Electronic properties of {_m63c-Si:H [myc-Si:H] layers investigated with Hall measurements [Elektronische Ressource] / vorgelegt von Torsten Bronger

Torsten Bronger

Description

E l ec tron i c p rop erti es of µc - Si : H l ay ersi n v es ti gated w i th Hal l m eas u rem en tsVon der Fakult t f r Mathematik,Informatik und Naturwissenschaften derRheinisch-Westf lischen TechnischenHochschule Aachen zur Erlangung desakademischen Grades eines Doktors derNaturwissenschaften genehmigteDissertationvorgelegt vonDipl.-Phys. Torsten Brongeraus AachenBerichter: Universit tsprofessor Dr. Matthias Wuttig Dr. Hans L thTag der m ndlichen Pr fung: 28.2.2007Diese Dissertation ist auf denInternetseiten der Hochschulbibliothekonline verf gbar.3C on ten ts1 I n trod u c ti on 52 F u n d am en tal s 1 12.1 Structural properties of our µc-Si:H layers . . . . . . . 112.2 The density of states . . . . . . . . . . . . . . . . . . . 132.2.1a-Si:H . . . . . . . . . . . . . . . . . . . . . . . . 142.2.2µc-Si:H . . . . . . . . . . . . . . . . . . . . . . . 152.3 Electronic transport in a-Si:H and µc-Si:H . . . . . . . 172.3.1General considerations and the situation in c-Si . 172.3.2Hall effect . . . . . . . . . . . . . . . . . . . . . 212.3.3Transport in a-Si:H . . . . . . . . . . . . . . . . . 252.3.4 in µc-Si:H . . . . . . . . . . . . . . . . 282.4 µc-SiC . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 E x p eri m en tal d etai l s 3 43.1 Preparation of the layers . . . . . . . . . . . . . . . . 343.1.1PECVD . . . . . . . . . . . . . . . . . . . . . . . . 343.1.2HWCVD . . . . . . . . . . . . . . . . . . . . . . . . 353.

Sujets

• Physik

Informations

• Publié par: rheinisch-westfalischen_technischen_hochschule_-rwth-_aachen
• Publié le: 01 janvier 2007
• Nombre de lectures: 3
• Langue: English
• Poids de l'ouvrage: 2 Mo

Extrait

E l ec tron i c p rop erti es of ?c - Si : H l ay ers
i n v es ti gated w i th Hal l m eas u rem en ts
Von der Fakult t f r Mathematik,
Informatik und Naturwissenschaften der
Rheinisch-Westf lischen Technischen
Hochschule Aachen zur Erlangung des
akademischen Grades eines Doktors der
Naturwissenschaften genehmigte
Dissertation
vorgelegt von
Dipl.-Phys. Torsten Bronger
aus Aachen
Berichter: Universit tsprofessor Dr. Matthias Wuttig Dr. Hans L th
Tag der m ndlichen Pr fung: 28.2.2007
Diese Dissertation ist auf den
Internetseiten der Hochschulbibliothek
online verf gbar.3
C on ten ts
1 I n trod u c ti on 5
2 F u n d am en tal s 1 1
2.1 Structural properties of our ?c-Si:H layers . . . . . . . 11
2.2 The density of states . . . . . . . . . . . . . . . . . . . 13
2.2.1a-Si:H . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.2?c-Si:H . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Electronic transport in a-Si:H and ?c-Si:H . . . . . . . 17
2.3.1General considerations and the situation in c-Si . 17
2.3.2Hall effect . . . . . . . . . . . . . . . . . . . . . 21
2.3.3Transport in a-Si:H . . . . . . . . . . . . . . . . . 25
2.3.4 in ?c-Si:H . . . . . . . . . . . . . . . . 28
2.4 ?c-SiC . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3 E x p eri m en tal d etai l s 3 4
3.1 Preparation of the layers . . . . . . . . . . . . . . . . 34
3.1.1PECVD . . . . . . . . . . . . . . . . . . . . . . . . 34
3.1.2HWCVD . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2 Implementation of defects . . . . . . . . . . . . . . . . 36
3.3 Raman measurements . . . . . . . . . . . . . . . . . . . . 37
3.4 Preparation of the Hall samples . . . . . . . . . . . . . 37
3.4.1Offset voltage, noise, and alignment . . . . . . . 39
3.4.2Clean room preparation and RIE etching . . . . . . 40
3.5 Hall setup . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.5.1Sample geometry and circuit diagram . . . . . . . . 42
3.5.2Cryostat and magnet . . . . . . . . . . . . . . . . . 44
3.5.3Measurement mode and noise suppression . . . . . . 45
3.5.4Analysis program . . . . . . . . . . . . . . . . . . 48
4 Hal l m eas u rem en ts 5 3
4.1 About the Arrhenius plots . . . . . . . . . . . . . . . . 534 Contents
4.2 ?c-Si matrix . . . . . . . . . . . . . . . . . . . . . . . 54
4.2.1Activation energies . . . . . . . . . . . . . . . . . 61
4.3 Electron-irradiated samples . . . . . . . . . . . . . . . 64
4.3.1Mobility vs. carrier density . . . . . . . . . . . . 67
4.4 SiC . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5 M od el of n orm al l y d i s tri bu ted barri ers (ND B)7 3
5.1 Activation and barrier distribution . . . . . . . . . . . 73
5.2 Inverse Laplace transform approach . . . . . . . . . . . 75
5.3 Basic properties of the transformation . . . . . . . . . 76
5.4 Numerical approach . . . . . . . . . . . . . . . . . . . . 78
5.5 Parallel vs. serial network . . . . . . . . . . . . . . . 78
5.5.1A building set for a network . . . . . . . . . . . . 79
5.5.2Topological scaling . . . . . . . . . . . . . . . . . 81
5.6 Assumption: Normal distribution of barriers (NDB) . . . 84
6 D i s c u s s i on of the ex p eri m en tal ?c - Si : H d ata87
6.1 Applying the NDB model to the data . . . . . . . . . . . 87
6.1.1Barrier height versus carrier density . . . . . . . 88
6.1.2 heights of samples of different crystallinity 89
6.1.3Data from previous works in the NDB model . . . . . 91
6.1.4Serial transport . . . . . . . . . . . . . . . . . . 94
6.2 High doping: second process as the upper limit . . . . 97
6.3 About the origin of the barriers . . . . . . . . . . . . . 99
6.3.1Donor agglomerations . . . . . . . . . . . . . . . . 101
6.3.2Heterojunction . . . . . . . . . . . . . . . . . . . 103
6.3.3Seto model . . . . . . . . . . . . . . . . . . . . . . 103
6.3.4Conclusion . . . . . . . . . . . . . . . . . . . . . . 104
6.4 Differential mobility . . . . . . . . . . . . . . . . . . . 105
6.4.1Conclusion . . . . . . . . . . . . . . . . . . . . . . 111
7 Su m m ary 1 1 4
A P rogram s ou rc e 1 2 45
C hap ter 1
I n trod u c ti on
Being the most promising form of renewable energy, the area of
photovoltaics has gained much interest over the last years, in
almost any respect: scientifically, ecologically, economically,
and politically. So far, this fruitful period has not cooled
down. On the contrary, this research area is still growing and
increasingly funded by governments and companies.
Although the underlying effects had been known in principle for a
long time (BØcquerel, 1839; Einstein, 1905), and viable cells have
been made since the 1950s (Chapin et al., 1954), the first
applications were in space travel, where no other energy source is
available (Green and Lomask, 1970). Silicon-based solar cells are
especially successfull with efficiencies of currently up to 24%.
Still, photovoltaics and solar power in general are minor energy
carriers.
The two main technologies used today are combustion of fossil
fuels and nuclear fission power. However, both methods have
serious disadvantages which cannot be overcome by technological
improvements. The most important issues are the pollution with
combustion products, in particular of the Earth’s atmosphere, and
the availability of raw material.
As far as mineral oil and gas is concerned, which are the most6 Introduction
Geothermal1,600
Other
renewables
Solar thermal
1,400 (heat only)
1,200
Solar power
(photovoltaics
1,000 and solar thermal
generation)
800
Wind
600
Biomass
(advanced)
Biomass
400 (traditional)
Hydroelectricity
Nuclear power
200 Gas
Coal
Oil
0
2000 2010 2020 2030 2040 2050 2100
Year
Figure 1.1: A possible transition path to sustainable energy
carriers recommended in Gra l et al. (2003). The diagram shows
the energy mix versus time, most of it extrapolated to the future.
Figure 1.2: Global oil production outside the OPEC and GUS area.
From Zittel and Schindler (2004).
important global energy carriers, Hubbert (1956) proposed his
famous ‘‘peak theory’’: For any given oil field, or group of
field, or even all global oil resources, the oil extraction versus
time vaguely resembles a bell curve. Since then, the Hubbert peak
has already been observed for most small-scale areas as well as
Great Britain and Norway (Zittel, 2001), see also fig. 1.2. It is
subject of heated discussion whether or not the global peak has
yet been reached.
Therefore, the world has to establish new sources of energy long

before the current raw materials are exhausted, since a decline --

however smooth -- is economically unbearable in the face of
increasing need for electrical and mechanical power.
Primary energy use [EJ/a]7
Photovoltaics combines many of the requirements for a sustainable
energy source: Sunlight has a potential of up to 1kW/m†, so that
theoretically, a 1=10000 of it equals the current global primary
energy consumption; it will be available virtually forever; it can
easily be applied in an environment-friendly way; it is more
homogeneously distributed over the globe than any other energy
carrier; its main device material silicon is abundantly available.
This explains the recent boom in photovoltaics. In order to be a
seriously competitive technology, solar cells must be cheap and
efficient. Thus, the technological development is continuously
looking for new production concepts or the optimisation of them,
while keeping a certain minimum efficiency in order to remain
economically viable.
The above mentioned silicon-based solar cells consist of
crystalline silicon (c-Si) modules. Commercially available
modules have an efficiency of up to 20% and long lifetimes of more
than 20 years. Unfortunately, their manufacturing still is too
costly for superseding conventional energy sources.
Therefore, thin-film solar cells have received increased interest.
While crystalline cells have a thickness of a couple of
millimeters, thin-film cells are only a few microns thick.
SuccessFul systems for thin-film solar cells are amorphous silicon
1(a-Si:H), CdTe, and CIGS , with silicon being the most popular,
because of its cheapness and non-toxicity.
Silicon-based thin-film solar cells are promising for three
reasons. First, the material is cheap and environmentally
friendly. Secondly, consumption is small. Note that
recently, the semiconductor industry suffered a serious shortage
of raw silicon, which hit the solar cell branch very hard. And
thirdly, the production temperatures are very low (in our case
<200?) in comparison with the production of crystalline silicon.
This too, leads to low production costs.
1copper-indium-gallium-diselenide8 Introduction
Purely a-Si:H solar cells have a serious drawback, however: Their
efficiency of about 6% is pretty low due to the poor electronic
transport in the material. Partly it is due to defects that catch
charge carriers, partly it is due to the intrinsic transport
properties of such structurally disordered materials in general.
Additionally, the Staebler-Wronski effect (Staebler and Wronski,
1977) causes amorphous silicon to degrade under illumination, so
that commercial mod