Incorporation of alkali metals in chalcogenide solar cells

Solar Energy Materials & Solar Cells 143 (2015) 9–20

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat

Review

Incorporation of alkali metals in chalcogenide solar cells P.M.P. Salomé a,n, H. Rodriguez-Alvarez a,b, S. Sadewasser a a b

International Iberian Nanotechnology Laboratory (INL), Av. Mestre José Veiga s/n, 4715-330 Braga, Portugal PVcomB, Helmholtz Zentrum Berlin, Hahn Meitner Platz 1, 14109 Berlin, Germany

art ic l e i nf o

a b s t r a c t

Article history: Received 2 March 2015 Received in revised form 8 May 2015 Accepted 6 June 2015

Since 1993 it has been recognized that for solar cells based on the chalcogenide absorber material Cu(In,Ga)Se2 (CIGSe) the incorporation of Na is crucial to obtain the highest values of power conversion efficiency. Since then, many reports have investigated the effects of Na in different chalcogenide solar cell materials. In the present review, we discuss the various sodium incorporation strategies for chalcogenide solar cells that have been reported. We briefly discuss the origin and the different interpretations of the positive effects that the presence of sodium provides in CIGSe and Cu2ZnSn(S,Se)4 (CZTSSe) solar cells. On the contrary, at the current stage of development of CdTe solar cells, Na is unwanted instead of incorporated. We review the various incorporation methods that have been reported: (a) diffusion from substrates like soda-lime glass, specialty glasses and sodium doped molybdenum (MoNa) layers; (b) external strategies by deposition of Na-containing compounds before, during and after the absorber growth; (c) other non-conventional methods. For each method, we present a literature review and critically present benefits and weak points. Finally, we provide an overview of characterization methods capable of directly probing the presence of sodium. & 2015 Elsevier B.V. All rights reserved.

Keywords: Solar cells Chalcopyrites Cu(In,Ga)Se2 (CIGS) Kesterites Cu2ZnSn(S,Se)4 CdTe Thin films

Contents 1. 2.

3.

4.

5.

n

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effects of Na in chalcogenide solar cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Sodium in CIGSe solar cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Sodium in CZTSSe solar cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Sodium in CdTe solar cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sodium incorporation strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Diffusion from the substrate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Soda-lime glass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2. Specialty glasses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.3. MoNa back contact layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. External strategies for incorporation of Na . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1. Incorporation of Na-compounds by means of thermal evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2. NaF as a precursor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.3. Na incorporation after the absorber growth process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.4. K incorporation after the absorber growth process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Unconventional methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Techniques for direct measurement of Na . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Glow discharge optical-emission spectroscopy (GD-OES) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Secondary ion mass spectrometry (SIMS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. X-ray photoelectron spectroscopy (XPS) and auger electron spectroscopy (AES). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Atom probe tomography (ATP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Corresponding author. Tel.: þ 351 253 140 112. E-mail address: [email protected] (P.M.P. Salomé).

http://dx.doi.org/10.1016/j.solmat.2015.06.011 0927-0248/& 2015 Elsevier B.V. All rights reserved.

10 10 10 11 12 12 12 12 13 13 13 13 14 15 15 15 16 16 16 16 16 17

10

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1. Introduction Na plays an important role in compound semiconductor thin film solar cells. Depending on the absorber material, the incorporation of Na into the solar cell stack can have a beneficial or harmful effect on the solar cell electrical performance. It also plays an important role how and at what time Na is introduced. For chalcopyrite Cu(In,Ga)Se2 (CIGSe) based solar cells, it has been shown that an incorporation of Na into the absorber is beneficial for the electrical performance of the devices. However, for CdTe based solar cells Na degrades the properties of the transparent conductive oxide (TCO) and causes unwanted changes to the morphology of the absorber and is therefore avoided. For the kesterite Cu2ZnSn(S,Se)4 (CZTSSe) compound, a more recent studied material, the full effects of Na are still rather unknown, but preliminary results indicate a positive effect in the electrical performance as well. For these three types of solar cells, the most studied case is the one of CIGSe since in this case it is very wellknown that the presence of Na within the CIGSe absorber layer is beneficial and is therefore desired. In this paper we review: (1) the general effects of Na on the different absorber materials, (2) the strategies that exist to incorporate Na in the absorber layer of thin film solar cells and (3) the characterization tools suitable to identify the presence of Na. The effects of Na in CIGSe, CZTSSe and CdTe, will be reviewed in the second chapter. We will describe how a doping concentration of Na in CIGSe was found to have a positive influence in CIGSe solar cells, briefly address some of these properties and explain why the amount of Na that is incorporated needs to be controlled. In the third chapter we will describe in detail the different approaches that have been presented for the incorporation of Na in CIGSe and CZTSSe solar cells. The discussed methods are: diffusion from the soda lime glass substrate or similar specialty glasses (Fig. 1a); the use of a precursor layer containing Na, i.e. a sodium doped Mo back contact (Fig. 1b) or a NaF precursor layer (Fig. 1c); external strategies for incorporation of Na as for example the evaporation of Na-compounds during the growth of the absorber layer (Fig. 1d) or after the growth of the absorber layer (Fig. 1e). In the fourth chapter we

describe the techniques and tools used to directly measure the presence of Na.

2. Effects of Na in chalcogenide solar cells 2.1. Sodium in CIGSe solar cells The importance of Na in the electrical performance of CIGSe was first observed in 1993 by Hedström et al. [1] and Holz et al. [2]. The work of Hedström et al. focused on solar cells properties: by comparing the growth of CIGSe solar cells on soda-lime glass (SLG), borosilicate, sapphire, and alumina substrates, the authors observed that CIGS solar cells grown on soda-lime glass leads to a better electrical performance, a stronger (112) texturing and a high concentration of Na throughout the CIGSe layer. Although the authors were not able to distinguish if the higher electrical performance was due to the (112) texturing or due to the presence of Na, the link between the presence of Na and high electrical performance was established. A more detailed study from the same authors showed that the improvement in the electrical performance was associated with an increase of the open voltage circuit (Voc) and the fill factor (FF) and that in addition to the texturing the grain size of the CIGSe also improved [3]. Holz et al. focused on the electric properties of CIGSe and concluded that the presence of Na, even at a level of 1015 atoms/cm  3 increased the electrical conductivity either if Na was implanted on CIGSe grown on Nafree substrates or if the CIGSe was grown on SLG [2]. The relation between growth temperature and diffusion of Na from the SLG was introduced in 1995 by Zweigart et al. [4] by showing that Na diffusion is connected to the order of deposition of the elements. Several subsequent studies confirmed that the presence of Na improved the electrical performance of CIGSe solar cells [4–7]. At that point, several studies pointed to the fact that the increase in the open circuit voltage, VOC, of solar cells was due to the higher built-in voltage (Vbi) of the junction created by Na increasing the free carrier concentration of CIGSe [8–10]. In 1997, a model was presented describing that by adding Na to CIGSe the number of acceptors does not change but the number of donors is

Fig. 1. Schematic representation of different Na incorporation strategies: (a) diffusion from the soda lime glass substrate; (b) diffusion from a sodium doped Mo layer; (c) diffusion from a NaF precursor layer; (d) introduction of Na during the growth of the absorber by evaporation of Na-containing compounds and (e) post-deposition evaporation of Na-containing compounds into the absorber.

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

11

Table 1 An overview of the most efficient reported and known power conversion efficiency values of CIGSe solar cells for the following Na incorporation strategies: diffusion from the glass substrate, post deposition of NaF; NaF precursor layer; sputtering of glass and a sodium doped Mo layer (MoNa). Na incorporation

Efficiency (%)

Substrate

CIGSe Fabrication Process

Efficiency of SLG reference (%)

Refs.

Glass Post-deposition NaF Precursor NaF sputtering of glass MoNa

20.8 20.4 18.7 17.7 16.6

SLG Flexible polyimide Corning glass Flexible zirconia Stainless steel

Co-evaporation Low temperature Co-evaporation Co-evaporation Co-evaporation Co-evaporation

– N/A 18.7 18.2 17.8

[61] [62] [63] [64] [65]

dramatically reduced, thereby changing the compensation level [11]. This model has since then been supported by several other studies [9,12–17]. After the initial studies on Na, the importance of adding Na to CIGSe to obtain solar cells with high open circuit voltage and fill factor was well documented and subsequently, many Na incorporation strategies were reported. However, with new incorporation strategies many contradictory reports of the effects of Na on CIGSe appeared. One of these contradictory issues is the relation of Na with grain size. It has been reported that Na increases the grain size [3,18,19], that it decreases the grain size [23], and also that there is no correlation between Na and the grain size [16,21,24,25]. Another contradictory topic is related to the preferentially oriented growth. Some studies indicate that the presence of Na promotes the (220)/(204) growth [26], while others show a (112) preferential orientation [2,20,27], yet other studies report that it is the way Na is provided that influences the preferential orientation [3]. Nevertheless, some of the effects that are widely accepted are:

 Na increases the free carrier concentration of CIGSe. This is due

   

to a decrease of the number of electrically active donors, reducing thus the compensation level and therefore the effective p-type doping, and consequently increasing the built-involtage. This effect leads to higher values of open circuit voltage and fill factor [10,26,28]. However, the exact physical mechanism by which this happens is still debated [29,30]. CIGSe solar cells that do not contain Na suffer from low values of Voc and a roll-over effect in the J–V behavior [10,31–33]. Na hinders the elemental intermixing of Ga and In [20,25,34,35]. The positive effects of Na on the electrical performance of solar cells occur when Na is present in the CIGSe with concentrations close or up to 1019 atoms/cm3 [1,8,34–36]. The amount of Na present in a CIGSe film is higher in films with a smaller grain size [37] and Na accumulates at the surface of the CIGSe films [6,21,38,39] and possibly in the grain boundaries [40–43].

Although it is known that Na increases the carrier concentration, there are still several models about how exactly Na affects the electronic properties of CIGSe. One of them is based on the reduction of the donor defect InCu [8,44] that leads to a change of the compensation ratio. Another, competing model is based on the premise that Na is only electrically active at the grain boundaries [45] and that it acts on Se vacancies in conjugation with O2 [5,38,46]. In addition to the above effects, there is a recent topic which has affected the CIGSe industry and has so far only been known from silicon modules: the potential induced degradation (PID). It was found that during operation and at the right temperature, humidity and electrical connections, CIGSe solar cells fail to work due to a possible excess of Na diffusion from the glass into the CIGSe [47,48]. It is therefore important to evaluate the effect of Na

also during the operation of solar modules. To what extent the effect of the PID is important to mitigate is still unknown but there are several CIGSe companies that already provide PID-free modules according to their data-sheets: tmsc solar, Stion, Solarion, just to name a few, showing that this is an issue relevant to the industry. There are other disputed topics regarding the influence of Na in CIGSe: the promotion of MoSe2 by Na [38,49–52]; the exact location and effect of Na in the grain boundaries or in the bulk [43,53]; and the relation between oxidized compounds with sodium [5,54– 57]. The reason for these numerous contradictory results is likely connected with the different growth processes used for the deposition of CIGSe. CIGSe is an intrinsically doped material, i.e. details of the growth process and conditions ultimately determine the material´s electronic properties. Naturally, also other properties than the electronic ones may depend on the way the material is grown. This review aims mainly at a discussion of the incorporation strategies for Na in CIGSe and CZTSSe, with only a brief discussion of the effects of Na and techniques to directly measure Na in these materials. For a detailed description of the influence of Na on the electrical properties of CIGSe, the reader is referred to Refs. [58–60]. In table 1, power conversion efficiency record values for some selected Na incorporation strategies are presented. The table only shows values for cases where there is significant information to clearly identify the Na incorporation strategy. The efficiency values reported by Solar Frontier (20.9%), Solibro (21.0%) and ZSW (21.7%) are not present since their processes are not entirely known. At this point it can also be noted that most of the research groups use SLG and thus it is a Na incorporation method much more optimized than possibly other methods. The table also shows that the other methods are capable of competing with SLG references and that most of them can be used in light-weight flexible substrates like polyimide, stainless steel and zirconia sheets. Each of these incorporation methods will be discussed further in the text. 2.2. Sodium in CZTSSe solar cells The research on CZTSSe thin film solar cells is less developed than that on CIGSe. The first solar cell based on CZTSSe was prepared by Ito et al. in 1988, and during the 1990s there was a small number of publications from Japan [66,67] and Germany [68]. It was not until the late 2000s that more research groups got interested in the material and started to study its properties in detail. As a consequence, the power conversion efficiency has now reached a value of 12.6% [69]. In spite of the large number of reports about CZTSSe, the topic of Na incorporation in CZTSSe is fairly undeveloped. Nevertheless, there are some preliminary reports that will be addressed. As is the case for CIGSe, CZTSSe is also an intrinsically doped, highly compensated semiconductor [70–72] and therefore comparing the effects of Na in CZTSSe grown by different techniques may lead to misinterpretations. Several reports claim that Na increases the grain size. Hlaing Oo et al. demonstrated that for Cu2ZnSnS4 (CZTS) prepared by co-

12

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

sputtering of binary sulfides and a posterior annealing process, the presence of Na increased the grain size of the final films [73], which was also confirmed by Bras et al. [74]. Prabhakar et al. prepared CZTS using a modified ultrasonic spray pyrolysis technique and compared films grown on SLG with films grown on Nafree glass [75]. The authors concluded that the presence of Na increases the grain size, the (112) texturing, and the free carrier concentration. This is in accordance with what was reported by Li et al. for co-evaporated Cu2ZnSnSe4 (CZTSe) where Na improved the hole density and mobility, increasing the built-in-potential of the junction and consequently also the Voc and the fill factor of the solar cell [76]. The same trend was observed by Nagaoka et al. where it was experimentally shown that an increase of the Na content in CZTS led to an increase in the effective hole concentration and a decrease in the thermal activation energy [77]. These findings suggest that Na has similar effects in CZTSSe as in CIGSe. However, it was also found by comparing secondary ion mass spectrometry (SIMS) that CZTSe films grown by co-evaporation on Mo-coated SLG incorporate half the amount of Na than the amount that CIGSe incorporates [78]. To counter this effect, a NaF precursor layer deposited on the Mo is needed so that CZTSe can incorporate more Na [79]. One of the most complete studies of Na in CZTSSe shows that Na is likely segregating along surfaces and grain boundaries, and that the amount of Na at the surface is correlated with the presence of oxygen [80]. Contrary to CIGSe, where the pure-selenide version of the material is much more studied and provides higher power conversion efficiency values than the pure-sulfide compound, in CZTSSe, it seems that the research is going at equal pace in the sulfide version, CZTS, and the selenide version, CZTSe. At this point it is hard to say if Na provides the same effect for CZTSe as for CZTS or even for the mixture, CZTSSe, and more studies are needed to understand if the findings depend on chalcogen type. It should be noted that the growth methods used for CZTSSe are similar to the ones used for CIGSe, and therefore the incorporation strategies reviewed in this article are as valid for CIGSe as for CZTSSe. 2.3. Sodium in CdTe solar cells Contrary to the previously mentioned absorber layers, in the current status of the development of CdTe solar cells, Na is not introduced, rather its incorporation is unwanted. CdTe solar cells are prepared in the superstrate configuration, which means that the CdTe layer is deposited on a stack of CdS deposited on top of a transparent conductive oxide (TCO). This stack is submitted to the CdTe high growth temperature. At those high temperatures, Na from the soda lime glass diffuses into the layers and it lowers the electrical performance of the resulting solar cells [81]. To prevent this degradation, most commercial CdTe solar cells incorporate a diffusion barrier layer made from Al2O3 or SiO2, deposited prior to the TCO [82]. In fact, for many years, the record CdTe solar cells were fabricated on Corning speciality boro-silicate Glass (7059 Corning glass) [83]. This glass is free of Na and therefore the influence of Na on the solar cell stack was avoided. The reason for the lower performance in CdTe solar cells where Na is allowed to diffuse is a debated topic in the literature. In 1982 it was identified that Na can act as a shallow acceptor with an activation energy of 59 meV [84] and thus it could potentially be used to increase the acceptor density and the resulting open circuit voltage [85] of solar cells, similar to what occurs in CIGSe. However, Romeo et al. have reported the appearance of NaCl crystallites during the CdCl2 treatment if SLG was used as a substrate [82], and in another report the appearance of CdSO4 and CdO during the same treatment was reported to be linked to Na as well [86]. Kranz et al. reported in 2012 that if present during the growth of CdTe, Na has an impact on its grain size and it has been confirmed that Na

increases the acceptor concentration of CdTe, which in turn reduces the width of the space charge region [87]. In the same report the authors speculate that a Na-post deposition treatment might be beneficial due to its improvement of the electrical properties but it would not affect the morphology or elemental distribution opening thus the doors to introduction of Na in CdTe solar cells. Due to the fact that most groups intentionally block Na, and that the research on intentional introduction of Na is very recent, we will not present Na incorporation methods specifically for CdTe in this article. Nevertheless, most of the presented incorporation strategies would also apply to CdTe solar cells, if the substrate configuration is chosen.

3. Sodium incorporation strategies 3.1. Diffusion from the substrate 3.1.1. Soda-lime glass Glass is used in laboratories and in the photovoltaic industry as a substrate for thin film solar cells and modules because of several reasons: it is relatively cheap, it is available in a large volume, is rather stable over time, can have several shapes and sizes, it is resistant, and easy to handle. The commonly used glass is sodalime-glass (SLG), the normal window glass. The composition of a standard SLG is shown in Table 2. Its coefficient of thermal expansion (CTE) is 90  10  7 K  1, the strain point is around 514 °C, the annealing point is around 550 °C, the glass transition temperature is around 570 °C and a softening Littleton Point is located around 720 °C [88]. During the growth of the absorber layer, at temperatures between 500 °C and 600 °C, the glass goes through some of these transitions, the mobility of the ions is greatly increased, and Na ions can diffuse from the glass substrate into the absorber layer. As described above, this is beneficial for the electrical performance of the resulting solar cells. Most of the CIGSe solar cell world records of the past were established using SLG substrates, being the only exception the 20.4% record in 2013 on flexible polyamide by EMPA [62]. A large number of companies also use SLG, but some, for instance Solar Frontier [89], sometimes report the use of SLG with a diffusion barrier to block and/or control the amount of Na that diffuses. It should be noted that for temperatures lower than 500 °C [90] the amount of Na released is not enough whereas for temperatures higher than 600 °C, soda lime glass is not rigid enough to be flat and thus the deposition of CIGSe is not possible. Alkali elements, as Na and K, are constituents of soda lime glass and at temperatures close to the strain point they become mobile enough to diffuse into the Mo and into the CIGSe layers. The outdiffusion of Na from the glass proceeds by means of Na þ ions and requires a counter-diffusion of positively charged ions to maintain charge balance [91]. It has been theoretically proposed [91] and experimentally verified [92] that the mechanism responsible for Na diffusion is a H þ /Na þ ion exchange process and that the H þ is being stored in the Mo layer during air exposure whereas the source of Na is Na2O present in the glass [93]. The diffusion of Na from the SLG through the Mo proceeds via oxygenated Mo grain boundaries [92,94] and it has also been shown that oxygenated Mo layers provide CIGSe with higher Na concentrations [57,95]. It is worth mentioning that the amount of Na that diffuses is of Table 2 Typical composition of SLG (only the main constituents are given) [88]. Compound

SiO2

Na2O

CaO

MgO

K2O

Al2O3

SO3

Fe2O3

Composition (at%)

72.20

14.30

6.40

4.30

1.20

1.20

0.03

0.03

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

13

Fig. 2. Schematic representation of the model explaining the diffusion of Na from SLG [91–93]. Prior to the absorber growth the Mo boundaries are oxidized and the layer stores H þ ions. During the absorber growth, a diffusion of Na þ ions from the glass into the CIGSe is countered by a diffusion of H þ into the glass substrate.

sufficient quantity to be beneficial to the CIGSe and also it does not appear to affect the formation of the Cu(In,Ga)Se2 phase [96]. However, also here contradictory results have been presented in the literature, stating that SLG provides an optimal amount of Na [16], that it provides too much Na [97], and that it provides too little Na [34]. The reason for this inconsistency is likely associated with the complexity of the Na out-diffusion process, originating from the glass and going through the Mo layer [57]. A schematic representation of the diffusion process based on the H þ /Na þ ion exchange model is shown in Fig. 2. Although SLG is widely used, there are some limitations that should be addressed. By using SLG as a Na source, it is impossible to control the amount of Na that diffuses since it depends strongly on the oxidation state and general properties of the Mo layer as well as on the capability to exactly measure and reproduce the substrate temperature during the growth of CIGSe. Along with Na, it is well known that also K diffuses into CIGSe, and it is possible that other elements contained in the SLG might also diffuse into the CIGSe. In view of the unknown nature of the electronic effect of such elements, this is generally unwanted. In summary, during the growth of the absorber layer and depending on many variables, Na is released from SLG. Two limitations regarding industry fabrication are the lack of control over the amount and time of Na release and the price of SLG which is becoming a sizable fraction compared with the other parts of the module, in view of the recent cost reduction achievements [98]. 3.1.2. Specialty glasses Specialty glasses have been explored by several CIGSe research groups and the focus was given to high-temperature glasses [99,100], i.e. glasses with higher strain and softening temperatures. Because these points are at higher temperatures than SLG, not only can the CIGSe layer be grown at higher temperatures, but also the alkali-diffusion will proceed at higher temperatures. Other high-temperature glasses that are commercially available are usually alkali-free. The number of requisites that a glass substrate has to comply with to be compatible with CIGSe processing is large. A few examples of these requisites are: a coefficient of thermal expansion similar to that of SLG, about 90  10  7 K  1, a high strain point, chemical inertness during the harsh growth conditions of CIGSe (high temperature and Se atmosphere), only releasing alkalis, lightweight, cheap per unit of area, easily produced, etc. With all of these requirements, the glass industry has yet to fabricate a glass that is better and cheaper than SLG. It should be noted that the current CIGSe world record of 21.7%, was achieved on a specialty alkali-aluminosilicate glass [101] that delivers Na during the CIGSe growth process.

3.1.3. MoNa back contact layer An alternative approach for the incorporation of Na, is to incorporate Na directly into the Mo layer from where it diffuses during the growth of CIGSe. This approach was tested as early as 1997 by Granath et al. [55] where Na was ion-implanted into the Mo layer. Additional reports of a Na-doped Mo layer by Ho Yun et al. [102] and Cheol Kim et al. [103] did not mention how Na was incorporated into the Mo Layer, but they reported working solar cell devices. Mansfield et al. prepared a CIGSe solar cell with 16.6% efficiency where Na was incorporated via a Mo layer doped with Na (MoNa) [65]. The sputtering target consisted of sodium molybdate, Na2MoO4, powder mixed with Mo powder giving a target with a 3% [wt] and 10% [at] content of Na. They showed that this approach was worse in terms of cell performance when compared with SLG (17.8%) but better when compared with a NaF precursor layer (13.7%). Similar results were obtained by Wuerz et al. but by using flexible substrates and lower CIGSe growth temperatures [104]. It has been shown that the Na content in a 350 nm MoNa layer with 10% Na content was sufficient to achieve the full effects in CIGSe, but that the effective release of Na from a 5% [at] MoNa layer was insufficient [31]. The targets are commonly prepared by mixing Mo with a certain percentage of sodium molybdate, with concentrations of Na that can vary from 2% to 15% [at] [105]. In two publications, Blosch et al. [106,107] have stated that Na easily diffuses from internal interfaces of two MoNa layers possibly because Na moves through grain boundaries. In addition they also found that the microstructure of the MoNa layers strongly influence this diffusion, in accordance with several findings [57,108]. Therefore, Blosch et al. introduced multi-stacked layers of MoNa to allow more Na to be released and to arrive to the CIGSe layer. A similar approach of intermixing Mo and MoNa layers was presented by Mansfield et al. [65]. The biggest advantage of the use of a MoNa layer is that the only additional cost of implementing this strategy is the cost difference between a Mo sputtering target and a MoNa target. The biggest limitation so far is still being capable of releasing the correct amount of Na from the MoNa layer into the CIGSe layer. 3.2. External strategies for incorporation of Na 3.2.1. Incorporation of Na-compounds by means of thermal evaporation Na can be introduced before, during or after the fabrication of the CIGSe layers by adding an adequate Na-source to the processing equipment. For the introduction of Na during the growth of the absorber layer, elemental Na [90,109], Na2S [110], Na2Se [111] or NaF [20,110] have been reported. The choice of material to be evaporated can be influenced by its hazardousness, the source temperatures required, the handling

Hygroscopic Hygroscopic

and storage limitations, and the targeted parallel effects (other than the introduction of Na), see Table 3 and Fig. 3. Introducing Na before or during the evaporation has effects on the microstructure of the resulting CIGSe layers and on the elemental gradients through the depth of the material, as already mentioned. 3.2.2. NaF as a precursor One of the strategies with the best control over the amount of Na that is incorporated in the absorber layer is the deposition of a sodium precursor layer on Mo prior to the deposition of the absorber (Fig. 4). For both CIGSe and CZTSSe the most used precursor is NaF, but other materials, like for instance NaCl, have been studied [97,116]. NaF has the advantage that it evaporates congruently simplifying thus the deposition of the compound. The control over the amount of Na that is incorporated by the absorber layer is made by defining the thickness of the NaF layer. Most studies report that for CIGSe the optimum NaF thickness is between 10 to 20 nm [8,16,52,118–120]. With thickness values lower than 10 nm the most evident result is a low value of Voc of 2

10

KCl

KF [8,62,117] KCl

Vapor Pressures /mbars

885 821

806 767

Not hazardous Causes skin and serious eye irritation. Very toxic to aquatic life Toxic. Causes severe skin burns and eye damage. Very toxic to aquatic life Toxic if swallowed or if inhaled. May cause damage to organs through prolonged or repeated exposure. Very toxic to aquatic life with long lasting effects Toxic if swallowed or inhaled. Toxic in contact with skin Not hazardous NaBr NaI Na2S [110] Na2Se [11,55]

439 1077 865

0

Na [24,90] Spontaneous ignition if in contact with water. Skin burns and eye damage NaF [17,20,21,25,34,56,104,112–114] Toxic. Causes skin and serious eye irritation NaCl [115,116] Not hazardous

Handling and storage issues Temperature of saturated vapor at 1 Torr pressure (°C) Hazard, according to the material safety datasheets Material

Table 3 Suitable compounds for the incorporation of Na into Cu(In,Ga)Se2 layers by means of thermal evaporation.

Avoid contact with water Hygroscopic Hygroscopic Sacrificial layer for water liftoff process Hygroscopic Hygroscopic Highly hygroscopic Sulfur doping Highly hygroscopic

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

Parallel effects in CIGSe

14

10

Na

NaCl

In NaF

-2

10

-4

10

-6

10

-8

10

S Ga Se Cu

-10

10

0

200

400 600 800 1000 Temperature / °C

Fig. 3. Vapor pressures of S, Se, Na, KCL, NaCl, In, NaF, Ga, and Cu.

Fig. 4. Schematic representation of the effect of a NaF precursor layer deposited on Mo before the growth of the absorber layer.

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

the resulting solar cells indicating a deficiency in the Na content as explained earlier. For a NaF thickness higher than 20 nm contradictory reports appear: cell degradation; peeling off of the CIGSe layer during the chemical bath deposition of CdS; or no effects. This approach has been used in several CIGSe growth processes: the traditional three-stage [8,121], low temperature co-evaporation processes [118,119], in-line co-evaporation processes [16,122,123], rapid thermal annealing [124] and both on flat and flexible substrates like steel [125] and polyimides [126,127]. For CZTSSe the same thickness range has been reported [78,80] but no detailed studies about the effects of thick or thin layers are yet known. After its deposition and during the absorber growth, the NaF layer is stable during the heating of the substrates and up to the deposition of the metals close to a substrate temperature of 500 °C [123]. As with the case for SLG, it is likely that the Mo layer could act as a sink of Na during the deposition of CIGSe since after the growth (Fig. 4), high contents of Na are found in the Mo layer [16,63]. One of the questions that is still open with regards to this approach is the presence of fluorine and its influence on CIGSe. No studies investigating the influence of F on CIGSe or CZTSSe have been reported, but since this approach seems to provide as good cells as traditional SLG [63], one could assume that fluorine F has no evident negative effects on the device characteristics. Sodium sulfide and selenide, Na2S and Na2Se, have also been studied as precursor layers for Na incorporation [11,55], however NaF remains a more common choice thanks to its stability and non-toxicity. 3.2.3. Na incorporation after the absorber growth process Na can also be incorporated in a finished CIGSe absorber layer through a post-deposition treatment (PDT). During 2003 and 2005 the EMPA group developed a method based on the evaporation of NaF after the 3-stage growth of CIGSe [20,21,25,34,112,115]. The further development of the process paved the way for the achievement of efficiency values higher than 20% on polyimide substrates [62]. The PDT can also be used on steel [128] and on glass, furthermore it is also commonly used by other groups [129]. The PDT method consists of growing the absorber without Na and let it cool down to room temperature or to 100 °C. At that point, and still under vacuum, a layer of NaF between 20 nm to 40 nm (usually 30 nm) is evaporated and then the substrate is heated to 400 °C during 20 min to allow Na to diffuse (Fig. 5) [20,21,25,34,112,115]. The experimental description of the method does not state if the annealing step is done in the presence of excess of Se or not. This method has the advantage that it uses Knudsen-type evaporation cells which are known for their

Fig. 5. Schematic representation of a post-deposition treatment where NaF is evaporated after the absorber is fully grown.

15

stability and reproducibility and thus there is a high control over the amount of Na introduced. In addition, it allows the growth of CIGSe at lower temperatures, where the limiting temperature is now 400 °C, the temperature of the annealing step. The PDT does not change the microstructural properties of CIGSe [21] but increases the electrical performance of resulting devices as explained in Section 2.1. 3.2.4. K incorporation after the absorber growth process In 2013 Chirilă et al. modified the Na post deposition process to also include a post deposition of potassium by using evaporation of potassium fluoride [130]. They showed that by evaporating KF at a rate of 1 nm/min during 20 min right after the Na PDT with the CIGSe substrate at a temperature of 350 °C, positive effects were observed in the electrical performance of solar cells. It was also shown that K was reducing the overall amount of Na, allowed for a higher Cd diffusion into the CdS without changing the microstructure of the CIGSe films and this increased the cell efficiency. At the same time, Laemmle et al. also predicted that a combination of a Na-PDT together with a K-PDT could bring benefits for the performance of solar cells [117]. After this discovery, other groups started also to work with K-PDT and most of the groups observed significant increases in Voc and FF which ultimately drives efficiency upwards [61,62,101,117,130]. This technique has helped to improve the values of power conversion efficiency of CIGSe solar cells from 20.3% [62] to values of 21.7% in 2015 [101]. Even though it was known from before that K had beneficial effects and that it is present in vast amounts in CIGSe when SLG is used [8,113] it is still not understood why only a combination of Na and K-PDT allows [61] for higher electrical performance than the use of a single alkali element. At the time of writing this paper, this topic was still being studied and the authors expect the understanding of these effects to increase, as well as novel methods to include K in both CIGSe and CZTSSe. 3.3. Unconventional methods In addition to the above described methods, some unconventional methods have been reported, which are not widely used, and there is only limited information available. Nevertheless, they are briefly mentioned in this section. A method that has been the topic of several reports was developed by the National Institute of Advanced Industrial Science and Technology from Japan and relies on rf-sputtering of SLG onto a Na-free substrate prior to the deposition of the Mo. The advantage of this method is that the precursor is not placed on the Mo/ CIGS interface and that this way the amount of Na that is available during the growth of the absorber is controlled and deposited from a compound that is safe to handle [22,35,64]. By tuning the sputtering parameters, and optimizing the thickness of the SLG to 120 nm, a CIGSe solar cell on flexible zirconia with an efficiency around 17.0% was achieved. Although the electrical performance is quite high and the control of Na appears to be very precise, sputtering from a glass target demands a very effective cooling of the magnetron and either dc-pulsed or rf sources have to be used since the target it not conductive. Another method based on sputtering of a Na-compound to be used as a Na-precursor was developed by Showa Shell Industries (SSI). However, the exact constitution of the compound, the deposition conditions and the final influence in the absorber were not revealed [131–133]. The CIGSe manufacturer Stion developed a strategy for Na incorporation that does not require any additional processing step. During the formation of the CIGSe by rapid thermal annealing, the CIGSe precursors deposited on SLG (coated with a diffusion barrier and Mo) are submitted to high temperatures and to an atmosphere of H2Se. The substrates are placed in such a manner that

16

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

the back part of a glass substrate is facing the CIGSe surface of another substrate in very close proximity. The H2Se promotes the formation of Na2Se that evaporates from the back part of one substrate and incorporates into the CIGSe film on the next substrate [134].

4. Techniques for direct measurement of Na Most studies that investigate the effect of Na on thin film solar cells focus on the comparison of samples with and without Na, or with different concentrations of Na. The effect of the presence or absence of Na is then studied by the changing solar cell characteristics, or other electronic or structural properties. As the Na influences several properties at the same time, it is difficult to determine the factor that ultimately is crucial for an efficiency improvement. As described above and as an example, the open circuit voltage increases with increasing Na content in the CIGSe layer and at the same time also the grain size may be affected. While there are numerous studies using such methods to investigate the effect of Na, we will focus here on the description of methods that can directly determine the presence and concentration of Na in CIGSe layers. Although many techniques can be used for the determination of Cu, In, Ga and Se [135], the detection and quantification of the presence of Na is more troublesome since it is present in values lower than 1% [at] [1,8,34–36] and that can go down to 1015 atoms/cm3 [2]. A technique used frequently for compositional analysis in CIGSe materials is energy-dispersive X-ray analysis (EDX or EDS), performed in a scanning electron microscope. Cross sectional characterization can provide depth resolution. However, for the characterization of Na, this technique is not well suited, on one hand due to the limitation in the detection of Na (the Na K-line is located at 1.041 keV, which is very close to the Cu K-line 0.930 keV) and on the other hand due to its relatively low resolution in concentration, typically on the order of 1 at%. The same limitations apply to EDX performed in transmission electron microscopy and to X-ray fluorescence. The technique can be used under a certain of circumstances and only for qualitative comparison purposes [136] and therefore, these techniques will not be discussed here. 4.1. Glow discharge optical-emission spectroscopy (GD-OES) In GD-OES the elemental composition of a sample is determined by means of the characteristic spectral emission lines of ionized and excited atoms, removed from the sample by sputtering. Depth resolution is achieved by slowly sputtering and following the signals over time. Due to a low concentration of the removed elements in the sputter gas (typically Ar), the concentration of a wide range of elements can be determined independently. The optical emission is detected with a spectrometer, requiring sufficient spectral resolution to separate emission from different elements. The intensity Iik of the emission line (at wavelength k) depends linearly on the concentration ci of element i in the sample, on the sputtering rate q (total sputtered mass of the sample per time), and on the emission yield Rik:Iik ¼ciqRik. The emission yield is typically considered independent of the pressure during sputtering and of the matrix. Quantitative analysis requires calibration using standard reference materials to establish the relationship between the emission intensity and the elemental concentration of an element in the sample. For the determination of Na, this is the biggest limitation of this technique; despite the possibility of determining the depth distribution of Na quite well [137], its absolute concentration will be hard to extract.

4.2. Secondary ion mass spectrometry (SIMS) In SIMS the elemental composition as a function of depth is also determined while sputtering the sample. The surface of the sample is sputtered, i.e. eroded, using a primary ion beam. The sputtered ions are analyzed by measuring their kinetic energy using an electric field and an aperture, and their mass is determined through time-of-flight, magnetic sector, or quadrupole mass spectrometry. For quantitative measurements, SIMS needs reference samples [138]; however, by using Cs þ as the primary ion and by looking at the secondary cluster ions of matrix-Cs þ , the measurements can lead to semi-quantitative results with a minor number of reference samples [138]. If in addition to Na, K is to be measured, then O2 þ needs to be used as primary ions. Note that the ionization process is connected to the chemistry of the host matrix and the chemistry of the element of interest in this matrix; thus the primary ion has to be selected with care. The sensitivity can reach parts-per-trillion resolution with lateral and depth resolution making this technique one of the most useful ones for the evaluation of the depth profile of alkalis and of its concentration [139]. 4.3. X-ray photoelectron spectroscopy (XPS) and auger electron spectroscopy (AES) Upon irradiation of a sample by high energy electrons or photons, electrons from the inner shells can be excited. In XPS, excitation of the sample uses X-ray photons and the energy spectrum of electrons leaving the sample can be correlated to the discrete binding energies of the electrons in the sample. The element-specific peaks in the energy spectrum are superimposed on a continuous background due to loss processes of the electrons before leaving the sample. XPS is a very surface sensitive technique due to the short mean free path of electrons in a solid [140]. To obtain depth profiles, XPS is combined with sputtering. Alternating cycles of sputtering and XPS analyze the composition in the newly exposed surface and thus provide depth information. AES is a related techniques and typically is performed using an electron gun for the excitation of the sample. However, in AES, the hole on an inner shell is filled with an electron from an outer shell. The energy released by this process is transferred to another electron, which leaves the sample, and its kinetic energy is analyzed. Again, the kinetic energy spectrum contains information of the binding energy, which is elemental specific and therefore provides information about the species and chemical environment of the analyzed element. As for XPS, AES is also surface sensitive and depth profiling can be realized in combination with sputtering cycles. 4.4. Atom probe tomography (ATP) In ATP the three dimensional (3D) distribution of elements is determined with sub-nanometer resolution [141]. For the analysis, the sample is prepared as a needle-shaped tip with a circular cross section with a radius of 10–100 nm. The specimen is typically prepared using focused ion beam (FIB) and can be characterized additionally by transmission electron microscopy (TEM) to identify and position a specific region of interest close to the end of the tip. The ATP experiment itself consists in removing atoms from the tip, and determining their species and position by time-of-flight measurement and a position sensitive detector, respectively. For the analysis of semiconductor materials, the atoms are evaporated one by one from the tip by picosecond laser pulsing. The evaporated ions are accelerated in an electric field and by measuring the time of flight to the detector, their mass to charge ratio is determined to identify the ion. With the (x,y) position of the ion impact

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

on the detector and the order of impact, the original position of the atoms on the tip can be reconstructed in 3D. Regarding the detection of Na in chalcogenide solar cell materials ATP has been mainly used to characterize the segregation of Na at grain boundaries [40–43].

[8]

[9]

5. Outlook Incorporation of Na is an established subject in the development of chalcogenide-based solar cells. In this review we briefly discussed the effects of Na, which can be negative (CdTe) or positive (CIGS and CZTSSe). Despite some specific improvements, the fundamental physical mechanism responsible for these effects is still not fully understood. The successful incorporation of Na can be achieved by several methods underlining the flexibility of the chalcogenides and their suitability for industrial applications. The reviewed methods include diffusion from the substrate (SLG, MoNa layer), incorporation prior, during or after the growth of the absorber layer, and some additional, more unconventional methods. We note that the discussed Na incorporation strategies can also be of importance for other chalcogenide materials for different applications, as for example thermoelectrics [142–144], energy storage devices [145] and other optoelectronic devices [146]. In terms of other dopants that are currently introduced in the absorber layer, the most important ones are bismuth [147], antimony [147] and most obviously potassium. According to the latest studies, potassium has a significant importance in passivation of the front interface [62,101,117,130], but only when incorporated after the growth of CIGSe and together with Na [148]. Having this in mind, future studies should address the best strategies to incorporate K, its effects on the solar cell properties and potential cross-effects with Na.

Acknowledgments P.M.P. Salomé acknowledges financial support from EU through the FP7 Marie Curie IEF 2012 Action No. 327367 and he also acknowledges fruitful discussions on the topic that he had with members from the Ångström Solar Center and from the Laboratory for Nanostructured Solar Cells (LaNaSC) at INL.

[10]

[11] [12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

References [1] J. Hedström, H. Ohlsen, M. Bodegård, A. Kylner, L. Stolt, ZnO/CdS/Cu(In,Ga)Se2 thin film solar cells with improved performance, in: Proceedings of the 23rd IEEE Photovoltaic Specialists Conference, 1993, pp. 364–371. [2] J. Holz, F. Karg, H. von Philipsborn, The effect of substrate impurities on the electronic conductivity in CIS thin films, Presented at the 12th European Photovoltaic Solar Energy Conference, Amsterdam, Netherlands, 1994, p.1592. [3] M. Bodegård, L. Stolt, J. Hedström, The influence of sodium on the grain structure of CuInSe2 films for photovoltaic application, Presented at the 12th European Photovoltaic Solar Energy Conference, Amsterdam, UK, 1994, p.1743. [4] S. Zweigart, D. Schmid, J. Kessler, H. Dittrich, H.W. Schock, Studies of the growth mechanism of polycrystalline CuInSe2 thin films prepared by a sequential process, J. Cryst. Growth 146 (1995) 233–238. [5] T. Walter, D. Hariskos, R. Heberholz, V. Nadenau, R. Schäffler, H.W. Schock, On the role of oxidation for the performance of Cu(In,Ga)(S,Se) 2 polycrystalline thin film solar cells, in: Proceedings of the 13th European Photovoltaic solar energy conference, 23–27 October 1995, Nice France, p. 1999. [6] C. Heske, R. Fink, E. Umbach, W. Riedl, F. Karg, Na-induced effects on the electronic structure and compositionof Cu(In,Ga)Se2 thin-film surfaces, Appl. Phys. Lett. 68 (24) (1996). [7] M. Bodegård, J. Hedström, K. Granath, A. Rockett, L. Stolt, Na precursors for coevaporated Cu(In,Ga)Se2 photovoltaic films, in: Proceedings of the 13th

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32]

17

European Photovoltaic solar energy conference, 23–27 October, 1995, Nice France, p. 2080. M.A. Contreras, B. Egaas, P. Dippo, J. Webb, J. Granata, K. Ramanathan, S. Asher, A. Swartzlander, R. Noufi, On the role of Na and Modifications to Cu(In,Ga)Se2 absorber materials using Thin-MF (M ¼Na, K, Cs) precursor layers, in Proceedings of the IEEE 26th Photovoltaic Specialists Conference, Sep. 30–Oct. 3, 1997, pp. 359–362. T. Nakada, D.I.H. Ohbo, A. Kunioka, Effects of sodium on Cu(In,Ga)Se2-based thin film and solar cells, Jpn. J. Appl. Phys. 36 (1997) 732–737. U. Rau, M. Schmitt, F. Engelhardt, O. Seifert, W. Parisi, J. Riedl, Rimmasch, F. Karg, Impact Of Na And S Incorporation On The Electronic Transport Mechanisms Of Cu(In, Ga)Se2 Solar Cells, Solid State Commun. 107 (2) (1998) 59–63. D.J. Schroeder, A.A. Rockett, Electronic effects of sodium in epitaxial CuIn1  x Gax Se2, J. Appl. Phys. 82 (1997) 4982. I. Dirnstorfer, D.M. Hofmann, D. Meister, B.K. Meyer, W. Riedl, F. Karg, Postgrowth thermal treatment of Cu(In,Ga)Se2: characterization of doping levels in In-rich thin films, J. Appl. Phys. 85 (1999) 1423. I. Dirnstorfer, Mt Wagner, D.M. Hofmann, M.D. Lampert, F. Karg, B.K. Meyer, Characterization of CuIn(Ga)Se2 Thin Films, Phys. Status Solidi A 168 (1) (1998) 163–175. M. Igalson, A. Kubiaczyk, P. Zabierowski, M. Bodegård, K. Granath, Electrical characterization of ZnO/CdS/Cu(In,Ga)Se2 devices with controlled sodium content, Thin Solid Films 387 (2001) 225–227. K. Timmo, M. Altosaar, J. Raudoja, E. Mellikov, T. Varema, M. Danilson, M. Grossberg, The effect of sodium doping to CuInSe2 monograin powder properties, Thin Solid Films 515 (2007) 5887–5890. P.M.P. Salomé, A. Hultqvist, V. Fjällström, M. Edoff, B. Aitken, K. Vaidyanathan, K. Zhang, K. Fuller, C. Kosik Williams, Cu(In,Ga)Se2 solar cells with varying Na content prepared on nominally alkali-free glass substrates, IEEE J. Photovolt. 3 (2) (2013). S. Puttnins, S. Levcenco, K. Schwarzburg, G. Benndorf, F. Daume, A. Rahm, A. Braun, M. Grundmann, T. Unold, Effect of sodium on material and device quality in low temperature deposited Cu(In,Ga)Se2, Sol. Energy Mater. Sol. Cells 119 (2013) 281–286. F. Hergert, S. Jost, R. Hock, M. Purwins, J. Palm, Formation reactions of chalcopyrite compounds and the role of sodium doping, Thin Solid Films 515 (2007) 5843–5847. Jessica Eid, Haifan Liang, Issam Gereige, Sang Lee, Jeroen Van Duren, Combinatorial study of NaF addition in CIGSe films for high efficiency solar cells, Prog. Photovolt.: Res. Appl. 23 (3) (2015) 269–280. D. Rudmann, G. Bilger, M. Kaelin, F.J. Haug, H. Zogg, A.N. Tiwari, Effects of NaF coevaporation on structural properties of Cu(In,Ga)Se2 thin films, Thin Solid Films 431–432 (2003) 37–40. D. Rudmann, A.F. da Cunha, M. Kaelin, F. Kurdesau, H. Zogg, A.N. Tiwari, G. Bilger, Efficiency enhancement of Cu(In,Ga)Se2 solar cells due to postdeposition Na incorporation, Appl. Phys. Lett. 84 (2004) 1129–1131. Shogo Ishizuka, Akimasa Yamada, Koji Matsubara, Paul Fons, Keiichiro Sakurai, Shigeru Niki, Development of high-efficiency flexible Cu(In,Ga)Se2 solar cells: a study of alkali doping effects on CIS, CIGS, and CGS using alkali-silicate glass thin layers, Curr. Appl. Phys. 10 (Suppl. 2) (2010) S154–S156. V. Izquierdo-Roca, R. Caballero, X. Fontané, C.A. Kaufmann, J. Álvarez-García, L. Calvo-Barrio, E. Saucedo, A. Pérez-Rodríguez, J.R. Morante, c, H.W. Schock, Raman scattering analysis of Cu-poor Cu(In,Ga)Se2 cells fabricated on polyimide substrates: effect of Na content on microstructure and phase structure, Thin Solid Films 519 (2011) 7300–7303. M. Lammer, R. Kniese, M. Powalla, In-line deposited Cu(In,Ga)Se2 solar cells: influence of deposition temperature and Na co-evaporation on carrier collection, Thin Solid Films 451–452 (2004) 175–178. D. Rudmann, M. Kaelin, F.-J. Haug, F. Kurdesau, H. Zogg, A.N. Tiwari, Impact of Na on structural properties and interdiffusion of CuInSe2 and CuGaSe2 thin films, in: Proceedings of the 3rd World Conference on Photovoltaic Energy Conversion May 11–18, 2003 Osaka, Japan, pp. 376–379. Miguel A. Contreras, Manuel J. Romero, R. Noufi, Characterization of Cu(In,Ga)Se2 materials used in record performance solar cells, Thin Solid Films 511–512 (2006) 51–54. M. Ruckh, D. Schmid, M. Kaiser, R. Schaffler, T. Walter, H.W. Schock, Influence of substrates on the electrical properties of Cu(In,Ga)Se2 thin films, Sol. Energy Mater. Sol. Cells 41–42 (1996) 335–343. Dae-Hyung Cho, Kyu-Seok Lee, Ju-Hee Chung, Yong-Duck Chung, JuHee Kim, Jeha Kim, Park and Jeha Kim, Electronic effect of Na on Cu(In,Ga)Se2 solar cells, Appl. Phys. Lett. 101 (2012) 023901. L.E. Oikkonen, M.G. Ganchenkova, A.P. Seitsonen, R.M. Nieminen, Effect of sodium incorporation into CuInSe2 from first principles, J. Appl. Phys. 114 (2013) 083503. Elif Selin Mungan, Xufeng Wang, Modeling the effects of Na incorporation on CIGS solar cells, Muhammad Ashraful, IEEE J. Photovolt. 3 (1) (2013) 451. P.M.P. Salomé, V. Fjällström, A. Hultqvist, M. Edoff, Na doping of CIGS solar cells using low sodium doped Mo layer, IEEE J. Photovolt. 3 (1) (2013) 509–513. Raquel Caballero, Christian A. Kaufmann, Tobias Eisenbarth, Thomas Unold, Reiner Klenk, Hans-Werner Schock, High efficiency low temperature grown Cu(In,Ga)Se2 thin films solar cells on flexible substrates using NaF precursor layers, Prog. Photovolt.: Res. Appl. 19 (5) (2011) 547–551.

18

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

[33] Bart Vermang, Viktor Fjällström, Jonas Pettersson, Pedro Salomé, Marika Edoff, Development of rear surface passivated Cu(In,Ga)Se2, thin film solar cells with nano-sized local rear point contacts, Sol. Energy Mater. Sol. Cells 117 (2013) 505–511. [34] D. Rudmann, D. Brémaud, A.F. da Cunha, G. Bilger, A. Strohm, M. Kaelin, H. Zogg, A.N. Tiwari, Sodium incorporation strategies for CIGS growth at different temperatures, Thin Solid Films 480–481 (2005) 55–60. [35] Shogo Ishizuka, Akimasa Yamada, Muhammad Monirul Islam, Hajime Shibata, Paul Fons, Takeaki Sakurai, Katsuhiro Akimoto, Shigeru Niki, Na-induced variations in the structural, optical, and electrical properties of Cu(In,Ga)Se2 thin films, J. Appl. Phys. 106 (2009) 034908. [36] A. Rockett, J.S. Britt, T. Gillespie, C. Marshall, M.M. Al Jassim, F. Hasoon, R. Matson, B. Basol, Na in selenized Cu(In,Ga)Se2 on Na-containing and Nafree glasses: distribution, grain structure, and device performances, Thin Solid Films 372 (2000) 212–217. [37] R. Caballero, C.A. Kaufmann, V. Efimova, T. Rissom, V. Hoffmann, H. W. Schock, Investigation of Cu(In,Ga)Se2 thin-film formation during the multi-stage co-evaporation process, Prog. Photovolt.: Res. Appl. 21 (1) (2013) 30–46. [38] D. Braunger, D. Hariskos, G. Bilger, U. Rau, H.W. Schock, Influence of sodium on the growth of polycrystalline Cu(In,Ga)Se2 thin films, Thin Solid Films 361–362 (2000) 161–166. [39] Mirjam Theelen, Vincent Hans, Nicolas Barreau, Henk Steijvers, Zeger Vroon, Miro Zeman, The impact of alkali elements on the degradation of CIGS solar cells, Prog. Photovoltaics Res. Appl. 23 (5) (2015) 537–545. [40] O. Cojocaru-Mirédina, P. Choi, R. Wuerz, D. Raabe, Atomic-scale distribution of impurities in CuInSe2-based thin-film solar cells, Ultramicroscopy 111 (6) (2011) 552–556. [41] O. Cojocaru-Miredin, P.P. Choi, D. Abou-Ras, S.S. Schmidt, R. Caballero, D. Raabe, Characterization of Grain Boundaries in Cu(In,Ga)Se2 films using Atom Probe Tomography, IEEE J. Photovolt. 1 (2) (2011) 207–212. [42] E. Cadel, N. Barreau, J. Kessler, P. Pareige, Atom probe study of sodium distribution in polycrystalline Cu(In,Ga)Se2 thin film, Acta Mater. 58 (2010) 2634–2637. [43] Pyuck-Pa Choi, Oana Cojocaru-Mirédin, Roland Wuerz, Dierk Dierk, Comparative atom probe study of Cu(In,Ga)Se2 thin-film solar cells deposited on soda-lime glass and mild steel substrates, J. Appl. Phys. 110 (2011) 124513. [44] S.H. Wei, S.B. Zhang, A. Zunger, Effects of Na on the electrical and structural properties of CuInSe2, J. Appl. Phys. 85 (1999) 7214. [45] W.J. Yin, Y. Wu, R. Noufi, M. Al-Jassim, Y. Yan, Defect segregation at grain boundary and its impact on photovoltaic performance of CuInSe2, Appl. Phys. Lett. 102 (2013) 193905. [46] L. Kronik, D. Cahen, H.W. Schock, Effects of Sodium on Polycrystalline Cu(In,Ga)Se2 and Its Solar Cell Performance, Adv. Mater. 10 (1998) 31–36. [47] V. Fjällström, P.M.P. Salomé, A. Hultqvist, M. Edoff, T. Jarmar, B.G. Aitken, K. Zhang, K. Fuller, C. Kosik William, Potential Induced degradation of CuIn1  xGaxSe2 thin film solar cells, IEEE J. Photovolt. 3 (3) (2013) 1090–1094. [48] V. Fjällström, P. Szaniawski, B. Vermang, P.M.P. Salomé, F. Rostvall, U. Zimmermann, M. Edoff, Recovery after potential induced degradation of Cu(In,Ga)Se2 solar cells with CdS and Zn(O,S) buffer layers, IEEE J. Photovolt. 5 (2) (2015) 664–669. [49] Ju-Heon Yoon, Jun-Ho Kim, Won Mok Kim, Jong-Keuk Park, Young-Joon Baik, Tae-Yeon Seong, Jeung-hyun Jeong, Electrical properties of CIGS/Mo junctions as a function of MoSe2 orientation and Na doping, Prog. Photovolt.: Res. Appl. 22 (1) (2014) 90–96. [50] T. Wada, N. Kohara, S. Nishiwaki, T. Negami, Characterization of the Cu(In,Ga)Se2/Mo interface in CIGS solar cells, Thin Solid Films 387 (2001) 118–122. [51] N. Barreau, J.C. Bernède, Low-temperature preparation of MoS2 thin films on glass substrate with NaF additive, Thin Solid Films 403–404 (2002) 505–509. [52] R. Caballero, M. Nichterwitz, A. Steigert, A. Eicke, I. Lauermann, H.W. Schock, C.A. Kaufmann, Impact of Na on MoSe2 formation at the CIGSe/Mo interface in thin-film solar cells on polyimide foil at low process temperatures, Acta Mater. 63 (2014) 54–62. [53] F. Couzinie-Devy, E. Cadel, N. Barreau, L. Arzel, P. Pareige, Atom probe study of Cu-poor to Cu-rich transition during Cu(In,Ga)Se2 growth, Appl. Phys. Lett. 99 (2011) 232108. [54] C. Heske, G. Richter, Zhonghui Chen, R. Fink, E. Umbach, Influence of Na and H2O on the surface properties of Cu(In,Ga)Se2 thin films, J. Appl. Phys. 82 (1997) 2411. [55] K. Granath, L. Stolt, M. Bodegård, A. Rockett, D.J. Schroeder, Sodium in sputtered Mo back contacts for Cu(In,Ga)Se2 devices: incorporation, diffusion, and relationship to oxygen, in: Proceedings of the 14th European Photovoltaic Solar energy conference, Barcelona, Spain, 30 June–4 July, 1997, pp 1278–1282. [56] D. Abou-Ras, G. Kostorz, D. Bremaud, M. K7lin, F.V. Kurdesau, A.N. Tiwari, M. Dfbeli, Formation and characterisation of MoSe2 for Cu(In,Ga)Se2 based solar cells, Thin Solid Films 480–481 (2005) 433–438. [57] Robert V. Forest, Erten Eser, Brian E. McCandless, Robert W. Birkmire, Jingguang G. Chen, Understanding the role of oxygen in the segregation of sodium at the surface of molybdenum coated soda-lime glass, AIChE J. 60 (6) (2014) 2365. [58] A. Rockett, The effect of Na in polycrystalline and epitaxial single-crystal CuIn1  xGaxSe2, Thin Solid Films 480–481 (2005) 2–7.

[59] Susanne Siebentritt, Malgorzata Igalson, Clas Persson and Stephan Lany, the electronic structure of chalcopyrites—bands, point defects and grain boundaries, Prog. Photovolt.: Res. Appl. 18 (6) (2010) 390–410. [60] Roland Scheer, Hans-Werner Schock, Chalcogenide Photovoltaics: Physics, Technologies, and Thin Film Devices, Wiley-YCH, Weinheim, Germany, 2011, ISBN: 978-3-527-31459-1. [61] Philip Jackson, Dimitrios Hariskos, Roland Wuerz, Wiltraud Wischmann, Michael Powalla, Compositional investigation of potassium doped Cu(In,Ga)Se2 solar cells with efficiencies up to 20.8%, Phys. Status Solidi (RRL) – Rapid Res. Lett. 8 (3) (2014) 219–222. [62] Adrian Chirilă, Patrick Reinhard, Fabian Pianezzi, Patrick Bloesch, Alexander R. Uhl, Carolin Fella, Lukas Kranz, Debora Keller, Christina Gretener, Harald Hagendorfer, Dominik Jaeger, Rolf Erni, Shiro Nishiwaki, Stephan Buecheler, Ayodhya N. Tiwari, Potassium-induced surface modification of Cu(In,Ga)Se2 thin films for high-efficiency solar cells, Nat. Mater. 12 (2013) 1107–1111. [63] P.M.P. Salomé, A. Hultqvist, V. Fjällström, M. Edoff, B.G. Aitken, K. Zhang, K. Fuller, C. Kosik Williams, Incorporation of Na in Cu(In,Ga)Se2 thin film solar cells: a statistical comparison between Na from soda lime glass and from a precursor layer of NaF, IEEE J. Photovolt. 4 (6) (2014) 1659–1664. [64] Shogo Ishizuka, Akimasa Yamada, Paul Fons, Shigeru Niki, Flexible Cu(In,Ga)Se2 solar cells fabricated using alkali-silicate glass thin layers as an alkali source material, J. Renew. Sustain. Energy 1 (2009) 013102. [65] L.M. Mansfield, I.L. Repins, S. Glynn, M.D. Carducci, D.M. Honecker, J.W. Pankow, M.R. Young, C. DeHart, R. Sundaramoorthy, C.L. Beall, B. To, Sodium-doped molybdenum targets for controllable sodium incorporation in CIGS solar cells, in: Proceedings of the 37th IEEE Photovoltaic Specialists Conference, Jun 2011, pp. 003636–003641. [66] Hironori Katagiri, Nobuyuki Sasaguchi, Shima Hando, Suguro Hoshino, Jiro Ohashi, Takaharu Yokota, Preparation and evaluation of Cu2ZnSnS4 thin films by sulfurization of E-B evaporated precursors, Sol. Energy Mater. Sol. Cells 49 (1997) 407–414. [67] H. Katagiri, K. Saitoh, T. Washio, H. Shinohara, T. Kurumadani, S. Miyajima, Development of thin film solar cell based on Cu2ZnSnS4 thin films, Sol. Energy Mater. Sol. Cells 65 (1–4) (2001) 141–148. [68] T.M. Friedlmeier, H. Dittrich, H.W. Schock, The 11th Conference on Ternary and Multinary Compounds, ICTMC-11, Salford, 8–12 September, 1997, p. 345. [69] W. Wang, M.T. Winkler, O. Gunawan, T. Gokmen, T.K. Todorov, Y. Zhu, D. B. Mitzi, Device Characteristics of CZTSSe Thin-Film Solar Cells with 12.6% Efficiency, Adv. Energy Mater. 4 (2014) 1301465. [70] J.P. Leitão, N.M. Santos, P.A. Fernandes, P.M.P. Salomé, A.F. da Cunha, J.C. Gonzalez, G.M. Ribeiro, F.M. Matinaga, Photoluminescence and electrical study of fluctuating potentials in Cu2ZnSnS4-based thin films, Phys. Rev. B 84 (2011) 024120. [71] J.C. González, G.M. Ribeiro, E.R. Viana, P.A. Fernandes, P.M.P. Salomé, K. Gutiérrez Z.-B., A.A. Garcia, F.M. Matinaga, J.P. Leitão, A.F. da Cunha, Hopping conduction and persistent photoconductivity in Cu2ZnSnS4, Thin Films J. Phys. D: Appl. Phys. 46 (2013) 155107. [72] V. Kosyak, N.B. Mortazavi Amiri, A.V. Postnikov, M.A. Scarpulla, Model of native point defect equilibrium in Cu2ZnSnS4 and application to one-zone annealing, J. Appl. Phys. 114 (2013) 124501. [73] W.M. Hlaing Oo, J.L. Johnson, A. Bhatia, E.A. Lund, M.M. Nowell, M.A. Scarpulla, Grain size and texture of Cu2ZnSnS4 thin films synthesized by co-sputtering binary sulfides and annealing: effects of processing conditions and sodium, J. Electron. Mater. 40 (11) (2011) 2214–2221. [74] Patrice Bras, Jan Sterne, Charlotte Platzer-Björkman, Influence of hydrogen sulfide annealing on copper–zinc–tin–sulfide solar cells sputtered from a quaternary compound target, Thin Solid Films 582 (2015) 233–238. [75] Tejas Prabhakar, Nagaraju Jampana, Effect of sodium diffusion on the structural and electrical properties of Cu2ZnSnS4 thin films, Sol. Energy Mater. Sol. Cells 95 (2011) 1001–1004. [76] Jian V. Li, Darius Kuciauskas, Matthew R. Young, Ingrid L. Repins, Effects of sodium incorporation in Co-evaporated Cu2ZnSnSe4 thin-film solar cells, Appl. Phys. Lett. 102 (2013) 163905. [77] Akira Nagaoka, Hideto Miyake, Tomoyasu Taniyama, Koichi Kakimoto, Yoshitaro Nose, Michael A. Scarpulla, Kenji Yoshino, Effects of sodium on electrical properties in Cu2ZnSnS4 single crystal, Appl. Phys. Lett. 104 (2014) 152101. [78] Ingrid Repins, Nirav Vora, Carolyn Beall, Su-Huai Wei, Yanfa Yan, Manuel Romero, Glenn Teeter, Hui Du, Bobby To, Matt Young, Rommel Noufi, Kesterites and Chalcopyrites: A Comparison of Close Cousins, NREL/CP-520051286, May 2011. [79] Ingrid Repins, Carolyn Beall, Nirav Vora, Clay DeHart, Darius Kuciauskas, Pat Dippoa, Bobby To, Jonathan Mann, Wan-Ching Hsu, Alan Goodrich, Rommel Noufi, Co-evaporated Cu2ZnSnSe4 films and devices, Sol. Energy Mater. Sol. Cells 101 (2012) 154–159. [80] Talia Gershon, Byungha Shin, Nestor Bojarczuk, Marinus Hopstaken, David B. Mitzi, Supratik Guha, The Role of Sodium as a Surfactant and Suppressor of Non-Radiative Recombination at Internal Surfaces in Cu2ZnSnS4, Adv. Energy Mater. 5 (2015) 1400849. [81] J. Perrenoud, L. Kranz, S. Buecheler, F. Pianezzi, A.N. Tiwari, The use of aluminium doped ZnO as transparent conductive oxide for CdS/CdTe solar cells, Thin Solid Films 519 (2011) 7444–7448. [82] N. Romeo, A. Bosio, R. Tedeschi, V. Canevari, Growth of polycrystalline CdS and CdTe thin layers for high efficiency thin film solar cells, Mater.s Chem. Phys. 66 (2000) 201–206.

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

[83] Xuanzhi Wu, High-efficiency polycrystalline CdTe thin-film solar cells, Sol. Energy 77 (2004) 803–814. [84] E. Molva, J.P. Chamonal, J.L. Pautrat, Shallow acceptors in cadmium telluride, Phys. Status Solidi B 109 (1982) 635–644. [85] James Sites, Jun Pan, Strategies to increase CdTe solar-cell voltage, Thin Solid Films 515 (2007) 6099–6102. [86] N. Romeo, A. Bosio, V. Canevari, M. Terheggen, L. Vaillant Roca, Comparison of different conducting oxides as substrates for CdS/CdTe thin film solar cells, Thin Solid Films 431–432 (2003) 364–368. [87] L. Kranz, J. Perrenoud, F. Pianezzi, C. Gretener, P. Rossbach, S. Buecheler, A.N. Tiwari, Effect of sodium on recrystallization and photovoltaic properties of CdTe solar cells, Sol. Energy Mater. Sol. Cells 105 (2012) 213–219. [88] MenZel-Glaser. Retrieved from: 〈http://www.menzel.de/ on 2014/07/16〉. [89] Shigeru Niki, Miguel Contreras, Ingrid Repins, Michael Powalla, Katsumi Kushiya, Shogo Ishizuka, Koji Matsubara, CIGS absorbers and processes, Prog. Photovolt.: Res. Appl. 18 (6) (2010) 453–466. [90] M. Lammer, U. Klemm, M. Powalla, Sodium co-evaporation for low temperature Cu(In,Ga)Se2 deposition, Thin Solid Films 387 (2001) 33–36. [91] N. Janke, O. Grassme, R. Wwissmann, Alkali ion migration control from flat glass substrates, Glass Sci. Technol. 73 (5) (2000) 143–155. [92] G. Hanna, S. Schleussner, G. Bilder, H.W. Schock, U. Rau, J. H. Werner, Na diffusion in the Cu(In,Ga)Se2/Mo/Glass system, in: Proceedings of the 3rd World Conference on Photovoltaic Energy Conversion, May 11–18, 2003, Osaka Japan, pp 368–371. [93] F.V. Natrup, H. Bracht, S. Murugavel, B. Roling, Cation diffusion and ionic conductivity in soda-lime silicate glasses, Phys. Chem. Chem. Phys. 7 (2005) 2279–2286. [94] M. Bodegård, K. Granath, L. Stolt, The behaviour of Na implanted into Mo thin films during annealing, Sol. Energy Mater. Sol. Cells 58 (1999) 199. [95] P.M.P. Salomé, V. Fjällström, A. Hultqvist, P. Szaniawski, U. Zimmermann, M. Edoff, The effect of Mo back contact ageing on CIGS thin film solar cells, Prog. Photovolt.: Res. Appl. 22 (1) (2014) 83–89. [96] Tokio Nakada, Hiroki Ohbo, Masakazu Fukuda, Akio Kunioka, Improved compositional flexibility of Cu(In,Ga)Se2-based thin film solar cells by sodium control technique, Sol. Energy Mater. Sol. Cells 49 (1997) 261–267. [97] Carolin M. Sutter-Fella, Josua A. Stückelberger, Harald Hagendorfer, Fabio La Mattina, Lukas Kranz, Shiro Nishiwaki, Alexander R. Uhl, Yaroslav E. Romanyuk, Ayodhya N. Tiwari, Sodium assisted sintering of chalcogenides and its application to solution processed Cu2ZnSn(S,Se)4 thin film solar cells, Chem. Mater. 26 (3) (2014) 1420–1425. [98] Keith Burrows, Vasilis Fthenakis, Glass needs for a growing photovoltaics industry, Sol. Energy Mater. Sol. Cells 132 (2015) 455–459. [99] P.M.P. Salomé, A. Hultqvist, V. Fjällström, B. Vermang, M. Edoff, B. Aitken, K. Zhang, K. Fuller, C. Kosik Williams, The effect of high growth temperature on Cu(In,Ga)Se2 thin film solar cells, Sol. Energy Mater. Sol. Cells 123 (2014) 166–170. [100] J. Haarstrich, H. Metzner, M. Oertel, C. Ronning, T. Rissom, C.A. Kaufmann, T. Unold, H.W. Schock, J. Windeln, W. Mannstadt, E. Rudigier-Voigt, Increased homogeneity and open-circuit voltage of Cu(In,Ga)Se2 solar cells due to higher deposition temperature, Sol. Energy Mater. Sol. Cells 95 (2011) 1028–1030. [101] Philip Jackson, Dimitrios Hariskos, Roland Wuerz, Oliver Kiowski, Andreas Bauer, Theresa Magorian Friedlmeier, Michael Powalla, Properties of Cu(In,Ga)Se2 solar cells with new record efficiencies up to 21.7%, Phys. Status Solidi (RRL) – Rapid Res. Lett. 9 (1) (2015) 28–31. [102] Jae Ho Yun, Ki. Hwan Kim, Min Sik Kim, Byung Tae Ahn, Se. Jin Ahn, Jeong Chul Lee, Kyung Hoon Yoon, Fabrication of CIGS solar cells with a Na-doped Mo layer on a Na-free substrate, Thin Solid Films 515 (2007) 5876–5879. [103] Sung Cheol Kim, Hyeonwook Park, Eun-Woo Lee, Jae Sung Han, Sung Ho Lee, Chan-Wook Jeon, Dongwook Jung, Jinju Jeong, Woo Kyoung Kim, Role of Na in reaction pathways and kinetics of CuInSe2 formation from stacked binary precursors, Thin Solid Films 519 (2011) 7250–7253. [104] R. Wuerz, A. Eicke, F. Kessler, P. Rogin, O. Yazdani-Assl, Alternative sodium sources for Cu(In,Ga)Se2 thin-film solar cells on flexible substrates, Thin Solid Films 519 (2011) 7268–7271. [105] http://dx.doi.org/http://www.plansee.com/en/Products-Thin-film-materialsSputtering-targets-and-arc-cathodes-Molybdenum-sodium-1343.htm (retrived in June 2015). [106] Patrick Blösch, Shiro Nishiwaki, Adrian Chirilă, Lukas Kranz, Carolin Fella, Fabian Pianezzi, Christoph Adelhelm, Enrico Franzke, Stephan Buecheler, Ayodhya Nath Tiwari, Sodium-doped molybdenum back contacts for flexible Cu(In,Ga)Se2 solar cells, Thin Solid Films 535 (2013) 214–219. [107] P. Blösch, S. Nishiwaki, L. Kranz, C.M. Fella, F. Pianezzi, T. Jäger, C. Adelhelm, E. Franzke, S. Buecheler, A.N. Tiwari, Sodium-doped molybdenum back contact designs for Cu(In,Ga)Se2 solar cell, Sol. Energy Mater. Sol. Cells 124 (2014) 10–16. [108] Charles Roger, Giovanni Altamura, Fabrice Emieux, Olivier Sicardy, Frédéric Roux, Raphaël Fillon, Pascal Faucherand, Nicolas Karst, H.élène Fournier, Louis Grenet, Frédérique Ducroquet, Arnaud Brioude, Simon Perraud, Sodium-doped Mo back contacts for Cu(In,Ga)Se2 solar cells on Ti foils: growth, morphology, and sodium diffusion, J. Renew. Sustain. Energy 6 (2014) 011405. [109] Wen-Jen Tsai, Chia-Hung Tsai, Chih-Hui Chang, Jyh-Ming Ting, RuiRen Wang, Addition of Na into CuInS2 thin film via co-evaporation, Thin Solid Films 519 (2010) 1712–1716.

19

[110] Young Min Shin, Dong Hyeop Shin, Ji. Hye Kim, Byung Tae Ahn, Effect of Na doping using Na2S on the structure and photovoltaic properties of CIGS solar cells, Curr. Appl. Phys. 11 (2011) S59–S64. [111] Tokio Nakada, Hiroki Ohbo, Masakazu Fukuda, Akio Kunioka, Improved compositional flexibility of Cu(In,Ga)Se2-based thin film solar cells by sodium control technique, Sol. Energy Mater. Sol. Cells 49 (1997) 261–267. [112] D. Rudmann, D. Brémaud, H. Zogg, A.N. Tiwari, Na incorporation into Cu(In,Ga)Se2 for high-efficiency flexible solar cells on polymer foils, J. Appl. Phys. 97 (2005) 084903. [113] M. Bodegård, K. Granath, L. Stolt, Growth of Cu(In,Ga)Se2 thin films by coevaporation using alkaline precursors, Thin Solid Films 361–362 (2000) 9–16. [114] D. Güttler, A. Chirilă, S. Seyrling, P. Blösch, S. Buecheler, X. Fontané, V. Izquierdo-Roca, L. Calvo-Barrio, A. Peréz-Rodriguez, J.R. Morante, A. Eicke, A.N. Tiwari, Influence of NaF incorporation during Cu(In,Ga)Se2 growth on microstructure and photovoltaic performance, in: Proceedings of the 35th IEEE Photovoltaic Specialists Conference, 20–25 June 2010, pp 003420– 003424. [115] D. Rudmann, F.J. Haug, M. Kaelin, H. Zogg, A.N. Tiwari, G. Bilger, Low temperature growth of CIGS thin films for flexible solar cells, MRS Proceedings, vol. 668, H3, 2001, p. 8. doi: 10.1557/PROC-668-H3.8. [116] A.N. Tiwari, M. Krejci, F.-J. Haug, H. Zogg, 12.8% efficiency Cu(In,Ga)Se2 solar cell on a flexible polymer sheet, Prog. Photovolt.: Res. Appl. 7 (5) (1999) 393–397. [117] Anke Laemmle, Roland Wuerz, Michael Powalla, Efficiency enhancement of Cu(In,Ga)Se2 thin-film solar cells by a post-deposition treatment with potassium fluoride, Phys. Status Solidi (RRL) – Rapid Res. Lett. 7 (9) (2013) 631–634. [118] R. Caballero, C.A. Kaufmann, T. Eisenbarth, A. Grimm, I. Lauermann, T. Unold, R. Klenk, H.W. Schock, Influence of Na on Cu(In,Ga)Se2 solar cells grown on polyimide substrates at low temperature: impact on the Cu(In,Ga)Se2 /Mo interface, Appl. Phys. Lett. 96 (2010) 092104. [119] Raquel Caballero, Christian A. Kaufmann, Tobias Eisenbarth, Thomas Unold, Susan Schorr, Raik Hesse, Reiner Klenk, Hans-Werner Schock, The effect of NaF precursors on low temperature growth of CIGS thin film solar cells on polyimide substrates, Phys. Status Solidi A 206 (5) (2009) 1049–1053. [120] Karin Granath, Marika Bodegård, Lars Stolt, The effect of NaF on Cu(In, Ga)Se2 thin film solar cells, Sol. Energy Mater. Sol. Cells 60 (2000) 279–293. [121] R. Sakdanuphab, C. Chityuttakan, A. Pankiew, N. Somwang, K. Yoodee, S. Chatraphorn, Growth characteristics of Cu(In,Ga)Se2 thin films using 3-stage deposition process with a NaF precursor, J. Cryst. Growth 319 (2011) 44–48. [122] Johan Lindahl, Uwe Zimmermann, Piotr Szaniawski, Tobias Törndahl, Adam Hultqvist, Pedro Salomé, Charlotte Platzer-Björkman, Marika Edoff, Inline Cu(In,Ga)Se2 Co-evaporation for high efficiency solar cells and modules, IEEE J. Photovolt. 3 (3) (2013) 1100–1105. [123] M. Edoff, P.M.P. Salomé, A. Hultqvist, V. Fjällström, Analysis of NaF precursor layers during the different stages of the Cu(In,Ga)Se2 co-evaporation process, in: Proceedings of the 2013 MRS Spring Meeting & Exhibit, April 1–5, 2013, San Francisco, California, USA. [124] Sung Cheol Kim, Jaseok Koo, Woo Kyoung Kim, Rapid thermal processing of Cu-rich (InGa)2Se3/CuSe bilayer precursors with an inset NaF layer, Thin Solid Films 531 (2013) 233–237. [125] R. Wuerz, A. Eicke, M. Frankenfeld, F. Kessler, M. Powalla, P. Rogin, O. Yazdani-Assl, CIGS thin-film solar cells on steel substrates, Thin Solid Films 517 (2009) 2415–2418. [126] X. Song, R. Caballero, R. Félix, D. Gerlach, C.A. Kaufmann, H.-W. Schock, R.G. Wilks, M. Bär, Na incorporation into Cu(In,Ga)Se2 thin-film solar cell absorbers deposited on polyimide: impact on the chemical and electronic surface structure, J. Appl. Phys. 111 (2012) 034903. [127] R. Caballero, C.A. Kaufmann, T. Eisenbarth, M. Cancela, R. Hesse, T. Unold, A. Eicke, R. Klenk, H.W. Schock, The influence of Na on low temperature growth of CIGS thin film solar cells on polyimide substrates, Thin Solid Films 517 (2009) 2187–2190. [128] F. Pianezzi, A. Chirilă, P. Blösch, S. Seyrling, S. Buecheler, L. Kranz, C. Fella, A.N. Tiwari, Electronic properties of Cu(In,Ga)Se2 solar cells on stainless steel foils without diffusion barrier, Prog. Photovolt.: Res. Appl. 20 (3) (2012) 253–259. [129] Xiao Hui Tan, Sheng Lin Ye, Ken Bin Fan, Tang, Xu Liu, Effects of Na incorporated at different periods of deposition on Cu(In,Ga)Se2 films, Appl. Opt. 9 (6) (2010) 3071–3074. [130] Adrian Chirilă, Stephan Buecheler, Fabian Pianezzi, Patrick Bloesch, Christina Gretener, Alexander R. Uhl, Carolin Fella, Lukas Kranz, Julian Perrenoud, Sieghard Seyrling, Rajneesh Verma, Shiro Nishiwaki, Yaroslav E. Romanyuk, Gerhard Bilger, Ayodhya N. Tiwari, Highly efficient Cu(In,Ga)Se2 solar cells grown on flexible polymer films, Nat. Mater. 10 (2011) 857–861. [131] D.E. Tarrant, R.R. Gay, Process R&D for CIS-Based Thin-Film PV Final Technical Report April 2002 – April 2005, Subcontract Report NREL/SR-520-38805, January 2006. [132] Jorg Palm, Volker Probst, Franz H. Karg, Second generation CIS solar modules, Sol. Energy 77 (2004) 757–765. [133] J. Palm, V. Probst, A. Brummer, Stetter, R. Tölle, T.P. Niesen, S. Visbeck, O. Hernandez, M. Wendl, H. Vogt, H. Calwer, B. Freienstein, F. Karg, CIS module pilot processing applying concurrent rapid selenization and sulfurization of large area thin film precursors, Thin Solid Films 431–432 (2003) 514–522.

20

P.M.P. Salomé et al. / Solar Energy Materials & Solar Cells 143 (2015) 9–20

[134] Robert D. Wieting, Stion Corporation, Method of Manufacture of Sodium Doped CIGS/CIGSS Absorber Layers for High Efficiency Photovoltaic Devices, Grant US 8436445 B2, May 7, 2013. [135] D. Abou-Ras, R. Caballero, C.-H. Fische, C.A. Kaufmann, I. Lauermann, R. Mainz, H. Mönig, A. Schöpke, C. Stephan, C. Streeck, S. Schorr, A. Eicke, M. Döbeli, B. Gade, J. Hinrichs, T. Nunney, H. Dijkstra, V. Hoffmann, D. Klemm, V. Efimova, A. Bergmaier, G. Dollinger, T. Wirth, W. Unger, A.A. Rockett, A. Perez-Rodriguez, J. Alvarez-Garcia, V. Izquierdo-Roca, T. Schmid, P.-P. Choi, M. Müller, F. Bertram, J. Christen, H. Khatri, R.W. Collins, S. Marsillac, I. Kötschau, Comprehensive comparison of various techniques for the analysis of elemental distributions in thin films, Microsc. Microanal. 17 (2011) 728–751. [136] J.C. González, P.A. Fernandes, G.M. Ribeiro, A. Abelenda, E.R. Viana, P.M. P. Salomé, A.F. da Cunha, Influence of the sulphurization time on the morphological, chemical, structural and electrical properties of Cu2ZnSnS4 polycrystalline thin films, Sol. Energy Mater. Sol. Cells 123 (2014) 58–64. [137] Sebastian W. Schmitt, Cornel Venzago, Björn Hoffmann, Vladimir Sivakov, Thomas Hofmann, Johann Michler, Silke Christiansen, Gerardo Gamez, Glow discharge techniques in the chemical analysis of photovoltaic materials, Prog. Photovolt.: Res. Appl. 22 (3) (2014) 371–382. [138] P. Peres, A. Merkulov, F. Desse, M. Schuhmacher, SIMS analytical technique for PV applications, Surf. Interface Anal. 43 (1–2) (2011) 643–645. [139] Ken Durose, Sally E. Asher, Wolfram Jaegermann, Dean Levi, Brian E. McCandless, Wyatt Metzger, Helio Moutinho, P.D. Paulson, Craig L. Perkins, James R. Sites, Glenn Teeter, Mathias Terheggen, Physical characterization of thin-film solar cells, Prog. Photovolt.: Res. Appl. 12 (2–3) (2004) 177–217.

[140] D. Briggs, M.P. Seah, Practical Surface Analysis by Auger and X-Ray Spectroscopy, John Wiley & Sons Ltd., Chichester, New York, Brisbane, Toronto, Singapore, 1983. [141] T.F. Kelly, M.K. Miller, Invited review article: atom probe tomography, Rev. Sci. Instrum. 78 (2007) 031101. [142] Jiaqing He, Li-Dong Zhao, Jin-Cheng Zheng, Jeff W. Doak, Haijun Wu, HuiQiong Wang, Yeseul Lee, Chris Wolverton, Mercouri G. Kanatzidis, Vinayak P. Dravid, Role of sodium doping in lead chalcogenide thermoelectrics, J. Am. Chem. Soc. 135 (12) (2013) 4624–4627. [143] Qinyong Zhang, Hengzhi Wang, Qian Zhang, Weishu Liu, Bo Yu, Hui Wang, Dezhi Wang, George Ni, Gang Chen, Zhifeng Ren, Effect of silicon and sodium on thermoelectric properties of thallium-doped lead telluride-based materials, Nano Lett. 12 (5) (2012) 2324–2330. [144] Min-Ling Liu, I-Wei Chen, Fu-Qiang Huang, Li-Dong Chen, Improved thermoelectric properties of Cu-doped quaternary chalcogenides of Cu2CdSnSe4, Adv. Mater. 21 (2009) 3808–3812. [145] Min-Rui Gao, Yun-Fei Xu, Jun Jiang, Shu-Hong Yu, Nanostructured metal chalcogenides: synthesis, modification, and applications in energy conversion and storage devices, Chem. Soc. Rev. 42 (2013) 2986–3017. [146] M. Popescu, Disordered chalcogenide optoelectronic materials: phenomena and applications, J. Optoelectron. Adv. Mater. 7 (4) (2005) 2189–2210. [147] T. Nakada, Y. Honishi, Y. Yatsushiro, H. Nakakoba, Impacts of Sb and Bi incorporations on CIGS thin films and solar cells, in: Proceedings of the 37th IEEE Photovoltaic Specialists Conference (PVSC), 2011, pp. 003527–003531. [148] A. Laemmle, R. Wuerz, M. Powalla, Investigation of the effect of potassium on Cu(In,Ga)Se2 layers and solar cells, Thin solid films 582 (2015) 27–30.