Project: “Hot Carrier solar cell: Implementation of the Ultimate PV Converter”

Annual Report April 2011 This project is a composite project involving four nodes PIs: Gavin Conibeer and Martin Green, University of New South Wales (UNSW) PI: Jean-François Guillemoles, Institut de Recherche et Développement sur

l'Energie Photovoltaïque, Paris (IRDEP) PI: Tim Schmidt, University of Sydney (USyd) PIs: Antonio Marti and Antonio Luque, Instituto de Energia Solar-UPM, Madrid

(IES-UPM) This third annual report details the work carried out on the project at each node, with this introduction acting as an overview of the whole project. As reported in previous reports, the disparate timings of the start of the project in the different nodes have meant that work at IES-UPM started nine months before the other nodes. Hence the report presented last year was for IES-UPM was their final report. This report is not reproduced in this compilation report, although it has been re-submitted separately. The inaugural meeting of all four nodes was held before the project started in September 2007 in Madrid. The first annual meeting was held in May 2009 in Strasbourg and the second annual meeting held in September 2010 in Valencia. Several other meetings between two of the partner nodes have taken place during the project. There now follows an overview of progress on the project followed by the individual detailed reports from each node. Contents Page Introduction 2 Overview report 3 UNSW node report 4 IRDEP node report 29 USyd node report 34 End 42 (IES-UPM node report was finalised last year and has been re-submitted separately.)

E

DOS

small Eg

Ef(n

Ef(p)

TA

h+ selective energy contact

Ef µA = qV

Hot carrier distribution

TH TA

e- selectiveenergy contact

ES

Fig 1: Band diagram of a Hot Carrier solar cell,

Introduction and Background Hot carrier solar cells offer the possibility of very high efficiencies (limiting efficiency 68% for unconcentrated illumination) but with a structure that could be conceptually simple compared to other very high efficiency PV devices – such as multi-junction monolithic tandem cells. For this reason, the approach lends itself to ‘thin film’ deposition techniques, with their attendant low costs in materials and energy usage and facility to use abundant, non-toxic elements. [1-5] The Hot Carrier cell device has the following stringent requirements:

A. Slowing of thermalisation of photogenerated electrons (and holes) in the absorber material.

B. Extraction of these ‘hot carriers’ to external contacts over a narrow range of energies, such that excess carrier energy is not lost to the cold contacts.

C. In addition, a working device would require integration of the structures used to tackle A and B without compromising their performance.

The absorber has a hot carrier distribution at temp TH . Carriers cool isoentropically in the mono-energetic contacts to TA: their kinetic energy being converted into useable potential energy [1,2]. The difference of the Fermi levels of these two contacts manifests as a difference in chemical potential of the carriers at each contact and hence an external voltage. The challenges to produce such devices fall into 2 categories: (i) keeping carriers hot without heating the lattice and (ii) achieving a fast extraction of hot carriers through a narrow allowed energy range [1,2]. While these challenges are tough, it seems they can be met using newly available materials and more specifically nanostructured semiconductors. The absorber is conceptually a single layer, homogeneous on the scale of a complete device. Similarly contacts are conceptually relatively simple effectively 2D structures. This conceptual simplicity of construction of the whole device would lend itself to thin film or related techniques. A great strength of this project is that it synergistically combines the significant experience in all three generations of photovoltaics at the University of New South Wales (UNSW), with that on intermediate band cells and QD characterisation at Instituto de Energia Solar (IES-UPM), with the pioneering work on Hot Carrier cells at IRDEP, Paris, in conjunction with UNSW and with the excellent optical expertise and facilities at Sydney University (USyd).

References: [1] Green MA, “Third Generation Photovoltaics”, (Springer Verlag, 2003). [2] Würfel P, Sol Energy Mats & Sol Cells, 46(1997) 43. [3] Ross R and Nozik AJ, J. Appl Phys, 53 (1982) 3318. [4] Conibeer G et al, Solar Energy Materials and Solar Cells, 2009, 93, 713. [5] Guillemoles J-F et al, PVSEC (Shanghai, 2005) 375.

Overview of Progress, May 2010 to April 2011

Progress towards both of the twin aims of developing a Hot Carrier absorber with slowed carrier cooling and characterising resonant tunnelling structures using quantum dots for energy selective contacts, has been made in the past year. Photogenerated hot carriers in an absorber lose their energy principally through emission of optical phonons. The subsequent blocking of the decay of these optical phonons by modification of an absorber material’s phonon dispersion maintains a ‘hot phonon’ population and reduces carrier cooling. Previous work in this project has shown this to be the optimal method to achieve slowed carrier cooling. In the past year, modelling of the efficiencies of hot carrier cells with indium niturde absorber has been continued. The efficiencies of devices with non-ideal selective contacts has also been continued by both the UNSW and IRDEP nodes, with 3D modelling of double barrier contacts and of non-ideal contacts, which show efficiencies over 50% can be achievable even if the contacts are only semi-selective. Illuminated I-V measurement of double barrier resonant structures at UNSW with silicon nanostructures grown by collaborators, have demonstrated improved resonance of enrgy at confined energy levels. Time resolved photoluminescence measurement of carrier cooling rates at the USyd and UNSW nodes has shown small increases in lifetimes of carriers in InN in addition to previously demonstrated slowed cooling in InP as compared to GaAS, demonstrating phonon bottleneck in wider phonon band gap materials. Further work at USyd on quantum well materials has shown long lifetimes of 2ns for hot carriers, corroborating earlier work and illuminating the mechanisms for slowing of carrier cooling. Modelling of nanostructure phononic properties has been used for real III-V nanostructures. Such structures have been fabricated by collaborators and are now being measured. Also colloidal quantum dot arrays have been fabricated at UNSW and are also now being measured for phonon dispersion. Further work at USyd on quantum well materials has shown long lifetimes of 2ns for hot carriers, corroborating earlier work and illuminating the mechanisms for slowing of carrier cooling. Device structures for complete devices are now being designed and initial structures for real devices are now being grown based on multiple quantum wells with energy selective contacts. Further work in the remaining period of the project will see further implementation and characterisation of these real device structures. It will also see consolidation of the nanostructure work and modelling of both III-V and colloidal core shell nanostructures as well as quantum well structure. Energy selective contacts using double barrier structures will be further refined and used in combination with absorbers, with time resolved PL, Raman and illuminated I-V characterisation of carrier cooling and electrical properties, with the aim of demonstrating a true hot carrier cell device. Full descriptions of these projects are given in the following detailed reports from each of the UNSW, IRDEP and USyd nodes. Lists of papers published in the last year is included at the end of each node report.

Project: “Hot Carrier solar cell: Implementation of the Ultimate PV Converter”

Institution (node): University of New South Wales, Sydney, Australia Annual Report April 2011 Abstract The UNSW node of the Hot Carrier Solar Cell project is investigating absorbers which slow the rate of carrier cooling - an essential element of a successful cell. It has also further developed the work on energy selective contacts for Hot Carrier cells. This work includes modelling the limiting efficiencies of cells using real material properties and modelling the phonon dispersions of ordered arrays of quantum dots. The object of the latter is to determine the potential to modify the allowed phonon energies such as to block the decay of optical phonons, this being the principle decay mechanism for hot photo-generated carriers in an absorber material. This will then offer the possibility of producing a material which substantially decreases the rate of carrier cooling in an absorber and hence allows carriers to be extracted whilst still hot and hence produce a higher external voltage. Thus also boosting the efficiency of such a ‘Hot Carrier solar cell’ very significantly. The investigation has three main themes: 1. Modelling of the phonon dispersions in three dimensions of these quantum dot

nanostructures. This will allow the phonon properties of various material combinations to be modelled and assessed.

2. Calculating the limiting efficiencies of cells with specific electronic, and later with specific phononic properties.

3. Fabrication and characterisation of ordered quantum dot arrays. Colloidal dispersion of capped nanocrystals to form 2D supercrystal arrays. These are being characterised for their modulation of phonon dispersions and for the effects of this modulation on the rate of carrier cooling.

4. Fabrication of double barrier resonant structures and characterising with illuminated I-V techniques.

In the past year modelling has progressed on the Hot Carrier efficiency with indium nitride as ab absorber. Measurement of double barrier resonant quantum dot structures grown in high quality by collaborators and improved 2/3D modelling has been carried out. For absorbers, modelling of nanocrystals superlattice arrays has been to III-V QD superlattices. Such superlattices have been grown by collaborators and are being measured and colloidal Langmuir-Blodgett growth of silicon Si nanocrsystals arrays has been carried out. Meausurement of carrier cooling rates has been carried out on indium nitride with its large phononic gap, demonstrating the importance of material quality. Finally design of complete hot carrier cell structures has been initiated with the important properties defined more precisely, with plans for fabricating such structures now being implemented.

Administration Personnel [All personnel are members of the ARC Photovoltaics Centre of Excellence (CoE), which is supported by the Australian Research Council.] Principal Investigators: Assoc/Prof. Gavin Conibeer – Deputy Director, Photovoltaics Centre of Excellence Prof. Martin A. Green – Research Director, Photovoltaics Centre of Excellence Investigators: Dr. Dirk König – Senior Research Fellow, GCEP, CoE Dr. Shujuan Huang – Research Fellow, GCEP, CoE Dr. Santosh Shrestha – Lecturer, CoE Dr. Ivan Perez-Wurfl – Research Fellow, CoE Dr. Yukiko Kamikawa – Visiting Research Fellow, CoE Students, CoE: Ms. Lara Treiber - PhD Mr. Pasquale Aliberti - PhD Mr. Andy Hsieh - PhD Mr. Robert Patterson - PhD Mr. Binesh Puthen Veettil - PhD Mr. Yu Feng - U/Grad thesis project (to Dec 2010) – PhD (from Feb 2011) Mr. Pengfei Zhang – PhD (from Feb 2011) Mr. Hongze Xia - U/Grad thesis project (from Feb 2011) Equipment (funded on other projects) Femtosecond laser for TR PL – located at JLUS, Sydney University Computer cluster – 650Gflops Langmuir-Blodgett deposition apparatus Organisation The UNSW node of the project is organised into sections working on:

Hot carrier dynamics and phonon dispersion modelling; Limiting Hot Carrier Cell efficiency modelling; Fabrication of quantum dot arrays for absorbers with slowed carrier cooling;

Fabrication and characterisation of energy selective contacts.

Introduction Hot carrier solar cells offer the possibility of very high efficiencies (limiting efficiency 65% for unconcentrated illumination) but with a structure that could be conceptually simple compared to other very high efficiency PV devices – such as multi-junction monolithic tandem cells. For this reason, the approach lends itself to ‘thin film’ deposition techniques, with their attendant low costs in materials and energy usage and facility to use abundant, non-toxic elements. An ideal Hot Carrier cell would absorb a wide range of photon energies and extract a large fraction of the energy to give very high efficiencies by extracting ‘hot’ carriers before they thermalise to the band edges. Hence an important property of a hot carrier cell is to slow the rate of carrier cooling to allow hot carriers collected whilst they are still at elevated energies (“hot”), and thus allowing higher voltages to be achieved from the cell and hence higher efficiency. A Hot Carrier cell must also only allow extraction of carriers from the device through contacts which accept only a very narrow range of energies (energy selective contacts or ESCs). This is necessary in order to prevent cold carriers in the contact from cooling the hot carriers, i.e. the increase in entropy on carrier extraction is minimized [1]. The limiting efficiency for the hot carrier cell is 65% at 1 sun and 85% at maximum concentration – very close to the limits for an infinite number of energy levels [2,3,4]. Figure 1 is a schematic band diagram of a Hot Carrier cell illustrating these two requirements.

Figure 1: Band diagram of the Hot Carrier cell. The device has four stringent requirements: a) To absorb a wide range of photon energies; b) to slow the rate of photogenerated carrier cooling in the absorber; c) To extract these ‘hot carriers’ over a narrow range of energies, such that excess carrier energy is not lost to the cold contacts; d) to allow efficient renormalisation of carrier energy via carrier-carrier scattering. Carrier cooling progresses by the interaction of hot carriers with the lattice to emit optical phonons. Interruption of the decay of these optical phonons I nto acoustic phonons can maintain the ‘hot phonon’ population and allow scattering back with the electron or hole gases, thus slowing down the rate of cooling. Methods to maximize this ‘hot phonon bottleneck’ effect are the subject for the work of the UNSW node of the project.

E

LO

ES

small Eg

Ef(n

Ef(p)

TA

h+ selective energy contact

Ef µA = qV

Hot carrier distribution

TH TA

e- selectiveenergy contact

ES

Progress May 2010 – April 2011 In the past year modelling of Hot Carrier efficiencies has progressed with implementation of real material properties to give more realistic efficiencies for InN which include Auger processes and more realistic contact structures. Significant progress has been made on demonstrating resonance in double barrier selective energy structures. This is complemented by improvements in 2/3D modelling of transport in these ESC structures. For absorbers, modelling of nanocrystals superlattice arrays has bee applied to real material systems. The growth of such systems in both III-V QD superlattices with collaborators and with colloidal Langmuir-Blodgett dispersion of Si nanocrsystals has produced structures which are now being characterised for their modulation of phononic properties. Also measurement of carrier cooling rates has been extended to other large phononic gap bulk materials including InN, demonstrating the importance of material quality. Design of structures for hot carrier cells which should be practical and realisable has developed, with the device properties more carefully specified and plans for fabricating such structures in real devices.

Energy Selective Contacts The requirement for a narrow range of contact energies can be met by an energy selective contact (ESC) based on double barrier resonant tunnelling. Tunnelling to the confined energy levels in a quantum dot layer embedded between two dielectric barrier layers, can give a conductance sharply peaked at the line up of the Fermi level on the ‘hot’ absorber side of the contact with the QD confined energy level. Conductance both below this energy and above it should be very significantly lower. This is the basis of the current work on double barrier resonant tunnelling ESCs. Modelling of Hot Carrier Solar Cell Efficiency with Energy Selective contacts Previous work was focused on developing a new model to calculate limiting efficiency of a real HCSC based on an Indium Nitride (InN) absorber layer. InN has been chosen as a potential material because of its narrow electronic band gap for absorption of a wide range of photon energies, whilst also having a wide phonon band gap. This is good for suppression of phonon decay, and hence for slowing carrier cooling [5,6]. Calculation of limiting efficiency was performed taking into account real optical and electronic properties of InN, removing most of the ideality assumptions used in other models [3,4,7,8]. The detailed band structure of wurtzite bulk InN has been considered in performing computation of carrier densities, pseudo-Fermi potentials and II-AR time constants [9]. Results have been calculated considering ideal energy selective contacts for the HCSC, which means that contacts have a very high conductivity and a discrete energy transmission level.

Figure 2: (a) Schematic representation of energy and particle fluxes interactions used in the model (Particle fluxes - full line arrow, energy fluxes- dotted line arrow). (b) HCSC efficiency as a function of carrier extraction energy level. Parameters used are: thermalisation time = 100 ps, concentration = 1000, lattice temperature = 300K and absorber layer thickness = 50 nm. [J, F and E are current density and particle and energy fluxes as denoted by subscripts for A absorption, E emission and IA Auger processes.] Recently the limiting efficiency for the hot carrier InN solar cell has been calculated considering non-ideal ESCs. In this case the carriers are not extracted on a single energy level, but in a finite energy window. Calculations have been performed taking into account contact resistance and entropy generation effects.

Figure 3: Carriers transmission probability versus energy for (a) ideal ESC, (b) non-ideal ESC.

The flux of current travelling through the ESCs towards the cold metal electrodes can be described using the following relation.

, ,, , , , 3

min 0 0

( ) ( )z

C e h rt e h

y ze h e h T T V y z

y z

dk dkeJ T f f d d

d d

(1)

The current density in this case is proportional to the occupation probability at the two sides of the ESC. Equation (1) has been derived assuming no correlation of energy of electrons in three different directions as shown in (2). This assumption is acceptable if there is a parabolic dispersion relation at minimum energy point along the three different directions.

zkykxkkzyx zyxzyx

; (2)

Based on energy and carrier conservation, Δμ and TC at steady state are calculated.

0;0 e

EEEE

e

JFFF J

THEAIAEA (3)

[quantities as defined in Figure 2(a).]

Figure 4: (a) HCSC efficiency as a function of extraction width of ESCs for different extraction energies ΔE. (b) HCSC efficiency as a function of extraction energy ΔE for different ESCs energy width. Thermalisation time is 100 ps, lattice temperature is 300 K. Absorber layer thickness is 50 nm.

The maximum efficiency has been found for a ΔE between 1.15 eV and 1.2 eV with a transmission energy window δE of 0.02 eV. The value of limiting efficiency is 9.6% compared to 43.6% calculated in the previous section using ideal ESCs. The drop in efficiency is mostly due to the decrease of open circuit voltage related to the

decreased extraction level, equation (3). This is partially compensated by an increase in extracted current due to increased II rate.

Figure 4 (a) shows calculated efficiency as a function of for several values of extraction energy. In all the curves two different trends can be identified. If the value of δE is too close to zero, the efficiency is very low due to low carrier extraction, thus a very small value of short circuit current. The conductivity of the contact in this case is indefinitely large. Enlarging δE, the number of carriers available for extraction increases, improving JSC, and so the maximum efficiency. In general the efficiency peak has been found for values of δE from 0.02 eV to 0.1 eV depending on the extraction energy ΔE. For the configurations which show higher efficiencies, ΔE < 1.35 eV, the optimum value of δE goes from 0.02 eV to 0.05 eV. This optimum value for δE of between is very close to kTRT. This represents the variation in energy in the contacts such that approximately the kTRT will inevitably be lost anyway by carriers thermalising within the contacts. Thus it sets a lower limit on a reasonable δE. Therefore this result indicates that the transmission energy range has to be very small and confirms once again the high selectivity requirements of ESCs for HCSC [10].

In Figure 4 (b) the value of maximum efficiency as a function of ΔE is reported for different values of δE. It can be observed that for small transmission energy window the extraction energy which allows maximum efficiency is lower compared to the one calculated using ideal ESCs. This effect is related to the higher occupancy at lower energies, which increases the value of JSC for contacts with a small transmission window. Modelling of QD structures for Energy Selective Contacts We have developed a robust 2 dimensional model for describing the transport properties through quantum dot structures and have used this model to understand the filtering characteristics of Energy Selective Contacts (ESCs). In this way we are able to compute the effective filtering in 2 dimensions by running numerical simulations. The model is developed from discretized Schrödinger equation by considering the sample volume as a collection of discrete points and using effective mass approximation method over the entire volume. During fabrication of the quantum dots in a dielectric matrix for selective energy contacts, different kinds of irregularities can be present in the structure, the major disorders being configurational (disorders in the position of the dots) and morphological (disorders in the size of the dots). The extent to which configurational and morphological disorders determine the electrical properties of the overall structure is investigated using simulation runs of the model. The disorders are assumed to follow a normal distribution from the mean position and size. The results show the outcome of an average of 1000 simulation runs with different standard deviation (σ) values.

Figure 5: Peak at 1.205eV of the conductance vs. energy plot for different values of σ for configurational disorder (left). Conductance vs. energy plot with peak at 1.573 eV for different values of σ for morphological disorder (right). Conductance given in units of 2e2/h. Figure 5 shows the simulation results for resonance in 2.6nm Si dot in SiO2 matrix under different orders of configurational disorders. As the disorders increases from σ =0 to σ =1 the conductance decreases by 53%, but the resonant energy remains the same at 1.205eV. This shows the confined energy in the QDs does not change as their size is fixed but the effective filtering reduces dramatically. Figure.2 shows the simulation results for a 2.2 nm Si dot in a SiO2 matrix under different orders of morphological disorders. As the disorder increases from σ =0 to σ = 1, the conductance decreases by 60% and the resonant peak remains the same at 1.573eV. But the morphological disorders cause major impact compared to configurational disorders because of the widening of the energy selection window. This is due to the distribution in size of the QDs, since QDs with different sizes have different resonant energies which are slightly different from the mean resonant energy, the average of them all increase the spread of the resonant peak and thus reducing the efficiency of the double barrier structure as energy selective contacts.

Figure 6: Planar view of the double barrier structure formed by Si QDs in SiO2 matrix, with SiC barriers of 2nm width [left]. Mean diameter of QDs is 1.8nm. I-V

characteristics for the double barrier structure with SiO2 and SiC barriers at temperature 10K and 300K [right]. A planar representation of the double barrier structure is shown as in Figure 6 (left). The barriers are usually high band gap dielectric materials like SiO2, Si3N4 or SiC. SiC barriers have advantages over SiO2 barriers that SiC barriers in double barrier structure makes a very good diffusion barrier for Silicon during processing, which can yield a structure with uniform QD size in the transport direction thus ensuring sharp resonances at the resonant energies of the structure. Since the conductance of the structure is more in the case of SiC barriers because of the lower barrier height of SiC, thicker barriers can be used to give the same conductance as SiO2, for which the very thin barriers are difficult to fabricate. Figure 6 (right) shows the I-V characteristics for the double barrier QD structure with SiO2 dielectric and SiC barriers at temperature 10K and 300K. As can be seen, the structures with SiC potential barriers have higher current density at both low temperature and at room temperature. As the temperature increases, the current density as well as the width of the resonant peak increases. Also, the negative differential resistances appear at slightly lower voltage in structures with SiC barrier than that in structures with SiO2 barrier. This is due to the increased leakage and thus the lower confinement energy for the SiC. Characterisation of Energy Selective Contacts using ‘Optically assisted I-V’ Optically assisted IV (oa-IV) is a new characterisation method which investigates an energy selective contact (ESC) fabricated on a Hot Carrier Absorber material with optical excitations in a pre-defined wave length range. If a very large optical generation rate Gopt exists adjacent to the ESC, it can probe this HC population immediately next to the contact. Alternatively, if the HC population within the Absorber is known, the ESC can be tested for its energy selectivity. The basic principle is to provide free carriers with additional energy by optical means, see Figure 7. The resonant level of the ESC can then be reached by carriers at a lower bias field, shifting the tunnelling resonance of the ESC – here a QD array – to lower bias voltages Vbias. Samples based on Si were provided by the University of Freiburg, see Figure 8. Si is very far from being an ideal Hot Carrier Absorber due to the complete lack of a gap between optical and acoustic phonon modes and hence no propensity to slow carrier cooling. Furthermore, Si has an indirect band gap in the spectral range of optical excitation which renders the interpretation of measured data to be rather complex due to the absorption coefficient Si changing over two orders of magnitude. However, for initial proof of concept it is adequate providing the excitation of hot electron-hole pairs in the material is sufficiently large and collection can be made from very close to the ESC. A Xenon (Xe) arc lamp is a continuous light source with one of the most constant photon fluxes h in the spectral range 1000 to 500 nm. Nevertheless, there are significant deviations in particular in the long wavelength region (> 800 nm). We therefore compare oa-IV curves which have the same Vbias for the current transition from forward to reverse direction, Vbias(j 0), indicating a constant carrier product (constant quasi-Fermi level positions) at the ESC. Under these conditions, the current density as a function of quasi-Fermi level position can be assumed to be identical for these curves, and this current density can be treated as a constant offset for each Vbias value.

Figure 7: Principle of oa-IV, shown for electron injection into the sample (left). If free carriers obtain additional energy by a massive photon flux significantly above the band gap, a smaller bias field is required to reach a resonant condition. Sample stage of oa-IV (right); wavelength range indicated. Every 50 nm spectral range has a photon flux of ca. 50 Suns.

Figure 8: Sample structure for oa-IV (top). Contact area on QD array is 1 mm2, with 1010 QDs. Optical generation rate Gopt within ballistic transport range to the ESC and its relative change for photon wave length intervals used in oa-IV (middle). For these, the absorption coefficient of c-Si Si and the photon flux of the Xe arc lamp h were used (bottom). Full green rectangles show range of measurements which can be compared. Dashed green rectangles show measurements which cannot be compared, but approach the high optical injection limit. Please observe breaks on left ordinate in both graphs.

250 – 500 nm 500 – 550 nm 550 – 600 nm

600 – 650 nm 650 – 700 nm 700 – 750 nm

For liquid Nitrogen (LN2, T = 77 K) measurements, resonant tunnelling behaviour (negative differential resistance – NDR) was detected for all spectral ranges. With the method mentioned above, only the oa-IV curves from 800 – 750 nm and 650 – 600 nm with h 50 Suns for each interval could be directly compared to each other. In addition, we evaluated the curves with optical excitations in the range of 1000 – 500 nm (h 600 Suns) and 550 – 250 nm (h 140 Suns). The former is near the practical high concentration limit for solar cells of 1000 Suns. As shown in Figure 9, for the curves in the 800 to 600 nm region at LN2 temperature, the potential difference between Vbias(j 0) and the average voltage of the resonance (VRes,avg) shows a decrease of 48 mV with increasing photon energy (decreasing range). As carrier densities (quasi-Fermi level positions) are identical for these curves, this potential drop corresponds to an increase in free hole energy of 48 meV. As this energy difference is equal to the effective kT, it can be interpreted as an increase in carrier temperature of T = 560 K when going from 800 – 750 nm (1.55 – 1.65 eV) to 650 – 600 nm (1.91 – 2.06 eV). Even at h 50 Suns, this effect is small. This is due to Si being a very unsuitable absorber material, further corroborated by the ratio of peak to valley current densities (quality factor of NDR – QFNDR) dropping from 2.2 to 1.5. Si does not have a phononic band gap so that optical phonons undergo a rapid decay which tremendously accelerates carrier cooling. For the 1000 – 500 nm and 550 – 250 nm ranges, the QFNDR increases significantly to 4.1 and 3.2, respectively. The potential difference Vbias(j 0) – VRes,avg is shifted by 365 and 200 mV, respectively; a quantitative comparison of both curves to each other or to the curves in the spectral range from 800 – 600 nm is thus not feasible. However, it shows that the energy selectivity may increase under high concentration h. For all oa-IV curves, a tunnelling feature for electron extraction can be seen at 760 mV forward bias. However, the background current density is much higher than the tunnelling portion of the current density in this forward bias range, so no NDR can be seen.

Figure 9: oa-IV results of a typical sample with optimum barrier thicknesses and QD size at LN2 temperature (left) and at room temperature (right). Curves in the wave length intervals from 800 to 600 nm can be compared due to their identical offset of Vbias to absolute zero Vbias., showing the free carrier product to be constant. Curves for 550 – 250 nm and 1000 to 500 nm show maximum optical injection limit. Both curve groups printed in bold. At room temperature (T = 300 K), NDR was detected only for ranges 600 nm, while the impact of resonant electron injection is still visible for curves in the 800 –

600 nm range. The background current density in the bias range around the tunnelling event increases by about an order of magnitude when going from LN2 to room temperature. This is due to thermal activation of trap states and increased thermal scattering during carrier transport through the QD array. It has the same effect on NDR suppression as explained above for electron injection under forward bias. The QD arrays were processed by segregation anneals, comprising some 1010 QDs as ESCs under the contact pad. It is very encouraging to see resonant tunnelling transport even at room temperature in a material system which was not grown epitaxially. At LN2 temperature (77 K), an increase in carrier temperature of T = 560 K was detected when increasing the photon energy from 1.6 0.05 eV to 1.985 0.075 eV under a constant photon flux of h 50 Suns. More suitable Hot Carrier Absorber materials will be investigated. Hot Carrier Absorbers: slowing of carrier cooling Carrier cooling in a semiconductor proceeds predominantly by carriers scattering their energy with optical phonons. This builds up a non-equilibrium ‘hot’ population of optical phonons which, if it remains hot, will drive a reverse reaction to re-heat the carrier population, thus slowing further carrier cooling. Therefore the critical factor is the mechanism by which these optical phonons decay into acoustic phonons, or heat in the lattice. The principal mechanism by which this can occur is the Klemens mechanism, in which the optical phonon decays into two acoustic phonons of half its energy and of equal and opposite momenta [11]. The build up of emitted optical phonons is strongly peaked at zone centre both for compound semiconductor due to the Frölich interaction and for elemental semiconducotrs due to the deformation potential interaction. The strong coupling of the Frölich interaction also means that high energy optical phonons are also constrained to near zone centre even if parabolicty of the bands is no longer valid [12]. This zone centre optical phonon population determines that the dominant optical phonon decay mechanism is this pure Klemens decay. Properties required of a Hot Carrier Absorber

The properties required for a good hot carrier absorber material are listed below in order of priority.

1. Large phononic band gap (EO(min) - ELA)* - in order to suppress Klemens decay of optical phonons this must be at least as large as the maximum acoustic phonon energy. Hence a large mass difference (or large force constant difference) between constituent elements is required [13].

2. Narrow optical phonon energy dispersion (ELO - EO(min))* – in order to minimise the loss of energy to TO phonons by Ridley decay. This requires a high symmetry atomic or nano-structure, preferably cubic with degenerate optical phonon energies at zone centre.

3. Small electronic band gap (Eg) and a reasonably high DOS for all energies above Eg – so as to allow a wide range of photon absorption. Eg should be less than 1eV. For 1 sun concentration the optimum is 0.7eV - as a band gap

* ELO , EO(min) & ELA are the maximum optical, minimum optical and maximum acoustic phonon energies, respectively.

below this energy gives no advantage in the balance between absorption and emission. As the concentration ratio increases this optimum band gap decreases to zero at maximum concentration [2].

4. A small LO optical phonon energy (ELO). This reduces the amount of energy lost per LO phonon emission, requiring a greater number to be emitted for a given energy loss [14]. However, it is difficult to have both a small ELO and the large phononic gap required in 1. This condition requires further investigation, but does still argue for a small ELO.

5. An absolute small maximum acoustic phonon energy (ELA). This maximises the phononic gap if ELO is also small. A small ELA requires a large mass for the heavy atom and/or small force constant for its bonds.

6. Good renormalisation rates in the material, i.e. good e-e and h-h scattering (e=electron, h=hole). This requires a reasonable DOS at all energies above Eg, or at least only very narrow gaps between energy levels, ≤ kT300K. This being similar to requirement 3. This in turn requires a good overlap of wavefunction for carriers through the material, i.e. poor electronic confinement or a reasonable conductor. This condition is met in all inorganic semiconductors quite easily, with e-e scattering rates of less than 100fs for reasonable carrier concentrations. It may not be met in organic semiconductors or in nanostructures with large barrier heights.

7. Good carrier transport in order to allow transport of hot carriers to the contacts. This is similar to 6, except that it only need be in the direction of the contacts, probably the z growth direction. A reasonably low resistance is probably good.

8. It should be possible to make good quality, highly ordered, low defect material. Preferably it should be easy and cheap to do this.

9. Earth abundant and readily available and processable constituent elements and processes.

10. No or low toxicity of elements, compounds and processes. [(ELO , EO(min) & ELA are the maximum optical, minimum optical and maximum acoustic phonon energies respectively.]

Suppression of phonon decay in bulk materials In some bulk semiconductors, with a large difference in their anion and cation masses, there can be a large gap between the highest acoustic phonon energy and the lowest optical phonon energy, possibly large enough to block operation of this Klemens mechanism, which can be termed a ‘phononic band gap’. Work previously presented [13] using a simple 1D force constant model and complemented by high accuracy DFT computation [14] indicated that GaN, InN and InP all have large phononic band gaps, which are close to those found experimentally [15]. We are using time resolved photoluminescence (tr-PL) to investigate the carrier cooling rates in these materials. Time resolved photoluminescence measurements of bulk phononic band gap materials (with USyd node) The potential efficiency boost, which can be achieved by Hot Carrier solar cells, is directly related to the possibility of extracting high energy carriers from the absorber layer before thermalisation, increasing the voltage and hence the conversion efficiency. The poor conversion efficiency of photons with energies above the band

gap of the absorber is the main loss mechanism in conventional single junction solar cells. The investigation of thermalisation time constants of hot carriers is a crucial step towards the engineering of Hot Carrier cells. The efficiency of an InN based hot carrier solar cell has been calculated using a complex theoretical model, (see section above on Hot Carrier cell efficiency modeling). It was found that the limiting efficiency is strongly related to hot carriers relaxation velocity in the absorber [9].

In a strong collaboration with the Sydney University (USyd) node, a comparison of femtosecond time resolved photoluminescence (tr-PL) spectroscopy between InP and GaAs was reported in last year’s annual report and is also now published [16]. This showed a distinctly longer carrier cooling time constant for the wide phononic gap InP as compared to almost zero phononic gap of GaAs. It also showed further evidence for excitation into higher side valleys for both GaAs and InP for appropriate excitation wavelengths. In 2010 the hot carrier cooling in InN has also been investigated using tr-PL.

Indium nitride (InN) also has a wide band gap between the optical and acoustic branches of its phonon dispersion relation (wider than that for InP) [15]. The main mechanism for carrier energy thermalisation in InN is the interaction of highly energetic electrons and holes with LO phonons (as it is in most semiconductors) [17]. However the large phonon band gap means that Klemens decay of optical phonons into acoustic phonons should be suppressed. Thus allowing the optical phonon population to increase and enhances the “Hot Phonon” or “Phonon Bottleneck Effect” thus promoting the re-absorption of optical phonons and slowing carrier cooling [18].

Figure 10: Schematic representation of time resolved photoluminescence arrangement.

The decay of hot carriers in InN for different excitation wavelengths has been investigated using tr-PL as in the measurement configuration shown in Figure 10. In this technique a laser pulse acts as a switching gate relating the photoluminescence signal to the time domain. The PL signal is collected from the sample, after a femtosecond laser excitation, and focused in a non linear crystal. The gate signal is generated from the same laser and is focused on the same crystal after passing through an optical delay stage. The signal is detected using a monochromator and a PMT. Our system configuration provides 150 fs pulses with tunable wavelength over a range of 256 nm (4.84 eV) to 2.6 µm (0.48 eV).

Figure 11: 3D representation of time resolved PL data for InN, showing the changing PL intensity spectrum with probe time.

Figure 11 is a three dimensional representation of PL as a function of time for all the probed wavelengths for the InN sample. It can be observed that the PL sharply rises when the carriers are photo-excited by the laser pulse. The fast decay of the PL shows the thermalisation of carriers towards the band edges. The decay is faster for highly energetic carriers compared to carriers closer to the bandgap. Thus the carrier population quickly degenerates towards the band edges during the thermalisation process.

To investigate the rate of the carrier cooling process, the effective temperature of the carrier population has been calculated fitting the high energy tail of the PL spectrum for each time during the cooling transient. The PL has been fitted assuming that carriers form a Boltzmann-like distribution within a few femtoseconds, using the following equation.

CB

G

Tk

EL

exp)()( 2

L(ε) represents the measured PL intensity at energy ε, α(ε) is the measured sample absorption coefficient, EG is the InN energy gap, 0.7 eV in this case, and kB is the Boltzmann constant. TC is the fitted parameter and represents the hot carriers’ temperature. Figure 12 shows the carrier temperature transient, which seems to follow a single exponential fit very well (blue line – fit).

Figure 12: Carrier relaxation curves for InN. The photoexcited carrier density is 1.5 x 1019 cm-3. The blue dashed curve is a single exponential fit.

The fitting of the calculated temperature data has been performed using a single exponential.

Kt

CtTTH

exp)(

Here τTH represents the carriers’ thermalisation time constant, whereas C and K are two constant parameters. The fitted value for τTH is 7 ps. This long cooling constant can be attributed to the hot phonon bottleneck effect due to the long lifetime of the A1(LO) phonon [16,17]. This hot carrier relaxation velocity is still too high to achieve a considerable efficiency gain in an InN hot carrier solar cell. However, it has been demonstrated that, for InN, the carrier cooling velocity is strictly related to the quality of the material and slower carrier cooling constants compared to the ones calculated in this study have been reported in the literature [19]. Hot Carrier Absorber: Choice of Materials

Indium nitride has most of the properties outlined above for a hot carrier absorber and also demonstrates some slowed carrier cooling rates. However it does lacjk three important properties: numbers 4. A small ELO , 8. Ease of making quality material and most importantly 9. Abundance of constituent elements. This last limitation makes the search for alternative materials to InN, which nonetheless retain its beneficial large phonon gap and other properties, important. Analogues of InN

As InN is a model material, but has the problems of abundance and bad material quality, another approach is to use analogues of InN to attempt to emulate its near ideal properties. These analogues can be II-IV-nitride compounds, large mass anion III-Vs, group IV compounds/alloys or nanostructures. Specific materials in each of these groups are being investigated further and will be measured for phonon dispersion and carrier cooling properties in the near future.

Nanostructures for absorbers Nanostructures offer another possibility for replacement of InN as an absorber material. Nanostructures allow modification of the phonon dispersion of a composite

material. III-V compounds or indeed most of the cubic and hexagonal compounds can be considered as very fine nanostructures consisting of ‘quantum dots’ of only one atom (say In) in a matrix (say N) with only one atom separating each ‘QD’ and arranged in two interpenetrating fcc lattices. Modelling of the 1D phonon dispersion in this way gives a close agreement with the phonon dispersion for zinc-blende InN extracted from real measured data for wurtzite material [13]. Similar ‘phonon band gaps’ should appear in good quality nanostructure superlattices, through coherent Bragg reflection of modes such that gaps in the superlattice dispersion open up [13]. There is a close analogy with photonic structures in which modulation of the refractive index in a periodic system opens up gaps of disallowed photon energies. Here modulation of the ease with which phonons are transmitted (the acoustic impedance) opens up gaps of disallowed phonon energies. Force constant modelling of III-V QD materials by SK growth 3D force constant modelling, using the reasonable assumption of simple harmonic motion of atoms in a matrix around their rest or lowest energy position, reveals such phononic gaps [20]. The model calculates longitudinal and transverse modes and can be used to calculate dispersions in a variety of symmetry directions and for different combinations of QD sub-lattice structure and super-lattice structure. III-V Stranski-Krastinov grown QD arrays of InAs in InGaAs and InGaAlAs matrices are fabricated by Professor Okada’s group at the University of Tokyo using MBE. We are investigating these for evidence of phonon dispersion modulation and potential slowed carrier cooling. In order to understand the expected phonon dispersions these are being modelled using the 3D force constant technique. Lattice matched and strain compensated material pairs that may produce large phonon band gaps are of interest. Previous iterations in the design of these structures indicated the importance of separating “light” and “heavy” atoms to different parts of the nanostructure. Initially, the lightest atom in the system, As, was present in both the QD and the matrix. This meant that the reduced mass of both regions (proportional to the sum of the inverses of each atomic mass) was very similar as the light element dominates in this case.

Figure 13: Schematic of the simple hexagonal superlattice of InAs QD disks, showing their relative dimensions. The red square indicates orthorhombic repeat unit for the simple hexagonal system. On this iteration structures with an In0.5Ga0.5-xAlxAs matrix (with x=0.4) and InAs QDs were grown. Significant Al content was introduced into these structures with the expectation that this light element, segregated to the matrix material, might produce

appreciable phonon bandgaps. A schematic of the structure is shown in Figure 13. Some images derived from characterisation of the structures grown at the University of Tokyo are presented in Figure 14. The superlattice of QDs has a simple hexagonal structure. Extraordinary periodic out-of-plane stacking is achievable and largely defect free structures can be grown on the order of microns.

Figure 14: Some images of the InAs QD structure as grown using S-K MBE growth at the University of Tokyo. (Above): the near-perfect stacking of the structure is shown. (Above right): in-plane InAs QD arrangement. (Right): simple hexagonal superlattice structure showing stacking. The QDs are similar to flattened disks with in plane dimensions of 40-50 nm and heights of 7 nm. Force constant modelling of this structure predicts an appreciable phonon band gap, as shown in Figures 15 & 16. This band gap is due almost entirely to the presence of the Al in the system. A small band gap is present due to the mass difference between In, Ga and As, but it is less than a quarter of the size shown in Figures 15 & 16. Due to computational constraints the size of the QDs is quite small, only about one nanometre in diameter. While actual sizes for these structures are too computationally intensive to model without extreme effort, recent modelling with gradually increasing size suggests that the dispersion relations scale linearly with size. That is, once the discrete distances (bond lengths) are small relative to the superlattice unit cell dimension, the dispersion should look exactly the same when scaled such that the relative dimensions are preserved. This will be investigated in greater detail in further work.

Figure 15: Dispersion curves for InAs QDs in a simple hexagonal SL. The matrix is In0.5Ga0.5-xAlxAs with (x = 0.4). There is a large phonon bandgap present in the system due to the presence of Al. The gap seems very tolerant to Al location. (Top): <100>, (Middle): <110>, (Bottom): <111> directions.

Figure 16: Densities of states (DOS) for the dispersion curve immediately to the left. Note small changes in the DOS could be due to pseudo-random locations of the Ga and Al particles or to differences in crystalline direction. (Top): <100>, (Middle): <110>, (Bottom): <111> directions respectively.

Fabrication and characterisation of highly ordered nanoparticle arrays for Hot Carrier absorbers As demonstrated by the work on modelling phonon modulation, periodic core-shell QD arrays offer a way to significantly change the phonon modulation in a superlattice because the core and shell can be of materials of very different force

constant, directly leading to a strong phonon confinement. Deposition methods are being investigated to fabricate such highly ordered QD arrays. The Langmuir-Blodgett (LB) technique fabricates thin films from colloidal dispersions of quantum dots (QDs). The LB technique allows for controllable, uniform film formation and subsequent transfer onto a solid substrate, such as quartz [21]. Through surface passivation of the QDs, interparticle spacing can also be controlled. Figure 17 illustrates the set-up of the LB apparatus and the functionalized QDs.

Figure 17: Langmuir-Blodgett set-up for film fabrication with surface functionalized Si NPs.

In 2010 work has focused on fabrication of periodic thin films of Si nanoparticles (NPs) provided by the University of Minnesota. The Si NPs under investigation are surface passivated with dodecane (C12H26), a long chain hydrocarbon rendering the surface hydrophobic. The particles were dispersed in chloroform and investigated using TEM, UV-Vis spectroscopy and photoluminescence (PL). Using TEM imaging the average diameter was found to be 3.0±0.5 nm. Figure 18 (left) shows a droplet of solution (dispersed in chloroform) and the inset highlights the lattice fringes confirming the presence of silicon - a 3.14 Å lattice spacing corresponding to the {111} planes.

Figure 18: (Left) TEM images of Si NPs dried on a grid. (Right) TEM image of Si NPs monolayer deposited at 50 mN/m of SP. Monolayer structures of Si NPs have been successfully deposited using the LB method and characterisation has been on-going using TEM, UV-vis, PL and Raman spectroscopy. Figure 18 (right) illustrates TEM image of the Si films fabricated at 50 mN/m surface pressures (SP) of the water trough. As noted, high contrast is difficult

to achieve due to the small density difference of Si and the carbon-passivating chains. As a result, to further quantify the coverage and regularity, computer recognition software is required; on which work is on-going. Nonetheless, Figure 18 (right) appears to show a dense packing of Si NPs. UV-Vis spectroscopy was used to study the optical absorption of the films. As illustrated in Figure 19 (left) for single layers of Si NPs at varying compression surface pressures (SP), the absorption clearly increases when SP increases, corresponding to higher packing densities of Si NPs. The PL spectra of the above films show narrow and symmetric peaks at 1.6 ~1.7 eV, as shown in Figure 19 (right), which are in good agreement with the estimated band gap using the ‘effective mass approximation’. The red shift in emission energy from Si NPs films as compared to the solution corresponds to the electronic energy transfer from small particles to larger ones when they are closely packed [22] and to the overlap of wavefunction and mini-band formation with subsequent energy level broadening of the closely packed thin film.

Figure 19: (Left) Optical absorption of Si monolayers deposited at various surface pressures (SP). (Right) PL of Si NPs films at varying SP, as compared to the Si NPs solution. Investigation of the electronic and phononic properties of the monolayer and multilayer structures are on-going using low-wavenumber Raman and time resolved PL spectroscopy. Implementation of nanostructures for Hot Carrier cells In order to use a nanoparticle array as a Hot Carrier absorber, and hence utilise slowed carrier cooling due to modified phonon dispersion, then a way to fabricate a complete structure with ESCs must be established. A structure in which nanoparticles are arranged in a uniform 3D array should give the required phononic band gap. The degree to which this needs to be ordered is still under investigation. Certainly a large difference between masses is beneficial between the nanoparticles and the matrix and this can be best achieved in an array of ‘core shell QDs’ in which the core and shell have very different acoustic impedance in order to promote coherent reflection and hence confinement of phonons [13]. The matrix in which such an array is embedded needs to allow transport of carriers to contacts and also electron-electron and preferably hole-hole scattering to renormalise carrier energies. Such a structure should also have a narrow electronic band gap so as to absorb a wide range of photon

0.5 1.5 2.5 3.5 4.5Energy (eV)

absorption (a.u)

40mN/m

60mN/m

65mN/m

collapse

energies. This combination of properties is challenging but not mutually exclusive because phononic properties are largely independent of electronic properties. Energy selective contacts are also required for such a structure. These would most likely be arranged at the top and bottom of the absorber. (A hole contact might not need to be selective because the hole population only contains a small fraction of the hot carrier energy, due to their larger effective mass, and hence thermalisation of holes is less important.) These ESCs would be QD or QW double barrier structures. (Addition of an extra layer to give double QD/QW triple barrier structures should also give rectification at the contact.) Such a QD array / double barrier QD ESC structure is shown schematically in Figure 20. Figure 20: Conceptual Integrated Device. Summary of work on Hot Carrier solar cells 2010 has seen significant development in most areas of the Hot Carrier solar cell work on the project. The modelling of efficiencies not only now includes real material parameters for highly promising absorber materials such as InN, but also is now extended to Energy Selective contacts of finite width. The transport across such contacts is further modelled for a range of QD matrix combinations. Work on ESCs has seen further more detailed demonstration of the necessary resonance in double barrier resonant tunnelling structures, with additional evidence for hot carrier populations, albeit very small, from illuminated I-V measurements with the ‘optically assisted I-V’ technique. Measurement of InN with time-resolved PL has indicated some evidence for slowed carrier cooling, further corroborating the importance of a large phonon band gap to block optical phonon decay, but also highlights the importance of material quality. Modelling of nanostructurs for absorber materials has focussed on real III-V QD structures, showing phonon bandgaps which will soon be measured in phonon dispersion measurements. Progress on the Langmuir-Blodgett approach to ordered nanoparticle arrays has seen development of ordered single layer arrays of Si nanoparticles. The potential application of nanostructures to fully integrated devices has started to be investigated conceptually, with various designs considered.

Selective energy contact: double (or triple) barrier QD resonant tunnelling

Selective energy contact: selecting opposite carrier type

Slowed carrier cooling absorber layer: Periodic QD array tuned to block optical phonon

decay by Klemens mechanism; Core shell QD for ease of tuning; Must also allow renormalization of carrier energy by

e-e & h-h scattering. Proof of concept with MOVPE/MBE III-V QD array. Transfer to thin film or self-assembled QD array.

External contact

External contact

Further Work Modeling work on efficiencies, energy selective contacts and phononic properties will be continued. Measurements on test structures for phonic properties will be extended to other materials. These will include time resolved PL for carrier cooling and high resolution Raman for phonon density of states. New nanostructures will be investigated, including III-V quantum well and quantum dot nanostructures with collaborators. These will build on the very promising results obtained for reduced carrier cooling measured in these nanostructures with the USyd node. Further work on colloidal nanostructures will investigate multi-layer uniform quantum dot arrays, and investigate their optical and electrical properties. Using both Si QD arrays and other core shell QD structures. Further work on energy selective contact materials will be carried out in which the quality of materials is improved, the quality of barriers and size uniformity of QDs, as well as dounble and triple barrier QW structures. Building on the initial work on complete devices, complete device structures incorporating both absorber and energy selective contacts will be fabricated, by collaborators and in-house, and measured for carrier cooling and electrical illuiminated I-V properties, with the aim of demonstrating a true Hot Carrier cell effect. Publications in past year of the project Refereed Journal papers:

1. P. Aliberti, Y. Feng, Y. Takeda, S. K. Shrestha, M. A. Green, G. Conibeer, Investigation of theoretical efficiency limit of hot carriers solar cells with a bulk indium nitride absorber, Journal of Applied Physics, 108 (2010) 094507.

2. R. Patterson, M. Kirkengen, B. Puthen Veettil, D. Konig, M.A. Green, G. Conibeer, Phonon lifetimes in model quantum dot superlattice systems with applications to the hot carrier solar cell, Solar Energy Materials and Solar Cells, 94 (Nov 2010) 1931-1935.

3. R. Clady, M.J.Y. Tayebjee, P. Aliberti, D. Konig, N.J. Ekins-Daukes, G.J. Conibeer, T.W. Schmidt, M.A. Green, Interplay between Hot phonon effect and Intervalley scattering on the cooling rate of hot carriers in GaAs and InP, Progress in Photovoltaics, accepted March 2011.

4. Y. Takeda, T. Motohiro , D. König , P. Aliberti, Y. Feng , S. Shrestha , G. Conibeer, Practical factors lowering conversion efficiency of hot carrier solar cells, Applied Physics Express, Applied Physics Express 3 (2010) 104301.

5. P. Aliberti, S.K. Shrestha, R. Teuscher, B. Zhang, M.A. Green, G.J. Conibeer, Study of Silicon Quantum Dots in a SiO2 Matrix for Energy Selective Contacts Applications, Solar Energy Materials and Solar Cells, 94 (2010)1936-1941.

6. G Conibeer, R Patterson, L Huang, J-F Guillemoles, D Konig, S Shrestha, MA Green, “Modelling of hot carrier solar cell absorbers”, Solar Energy Materials and Solar Cells, 94 (2010) 1516-1521. doi:10.1016/j.solmat.2010.01.018.

7. Santosh K. Shrestha, Pasquale Aliberti, Gavin J. Conibeer, Energy selective contacts for hot carrier solar cells, Solar Energy Materials and Solar Cells, 94 (Sept 2010) 1546-1550.

8. D. König, K. Casalenuovo, Y. Takeda, G. Conibeer, J.F. Guillemoles, R. Patterson, L.M. Huang, M.A. Green, “Hot Carrier Solar Cells: Principles, Design and Materials”, Physica E: Low-dimensional Systems and Nanostructures, 42 (2010) 2862-2866.

Refereed Conference papers:

9. P. Aliberti, B. Zhang, M.A. Green. G. Conibeer, S. Shrestha, A. Hsieh, R. Li, B. Puthen-Veettil, Australian Solar Energy Society conference, Solar2010 (Canberra Dec 2010), “Investigation of single layers of silicon quantum dots in SiO2 matrix for energy selective contacts in hot carriers solar cells”, December 2010.

10. G. Conibeer, Australian Solar Energy Society conference, Solar2010 (Canberra Dec 2010), “Third Generation Photovoltaics”, December 2010.

11. A. Hsieh, S. Shrestha, M.A. Green. G. Conibeer, Australian Solar Energy Society conference, Solar2010 (Canberra Dec 2010), “Investigation of Si/SiO2 quantum well structures for selective energy contacts in hot carrier solar cells”, December 2010.

12. F. QI, G. Conibeer, S. Shrestha, S. Huang, Australian Solar Energy Society conference, Solar2010 (Canberra Dec 2010), “The study of potential application of Group IV alloys in photovoltaic field”, December 2010. Un-refereed Conference papers:

13. M. A. Green, G. Conibeer, D. König, S. Shrestha, S. Huang, P. Aliberti, L. Treiber, R. Patterson, B. Puthen Veettil, A. Hsieh, A. Luque, A. Marti, P.G. Linares, E. Cánovas, E. Antolín, D. Fuertes Marrón, C. Tablero, E. Hernández, J-F. Guillemoles , L. Huang, T. Schmidt, R. Clady and M. Tayebjee, “Recent Progress with Hot Carrier Solar Cells”, 25th European Photovoltaic Solar Energy Conference and Exhibition (25th EU PVSEC) / 5th World Conference on Photovoltaic Energy Conversion (WCPEC-5), Session: 1AP.1.1, Abstract No. [1191], Valencia, Spain, 6-10 September, 2010 (plenary).

14. P. Aliberti, Y. Feng, R.Clady, M.J.Y. Tayebjee, T.W. Schmidt, S.K. Shrestha, M.A. Green, G.J. Conibeer, “On efficiency of hot carriers solar cells with a Indium nitride absorber layer”, Oral Presentation, Proceedings of European photovoltaic conference, Valencia, Spain, 6-10 September 2010.

15. A. Hsieh, S. Shrestha, M.A. Green, G. Conibeer, ‘Investigation of Si/SiO2 quantum well structures for energy selective contacts in hot carrier solar cells’, 5th World Photovoltaic Solar Energy Conference, Valencia, Spain, 6-10 September 2010.

16. R. Patterson, B. Puthen Veettil, M. Kirkengen, D.Koenig, M.A. Green, G.J. Conibeer, Poster Presentation 1DV.3.77, 5th World PVSEC, Valencia, Spain, 6 - 10 September, 2010.

17. B. Puthen Veettil, R. Patterson, D. König, G. Conibeer and M. A. Green, “Modelling the Effects of Doping in Quantum Dot Structures for Tandem Cells”, Poster Presentation 1DV.3.64, 5th World PVSEC, Valencia, Spain, 6 - 10 September 2010.

18. M.A. Green, G. Conibeer, D. König, S. Shrestha, S. Huang, P. Aliberti, L. Treiber, R. Patterson, B. Puthen-Veettil, A. Hsieh, Y. Feng, A. Luque, A. Marti, P.G. Linares, E. Cánovas, E. Antolin, D. Fuertes Marrón, C. Tablero, E.Hernandez, J-F. Guillemoles, L. Huang, A. Le Bris, T. Schmidt, R. Clady and M. Tayebjee, “Hot Carrier Solar Cells: Challenges and Recent Progress”, 35th IEEE Photovoltaic Specialists Conference, Honolulu, Hawaii, 20-25 June, 2010.

19. R. Patterson, Z. Wan, B. Puthen-Veettil, D. König, G. Conibeer, “Phonon Decay in Naostructure: Computational Study”, Proc. of 35th IEEE PVSC, Honolulu, Hawaii, 20 - 25 June 2010, proc. pp. 1843 (art.-no. 5615953).

Contacts Emails of the PIs are: g.conibeer@unsw.edu.au m.green@unsw.edu.au References 1. G.J. Conibeer, C-W. Jiang, D. König, S. Shrestha, T. Walsh, M.A. Green “Selective energy contacts for hot carrier solar cells”, Thin Solid Films, 516 (July 2008) 6968.

2. M. A. Green, “Third Generation Photovoltaics: Ultra-High Efficiency at Low Cost” (Springer- Verlag, 2003). 3. P. Würfel, “Solar Energy Conversion with Hot Electrons from Impact Ionisation”, Sol. Energy

Mats. and Sol. Cells. 46 (1997) 43. 4. Ross R and Nozik AJ, J. Appl Phys, 53 (1982) 3318. 5. V.Y. Davydov, et al., “Absorption and emission of hexagonal InN: Evidence of narrow

fundamental band gap”, Physica Status Solidi (B), 229 (2002) p. r1-r3.

6. D. Fritsch, H. Schmidt, M. Grundmann, “Band dispersion relations of zinc-blende and wurtzite

InN”, Physical Review B, 69 (2004) 165204. 7. Y. Takeda, T. Ito, T. Motohiro, D. König, S.K. Shrestha, G. Conibeer, “Hot carrier solar cells

operating under practical conditions”, Journal of Applied Physics, 105 (2009) 074905-10. 8. P. Wurfel, A.S. Brown, T.E. Humphrey, and M.A. Green, Particle conservation in the hot-carrier

solar cell”, Progress in Photovoltaics, 13 (2005) 277-285. 9. P. Aliberti, Y. Feng, Y. Takeda, S.K. Shrestha, M.A. Green, G.J. Conibeer, “Investigation of

theoretical efficiency limit of hot carrier solar cells with bulk InN absorber”, Journal of Applied Physics, 108 (2010) 094507-10.

10. G. Conibeer, C.W. Jiang, D. König, S.K. Shrestha, T. Walsh, M.A. Green, “Selective energy contacts for hot carrier solar cells”, Thin Solid Films, 216 (2008) 6968-6973.

11. P.G. Klemens, Phys. Rev. 148 (1966) 845. 12. G. Conibeer, R. Patterson, P. Aliberti, L. Huang, J-F. Guillemoles, D. König, S. Shrestha, R.

Clady, M. Tayebjee, T. Schmidt, M.A. Green “Hot Carrier solar cell Absorbers”, 24th European PVSEC, Hamburg (Sept. 2009).

13. G. Conibeer, N.J. Ekins-Daukes, J-F. Guillemoles, D. König, E-C. Cho, C-W. Jiang, S. Shrestha and M.A. Green, “Progress on Hot Carrier Cells”, Solar Energy Materials and Solar Cells, 2009, 93, 713.

14. D. König, K. Casalenuovo, Y. Takeda, G. Conibeer, J.F. Guillemoles, R. Patterson, L.M. Huang, M.A. Green, “Hot Carrier Solar Cells: Principles, Materials and Design”, Physica E, 42 (2010) 2862-2866.

15. H. Bilz, W. Kress, “Phonon dispersion relations in Insulators”, Springer-Verlag (1979). 16. R. Clady, M.J.Y. Tayebjee, P. Aliberti, D. Konig, N.J. Ekins-Daukes, G.J. Conibeer, T.W.

Schmidt, M.A. Green, “Interplay between Hot phonon effect and Intervalley scattering on the cooling rate of hot carriers in GaAs and InP”, Progress in Photovoltaics, accepted March 2011.

17. Y.C. Wen, Y.C., C.Y. Chen, C.H. Shen, S. Gwo, C.K. Sun, “Ultrafast carrier thermalization in InN”, Applied Physics Letters, 89 (2006) 3; M.D. Yang, et al., “Hot carrier photoluminescence in InN epilayers”, Applied Physics A - Materials Science & Processing, 90 (2008) 123-127.

18. K.T. Tsen, J.G. Kiang, D.K. Ferry, H. Lu, W.J. Schaff, H.W. Lin, S. Gwo, “Direct measurements of the lifetimes of longitudinal optical phonon modes and their dynamics in InN”, Applied Physics Letters, 90 (2007) 3.

19. F. Chen, A.N. Cartwright, H. Lu, W.J. Schaff, “Ultrafast carrier dynamics in InN epilayers”, Journal of Crystal Growth, 269 (2004) 10-14.

20. R. Patterson, M. Kirkengen, B. Puthen Veettil, D. Konig, M.A. Green, G. Conibeer, Sol. Ener. Mats. And Sol. Cells, 94 (2010) 1931-1935.

21. S. Huang, K. Minami, H. Sakaue, S. Shingubara, T. Takahagi, “Effects of the surface pressure on the formation of Langmuir-Blodgett monolayer of nanoparticles”, Langmuir, 20 (2004), 2274.

22. C. R. Kagan, C. B. Murray, M. Nirmal, M. G. Bawendi, Phys. Rev. Lett. “Electronic Energy Transfer in CdSe Quantum Dot Solids”, 76 (1996) 1517.

Hot Carrier solar cell: Implementation of the Ultimate PV Converter; Subcontract Number: 15618900-34775 IRDEP Investigators Note: all investigators are member of IRDEP but not necessarily under contract with CNRS, in which case they are not supported by the grant. Jean-François Guillemoles, Research director, CNRS, PI Christophe Domain, EDF research engineer Par Olsson, EDF research engineer Sana Laribi, EDF research engineer Laurent Lombez, Researcher, CNRS Marie Jubault, research engineer, CNRS Arthur LeBris, Graduate researcher, CNRS Hanane El Begheti, Graduate researcher, CNRS Myriam Paire, Graduate researcher, CNRS Artabaze Shams, Graduate researcher, CNRS Abstract

In this third year of the project, we have worked on the extraction and collection of hot carriers on two aspects: (i) modeling of the thermalisation in the contacts showed that the contact selectivity may not be as drastic as initially thought, and that 100 meV or even semi-selective contacts only might be enough., and (ii) devising a structure with micro-contacts able to handle the huge current densities involved. We have also continued work on measuring, simulating and reducing thermalization rates in HCSC. The modeling of the thermalization in bulk samples gave access to the experimental value of thermalization rates. Si Bulk phonon lifetimes could be well reproduced by molecular dynamics modeling of the lattice and we have simulated Si based nanostructures. This will help for searching new materials based on Si with increased phonon lifetimes. On the experimental side, we have designed new heterostructures and have worked with U. Syd. to extract thermalization data from the TR PL experiments.

Introduction

The original plan consisted, in short, to contribute to the making of a new, very efficient, type of solar cell: the hot carrier device. Ideally this device should also be made from non-toxic and earth abundant materials (e.g. Si).

Important progress has been made towards the realisation of this objective, albeit, realisation of device based on Silicon seems still difficult, meanwhile a structure and materials (III-V) have emerged that could be a proof of concept. The comparison of the SOW and the status of the item, plus the list of publication (appendix) help to demonstrate this.

Hot Carrier solar cell: Implementation of the Ultimate PV Converter; Subcontract Number: 15618900-34775 Background In previous years, we had made experimental structures and compared the advantages of several nanostructures, Quantum dots or Quantum wells based, and of various materials in terms of thermalization rates. We have also devised a method to measure those. We have also developed a simulation of phonon decay in nanostructures (based on molecular dynamics) and of electron-electron and electron-phonon interactions (based on a Monte Carlo algorithm). On our experimental samples we found that MQW samples based on antimonides were compatible with efficiencies above 50% at high solar concentration. On the simulation side, we could validate the approaches. Phonon dispersion and LO phonon decay were well reproduced in Si and several other semiconductors. Results The main thrust of this third year was devoted to understanding and designing a workable and efficient approach for the hot carrier collection.

Especially notable are the following results: - Measurements of thermalisation times in III-V heterostructures (multiple quantum wells): these measurements are the key for determining the potential efficiency of the materials and the optimal device structures. They supported claims for solar energy conversion efficiency of up to 54% with the type of material experimentally available to us under very high solar concentration (x40,000). - Modelling of phonons modes and their lifetimes in nanostructures using molecular dynamics: these models are necessary to help design materials where thermalisation can be even slower as found experimentally. We need such reduced thermalisation to enable device operation under lower solar concentration levels (below x1000, the current practical value). Modeling of the modes has been done for many nanostructures and computation of the lifetimes is being done. The method developed is very versatile and will help us to conduct a systematic search of low thermalisation rate compounds. We have recently found by this method that Si nanostructures could have LO phonon lifetimes, and therefore thermalisation times, 4 times longer (present value, certainly not a maximum) as compared to Silicon, this property directly translates in more efficient devices. - Definition of a scheme for carrier extraction: initially it was thought that the device would need very selective contacts. We proposed a structure based on semi-selective contacts, much simpler to realise in practice (it is simply a barrier of a well adjusted height) that would yield most of the efficiency gain of a fully selective contact (APL 2010). This is a very significant contribution towards the making of these devices.

Hot Carrier solar cell: Implementation of the Ultimate PV Converter; Subcontract Number: 15618900-34775 - definition of device geometry: in these devices, operating under very high solar concentration, one is fighting against series resistance for current extraction. We have proposed a device architecture (microscale solar cells) enabling to handle current densities up to and beyond x40,000 suns (Paire et al. ,JAP 2010) and demonstrated this experimentally (Paire et al. , Valencia conference 2010). These points are instrumental in enabling us to make a proof-of -concept device based on III-V heterostructures.

Hot Carrier solar cell: Implementation of the Ultimate PV Converter; Subcontract Number: 15618900-34775 Appendix:

Scientific production Refereed papers:

1. Martin A. Green Gavin Conibeer, Dirk König, Santosh Shrestha, Shujuan Huang, Pasquale Aliberti, Lara Treiber, Robert Patterson, Binesh Puthen Veettil, Andy Hsieh, A. Luque A. Marti, P.G. Linares, E. Cánovas, E. Antolín, D. Fuertes Marrón, C. Tablero, E. Hernández, J-F. Guillemoles, L. Huang, T. Schmidt, R. Clady, M. Tayebjee, “ RECENT PROGRESS WITH HOT CARRIER SOLAR CELLS ”, Progress in Photovoltaics , in press

2. Arthur LE BRIS and Jean-François GUILLEMOLES , “ Hot Carrier Solar cells: achievable efficiencies accounting for main thermalisation paths” Applied Phys. Letters, 97, (2010) 113506

3. D. Konig, K.Casalenuovo, Y.Takeda, G.Conibeer, J.F.Guillemoles, R.Patterson, L.M. Huang, M.A.Green, "Hot carrier solar cells: Principles,materials and design", Physica E, in press

4. G. Conibeer, L. Huang, J-F. Guillemoles, R. Patterson, A. Hsieh, D. Kőnig, S. Shrestha, M.A. Green, Modelling of hot carrier solar cell absorbers, Solar Energy Materials and Solar Cells 94 (2010) 1516-1521

5. Myriam PAIRE, Laurent LOMBEZ, Jean-François GUILLEMOLES and Daniel LINCOT, “Towards microscale CIGS solar cells for efficient conversion and optimized material usage : theoretical evaluation”, Jounal of Applied Physics, 108 (2010) 034907

6. Gavin Conibeer, Robert Patterson, Lunmei Huang, Jean-Francois Guillemoles, Dirk Kőnig,Santosh Shrestha, Martin A. Green, "MODELLING OF HOT CARRIER SOLAR CELL ABSORBERS", Solar Energy Materials and Solar Cells, 94 (2010) 1516–1521

7. JF. Guillemoles, "The quest for very high efficiency in Photovoltaic energy conversion" Europhysics news, 2010, 41/2, p 19-22.

8. A. Slaoui, JF Guillemoles, "Nanostructured solar cell materials for high efficiency photovoltaics: dream or reality?" Actualités chimiques, l’actualité chimique - juin 2009 - n° 331

9. Gavin Conibeer, Nicholas Ekins-Daukes, Jean-François Guillemoles, Dirk Kőnig, Eun-Chel Cho, Chu-Wei Jiang, Santosh Shrestha and Martin Green, "Progress on hot carrier cells", Solar Energy Materials and Solar Cells, 2009, 93, 713.,

10. Conibeer, G., D. Konig, M.A. Green, and J.F. Guillemoles "Slowing of carrier cooling in hot carrier solar cells". Thin Sol. Films, 516 (2008), 6948-6953

Proceedings:

1. Jean-Francois Guillemoles, A. Le Bris, M. Paire, L. Lombez, S. Laribi, D. Lincot, C.

Colin, S. Collin, JL Pelouard, M. Laroche, R. Esteban, J.J. Greffet, G. Boissier, P. Christol, "Hot carrier solar cells: the ultimate photovoltaic conversion in practice", SPIE Photonics West 2011 newsroom (invited communication)

2. Martin A. Green Gavin Conibeer, Dirk König, Santosh Shrestha, Shujuan Huang, Pasquale Aliberti, Lara Treiber, Robert Patterson, Binesh Puthen Veettil, Andy Hsieh, A. Luque A. Marti, P.G. Linares, E. Cánovas, E. Antolín, D. Fuertes Marrón, C. Tablero, E. Hernández, J-F. Guillemoles, L. Huang, T. Schmidt, R. Clady, M. Tayebjee, “ RECENT PROGRESS WITH HOT CARRIER SOLAR CELLS ”, Proceedings of the 25th European Photovoltaic Solar Energy Conference / 5th World Conference on Photovoltaic Energy Conversion , held in Valencia, Spain , 6 - 10 September 2010 , Edited by G.F. DE SANTI , H. OSSENBRINK , P. HELM , publisher WIP - Renewable Energies Munich, Germany , p 1.

3. A. Le Bris, L. Lombez, JF. Guillemoles , R. Esteban, M. Laroche, JJ. Greffet , G. Boissier , P. Christol, S. Collin, JL. Pelouard, P. Aschehoug, F. Pellé , “HOT CARRIER SOALR CELLS : IN

Hot Carrier solar cell: Implementation of the Ultimate PV Converter; Subcontract Number: 15618900-34775

THE MAKING ? “, 25th European Photovoltaic Solar Energy Conference and Exhibition / 5th World Conference on Photovoltaic Energy Conversion, 6-10 September 2010, Valencia, Spain , p 683

4. Myriam Paire, Laurent Lombez, Nicolas Péré-Laperne, Stephane Collin, Jean-Luc Pelouard, Jean-François Guillemoles, Daniel Lincot , “MICROSCALE CU(IN,GA)SE2 SOLAR CELLS FOR CONCENTRATION : A PATH TO 30% EFFICIENCY – THEORETICAL EVALUATION AND FIRST EXPERIMENTAL RESULTS”, 25th European Photovoltaic Solar Energy Conference and Exhibition / 5th World Conference on Photovoltaic Energy Conversion, 6-10 September 2010, Valencia, Spain , p 2872

5. Martin A. Green, Gavin Conibeer, Dirk König, Santosh Shrestha, Shujuan Huang, Pasquale Aliberti , Lara Treiber , Robert Patterson , Binesh Puthen Veettil , Andy Hsieh , Yu Feng , A. Luque , A. Marti , P.G. Linares , E. Cánovas , E. Antolín , D. Fuertes Marrón , C. Tablero , E. Hernández , J-F. Guillemoles , L. Huang , A. Le Bris , T. Schmidt , R. Clady , M. Tayebjee , “HOT CARRIER SOLAR CELLS: CHALLENGES AND RECENT PROGRESS ”, proceedings of the 35th IEEE PV specialist Conference, Hawai 2010, p 57

6. A. Le Bris, J.F. Guillemoles, “HOT CARRIER SOLAR CELL EFFICIENCY SIMULATION WITH CARRIER EXTRACTION THROUGH NON IDEAL SELECTIVE CONTACTS ”, proceedings of the 35th IEEE PV specialist Conference, Hawai 2010, p 61

7. A. Le Bris, J.F. Guillemoles, M. Laroche, R. Esteban, J.J. Greffet, P. Christol, P. Aschehoug, S. Ivanova, "Hot carrier solar cell: from efficiency simulation to hot carrier detection by photoluminescence", Quantsol 2009, Rauris

8. A. Le Bris, J-F. Guillemoles, P. Christol, G. Boissier, S. Ivanova, P. Aschehoug, "Hot carrier solar cell efficiency simulation with partial thermalization", 24th European Photovoltaic Solar Energy Conference, Hambourg , sept. 2009

9. Gavin Conibeer, Robert Patterson, Pasquale Aliberti, Lunmei Huang, Jean-Francois. Guillemoles, Dirk König, Santosh Shrestha, Raphael Clady, Murad Tayebjee, Tim Schmidt, Martin Green, "HOT CARRIER SOLAR CELL ABSORBERS" , 24th European Photovoltaic Solar Energy Conference, Hambourg , sept. 2009

10. Lunmei Huang, Jeremie Zafran, Arthur Le Bris, Pär Olsson, Christophe Domain, Jean-François Guillemoles, "PHONON MODES IN SI-GE NANO-STRUCTURES FOR HOT CARRIER SOLAR CELLS", 23rd European Photovoltaic Solar Energy Conference, Valencia, sept. 2008

Note: papers from previous periods are italicized.

Contacts J-F. Guillemoles : jf-guillemoles@chimie-paristech.fr

Hot Carrier Solar Cell: Implementation of the Ultimate Photovoltaic Converter (University of Sydney node) Investigators A/Prof. Timothy W. Schmidt, Associate Professor Dr Raphaël Clady, Postdoctoral Fellow Murad J. Y. Tayebjee, Graduate Researcher, Chemistry Dr N. J. Ekins-Daukes1, Senior Lecturer Abstract

The capabilities of the Joint Laboratory for Ultrafast Spectroscopy are outlined. Descriptions of the transient photoluminescence and transient absorbance experiments are given, followed by a number of applications relating to the project. GaAs and InP have been compared with regard to electron cooling phenomena, with the results accepted for publication in Progress in Photovoltaics. A short description is given of the emerging collaboration between the Sydney node and the group of Paul McIntyre (Stanford). Germanium nanowires produced by McIntyre’s group have been investigated in the Sydney laboratory. Oscillations observed in the nanowires’ transient absorption signatures are interpreted as being due to low frequency breathing vibrations. InN is investigated as a prototype hot carrier absorber. Though the cooling rates are observed to be fast, a hot carrier effect should be observable with energy selective contacts. Finally, the results of transient photoluminescence of quantum well devices shows that in these materials hot carriers can persist for more than 2 ns.

1 Department of Physics and Grantham Centre for Climate Change, Imperial College, London, UK.

Ultrafast spectroscopy Laboratory Of the course of the GCEP Project, The University of Sydney node has built up a

capable spectroscopy laboratory for the investigation of ultrafast phenomena. Processes occurring on a time-scale of several picoseconds is too fast for electronic instrumentation and thus specialist techniques must be employed. By controlling the timing between two laser pulses using a optical delay stage (1 m travel = 6 femtosecond), phenomena can be instigated and probed with the temporal precision of the laser pulses. We employ a Clarke MXR femtosecond laser to produce 150 fs pulses of 780 nm light at a repetition rate of 1 kHz. This laser beam is split to pump two independently tunable optical parametric oscillators which access the full solar spectrum, from 2.2 m to 190 nm. Two experiments have been built to measure ultrafast phenomena in candidate hot carrier materials. Time-resolved photoluminescence

The time-resolved photoluminescence TRPL experiments are undertaken using the Haclyone spectrometer (Ultrafast Systems). The experiment is illustrated in Figure 1. In this experiment, one laser pulse impinges on a sample and creates excited states of matter which then commence to luminesce. The luminescence is imaged onto a crystal which exhibits a quadratic term in its polarizability. Such a crystal is capable of adding the frequency of contemporaneous light pulses. When the PL coincides with a gate pulse of light, then the upconverted light with the sum of the frequency of the gate and the PL is detected. Scanning the delay of the gate pulse reveals the temporal characteristics of the PL, while analysis of the upconverted light reveals the spectral characteristics.

non-linear crystal

ps luminescence

pump beam

monochromator

& photomultiplier

ωL + ωG

ωL

gate ωG

sample

delay stage150 fs 150 fs

0 5 10 15 20

0

100

200

300

400

500

600

co

un

ts s

-1

delay (ps)

0 1 2 3

Figure 1: Experimental setup for ultrafast TRPL investigations.

They key advantage of TRPL as it applies to the hot carrier cell project is that we can directly ascertain the temperature of the electrons and holes in the sample by analysis of the emission spectrum. Such a direct measurement is immensely valuable and has been used to demonstrate the hot phonon effect which underpins this project.

Transient absorption spectroscopy

A less sophisticated experiment is the measurement of transient absorption, A(). In this experiment, the sample is assumed to recover fully between laser shots which arrive every millisecond. Absorbance, A, is quantified by

I

IA 0

10log ,

where I0 is the incoming light intensity at a given wavelength, and I is the intensity after the sample. Intensities are proportional to the signals recorded on the home-built integrating photodiodes (IPD). The pump laser is chopped, so that consecutive measurements represent absorbance in presence and absence of the pump beam respectively. The change in absorbance is given with reference to the unpumped sample, the absorbance of which is A’.

12

21log

loglog'

10

010

010

IPDIPD

IPDIPD

I

I

I

IAAA

The signal can be normalized for the strength of the pump pulse as measured on

IPD3 (not shown). Typical sensitivity is in the A range of 10-5.

pump beam

sample

scanned

delay stage

150 fs150 fs

probe beam

chopper

IPD3

IPD2

IPD1

Figure 2: The transient absorption spectroscopy set-up.

Time-Resolved Photoluminescence of GaAs and InP (with UNSW node) GaAs and InP were co-investigated to elucidate the effect of the phononic bandgap

of InP on the rate of electron cooling. In both materials, a broad photoluminescence signal which extends from the bandgap energy to values larger than the pump pulse energy was observed during the first few picoseconds after photoexcitation. This demonstrates fast carrier-carrier interactions (< 100 fs) for each pump energies at carrier densities similar to that of photovoltaic devices operating under solar concentration. For each excitation energies, the time resolved spectra of GaAs and InP are described in Figure 3(a,b,c). The longer hot carrier relaxation observed in InP is direct evidence of an enhanced hot phonon effect, which is due to the larger phononic bandgap of InP.

Figure 3: Sub-ps time resolved PL spectra of GaAs and InP after three different excitation energies.

In order to ascertain the temperatures of the electron and hole populations, a temperature-dependent photoluminescence spectrum was simulated for both InP and GaAs. The hot phonon effect is confirmed in Fig. 4, where the temperature evolution of the electron population is shown for all pump energies. For a 730nm (1.7eV) pump, where all excited electrons remain in the valley, the rate of InP carrier cooling is lower than in GaAs, demonstrating the influence of the hot phonon effect, as depicted in Fig. 4(a). This study has been accepted for publication in Progress in Photovoltaics.1

GaAs 1.7eV

GaAs 1.88eV

GaAs 2.4eV

InP 1.7eV

InP 1.88eV

InP 2.4eV

(c)

(b)

(a)

Figure 4: Carrier temperatures as a function of time after excitation with (a) 730 nm; (b) 660 nm; and (c) 515 nm pump pulses.

Transient absorption of Ge nanowires (with McIntyre, Stanford)

Quantum confinement effects on carrier cooling have been investigated in one-dimensional indirect bandgap materials. Germanium nanowire samples were obtained from Professor Paul McIntyre (Stanford). Absorption transients for 40nm and 20nm nanowires are shown in Figure 5.

0 20 400.0

0.5

1.0

ΔR

/R

delay (ps)

40nm

20nm

3010

Figure 5: Normalized transient refectance of Ge nanowires of 40nm and 20nm thickness.

Immediately noticeable are the oscillations observed in the 40nm samples. This

indicates population of the symmetric breathing mode of the nanowire. It is known that rapid heating of a material will cause coherent vibrations, populating low frequency symmetric modes such as the breathing and extension modes. In long nanowires, the extension modes have periods exceeding the lifetime of the electronic excitation, However, in wires of nanoscale widths, the breathing mode is observed to have a picosecond scale period. The frequency of the fundamental breathing mode for a solid cylinder (nanowire) is given by

vv

vE

d

xf

211

11

where E and v are respectively Young’s modulus and the Poisson ratio of the material; and d is the diameter of the nanowire. The parameter x1 is the first non-trivial root of

xJv

vxxJ 10 1

21

where Jm are the mth Bessel function of the first kind. for germanium, E = 102:66 GPa, v = 0:278 and = 5330 kgm-3. By numerical means, we obtain x1 = 2.10. Using these parameters we obtain an expression for the oscillation period, =Cd, where C = 0.302 ps/nm. As such, for 40 nm wide nanowires, we expect a period of about 12 ps, and about half this for nanowires of half the width. While the 40 nm wires clearly display an oscillatory motion, that of the 20 nm wires is less clear. The observed period of vibration is consistent with modelling the nanowires using bulk parameters (C = 0.302 ps/nm). A fit to the data yields C = 0.344±0:029 ps/nm, which encapsulates the experiment at the 2

level of confidence. However, it may be of interest to do more careful measurements with better characterization. It may be that, possibly due to surface effects, germanium nanowires do not exhibit a bulk Young’s modulus. Transient luminescence of Indium Nitride (with UNSW node)

The large mass-mismatch between Indium and nitrogen engenders a large phononic bandgap in InN. As such, it is an obvious choice of material in which to study the effect of a phononic bandgap on attenuation of the Klemens phonon decay mechanism.

InN samples were obtained from the UNSW node, and transient photoluminescence was studied at a number of pump and emission wavelengths in order to ascertain the rate of electron cooling in the material.

Figure 6: Temperature transients for InN.

For the three different samples the temperature transients are quite similar. This

testifies to the quality and accuracy of the measurements. However, the transients are quite fast. The fitted relaxation time constants are in the order of ~5ps. To have a consistent gain in efficiency in a HCSC one needs a slightly longer value. However, it is expected that with energy selective contacts one should be able to observe the hot carrier effect.

Quantum Wells for Hot Carrier Solar Cells (with Ekins-Daukes, Imperial)

The suitability of InGaAs/GaAsP quantum wells (QWs) as absorbers for hot carrier solar cells (HCSC) is considered. Ultra-fast time resolved photoluminescence (TRPL) spectroscopy was used to measure carrier distribution temperature as a function of time after excitation. It was shown that the carrier distribution remained hot relative to the surrounding lattice in excess of 2ns after excitation. This shows slowed carrier cooling relative to GaAs bulk, making the proposed material system an interesting prospect for further study and the development of a proof of concept device.

A dual OPA upconverting laser system was employed to perform ultrafast time resolved PL spectroscopy on a 10 well In0.20Ga0.80As/Ga0.90AsP0.10 strain balanced sample. This deeper well sample was required because of the reduced spectral resolution of the ultra-fast system compared to the CW system. In addition, measurements were conducted at 77K to further spectrally resolve bulk and well peaks. Figure 7 shows the PL from the well region peaking at approximately 1.32 eV. At higher energies the onset of the GaAs bulk PL is also observed. As the time after excitation is increased the shape of the well region PL changes, with the PL peak becoming sharper and the slope of the high energy tail becoming steeper.

Figure 7: Ultra-fast time resolved PL spectroscopy showing PL as a function of energy and time after excitation.

Figure 8: Carrier distribution temperature as a function of time after excitation. The lattice temperature (77K) is marked with a dotted line.

Figure 8 shows the temperature of the carrier distribution as a function of time after

excitation. The carrier distribution temperature is still greater than that of the surrounding lattice beyond 2ns after excitation. Previous Ultra-fast TRPL studies have shown carrier distributions in undoped GaAs bulk equilibrating with the surrounding lattice in < 500ps.

This slowed carrier cooling in strain balanced InGaAs/GaAsP QWs compared to GaAs bulk make this material system a suitable candidate for further hot carrier studies. Summary

The University of Sydney node comprises a well staffed and equipped ultrafast spectroscopy laboratory which is capable of performing a number of experiments relating to hot carrier phenomena, as well as other studies.

As shown above, the node is involved in a number of projects and collaborations involving this GCEP project. InP and GaAs were compared with regard to carrier cooling dynamics, and the comparison is published in Reference 1. InN samples were obtained from UNSW and have been demonstrated to exhibit a hot carrier effect. Importantly, III-V materials under investigation have been shown to demonstrate the hot phonon effect which underpins the hot carrier cell design.

Germanium nanowire samples are under investigation as part of a collaboration with Prof. Paul McIntyre of Stanford University. These have been shown to exhibit oscillatory absorbance which is attributed to the breathing mode of the highly monodisperse sample.

Quantum-wells were observed to maintain above background electron temperatures for up to 2ns, demonstrating the longest lived hot carrier effect so far observed in our laboratory. 1 Interplay between the hot phonon effect and intervalley scattering on the cooling rate of hot carrier in GaAs and InP, R. Clady, M.J.Y. Tayebjee, P. Aliberti, D. König, N.J. Ekins-Daukes, G.J. Conibeer, T.W. Schmidt and M.A. Green, Progress in Photovoltaics 2011.