Quantum dot solar cells^[1]

A. J. Nozik, “Quantum dot solar cells,” Physica E: Low-dimensional Systems and Nanostructures, vol. 14, no. 1–2, pp. 115–120, Apr. 2002.

Background: Hot carrier and impact ionization Hot carrier @ solar cells: Electrons/holes receives photons with energy much higher than their band gap will carry kinetic energy create effective temperature much higher than lattice temperature (3000K of carrier @ 300K of lattice etc.). Primary loss for this:

• Heat through scattering (KN is transferred into band vibration?)
• Photon emission (electrons occupied holes to let out photons?). Approach to stop this loss:
• Stacked cascaded multiple p-n junctions to match multiple band of solar spectrum. This reduce carrier relaxation via photon emission. Efficiency increase to 66%.
• Reduce thermal relaxation – utilize hot carriers before their relaxation.
  • Enhanced photo-voltage: being extracted before cooling; Requirement: transportation of hot carriers faster compare to the cool down rate – which should be related to the material itself.
  • Enhanced photo-current: create second or more electron-hoe pair(s) by impact ionization (Auger effect: one electron hit a band create one e-h pair. What will happen to this electron?); Requirement: impact ionization rate faster compare to the cool down rate & e-h pair transport rate faster than cool rate?.
  • Quantization confinement will dramatically reduce hot carrier’s cooling rates. (So this confinement will also reduce the cool down rate in Auger process which is not desired?)
• What about some part of solar spectrum is absorbed to stimulate electron from En to En+1 or higher and the left spectrum will cause these electrons to emission from En+1 to En-1 to let out more photons with same lambda. Then one layer will be used to specifically to absorb these electrons.

Hot electrons and hot holes cool down rate are different because:

• Mass difference (Which one is faster?)
• Hot carrier cool rate depends on density of photo-generated hot carriers (How this makes diff?).

Predicted way for QDs: hot electrons with slowed cooling rate in QD --- Auger process, fast cool down hole due to mass & closer quantized space --- photo-currents. Bottleneck due to fast hole trap at the surface will slow cooling rate (7 ps at CdSe, InP QDs compare to 0.3 ps cooling rate without hole trap). This will prevent Auger process.
Suggested Syetems:

• QD arrays in p-i-n cells. (Quasi-1D system)
  • Advantages:
    • Delocalized quantized 3-D miniband states might slow the carrier cooling rate to allow hot electron transportation and collection.
    • Impact ionization can occur, however not in the same time with hot electron transportation/collection.
  • Fabrication:
    • Colloidal
    • Epitaxial
  • Challenge:

Disorder of shape, surface state etc.

• QD sensitized DSSC (Dye-sensitized solar cell)
  • Advantage:
    • Tunability of optical properties with size of QDs.
    • Quantum yield might be greater than 1, meaning one photon might generate more than one electron-hole pairs.
  • Fabrication:

Absorption from QD colloidal.

• Quantum dots dispersed in organic semiconductor polymer matrices.

CdSe QDs are formed in hole conducting polymer. Structure of QDs affects the efficiency. Different layers can be applied to conduct electrons (TiO2) or holes (MEH-PPV). Challenge: electron-hole pair might recombine in surface of polymers. Refer to CdSe Nanocrystal Rods/Poly (3-hexylthiophene) Composite Photovoltaic Devices for rodlike CdSe QDs structure. Refer to Charge transfer in photovoltaics consisting of interpenetrating networks of conjugated polymer and TiO2 nanoparticles for multiple polymer structure.

CdSe Nanocrystal Rods/Poly (3-hexylthiophene) Composite Photovoltaic Devices^[2]

W. U. Huynh, X. Peng, and A. P. Alivisatos, “CdSe Nanocrystal Rods/Poly(3-hexylthiophene) Composite Photovoltaic Devices,” Adv. Mater., vol. 11, no. 11, pp. 923–927, Aug. 1999.

Important factors affect the performance of a PV device includes:

• Overlap of the absorption of solar spectrum;
• Efficiency of carrier separation and transportation;
• Enhanced charge separation occurs at interface between two materials with disparate electron affinities will transport in two materials and have low probability of recombination.

Paper demonstrated:

• 5 nm PbSe nanocrystal: 5% external quantum efficiency (EQE, electrons/incident-photons per-second; internal QE: electrons/absorbed-photons per-second. 0.25% power conversion efficiency. Poor behavior due to poor transportation.
• Elongated particles have a tendency, that increases with size, to form *chains of particles connected along the c-axis
• 4*7 nm PbSe nanocrystal after forming chain, can reach EQE as high as 4%
• 8*13 nm PbSe nanocrystal after forming chain, can reach EQE as high as 16%

Larger size of PbSe nanocrystal are hard to fabricated, lead to question of what influence of nanocrystal’s shape has.

Charge transfer in photovoltaics consisting of interpenetrating networks of conjugated polymer and TiO2 nanoparticles^[3]

A. C. Arango, S. A. Carter, and P. J. Brock, “Charge transfer in photovoltaics consisting of interpenetrating networks of conjugated polymer and TiO2 nanoparticles,” Applied Physics Letters, vol. 74, no. 12, pp. 1698–1700, Mar. 1999.

Use of interpenetrating donor-acceptor heterojunctions and interpenetrating polymer network can substantially improve the photoconductivity. Layer fabrication: Opaque TiO2 layer is fabricated in thickness of 4-6 um, at particle size of 80 nm. Polymer is penetrated into TiO2 particle pores with size of 20 nm. Polymer is used as charge transporter. ITO-(Layer)-Calcium sandwich. For current flow:

• Calcium acts like electron transfer;
• ITO acts like hole transfer. Hole should transfer from ITO to TiO2 valence band;
• Hole transfer from TiO2 valence band to ITO. However, insulation property of TiO2 prevent upper two situation from happening. But, under illumination, TiO2 presents property of accepting and conducting photo-excited electrons.

Advantage: TiO2 has low conduction band, and has low saturation voltage (achieved by using stable and high-working function contact)

Quantum Dot Solar Cells. Tuning Photoresponse through Size and Shape Control of CdSe-TiO2 Architecture^[4]

A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno, and P. V. Kamat, “Quantum Dot Solar Cells. Tuning Photoresponse through Size and Shape Control of CdSe−TiO2 Architecture,” J. Am. Chem. Soc., vol. 130, no. 12, pp. 4007–4015, Mar. 2008.

Basic principle for QD solar cells: Short-band-gap semiconductors such as CdS, PbS, Bi2S3, CdSe and InP can be used as sensitizers (to light?) due to their ability to transfer electrons to large-band-gap semiconductors such as TiO2 or SnO2 under visible light excitation. Short-band-gap also plays roll of extend TiO2 etc. to visible region.
Chemical and electrochemical deposited CdS and CdSe nanocrystal on TiO, SnO2 and ZnO surface can inject excited electrons and generate photocurrent under visible irradiation. But the efficiency is low due to fast charge recombination (where?) Refer to: The influence of dye structure on charge recombination in dye-sensitized solar cells for details.
TiO2 nanotubes scales: 80-90 nm in diameter and ~8 um in length. Hollow nature make tubes both inner and outer surface accessible for modification with sensitizing dyes or quantum dots. Tubes are upon Ti base, which is used as charge collector. Roughness factor: 48 for nanotubes and 75 for particles.
Smaller-sized CdSe quantum dots show greater charge injection rates and also higher IPCE at the excitonic band. Larger particles have better absorption in the visible region but cannot inject electrons into TiO2 as effectively as smaller-sized CdSe quantum dots. 3.0 nm CdSe quantum dots were observed to be able to generate the highest photocurrent density @ 2 mA/cm^2 at response time between 20 sec and 45 sec.
Incident Photo to Charge-carrier Efficiency (IPCE): 1240*short-circuit-photo-current@(A/cm^2) / (wavelength@nm*mono-incident-power@(W/cm^2))
Through observation, smaller sized QD @2.3 nm is believed to give faster rate of electron transfer, may due to that they are more energetic in excited state to be capable of injecting (hot) electrons into TiO2 at faster rate.
Absorption: TiO2 nanotube 5% higher than nanoparticle; IPCE: TiO2 nanotube 10% higher than nanoparticle.—Represent structure influence charge transfer.
Electrons in TiO2 nanoparticles are more likely to loss at grain boundaries than those in TiO2 nanotubes.
Observation of open circuit voltage shows that: with CdSe particles in nanotubes, electrons injected into nanotubes can survive longer thus to improve photocurrent generation efficiency of solar cells.
Higher vacant energy levels of TiO2 facilitate direct electron transfer from the excited sensitizer in sub-psec time scale (what’s the ordinary time scale?)
Rainbow solar cell: ordered assemblies of CdSe QDs of different diameters decorate TiO2 nanotubes.

• Downside: excess energy of electrons of small-sized particles is lost once they are transferred to TiO2.
• Advantage: faster electron injection rate and greater absorption range.

Future work: rainbow solar cell maximize light absorption of QD solar cell.

Dye-Sensitized solar cells^[5]

M. Grätzel, “Dye-sensitized solar cells,” Journal of Photochemistry and Photobiology C: Photochemistry Reviews, vol. 4, no. 2, pp. 145–153, Oct. 2003.

Light absorption and carrier transportation are separated in DSSC, light is absorbed by sensitizer, which has a range from UV to near IR. Solar to current efficiency (IPCE) @ AM 1.5 reaches 10% here.
Structure:

• TiO2: mesoporous oxide layer, nano particles, where electronic transportation takes place. ZnO, Nb2O5 can also be used.
• Dye: where electrons are excited by photons and injected into conduction band of oxide.
• Electrolyte: provide electron donation to restore dye. Usually an organic solvent containing redox system.

Future study: Dye should have these properties:

• Panchromatic
• Attachment group such as carboxylate or phosphonate.
• Upon excitation dye can inject electrons into oxide layer
• The energy level should match the lower bound of conduction band of oxide.
• Redox potential should be high enough to gain donation from electrolyte.
• Stable enough: 10^8 correspond 20 years of nature light

Organic dyes: coumarine or polyene solar-electric power conversion reaching up to 7.7% @ full sunlight Problem for QD dye: photo-corrosion: unstable
Mesoporou oxide film:

• Inherent conductivity is low
• Does not support built in E-field
• Three dimension transportation. Future study will focus on structure of higher degree of order.

The influence of dye structure on charge recombination in dye-sensitized solar cells^[6]

J. R. Jennings, Y. Liu, Q. Wang, S. M. Zakeeruddin, and M. Grätzel, “The influence of dye structure on charge recombination in dye-sensitized solar cells,” Phys. Chem. Chem. Phys., vol. 13, no. 14, pp. 6637–6648, Apr. 2011.

Bipyridyl ruthenium failed to perform well in complete solar cells as sensitizer as expected. New sensitizer dyes with improved absorption of red part of solar spectrum should be developed.
Z-907 has poorer PV performance when nonyl groups are replaced with amino groups due to higher rate of recombination (at surface or bulk?)