  • To obtain a smooth n-GaN surface for the subsequent metal deposition as the n-type contact electrode, two-step etching sequences are used to form the mesa
  • The group-III nitride epilayers were grown on c-plane sapphire by metal–organic chemical vapor deposition
  • The material structure is composed of a 1.5-μm GaN buffer layer, a 2-μm Si-doped n-type GaN layer, a 150-nm undoped lower bandgap In0.06Ga0.94N, and a 150-nm p-type GaN layer
  • Vertically oriented nanorod arrays were prepared by the inductively coupled plasma (ICP) etching of the p-GaN film with self-assembled Ni cluster as the etching mask. The depth of etching is about 150 nm, which is approximately equal to the thickness of the p-GaN layer
  • The diameter and the density of the nanorods, which are controlled by the Ni annealing conditions, are 100–200 nm and 1X109 cm−2, respectively
  • A smoother mesa for an electrical contact with the n-GaN layer was formed by the improved two-step ICP processing of the n-GaN mesa
  • After the formation of the p-GaN nanorod arrays, the indium tin oxide was directly deposited on the top and the sidewall of the nanorods. Finally, the p-type and n-type ohmic contact layers were formed by depositing 25-nm Ti/200-nm Al/25-nm Ti/300-nm Au to complete the nanostructure solar cells (SC-II) fabrication
  • solely formed p-GaN nanostructures are proposed and explored to reduce the reflectance loss.
  • p-GaN nanorod solar cell exhibits a considerably small reflectance of up to less than 1% within the entire wavelength range
  • Since no foreign material is involved, the p-GaN nanorod arrays are intrinsically more stable and durable than dielectric coatings. Furthermore, the nanorod arrays can maintain a low reflectance at a variety of incident angles
  • The observed significant decrease in the optical reflectance of p-GaN nanorod solar cell is mainly due to the higher absorption in nanorod arrays
  • A part of the light can transmit through the nanorods or reach the layer underneath via the space between the p-GaN nanorods, which may be another reason for the decrease in reflectance
  • The indium content is determined to be 6% from the distance between the InGaN and GaN peaks in the X-ray scanning curves
  • The EL emission peaks of the InGaN layers were observed at 394 and 389 nm for p-GaN nanorod and p-GaN solar cells, respectively
  • The emission of p-GaN nanorod solar cell shows a blue-shift phenomenon, which is caused by a partial reduction of the piezoelectric field
  • The strain in the InGaN/GaN layer caused by the lattice mismatch between the GaN and InGaN layers is partially released because the nanorods can accommodate the misfit more effectively
  • vertically aligned p-GaN nanorod arrays may be useful in improving the conversion efficiency and reducing material consumption.
  • p-GaN nanorod arrays can effectively reduce optical loss in PV applications
  • p-GaN nanorod arrays applied in InGaN/GaN heterojunctional solar cells (author(s) prototype) yielded up to 55.5% peak EQE.
  • low reflectance of p-GaN nanorod arrays originated from high surface area and subwavelength scale of the nanorods, can be used effectively for their enhanced antireflection ability.
  • p-nanorod type solarcell exhibited considerably small reflectance of up to less than 1% within entire wavelength range whereas for p-GaN (planar type) showed reflectivity of 18% as determined by Fresnel’s law for air/GaN interface
  • reduction in the piezoelectric field caused by partial strain release induces a blue-shift in p-nanorod solar cell
  • This reduction in the piezoelectric field caused by the partial strain release then induces a blue-shift value of 5 nm
  • The response of p-GaN nanorod solar cells in the shorter wavelength region (330–360 nm) is enhanced, the maximum EQE value at 380 nm is 55.5%
  • The p-GaN nanorod solar cell has a substantial cost benefit if the nanorod arrays are grown by a bottom up process, such as vapor-liquid-solid growth

Improved Conversion Efficiency of GaN/InGaN Thin-Film Solar Cells[13][edit]

Abstract: In this letter, Author(s) report on the fabrication and photovoltaic characteristics of p-i-n GaN/InGaN thin-film solar cells. The thin-film solar cells were fabricated by removing sapphire using a laser lift-off technique and, then, transferring the remaining p-i-n structure onto a Ti/Ag mirror-coated Si substrate via wafer bonding. The mirror structure is helpful to enhance light absorption for a solar cell with a thin absorption layer. After the thin-film process for a conventional sapphire-based p-i-n solar cell, the device exhibits an enhancement factor of 57.6% in current density and an increment in conversion efficiency from 0.55% to 0.80%. The physical origin for the photocurrent enhancement in the thin-film solar cell is related to multireflection of light by the mirror structure.

  • crystalline defects commonly observed include v-shaped pits, phase separation, and dislocations, which have been shown to deteriorate device performance by increasing leakage current
  • conventional solar cell with a thin InGaN absorption layer exhibits smaller photocurrent than that with a thick InGaN layer does, the thin-InGaN solar cell shows better performance in open-circuit voltage, fill factor, and shunt resistance
  • Growth of high-In-content InGaN alloy on GaN typically results in the formation of an InGaN film with high defect density. The crystalline defects commonly observed include v-shaped pits, phase separation, and dislocations, which deteriorate device performance by increasing leakage current
  • The superlattice and MQWs schemes showed improved solar-cell performance, but both schemes still have the drawback that the optical absorption layer is too thin to sufficiently absorb the solar spectrum
  • GaN/InGaN p-i-n solar cells were grown on (0001) sapphire substrates by metal–organic chemical vapor deposition (MOCVD)
  • The p-i-n structure consisted of a 3-μm n-GaN bottom layer, a thin 150-nm intrinsic In0.1Ga0.9N absorption layer, and a 150-nm p-GaN top layer
  • Thin-film technique is used to remove the sapphire substrate and transfer the p-i-n structure onto a mirror-coated Si substrate
  • For enhancing light absorption, a highly reflective mirror is employed in thin-InGaN solar cell, which improved current density from 0.33 to 0.52 mA/cm2 and an increment in conversion efficiency from 0.55% to 0.80%
  • Mirror coated Si substrate has another advantage of good heat dissipation as compared with conventional sapphire-based devices
  • The high fill factor in this letter could result from negligible leakage current or large shunt resistance, which could be thought of as the result of good crystalline quality of InGaN
  • The physical origin for the photocurrent enhancement in the thin-film solar cell is related to multireflection of light by the mirror structure
  • In addition to the enhanced light absorption, the mirrorcoated Si substrate still has another advantage of good heat dissipation as compared with the conventional sapphire-based devices

High-quality InGaN/GaN heterojunctions and their photovoltaic effects[14][edit]

Abstract: High-quality p-GaN/i-In0.1Ga0.9N/n-GaN heterojunctional epilayers are grown on (0001)-oriented sapphire substrates by metal organic chemical vapor deposition. The Pendellösung fringes around the InGaN peak in high-resolution x-ray diffraction (HRXRD) confirm a sharp interface between InGaN and GaN films. The corresponding HRXRD and photoluminescence measurements demonstrate that there is no observable phase separation. The improvement in crystal quality yields high-performance photovoltaic cells with open-circuit voltage of around 2.1 eV and fill factor up to 81% under standard AM 1.5 condition. The dark current-voltage measurements show very large shunt resistance, implying an insignificant leakage current in the devices and therefore achieving the high fill factor in the illuminated case.

  • crystalline defects commonly observed include v-shaped pits, phase separation, and dislocations, which have been shown to deteriorate device performance by increasing leakage current.
  • fabricated high quality p-i-n type solar cell using MOCVD by adjusting the layer thickness using critical thickness calculations.
  • leakage current density increases with area/periphery ratio of diode, reavealing leakage current as one of main components degrading photovolatic performance
  • A high-quality p-GaN/ i-In0.1Ga0.9N/n-GaN double heterojunction with no observable phase separation and relaxation is obtained using metal organic chemical vapor deposition (MOCVD)
  • Epitaxial layers of GaN and InGaN were grown on c-plane sapphire substrates by MOCVD using the conventional two-step growth process
  • The p-i-n junction for photovoltaic effect consists of a 3 um thick bottom Si-doped n-type GaN (n-GaN), 0.15 um thick intrinsic In0.1Ga0.9N layer (i-InGaN), and 0.15 um thick top p-GaN
  • Pendellösung fringes phenomenon is frequently used to judge heterojunction quality as the figure-of-merit
  • A sharp single peak corresponding to near-band edge transition in the epitaxially grown InGaN phase appears at 393 nm (~3.15 eV) in the PL spectrum. Induced broadening of the primary InGaN peak which implies phase separation is not observed. These results confirm a good suppression of phase separation in the InGaN epilayers
  • Due to the good crystal quality and interface property in InGaN/GaN heterojunction, devices may exhibit low reversed saturation current density
  • The limited Voc in this study could be ascribed to the relatively smaller built-in potential caused by the lower doping concentration in GaN and/or less incident light due to the adsorption in the Ni/Au contact layer
  • The relatively high fill factor (81%) could result from the negligible leakage current or large shunt resistance in the fabricated solar cells. The cause of the improved shunt resistance or leakage current could be thought of as the reduction of defects such as dislocations in the bulk portion or interface of the heterojunctions
  • The devices display the poorer photovoltaic effect (especially the low fill factor, data are not shown here), which could be attributed to worse crystal quality caused by a relaxation of InGaN layer which generates dislocations (increasing leakage current and recombination) because the growth thickness exceeds the critical thickness of InGaN on GaN

Growth, fabrication, and characterization of InGaN solar cells[15][edit]

Abstract: The InGaN alloy system offers a unique opportunity to develop high efficiency multi-junction solar cells. In this study, single junction solar cells made of Inx Ga1–xN are successfully developed, with x = 0, 0.2, and 0.3. The materials are grown on sapphire substrates by MBE, consisting of a Si-doped InGaN layer, an intrinsic layer and an Mg-doped InGaN layer on the top. The I –V curves indicate that the cell made of all-GaN has low series resistance (0.12 Ω cm2) and insignificant parasitic leakage. Contact resistances of p and n contacts are 2.9 × 10–2 Ω cm2 and 2.0 × 10–3 Ω cm2, respectively. Upon illumination by a 200 mW/cm2, 325 nm laser, Voc is measured at 2.5 V with a fill factor of 61%. Clear photo-responses are also observed in both InGaN cells with 0.2 and 0.3 Indium content when illuminated by outdoor sunlight. But it is difficult to determine the solar performance due to the large leakage current, which may be caused by the material defects. A thicker buffer layer or GaN template can be applied to the future growth process to reduce the defect density of InGaN films.

  • good control of p-type doping in InGaN is one of most critical issues. Due to unusual low position of the conduction band edge at 0.9 eV below Fermi level stabilization energy (EFs), p-type doping of InGaN has proved extremely difficult. evaluation of p-type doping with Magnesium (Mg) as an acceptor still remains a challenge because strong surface accumulation of electrons exists throughout most of the Indium (In) composition range.
  • surface accumulation represents a possible significant parasitic conductivity path between p and n contacts on solar cell structures.
  • suggested that a thicker buffer layer or GaN template can be applied to reduce defect density of InGaN
  • devices with high In compositions have much higher leakage current, which cause difficulty to determine turn-on voltage and output power of cells
  • high In mole fraction alloy has a strong surface electron accumulation which can also contribute to the leakage in InGaN cells
  • leakage current density goes up as area/periphery ratio of the diode increases, which tell that bulk leakage is one of main component in leakage current in devices.
  • high defect density in InGaN materials may cause the abnormal bulk leakage current in devices

Due to the unusual low position of the conduction band edge at 0.9 eV below the Fermi level stabilization energy (EFs), p-type doping of InGaN has been proved extremely difficult

  • The surface accumulation represents a possible significant parasitic conductivity path between p and n contacts on solar cell structures
  • The buffer layer used in the design is a 250 nm AlN layer grown at about 800 °C followed by a 1 μm GaN layer
  • Due to the surface electron accumulation, most of the Mg-doped InGaN films exhibit strong n-type Hall polarity
  • The devices with high In compositions have much higher leakage current, which cause the difficulty to determine the turn-on voltage and output power of the cells
  • It is anticipated that high In mole fraction alloy has a strong surface electron accumulation which can also contribute to the leakage in InGaN cells
  • The electroluminescence peak of In0.2Ga0.8N cell overlaps with the absorption edge of the device, which indicates relative lack of localized states
  • The cell consisted of all-GaN composition has low series resistance and insignificant parasitic leakage
  • A thicker buffer layer or GaN template can be applied to the future growth to reduce the defect density of InGaN

Characteristics of InGaN designed for photovoltaic applications[16][edit]

Abstract: This work addresses the required properties and device structures for an InGaN solar cell. Homojunction InGaN solar cells with a bandgap greater than 2.0 eV are specifically targeted due to material limitations. These devices are attractive because over half the available power in the solar spectrum is above 2.0 eV. Using high growth rates, InGaN films with indium compositions ranging from 1 to 32% have been grown by Molecular Beam Epitaxy with negligible phase separation according to X-ray diffraction analysis, and better than 190 arc-sec ω-2θ FWHM for ∼0.6 μm thick In0.32Ga0.68N film. Using measured transmission data, the adsorption coefficient of InGaN at 2.4 eV was calculated as α ≅ 2×105 cm–1 near the band edge. This results in the optimal solar cell thickness of less than a micron and may lead to high open circuit voltage while reducing the constraints on limited minority carrier lifetimes.

  • suggested that InN films must be grown at low temperatures, such as 360–550 °C because of low dissociation temperature of InN
  • at low substrate temperatures, it is difficult to achieve p-type InGaN while maintaining good crystalline structures
  • InGaN has phase separation for high In composition due to the immiscibility of InN in GaN
  • phase separation also affects the bandgap of material, creating localized regions of different composition materials, seriously limiting efficiency of solar cell
  • phase separation can be suppressed by increasing growth rate, author(s) successfully grew InGaN films with various In compositions by Molecular Beam Epitaxy (MBE) at rates in excess of 0.6–1.3 μm/hr.
  • absorption coefficient of InGaN at 2.4 eV was calculated to be α ~ 2×105 cm–1, near the band edge
  • InN films must be grown at low temperatures, such as 360–550 °C because of the low dissociation temperature of InN
  • At low substrate temperatures, it is difficult to achieve p-type InGaN while maintaining good crystalline structures
  • InGaN has a well documented phase separation for high In composition due to the immiscibility of InN in GaN
  • The phase separation also affects the bandgap of the material, creating localized regions of different composition materials, seriously limiting the efficiency of the solar cell. Such localization of the electron and hole wave functions makes current collection more difficult and to date has lead to decreased conversion efficiency
  • A polarization induced discontinuity in the conduction-band edge of GaN – GaN heterojunctions is formed, which negatively affects the photocurrent of the device by preventing minority carrier collection
  • A heterojunction device has defects arising from lattice mismatch directly at the collection junction – the most sensitive region of the device
  • Phase separation can be suppressed by increasing the growth rate
  • As the III/V ratio is increased in InGaN, phase separation minimizes. The growth rate increased with increasing III/V ratio and phase separation diminished
  • Optical transmission measurements of MBE grown samples with a film thickness of ~1 um found very sharp bandgap edges, indicative of minimal phase separation
  • Relatively sharp transition for lower In composition materials but slightly less sharp transition for higher In composition InGaN, resulting from slight phase separation was observed

Growth of InGaN self-assembled quantum dots and their application to photodiodes[17][edit]

Abstract: Nanometer-scale InGaN self-assembled quantum dots (QDs) have been prepared by growth interruption during metalorganic chemical vapor deposition growth. With a 12 s growth interruption, author(s) successfully formed InGaN QDs with a typical lateral size of 25 nm and an average height of 4.1 nm. The QD density was about 2×1010 cm−2. In contrast, much larger InGaN QDs were obtained without growth interruption. InGaN metal-semiconductor-metal photodiodes with and without QDs were also fabricated. It was found that the QD photodiode with lower dark current could operate in the normal incidence mode, and exhibit a stronger photoresponse.

  • Author(s) could achieve a much larger photocurrent to dark current contrast ratio from MSM photodiodes with nanoscale InGaN SAQDs

III–V nitride semiconductor materials have a wurtzite crystal structure

  • At room temperature, the band gap energy of AlInGaN varies from 0.7 to 6.2 eV depending on its composition
  • MSM photodiodes have an ultra low intrinsic capacitance and their fabrication process is also compatible with field-effect transistor (FET)-based electronics
  • Nitride quantum dots (QDs) can be self-organized using the strain-induced Stranski–Krastanov growth mode and also by applying growth interruption during metalorganic chemical vapor deposition (MOCVD) growth
  • Small circular InGaN self assembled quantum dots (SAQDs) were formed by the interrupted growth mode with diameter in the range of 20–38 nm, average height of 4.1 nm and density to be around 2X1010 cm-2
  • Large oval InGaN islands were found in sample grown without growth interruption with width in the range of140 nm-70 nm, average height of 1.7 nm density about 3.5X108 cm-2
  • The surface morphology of MOCVD grown InGaN can be varied by using interrupted growth rate
  • Growth interruption could release the partial strain energy of InGaN epitaxial layer when InGaN was grown more than the critical thickness 0.8 nm
  • The introduction of growth interruption resulted in a PL blueshift as large as 67 meV that can be attributed to the lateral size effect of nanostructures
  • The subband transition energies increase when the size of nanostructure becomes smaller
  • The QW (2.4nm thick InGaN) detectors are not sensitive to radiation that is incident perpendicularly, while the QD detector can operate in the normal incidence mode showing better photoresponse
  • A larger photoresponse was observed from photodiodes with In- GaN nanostructures

InGaN/GaN multiple quantum well concentrator solar cells[18][edit]

Abstract: Author(s) present the growth, fabrication, and photovoltaic characteristics of Inx Ga1−xN/GaN(x ~ 0.35) multiple quantum well solar cells for concentrator applications. The open circuit voltage, short circuit current density, and solar-energy-to-electricity conversion efficiency were found to increase under concentrated sunlight. The overall efficiency increases from 2.95% to 3.03% when solar concentration increases from 1 to 30 suns and could be enhanced by further improving the material quality.

  • requirements of an active material system for obtaining solar cells with a conversion efficiency greater than 50% can be fulfilled by InGaN alloys with In-content of about 40%.
  • large lattice mismatch between InN and GaN, results in phase separation and as a consequence reported values of open circuit voltages (Voc) for different In contents in general are significantly lower than theoretical values (thermodynamic limit)
  • strain could suppress phase separation in InGaN
  • used strain to suppress phase separation for In(x)Ga(1-x)N (x~0.3) based multiple quantum well
  • advantages of low dimensional InGaN MQW solar cells include (i) crystalline quality of thin light absorption layers (InGaN wells) embedded between GaN barriers is higher than that of InGaN epilayers with thickness exceeding the critical thickness, (ii) with incorporation of MQW structure in the i-region, Voc and Jsc can be independently optimized. (iii) MQW solar cells are expected to outperform bulk i-layer solar cells under concentrated sunlight
  • described Voc as a function of concentrated sun light by mathematical expression

Lower Voc values in InGaN solar cells with higher In contents are not only caused by the lowering of the band gaps but are also related to reduced crystalline quality By directly depositing on GaN or AlN epitemplates without buffer layers, single phase InGaN epilayers across the entire alloy range can be produced by MOCVD

  • Strain can suppress phase separation in InGaN
  • In the MQW model the crystalline quality of the thin light absorption layers (InGaN wells) embedded between GaN barriers is higher than that of InGaN epilayers with thickness exceeding the critical thickness. Also in the i-region, Voc and Jsc can be independently optimized
  • Voc is primarily determined by the wider band gap barrier material while spectral response is determined by the width and depth of the lower band gap material of QWs
  • MQW solar cells are expected to outperform bulk i-layer solar cells under concentrated sunlight
  • The overall solar to electrical power conversion efficiency of the device is 2.95%, which is still much lower than the theoretically expected value of a single junction solar cell of about 8% at the experimented optical energy band gap can be attributed to the insufficient thickness of the light absorbing layer in InGaN wells
  • The decrease in FF with increasing C is related to the enhanced carrier recombination at the interface region due to high carrier densities under concentrated sunlight

Optimization of GaN window layer for InGaN solar cells using polarization effect[19][edit]

Abstract: The III-nitride material system offers substantial potential to develop high-efficiency solar cells. Theoretical modeling of InGaN solar cells indicate strong band bending at the top surface of p-InGaN junction caused due to piezoelectric polarization-induced charge at the strained p-GaN window interface. A counterintuitive strained n-GaN window layer is proposed, modeled and experimentally verified to improve performance of InGaN solar cells. InGaN solar cells with band gap of 2.9 eV are grown using MOCVD with p-type and n-type strained GaN window layers, and fabricated using variable metallization schemes. Fabricated solar cells using n-GaN window layers yield superior VOC and FF compared to those using p-GaN window layers. The VOC's of InGaN solar cells with n-GaN window layers are further enhanced from 1.5 V to 2 V by replacing the conventional NiOX top contact metal with Ti/Al, which also verifies the tunneling principle.

  • net polarization and consequent internal electric fields have detrimental effect on performance of optoelectronic devices due to polarization-induced potential barriers and band bending.
  • demonstrated counterintuitive design, fabrication and optimization of the GaN window layer for InGaN solar cells mediated by polarization effects
  • thin GaN window layers, designed in a 2 – 10 nm thickness range, are typically strained due to lattice mismatch with the underlying InGaN layer and hence, generate substantial piezoelectric polarization.
  • simulations indicated that tunneling contacts using n-type material can potentially provide superior Ohmic characteristics to p-type GaN or InGaN contacts.
  • due to lower resistivity and ease of forming Ohmic contacts to n-GaN compared to p-GaN, tunneling contact using n-type material indicates a viable alternative
  • Strained n-GaN window layers enhance tunneling of holes from the p-InGaN junction due to piezoelectrically induced sheet charge and strong band bending at heterointerface
  • Fabricated InGaN solar cells with band gap of 2.9 eV and n-GaN window layer demonstrate a superior performance compared to those with p-GaN window layers
  • The III-nitrides are highly polar molecules due to non-centrosymmetry of charge in the wurtzite structure and the large ionicity of the covalent bonds
  • The net polarization and consequent internal electric fields have been shown to be detrimental to the performance of optoelectronic devices due to polarization-induced potential barriers and band bending
  • Window layers serve to passivate the top junction surface and generate a front-surface field to minimize front surface recombination
  • Thin GaN window layers, designed in a 2 - 10 nm thickness range, are typically strained due to lattice mismatch with the underlying InGaN layer and hence, generate substantial piezoelectric polarization
  • The downward bending of bands at the strained-p-GaN/p- InGaN heterointerface tends to develop a potential well that accumulates a 2-Dimensional Electron Gas (2DEG)
  • InGaN solar cells with p-GaN window layers demonstrate very low Open-Circuit Voltages (Voc) and Fill Factors (FF)
  • The combination of strong polarization charge at the p-GaN/p-lnGaN interface with a non-ideal NiOx/p-GaN contact substantially reduces the Voc and FF of the solar cell

Simulation of In0.65Ga0.35N single-junction solar cell [20][edit]

Abstract: The performances of In0.65Ga0.35N single-junction solar cells with different structures, including various doping densities and thicknesses of each layer, have been simulated. It is found that the optimum efficiency of a In0.65Ga0.35N solar cell is 20.284% with 5 × 1017 cm−3 carrier concentration of the front and basic regions, a 130 nm thick p-layer and a 270 nm thick n-layer.

  • When carrier concentrations of front and basic regions are 5×1017 cm−3, thickness of p-layer and n-layer are 130 nm and 270 nm, respectively, optimum efficiency calculated is 20.284% (AM1.5G, 100mWcm−2, 0.32–1.32μm)
  • summarized expression for efficiency, minority carrier, diffusion length, open circuit voltage short-circuit current and other experimentally measured properties of GaN and InGaN
  • Authors used the analysis of microelectronic and photonic structures (AMPS) to analyses InGaN PV cells. It uses the first-principles of continuity and Poisson’s equations approach to analyze the transport behaviour of semiconductor electronic and optoelectronic device structures

Photoconductivity in nanocrystalline GaN and amorphous GaON[21][edit]

Abstract: In this work Author(s) present a study of the optoelectronic properties of nanocrystalline GaN (nc-GaN) and amorphous GaON (a‐GaON) grown by ion-assisted deposition. The two classes of film show very distinct photoconductive responses; the nc-GaN has a fast small response while the a‐GaON films have a much larger response which is persistent. To describe the observed intensity, wavelength, and temperature dependence of the photoconductivity in each class of film, Author(s) build a model which takes into account the role of a large density of localized states in the gap. The photoconductivity measurements are supplemented by thermally stimulated conductivity, measurement of the absorption coefficient, and determination of the Fermi level. Using the model to aid our interpretation of this data set, Author(s) are able to characterize the density of states in the gap for the two materials.

  • Polycrystalline GaN has recently been shown to be capable of blue light emission
  • Morphous GaN can be stabilized into an amorphous form during growth by incorporating at least 12 at. % of O- accomplished by depositing in the presence of a carefully controlled partial pressure of water vapor
  • The release of carriers during an anneal permits an estimate of the energy dependence of the trap depth distribution
  • The temperature dependence during the sample cooling allows determination of the depth of the Fermi level in the dark configuration
  • Covalent materials, even in a strongly disordered configuration can have a clear gap between valence and conduction bands. They commonly have a significant density of localized carrier states between the bands of extended states, inside what is called the mobility gap
  • The interplay of trapping, thermal detrapping, and recombination from extended states and/or tail states ultimately determines the behavior of the photoconductivity
  • Oxygen incorporation at high concentrations effectively changes the band structure of the material, and does not act as an impurity which introduces states in the gap
  • The position of the Fermi level was determined from the temperature dependence of the dark conductivity and it lies at about 1 eV away from the mobility edge
  • The distribution of localized states in the gap was found via spectral measurements: optical absorption, spectral photoconductivity, and TSC. Experimental results support that it is more likely that the Fermi level simply lies somewhere within the band tails
  • At low energies the absorption coefficient in the GaON continues to drop away sharply, while in the nc-GaN film there is a clear change in slope at these low energies
  • The recombination and trapping rates are proportional to the densities of carriers in the conduction band multiplied by the density of recombination centers and the density of unoccupied traps
  • The trapping/detrapping rates are larger than the recombination rate, resulting in quasithermal equilibrium between the free and trapped electrons
  • nc-GaN film exhibits a weak photoconductivity (approximately 50 fA at 9 V, corresponding to 10−7 A/W) which is fast, switching on and off
  • GaON film has a response several orders of magnitude greater (several nanoamperes at 9 V, corresponding to 10−2 A/W) which is very slow, decaying nonexponentially over the course of hours, i.e., showing PPC (persistent photoconductivity)
  • By examining intensity dependence of photoconductivity under 325 nm illumination, photoresponse was found to be linear in generation rate in nc-GaN films and had a squareroot dependence in GaON films.
  • Spectral dependence of photoconductivity showed an exponential drop-off in combined densities of states of conduction- and valence-band tails with decreasing energy for both types of film.
  • Author(s) developed a model which considers the effect of a large density of localized states in the gap to describe the photoconductive behavior observed in these two types of films

Theoretical possibilities of InxGa1−xN tandem PV structures[22][edit]

Abstract: Author(s) designed a model of InxGa1−xN tandem structure made of N successive p–n junctions going from two junctions for the less sophisticated structure to six junctions for the most sophisticated. Author(s) simulated the photocurrent density and the open-circuit voltage of each structure under AM 1.5 illumination in goal to optimize the number of successive junctions forming one structure.

For each value of N, Author(s) assumed that each junction absorbs the photons that are not absorbed by the preceding one. From the repartition of photons in the solar spectrum and starting from the energy gap of GaN, Author(s) fixed the gap of each junction that gives the same amount of photocurrent density in the structure. Then Author(s) calculated the current density accurately and optimized the thicknesses of p and n layers of each junction to make it give the same output current density. The evaluation of ni: the intrinsic concentration permitted to calculate the saturation current density and the open-circuit voltage of each junction. Assuming an overall fill factor of 80%, Author(s) divided the output peak power by the incident solar power and obtained the efficiency of each structure.

The numerical values for InxGa1−xN were taken from the relevant literature. The calculated efficiency goes from 27.49% for the two-junction tandem structure to 40.35% for a six-junction structure. The six-junction InxGa1−xN tandem structure has an open-circuit voltage of about 5.34 V and a short circuit current density of 9.1 mA/cm2.

  • short-circuit current density of a tandem cell Jsc is given by the least of the photocurrent densities produced by the junctions of the tandem cell.
  • The open-circuit voltage of a tandem cell is is calculated to be equal to the sum of the open-circuit voltages of the tandem junctions.
  • Simulations show that six-junctions InxGa1−xN tandem cell could reach an efficiency of more than 40% with a short-circuit current density of 9.1 mA/cm2 and an open-circuit voltage of 5.3 V.
  • output power saturates when the number of junctions increases. More than six junctions would not give more power proportional to the complication of adding more cells.
  • achievable short-circuit current density decreases as the tandem cell contains more layers, however, the increase of the open-circuit voltage as a function of the number of junctions is almost linear; hence, it compensates the decrease of the short-circuit current density, which is a function of the inverse of the same variable; this explains the increase of the output maximum power and the cell efficiency with the number of junctions.
  • The used materials in tandem solar cells should have some similar properties like the thermal expansion coefficient, the electron affinity and the lattice mismatch
  • Assuming a perfect quantum response of the materials and equal photocurrent densities for all the junctions of the tandem cell the energy gaps of the InxGa1-xN alloys were identified
  • The bowing parameter is composition dependent
  • The short-circuit current density is given by the least of the photocurrent densities produced by the junctions of the tandem cell
  • A method for calculating the open circuit voltage and short circuit current density of individual and therefore all the junctions of tandem solar cells has been proposed
  • The open-circuit voltage produced by the junctions of the tandem cell decrease almost linearly from the top to the bottom and is mainly produced by the first three junctions- can be attributed to the increase of saturation current density
  • The logarithm of the saturation current density increases almost linearly from the top to the bottom
  • The output power saturates when the number of junctions increases

The impact of piezoelectric polarization and nonradiative recombination on the performance of (0001) face GaN/InGaN photovoltaic devices[23][edit]

Abstract: The impact of piezoelectric polarization and nonradiative recombination on the short-circuit current densities (Jsc) of (0001) face GaN/InGaN photovoltaic devices is demonstrated. P-i-n diodes consisting of 170 nm thick intrinsic In0.09Ga0.91N layers sandwiched by GaN layers exhibit low Jsc ~ 40 μA/cm2. The piezoelectric polarization at the GaN/InGaN heterointerfaces creates drift currents opposite in direction needed for efficient carrier collection. Also, nonradiative recombination centers produce short carrier lifetimes, limiting Jsc. Alternative structures with intrinsic InGaN layers sandwiched by n-type InGaN or graded InyGa1−yN (y = 0–0.09) layer and a p-type In0.015Ga0.985N layer have favorable potentials, longer carrier lifetimes, and improve Jsc to ~ 0.40 mA/cm2.

  • NRCs or Hall–Shockley–Read recombination centers can also affect the performance of PV devices, with an increase in NRCs reducing the carrier lifetimes.
  • structural differences such as doping, layer thickness and composition, and growth conditions affect piezoelectric polarization and NRC formation within the structure.
  • structures consisting of n-type and p-type InGaN layers sandwiching intrinsic InGaN layer are used to control the piezoelectric polarization in InGaN/GaN structures but other methods could be used such as delta doping to reduce polarization fields, growth on non-(0001) substrates, or growth of low dimensional materials such as nanowires
  • InGaN is typically grown strained on GaN and the lattice-mismatch strain results in the formation of defects in the InGaN layer that may act as nonradiative recombination centers
  • InGaN layer growth is typically performed at low growth temperatures to incorporate sufficient indium, resulting in increased incorporation impurities or other point and V-defect related NRCs
  • Trimethylgallium, trimethylindium, and ammonia, are used for the Ga, In, and N sources in H2 and N2 carrier gas, and SiH4 and bis-methylcyclopentadienylmagnesium are used for n-type (Si) and p-type (Mg) doping
  • Structural differences such as doping, layer thickness and composition, and growth conditions affect the piezoelectric polarization and NRC formation within the structure
  • Piezoelectric polarization and nonradiative recombination affect the performance of (0001) face GaN/InGaN heterojunction p-i-n PV devices
  • Placing InGaN layers above and below the intrinsic InGaN layer reduces the piezoelectric polarization and nonradiative recombination centers and increases Jsc

Superior radiation resistance of In1−xGaxN alloys: Full-solar-spectrum photovoltaic material system[24][edit]

Abstract: High-efficiency multijunction or tandem solar cells based on group III–V semiconductor alloys are applied in a rapidly expanding range of space and terrestrial programs. Resistance to high-energy radiation damage is an essential feature of such cells as they power most satellites, including those used for communications, defense, and scientific research. Recently author(s) have shown that the energy gap of In1−xGaxN alloys potentially can be continuously varied from 0.7 to 3.4 eV, providing a full-solar-spectrum material system for multijunction solar cells. We find that the optical and electronic properties of these alloys exhibit a much higher resistance to high-energy (2 MeV) proton irradiation than the standard currently used photovoltaic materials such as GaAs and GaInP, and therefore offer great potential for radiation-hard high-efficiency solar cells for space applications. The observed insensitivity of the semiconductor characteristics to the radiation damage is explained by the location of the band edges relative to the average dangling bond defect energy represented by the Fermi level stabilization energy in In1−xGaxN alloys.

  • primary cause for the degradation of solar cells in space are due to bombardment by protons and electrons in the energy range of electron volts to hundreds of million electron volts.
  • solar cell degradation in a space radiation environment can be successfully modeled using displacement damage dose methodology.
  • PL intensities of GaAs and Ga0.51In0.49P are drastically suppressed by the irradiation whereas the PL signal of InN does not show any reduction in intensity after being subjected to a similar radiation dose.
  • in general, nitrides are less sensitive to the irradiation as compared to GaAs and GaInP. GaN and Ga-rich InGaN materials maintain excellent optical properties even in the presence of high densities of structural defects.
  • No strong reduction in PL intensity is observed in InN and In1–xGaxN with small Ga fractions. However, it is interesting to note the trend that as the Ga fraction increases, the irradiation-induced degradation in optical emission becomes increasingly severe.
  • radiation sensitivity of GaAs and GaInP is far above that of InN, GaN, and InGaN alloys.
  • insensitivity of optical properties of InGaN alloys to radiation damage is a good indicator of expected radiation hardness of photovoltaic devices made with these n-type alloys.
  • Compared with conventional solar cell materials (GaAs and GaInP), InN and In1–xGaxN alloys show much less deterioration in their optical and transport properties. The different behaviors are explained in terms of the different energy configurations of the irradiation-induced defect states in the band diagram of the host materials.
  • superior resistance against irradiation damage In1–xGaxN alloys present a great potential for applications in photovoltaic and other optoelectronic devices
  • Authors investigated the effects of high-energy proton irradiation on the optical and electrical properties of In-rich In1-xGaxN alloys as a function of the displacement damage dose
  • InGaN can withstand extraordinarily high dose of radiation and nitrides are less sensitive to the irradiation
  • With Ga fraction increament, the irradiation-induced degradation in optical emission becomes increasingly severe in In1-xGaxN
  • Defect levels located close to the midgap position are the most effective nonradiative recombination centers
  • Nonradiative recombination rates decrease in semiconductors in which EFS lies closer to the conduction band edge
  • Radiation hardness of InN and InGaN alloys is an intrinsic property of these materials, rather than an extrinsic insensitivity caused by the high density of defects that already exist in the starting materials
  • The irradiation-generated defect states are efficient electron traps in GaAs and GaInP, while in InN and In-rich InGaN they are not

Plasmonic nanoparticle enhanced photocurrent in GaN/InGaN/GaN quantum well solar cells[25][edit]

Abstract: Author(s) demonstrate enhanced external quantum efficiency and current-voltage characteristics due to scattering by 100 nm silver nanoparticles in a single 2.5 nm thick InGaN quantum well photovoltaic device. Nanoparticle arrays were fabricated on the surface of the device using an anodic alumina template masking process. The Ag nanoparticles increase light scattering, light trapping, and carrier collection in the III-N semiconductor layers leading to enhancement of the external quantum efficiency by up to 54%. Additionally, the short-circuit current in cells with 200 nm p-GaN emitter regions is increased by 6% under AM 1.5 illumination. AFORS-Het simulation software results were used to predict cell performance and optimize emitter layer thickness.

  • Plasmonic nanoparticle scattering offers a unique way to circumvent the inherent trade-off between absorption and carrier collection in the design of solar cells. Optically thick cells can absorb all nonreflected incident light but incomplete carrier collection can limit cell internal quantum efficiency. The thickness reduction of high efficiency, low cost, thin film solar cells is limited by the absorption properties of the active layer.
  • In case of thin film cells, there is an additional benefit: metal nanostructures can couple incident light into guided modes that propagate through the active region, thereby increasing absorption and photocurrent.
  • The cell with a 50 nm p-GaN emitter layer shows a 6% overall increase in EQE with the addition of nanoparticles.
  • notable difference in EQE between the two cells (author(s) prototype) suggests that light absorption is most strongly enhanced in region closest to Ag nanoparticles. photocurrent enhancement may result from a combination of scattering, local field enhancement, and antireflection coating effects.
  • presence of nanoparticle arrays on the cells increases the short circuit current appreciably, increase in short circuit current (Jsc) is attributed to absorption enhancement evident from spectral response measurements.
  • enhanced Jsc could also be attributed to an increase in carrier collection efficiency at surface of the cell
  • Metal nanoparticles can increase optical path length in very thin cells, where surface texturing is not possible, due to their large scattering crosssections and enhanced local fields
  • The metal nanostructures in thin film cells can couple incident light into guided modes that propagate through the active region, thereby increasing absorption and photocurrent
  • Metal layers improve the contact to both p-GaN and p-AlGaN layers causing an increase in the depletion width at the surface of the device
  • The single QW system is a useful platform for the exploration of absorption enhancement from plasmonic scatterers on ultrathin cells designed for optimum photovoltaic response

Effect of indium fluctuation on the photovoltaic characteristics of InGaN/GaN multiple quantum well solar cells[26][edit]

Abstract: Severe In fluctuation was observed in In0.3Ga0.7N/GaN multiple quantum well solar cells using scanning transmission electron microscopy and energy dispersive x-ray spectroscopy. The high In content and fluctuation lead to low fill factor (FF) of 30% and energy conversion efficiency (η) of 0.48% under the illumination of AM 1.5G. As the temperature was increased from 250 to 300 K, FF and η were substantially enhanced. This strong temperature-dependent enhancement is attributed to the additional contribution to the photocurrents by the thermally activated carriers, which are originally trapped in the shallow quantum wells resulting from the inhomogeneous In distribution.

  • It has been reported that the composition fluctuation in InxGa1−xN MQWs tends to be severe for x>0.15. The fluctuation of In content is mainly due to the low miscibility of InN in GaN.
  • for high In content (30%), severe composition fluctuation occurred in the InxGa1−xN layers, leading to various quantum wells with different barrier heights. The low EQE may result from the fact that many shallow quantum wells, containing band gap energies close to that of barriers, were formed by In fluctuation, and made most of the incident optical energies insufficient to excite carriers, leading to limited photocurrents.
  • poor device performance is caused by compromised crystal quality in active region due to high In content. As lattice strain is introduced in InxGa1−xN/GaN structures, undesired defects are likely to form with increasing x, these defects reduce the electrical field across the intrinsic regions and hinder photogenerated carriers escaping from MQWs, leading to low FF and η. The suspected composition fluctuation could also reduce FF and η by preventing the carriers from being optically excited.
  • HAADF STEM image revealed that the In distribution inside quantum wells is not completely homogeneous. A depletion of In is observed in some quantum well regions. The EDS attached to the STEM was utilized to characterize the distribution of In composition further. For the EDS analysis, the electron beam focused down to a diameter of 1.5 nm shows the intensity profile of In( Kα) measured by EDS scan, indicating the severe fluctuation of In atom distribution. The depletion regions of quantum wells with the lower In intensities imply that quantum wells were formed with shallow potential depths.
  • The increase in FF and η with temperature indicates that additional photocurrents contributed by thermally activated carriers from shallow wells outweighed the reduction in Voc.
  • The critical thicknesses (Tc) of InxGa1−xN grown on GaN are generally less than 100 nm for x>0.1, and decrease very rapidly with increasing x
  • The composition fluctuation in InxGa1−xN MQWs tends to be severe for x>0.15
  • For InGaN/GaN MQW solar cells the peak wavelength of EL (522 nm) is longer than that of EQE (375 nm)- can be attributed to the fact that in emission process carriers are relaxed to lower energy states before being recombined radiatively, whereas in absorption process electrons (holes) can be excited to higher states as long as the absorbed photon energy matches the energy level separation
  • The low EQE is due to many shallow quantum wells, containing the band gap energies close to that of barriers- were formed by the In fluctuation
  • Defects reduce the electrical field across the intrinsic regions and hinder the photogenerated carriers escaping from MQWs, leading to low FF and efficiency
  • The depletion regions of quantum wells with the lower In intensities imply that quantum wells were formed with shallow potential depths
  • Thermal activation increases photocurrent and hence Jsc
  • Increasing saturation currents and recombinations causes Voc to drop

Temperature dependences of InxGa1−xN multiple quantum well solar cells[27][edit]

Abstract: In this work, high open circuit voltages (Voc) of In0.2Ga0.8N and In0.28Ga0.72N multiple quantum well solar cells (MQWSCs) are experimentally obtained (2.2 V and 1.8 V, respectively). The Voc of In0.28Ga0.72N MQWSCs is lower than the expected value due to serious indium segregation problems causing more defects in In0.28Ga0.72N films, which is consistent with the observation of a high ideality factor in dark current measurement. The temperature dependence of the Voc and the short circuit current (Jsc) in In0.2Ga0.8N MQWSCs is found to be larger than the corresponding values in In0.28Ga0.72N MQWSCs. It is also observed that higher quantum well energy barrier exhibits a low fill factor of 0.52 due possibly to the loss of electric field and the higher energy barrier. This obtained efficiency increases with temperatures up to 100 °C and then decreases due to competing results between the reduction in Voc and an increase in Jsc.

  • Wider bandgap barrier materials determine the open circuit voltage
  • The short circuit current is determined by the narrower bandgap well materials
  • A single bandgap solar cell has the maximum efficiency at an energy bandgap of around 1.35–1.5 eV
  • The quantum well with lower bandgap materials can absorb more light spectrum and contribute additional photocurrent
  • The electric field across the intrinsic region is an important factor affecting the photogenerated carrier’s escape from QWs
  • At higher temperatures, short circuit current (Jsc) increases slightly and open circuit voltage (Voc) drops significantly in MQWSCs. solar cell efficiency is seriously degraded due to a large voltage drop at high temperature.
  • large series resistance of solar cells may also reduce the fill factor
  • loss of electric field can be attributed to higher background impurities in the intrinsic region and defects caused by lattice mismatch. In addition, it is noted that temperature effect on photocurrent in low indium content alloy MQWSCs is more obvious than that in low indium content alloy MQWSCs. It is possible that when temperature becomes higher, the photogenerated carriers with lower energy barriers have a larger probability to escape out of QW than those with higher energy barriers.
  • large number of defects in high In composition film could result in higher background doping, which in turn effects the loss of electric field and thus seriously degrades the fill factor of solar cells. Furthermore, higher energy barrier can also degrade the fill factor of solar cells.
  • large drop in fill factor of MQWSCs is observed with an increasing energy barrier in QWs.
  • To achieve a high efficiency in MQWSCs, a compromise between Jsc and Voc is necessary. If the electric field can be maintained, addition of quantum well in solar cell will result in an increase in Jsc and a reduction in the Voc. The efficiency depends mainly on the compromise between Jsc and Voc.
  • increase in Jsc is due to bandgap narrowing with increasing temperature. More carriers are generated in QWs and escape of these carriers is enhanced by thermal contribution.

Substantial photo-response of InGaN p–i–n homojunction solar cells[28][edit]

Abstract: InGaN p–i–n homojunction structures were grown by metal-organic chemical vapor deposition, and solar cells with different p-contact schemes were fabricated. X-ray diffraction measurements demonstrated that the epitaxial layers have a high crystalline quality. Solar cells with semitransparent p-contact exhibited a fill factor (FF) of 69.4%, an open-circuit voltage (Voc) of 2.24 V and an external quantum efficiency (EQE) of 41.0%. On the other hand, devices with grid p-contact showed the corresponding values of 57.6%, 2.36 V, 47.9% and a higher power density. These results indicate that significant photo-responses can be achieved in InGaN p–i–n solar cells.

  • heterojunction may cause poor quality of InGaN for future practical application due to large latticemismatch betweenGaNand InGaN, especially for full-spectrum-response devices in which higher In content is necessary. Therefore, it is important to develop InGaN solar cells with homojunction structures.
  • there is a 38% increase of the light power reaching absorption region for the devices with grid contact, compared with devices with semitransparent contact. However, the observed 26% increase in Jsc for the devices with grid contact is much lower, which is currently attributed to the larger series resistance of these devices. This larger resistance may be caused by the transverse flow of current in p-GaN to the grids with a spacing of 45 μm.
  • devices with semitransparent contact have a high fill factor (FF) of 69.4% while the devices with grid contact have a smaller FF of 57.6%. This reduction of the FF value is currently attributed to large resistance which may deteriorate J–V performance of devices with grid contact
  • devices with grid contact exhibit a higher maximum output power density (Pmax) of 2.32 mW cm−2, compared with 2.12 mW cm−2 of devices with semitransparent contact.
  • for devices with grid contact, major loss mechanisms include reflection at the p-GaN surface and absorption of grid lines.
  • InGaN alloys with a length of a few hundred nanometers can absorb a substantial fraction of the incident light
  • Thomas Swan low-pressure MOCVD technique was used
  • The barrier at the p-InGaN/p-GaN interface is of benefit in keeping the electrons generated in p-InGaN away from the p-type ohmic contact and contributes to the light-generated current
  • The observed 26% increase in Jsc for the devices with grid contact is much lower- attributed to the larger series resistance of these devices
  • Devices with grid contact showed a higher power density, which indicated that more incident light can be absorbed and then converted into electric energy, compared with devices with semitransparent contact

Analytical model for the optical functions of indium gallium nitride with application to thin film solar photovoltaic cells[29][edit]

Abstract: This paper presents the preliminary results of optical characterization using spectroscopic ellipsometry of wurtzite indium gallium nitride (InxGa1−xN) thin films with medium indium content (0.38 < x < 0.68) that were deposited on silicon dioxide using plasma-enhanced evaporation. A Kramers–Kronig consistent parametric analytical model using Gaussian oscillators to describe the absorption spectra has been developed to extract the real and imaginary components of the dielectric function (ɛ1, ɛ2) of InxGa1−xN films. Scanning electron microscope (SEM) images are presented to examine film microstructure and verify film thicknesses determined from ellipsometry modeling. This fitting procedure, model, and parameters can be employed in the future to extract physical parameters from ellipsometric data from other InxGa1−xN films.

  • film composition was determined by the Tauc method using absorption data obtained from the spectroscopic ellipsometry models.
  • Ellipsometric measurements of the InxGa1−xN films yield two parameters for each wavelength and angle of incidence: Ψ and Δ, representing the change in amplitude ratio and change in phase shift, respectively, between the p- and s-components of the light beam's electric field. Through this method important information such as film thickness, surface roughness, optical constants (dielectric constants: ɛ1 and ɛ2; index of refraction: η and extinction coefficient: κ) and absorption coefficients, α, can be obtained.
  • 204 nm-thick In0.64Ga0.36N and 221 nm-thick In0.68Ga0.32N films show single-phase isolated platelet/nanocolumnar grains with average diameters of 85 nm and 105 nm, respectively. The InxGa1−xN films display a clear relationship of larger grain or nanocolumn sizes with increasing indium content, which has been reported previously by author(s).
  • Typically, transmission and reflection (T&R) measurements are performed to obtain absorption information. However, spectroscopic ellipsometry is a good alternative if T&R is not possible due to opaque substrates such as the Si/SiO2 used by author(s). Additionally, ellipsometry can be used to determine the optical band gap of a thin film if photoluminescence and spectrophotometry are not options.
  • Kramers–Kronig consistent parametric model has been developed for the optical functions of wurtzite InxGa1−xN alloy films of medium indium contents (0.38 < x < 0.68) deposited by a novel peed system. This model employing simple Gaussian oscillators is used to fit spectroscopic ellipsometric data over the 0.8–4.5 eV range to obtain film thicknesses, dielectric functions and absorption coefficients.

InGaN-Based p–i–n Solar Cells with Graphene Electrodes[30][edit]

Abstract:InGaN-based p–i–n solar cells with graphene electrodes were fabricated and compared with solar cells using indium tin oxide (ITO) electrodes. In particular, authors analyzed the properties of graphene film by means of high-resolution transmission electron microscopic (HRTEM) and Raman spectroscopy, also comparing optical properties with those of ITO, conventionally used as transparent electrodes. The solar cells using graphene revealed a short circuit current density of 0.83 mA/cm2, an open circuit voltage of 2.0 V, a fill factor of 75.2%, and conversion efficiency of 1.2%, comparable to the performance of solar cells using ITO.

  • ITO also has limitations such as increasing cost due to indium scarcity, limited transparency at near UV and IR regions, and a sensitivity to acidic and base chemical sources. Furthermore, ITO can easily crack and break when deposited on bending substrates, making it difficult to use for applications such as flexible displays and touch screens
  • Graphene has outstanding optical and electrical properties, i.e., high mobility, transparency, flexibility, and mechanical and chemical stability
  • Graphene, having 4.5 eV work function, reveals slightly lower Schottky barrier with p-GaN than ITO which shows 4.3 eV work function
  • The approximately 10% increase in the Jsc of the graphene sample could be explained by the overlapping EQE region of the graphene sample with the solar spectrum, which had a larger area than that of the ITO sample
  • ITO transmittance varies with thickness and wavelength, while the thin-layered graphene maintains almost the same transmittance

References[edit]

  1. D.V.P. McLaughlin & J.M. Pearce, "Progress in Indium Gallium Nitride Materials for Solar Photovoltaic Energy Conversion", Metallurgical and Materials Transactions A 44(4) pp. 1947-1954 (2013).
  2. X.-mei Cai, S.-wei Zeng, and B.-ping Zhang, “Fabrication and characterization of InGaN p-i-n homojunction solar cell,” Applied Physics Letters, vol. 95, no. 17, pp. 173504-173504-3, Oct. 2009
  3. E. A. Berkman, N. A. El-Masry, A. Emara, and S. M. Bedair, “Nearly lattice-matched n, i, and p layers for InGaN p-i-n photodiodes in the 365–500 nm spectral range,” Applied Physics Letters, vol. 92, no. 10, p. 101118, 2008.
  4. X. Cai, S. Zeng, and B. Zhang, “Fabrication and characterization of InGaN p-i-n homojunction solar cell,” Applied Physics Letters, vol. 95, no. 17, p. 173504, 2009.
  5. R. Dahal, B. Pantha, J. Li, J. Y. Lin, and H. X. Jiang, “InGaN/GaN multiple quantum well solar cells with long operating wavelengths,” Applied Physics Letters, vol. 94, no. 6, p. 063505, 2009.
  6. C. J. Neufeld, N. G. Toledo, S. C. Cruz, M. Iza, S. P. DenBaars, and U. K. Mishra, “High quantum efficiency InGaN/GaN solar cells with 2.95 eV band gap,” Applied Physics Letters, vol. 93, no. 14, p. 143502, 2008.
  7. X. Sun, W. B. Liu, D. S. Jiang, Z. S. Liu, S. Zhang, L. L. Wang, H. Wang, J. J. Zhu, L. H. Duan, Y. T. Wang, D. G. Zhao, S. M. Zhang, and H. Yang, “Photoelectric characteristics of metal/InGaN/GaN heterojunction structure,” Journal of Physics D: Applied Physics, vol. 41, no. 16, p. 165108, Aug. 2008.
  8. C. Boney, I. Hernandez, R. Pillai, D. Starikov, A. Bensaoula, M. Henini, M. Syperek, J. Misiewicz, and R. Kudrawiec, “Growth and characterization of ingan for photovoltaic devices,” 2010, pp. 003316–003321.
  9. E. Matioli, C. Neufeld, M. Iza, S. C. Cruz, A. A. Al-Heji, X. Chen, R. M. Farrell, S. Keller, S. DenBaars, U. Mishra, S. Nakamura, J. Speck, and C. Weisbuch, “High internal and external quantum efficiency InGaN/GaN solar cells,” Applied Physics Letters, vol. 98, no. 2, p. 021102, 2011.
  10. O. Jani, I. Ferguson, C. Honsberg, and S. Kurtz, “Design and characterization of GaN∕InGaN solar cells,” Applied Physics Letters, vol. 91, no. 13, p. 132117, 2007.
  11. L. Hsu and W. Walukiewicz, “Modeling of InGaN/Si tandem solar cells,” Journal of Applied Physics, vol. 104, no. 2, p. 024507, 2008.
  12. D.-Y. Zhang, X.-H. Zheng, L.-J. Tang, J.-R. Dong, H. Wang, and H. Yang, “Photovoltaic Effects of InGaN/GaN Double Heterojunctions With p-GaN Nanorod Arrays,” IEEE Electron Device Letters, vol. 31, no. 12, pp. 1422–1424, Dec. 2010.
  13. Ray-Hua Horng, Shih-Ting Lin, Yu-Li Tsai, Mu-Tao Chu, Wen-Yih Liao, Ming-Hsien Wu, Ray-Ming Lin, and Yuan-Chieh Lu, “Improved Conversion Efficiency of GaN/InGaN Thin-Film Solar Cells,” IEEE Electron Device Letters, vol. 30, no. 7, pp. 724–726, Jul. 2009.
  14. X. Zheng, R.-H. Horng, D.-S. Wuu, M.-T. Chu, W.-Y. Liao, M.-H. Wu, R.-M. Lin, and Y.-C. Lu, “High-quality InGaN∕GaN heterojunctions and their photovoltaic effects,” Applied Physics Letters, vol. 93, no. 26, p. 261108, 2008.
  15. X. Chen, K. D. Matthews, D. Hao, W. J. Schaff, and L. F. Eastman, “Growth, fabrication, and characterization of InGaN solar cells,” physica status solidi (a), vol. 205, no. 5, pp. 1103–1105, 2008.
  16. E. Trybus, O. Jani, S. Burnham, I. Ferguson, C. Honsberg, M. Steiner, and W. A. Doolittle, “Characteristics of InGaN designed for photovoltaic applications,” physica status solidi (c), vol. 5, no. 6, pp. 1843–1845, 2008.
  17. L. W. Ji, Y. K. Su, S. J. Chang, S. T. Tsai, S. C. Hung, R. W. Chuang, T. H. Fang, and T. Y. Tsai, “Growth of InGaN self-assembled quantum dots and their application to photodiodes,” Journal of Vacuum Science Technology A: Vacuum, Surfaces, and Films, vol. 22, no. 3, pp. 792 –795, May 2004.
  18. R. Dahal, J. Li, K. Aryal, J. Y. Lin, and H. X. Jiang, “InGaN/GaN multiple quantum well concentrator solar cells,” Applied Physics Letters, vol. 97, no. 7, pp. 073115–073115–3, Aug. 2010.
  19. O. Jani, B. Jampana, M. Mehta, Hongbo Yu, I. Ferguson, R. Opila, and C. Honsberg, “Optimization of GaN window layer for InGaN solar cells using polarization effect,” in 33rd IEEE Photovoltaic Specialists Conference, 2008. PVSC ’08, 2008, pp. 1 –4.
  20. X. Zhang, X. Wang, H. Xiao, C. Yang, J. Ran, C. Wang, Q. Hou, and J. Li, “Simulation of In0.65Ga0.35N single-junction solar cell,” Journal of Physics D: Applied Physics, vol. 40, no. 23, pp. 7335–7338, Dec. 2007.
  21. A. Koo, F. Budde, B. J. Ruck, H. J. Trodahl, A. Bittar, A. Preston, and A. Zeinert, “Photoconductivity in nanocrystalline GaN and amorphous GaON,” Journal of Applied Physics, vol. 99, no. 3, pp. 034312–034312–7, Feb. 2006.
  22. H. Hamzaoui, A. S. Bouazzi, and B. Rezig, “Theoretical possibilities of InxGa1−xN tandem PV structures,” Solar Energy Materials and Solar Cells, vol. 87, no. 1–4, pp. 595–603, May 2005.
  23. J. J. Wierer, A. J. Fischer, and D. D. Koleske, “The impact of piezoelectric polarization and nonradiative recombination on the performance of (0001) face GaN/InGaN photovoltaic devices,” Applied Physics Letters, vol. 96, no. 5, pp. 051107–051107–3, Feb. 2010.
  24. J. Wu, W. Walukiewicz, K. M. Yu, W. Shan, J. W. Ager, E. E. Haller, H. Lu, W. J. Schaff, W. K. Metzger, and S. Kurtz, “Superior radiation resistance of In1−xGaxN alloys: Full-solar-spectrum photovoltaic material system,” Journal of Applied Physics, vol. 94, no. 10, pp. 6477–6482, Nov. 2003.
  25. I. M. Pryce, D. D. Koleske, A. J. Fischer, and H. A. Atwater, “Plasmonic nanoparticle enhanced photocurrent in GaN/InGaN/GaN quantum well solar cells,” Applied Physics Letters, vol. 96, no. 15, pp. 153501–153501–3, Apr. 2010.
  26. K. Y. Lai, G. J. Lin, Y.-L. Lai, Y. F. Chen, and J. H. He, “Effect of indium fluctuation on the photovoltaic characteristics of InGaN/GaN multiple quantum well solar cells,” Applied Physics Letters, vol. 96, no. 8, pp. 081103–081103–3, Feb. 2010.
  27. M.-J. Jeng, Y.-L. Lee, and L.-B. Chang, “Temperature dependences of InxGa1−xN multiple quantum well solar cells,” Journal of Physics D: Applied Physics, vol. 42, no. 10, p. 105101, May 2009
  28. S. W. Zeng, B. P. Zhang, J. W. Sun, J. F. Cai, C. Chen, and J. Z. Yu, “Substantial photo-response of InGaN p–i–n homojunction solar cells,” Semiconductor Science and Technology, vol. 24, no. 5, p. 055009, May 2009
  29. D. V. P. McLaughlin and J. M. Pearce, “Analytical model for the optical functions of indium gallium nitride with application to thin film solar photovoltaic cells,” Materials Science and Engineering: B, vol. 177, no. 2, pp. 239-244, Feb. 2012
  30. J.-P. Shim, M. Choe, S.-R. Jeon, D. Seo, T. Lee, and D.-S. Lee, “InGaN-Based p–i–n Solar Cells with Graphene Electrodes,” Applied Physics Express, vol. 4, no. 5, p. 052302, 2011.