«c SIMON LINDBERG, 2013 Cover: To the left is an image of a mono crystalline silicon solar cell , In the middle is an image of three samples of ...»
Fabrication and characterization of Mg-Ni
hydride thin ﬁlms for photovoltaic
Master of Science Thesis in the Master Degree Programme, Applied Physics
Department of Applied Physics
CHALMERS UNIVERSITY OF TECHNOLOGY
Gothenburg, Sweden 2013
Fabrication and characterization of Mg-Ni hydride thin ﬁlms for photovoltaic applications
c SIMON LINDBERG, 2013 Cover: To the left is an image of a mono crystalline silicon solar cell , In the middle is an image of three samples of Mg2.4 NiH4 annealed at different temperatures and to the right is a picture taken of the Raman equipment used in the thesis.
Department of Applied Physics Chalmers University of Technology SE-412 96 Gothenburg, Sweden Fabrication and characterization of Mg-Ni hydride thin ﬁlms for photovoltaic applications Master of Science Thesis conducted in collaboration with Institute for Energy Technology, Lillestr¨ m.
SIMON LINDBERGDepartment of Applied Physics Chalmers University of Technology i Abstract We prepared and characterized three different compositions of the novel semiconductor material Mg2.0 NiH4, Mg2.2 NiH4 and Mg2.4 NiH4 as thin ﬁlms for photovoltaic applications.
The thin ﬁlms were deposited using reactive sputtering at room temperature, to achieve crystallization the samples were annealed at 220C and 290C. The composition of the samples were determined with EDS and the structural properties of the samples were then investigated using XRD and Raman spectroscopy. The band gap for the different compositions and annealing processes were calculated using a Tauc-plot.
We did Hall-measurements to investigate the electronic properties such as mobility, density and charge-carrier in the samples while the resisitivity were measured using a four point probe.
Also during the annealing process an unexpected color change appeared in the crystalline Mg2.4 NiH4 samples. This thermochromic effect were investigated with transmission and reﬂection spectroscopy, which indicates a band gap narrowing. XRD measurements with variable temperature were also performed to investigate if any structural change occurs during heating.
Acknowledgments First of all I would like to thank the Institute for Energy Technology for this opportunity to work with this interesting subject. Thanks to Trygve, Chang, Zeljka, Erik, Smagul and all other people at IFE for all the help and interesting discussion making these 6 last months very rewarding and successful.
I would also like to thank Dinko Chakarov for all his help and suggestions in writing the thesis and planing my work. My thanks goes also to Maths Karlsson for letting me use the Raman equipment at his department and valuable discussions regarding the results.
Finally I would like to deeply thank my friends and family for all the support and help during all of these years.
G¨ teborg, Mars 2013 o
• TWh - Terawatt hour
• eV - Electron volt
• Mg - Magnesium
• Ni - Nickel
• XRD - X-ray diffraction
• mbar - Millibar
• DC - Direct current
• RF - Radio frequency
• EDS - Energy dispersive x-ray spectroscopy
• GIXRD - Grazing incidence x-ray diffraction
• UV - Ultraviolet
• VIS - Visible
• IR - Infrared
• nm - Nanometer
• TEM - Transmission electron microscopy
• K - Kelvin
• I-V - Current-voltage
• µm - Micrometer
• RBS - Rutherford backscattering spectrometry
1.1 Background and motivation Today the ever increasing concern for global climate change and other environmental issues has made new, renewable, energy sources increasingly interesting for research. But a majority of the energy today is still produced using fossil fuels so the renewable energy sources need to increase dramatically to meet the emission goals. One off these renewable sources are photovoltaics, Figure 1.1: The historical contributions of different energy sources to the total energy production in TWh.
which are semiconductor based devices that convert sunlight to electricity. The ﬁrst photovoltaic cell based on silicon was constructed in the 1950’s with a efﬁciency of around 6% by Chapin et al at Bell laboratories , which was a new world record in 1954 for solar to electric conversion. Since then signiﬁcant progress has been accomplished and today commercial cells based on silicon have an efﬁciency around 17-23% which produces electricity at a cost around 2$/watt which is quite an improvement compared to the price of 1500$/watt in 1950. But it is only in the last couple of years that the interest in photovoltaics has done a major breakthrough, the market share has increased from 4 TWh installed in 2005, to 65 TWh 2011. Yet this must be compared to the total energy production in the world which has reached 20000 TWh so the the electricity from photovoltaics is still only a small part of the total energy production, which also is seen in Figure 1.1.
However solar cells have several advantages to other energy sources such as low maintenance and direct conversion from sunlight to electricity, but there are also some disadvantages with the technique where the biggest obstacle preventing a big expansion in the solar sector today is the production cost. So to make solar cells more attractive the cost must be decreased, this can either be done by making the solar cells more efﬁciently thus decreasing the area needed. Or to research on cheaper materials and processes to reduce the price per m2 at the cost of slightly lower efﬁciency.
An important area of research to both achieve a higher efﬁciency or cheaper materials this is to ﬁnd new materials to be used in solar cells. But these new materials also needs to fulﬁll certain criteria based on the physics behind the solar cells.
1.2 Physical properties of semiconductors and solar cells To explain how solar cells work a basic understanding of semiconductors is necessary. Semiconductors are a class of materials which exhibit electrical properties in between insulators and conductors; the simplest way to explain these properties is to introduce the concept of band structure. By combining the different discrete energy level for the electrons orbiting the atoms which constitutes the crystal and their positions is it possible to create a structure of allowed and forbidden energies of the electrons in the structure and the relationship between energy and momentum of the electrons.
The forbidden energies constitute a so-called band gap which separate the electrons strongly bound to the core (valence band) and electrons moving around in the lattice (conduction band).
To explain the big difference in physical characteristics between metals and insulators/semiconductors the distribution of electrons in the band structure needs to be explained. The distribution of electrons is governed by the so called Fermi-level which says that until this level all allowed states are occupied by electrons according to the Fermi-distribution. For metals the Fermi-level lies inside the conduction band which makes the band half-ﬁlled, this enables the electrons to move freely in the conduction band, which is the case for metals where the conductivity is high and the resistance low.
In semiconductors and insulators the Fermi-level lies in the band gap which means that only the valence band is ﬁlled and no electrons are in the conduction band at 0K. The difference between semiconductors and isolators lies instead in the magnitude of the band gap; there exists semiconductors which ranges from 0.17eV for InSb up to about 6.2eV for AlN while a typical insulator such as SiO2 have an even higher band gap of around 8.9eV. The magnitude of the band gap is what gives the semiconductor its speciﬁc properties and usages. At 0K a pure semiconductor behaves like an insulator due to that the conduction band is empty, but at higher temperatures the Fermi-distribution of the electrons makes the conduction band partially occupied, in an insulator the band gap is so large that the electrons require large amount of energy to be excited to the conduction band. A schematic overview of the difference is found in Figure 1.2. More
background on the theory of semiconductors can be found in the book ’Semiconductor materials:
an introduction to basic principles’.
Figure 1.2: Difference in band structure between a metal, semiconductor and insulator.
 In a solar cell the band should match the region where most of the photons in the solar spectrum are located, if the band gap is too large the photons will not be able to excite any electrons and if the band gap is too small the excess energy of the photons will convert into heat which reduces the overall efﬁciency.
When an electron is excited in a semiconductor it leaves behind a hole in the valence band, which will eventually recombine with the electron so that it falls back to the valence band, thus in a solar cell the hole and electron needs to be separated to create a potential which can be used.
A term often used in semiconductor physics is ’charge-carrier’, which is a particle that carries electrical charge in the semiconductor that can be either a hole or an electron in semiconductors and to determine which carrier type that dominates in a material the term majority charge-carrier is used. The separation of holes and electrons in a ’traditional’ solar cell is achieved by changing the concentrations of charge carriers in a material by adding appropriate atoms, creating a so called ’pn-junction’ were the electrons and holes are separated through diffusion. With ’appropriate atoms’ this means that they should be similar to the semiconductor atom but with one surplus or deﬁcit electron. So the atoms will either donate its extra electron to the conduction band (n-type) or remove one from the valence band (p-type), changing the majority charge carrier. The concentration difference between the two different regions induces an electrical ﬁeld which drives the charge carriers across the interfaces creating a region which neutralizes the charge difference and makes it possible for the electrons and holes to drift in opposite directions effectively separating the charges, Figure 1.3 illustrates the processes in such a junction.
So which material properties are necessary for a high efﬁciency photovoltaic cell? First of all the band gap needs to be narrow enough to absorb in the most intensive part of the solar spectrum.
The spectra from the sun can be found in Figure 1.4, from this spectra and knowledge of the physical processes in the solar cell Shockley and Quessier calculated the optimal band gap to be 1.4eV . The electrical current is maximized by small band gap but the electrical ﬁeld is Figure 1.3: The effect of carrier concentration and movement in a pn-junction. P denotes the holes and n denotes the excited electrons. maximized by a big band gap. Other important optical properties are low reﬂectivity and low transmission through the material which in other words mean high absorption of photons in the material to reduce the material needed.
Figure 1.4: The wavelength distribution from the sun at the surface of the earth.
 The ﬁnal material property that needs to mentioned is the transport of charge carriers in the material. As mentioned before the excited electrons have a certain lifetime before they recombines with a hole in the valence band thus loosing a charge carrier which reduces the current. So the conductivity of charge carriers in the material is of outermost importance. To improve the transport of electrons the obstacles for electrons should be few, in other words the resistivity should be low. Physical properties of the material which increases the conductivity are few impurities, grain boundaries and other dislocations. But conductivity is also strongly dependent on the number of charge carriers in the material, more charge carriers gives a better conductivity.
1.3 Goals speciﬁcations The goal of this work is to penetrate the area of the novel semiconductor material Mg2 NiH4 for photovoltaic applications. First to deposit thin ﬁlms with uniform composition using reactive sputtering, the structure of these ﬁlms will be characterized with XRD, Raman spectroscopy and
EDS to answer the following questions:
• What is the structure of the deposited ﬁlms and how does it change upon heating?
• Are there any impurities or defects in the deposited samples?
• How are the hydrogen atoms distributed in the material?
And to investigate whether or not the physical properties, such as band gap and electrical transport properties, makes Mg2.x NiH4 a suitable material in solar cells. The physical properties will
be investigated using spectrophotometry, four point probe resistivity measurements and Hallmeasurements to investigate the following questions:
• What is the band gap for the different materials, how does it change with composition and annealing?
• What is the resistivity of the samples compared to previous results and how does the resistivity depend on magnesium content and structure of the material?
• What are the physical properties such as mobility and density of the charge-carriers and how does it compared to previous measurements?
Finally the relationship between the structure and the physical properties will be discussed, and also compare these results with previous experiments.
2 Methods & Materials