A plasma approach to rare-earth based solar spectrum conversion

(1)

conversion

Citation for published version (APA):

Petcu, M. C. (2010). A plasma approach to rare-earth based solar spectrum conversion. Technische Universiteit Eindhoven. https://doi.org/10.6100/IR692937

DOI:

10.6100/IR692937

Document status and date: Published: 01/01/2010 Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

(2)

A plasma approach to rare-earth-based

solar spectrum conversion

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de rector magnificus, prof.dr.ir. C.J. van Duijn, voor een

commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op donderdag 9 december 2010 om 14.00 uur

door

Maria Cristina Petcu

geboren te Corabia, Roemenië

(3)

Copromotor: dr. M. Creatore

This research was financially supported by Energie Onderzoek Subsidie Lange Termijn (EOS-LT) of Agentschap NL (formerly SenterNovem).

Printed and bound by universiteitsdrukkerij Technische Universiteit Eindhoven The cover was painted and designed by Corina Popa. The painting was inspired by the idea of “light at the end of the tunnel”. The silhouette presented in the painting portrays every PhD student caught in the research whirl, trying to get the best results to be presented in the final manuscript – the PhD thesis. At the end, the student is the only person who ‘holds’ the solution to all the challenges and struggles – represented by the light in the center of the painting.

A catalogue record is available from the Eindhoven University of Technology Library

ISBN: 978-90-386-2397-9 NUR 926

A plasma approach to rare-earth based solar spectrum conversion by Maria Cristina Petcu – Eindhoven University of Technology, 2010

Subject headings: plasma deposition/ capacitive probe/ ion flux measurements/ amorphous silicon nitride/ rare-earth ions/ europium/ photoluminescence/ solar cells.

Trefwoorden: plasmadepositie/ capacitieve sonde/ ionenfluxmetingen/ amorf siliciumnitride/ zeldzame-aardionen/ europium/ fotoluminescentie/ zonnecellen

(4)

(5)

(6)

Chapter 1

(9)

(10)

I. Framework and overview of the research

I. A. Towards renewable energies

The necessity of energy production in a sustainable way has become the first priority among the majority of renewable energy technologies, e.g. wind energy, solar energy, geothermal and marine energy, all of them making use of the natural resources, such as sunlight, wind, etc. As reported by N. S. Lewis and D. G. Nocera1_{the world energy consumption rate is predicted to double from 13 TW}

in 2001 to 27 TW by 2050 and to triple to 43 TW by 2100. Within the world energy consumption of 13 TW reported in 2001, only 2% (0.286 TW) of the energy originates from renewable sources2_{(cf. Fig. 1).}

Figure 1 - Global energy consumption in 2001 as reported by N. S. Lewis2_.

By looking at the scenarios of the global transformation energy systems3

pro-jected over the next 100 years the renewable energy options are expected to be-come more important in the future. In particular, one of the oldest in the renew-able energy generation, i.e. solar energy, is expected to dominate the energy pro-duction in 2100, as shown in Figure 2. The challenge here is to dramatically re-duce the cost per kWh of delivered solar electricity.

Recently, Green et al4_{reported an extensive overview of the highest}

(11)

differ-ent cells and modules tested and commercialized on the market. New results for 17 cm2_{dye sensitized sub-module with an efficiency of 8.4%, whereas for an}

or-ganic stack of two cell of 2 cm2_{an efficiency of about 6% have been achieved.}

The highest efficiency solar cell of about 42% from a GaInP/GaInAs/Ge multi-junction device fabricated and measured under the new reference solar spectrum (ASTM G173-03) has been obtained.

Figure 2 - The future energy scenario on Global Change until 2100 as projected by German Advisory Council3_.

(12)

A lot of progress has been made in the solar technology since 1941, when for the first time a “Light sensitive electric device” with an efficiency of 1% was dem-onstrated by R. S. Ohl6_{. Out of the total electrical power capacity of 24 GW}

produced in 2008 by European countries which support programmes for energy production (cf. Fig. 3), 19% (4.59 GW) represents the electricity obtained from PV installations5_{. In 2008, the PV production grew by more than 80% which}

represents approximately 7.35 GW, as reported by A. Jager-Waldau5_{. This}

im-pressive growth was mainly due to the Spanish development market with a con-tribution of 2.5 - 2.7 GW, followed by the second and most stable market of Germany with a 1.5 GW contribution.

Considering their efficiencies, costs, fabrication technologies and materials used until now, solar cells have been divided in three categories, so-called “gen-erations”, as reported by M. A. Green7_{: first generation PV devices - with}

rela-tively high efficiencies but very high material costs, i.e. Si based wafer - mono and multi-crystalline silicon diffused emitter, ribbon Si; second generation solar cells based on low cost, low efficiency thin film technologies - amorphous Si (a-Si:H), microcrystalline Si (μc-(a-Si:H), Cadmium Telluride (CdTe), Copper Indium (Gallium) diselenide (CI(G)S); the third generation which would benefit of both low cost concepts (organic, dye sensitized cells) and high efficiency concepts (cf. Fig. 4)

Figure 4 - Solar cell classification (free after reported by M. A. Green); the IIIa and IIIb repre-sent the third generation solar cells combining the low production costs and intermediate con-versions efficiencies (IIIa) and intermediate production cost (IIIb).

(13)

Historically, crystalline silicon (c-Si) has been used as light-absorbing semi-conductor in most solar cells, even though it is a relatively poor light absorber and requires a considerable thickness (~100 - 250 μm) due to breakage process-ing problem. Nevertheless, it yields stable solar cells with good efficiencies and uses robust process technology developed already within the semiconductor in-dustry. The best recent achievement for crystalline Si with an efficiency of 25%, which represents a 57% improvement as compared to the results obtained in 1983, has been reported for Passivated Emitter and Rear-Locally-diffused device (PERL)8_{. Although crystalline silicon based solar cells technologies are}

dominat-ing the PV production with about 85% in 20095_{, thin film solar cells}

technolo-gies are starting in parallel to show significant results on the market, as depicted in Figure 5 a.

In particular, it is expected that thin film production, i.e. CdTe, CIS, dye sensitized cells could increase from 30% in 2010 (cf. Figure 5 a) to 37% in 2012 (cf. Fig. 5 b)5,9_{. According to these predictions}5_{, thin film technologies would}

double their production in 2015 when compared to 2010 (cf. Fig. 5b).

Figure 5 - a) Production capacity for thin film solar cell technologies9_{b) Present and projected}

PV production capacities of thin films and crystalline silicon based solar modules6_.

In general, one of the limits in getting higher efficiencies is the existent in-trinsic inherent losses in solar cells. In the following discussion, we will address specifically the case of c-Si solar cell losses. The first process to generate current in a solar cell is the generation of electron-hole pairs by light absorption

(14)

mecha-nism. However, a certain amount of light is reflected from the front surface or is not absorbed by the cell and thus, lost.

The first group of losses are called optical losses. One approach to reduce the optical losses consists of using antireflection (AR) coatings (i.e. amorphous sili-con nitride) on the top surface of the cell. Furthermore, in order to guarantee an enhanced light absorption path within the cell, the top surface of the crystalline silicon undergoes texturing. Another new and more generally applied light trap-ping approach is based on metallic nanostructures, i.e. plasmonics, intensively explored as promising solution towards higher efficiency solar cells10,11_.

Next to the optical losses, recombination losses are related to bulk and surface

recombination at the contacts. The defects existent in the structure of the bulk materials contributing to the recombination can be reduced by passivation whereas the surface recombination can be reduced either by placing a heavily doped layer under the front contact (e.g. n++ doped layer to keep the minority carriers, in this case holes, away from the recombination front contact) or at the rear of the cell (e.g. p++ doped layer meant to reflect, via the electric field, the electrons back towards the junction), or by placing a thin amorphous silicon ox-ide (a-SiO2) or aluminium oxide (Al2O3) layers under the top solar cell contacts,

as shown in Figure 6 a. The resistive losses can be reduced by different design of the solar cell which involves new contacting methods (cf. Fig. 6 a), i.e. interdigi-tated contacts at the back.

Figure 6 - Overview of the existing inherent losses in a solar cell: a) resistive and recombina-tion losses; b) spectral mismatch – the blue region of the solar spectrum efficiently used by the c-Si solar cell, while the orange shaded region is the lost solar irradiance.

(15)

The spectral mismatch is generated by the photons having higher energy than the bandgap of the solar cell creating an excess energy which will be lost as heat and by the photons with energy lower than the bandgap of the solar cell which will not be absorbed by the solar cell and therefore, lost, as depicted in Figure 6 b. A solution for thin film based solar cells, to reduce these losses could be either by building up a stack of cells, so-called tandem solar cells consisting of absorber layers with different bandgaps which will absorb different photons of the solar spectrum or by using convertors based on rare earth (RE) materials as it will be explained in section I. B.

Figure 7 - Existent losses in a c-Si solar cell and solutions used to minimize them. For example, optical losses could be reduced by using i.e. anti-reflection coatings8,12_{, backside reflectors}13-15_or

surface texturing8,16-21_{, recombination losses can be reduced by passivating both bulk and}

sur-face, while resistive losses could be reduced either by better design of the solar cell or by improv-ing the device contacts22,23_{. The spectral mismatch, discussed in detail in section I. B., could be}

minimized by using light conversion materials as previously reported24,25_{. An example of}

opti-mum efficiency of about 25% has been recently obtained by a PERL8_{solar cell by texturing}

the front surface of the cell, placing a double layer of antireflective coating and finally adding a SiO2 layer at the rear side of the c-Si cell and more advanced contacting.

As shown in Figure 7, within the losses existent in a c-Si solar cell, about 56% of the energy incident on the solar cell is lost due to the spectral mismatch.

(16)

Therefore, in order to increase the solar cells efficiency, one of the most impor-tant challenges of the solar cell research is to reduce the spectral mismatch. Cur-rently, research is oriented to decrease the losses in a solar cell and enhance the device efficiency by changing the structure of the device, or by using different types of materials for light and spectrum manipulation.

I. B. Light management approaches

A novel generic concept to increase the solar cell efficiency by manipulating the solar spectrum has been investigated in this thesis. Within the light manage-ment concept, there are two approaches which can lead to an increase of the solar cell efficiency, either by modifying the solar spectrum via light conversion mechanisms (i.e. solar convertors), or by modifying the bandgap of the absorber either via quantum confinement principle (i.e. quantum dots) or by combining different bandgap absorbers using a stack of different bandgap solar cells (i.e. multi-junction solar cells).

I. B. 1. Solar spectrum conversion

Solar spectrum conversion mechanisms to further enhance the solar cell effi-ciency have been extensively discussed by Richards et al24_{and Strümpel et al}25_.

According to this approach, the solar spectrum can be efficiently modified by shifting the photons towards a wavelength range where the solar cell has a better or higher response (cf. Fig. 6 b). The light conversion processes, i.e. down con-version (DC), down shifting (DS) or luminescent down shifting (LDS) and up-conversion (UC) are schematically depicted in Figure 8. Briefly, thermalization losses arising due to the excess of energy when an electron-hole pair with energy higher than the bandgap of the solar cell is created, could be reduced by using down-convertors, thus by converting one high energy photon into two low en-ergy photons or by down shifting the light into a region where the solar cell has a better or high response. Transmission losses arising as a result of photons with energy lower than the bandgap of the solar cell can be reduced by up-convertors combining two low energy photons into one high energy photon.

(17)

a) b)

c)

Figure 8 - Light conversion mechanisms: a) down conversion, b) up-conversion, c) down shifting.

A theoretical principle for solar energy conversion based on light conversion mechanism was proposed and evaluated in 1977 by A. Goetzberger and W. Greubel26_{. The authors had proposed a new design to collect and convert the}

light by using a stack of transparent sheets of material doped with fluorescent dyes. This configuration offered the advantage of separating and converting the light in different regions of the solar cell. Theoretical conversion efficiencies of 32% under optimum conditions and using a stack of 4 semiconductors with dif-ferent bandgap energy had been calculated. Substantial theoretical efficiency in-creases from 31% (Schockley-Queisser limit) to 47.6% and 38.6% for a conven-tional bifacial solar cell used in a combination with an up- and down-convertors, respectively, have been previously calculated by Trupke et al27,28_{. Based on these}

theoretical calculations insights into the up and down processes and materials have been obtained29-32_{. The luminescent down shifting concept (LDS), has been}

applied for the first time in 1979 by Hovel et al33_{. The authors have used two}

types of luminescent layers on top of a PV cell: a plastic sheet doped with organic dyes and a ruby crystal doped with chromium, resulting in a relative efficiency increase of 2%.

A class of materials which proved to be a good candidate for the light conver-sion processes and which will be discussed later in Section II. C. is the one of rare earth (RE) elements.As reported by A. Blasse and B. C. Grabmaier34_{, “A}

(18)

lumi-nescent material, also called a phosphor, is a solid which converts certain types of energy into electromagnetic radiation over and above thermal radiation”.

The RE ions, also known as lanthanide ions, are formed by ionization of a number of atoms which are usually incorporated and mostly stable as divalent or trivalent cations. They can be embedded in a lattice and emit radiation known as luminescence in a broad spectral range when different excitation radiation is used. It has been demonstrated that RE ions such as Europium (Eu), Samarium (Sm), Cerium (Ce) and Terbium (Tb) inserted in a specific oxide/nitride envi-ronment, can very efficiently convert UV radiation into visible light, as observed in the fluorescent lamps and displays35,36_{, Light Emitting Diodes (LEDs)}37-41_and

Organic LEDs (OLEDs)42_{. The incorporation of RE into large area polymer}

sheets for application in silicon solar cells to increase their efficiency has been described by B. S. Richards and A. Shalav29_{. The authors present initial results for}

UC and LDS but limited options for DC of high energy sunlight, for the con-figurations used in this research. In 1996 Gibart et al43_{have developed a device}

based upon two photons up-conversion concept to use the IR photons lost in a GaAs solar cell. The cell is placed on top of a 100 μm vitroceramic doped with RE elements Yb3+_{and Er}3+_{, respectively. The process is based on sequential}

ab-sorption followed by the energy transfer of two IR photons from Yb3+_{to Er}3+

which then emits one photon in the green part of the spectrum. Although the efficiency obtained has been reported to be very low for practical application in photovoltaics, i.e. 2.5% for an input excitation of 1W, the up-conversion princi-ple had been demonstrated. Two years later, D. Diaw44_{reported the results on}

LDS mechanism showing the effect of Eu3+_{ions incorporated in the}

antireflec-tion coating (SiO2) of an amorphous Si solar cell. The author used the ion

im-plantation technique to incorporate the Eu ions in the SiO2 coating. Therefore,

by using a dose of 2.5 x 1011_ions/cm2_{a relative efficiency improvement of 58%}

has been obtained. A clustering effect of Eu ions at the surface of the target was reported which is an important parameter which can decrease the light transmit-ted to the solar cell. The results of LDS process on crystalline Si solar cells encap-sulated with poly-vinylacetate (PVA) doped with different Eu3+_organic

com-plexes have been studied and reported by Le Donne et al45_{. The authors have}

ob-served light enhancement for low dopant concentration, while at higher concen-trations a relevant amount of light is absorbed without showing LDS. The

(19)

conse-quence of the strong quenching of the energy transfer from the organic host lat-tice to the Eu ions was a 1.1% decrease of the short circuit current density ob-served in the I-V curves of doped PVA coated cells as compared to the undoped cells. Recently, the application of LDS as a method to improve the poor spectral response of the solar cell for short wavelengths region has been discussed by Klampaftis et al46_{. The authors review the area of LDS describing the}

develop-ment in terms of materials over the last three decades, starting from early experi-ments where the choice materials available at that time were limited, to the mate-rials used nowadays which exhibit better efficiency and have better photostability.

I. B. 2 Bandgap engineering

The second approach within the light management concept is the bandgap modification of the absorber. Two ways towards this approach can be high-lighted: either increasing the number of bandgaps by means of tandem or multi-junction solar cell, or creating extra electron-hole pairs using nanostructures, e.g. QDs, quantum wells, etc..The multi-junction solar cell concept consists of com-bining several layers absorbing different energy light of the solar spectrum. An example of a multi-junction solar cell combining two types of thin films, i.e. amorphous silicon (a-Si:H) and microcrystalline silicon (μc-Si:H) both obtained by plasma techniques which will be later addressed, is presented in Figure 9 a.

Both films will absorb mostly different parts of the solar spectrum corre-sponding to their bandgaps, i.e. 1.7 eV for a-Si:H and 1.1 eV for μc-Si:H (cf. Fig. 9 b), such solar cell configuration achieving an efficiency of ~ 12% as previ-ously reported 4_{. However, besides the advantage of low cost production for thin}

films fabrication, other issues, such as thickness optimization of the layers to achieve the current matching, device stability etc., need to be addressed. Never-theless, other types of semiconductors can be used in multi-junction solar cells, e.g. GaAs, InGaP, etc, which have achieved the highest efficiency of ~ 42% re-ported at this moment4_.

As mentioned earlier, another way towards the bandgap engineering ap-proach to further increase the efficiency of solar cell is to use a group of materials currently under investigation for solar cell applications, namely nanostructures, i.e. quantum dots (QDs), nanocrystals (NCs), nanotubes/nanowires, etc. As

(20)

de-scribed by L. Tsakalakos47_{, within the four nanostructures categories, QDs have}

been investigated and implemented in different configurations for PV applica-tions. Therefore, since all the PV industry is already based on Si, Si QDs or Si NCs are rather promising to be investigated.

Figure 9 - a) Example of a tandem solar cell structure containing 2 intrinsic absorber layers, i.e. a-Si:H and μc-Si:H; b) External quantum efficiency spectra of both a-Si:H and μc-Si:H showing different bandgap values and the solar spectrum.

Efficient generation of multiple electron-hole pairs (i.e. excitons) per ab-sorbed photon with energy greater than twice the bandgap of Si (1.1 eV) repre-sents one possible route towards enhanced light conversion efficiency. Multiple exciton generation (MEG) whereby multiple excitons are produced upon absorp-tion of a photon has been observed for more than 50 years in bulk semiconduc-tors, e.g. Si, Ge, PbS, PbSe. Although several mechanisms have been proposed for MEG, most authors agree that MEG occurs as a result of Coulomb interac-tion between charge carriers, commonly known as impact ionizainterac-tion. This mechanism relies on a single, high-energy exciton, which produces a second exci-ton through a strong Coulomb interaction between electrons and holes. The same principle has been proposed for QDs to explain the electron-hole pair crea-tion, as reported by A. J. Nozik48_{and shown in Figure 10 a.}

(21)

By tuning the quantum dots sizes which results in a badgap tailoring, their conductive properties and PL emission can be controlled very precisely. There-fore, they have been studied for applications in transistors and diode lasers. Cur-rently, a lot of solar cell research has been focused on investigating different QDs materials and production methods to increase the cells efficiencies.

Figure 10 - a) Multiple electron-hole pair generation in quantum dot as reported by A. Nozik48_{; b) tandem solar cell based on QDs with different sizes.}

Several research studies regarding the investigation of Si QDs embedded in silicon nitride or silicon oxide layers by using different techniques have been re-ported49,49-61_{. Zacharias et al}60_{have presented a method for the preparation of Si}

NCs embedded in amorphous SiOx/SiO2 superlattices compatible with Si

tech-nologies which enable independent control of particle sizes, density and spatial distribution. An example of such QDs superlattices structure containing several layers placed on top of each other with different QDs sizes will absorb different parts of the solar spectrum as presented in Figure 10 b. Švrček et al57_have

re-ported the potential use of Si NC as possible candidate for PV industry to in-crease the solar cells efficiencies. The Si-NC samples were embedded into a spin-on-glass antireflecting SiO2. The authors have showed that using the layers as

downshifters, the internal quantum efficiency (IQE) of standard Si solar cell is enhanced. The experimental data have shown an increase of 0.4% of the

(22)

effi-ciency, while from the model, when using a convertor layer with 100% down conversion efficiency, an approximately 1.2% absolute improvement efficiency has been obtained.

Remarks

Several technologies used to process materials necessary for light manage-ment approaches have been proposed. Sol gel based technologies with lumines-cent materials dissolved as small particles, i.e. spin-on glass (SOG) methods or plasma based technologies, e.g. laser processing, magnetron sputtering, PECVD are the most common techniques used at the moment. Besides the several advan-tages of sol gel processes, such as easy to implement, application on a variety of substrates, low processing temperatures needed, there are some disadvantages such as thickness control or limited lifetime of sol-gel solutions (~12 months). Plasma based technologies have become essential for their advantages, e.g. dry process, high growth and etching rates, low substrate temperatures (100 - 400 °C), deposition on large substrate areas (> 1 m2_{) as well as low costs in thin film}

manufacturing, e.g. a-Si:H, a-SiNx, μc-Si:H are widely used materials in devices such as solar cells (2nd_{generation), LEDs, OLEDs, etc.}

In this thesis, both free-standing QDs and RE doped amorphous nitride and oxide matrices have been investigated and deposited by means of plasma based techniques briefly discussed in section II.

(23)

II. Plasma technologies for solar cell industry

As mentioned earlier, one of the plasma deposition methods used to incorpo-rate RE material in amorphous matrices is magnetron sputtering. This method has a high impact in a broad field of industrial applications and can be used to gener-ate small Si nanoparticles with controlled size and to sputter RE mgener-aterials. Basi-cally, in the magnetron process a target of interest is bombarded by energetic ions generated in a plasma discharge created in front of the target. The species resulted from the ion bombardment, i.e. individual atoms or clusters will be deposited on a substrate as a thin film or incorporated in a matrix, as schematically shown in Figure 11.

Figure 11 - Principle of magnetron sputtering63_.

As will be shown in Chapter 5, the presence of clusters may be the cause of non radiative energy transfer in RE- doped amorphous matrices. For the produc-tion of QDs this method requires a post-treatment to create small crystallites (~ 2 nm) which may require further parameter optimization60,62_{. In this thesis we have}

used a magnetron sputtering source to sputter RE metals, i.e. Eu and Sm used to incorporate them in a-SiNx and a-SiO2.More technical details are given in

sec-tion II. A., frame I, whereas the experimental results are extensively reported in Chapter 4 and 5.

(24)

Another plasma technique used for QDs production is by means of a

non-thermal plasma, i.e. Inductively Coupled Plasma (ICP). The main characteristic

of this type of plasma is the high electron temperature, typically 2-5 eV which is used to dissociate the gas phase precursors, while the atoms and ions are at room temperature. Amorphous to crystalline Si nanoparticles, free-standing or embed-ded in an amorphous matrix can be produced using an Ar/SiH4 plasma mixture.

One important parameter to control the size and crystallinity of the nanoparticles is the residence time, which is influenced by the reactor pressure, geometry and gas flow rates. A longer residence time leads to larger particles, since the time to react with SiH4 gas and enhance the growth is larger. In this work, an ICP

tech-nique to produce free-standing QDs explained in detail in Section II. A., frame II has been used, while the results are presented in section II. B.

A third technique for high deposition rate thin films is the expanding thermal plasma (ETP) developed in our group64_{, Plasma and Materials Processing group}

(PMP). This technique has been extensively used to deposit a-Si:H and a-SiNx

thin films. Due to the high deposition rates (2-20 nm/s) the ETP technique has been implemented in an industrial system, the so-called DEPx system by OTB-Solar and used to deposit a-SiNx thin films for the in-line manufacturing of c-Si

solar cells. The a-SiNx:H thin films are known for their anti-reflective properties

and bulk and surface passivation12,65_.

In this thesis, the ETP setup, explained in detail in section II. A, has been used in combination with a magnetron sputtering unit to deposit RE doped amorphous materials. An overview of the materials deposited using an ETP reac-tor in PMP group is presented in Table 1.

Currently, at PMP, the ETP reactor is used in a project related to the Si nanoparticles production, obtained from Ar/SiH4 plasma mixtures, embedded in

a-Si:H matrix, namely “Novel Synthesis and Passivation Routes of Silicon Nanocrystals for Photovoltaic Applications”. The aim of the project is to investi-gate novel plasma processing routes for improving the growth and surface func-tionality of silicon nanocrystals which will be formed in plasma and without any post processing, such as annealing. The silicon nanocrystals will be used to fabri-cate downshifters used for high efficient solar cells.

(25)

Material Technique Application Ref Zinc Oxide (ZnO) Plasma-Enhanced Metalorganic

Chemical Vapor Deposition - Expanding thermal Plasma

(MOCVD-ETP)

Solar cells (TCO) 66-69

Amorphous Silicon Nitride (a-SiNx)

Plasma Enhanced Chemical Vapor Deposition -Expanding thermal

Plasma (PECVD-ETP)

Solar cells (ARC), encapsulation (multilayers)

12,65

Amorphous Silicon Oxide (a-SiO2)

PECVD-ETP Solar cells (passivation),

encap-sulation (multilayers) 70-73 Amorphous, poly-, nano- crystalline Silicon (a-Si:H, NC-Si)

PECVD-ETP Solar cells 74-80

Metal oxides (e.g. Al2O3)

Plasma Assisted Atomic Layer Deposition (ALD)

Solar cells (passivation), encap-sulation (multilayers)

81-89

Table 1 - Materials overview used in PV obtained at PMP

II. A. Experimental setups

The results presented in this thesis have been obtained using an Expanding Thermal Plasma (ETP) in combination with a magnetron sputtering system, as shown in Frame I - Figure 12 a. An experimental study, to obtain and character-ize free-standing Si NPs using a Capacitively Coupled Plasma (CCP) reactor pre-sented in Frame II - Figure 13 a, has been performed and the results will be ex-plained in detail in the section II. C. of this thesis. The experiments were carried out at Colorado School of Mines, Chemical Engineering Department in the group of prof. dr. Sumit Agarwal. The project was aiming to control the emission intensity by changing the size of the Si NPs obtained in Ar/SiH4 plasma

mix-tures, but also to understand the formation mechanisms of Si NPs and how the residence time is influencing the structure and size of the particles (from amor-phous to crystalline).

(26)

Frame I – ETP-CVD and -PVD techniques

Figure 12 a) schematic view of the experimental setup which includes both the plasma source, namely a cascaded arc, and the deposition chamber with the expanding thermal plasma and gas recirculation cells, the rings for the injection of both precursors, i.e. NH3 and SiH4, the plasma

characterization tools, i.e. ion probe, ‘cold finger’ (CF), mass spectrometer, rf PP 13.56 MHz sputtering source; b) image of the plasma setup in the lab; c) picture of the ion probe showing the inner and outer rings; the flow direction represents the plasma direction; d) an image of the ‘cold finger’ mounted on the setup; behind the CF the substrate holder, where the samples are deposited, can be observed; e) picture of the plasma generated in front of the substrate holder generated by the magnetron sputtering source working in Ar, while sputtering the RE metal1_;

the picture is taken through the window placed on the lateral side of the substrate holder; f) picture of the 1”magnetron source for the sputtering of the RE metals; the source was used simultaneously with the ETP source in Ar/NH3/SiH4 plasma mixtures. Using this expertise, a

project on RE doped a-SiNx in collaboration with Chemistry Department at Eindhoven Uni-versity of Technology, has been started in 2006.

1_{In this work both Eu and Sm targets used were provided by Kurt J. Lesker Company}

b

f c

a

e _d

(27)

5 W b RF To Pump _Ar/SiH 4 Grid a 15 W c

Figure 13 – a) schematic view of the PECVD reactor with external electrodes used for synthesis of Si QDs; b) and c) pictures of Ar/SiH4 downstream plasmas powered at 5 and 15 W.

The reactor consists of a quartz tube, placed horizontal between two flanges, of approximately 8 mm inner diameter and 10 mm outer diameter. Two copper electrodes, powered by a rf 13.56 MHz generator provided with a matching unit, are placed externally on the tube. The argon gas is injected through one of the heads of the quartz tube (right side, Figure 13 a) while the pump is connected at the opposite side (left side, Figure 13 a). The SiH4 precursor is introduced

through a lateral connection from the same direction as Ar flow. During the experiments the Ar and SiH4 flows have been kept constant at 275 and 1 sccm, respectively. Using the 1 cm

dis-tance between the electrodes and varying their position on the tube we have changed the plasma coupling, i.e. downstream (close to the collection region) or upstream (close to the gases injec-tion) resulting in a change of particles structure and size, from amorphous to crystalline.

(28)

II. B. Investigation of Si nanoparticles synthesized by PECVD-CCP

technique

As mentioned in the previous section (I. B. 2), Si nanoparticles are one of the interesting materials to increase the solar cell efficiency. Previously, Mangolini et

al90_{have been investigating the production of Si NCs using a CCP system. The}

authors reported the formation of Si particles with nanometre size dimensions between 2 and 8 nm in timescales of milliseconds. A complementary work, by R. Anthony and U. Kortshagen91_{, reports a comparison of the PL efficiency between}

free standing Si a-NPs and Si NCs. They have studied the effects of plasma pa-rameters on the particle structure transition from amorphous to crystalline.

In this work we have investigated and characterized the properties of Si nanoparticles obtained from RF Ar/SiH4 plasma in a similar reactor used by R.

Anthony and U. Kortshagen described in detail in section II. A., frame II. In or-der to change the particles size and structure, i.e. amorphous vs. crystalline, we have optimized the reactor geometry, plasma power and precursor flow rates. Pre-viously, the formation mechanisms of the particles in RF SiH4 plasmas have been

extensively discussed92-96_{. A four step process to explain the particle formation,}

initiated by dissociative attachment followed by the formation of a negative ion i.e. SiH3-and SiH2-, has been proposed96.

Two cases will be discussed in this section: the particles and/or film deposited when the plasma is coupling in the downstream region, near the collec-tion/deposition area and the particles formed in the upstream region near the in-jection flows area. In both cases the particles are collected downstream, on fine stainless steel meshes or deposited on silicon (100) substrates. By tuning the RF power delivered to the reactor from 3 to 15 W and keeping all the parameters constant, several samples have been deposited in the downstream region. The samples have shown a strong blue PL emission shift with the RF power decrease as shown in Figure 14. The PL intensity signal with a maximum around 800 nm for the highest power (15 W) shifted down to 680 nm for the sample obtained at the lowest RF power applied (3 W) has been observed. The XRD results showed either no crystalline evidence or a very low crystallinity fraction.

(29)

Figure 14 - Normalized PL emission spectra for samples obtained in the downstream plasma region at different RF powers from 15 W down to 3 W using an excitation wavelength of 365 nm from a UV lamp.

The particles obtained when the plasma is generated in the upstream reactor region showed PL emission with a maximum peak intensity around 690 nm un-der the same 254 nm excitation wavelength from a mercury lamp. An example of the PL emission for the sample produced at 10 W is presented in Figure 15 a. Moreover, clear evidence of the crystalline structure from the XRD measure-ments has been observed, as shown in Figure 15 b. We can observe the (111) ori-entation of Si as being the main contribution to the crystalline structure of the particles. From the XRD measurements an average particles dimension using the Scherrer formula has been estimated:

θ β XRD Dp cos 89 . 0 ) (  λ , (1)

where λ is the wavelength and β is the line broadening at full width half maxi-mum intensity in radians and θ the Bragg angle. By using Eq. 1 we have ob-tained an average size of the particles of ~3 nm. However, if we look at the parti-cles collected on a TEM carbon grid in the first second of partiparti-cles production we can observe the tendency of the particles to agglomerate (cf. Fig 15 c). This

(30)

parti-cle agglomeration could be a serious drawback for the PL measurements due to the non-radiative emission at the grain boundaries. As shown in Figure 15 d, sin-gle particles with dimensions around 7 nm which clearly present the crystallographic planes have been observed.

b)

c) d)

a)

Figure 15 - a) PL emission spectrum of the sample obtained in the upstream plasma region at 10 W RF power using an excitation wavelength of 254 nm from mercury lamp; b) X Ray Diffraction (XRD) spectrum of the same sample obtained as a result of upstream coupling plasma; c) Transmission Electron Microscopy (TEM) image of the particles collected on a car-bon grid with the selected area electron diffraction patterns inset; d) zoom-in image on one of the particles of about 7 nm diameter.

(31)

Therefore, smaller particle dimensions have been calculated from XRD com-pared with TEM which could be related to technical issues, such as instrument broadening. Nevertheless, from PL emission spectra the diameter of the particles can be estimated using the following equation97_:

72 . 0 , , 73 . 3 ) ( _         Si g QD g p E E PL D (2)

where E_g_,_QD is the energy of the particles determined from PL and E_g_,_Siis the bandgap of bulk Si (1.12 eV).

By using Eq. 2 for particle energy in the range 1.6-1.9 eV as determined from Figure 13 a diameter values of 3-5 nm has been calculated. An oxidation study using FTIR has been performed. The formation of a native oxide shell around the Si NP is known to enhance the PL efficiency98_{since the oxides}

re-move the non-radiative surface states, as discusses previously by Ledoux et al97_.

Figure 16 – FTIR time dependence of a sample exposed to air for several hours.

The sample was exposed to the air for several hours and the chemical compo-sition was determined. As shown in Figure 16, the Si-O-Si stretching bond is increasing with the air exposure time, indicating the formation of the oxide shell around the NC structure.

(32)

II. C. Investigation of RE doped amorphous thin films obtained by

ETP-CVD technique

As mentioned in section I. B. 1, a promising group of materials, on the basis of their special optical properties, i.e. UV absorption followed by visible emis-sion, are RE ions. In this thesis the incorporation of Eu and Sm in amorphous silicon nitride (a-SiNx) and silicon dioxide (a-SiO2) using the combination of two

techniques described in section II. A., Frame I, will be discussed. Eu and Sm can be both incorporated in their divalent (Eu2+_{, Sm}2+_{) and/or trivalent (Eu}3+_{, Sm}3+₎

state (cf. Table 2). The emissions of both divalent and trivalent RE are strongly correlated to the 4f electrons in the excited state. Briefly, most of divalent RE ions have an excited state which consists of a broad band, i.e. of Eu2+_,

whereby the 5d electrons are not shielded. Therefore, the position of the band emission of divalent ions is strongly influenced by the surrounding lattice/matrix and also by the absorption cross sections at specific excitation wavelengths99_{. For}

example, for Eu2+_{incorporated in different host lattices the absorption cross}

sec-tion coefficient, , can vary between and 1 .

1 6 5 4f d 2 17 10 cm abs σ 18 2 10 7 . 1   cm  Ion Number of 4f electrons (ground state) First excited state Emission Eu2+ _{7 (}8_S 7/2) 4f65d1 band Eu3+ _{6 (}7_F 0) 5D0 line Sm2+ _{6 (}7_F 0) 5D0 line Sm3+ _{5 (}6_H 5/2) 5G5/2 line

Table 2 - Characteristics of Eu and Sm RE ions

In the case of trivalent ions, the 4f electrons are strongly shielded by the filled

5s2_{orbitals, thus a number of discrete energy levels are present which will lead to}

line emission at fixed wavelength position. For trivalent ions the emission is not anymore host lattice- dependent, but influenced by the absorption cross sections at specific excitation wavelengths100_{. For example, for Eu}3+_{incorporated in}

(33)

differ-ent complexes (e.g. methyl, trifluoromethyl, phenyl, naphthyl, etc) can range from to for a 355 nm excitation wavelength.

abs σ 2 19 10 2  cm 8501019cm2

Recently, Li et al101_{have reported the incorporation of Eu}2+_{in crystalline silicon}

nitride (Si3N4) based compounds to be highly promising red-emitting conversion

phosphor for white-LED applications. The authors reported visible emission shift towards high wavelengths of Eu2+_{incorporated in Sr}

2-xEuxSi5N8 with increasing

the Eu concentration using a 465 nm excitation wavelength, as shown in Figure 17.

Nowadays, amorphous matrices are investigated in the field of solar cells: in particular thin amorphous silicon nitride layers are already successfully applied as bulk passivating and antireflection coatings (ARC) and amorphous silicon diox-ide as surface passivating layer to reduce surface recombination.

Figure 17 - Emission spectra of Sr2-xEuxSi5N8 as reported by Li et al.101

Moreover, due to their high quality and deposition rate (i.e. ~ 5 nm/s ) achieved by using a plasma processing technique, silicon nitride layers are already implemented in the manufacturing process of c-Si solar cells. Therefore, a novel challenge of combining two functionalities, i.e. antireflection properties of a-SiNx

or a-SiO2 and spectral conversion properties of Eu, in one amorphous material

could open up a new route towards further improvement of the solar cell effi-ciency102_.

In this thesis, we investigate the properties of Eu- doped amorphous layers, i.e. a-SiNx and a-SiO2 deposited by means of two plasma- aided techniques,

(34)

namely sputtering (PVD) of Eu in combination with a remote expanding ther-mal plasma (ETP-CVD) fed with Ar/SiH4/NH3 and Ar/O2/hexamethyldisiloxane

(HMDSO) mixtures. However, the application of Eu doped a-SiNx and a-SiO2 is

not restricted only to c-Si solar cells. As shown in Figure 18, various PV tech-nologies exhibit different performance as function of the solar spectrum, as re-ported by Klampaftis et al46_{. The authors report on the external quantum}

effi-ciencies (EQE) defined as the ratio of the electron-hole pairs generated to the number of incident photons on the front surface of the solar cell. However, a common characteristic of all solar cells is that most of them present poor UV absorption where interaction with high energy photons occurs, as observed from EQE results (cf. Fig. 18). This problem could be partially solved by placing an Eu doped a-SiNx on top of the solar cell which will shift the UV photons into the

visible part of the solar cell.

Figure 18 - External Quantum Efficiencies of different solar cells as reported by Klampaftis34_:

Copper Indium Gallium Diselenide (CIGS), Cadmium Telluride (CdTe), multicrystalline silicon (mc-Si), Gallium Arsenide (GaAs).

The results of Eu doped a-SiNx and a-SiO2 will be extensively explained in

Chapters 4 and 5 of this thesis. In Figure 19 a and b two key results of the PL emission spectra of Eu and Sm doped a-SiNx are presented. For Eu doped a-SiNx

both Eu2+₍ _{) the band emission and Eu}3+_{line emissions}

( , J = 1, 2, 3, 4) under 266 nm excitation radiation have been ob-served. 7 1 6 4 5 4f d  f J F D0 7 5 _

(35)

For Sm it was much difficult to see emission from the divalent state, Sm2+_{, as}

it is harder to reduce Sm to the divalent state and the Sm2+_{first excited state is}

positioned higher than in case of Eu3+_{. Nevertheless, we observed Sm}3+_emission

(4G ₅₂ 6H_J₂, J = 5, 7, 9) when Sm was incorporated in a-SiNx matrix, as

shown in Figure 19 b. Moreover, Sm3+_{is more sensitive to concentration}

quench-ing than Eu3+_{, since Sm}3+_{suffer from concentration quenching due to}

cross-relaxation between neighbouring Sm3+_ions.

Figure 19 - PL emission spectra under 266 nm excitation wavelength of a) Eu incorporated in a-SiNx and b) Sm incorporated in a-SiNx matrix obtained at 40W, for different laser input

powers; Sm% (RBS):~0.4%.

II. D. Plasma diagnostics

To gain more insight into the plasma chemistry and a better understanding of the film growth, which will eventually lead to the improvement of the material properties, comprehensive information about the composition of the plasma and plasma-surface reactions is essential. Therefore, besides the electron density and temperature, a very important parameter to understand several processes occur-ring in plasmas which lead to the formation of a thin film, is the ion density. It has been demonstrated that ions have a high impact on the species contributing to the film growth, such as radicals. Previous ETP studies on the Ar/H2/N2/SiH4

mixtures used to deposit a-SiNx:H films have shown that the Ar+ ions produced

(36)

mainly create NH3+ ions which will later on be indirectly involved in the SiH4

dissociation as mentioned previously103,104_{. The authors have shown that SiH} 3

radicals are mainly responsible for the Si incorporation, while the N atoms in the film originate from N radicals present in the plasma105,106_{. Moreover, the}

reac-tions between SiHx+ ions and SiH4 molecules have been involved in a growth

mechanism of a-Si:H thin films obtained in Ar/H2/SiH4 ETP plasma, as

previ-ously proposed by Kessels et al75,107_{. The authors have investigated the}

contribu-tion of SiH3 radicals to the a-Si:H growth, using a combination of two plasma

diagnostics, cavity ring down absorption spectroscopy (CRDS) and threshold ionisation mass spectrometry (TIMS).

Within the framework of this thesis, namely, explorative studies of RE incor-poration in amorphous matrices through a gas phase plasma process, comple-mentary research regarding the design of a plasma diagnostic tool to work in harsh environments for ion flux measurements and cluster detection, has been performed. In order to get an estimation of the total amount of ions left after precursor dissociation and available for the sputtering of the RE target, a new plasma diagnostic technique has been developed. A pulse-shaped capacitive probe, inspired from earlier work of Braithwaite et al108_{, able to measure spatially}

resolved ion fluxes in depositing and non-depositing plasmas has been developed and implemented in a high deposition rate expanding thermal plasma system. The probe is tolerant of insulating films deposited on the probe surface and has been tested in Ar/NH3/SiH4 plasma mixture. The results are presented and

dis-cussed in detail in Chapter 2.

Moreover, as a result of the peculiar off axis ion flux peak found in Ar/NH3/SiH4 plasma mixtures, positioned at 5 cm from the axial position of the

reactor, the ion probe has been proposed to be a suitable tool to detect clusters formed at the boundaries between plasma beam and the recirculation cells, as shown in Chapter 3. A preliminary result of ion flux using the ion probe tech-nique measured in Ar/SiH4 plasma mixtures, used to produce nanoparticles for

higher efficiency solar cells applications, is presented in Figure 20. The results are correlated with the particles/clusters locally formed in the recirculation cells at the boundary between plasma jet and background gas, as discussed in detail in Chapter 3.

(37)

Figure 20 - Ion fluxes profiles determined from the ion probe measurements for SiH4 flow rates

of a) 3, b) 5, c) 7 and d) 10 sccs; the Ar flow and the pressure were kept constant to 33 sccs and 0.15 mbar, respectively.

III. Goal and outline of the thesis

Two important challenges in the solar cell field, namely efficiency increase and cost reduction have drawn the PV research attention towards novel materials and concepts. In the research presented in this thesis, we choose to focus on ex-plorative studies aimed to develop a down convertor material synthesized via gas phase methods, i.e. plasma-enhanced CVD and PVD processes, supported by process control tools, i.e. the ion probe, which provide further understanding of the plasma mechanisms and processes.

In this thesis we investigate the properties of RE materials incorporated in amorphous matrices, i.e. a-SiNx, a-SiO2, by means of two plasma techniques,

namely PVD of a metal target (Eu, Sm) in combination with a remote PECVD-ETP system, to achieve down shifting mechanism for higher efficiency solar cells. In order to get an estimation of the total amount of ions left after precursor dis-sociation and available for the RE sputtering, a complementary research

(38)

regard-ing a new design of a plasma diagnostic tool, namely a pulse-shaped capacitive probe, has been developed and implemented for the first time in a high deposi-tion rate expanding thermal plasma system. The novelty of this technique with respect to other plasma characterization techniques, e.g. the Langmuir probe, is its tolerance towards insulating film deposited on the probe surface. The probe was used to measure the ion flux in an Ar/NH3/SiH4 plasma mixture to deposit

silicon nitride thin films used as ARC in solar cells. Moreover, the probe has been proposed to detect clusters in harsh plasma environments.

Figure 21 – Thesis structure

This thesis is structured in two main parts, as depicted in Figure 21: the first part presents the framework of the research – a short overview of the PV industry and the challenges (section I. A.) in getting higher efficiencies, an overview of the

(39)

novel concepts, i.e. light conversion mechanisms, and the materials defined as good candidate to increase the solar cell efficiencies (section I. B.); section II. highlights the role of the plasma technologies for solar cell manufacturing. In section II. A. a summary of the plasma techniques used in this work is given, fol-lowed by an overview of the deposited layers (sections II. B. and II. C). Chapters 2-5 contain papers published in international peer-reviewed journals or submit-ted for publication.

IV. Conclusions and outlook

The research of this thesis is dedicated to investigate new amorphous RE doped materials, i.e. a-SiNx:Eu, a-SiO2:Eu, obtained by plasma assisted

tech-niques and to demonstrate whether the LDS mechanism proposed to increase the solar cells efficiency can be achieved. We have succeeded in the incorporation of RE elements, i.e. Eu and Sm, in amorphous silicon nitride and silicon dioxide host matrices already used for the 2nd_{generation solar cells, by combining two}

plasma techniques PECVD and PVD. In order to determine the ion flux avail-able for sputtering the RE target, ion flux measurements have been carried out.

A pulse-shaped capacitive probe has been implemented for the first time in a high rate deposition (2-20 nm/s) ETP system fed with Ar/NH3/SiH4 mixtures,

showing the probe tolerance towards the presence of insulating layers, as ex-plained in Chapter 2. The method is based on the discharge of an external ca-pacitor, Cp, while the bias voltage is maintained constant by changing the slope of

the applied bias signal. This method makes the ion probe a feasible tool which can be adapted to a sensor in order to monitor the film growth for different in-dustrial applications or to control the ion flux in various plasma discharges.

An interesting behaviour in the downstream region of the plasma at 5 cm off the axial position of the reactor, for higher SiH4 flows (>1 sccs) has been

ob-served, as explained in Chapter 3. We have proposed a hypothesis which explains the presence and position of the local enhancement of the ion flux. This hy-pothesis is based on cluster formation in the recirculation cells of the remote ETP at the boundaries between the plasma jet and the reactors walls. We have investi-gated the response of the clusters to pressure, thermophoretic and electrostatic forces. A shift of the off- axis ion flux peak position towards the axial position of

(40)

the reactor when increase the reactor pressure, has been observed. Moreover, we have investigated the influence of a cold finger element on the behaviour of the clusters. We have measured a substantial reduction of 80% of the off-axis peak from 3·1015_cm-2_s-1_{to 0.8 10}15_cm-2_s-1_{when the cold finger temperature was}

de-creased down to -180 o_{C. Therefore, the ion probe technique has been proposed}

as suitable tool to detect clusters in depositing plasmas.

The successful incorporation of Eu in both oxide and nitride matrices has been confirmed by X-ray Photoelectron Spectroscopy (XPS) and Rutherford Backscattering (RBS) measurements. The Eu incorporated is optically active showing the presence of both Eu2+_{and Eu}3+ _{emission, as confirmed by}

Photolu-minescence (PL) measurements. We have observed Eu3+_{peak emission}

corre-sponding to (J = 1, 2, 3, 4) transitions superimposed on Eu2+_band

emission corresponding to transition from 400 to 800 nm. However, the faster decay time values measured on both materials indicate the presence of both radiative and non-radiative transitions. Nevertheless, Transmis-sion Electron Microscopy (TEM) cross sectional investigations shows the pres-ence of Eu rich domains, so called ‘clusters’, with dimensions of approximately 3 nm. Furthermore, by combining the short decay time values we measured i.e. 12 to 18 ns for Eu2+ _{and 0.3 ms for Eu}3+_{, and the ‘clusters’ observed in the TEM}

im-ages, a hypothesis for the non-radiative decay has been drawn: the energy transfer between the luminescent active centers of the ‘cluster’ becomes very efficient, which will then be transferred to a defect in the vicinity of a ‘cluster’. More com-plete, in depth analysis on the investigation of the defects to improve the material quality and further enhance the PL signal will be necessary in the future. One way to reduce the defects is by annealing the layers at high temperature. Another solution proposed to reduce the density of defects is passivation via hydrogen or by H2 plasma. These procedures will lead to a reduction of the non-radiative

transitions and, therefore, an increase of the PL signal. Moreover, a challenging problem which needs to be further investigated is to control the Eu clustering. A possible solution would be to make use of metallorganic- based Eu precursors injected in the vacuum setup upon sublimation in an Ar/NH3/SiH4 plasma.

However, the presence of carbon in the precursor will challenge the development and control of the plasma chemistry to obtain carbon- free optically transparent layers, which still meet the anti-reflection and passivation requirements.

J F D₀ 7 5 _ 7 1 6 4 5 4f d  f

(41)

References

[1] N. S. Lewis and D. G. Nocera, Proceedings of the National Academy of Sciences of the United States of America 103, 15729 (2006).

[2] N. S. Lewis, Mrs Bulletin 32, 808 (2007).

[3] H.Graßl, J.Kokott, M.Kulessa, F.Nuscheler, R.Sauerborn,

H.J.Schellnhuber, R.Schubert, and E.D.Schulze, http://www. wbgu. de (2003).

[4] M. A. Green, K. Emery, Y. Hishikawa, and W. Warta, Progress in Photovoltaics 18, 144 (2010).

[5] A.Jäger-Waldau, http://re. jrc. ec. europa. eu/refsys/ (2009). [6] S.Ohl Russels, U. S. Patent 2 469 (1949).

[7] M. A. Green, Progress in Photovoltaics 9, 123 (2001). [8] M. A. Green, Progress in Photovoltaics 17, 183 (2009). [9] A.Jäger-Waldau, http://www. jrc. ec. europa. eu (2007).

[10] H. A. Atwater, S. Maier, A. Polman, J. A. Dionne, and L. Sweatlock, Mrs Bulletin 30, 385 (2005).

[11] H. A. Atwater and A. Polman, Nature Materials 9, 205 (2010). [12] B. Hoex, A. J. M. van Erven, R. C. M. Bosch, W. T. M. Stals, M. D.

Bijker, P. J. van den Oever, W. M. M. Kessels, and M. C. M. van de Sanden, Progress in Photovoltaics 13, 705 (2005).

[13] R. Bilyalov, C. S. Solanki, J. Poortmans, O. Richard, H. Bender, M. Kummer, and H. von Kanel, Thin Solid Films 403, 170 (2002). [14] L. Zeng, Y. Yi, C. Hong, J. Liu, N. Feng, X. Duan, L. C. Kimerling,

and B. A. Alamariu, Applied Physics Letters 89, (2006).

[15] J. Zettner, M. Thoenissen, T. Hierl, R. Brendel, and M. Schulz, Pro-gress in Photovoltaics 6, 423 (1998).

[16] Y. Inomata, K. Fukui, and K. Shirasawa, Solar Energy Materials and Solar Cells 48, 237 (1997).

[17] Y. Nishimoto and K. Namba, Solar Energy Materials and Solar Cells 61, 393 (2000).

[18] V. Y. Yerokhov, R. Hezel, M. Lipinski, R. Ciach, H. Nagel, A. Myly-anych, and P. Panek, Solar Energy Materials and Solar Cells 72, 291 (2002).

[19] D. H. Macdonald, A. Cuevas, M. J. Kerr, C. Samundsett, D. Ruby, S. Winderbaum, and A. Leo, Solar Energy 76, 277 (2004).

(42)

[20] H. Sai, Y. Kanamori, K. Arafune, Y. Ohshita, and M. Yamaguchi, Pro-gress in Photovoltaics 15, 415 (2007).

[21] M. Moreno, D. Daineka, and P. R. I. Cabarrocas, Solar Energy Materi-als and Solar Cells 94, 733 (2010).

[22] S. J. C. Irvine, V. Barrioz, D. Lamb, E. W. Jones, and R. L. Rowlands-Jones, Journal of Crystal Growth 310, 5198 (2008).

[23] I. Zumeta, J. A. Ayllon, B. Gonzalez, X. Domenech, and E. Vigil, Solar Energy Materials and Solar Cells 93, 1728 (2009).

[24] B. S. Richards, Solar Energy Materials and Solar Cells 90, 2329 (2006). [25] C. Strumpel, M. McCann, G. Beaucarne, V. Arkhipov, A. Slaoui, V.

Svrcek, C. del Canizo, and I. Tobias, Solar Energy Materials and Solar Cells 91, 238 (2007).

[26] A. Goetzberger and W. Greubel, Applied Physics 14, 123 (1977). [27] T. Trupke, M. A. Green, and P. Wurfel, J. Appl. Phys. 92, 1668

(2002).

[28] T. Trupke, M. A. Green, and P. Wurfel, J. Appl. Phys. 92, 4117 (2002).

[29] B. S. Richards and A. Shalav, Synthetic Metals 154, 61 (2005).

[30] B. S. Richards and A. Shalav, Ieee Transactions on Electron Devices 54, 2679 (2007).

[31] A. Shalav, B. S. Richards, and M. A. Green, Solar Energy Materials and Solar Cells 91, 829 (2007).

[32] T. Trupke, A. Shalav, B. S. Richards, P. Wurfel, and M. A. Green, Solar Energy Materials and Solar Cells 90, 3327 (2006).

[33] H. J. Hovel, R. T. Hodgson, and J. M. Woodall, Solar Energy Materials 2, 19 (1979).

[34] A. Blasse and B. C. Grabmaier, "Luminescent materials" 1994.

[35] C. R. Ronda, Journal of Alloys and Compounds 225, 534 (1995). [36] C. R. Ronda, Journal of Luminescence 72-4, 49 (1997).

[37] V. Bachmann, A. Meijerink, and C. Ronda, Journal of Luminescence 129, 1341 (2009).

[38] T. W. Chou, S. Mylswamy, R. S. Liu, and S. Z. Chuang, Solid State Communications 136, 205 (2005).

[39] Y. Q. Li, G. De With, and H. T. Hintzen, J. Electrochem. Soc. 153, G278-G282 (2006).

[40] M. R. Robinson, J. C. Ostrowski, G. C. Bazan, and M. D. McGehee, Advanced Materials 15, 1547-+ (2003).

(43)

[41] X. M. Zhang, L. F. Liang, J. H. Zhang, and Q. Su, Materials Letters 59, 749 (2005).

[42] T. Oyamada, H. Yoshizaki, H. Sasabe, and C. Adachi, Chemistry Let-ters 33, 1034 (2004).

[43] P. Gibart, F. Auzel, J. C. Guillaume, and K. Zahraman, Japanese Jour-nal of Applied Physics Part 1-Regular Papers Short Notes & Review Pa-pers 35, 4401 (1996).

[44] D. Diaw, Solar Energy Materials and Solar Cells 53, 379 (1998). [45] A. Le Donne, M. Acciarri, D. Narducci, S. Marchionna, and S. Binetti,

Progress in Photovoltaics 17, 519 (2009).

[46] E. Klampaftis, D. Ross, K. R. McIntosh, and B. S. Richards, S 93, 1182 (2009).

[47] L. Tsakalakos, Materials Science & Engineering R-Reports 62, 175 (2008).

[48] A. J. Nozik, Chem. Phys. Lett. 457, 3 (2008).

[49] G. Conibeer, M. Green, R. Corkish, Y. Cho, E. C. Cho, C. W. Jiang, T. Fangsuwannarak, E. Pink, Y. D. Huang, T. Puzzer, T. Trupke, B. Richards, A. Shalav, and K. L. Lin, Thin Solid Films 511, 654 (2006). [50] B. H. Kim, C. H. Cho, T. W. Kim, N. M. Parka, G. Y. Sung, and S. J.

Park, Applied Physics Letters 86, (2005).

[51] L. V. Mercaldo, P. li Veneri, E. Esposito, E. Massera, I. Usatii, and C. Privato, Materials Science and Engineering B-Advanced Functional Solid-State Materials 159-60, 74 (2009).

[52] L. V. Mercaldo, P. li Veneri, E. Esposito, E. Massera, I. Usatii, and C. Privato, Materials Science and Engineering B-Advanced Functional Solid-State Materials 159-60, 77 (2009).

[53] S. Mirabella, R. Agosta, G. Franzo, I. Crupi, M. Miritello, R. Lo Savio, M. A. Di Stefano, S. Di Marco, F. Simone, and A. Terrasi, J. Appl. Phys. 106, (2009).

[54] A. K. Panchal and C. S. Solanki, Thin Solid Films 517, 3488 (2009). [55] B. Rezgui, A. Sibai, T. Nychyporuk, M. Lemiti, and G. Bremond,

Jour-nal of Luminescence 129, 1744 (2009).

[56] B. Rezgui, A. Sibai, T. Nychyporuk, M. Lemiti, and G. Bremond, Jour-nal of Vacuum Science & Technology B 27, 2238 (2009).

[57] V. Svrcek, A. Slaoui, and J. C. Muller, Thin Solid Films 451-52, 384 (2004).

(44)

[59] Y. Q. Wang, W. D. Chen, X. B. Liao, and Z. X. Cao, Nanotechnology 14, 1235 (2003).

[60] M. Zacharias, J. Heitmann, R. Scholz, U. Kahler, M. Schmidt, and J. Blasing, Applied Physics Letters 80, 661 (2002).

[61] G. Conibeer, M. Green, E. C. Cho, D. Konig, Y. H. Cho, T. Fangsu-wannarak, G. Scardera, E. Pink, Y. D. Huang, T. Puzzer, S. J. Huang, D. Y. Song, C. Flynn, S. Park, X. J. Hao, and D. Mansfield, Thin Solid Films 516, 6748 (2008).

[62] S. T. H. Silalahi, Q. V. Vu, H. Y. Yang, K. Pita, and Y. Mingbin, Ap-plied Physics A-Materials Science & Processing 98, 867 (2010). [63] M.Mikula, http://www. umms. sav. sk (2010).

[64] D.C.Schram and G.M.W.Kroesen, U. S. Patent No. 4. 871. 580 (1989).

[65] J. Hong, W. M. M. Kessels, F. J. H. van Assche, H. C. Rieffe, W. J. Soppe, A. W. Weeber, and M. C. M. van de Sanden, Progress in Photo-voltaics 11, 125 (2003).

[66] I. Volintiru, M. Creatore, W. H. van Helvoort, J. L. Linden, and M. C. M. van de Sanden, Applied Physics Letters 89, (2006).

[67] I. Volintiru, M. Creatore, B. J. Kniknie, C. I. M. A. Spee, and M. C. M. van de Sanden, J. Appl. Phys. 102, (2007).

[68] I. Volintiru, M. Creatore, J. L. van Hemmen, and M. C. M. van de Sanden, Plasma Processes and Polymers 5, 645 (2008).

[69] I. Volintiru, M. Creatore, and M. C. M. van de Sanden, J. Appl. Phys. 103, (2008).

[70] M. Creatore, W. M. M. Kessels, Y. Barrell, J. Benedikt, and M. C. M. van de Sanden, Materials Science in Semiconductor Processing 7, 283 (2004).

[71] M. Creatore, J. C. Cigal, G. M. W. Kroesen, and M. C. M. van de San-den, Thin Solid Films 484, 104 (2005).

[72] M. Creatore, Y. Barrell, J. Benedikt, and M. C. M. van de Sanden, Plasma Sources Sci. Technol. 15, 421 (2006).

[73] B. Hoex, F. J. J. Peeters, M. Creatore, M. A. Blauw, W. M. M. Kessels, and M. C. M. van de Sanden, Journal of Vacuum Science & Technol-ogy A 24, 1823 (2006).

[74] A. H. M. Smets, D. C. Schram, and M. C. M. van de Sanden, J. Appl. Phys. 88, 6388 (2000).

(45)

[75] W. M. M. Kessels, M. G. H. Boogaarts, J. P. M. Hoefnagels, D. C. Schram, and M. C. M. van de Sanden, Journal of Vacuum Science & Technology A-Vacuum Surfaces and Films 19, 1027 (2001).

[76] W. M. M. Kessels, M. G. H. Boogaarts, J. P. M. Hoefnagels, D. C. Schram, and M. C. M. van de Sanden, Journal of Vacuum Science & Technology A-Vacuum Surfaces and Films 19, 1027 (2001).

[77] W. M. M. Kessels, R. J. Severens, A. H. M. Smets, B. A. Korevaar, G. J. Adriaenssens, D. C. Schram, and M. C. M. van de Sanden, J. Appl. Phys. 89, 2404 (2001).

[78] A. H. M. Smets, J. H. van Helden, and M. C. M. van de Sanden, Journal of Non-Crystalline Solids 299, 610 (2002).

). ).

[79] A. H. M. Smets, W. M. M. Kessels, and M. C. M. van de Sanden, Applied Physics Letters 82, 1547 (2003

[80] A. H. M. Smets, W. M. M. Kessels, and M. C. M. van de Sanden, Applied Physics Letters 82, 865 (2003

[81] S. B. S. Heil, P. Kudlacek, E. Langereis, R. Engeln, M. C. M. van de Sanden, and W. M. M. Kessels, Applied Physics Letters 89, (2006). [82] B. Hoex, S. B. S. Heil, E. Langereis, M. C. M. van de Sanden, and W.

M. M. Kessels, Applied Physics Letters 89, (2006).

[83] E. Langereis, M. Creatore, S. B. S. Heil, M. C. M. van de Sanden, and W. M. M. Kessels, Applied Physics Letters 89, (2006).

[84] E. Langereis, M. Creatore, S. B. S. Heil, M. C. M. van de Sanden, and W. M. M. Kessels, Applied Physics Letters 89, (2006).

[85] E. Langereis, J. Keijmel, M. C. M. V. de Sanden, and W. M. M. Kessels, Applied Physics Letters 92, (2008).

[86] S. E. Potts, W. Keuning, E. Langereis, G. Dingemans, M. C. M. van de Sanden, and W. M. M. Kessels, Journal of the Electrochemical Society 157, 66 (2010).

[87] G. Dingemans, P. Engelhart, R. Seguin, F. Einsele, B. Hoex, M. C. M. van de Sanden, and W. M. M. Kessels, J. Appl. Phys. 106, (2009). [88] G. Dingemans, R. Seguin, P. Engelhart, M. C. M. van de Sanden, and

W. M. M. Kessels, Physica Status Solidi-Rapid Research Letters 4, 10 (2010).

[89] G. Dingemans, M. C. M. van de Sanden, and W. M. M. Kessels, Electrochemical and Solid State Letters 13, H76-H79 (2010). [90] L. Mangolini, E. Thimsen, and U. Kortshagen, Nano Letters 5, 655