Modulation of atomic-layer-deposited Al2O3 film passivation of silicon surface by rapid thermal processing

Article abstract of DOI:10.1063/1.3616145

We have investigated the effect of a thin interfacial silicon oxide on the atomic-layer-deposited Al₂O₃ film passivating the silicon surface based on rapid thermal process (RTP). It is found that the effective carrier lifetime of sam

Full text of DOI:10.1063/1.3616145

Modulation of atomic-layer-deposited Al2O3 film passivation of silicon surface by rapid
thermal processing
Dong Lei, Xuegong Yu, Lihui Song, Xin Gu, Genhu Li, and Deren Yang
Citation: Applied Physics Letters 99, 052103 (2011); doi: 10.1063/1.3616145
View online: http://dx.doi.org/10.1063/1.3616145
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/99/5?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Deposition temperature dependence of material and Si surface passivation properties of O3-based atomic layer
deposited Al2O3-based films and stacks
J. Vac. Sci. Technol. A 32, 01A128 (2014); 10.1116/1.4852855
Surface passivation of nano-textured silicon solar cells by atomic layer deposited Al2O3 films
J. Appl. Phys. 114, 174301 (2013); 10.1063/1.4828732
Si surface passivation by Al2O3 thin films deposited using a low thermal budget atomic layer deposition process
Appl. Phys. Lett. 102, 131603 (2013); 10.1063/1.4800541
Plasma-enhanced and thermal atomic layer deposition of Al2O3 using dimethylaluminum isopropoxide,
[Al(CH3)2(μ-O i Pr)]2, as an alternative aluminum precursor
J. Vac. Sci. Technol. A 30, 021505 (2012); 10.1116/1.3683057
Controlling the fixed charge and passivation properties of Si(100)/Al2O3 interfaces using ultrathin SiO2
interlayers synthesized by atomic layer deposition
J. Appl. Phys. 110, 093715 (2011); 10.1063/1.3658246
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
128.233.210.97 On: Wed, 15 Oct 2014 04:41:40

APPLIED PHYSICS LETTERS 99, 052103 (2011)
Modulation of atomic-layer-deposited Al O film passivation of silicon
2
3
surface by rapid thermal processing
a)
Dong Lei, Xuegong Yu, Lihui Song, Xin Gu, Genhu Li, and Deren Yang
State Key Lab of Silicon Materials and Department of Materials Science and Engineering, Zhejiang University,
3
10027 Hangzhou, People’s Republic of China
(Received 30 May 2011; accepted 5 July 2011; published online 1 August 2011)
We have investigated the effect of a thin interfacial silicon oxide on the atomic-layer-deposited
Al O film passivating the silicon surface based on rapid thermal process (RTP). It is found that the
effective carrier lifetime of samples strongly depends on the RTP temperature and reaches the
2
3
ꢀ
maximum value at 550 C. Both capacitance-voltage measurements and theoretical simulation
have revealed that the RTP treatment cannot only modulate the charges in the Al O film but also
2
3
reduce the density of interface states responsible for the surface recombination. These results are
interesting for the fabrication of high efficiency silicon solar cells in photovoltaics. VC 2011
American Institute of Physics. [doi:10.1063/1.3616145]
Surface passivation of crystalline silicon is becoming a
decisive factor for the efficiency of silicon solar cells. The
reason is that the electronic states induced by surface dan-
gling bonds can recombine the carriers via a Schockley-
Read-Hall (SRH) process and therefore increase the surface
titatively investigated. The results are of interest for passiva-
tion engineering of high efficiency silicon solar cells.
The starting material was 500 lm thick p-type h100i
Czochralski silicon wafers, with a boron concentration of
15
3
1.25 ꢁ 10 /cm . After chemical-mechanical double-surface
polishing, the samples were subjected to standard Radio
Corporation of America (RCA) cleaning and hydrofluric acid
(HF) dipping. A 35 nm thick Al O thin film was deposited
1
recombination velocity (SRV) of minority carriers. One
standard way of silicon surface passivation is the thermal ox-
idation at elevated temperatures, which has been used for the
fabrication of solar cells with a world record efficiency in the
2
3
on the double surfaces of samples by an ALD system (Cam-
ꢀ
2
lab. However, the high temperature process risks the metal
contamination, degrading the minority carrier lifetime. The
bridgenanotech, S200) at 200 C, with the trimethylaluminium
(Al(CH ) ) and water vapor as reactant gas. Then, the samples
3
3
amorphous hydrogenated silicon nitride (a-SiN :H) thin
x
were subjected to a RTP post-anneal for 30 s at 450, 550,
ꢀ
3
film cannot only be used as an anti-reflective film for solar
650 C with nitrogen ambient protection. High resolution
transmission electron microscopy (HRTEM) was used to
observe the Al O /Si interface. The effective carrier lifetimes
cells but also passivate the surface of n-type silicon. It is
mainly associated with the modulation of the energy band
bending by the positive charges in the a-SiN :H film. But,
2
3
(s) of samples were measured by a microwave photoconduc-
tance decay (MW-PCD) system (WT2000, Semilab) before
and after RTP treatments. Meanwhile, the capacitance-voltage
(C-V) measurements were performed on the samples by a
Keithly 4200 system after preparing a 1 mm diameter Al dot
on the Al O film by thermal evaporation and scratching In/
x
the passivation of a-SiN :H film is less efficient for the sur-
x
face of p-type silicon, not suitable for the fabrication of high
4
efficiency silicon solar cell with double surface passivation.
Recently, it has been recognized that the atomic-layer-depos-
ited (ALD) Al O thin film can be used as a more efficient
2
3
2 3
5–7
passivation film for both p-type and n-type silicon surface.
Benick et al. has fabricated the silicon solar cells with an ef-
ficiency of 23.2% by employing Al O passivation technol-
Ga eutectic solution at the backside.
Figure 1 shows the cross-sectional HRTEM diagrams of
ꢀ
the Al O /Si interfaces before and after the 550 C RTP
2 3
2
3
8
ogy. The Al O passivation efficiency is strongly dependent
treatment. Note that a SiO interlayer is present for both
x
2
3
on the structure of Al O /Si interface. A silicon oxide (SiO )
2
samples. The thickness of SiO interlayer in the as-deposited
x
3
x
layer formed at the interface might play a critical role in the
origin of the negative charge.
sample is about ꢂ1.5 nm, resulting from the natural oxida-
9
,10
However, it is still neces-
sary to understand the mechanism about the role of interfa-
cial SiO layer in the Al O passivation of silicon surface.
tion of the silicon prior to the Al O deposition. After the
3
2
x
2 3
The purpose of this work is to understand the influence
of SiO interlayer on the efficiency of Al O thin film passi-
vating the silicon surface. After depositing the Al O thin
x
2 3
2
3
films on the silicon surface, rapid thermal process (RTP)
treatments were employed to form an interfacial SiO layer
x
between Al O and silicon. The evolution of carrier lifetimes
3
2
and the density of interface states in the samples were quan-
FIG. 1. (Color online) High resolution TEM diagrams of Al
ꢀ
2
O
3
/Si interface.
a)
Electronic mail: yuxuegong@zju.edu.cn.
(a) As-deposited and (b) 550 C RTP.
0003-6951/2011/99(5)/052103/3/$30.00
99, 052103-1
VC 2011 American Institute of Physics
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
28.233.210.97 On: Wed, 15 Oct 2014 04:41:40
1

052103-2
Lei et al.
Appl. Phys. Lett. 99, 052103 (2011)
FIG. 2. (Color online) (a) High fre-
quency C-V curve of different samples
and the correspondingly simulated
curves and (b) Energy band diagram of
the samples.
RTP treatment, the SiO interlayer becomes much thicker
x
density in the neutral region. Note that Qs is positive for ws
and increases to ꢂ4 nm. This increase in the thickness of the
> 0, and it is negative for w < 0. If the silicon surface is not
s
SiO interlayer should be attributed to the silicon oxidation
x
in strong inversion, the silicon surface capacitance (C ) can
s
11
during the RTP treatment.
Figure 2(a) shows the high-frequency C-V characteris-
be expressed as
ꢀ
dQs
tics of the samples before and after 550 C RTP treatment.
s
C ¼
dws
Note that the C-V curves show three characteristic regions
for both samples, i.e., accumulation, depletion, and inver-
sion. It can be seen that the flat-band voltage in the C-V
curve of the as-deposited sample is negative, indicating that
the net charges stored at the interface are positive. However,
the flat-band voltage becomes positive after the RTP treat-
ment, indicating the type of charge in the Al O film has
es 1 ꢄ expðꢄbw Þ þ ðnpo=ppoÞ½expðbw Þ ꢄ 1ꢅ
s
s
¼
pffiffi
:
2
LD
Fðbws; npo=ppoÞ
(5)
However, when the silicon is in strong inversion, the genera-
tion and recombination rate in the inversion layer cannot
keep up with high frequency. The capacitance can be
2
3
been changed from positive to negative. Figure 2(b) shows
the energy band diagram of the sample under a bias voltage
U. The applied voltage is divided into three parts (see V1,
V , and V in Fig. 2(b)) distributed on the Al O , SiO , and
12
expressed by
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
es
4eskTlnðNA=niÞ
2
3
2
3
2
Cs ¼
ꢆ
;
(6)
2
silicon, respectively,
Wmax
q NA
qV þ qV þ qV ¼ qU þ DW;
(1)
1
2
3
where Wmax is the maximum width of depletion region in the
silicon, N is the boron concentration and n is the intrinsic
A
i
where q is the elemental charge and DW is the difference of
carrier concentration in silicon. The total capacitance, Ctotal,
consisting of Cs, CSiO , and CAl O , can be expressed as
work functions between Al and silicon. V is actually equal
3
2
2
3
to the silicon surface potential, w . Based on the Gauss theo-
s
1
1
rem, one can also obtain the following equations:
1
1
1
1
¼
þ
þ
:
(7)
ꢀ
ꢁ
Ctotal CAl2O3 CSiO2 Cs
Qs þ qD V3
it
V2 ¼
d2;
(2)
(3)
esiO
2
Based on Eqs. (1)–(7), the high-frequency C-V curves of our
sample can be simulated, see the solid line in Fig. 2(a), and
then the values of Qfixed and Dit can be derived. It shows that
the charge in the as-deposited Al O film is indeed positive
ꢀ
ꢁ
Q þ qD V þ Q
s
it
3
fix
V1 ¼
d1;
2
ꢄ2
3
eAl O
2
3
12
with a density of 3.4 ꢁ 10 cm . However, the type of
ꢀ
charge becomes negative after 550 C RTP treatment, with a
where Q is the charge density in silicon, e
is the SiO2
s
siO2
12
ꢄ2
value of 1.7 ꢁ 10 cm . The SiO interlayer should be re-
2
permittivity, Dit is the density of interface states, d2 is the
sponsible for the evolution of the charges in the dielectric
SiO thickness, Qfix is the fixed charges in Al O film, eAl O
2
2
3
2
3
9
film. The negative charges are usually present at unique tet-
is the Al O permittivity and d is the Al O thickness.
2
3
1
2 3
rahedrally coordinated Al site at the interface, in contrast to
the octahedral coordination site where Al has a charge of
According to the Poisson’s equation, the value of Qs can be
1
1
further obtained by using the F function
pffiffi
13
þ3. Therefore, the interfacial negative charges could be
14
ꢀ
ꢁ
obtained by the local reconstruction of Al O at the SiO /
2
2
eskT
npo
ppo
3
2
Qs ¼ ꢃ
F bw ;
;
(4)
s
Al O interface due to the RTP treatment. More interest-
2 3
qL
D
ingly, the density of states at the interface in the sample with
qffiffiffiffiffiffiffiffi
ꢀ
11
ꢄ2
ꢄ1
the 550 C RTP treatment is 7.0 ꢁ 10 cm eV , much
smaller than that in the as-deposited sample. It is believed
that the hydrogen atoms contained in the Al O thin films
q
kT
es
qp
po
where b ¼ , LD ¼
b, es are the silicon permittivity, k
is the Boltzmann constant, n , p are the electron and hole
po
po
2 3
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
28.233.210.97 On: Wed, 15 Oct 2014 04:41:40
1

052103-3
Lei et al.
Appl. Phys. Lett. 99, 052103 (2011)
ꢀ
decreases, e.g., 650 C. Obviously, the s values in the sam-
ples after the RTP treatments are dependent on the surface
recombination, determined by the charges in the Al O film
2
3
and the density of interface states.
In summary, we have quantitatively investigated the
effect of an interfacial silicon oxide layer on the passivation
efficiency of Al O thin films at the silicon surface. It is
2
3
found that the RTP temperature for the formation of silicon
oxide layer is critical for the silicon surface passivation. It
will not only change the type of charges in the Al O thin
2
3
film from positive to negative but also reduce the density of
interface states. The effective carrier lifetime of samples
therefore can be significantly increased due to both the field
effect and hydrogenation. Optimizing the RTP scheme will
be the future interesting work for a best passivation of Al O3
2
thin film on the surface of photovoltaic silicon.
This work is supported by National Natural Science
Foundation of China (Nos. 60906002 and 50832006), “973
Program” (No. 2007CB613403), and Research Fund for
Doctoral Program of Higher Education of China (No.
20090101120165).
FIG. 3. Evolution of (a) interface fixed charge, (b) density of interface
states, and (c) effective carrier lifetime for the samples annealed at different
temperatures.
1
W. Shockley and W. T. Read, Phys. Rev. 87, 835 (1952).
J. Zhao, A. Wang, P. Campbell, and M. Green, in Proceedings of the 26th
2
15
may diffuse to the interface and reduce the surface states.
The thickening of SiO interlayer should also be beneficial
IEEE PVSC (Anaheim, CA, USA, 1997), p. 1133.
T. Lauinger, J. Schmidt, A. Aberle, and R. Hezel, Appl. Phys. Lett. 68,
3
x
1
232 (1996).
for the reduction of interface states since the SiO itself has
x
4
5
6
7
8
9
0
1
2
S. Dauwe, L. Mittelstadt, A. Metz, and R. Hezel, Prog. Photovoltaics 10,
17 (2002).
strong passivation effect on the silicon surface.
2
Figures 3(a) and 3(b) show the values of Qfixed and Dit
obtained for the samples subjected to different RTP treat-
ments. It can be seen that the density of positive charges
decreases with an increase of the annealing temperature and
K. Jaeger and R. Hezel, in Proceedings of the 18th IEEE Photovoltaic Spe-
cialist Conference, Las Vegas (IEEE, Piscataway, NJ, 1985), p. 578.
G. Agostinelli, A. Delabie, P. Vitanov, Z. Alexieva, H. Dekkers, S. Wolf,
and G. Beaucarne, Sol. Energy Mater. Sol. Cells 90, 3438 (2006).
B. Hoex, S. Heil, E. Langereis, M. van de Sanden, and W. Kessels, Appl.
Phys. Lett. 89, 042112 (2006).
J. Benick, B. Hoex, M. van de Sanden, W. Kessels, and O. Schultz, Appl.
Phys. Lett. 92, 253504 (2008).
B. Hoex, J. Gielis, M. van de Sanden, and W. Kessels, J. Appl. Phys. 104,
ꢀ
there is a change of charge type at 550 C. The density of
interface states also decreases with an increase of annealing
ꢀ
temperature up to 550 C. This indicates that a relatively
high temperature is beneficial for the reconstruction of
Al O at the SiO /Al O interface as well as hydrogen diffu-
1
13703 (2008).
1
1
1
K. Kimoto, Y. Matsui, T. Nabatame. T. Yasuda. T. Mizoguchi, I. Tanaka,
and A. Toriumi, Appl. Phys. Lett. 83, 4306 (2003).
S. M. Sze and K. K. Ng, Physics of Semiconductor Devices, 3rd ed.
2
3
2
2 3
ꢀ
sion. However, it is unexpected that in the case of 650 C
RTP annealing, the charge type in the Al O becomes posi-
tive and the density of interface states increases. The reason
2
3
(
Wiley, Hoboken, NJ, 2007), p. 200.
C. Garrett and W. Brattain, Phys. Rev. 99, 376 (1955).
13
R. Johnson, G. Lucovsky, and I. Baumvol, J. Vac. Sci. Technol. A 19,
1
6
could be related to the formation of silicate, destroying the
structures responsible for negative charges. But, this point
still needs more experiments for further understanding. Fig-
ure 3(c) shows the variation of s value for the samples
annealed at different temperatures. It can be seen that the
value of s in the samples can significantly be improved
by the RTP treatment until it reaches the maximum value
1353 (2001).
1
4
G. Lucovsky, J. C. Phyllips, and M. F. Thorpe, in Proceedings of the
Characterisation and Metrology for USLI Technology, edited by D. G.
Seiler, A. C. Diebold, T. J. Shaffner, R. McDonald, W. M. Bullis, P. J.
Smith, and E. M. Secula, AIP Conf. Proc. No. 550 (AIP, New York,
2001), p. 154.
1
5
H. M a¨ ckel and R. L u¨ demann, J. Appl. Phys. 92, 2602 (2008).
M. Cho, H. Park, J. Park, C. Hwang, J. Lee, K. Hyun, H. Kang, Y. Kim,
and J. Lee, J. Appl. Phys. 94, 2563 (2003).
16
ꢀ
at 550 C. Afterwards, the carrier lifetime of the sample
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
28.233.210.97 On: Wed, 15 Oct 2014 04:41:40
1