J Mater Sci: Mater Electron (2016) 27:705–710 DOI 10.1007/s10854-015-3806-5

Annealing induced amorphous/crystalline silicon interface passivation by hydrogen atom diffusion Xiaowan Dai1 • Hongkun Cai2 • Dexian Zhang2 • Guifeng Chen3 • Yong Wang4 Wei Liu1 • Yun Sun1

•

Received: 25 May 2015 / Accepted: 29 September 2015 / Published online: 7 October 2015 Ó Springer Science+Business Media New York 2015

Abstract Post-annealing is an efficient method to improve passivation quality of the amorphous/crystalline silicon configuration and it’s been widely used in fabrication of amorphous/crystalline silicon heterojunction solar cells. In this study, hydrogenated amorphous silicon thin films are deposited on n type monocrystalline silicon, using a single chamber radio-frequency plasma-enhanced chemical vapor deposition system. After passivation the best result with effective minority carrier lifetime (seff) exceeding 1600 ls is achieved. Fourier transform infrared spectrum is utilized to investigate the structure evolution after annealing. The result shows that, the improved passivation quality is attributed to interface dangling bonds saturated by H atoms diffusion during annealing, the reason of passivation quality affected by annealing and film thickness is also proposed.

& Hongkun Cai [email protected] 1

Institute of Photo-Electronic Thin Film Devices and Technology, Nankai University, Tianjin 300071, China

2

Department of Electronics and Microelectronics, College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300071, China

3

School of Materials Science and Engineering, Hebei University of Technology, Tianjin 300130, China

4

Haitai New Energy Technology Co. Ltd., No.88 Haomen Road, Yutai Industrial Park, Yutian County, Tangshan City, Hebei Province 064100, China

1 Introduction As the large area silicon based solar cells with the highest efficiency to date, silicon heterojunction solar cells with intrinsic thin film (HIT) attracted a lot of investigation [1– 7]. With superior interface passivation and wide bandgap of amorphous silicon, amorphous/crystalline silicon solar cells could achieve higher open-voltage (Voc), which levels up the short-circuit current density (Jsc) and fill-factor (FF), even similar to traditional monocrystalline silicon solar cells, thus achieving a higher efficiency (25.6 %) [1]. In contrast to PERL (passivated emitter, rear locally-diffused) solar cells developed by UNSW [8], amorphous/crystalline silicon solar cells exhibit the feature of easier processing, lower temperature and thinner silicon wafer, and promising application. Superior interface passivation is necessary to achieve high efficiency amorphous/crystalline heterojunction solar cells [9–11]. Carrier recombination velocity in pn junction is drastically decreased after passivation, which ensures the high Voc and FF in final devices. Many researchers have revealed that, abrupt interface between amorphous and crystalline silicon is crucial in order to achieve superior surface passivation, which is the transform from amorphous to monocrystalline without too much mico/nanocrystal transition state [12–14]. In order to achieve higher carrier lifetime, post-annealing is widely used to optimize surface passivation quality during layer deposition. Generally, carrier lifetime of experimental samples can be significantly enhanced after annealing (without any crystallization before or after annealing), this phenomenon has a general explanation: the film structure changes during annealing, film quality is improved, therefore improved passivation quality is achieved [15, 16].

123

706

J Mater Sci: Mater Electron (2016) 27:705–710

However, there are still many questions about the mechanism which influences the passivation result during annealing. Is the improved film quality really the key issue of improved passivation? Yet, the influence of film thickness on the final passivation quality is not clear. In this article, we use FTIR method to investigate the annealed amorphous passivation layer on silicon wafer. Hopefully, results could reveal that, the improved passivation is not entirely attributed to the optimization of bulk film quality, but mainly attributed to the saturation of interface dangling bonds by hydrogen atoms.

2 Experimental Commercial Cz Si wafers with the parameters of 1–3 X cm resistivity, n-type, 250 lm thickness, (111) oriented, are utilized. Symmetrically deposited intrinsic amorphous silicon (i-a-Si:H) is double-side polished, prior to the i-a-Si:H deposition, silicon wafers are ultrasonically cleaned under the temperature of 80 °C for 10 min in RCA1 solution (NH4OH:H2O2:DI water = 1:1:5). After cleaning, substrates were dipped in 1 % hydrofluoric acid for 2 min to remove native oxide, then blow-dried with nitrogen before transported into PECVD system, which is equipped with load-lock chamber and shower-head parallel plate reactor. As to the film deposition, a plasma excitation frequency of 13.56 MHz is used and post-deposition annealing is still conducted under vacuum condition. The deposition and annealing parameters are listed in Table 1. To determine the passivation quality, effective minority carrier lifetime is measured by l-PCD method with the Semilab WT-1200 at an injection level of 1013 cm-3. Film thickness is estimated by testing a-Si:H deposited on glass substrate using profilometer. Surface morphology and bulk microstructure of passivation films are studied with AFM and FTIR methods.

3 Results and discussion Silicon wafers with 30 nm passivation layers deposited under 900 mTorr are annealed under 250 °C with different time. Figure 1 shows the difference of effective minority carrier lifetime of these samples. It is clear that seff increases first and then decreases with increased annealing time. To understand this phenomenon, their FTIR spectra are investigated. Figure 2a, b are the FTIR spectra of these passivation layers in the range of 500–800 and 1800–2300 cm-1 respectively. In Fig. 2b, it is noted that absorption peak from 1900 to 2200 cm-1 doesn’t change much until the annealing time exceeds 60 min, which indicates SiH2 (with appearance at the wavenumber of 2075 cm-1), the stretching mode of these films does not change significantly after short time annealing(will be discussed later). In Fig. 2a, the absorption peaks at around 640 cm-1 represent the decrease of film H content as annealing time increases. In the case of 90 min, extensively reduced H content deteriorates the passivation quality. Figure 3 depicts FTIR spectra of the 30 nm passivation layers after 30 min annealing, a double-Gauss function fitting is used to analyze such FTIR spectra. The silicon thin films have two main absorption modes in this range. One mode is the Si–H stretching mode which is at 2000 cm-1, the other is SiH2 stretching mode at 2100 cm-1 [17]. In FTIR analyzing, microstructure factor (R) is used to character the film feature of compactness and quality. R is defined as R = ISH2/(ISH ? ISH2), where I is integral intensity of corresponding Gauss peak. The SH2 stretching mode represents loose and defective tissue in the silicon films, so a larger R value should be able to indicate that the film is of bad quality. In Fig. 3a, b, R value of the films deposited under same condition does not change obviously before and after the annealing, implying a negligible influence on film structure after annealing in this 1800 250°C annealing

Table 1 Deposition parameter of intrinsic a-si films Deposition parameter Hydrogen (sccm)

9

Silane (sccm)

21

Deposition temperature (°C)

160/200

Pressure (mTorr)

500/900

Power density (mw/cm2)

10

Annealing temperature (°C)

250

Annealing time (min)

30

Base vacuum (Pa)

10-5

123

Minority carrier lifetime(μs)

1600 1400 1200 1000 800 600 400 0

20

40

60

Annealing time(min)

Fig. 1 seff as a function of annealing time

80

100

J Mater Sci: Mater Electron (2016) 27:705–710

707

a

b

500

2075cm

0 min 10min 30min 60min 90min

-1

Intensity(a.u.)

Intensity(a.u.)

0min 30min 60min 90min

550

600

650

700

750

800

1800

1900

2000

-1

2100

2200

2300

-1

Wavenumber(cm )

wavenumber(cm )

Fig. 2 FTIR spectra of 30 nm passivation films annealed for different time a in the range of 500–800 cm-1, b normalized, in the range of 1800–2300 cm-1

a

τeff : 502us

b

τeff : 1621us

R : 0.398

c

τeff : 1363us

R : 0.420

Intensity(a.u.)

R : 0.678

1800

1900

2000

2100

Wavenumber(cm-1)

2200 1800

1900

2000

2100

Wavenumber(cm-1)

2200 1800

1900

2000

2100

2200

Wavenumber(cm-1)

Fig. 3 FTIR spectra of the passivation layers a 30 nm as deposited, b 30 nm annealed, c 30 nm (50 % silane concentration) annealed

case. But the carrier lifetime is drastically changed, from 502 ls before to 1621 ls after annealing, implying the increase of carrier lifetime exceeds 300 %. This also demonstrated that the annealed film is almost completely amorphous state [18]. In order to make further contrast, films with the same thickness are deposited using a silane concentration of 50 %. As shown in Fig. 3c, although the microstructure factor reaches a high value of 0.678, i.e. very bad bulk quality, high carrier lifetime of 1393 ls can still be achieved. So we can deduce that, high film bulk quality is not necessary to obtain high carrier lifetime, and annealing induced carrier lifetime optimization is not attributed to the improvement of film bulk quality. Actually, hydrogen rich passivation layer contains many weak-boding H atoms, during annealing, these weak-boding H can diffuse and saturate interface dangling bonds, decrease the interface state density and cause no detectable change at the same time [19]. To investigate the influence of bulk structure upon surface passivation during annealing, the film thickness is decreased down to 15 nm. To make sure the film thickness more controllable, the deposition pressure is set to 500 mTorr. FTIR

spectra of films with different thickness 5, 8, 11 and 15 nm are shown in Fig. 4 respectively, their minority carrier lifetime and microstructure factor are listed in Table 2. Usually, in a certain range, the carrier lifetime become saturated along with increasing film thickness, i.e. passivation quality deteriorates with decreased film thickness [20–22]. Nevertheless, surface dangling bond saturated by hydrogen is generally thought as the reason why amorphous silicon can passivate monocrystalline silicon. In another word, if the surface is 100 % covered by passivation layer, carrier lifetime shouldn’t have obvious variation when the thickness increases. This is apparently different from the real case. De Wolf and Beaucarne [23] reported that super thin film caused the overflow of electron wave function from interface, but without any detailed discussion. From Table 2, it’s obvious that carrier lifetime varies with the passivation layer thickness, besides, compared with the prior film deposited under 900 mTorr, the deposited film here shows a lower R value, indicating the film deposited under higher pressure tends to become more disordered. In Table 2, all films show the low passivation ability as deposited. After annealing, the carrier lifetime increases with increased film thickness, meanwhile the microstructure factor decreases. All samples are

123

708

a

b Intensity(a.u.)

Intensity(a.u.)

Fig. 4 FTIR spectra of passivation layers with different thickness a 5 nm, b 8 nm, c 11 nm, d 15 nm

J Mater Sci: Mater Electron (2016) 27:705–710

1800

1900

2000

2100

2200

2300

1800

1900

c

2000

2100

1800

2200

1900

2000

2100

2200

Wavenumber(cm-1)

Thickness (nm)

5

8

11

15

Microstructure factor after annealed (R)

0.454

0.348

0.234

0.252

Lifetime before annealed (ls)

177

136

151

197

Lifetime after annealed (ls)

169

228

674

741

deposited under the same condition and, the changes in microstructure factor reflect the difference of films with different thickness upon annealing. Annealing doesn’t have an obvious effect on film bulk microstructure especially in 5 nm case, after annealing, the film still maintains a high microstructure factor. The film tends to form SHn[1 rich structure at the initial stage of passivation layer growth [24– 26], some researchers showed that SH2 configuration tends to transform to SH during annealing, in this processing, removable hydrogen is released to passivate the interface and reduce the recombination velocity [9, 27]. In our case, the microstructure factor of both 11 and 15 nm films are improved after annealing and, the surface passivation quality is also optimized. But when it comes down to the ultra-thin film case, annealing has a weak effect on film microstructure, it is like the film was adhered on the silicon substrate and can’t be moved, therefore there is not enough hydrogen to passivate interface after annealing in ultra-thin passivation layer. Figure 5 depicts AFM surface morphology of amorphous thin films with thickness of 8 and 15 nm on silicon substrate before and after annealing. 8 nm film shows higher root-mean-square roughness (RMS) than 15 nm film, implying the island growth feature at initial stage of

123

2200

Intensity(a.u.) 1900

Wavenumber(cm-1)

Table 2 Test results of passivation films with different thickness

2100

d

Intensity(a.u.) 1800

2000

Wavenumber(cm-1)

Wavenumber(cm-1)

silicon films. The bright spike structure on 15 nm film is a place where the crystallization occurred (demonstrated in glass substrate case and not shown here). In the 15 nm case, RMS value of passivation film shows slight improvement after annealing, while the 8 nm film doesn’t have any change. This indicates that, in this work the network architecture of passivation layer don’t have significant change, the change mainly take place when it is Si– H bonding mode. We assume that the improved minority carrier lifetime of 11 nm and 15 nm films should be attributed to H released from upper part of the film during annealing, which permeate underneath part of the film towards the a-Si/c-Si interface, and saturate the si dangling bonds. For ultra-thin (below 10 nm) films, as mentioned above, the a-Si network is hardly transformed during annealing, for it is adjacent to silicon surface, therefore lacking H atoms to passivate the surface dangling bonds. Other parameters are kept constant while deposition temperature is increased to 200 °C, Fig. 6 depicts FTIR spectra of 15 nm passivation layer deposited under 200 °C. The value of microstructure factor is low, SH2/SH ratio is low for passivation layer, implying that annealing has a weak effect on Si– H bonding mode, therefore, lack of removable hydrogen causes small improvement on carrier lifetime, it is also a

J Mater Sci: Mater Electron (2016) 27:705–710

709

Annealing 15nm

RMS:4.972

RMS:4.445

Annealing 8nm

RMS:5.789

RMS:5.801

Fig. 5 AFM morphology of passivation films as-deposited and post-annealed

a

b

1800

Intensity(a.u.)

Intensity(a.u.)

τ=314μs R=0.187

1900

2000

2100

2200

1800

τ=376μs R=0.166

1900

2000

2100

2200

Wavenumber(cm-1)

Wavenumber(cm-1)

Fig. 6 FTIR spectra of passivation films deposited under 200 °C a as-deposited, b post-annealed

convincing proof that SH2 can be useful when improving passivation quality during annealing in amorphous silicon.

4 Conclusions In summary, we deposit amorphous silicon thin films as the passivation layer on n-type (111) crystal-silicon and, the highest carrier lifetime which exceeds 1600 ls is achieved after annealing using l-PCD test method. Also FTIR is utilized to investigate the structure change in passivation layer after annealing. The results show that, the transform of S–H bonding mode is responsible for the promotion of effective

minority carrier lifetime, hydrogen released in this processing has a surface passivation effect. However, when the film is ultrathin (under 10 nm), the effect of annealing on Si–H bonding mode is weakened and, the film with lower thickness shows low effective carrier lifetime because of lack of removable hydrogen to passivate interface.

References 1. K. Masuko, M. Shigematsu, T. Hashiguchi, D. Fujishima, M. Kai, N. Yoshimura, T. Yamaguchi, Y. Ichihashi, T. Mishima, N. Matsubara, T. Yamanishi, T. Takahama, M. Taguchi, E.

123

710

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

J Mater Sci: Mater Electron (2016) 27:705–710 Maruyama, S. Okamoto, Achievement of more than 25% conversion efficiency with crystalline silicon heterojunction solar cell. IEEE J. Photovolt. 4(6), 1433–1435 (2014) M. Filipicˇ, Z. Holman, F. Smole, S. De Wolf, C. Ballif, M. Topicˇ, Analysis of lateral transport through the inversion layer in amorphous silicon/crystalline silicon heterojunction solar cells, in Proceedings of 29th International Conference on Microelectronics. (2014), pp. 227–230 S.Y. Herasimenka, W.J. Dauksher, S.G. Bowden, [750 mV open circuit voltage measured on 50 lm thick silicon heterojunction solar cell. Appl. Phys. Lett. 103, 053511 (2013) M. Schaper, J. Schmidt, H. Plagwitz, R. Brendel, 20.1 %-Efficient crystalline silicon solar cell with amorphous silicon rearsurface passivation. Prog. Photovolt. Res. Appl. 13, 381–386 (2005) A. Descoeudres, Z.C. Holman, L. Barraud, S. Morel, S. De Wolf, C. Ballif, [21% Efficient silicon heterojunction solar cells on nand p-type wafers compared. IEEE J. Photovolt. 3(1), 83–89 (2013) L. Barraud, Z.C. Holman, N. Badel, P. Reiss, A. Descoeudres, C. Battaglia, S. De Wolf, C. Ballif, Hydrogen-doped indium oxide/ indium tin oxide bilayers for high-efficiency silicon heterojunction solar cells. Sol. Energy Mater. Sol. Cell s 115, 151–156 (2013) D. Chen, L. Zhao, H. Diao, W. Zhang, G. Wang, W. Wang, Lowtemperature sintering properties of the screen-printed silver paste for a-Si:H/c-Si heterojunction solar cells. J. Mater. Sci. Mater. Electron. 25, 2657–2664 (2014) J. Zhao, A. Wang, M.A. Green, 24.5% Efficiency silicon PERT cells on MCZ substrates and 24
7% efficiency PERL cells on FZ substrates. Prog. Photovolt. Res. Appl. 7, 471–474 (1999) J.A. Schu¨ttauf, K.H.M. van der Werf, I.M. Kielen, W.G.J.H.M. van Sark, J.K. Rath, R.E.I. Schropp, Excellent crystalline silicon surface passivation by amorphous silicon irrespective of the technique used for chemical vapor deposition. Appl. Phys. Lett. 98, 153514 (2011) U.K. Das, M.Z. Burrows, M. Lu, S. Bowden, R.W. Birkmire, Surface passivation and heterojunction cells on Si (100) and (111) wafers using dc and rf plasma deposited Si:H thin films. Appl. Phys. Lett. 92, 063504 (2008) Q. Zhang, M. Zhu, F. Liu, Y. Zhou, The optimization of interfacial properties of nc-Si:H/c-Si solar cells in hot-wire chemical vapor deposition process. J. Mater. Sci. Mater. Electron. 18, S33– S36 (2007) S. De Wolf, M. Kondo, Abruptness of a-Si:H/c-Si interface revealed by carrier lifetime measurements. Appl. Phys. Lett. 90, 042111 (2007) H. Fujiwara, T. Kaneko, M. Kondo, Optimization of interface structures in crystalline silicon heterojunction solar cells. Sol. Energy Mater. Sol. Cell s 93, 725–728 (2009) M.R. Page, E. Iwaniczko, Y.-Q. Xu, L. Roybal, F. Hasoon, Q. Wang, R.S. Crandall, Amorphous/crystalline silicon heterojunction solar cells with varying i-layer thickness. Thin Solid Films 519, 4527–4530 (2011)

123

15. A. Descoeudres, L. Barraud, S. De Wolf, B. Strahm, D. Lachenal, C. Gue´rin, Z.C. Holman, F. Zicarelli, B. Demaurex, J. Seif, J. Holovsky, C. Ballif, Improved amorphous/crystalline silicon interface passivation by hydrogen plasma treatment. Appl. Phys. Lett. 99, 123506 (2011) 16. M. Jeon, S. Yoshiba, K. Kamisako, Hydrogenated amorphous silicon film as intrinsic passivation layer deposited at various temperatures using RF remote-PECVD technique. Curr. Appl. Phys. 10, S237–S240 (2010) 17. L. Zhao, H. Diao, X. Zeng, C. Zhou, H. Li, W. Wang, Comparative study of the surface passivation on crystalline silicon by silicon thin films with different structures. Phys. B 405, 61–64 (2010) 18. H.J. Yang, K. Ji, J. Choi, H.M. Lee, Annealing effect on surface passivation of a-Si:H/c-Si interface in terms of crystalline volume fraction. Curr. Appl. Phys. 10, S375–S378 (2010) 19. T.F. Schulze, H.N. Beushausen, T. Hansmann, L. Korte, B. Rech, Accelerated interface defect removal in amorphous/crystalline silicon heterostructures using pulsed annealing and microwave heating. Appl. Phys. Lett. 95, 182108 (2009) 20. D. Pysch, M. Bivour, K. Zimmermann, C. Schetter, M. Hermle, S.W. Glunz, Comprehensive study of different PECVD deposition methods for deposition of thin intrinsic amorphous silicon for heterojunction solar cells, in Proceedings of 24th European Photovoltaic Solar Energy Conferrence, Hamburg, Germany (2009) 21. J. Ge, Z.P. Ling, J. Wong, T. Mueller, A.G. Aberle, Optimisation of intrinsic a-Si:H passivation layers in crystalline-amorphous silicon heterojunction solar cells. Energy Procedia 15, 107–117 (2012) 22. R. Ferre, I. Martı´n, M. Vetter, M. Garı´n, R. Alcubilla, Effect of amorphous silicon carbide layer thickness on the passivation quality of crystalline silicon surface. Appl. Phys. Lett. 87, 202109 (2005) 23. S. De Wolf, G. Beaucarne, Surface passivation properties of boron-doped plasma-enhanced chemical vapor deposited hydrogenated amorphous silicon films on p-type crystalline Si substrates. Appl. Phys. Lett. 88, 022104 (2006) 24. H. Fujiwara, M. Kondo, Real-time monitoring and process control in amorphous/crystalline silicon heterojunction solar cells by spectroscopic ellipsometry and infrared spectroscopy. Appl. Phys. Lett. 86, 032112 (2005) 25. S.N. Granata, T. Bearda, F. Dross, I. Gordon, J. Poortmans, R. Mertens, Effect of an in situ H2 plasma pretreatment on the minority carrier lifetime of a-Si:H (i) passivated crystalline silicon. Energy Procedia 27, 412–418 (2012) 26. S. De Wolf, S. Olibet, C. Ballif, Stretched-exponential a-Si:H/cSi interface recombination decay. Appl. Phys. Lett. 93, 032101 (2008) 27. M.Z. Burrows, U.K. Das, R.L. Opila, S. De Wolf, R.W. Birkmire, Role of hydrogen bonding environment in a-Si:H films for c-Si surface passivation. J. Vac. Sci. Technol. A 26, 683 (2008)