• No results found

Citation for the original published paper (version of record):

N/A
N/A
Protected

Academic year: 2022

Share "Citation for the original published paper (version of record):"

Copied!
10
0
0

Loading.... (view fulltext now)

Full text

(1)

http://www.diva-portal.org

Postprint

This is the accepted version of a paper published in Solar RRL.

Citation for the original published paper (version of record):

Ericson, T., Larsson, F., Törndahl, T., Frisk, C., Larsen, J. et al. (2017)

Zinc-Tin-Oxide Buffer Layer and Low Temperature Post Annealing Resulting in a 9.0%

Efficient Cd-Free Cu2ZnSnS4 Solar Cell.

Solar RRL

https://doi.org/10.1002/solr.201700001

Access to the published version may require subscription.

N.B. When citing this work, cite the original published paper.

Permanent link to this version:

http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-321006

(2)

Supporting Information

Zinc-Tin-Oxide Buffer Layer and Low Temperature Post Annealing Resulting in a 9.0 % Efficient Cd-free Cu

2

ZnSnS

4

Solar Cell

Tove Ericson, Fredrik Larsson, Tobias Törndahl, Christopher Frisk, Jes Larsen, Volodymyr Kosyak, Carl Hägglund, Shuyi Li, Charlotte Platzer-Björkman*

Department of Engineering Sciences, Solid State Electronics, Ångström Solar Center, Box 534, 751 21, Uppsala, Sweden.

E-mail: chpl@angstrom.uu.se, toer@angstrom.uu.se.

(3)

S.1 TEM analysis

Sample D was investigated by transmission electron microscopy (TEM) as described in the

Experimental section of the article. Below some of the results are presented. For this sample the

estimated thickness was 5-10 nm from TEM. The TEM investigation showed no clear indication

of interdiffusion between buffer and absorber, however, due to the thin layer thickness more

thorough characterization would have to be done to rule this out.

(4)

Figure S1. Scanning transmission electron microscopy (STEM) image of the ZnO- ZTO- CZTS

interface.

(5)

Figure S2. High resolution transmission electron microscopy (HR TEM) image of the ZnO-

ZTO- CZTS interface.

(6)

Figure S3. Energy dispersive X-ray spectroscopy (EDS) line scan of the ZnO- ZTO- CZTS interface.

0 10 20 30 40

[nm]

[arb. u nit s]

(7)

S.2 Post annealing effects

Below a more thorough discussion about three of the different possible effects of post annealing on these samples is given.

S.2.1 Cu-Zn disorder

For the Cu-Zn disorder our previous experiments indicates that major changes only occur for longer times or higher temperatures.

[1]

Additionally, the absorber bandgap should be affected by the ordering process. Looking at the QE bandgap for the two extra series in this paper, the two unheated CdS-references show a bandgap of 1.50 and 1.49 eV respectively, while the heated samples show 1.49 and 1.50 eV, indicating that the changes in Cu-Zn-disorder for these samples are small. This was also confirmed by Raman spectroscopy, where no significant changes were observed.

S.2.2 Zn-enrichment of surface

Generally the papers showing a change in cation surface composition have used higher annealing temperatures, for example, Tajima et al. made anneals at 300 and 400 °C which showed a Zn- enriched surface while the sample annealed at 200 °C seemed unaffected in this sense.

[2]

However, the results from Dimitrievska et al. points to that the surface composition in CZTSe

starts to be affected already from 150 °C.

[3]

In this case Raman spectroscopy was used to evaluate

air-annealing at different temperatures and the changes in Raman spectra was connected to

changes in the amount of certain defect complexes. There are also reports where no change in Zn

is seen even for high temperatures, as for example reported by Hironiwa et al., for temperatures

up to 300 °C.

[4]

(8)

We recently presented a study of air-annealing, performed before etching and buffer deposition, with similar samples as the ones used in the present paper.

[5]

For a sulfo-selenide sample that was air-annealed at 100 °C, changes in the surface composition of Zn were seen by x-ray photoelectron spectroscopy (data not shown in the article). Somewhat similar trends in the

Raman spectra, as the ones presented by Dimitrievska et al., were observed.

[3]

However, in the air annealing series, the sample heated in air at 100 °C showed a large improvement in electrical properties, opposite to what was seen for the sample heated at 105 °C in N

2

in the present series.

CV profiling was also done on the air-annealing series (data not shown in the article). The sulfo- selenide air-annealed at 300 °C and its non-heated reference showed no significant difference in doping level, indicating that surface Zn-enrichment would not affect the doping value measured by CV.

The change in QE seen for the Zn-treated sample is in agreement with the observation that the Zn-composition at the surface having an influence on the device properties. Regarding changes in Raman spectra, we do not find a reference for how the spectra changes with surface Zn-

enrichment, for the sulfide case. As mentioned above, in the present study no clear trends with temperature was seen in the Raman spectra, but if this can be said to be an indication of a lack of Zn-surface composition change is not possible to conclude at the moment.

S.2.3 Na-redistribution

Redistribution of Na in CZTS should be possible at low temperatures and short times, due to the

small size of the Na-atoms. For CIGS, very quick diffusion of Na has been observed, mainly

through grain boundaries but also in the bulk.

[6]

Laemmle et al. suggest the direct interstitial

mechanism as most likely for Na diffusion in CIGS and from the obtained data it is, for example,

(9)

calculated that Na could diffuse 758 nm just during a chemical bath deposition process at 65 °C for 10 min.

[6]

Due to the similarity between CIGS and CZTS crystal structure it is not far-fetched to assume a similar diffusion behavior in CZTS. The Na-explanation could agree with a

phenomenon that is varying heavily, depending on deposition method and the exact conditions for the post annealing. Different ways of creating the absorber will leave different amount of Na inside the film. Additionally, the removal of Na from the solar cell surface could differ widely depending on at which step the solar cell is annealed, at which temperature and in which kind of atmosphere, both type and pressure. The latter was shown by Xie et al. where the Na spatial distribution was connected to the electrical performance of the devices, and an N

2

anneal at 200 °C completely deteriorated the device performance when it was performed in 1 bar, but barely affected the electronic properties when it was carried out at 1x10

-6

bar.

[7]

The theory presented in that paper, that there are two competing processes, Na supply (from the glass to the absorber) and removal (phase transformation and evaporation from the surface) could fit our results, but due to our lower pressure the evaporation from the surface would then be increased and the optimal point would occur at a lower temperature, which could agree with the results seen.

[7]

When the Na-redistribution is occurring after the buffer has started to be deposited, also the type of buffer may have a large influence, as seen for example in the case of CdS compared to

Zn(O,S) on CIGS.

[8]

Here voltage and heat was used to degrade and revive solar cell devices and it was seen that Na was accumulated in CdS but blocked by Zn(O,S) resulting in that the CdS device was possible to recover while the device with Zn(O,S) was not. Also for CZTS it has been shown that the Na-content in the buffer layer may have an influence on the electronic

properties.

[9]

In this case only heating (300 °C for 30 min) was used as treatment.

(10)

A main difference between our two supplementary series is that the surface was etched before in one case and after in the other case. This will affect which compounds are present on the surface, which in turn could affect for example the progression of Na-diffusion. In the case of samples A- D the situation is even more complex since a film is continuously growing, and additionally, the atmosphere is varying with the ALD pulses.

[1] J.J.S. Scragg, L. Choubrac, A. Lafond, T. Ericson, C. Platzer-Björkman, Appl. Phys. Lett.

2014, 104, 041911.

[2] S. Tajima, R. Asahi, D. Isheim, D.N. Seidman, T. Itoh, M. Hasegawa, K. Ohishi, Appl.

Phys. Lett. 2014, 105, 093901.

[3] M. Dimitrievska, S. Giraldo, P. Pistor, E. Saucedo, A. Pérez-Rodríguez, V. Izquierdo- Roca, Sol. Energy Mater. Sol. Cells 2016, 157, 462.

[4] D. Hironiwa, N. Sakai, T. Kato, H. Sugimoto, Z. Tang, J. Chantana, T. Minemoto, Thin Solid Films 2015, 582, 151.

[5] J.K. Larsen, Y. Ren, N. Ross, E. Särhammar, S.Y. Li, C. Platzer-Björkman, Thin Solid Films, DOI: 10.1016/j.tsf.2016.08.030.

[6] A. Laemmle, R. Wuerz, T. Schwarz, O. Cojocaru-Mirédin, P.-P. Choi, M. Powalla, J.

Appl. Phys. 2014, 115, 154501.

[7] H. Xie, S. López-Marino, T. Olar, Y. Sánchez, M. Neuschitzer, F. Oliva, S. Giraldo, V.

Izquierdo-Roca, I. Lauermann, A. Pérez-Rodríguez, E. Saucedo, ACS Applied Materials

& Interfaces 2016, 8, 5017.

[8] V. Fjällström, P. Szaniawski, B. Vermang, P.M.P. Salomé, F. Rostvall, U. Zimmermann, M. Edoff, IEEE Journal of Photovoltaics 2015, 5, 664.

[9] K.-J. Yang, J.-H. Sim, D.-H. Son, D.-H. Kim, J.-K. Kang, Current Applied Physics 2015,

15, 1512.

References

Related documents

Och i flera fall är det ju också så att det blir, faktiskt ses som en liten fjäder i hatten om man jobbar, alltså att det är någonting man faktiskt kan lyfta fram i relation till

Denna tudelning, med formen av en arbetsdelning, genererar ett särskilt förhållande till idén om dikten vars autonomi på en gång utmanas och upprättas – ett förhållande som i

Haukås (2016) studie om lärares uppfattningar om flerspråkighet visar bland annat att lärarna anser att det är viktigt att elever är medvetna om hur de kan dra nytta av sina

This is the published version of a paper published in Arkitektur: byggnad, interiör, plan, landskap.. Citation for the original published paper (version

Surface enrichment of polymer at the free surface in polymer:fullerene blend lms has previously been observed by dynamic secondary ion mass spectrometry (d- SIMS), 19–21

Kursen riktar sig till lärare från samtliga ämnesområden med målet att lära sig mer om hållbar utveckling, att stärka kompetenser för att undervisa om/i hållbar utveckling

Den röst man får istället är en mer indirekt sådan, då den fria versen i första hand är vers för ögat och inte örat: den skiljer sig från prosan endast genom den

Samtidigt behövs stora grepp för att de gröna miljöerna i våra städer ska bli ännu mer gynnsamma för biologisk mångfald inklusive pollinatörer, och för att