Conventional methods for H2 isotope separation, such as cryogenic distillation or thermal diffusion, are complex and energy intensive. From a classical viewpoint,

Fig. 2.9 QENS self — 10-7

diffusivities for H2 in ZIF-8 at

77 K (circles), compared with

values computed with a rigid

(squares) and flexible

(triangles) framework model 10

‘со

CM ‘

E,

CO

10-9 10-10

H2 and D2 are similar in terms of size and shape as well as energetic considerations. However, H2 and its isotopes can no longer be treated as classical molecules at low temperatures, as quantum effects become important due to their low mass. Quantum separation should be facilitated by the larger de Broglie wavelength of the lighter isotope. Kinetic molecular sieving, based on differences in diffusivity of the two isotopes, is attractive for porous materials with pore sizes comparable to the de Broglie wavelength.

MD simulations, with quantum corrections incorporated via the Feynman-Hibbs approach, showed that D2 should diffuse faster than H2 below 150 K [27], in a hypothetical zeolite rho having a window diameter of 5.43 A (this corresponding to the distance between the O centres, upon subtraction of the O diameter a free diameter of 2.73 A is obtained). QENS experiments were performed to test this prediction. However the zeolite rho used for the measurements had a different chemical composition, so that the free diameter of the windows was approximately 3.26 A [28]. A new set of potential parameters was defined for the MD simulations, which yielded excellent agreement with the QENS diffusion coefficients of H2 and D2, but no quantum effect was observed. This was attributed to the larger pore size in the rho sample used experimentally.

Another series of QENS experiments performed with a carbon molecular sieve (CMS) confirmed the predictions made by the simulations [29]. A CMS can be obtained with different pore sizes, and the Takeda 3 A CMS has pores below 3 A. In this material, it was indeed found that D2 diffuses faster than H2, below 100 K (Fig. 2.10). This shows the extreme sensitivity of the reverse kinetic-selectivity on the window dimension. Transition-state theory and MD calculations indicated that to achieve high flux the cages interconnected by these windows must be small.

Fig. 2.10 Diffusion 10-7

coefficients of H2 (squares)

and D2 (rounds) in Takeda 3

A CMS, as a function of

temperature (loading

0. 5 mmol/g)

10-8

Neutron characterization techniques are particularly important in understanding materials for the nuclear-energy sector for many reasons. In structural components the combination of radiography, residual stress, and texture measurements allows assessment of structural integrity and optimization of manufacturing, while SANS provides a unique tool to track the effects of radiation damage. In fuels and waste forms, particularly when heavy and light elements are combined, neutron diffraction has inherent advantages over X-rays.

Neutron characterization techniques will be increasingly important for under­standing materials for the nuclear sector. The increased functionality comes partly from new techniques, and partly from improvements in existing techniques.

New techniques such as energy-dispersive neutron radiography and Bragg-edge strain imaging offer new insights into materials. As the technologies are new, large improvements are to be expected in technique and analysis.

Well established techniques will benefit from improvements in neutron flux, detectors, and analysis, enabling in situ or kinetic studies, and smaller samples. Sample environments will increase in complexity to better mimic the studied operating conditions. Shielding or dedicated beamlines will allow characterization of active materials.

Combinations of techniques, such as diffraction and radiography, will provide detailed crystallographic information in combination with a spatially-resolved distribution of the properties of interest.

The workhorse of ND is powder diffraction, which has been developed to the point where complete structural information can be obtained from polycrystalline sam­ples. Even modest dynamical information, such as diffusion pathways, can be deduced from atomic displacement parameters. The technique relies on the well — known coherent-interference pattern that is scattered from well-defined lattice planes, which, due to rotational averaging in a powder, collapse to a simple one­dimensional powder pattern. The level of detail that is available from this pattern is largely dependent on the resolution and range of the diffractometer, plus the availability of refinement algorithms, all of which continue to improve. The tech­nique is not limited to a single compound, and measured diffraction patterns are often used to establish the phase composition in complex materials, for example where doping is used to modify electronic structure in solar-cell materials (such as in Chaps. 5 and 6).

Although very detailed structural information can be obtained from powders, the comparative simplicity of the technique also makes it the prime candidate for in situ studies, which also profit from the penetration and isotope selectivity of neutron scattering. In the context of this book, the technique is commonly used to follow the evolution of structure with temperature, or composition, for example in charging and discharging electrodes (such as in Chap. 7).

The main constraint on neutron powder diffraction (NPD) is the larger samples required compared with X-rays, and the large incoherent neutron-scattering from

the 1H nucleus that causes a high background. In general this background can be

eliminated by deuteration, with the added advantage that the crystallographic positions of these atoms can then be more-easily established.

Silicalite membranes exhibit an interesting selectivity for pure CO2, which flows through the material faster than He or H2, notwithstanding the larger kinetic diameter of CO2 [58]. The diffusion of CO2 and N2 in silicalite was studied using QENS, with combined coherent QENS-MD used to determine diffusivities for CO2 and more traditional QENS used to determine the N2 diffusivity [59, 60]. The work obtained the diffusivity as a function of intra-crystalline occupancy. A direct comparison between computed and measured transport-diffusion allowed a better understanding of the molecular factors governing the occupancy dependence of the corrected diffusivity, a “pure” kinetic parameter that is largely free of the influence of the isotherm. The corrected diffusivity has often been assumed to be independent of sorbate concentration and approximately equal to the self-diffusivity in the limit of zero occupancy, an assumption that was reassessed in light of these results. These measurements pointed to a significant difference in the occupancy depen­dence of the corrected diffusivity for N2 and CO2, where at 300 K the corrected diffusivity for CO2 was found to decrease with guest loading of the host and for N2 at 200 K it was fairly constant, exhibiting a weak maximum. At the practical level this work found that interactions are considerably stronger (more attractive) for CO2 than for N2 in silicalite, explaining the strong preference of the material for CO2.

The performance of an OPV is characterized by the current-voltage dependence (I-V curve) depicted in Fig. 6.2.

The directly-measurable parameters that allow a quantitative analysis of the performance of a PV cell are:

(i) The open-circuit voltage Voc corresponding to zero-current in the cell, which is dependent on the offset of the energy levels (highest occupied molecular orbital of the donor: HOMOD and lowest unoccupied molecular orbital of the acceptor: LUMOA) of the DA interface of the cell (Fig. 6.3). The short-circuit current Isc which is related to the photo-generated charges. The fill factor (FF) which is related to the ratio of the red and grey areas in Fig. 6.2, which reflects the quality of a photovoltaic (PV) cell, or more precisely, it is determined by the mobilities of charge carriers [4, 9].

(ii) The incident-photon-to-current efficiency (IPCE) is defined as the ratio of the number of incident photons and the number of photo-induced charge carriers which can be generated. Unlike the internal quantum efficiency, IPCE accounts for losses by reflection, scattering, and recombination.

(iii) The power conversion efficiency (PCE), mentioned above, depends on the FF, Voc, Isc, and the incident light intensity. PCE is maximal for larger values of the first three parameters. A standard AM1.5 spectrum is used for PV characterizations.

(iv) The PCE can be decomposed in terms of different “sub-efficiencies” related to each step towards the generation of the photocurrent. The overall per­formance of an OPV depends on the sequence of steps mentioned above and illustrated in Fig.

p

Fig. 6.2 The I-V curves characteristics of PV cells with (short dash) and without (long dash) solar illumination. A diode-like behaviour is observed in the absence of illumination. The short — circuit current Isc (V = 0) and the open-circuit voltage Voc (I = 0) are shown. MPP and Pmax denote the maximum power point and maximum power output of the PV cell, respectively

ПР = Па X nd X П x Пс

Fig. 6.3 Schematic representation of the main fundamental steps describing the photo-current conversion by an organic solar cell along with the related efficiencies in terms of which the power conversion efficiency (PCE) parameter, np, is shown

The incident light, with energy falling inside the HOMO-LUMO gap, is absorbed, leading to the creation of a strongly-bound Frenkel exciton by promoting an electron from its ground state to a higher energy level. The exciton is of a strongly “bound” nature because the electron-hole pair is subject to a strong mutual attractive interaction [6]. In OPVs additional steps are then required for electron- hole extraction, which is in contrast to inorganic PVs where the free carriers are released immediately. Therefore the transport processes are much more complex in OPVs due to the polaronic nature of the carriers. In the absence of an applied electric field, the exciton diffuses inside the organic semiconductor and becomes spatially dissociated into bound, positive and negative carriers (a polaron-holon pair). The effective dissociation occurs at the DA interface followed by transport of the polarons and holon to the relevant electrodes where they are extracted gener­ating the final photocurrent (Fig. 6.4). During its diffusion the exciton may recombine before it reaches the DA interface, with a consequent loss of the absorbed energy by dissipation. Indeed, the diffusion length of excitons to the DA interface is much shorter than the optical absorption length. This is of prime importance in understanding the different loss mechanisms that should be avoided to improve the dissociation and collection of charges before recombination.

Figure 6.5 shows different loss mechanisms, either by exciton recombination or by charge-carrier trapping, which may occur during the different steps towards photocurrent production explained above. Loss by recombination can even occur after a successful exciton dissociation recombination, and trapping during the charge transport or collection at the DA interface and electrodes, is a further loss. After a recombination, the loss leading to dissipation of the absorbed solar-energy can be either radiative or non-radiative, the former being detected via fluorescence and/or phosphorescence, whilst the later via a phonon creation. When compared with inorganic devices, the organic analogous have two important limitations: first, their narrow absorption-window, and second, the tendency for thermal motion to perturb

the rather soft charge-transport system. Therefore, efficient energy-conversion for practical applications requires significant improvements of charge transport and transfer processes. Energy loss during the photogeneration process should be min­imized by maximizing the number of excitons that dissociate into free charge — carriers, rather than simply recombining without contributing to the photocurrent. A rich research field has been stimulated aimed at extending the spectral sensitivity of OPVs to cover a broader wavelength region and therefore the band-gap tuning.

In this context, it has been proposed that stacking different OPVs would help to achieve this goal. The problem of inefficient exciton-dissociation could be solved thanks to the introduction of tandem[3] device architectures and BHJ [7-11]. The interpenetrating networks of n-and p-semiconductors (Fig. 6.1) improve the charge separation and transport to their respective electrodes for collection.

The main classes of heterogeneous catalysts are: (i) metals and alloys (supported or not), (ii) metallic oxides (including mixed oxides, heteropolyacids, superacids),

(iii) zeolites and molecular sieves in general, and (iv) sulfides. It will take some more years before deciding if metal-organic frameworks (MOFs) become a new member of the catalysts family.

Several neutron techniques are used to study catalytic systems: neutron dif­fraction (ND), small-angle neutron scattering (SANS), inelastic neutron scattering (INS), and quasi-elastic neutron scattering (QENS). We will limit ourselves here to INS and QENS of hydrogen species and dihydrogen molecules adsorbed on the surface of catalyst particles or inside porous materials.

H. Jobic (H)

Centre National de la Recherche Scientifique, Institut de Recherches sur la Catalyse et l’Environnement de Lyon, Lyon, France e-mail: herve. jobic@ircelyon. univ-lyon1.fr

© Springer International Publishing Switzerland 2015 17

G. J. Kearley and V. K. Peterson (eds.), Neutron Applications in Materials for Energy, Neutron Scattering Applications and Techniques,

DOI 10.1007/978-3-319-06656-1_2

Hydrogen has recently been associated with the words fuel cells and energy storage, but it is also an essential component in catalytic reactions and the hydrogen produced is mainly used in petroleum refining and ammonia production for fer­tilizer. Nowadays, about 90 % of the H2 production comes from catalytic steam reforming of natural gas at high temperatures (subsequent reactions of water-gas shift and preferential oxidation are required to decrease the CO level of the gas mixture to a few ppm before it can be used in a fuel cell). At the time being, one is facing a huge increase of H2 needs (with related CO2 emissions), so that until Jules Verne’s predictions are realized (water: the coal of the future), we may reach an H2 deficit, as predicted by some experts.

The applications of INS to catalysis have been mainly focused to systems which are either difficult or impossible to study by other spectroscopies such as trans­mission or reflection-absorption infrared, and Raman. The kind of catalyst which is studied in INS has generally an inhomogeneous surface, e. g. oxides, sulfides, and metals, although zeolites, which are well-crystallized materials, are well suited. These substrates can be almost transparent to neutrons if they contain a small quantity of hydrogen, in which case the neutron spectrum will be fairly flat and it will be possible to observe all the vibrational modes of the adsorbate.

QENS has been mainly used to measure the diffusion of H or hydrogenated molecules, although the transport of deuterated molecules and of molecules which do not contain hydrogen atoms can now be followed. The dimensionality of dif­fusion has been studied, even if the samples are in powder form. On zeolites [1], MOFs [2], or clays [3], anisotropic diffusion (one or two dimensional) has been evidenced. The technique allows us to probe diffusion over length scales ranging from an A to hundreds of A. The mechanism of diffusion can thus be followed from the elementary jumps between adsorption sites to Fickian diffusion.

Modern steels are strengthened by finely dispersed nano-particulates, particularly in high-temperature structural materials. Radiation damage and temperature can cause these to change their shape, size, and distribution, leading to embrittlement. Oxide dispersion strengthened (ODS) steels are designed for high-temperature operation; they contain oxide nano-clusters (e. g. yttrium-titanium oxide) in a ferritic-steel matrix.

The size and morphology of these particles may change with exposure to stress, temperature, or radiation; all conditions which occur in nuclear-power systems, typi­cally increasing the size and decreasing the number of strengthening particles. Changes in the volume fraction, surface interface and size distribution of these particles can cause significant changes to the strength, ductility, and toughness of these materials. Advanced reactor-materials use nano-particulates to act as sinks for radiation-induced phenomena such as vacancies, self-interstitials, and as trapping sites for He bubbles from nuclear reactions in the material. Small-angle neutron scattering (SANS) can non­destructively analyse these nanoscale features over a large sample volume.

The use of a strong magnetic field can differentiate between magnetic and non­magnetic scattering at right angles to each other, and the strengthening nano-par­ticulate oxides are typically non-magnetic. With ODS steel samples the SANS curves for no magnetic field (nuclear scattering only) and with field (nuclear + magnetic scattering) are compared. If they have no difference then the scattering is entirely non-magnetic and it can be assumed that the scattering is from the non-magnetic oxide nano-clusters. If there is a difference then the contribution from iron-based magnetic scatterers can be removed. This experimental technique should be used in conjunction with an oxide-free reference sample of the same manufacturing route (if possible), for correct de-convolution of the SANS from oxide nano-clusters.

The size of strengthening nano-particles is in the range 1-20 nm, and the few techniques available for studying these are transmission-electron microscopy (TEM), atom-probe tomography (APT), and SANS, however TEM and APT only sample a small volume. As a result, SANS has become a critically important technique for the development, understanding and characterisation of these irradi­ation-resistant materials [16, 17].

The success of Li-ion batteries is based on their high volumetric and gravimetric energy-density available for storage. This has enabled the realization of small portable devices like mobile telephones and laptops. However, important demands also include material and production costs, safety, cycle life (number of charge/ discharge cycles), and high (dis)charge rates. Addressing all these demands makes it very challenging to find better electrode and electrolyte materials to improve Li-ion batteries [2, 4]. To gain more insight into the challenges in battery research it is useful to express the performance-related demands in more specific material properties of the electrodes and electrolytes. With respect to the general demands of cost and safety we merely state that batteries require abundant and cheap materials that are intrinsically stable during battery operation. For more extensive literature we refer the reader to some excellent reviews [2, 4—6].

The gravimetric and volumetric energy density of a battery is CVOCP, where C is the specific or volumetric capacity (mAhg-1 or mAgcm-3), respectively, and VOCP is the open circuit, or equilibrium potential of the battery. Large gravimetric energy — density is required for automotive applications, and large volumetric energy den­sities are essential in mobile electronic equipment. It is important to realize that the specific Li capacity of the electrode materials is not necessarily the only decisive factor, and that the actual capacity of an electrode also depends significantly on the electrode morphology [7—9]. Large gravimetric energy-density requires dense electrodes and hence low porosities. In this context nano-sizing of electrode materials, aimed at higher storage capacities and higher rates, typically carries the disadvantage of low tap densities (powder packing) leading to less dense electrodes and compromising both volumetric and gravimetric energy-density.

The power density is the product of the specific current and the voltage that the battery can deliver. The battery voltage is defined by the difference in potential between the electrodes and the current via the internal resistance of the battery. High power densities, allowing for fast (dis)charge, require low internal resistance of the battery. The various charge-transport phenomena inside a battery contribute to this internal resistance, including the electronic conductivity through the elec­trodes, the ionic conductivity through the electrode and electrolyte, and finally the charge transfer through the interface between the electrode and electrolyte. Gen­erally, the Li ion and electronic conductivity through the electrodes are assumed to be rate limiting in Li-ion batteries. Note that because the electrode is porous the ionic conductivity of the electrodes includes both the transport of ions through the solid-state electrode as well as through the electrolyte dispersed in the electrode pores. The solid-state transport of ions through electrodes depends on the specific electrode host material and is typically orders of magnitude slower than in liquid electrolytes. The overall Li-ion conductivity within the electrodes also depends strongly on the electrode morphology, such as characterized by the porosity [10] and the interconnectivity of the pores [11, 12].

High energy and power density are conflicting demands, as evidenced by the impact of electrode morphology and electrode thickness on these. High porosities, such as in nanostructured electrodes, generally lead to fast ionic transport throughout the electrodes responsible for high (dis)charge rates, and hence high power densities. However, the downside of large porosities is the larger volume required to store the same amount of energy, resulting in smaller volumetric energy- density. Because in most cases either electronic or ionic conductivity through the electrodes is rate limiting [7—9], thin electrodes can be charged faster than thick electrodes. Hence, the power density of a battery can be improved by the use of thinner electrodes. However, building a battery from thin electrodes leads to a smaller amount of active electrode material per gram of battery because of the relatively larger amount of current collector, electrolyte, and packing materials.

The cycle life of batteries is determined by a combination of the chemical and physical processes that occur during (dis)charging in the electrolyte and electrodes of a battery. Generally, electrodes undergo structural changes upon Li insertion and extraction. Large volumetric changes upon Li insertion and extraction lead to mechanical failure of the electrode. The result is generally that part of the active material is electronically disconnected from the electrode and therefore inactive. Consequently, small structural changes, such as occurring in Li4Ti5O12 spinel (where the change in the unit-cell volume, AVunit_ceU, & 0.1 %) [13], and moderate structural changes, such as occurring in LiFePO4 (AVunit_ceu & 6.5 %) [14], contribute to batteries with long cycle-life. The other extreme is the alloying reaction of Li with silicon up to the Li4.4Si composition resulting in a volume expansion >300 %, which leads to mechanical failure of the electrodes containing relatively large silicon particles within only a few charge/discharge cycles [15]. Nevertheless, the very high capacity associated with this type of alloying reaction has motivated chemists and material scientists to develop smart strategies to maintain the mechanical coherence of these electrodes [15—17].

The second factor that is important for the cycle life, and also to safety, is the thermodynamic stability of the electrolyte with respect to the positive and the negative electrodes (Fig. 7.2). The electrodes have electrochemical potentials (anode) and ^C (cathode) equal to their Fermi energies sF. If ^A is above the lowest unoccupied molecular orbital (LUMO) of the electrolyte, the anode will reduce the electrolyte. This reduction decomposes the electrolyte unless a passivating layer, generally referred to as the solid electrolyte interface (SEI) layer, is formed. The SEI layer often forms in the first few cycles and can act to electronically insulate the anode from the electrolyte, preventing further reduction. The SEI growth tends to stop after a few cycles, resulting in stable battery performance. Similarly, if ^C is below the highest occupied molecular orbital (HOMO) of the electrolyte, the cathode will oxidize the electrolyte unless a passivating SEI layer is formed, electronically insulating and preventing further oxidation. Therefore, not only should the

electrolyte provide a wide stability window in terms of voltage, but the gap between its LUMO and HOMO must be larger than the difference between the chemical potentials of the electrodes. That is, the electrolyte voltage window should be positioned such that LUMO > and HOMO < fic. The large open-cell voltage (VCeld of Li-ion batteries (e. g. 3.6 V) that is responsible for their high energy and power densities requires electrolyte stabilities that exceed that of water (Elumo-Ehomo ^ 1.3 V) or water-containing electrolytes, leading to the application of non-aqueous electrolytes. Such stability demands do not apply if an electroni­cally-insulating layer is formed upon electrolyte reduction or oxidation which can passivate further reactions. An example of passivation of SEI layers is the graphite or carbon-based anodes that work at about 0.5 V below the stability of typical carbonate electrolyte solutions, and hence 0.5 V above the LUMO of the electrolyte. The first few cycles with these anodes and carbonate-based electrolyte solutions results in the reduction of the electrolyte which leads to the formation of a stable SEI layer that prevents further reduction of the electrolyte. Importantly, the SEI layer does not grow significantly in subsequent cycles. Therefore, strategic combinations of elec­trodes, electrolytes, and SEI layers can result in better-performing batteries.

Improving Li-ion batteries with respect to energy and power density, manu­facturing costs, safety, and cycle life is clearly a formidable challenge. A common starting point is to lower costs and use environmentally-benign materials for both the electrodes and the electrolyte, noting that the cathode can account for as much as 40 % of the cost of the battery. The electrolyte should have high Li-ion conductivity over a practical ambient-temperature range and a large stability (potential) window allowing use of large differences in electrode potentials. The electrodes should work within the stability window of the electrolyte and allow fast (dis)charge of a large reversible capacity. This not only depends on the electrode material and composition but also on the electronic and ionic ‘wiring’ of the electrodes determined by the electrode morphology. The development of new electrode and electrolyte materials and the improvement of existing materials requires a fundamental understanding of the Li insertion and transport mechanisms that involve both structural and kinetic phenomena. Probing these is a tremendous challenge given the complexity of the system and the difficulty of probing a light element such as Li, particularly under in situ conditions.

Sensitivity is often a problem when studying catalysts with neutrons, only solids which have high surface areas can be studied. One still needs more neutron flux to widen the range of applications.

With INS, the advent of the VISION instrument at the Spallation Neutron Source (SNS) at the Oak Ridge National Laboratory (ORNL), the LAGRANGE instrument being operational at the ILL, and possible improvements on the instrument TOSCA at ISIS should allow us to tackle in the near future grafted catalysts, fuel cells, small supported metal particles (metal loading <1 %), etc. The large Q values which are inherent to these spectrometers at large energy transfers is a serious limitation. The detrimental influence of the Debye-Waller factor on the INS spectra has been known for a long time [30]. This can be partially resolved using direct geometry instruments which allow access of lower Qs [31], but different incident energies have to be selected so that various parts of the spectrum can be combined. On the other hand, the technique is particularly suited to study the different hydrogen species which are formed after dissociation of dihydrogen, during reduction or activation of the cat­alyst. The problem of characterizing active hydrogen is still a big issue in catalysis.

In the same way as ab initio methods are increasingly being used in INS, molecular simulations are now combined with QENS experiments. Since the space and time scales of the neutron techniques match closely the ones covered by molecular simulations, one expects, and usually finds, good agreement between neutrons and simulations. QENS constitute a benchmark to validate and further develop the modelling work, and the computed trajectories of the sorbate guest molecules within the host matrices are invaluable to understand QENS observables. Discrepancies between experiment and simulation do happen and require in such cases the consideration of a flexible lattice or an improved force field.

References

1. H. Jobic, J. Mol. Catal. A: Chem. 158, 135 (2000)

2. F. Salles, H. Jobic, G. Maurin, M. M. Koza, P. L. Llewellyn, T. Devic, C. Serre, G. Ferey, Phys. Rev. Lett. 100, 245901 (2008)

3. N. Malikova, S. Longeville, J.-M. Zanotti, E. Dubois, V. Marry, P. Turq, J. M. Ollivier, Phys. Rev. Lett. 101, 265901 (2008)

4. H. Jobic, in Catalysis by Metals, A. J. Renouprez, H. Jobic (eds.) (Les Editions de Physique — Springer, Berlin 1997), p. 181

5. P. W. Albers, S. F. Parker, Adv. Catal. 51, 99 (2007)

6. A. A. Khassin, G. N. Kustova, H. Jobic, T. M. Yurieva, Y. A. Chesalov, G. A. Filonenko, L. M. Plyasova, V. N. Parmon, Phys. Chem. Chem. Phys. 11, 6090 (2009)

7. R. Juarez, S. F. Parker, P. Concepcion, A. Corma, H. Garcia, Chem. Sci. 1, 731 (2010)

8. C. Pirez, M. Capron, H. Jobic, F. Dumeignil, L. Jalowiecki-Duhamel, Angew. Chem. Int. Ed. 50, 10193 (2011)

9. B. Bogdanovic, M. Schwickardi, J. Alloys Compd. 1, 253 (1997)

10. C. H. Liu, B. H. Chen, C. L. Hsueh, J. R. Ku, M. S. Jeng, F. Tsau, Int. J. Hydrogen Energy 34, 2153 (2009)

11. D. Colognesi, A. Giannasi, L. Ulivi, M. Zoppi, A. J. Ramirez-Cuesta, A. Roth, M. Fichtner, J. Phys. Chem. A 115, 7503 (2011)

12. K. Skold, G. Nelin, J. Phys. Chem. Solids 28, 2369 (1967)

13. A. Renouprez, P. Fouilloux, R. Stockmeyer, H. M. Conrad, G. Goeltz, Ber. Bunsenges. Phys. Chem. 81, 429 (1977)

14. A. Renouprez, R. Stockmeyer, C. J. Wright, J. Chem. Soc. Far. Trans. I 75, 2473 (1979)

15. N. K. Bar, H. Ernst, H. Jobic, J. Karger, Magn. Reson. Chem. 37, S79 (1999)

16. R. Hempelmann, D. Richter, D. A. Faux, D. K. Ross, Z. Phys, Chem. Neue Folge 159, 175 (1988)

17. L. Onsager, Phys. Rev. 37, 405 (1931)

18. H. Jobic, D. N. Theodorou, Microporous Mesoporous Mater. 102, 21 (2007)

19. H. Jobic, J. Karger, M. Bee, Phys. Rev. Lett. 82, 4260 (1999)

20. E. Pantatosaki, G. K. Papadopoulos, H. Jobic, D. N. Theodorou, J. Phys. Chem. B 112, 11708 (2008)

21. P. G. de Gennes, Physica 25, 825 (1959)

22. F. Salles, D. I. Kolokolov, H. Jobic, G. Maurin, P. L. Llewellyn, T. Devic, C. Serre, G. Ferey, J. Phys. Chem. C 113, 7802 (2009)

23. H. Jobic, K. Hahn, J. Karger, M. Bee, A. Tuel, M. Noack, I. Girnus, G. J. Kearley, J. Phys. Chem. B 101, 5834 (1997)

24. A. I. Skoulidas, D. M. Ackerman, J. K. Johnson, D. S. Sholl, Phys. Rev. Lett. 89, 185901 (2002)

25. R. Banerjee, A. Phan, B. Wang, C. Knobler, H. Furukawa, M. O’Keeffe, O. M. Yaghi, Science 319, 939 (2008)

26. E. Pantatosaki, H. Jobic, D. I. Kolokolov, S. Karmakar, R. Biniwale, G. K. Papadopoulos, J. Chem. Phys. 138, 34706 (2013)

27. A. V. Anil Kumar, S. K. Bhatia, Phys. Rev. Lett. 95, 245901 (2005)

28. A. V. Anil Kumar, H. Jobic, S. K. Bhatia, J. Phys. Chem. B 110, 16666 (2006)

29. T. X. Nguyen, H. Jobic, S. K. Bhatia, Phys. Rev. Lett. 105, 085901 (2010)

30. H. Jobic, Chem. Phys. Lett. 106, 321 (1984)

31. S. F. Parker, D. Lennon, P. W. Albers, Appl. Spectrosc. 65, 1325 (2011)

Susan Schorr, Christiane Stephan and Christian A. Kaufmann

Abstract In order to understand the importance of the structural properties of compound semiconductors for the operation of a thin-film solar cell, this section aims to explain the operation principle using the example of a Cu(In, Ga)Se2 (CIGSe) thin-film solar cell. For detailed information the reader is kindly referred to the literature for a more extensive overview of the recent developments [1], device operation [2] and material preparation [3].

5.1 Introduction

CIGSe thin-film solar cells are made of a stack of metal and semiconductor thin films in the following sequence: a molybdenum back contact (metal), a polycrys­talline CIGSe absorber layer (p-type semiconductor), CdS buffer layer (n-type semiconductor), and ZnO front contact (n-type semiconductor). Together, the CdS and the ZnO are often referred to as the ‘window’ of the device. The core of the device is the p-n-heterojunction between the p-type absorber layer and the n-type window layers. The resulting energy band line-up and a cross sectional scanning electron microscope view of a complete device are shown in Fig. 5.1. Due to the high doping of the ZnO front contact layer the field, which develops upon contact of the n — and p-type materials in the interface region, is located almost entirely inside the absorber layer. In comparison to homojunctions, the heterojunction has the advantage that the n-type component can be chosen such that its band gap is

S. Schorr (H) • C. Stephan

Institut Fur Geologische Wissenschaften, Freie Universitat Berlin,

Malteserstr. 100, 12249 Berlin, Germany e-mail: susan. schorr@fu-berlin. de

C. Stephan

e-mail: christiane. stephan@fu-berlin. de S. Schorr • C. A. Kaufmann

Helmholtz-Zentrum Berlin Fur Materialien Und Energie, Berlin, Germany

© Springer International Publishing Switzerland 2015 83

G. J. Kearley and V. K. Peterson (eds.), Neutron Applications in Materials for Energy, Neutron Scattering Applications and Techniques,

DOI 10.1007/978-3-319-06656-1_5

large and the photoactive band-gap Eg of the solar-cell device, determined by the band gap of the absorber-layer material, is optimized to reach high conversion — efficiencies.

Figure 5.2 illustrates the basic working principle of a CIGSe thin-film solar cell. Electron-hole pairs are generated by light absorption within the absorber thin-film. Absorption of photons with an energy higher than Eg results in the loss of excess energy via thermalization. If an electron-hole pair is excited within the depletion

region near the font interface of the device, it is immediately spatially separated by the electric field that is present. If generation happens outside the depletion region, the minority charge carrier has to diffuse into the depletion region in order to contribute to the photo current. Recombination of the electron-hole pairs can occur by a number of processes: radiative and non-radiative, band-to-band or via one or several defect levels located in the band gap of the absorber material, within the bulk of the thin-film material, within the depletion region, and at the front interface of the device or possibly at the grain boundaries of the polycrystalline CIGSe thin — film. The nature of grain boundaries in CIGSe thin-films however, has proven to be considerably more benign than in other semiconductor materials such as for example silicon, the exact reasons for which are still under discussion [4].

CIGSe absorbers for photovoltaic application are mostly fabricated slightly Cu-poor and with an overall Ga content of [Ga]/([Ga] + [In]) * 0.3. The resulting CIGSe material is ap-type semiconductor material (Eg * 1.15 eV), which is highly compensated. This means that there are acceptor — and donor-type defects present within the material and p-type conductivity is established due to the exceptionally-low formation energy of Cu vacancies. In addition, the formation of defect complexes, such as 2VCu + InC+,, seems to play an important role in terms of the electronic properties of the material, and also regarding phase formation and crystal structure [5] (Figs. 5.2 and 5.3).

For most applications rigid soda-lime glass is used as a substrate material, but flexible metal or polyimide foils have been used successfully. Working on soda — lime glass, it has been established that Na plays an important role in increasing the carrier concentration within the CIGSe absorber [6]. It diffuses at elevated process temperatures from the glass substrate through the Mo back contact into the growing CIGSe thin-film [7], and also has an effect on the morphology and material inter­diffusion in the growing layer [8, 9]. On samples, which do not intrinsically contain sodium, it has to be externally supplied.

Fig. 5.3 Elemental flux and substrate-temperature profiles for a typical 3-stage co-evaporation process

While ZnO and Mo are usually sputtered, CdS is most widely deposited via chemical-bath deposition. As the use of Cd-containing components is viewed critically, Cd-free alternatives are in development and, in some cases, already incorporated in the commercially-available product [10]. The methods for CIGSe thin-film fabrication can be grouped into two main categories: co-evaporation and sequential processing [10]. The method of co-evaporation relies on the more or less simultaneous evaporation of the elements Cu, In, Ga, and Se to form a CIGSe thin — film on a heated, Mo-coated substrate in vacuum. For sequential processing on the other hand, a so-called precursor layer in either metallic, binary, or nanoparticle form is treated in a reactive atmosphere in order to make the CIGSe thin-film absorber. The current record conversion-efficiency of 20.3 % [11] is reached by a CIGSe thin-film device that was deposited via 3-stage co-evaporation [12].

The 3-stage co-evaporation process for CIGSe deposition leads to a thin film, which has a characteristic compositional in-depth profile, shown in Fig. 5.4. While Ga accumulates near the Mo/CIGSe interface during the inter-diffusion of Cu-Se in stage 2, the last stage of the process also provides for a slightly increased Ga content near the surface. Of course the latter depends on the Ga flux during stage 3. The Ga profile is of relevance to the resulting solar-cell device, as the Ga content determines the energy band-gap of the CIGSe material. A graded compositional profile is therefore equivalent with an in-depth band-gap grading within the device. Compositional gradients, as shown in Fig. 5.4, can be observed even within single grains. Not only does the Ga content have an impact on the resulting energy band — gap of CIGSe, but also on the Cu-deficiency, in particular near the absorber surface. This is possibly caused by the nature of the last stage of the 3-stage process, which can affect its band gap. Cu-deficient CIGSe phases, such as Cu(In, Ga)3Se5, show slightly-larger band gaps than the stoichiometric compound. The presence of such Cu-poor phases at the front interface has been argued to be of relevance for efficient CIGSe thin-film devices [13]. Other evidence points towards entirely Cu-free film surfaces, which ensure favourable interface formation between the CIGSe and the buffer layer [14]. It is most likely that the exact process-parameters determine which

of the different scenarios holds in each case. CIGSe thin-films from other manu­facturing routines, such as the sequential processing, will have grown under very different thermodynamic and kinetic growth conditions and, so far, only little is known of the extent to which the resulting thin films can be considered identical.

In order to understand the correlation between growth conditions, material properties and final device quality, much has to be understood regarding the basic material properties of the absorber material in question. Topics such as the occu­pation density of the different sites within the crystal lattice, i. e. defect formation under certain growth conditions, at certain material compositions, or in the presence of foreign elements are areas where neutron diffraction can provide valuable input for the design of growth models, analytical material science, and also for compu­tational methods.