Electrical Characteristics of n-ZnO/p-Si Heterojunction Diodes
4
Grown by Pulsed Laser Deposition at Different Oxygen
5
Pressures
6
7
R.S. AJIMSHA,
1
M.K. JAYARAJ,
1,3
and L.M. KUKREJA
2
8
1.”Optoelectronics Devices Laboratory, Department of Physics, Cochin University of Science and
9
Technology, Kochi 682022, India. 2.”Laser Materials Processing Division, Raja Ramanna Centre
10
for Advanced Technology, Indore 452013, India. 3.”e-mail: mkj[at]cusat.ac.in
11
Heterojunction diodes of n-type ZnO/p-type silicon (100) were fabricated by
12
pulsed laser deposition of ZnO films on p-Si substrates in oxygen ambient at
13
different pressures. These heterojunctions were found to be rectifying with a
14
maximum forward-to-reverse current ratio of about 1,000 in the applied
15
voltage range of -5 V to +5 V. The turn-on voltage of the heterojunctions was
16
found to depend on the ambient oxygen pressure during the growth of the ZnO
17
film. The current density“voltage characteristics and the variation of the
18
series resistance of the n-ZnO/p-Si heterojunctions were found to be in line
19
with the Anderson model and Burstein-Moss (BM) shift.
20
Key words: Heterojunctions, ZnO, p-Si, pulsed laser deposition
21
22
INTRODUCTION
23
Currently there is significant interest in ZnO as a
24
candidate for various future optoelectronic devices.
25
ZnO is a rugged semiconductor with direct wide
26
band gap and it exhibits significant n-type conduc-
27
tivity even without any intentional doping. This
28
n-type conductivity can be further enhanced by
29
doping with Al or Ga.
1“3
This property and the
30
transparency in the visible spectral region have
31
prompted extensive investigations of ZnO films as
32
transparent electrodes in flat-panel displays,
4
p“n
33
heterojunction diodes,
5“7
thin-film transistors,
8
34
multiple-quantum-well
structures,
9
and
solar
35
cells.
10
Recently we have reported ZnO based all-
36
transparent conducting p“n heterojunction diodes
37
with p-type AgCoO
2
.
11,12
Although ZnO films can be
38
grown by a variety of methods, including radiofre-
39
quency (RF) and direct-current (DC) sputter-
40
ing,
3,13,14
chemical vapor deposition,
15
spray
41
pyrolysis,
16
and electron cyclotron resonance-as-
42
sisted molecular-beam epitaxy,
17
we used pulsed
43
laser deposition (PLD)
1,18,19
to deposit high-quality
44
ZnO films because of its effectiveness and amena-
45
bility to different growth conditions.
20
For the
46
present study we fabricated heterojunctions of
47
n-type ZnO on p-type Si, which has many advanta-
48
ges such as low cost, large wafer size, and the pos-
49
sibility of integrating oxide semiconductors with
50
already highly matured silicon technology.
51
The growth of ZnO on Si substrates has been
52
studied extensively including the epitaxial growth
53
of ZnO on Si (100) substrates,
21
ZnO/p-Si
54
diodes,
22“24
ZnO:N/p-Si heterostructures
25
etc.
55
Studies on the electrical transport properties of
56
ZnO/p-Si heterojunctions with different dopants in
57
the p-Si
26
and ZnO
27
have also been reported
58
recently. However, due to the complex nature of
59
the carrier transport across the interfaces of the
60
n-ZnO/p-Si heterojunction, the transport proper-
61
ties of these heterostructures are not yet well
62
understood and are even debatable. We have fur-
63
thered these studies on n-ZnO/p-Si heterojunction
64
diodes fabricated by pulsed laser deposition at
65
different oxygen pressures. These heterojunction
66
diodes are found to have highly favorable forward-
67
to-reverse current ratio. We have also studied the
68
parametric dependence of the electrical charac-
69
teristics of these heterojunctions. The results of
70
these studies are presented and discussed in this
71
communication.
(Received August 15, 2007; accepted December 3, 2007)
Journal of ELECTRONIC MATERIALS
Special Issue Paper
DOI: 10.1007/s11664-007-0365-4
Ó
2007 TMS
Journal :
11664_JEM
Dispatch :
18-12-2007
Pages : 6
h LE
h TYPESET
Article No.: 365
h CP
h DISK
4
4
Author ProofPage 2

UNCORRECTE
D
PROOF
72
73
EXPERIMENTAL
74
The pulsed laser deposition (PLD) of the ZnO
75
films was carried out in a growth chamber, which
76
was first evacuated to a base pressure of 10
-6
mbar.
77
A polycrystalline, stoichiometric, sintered (for 5 h at
78
1200°C) pellet of ZnO with a purity of 99.999% was
79
used as the target for PLD. The third harmonics
80
(355 nm) of a Q-switched Nd:YAG laser with a
81
repetition rate of 10 Hz, pulse width of 9 ns, and
82
fluence of about 3 J/cm
2
per pulse was used for
83
ablation of the ZnO target. Cleaned p-type silicon
84
(100) wafers with a carrier concentration of about
85
1 · 10
15
cm
-3
were used as substrates. The growth
86
chamber was filled with flowing oxygen ambient
87
and its pressure was varied from 0.003 mbar to
88
0.007 mbar during the growth of different samples.
89
The substrate-to-target distance was kept at about
90
4.5 cm. The ZnO films were deposited for about
91
30 min on the Si substrates at room temperature.
92
To measure the conductivity and band gap of the
93
ZnO films those were separately deposited on silica
94
substrates under identical experimental conditions
95
as those used for the growth on the Si substrates.
96
For electrical measurements, indium metal contacts
97
were made on both p-type silicon surface and n-type
98
ZnO films, which were found to be ohmic in nature.
99
The room-temperature electrical measurements of
100
the ZnO thin films grown on the silica substrates
101
were carried out using the four-probe van der Pau
102
configuration in the Hall geometry.
103
RESULTS AND DISCUSSION
104
The thickness of the deposited ZnO films, mea-
105
sured using a stylus profiler (Dektak 6 M Stylus
106
profiler) was found to be about 250 nm. The X-ray
107
diffraction patterns of all the ZnO films showed only
108
(002) peaks along with the Si (200) peak. A typical
109
XRD pattern of these films is shown in Fig. 1a. This
110
confirmed the highly c-axis-oriented growth of the
111
ZnO films. The full-width at half-maximum
112
(FWHM) of the (002) X-ray diffraction peak of the
113
ZnO films was found to be about 0.34°, indicating a
114
reasonably good crystalline quality of these films.
120
X-ray diffraction pattern of the ZnO films deposited
121
on the silica substrates is shown in Fig. 1b. This
122
also showed only a (002) peak of ZnO, confirming
123
the same c-axis-oriented growth as in the case of
124
ZnO films grown on the p-Si substrates. However
125
the FWHM of this peak was found to be about 0.36°,
126
which is slightly higher than that of the films grown
127
on the Si substrates, as expected.
128
Figure 2a shows the band gap of the ZnO thin
129
films grown on silica substrates, estimated from the
130
plot of (ahm)
2
versus hm. It can be seen from this
131
figure that the band gap decreased from 3.36 eV to
132
3.257 eV with an increase of the oxygen pressure
133
from 0.003 mbar to 0.007 mbar. Series resistance,
134
an inherent resistance of the depletion region in
135
N-ZnO/p-Si heterojunction of all the diodes grown at
136
different oxygen pressures was calculated from the
137
plot of log (I) versus V,
28
which is also shown in
138
Fig. 2a. As can be seen in this figure the series
139
resistance
increased
from
3.45 · 10
5
ohm
to
Fig. 1. XRD pattern of ZnO films deposited on (a) p-silicon (100) and
(b) silica substrates.
Fig. 2. (a) The series resistance and the variation of the optical band
gap with oxygen pressure and (b) a plot of resistivity and mobility
with oxygen pressure.
Ajimsha, Jayaraj, and Kukreja
Journal :
11664_JEM
Dispatch :
18-12-2007
Pages : 6
h LE
h TYPESET
Article No.: 365
h CP
h DISK
4
4
Author ProofPage 3

UNCORRECTE
D
PROOF
145
5.6 · 10
5
ohm with increasing oxygen partial pres-
146
sure from 0.003 mbar to 0.007 mbar. Figure 2b
147
shows the variation of resistivity and the electron
148
mobility for the ZnO thin films with respect to the
149
oxygen pressure. It can be seen from this figure
150
that, while the resistivity increased, the mobility
151
decreased when the oxygen pressure used during
152
the deposition was increased. Hall measurements
153
confirmed the n-type conductivity of the ZnO films.
154
Using these Hall measurements, the carrier con-
155
centration was found to decrease from about
156
3.2 · 10
19
cm
-3
to 1.32 · 10
18
cm
-3
when the oxygen
157
pressure was increased from 0.003 mbar to
158
0.007 mbar, as shown in Fig. 3. A theoretical curve
159
based on the calculated values of the carrier con-
160
centration from the Burstein-Moss (BM) shift
29
is
161
also shown in this figure. With a small gap between
162
the two curves, the trend of experimental data and
163
that of the calculated ones coincide reasonably well.
164
As seen from Fig. 2a the band gap of the ZnO
165
films decreased with increasing oxygen pressure
166
during growth, as did the electron concentration.
167
This means that films grown at lower oxygen pres-
168
sure had a larger band gap due to the enhanced
169
carrier concentration in the film. This increase in
170
the band gap accompanied by an enhanced carrier
171
concentration can be explained using the BM
172
shift.
29
As is well known, this model relies on the
173
effective mass approximation (EMA) in which the
174
wavefunctions are represented by plane waves and
175
the conduction and valance bands are taken to be
176
parabolic near the Brillouin zone. The BM shift in
177
band gap, DE
g
according to this model
29
is given by:
DE
g
¼
h
2
8p
2
1
m
e
þ
1
m
h
3p
2
n
Ã
Á
2=3
(1)
179
180
where m
e
= 0.28 m
e
, m
h
= 0.59 m
e
, h, and n are the
181
effective electron mass, effective hole mass, Planck
182
constant, and electron density per unit volume,
183
respectively.
184
This leads to a total band gap of
E
g
¼ E
go
þ D
E
g
(2)
186
187
We took the band gap of ZnO without BM shift
188
to be E
go
= 3.25 eV, which is that of the ZnO bulk
189
crystal at room temperature.
30
The BM shift in the
190
band gap (DE
g
) was obtained from Eq. 2 using the
191
total band gap (E
g
) estimated from the optical
192
transmission spectra. Then electron concentrations
193
(n) were calculated using Eq. 1. These calculated
194
values of the electron concentration are plotted as
195
a function of the oxygen partial pressure in Fig. 3.
196
Experimental values of the electron concentrations
197
obtained from the Hall measurements are also
198
shown in Fig. 3. It can be seen in this figure that
199
the electron concentrations obtained from the Hall
200
measurements match well with those obtained
201
from the theoretical BM shift except at the lowest
202
oxygen pressure. This might be due to the strain
203
resulting from the increased oxygen vacancies in
204
the film.
205
The physical basis for the concentration of oxygen
206
incorporation in the ZnO films was investigated by
207
X-ray photoelectron spectroscopy (XPS) of the films
208
grown at oxygen pressures of 0.003 mbar and
209
0.007 mbar using an Al K
a
radiation source
210
(1486.6 eV). The results are shown in Fig. 4. The
211
intensity of the oxygen 1s XPS peak showed greater
212
oxygen incorporation in the ZnO films grown at
213
0.007 mbar oxygen pressure. It was also observed
214
from the XPS data that increase of oxygen pressure
215
during deposition enhanced the O/Zn ratio in the ZnO
216
thin films. From the XPS and Hall measurement data
Fig. 3. The variation of the electron concentration in the ZnO films
(obtained from the Hall measurement and theoretical model using
the BM shift) with oxygen pressure.
Fig. 4. XPS of O 1s ZnO thin films deposited at 0.007 mbar and
0.003 mbar oxygen pressures.
Electrical Characteristics of n-ZnO/p-Si Heterojunction Diodes Grown
by Pulsed Laser Deposition at Different Oxygen Pressures
Journal :
11664_JEM
Dispatch :
18-12-2007
Pages : 6
h LE
h TYPESET
Article No.: 365
h CP
h DISK
4
4
Author ProofPage 4

UNCORRECTE
D
PROOF
222
it can be elicited that, the greater the level of oxygen
223
incorporation in the films, the lower the electron
224
concentration. This is also in agreement with the
225
earlier study of Look et al.
31
226
Figure 5 shows the J“V characteristics of five
227
different n-ZnO/p-Si heterojunctions with ZnO films
228
grown at different oxygen pressures. All of the five
229
heterojunctions were found to be rectifying and the
230
turn-on voltage of the heterojunctions increased as
231
shown in the inset of Fig. 5 with increasing oxygen
232
pressure during the growth of the ZnO films. The
233
J“V characteristics of the n-ZnO/p-Si heterojunc-
234
tion diode with the lowest turn-on voltage is plotted
235
on a logarithmic scale in Fig. 6. The maximum for-
236
ward-to-reverse current ratio was found to be about
237
1,000 in the range of applied voltage from -5 V to
238
+5 V. The inset to Fig. 6 shows the ohmic nature of
239
the In/ZnO contact. The room-temperature leakage
240
current at -5 V was of the order of 10
-7
A. The
241
ideality factor was found to be greater than 10 for all
242
the heterojunctions fabricated.
243
The band structure of n-ZnO/p-Si at the hetero-
244
junction can be constructed using the Anderson
245
model
32
by assuming continuity of vacuum levels
246
and neglecting the effects of dipole and interfacial
247
states. A similar band structure has been suggested
248
for doped and pure ZnO/Si heterojunction by P Chen
249
et al.
26,33
Figures 7a and 8 show the constructed
250
band structure of the n-ZnO/p-Si heterojunction
251
fabricated at the 0.007 mbar oxygen pressure under
252
zero and forward bias, respectively. Values of the
253
band gaps of E
g
(ZnO) = 3.257 eV and E
g
(Si) =
254
1.12 eV, and of the electron affinities of v(ZnO) =
255
4.35 eV and v(Si) = 4.05 eV, were used.
26
The
256
valance-band offset (DE
v
) and conduction-band
257
offset (DE
c
) are equal to 2.43 eV and 0.3 eV
Fig. 5. Current density“voltage (J“V) plot of ZnO/p-Si heterojunc-
tions. The inset shows the variation of the turn-on voltage with
oxygen pressure, P(O
2
).
Fig. 6. Current density“voltage (J“V) plot of ZnO/n-Si heterojunc-
tions on a logarithmic scale. The inset shows the current“voltage
(I“V) plot of the In/ZnO contact.
Fig. 7. (a) The band structure of the ZnO/p-Si heterojunction (grown
at 0.007 mbar oxygen pressure) under zero bias. (b) The variation of
DE
v
with oxygen pressure during PLD of ZnO films.
Ajimsha, Jayaraj, and Kukreja
Journal :
11664_JEM
Dispatch :
18-12-2007
Pages : 6
h LE
h TYPESET
Article No.: 365
h CP
h DISK
4
4
Author ProofPage 5

UNCORRECTE
D
PROOF
263
respectively. The variation of DE
v
with oxygen
264
pressure during PLD of ZnO films is shown in
265
Fig. 7b. Both DE
v
and DE
c
are emerging out of the
266
difference in the electron affinities and band gaps of
267
the two materials forming the junction. It can be
268
noted that the valance-band offset DE
v
is much
269
higher than the conduction-band offset DE
c
.
270
Since the carrier concentration in the p-Si side is
271
about 3 orders of magnitude lower than that in the
272
ZnO side, all the depletion region within the p-Si/
273
ZnO heterojunction is extended into the p-Si side.
274
Figure 7a shows that the bottom of the conduction
275
band on the ZnO side lies lower in energy than that
276
on the p-Si side. Hence under relatively low forward
277
bias, the chance of electron flow from the ZnO side
278
to the p-Si side is negligible due to the higher bar-
279
rier difference felt by the electrons at the bottom of
280
the conduction band on the ZnO side. This resulted
281
in a higher turn-on voltage for the p-Si/ZnO junction
282
grown at 0.007 mbar oxygen pressure. However,
283
under higher forward bias, the barrier difference is
284
lowered and the injection of electrons from the bot-
285
tom of the conduction band on the ZnO side to the
286
p-Si increased considerably (as shown in Fig. 8).
287
Thereby the forward current rapidly increased
288
under a higher voltage bias. When the oxygen
289
pressure during the deposition of ZnO was
290
decreased, the carrier concentration increased and
291
hence the Fermi level shifted towards the bottom of
292
the conduction band. This means that, upon
293
decrease of the oxygen pressure, the Fermi level
294
may even move into the conduction band, resulting
295
in the easy flow of electrons from the ZnO side to the
296
p-Si side. Hence the forward voltage required for
297
considerable forward current decreased and thereby
298
the turn-on voltage decreased. This seems to explain
299
the decrease of the turn-on voltage for the n-ZnO/
300
p-Si heterojunction fabricated at the lower oxygen
301
pressure.
302
The variation of the turn-on voltage with oxygen
303
pressure can also be explained with calculated
304
values of series resistance. Due to series resistance,
305
a part of the applied voltage is effectively wasted
306
and hence a larger applied voltage is necessary to
307
achieve the same level of current compared to the
308
ideal value. Hence the turn-on voltage will increase
309
with the increase of series resistance in the quasi-
310
neutral region of p-Si/ZnO. It is noticed that the
311
calculated values of series resistance thus obtained
312
increased with increasing oxygen pressure, thereby
313
increasing the turn-on voltage.
314
CONCLUSION
315
In conclusion c-axis-oriented crystalline ZnO
316
films deposited on p-type Si (100) at different oxy-
317
gen pressures using PLD form effective n-ZnO/p-Si
318
heterojunctions, which were found to be rectifying.
319
The maximum forward-to-reverse current ratio was
320
found to be 1000 in the applied voltage range from
321
-5 V to +5 V. The variation of the turn-on voltage
322
with oxygen pressure was modeled with the
323
Anderson model and the BM shift, which is in
324
agreement with the values of the series resistance
325
calculated across the n-ZnO/p-Si heterojunction.
326
ACKNOWLEDGEMENTS
327
We thank DAE-BRNS for a financial grant to
328
carry out this work. Thanks are also due to Drs.
329
B. N. Singh and P. Misra for their help with this
330
work. We would also like to thank Dr. Shripati from
331
IUC-DAEF, Indore for the XPS measurements.
332
REFERENCES
333
1. D.C. Look, D.C. Reynolds, C.W. Litton, R.L. Jones, D.B. Eason,
334
and G. Cantwell, Appl. Phys. Lett. 81, 1830 (2002).
335
2. M. Hiramatsu, K. Imaeda, N. Horio, and M. Nawata, J. Vac.
336
Sci. Technol. A 16, 669 (1998).
337
3. H. Kim, C.M. Gilmore, J.S. Horwitz, A. Pique, H. Murata,
338
G.P. Kuhsto, R. Schlaf, Z.H. Kafafi, and D.B. Chrisey, Appl.
339
Phys. Lett. 76, 259 (2000).
340
4. C.J. Lee, T.J. Lee, S.C. Lyu, Y. Zhang, H. Ruh, and H.J. Lee,
341
Appl. Phys. Lett. 81, 3648 (2002).
342
5. R.L. Hoffman, J.F. Wager, M.K. Jayaraj, and J. Tate,
343
J. Appl. Phys. 90, 5763 (2001).
344
6. H. Ohta, M. Orita, and M. Hirano, J. Appl. Phys. 89, 5720
345
(2001).
346
7. H. Sato, T. Minami, S. Takata, and T. Yamada, Thin Solid
347
Films 236, 27 (1993).
348
8. R.L. Hoffman, B.J. Norris, and J.F. Wager, Appl. Phys. Lett.
349
82, 733 (2003).
350
9. P. Misra, T.K. Sharma, S. Porwal, and L.M. Kukreja, Appl.
351
Phys. Lett. 89, 161912 (2006).
352
10. J. Aranovich, A. Ortiz, and R.H. Bube, J. Vac. Sci. Technol.
353
16, 994 (1979).
354
11. K.A. Vanaja, R.S. Ajimsha, A.S. Asha, and M.K. Jayaraj,
355
Appl. Phys. Lett. 88, 1 (2006).
356
12. R.S.Ajimsha,K.A. Vanaja,M.K.Jayaraj,P.Misra,V.K.Dixit,
357
and L.M. Kukreja, Thin Solid Films 515, 7352 (2007).
358
13. B. Lin, Z. Fu, and Y. Jia, Appl. Phys. Lett. 79, 943 (2001).
359
14. F.S. Hickernell, J. Appl. Phys. 44, 1061 (1973).
360
15. G.H. Lee, Y. Yamamoto, M. Kourogi, and M. Ohtsu, Thin
361
Solid Films 386, 117 (2001).
362
16. A.S. Riad, S.A. Mahmoud, and A.A. Ibrahim, Physica B 296,
363
319 (2001).
364
17. S.H. Lim, J. Washburn, Z.L. Weber, and D. Shindo, J. Vacc.
365
Sci. Technol. A 19, 2601 (2001).
366
18. X.W. Sun and H.S. Kwok, J. Appl. Phys. 86, 408 (1999).
Fig. 8. The band structure of the ZnO/p-Si heterojunction (grown at
0.007 mbar oxygen pressure) under forward bias.
Electrical Characteristics of n-ZnO/p-Si Heterojunction Diodes Grown
by Pulsed Laser Deposition at Different Oxygen Pressures
Journal :
11664_JEM
Dispatch :
18-12-2007
Pages : 6
h LE
h TYPESET
Article No.: 365
h CP
h DISK
4
4
Author ProofPage 6

UNCORRECTE
D
PROOF
367
19. H. Cao, J.Y. Wu, H.C. Ong, J.Y. Dai, and R.P.H. Chang,
368
Appl. Phys. Lett. 73, 572 (1998).
369
20. P. Misra and L.M. Kukreja, Thin Solid Films. 485, 42
370
(2005).
371
21. S.M. Jejurikar, A.G. Banpurkar, A.V. Limaye, S.K. Date,
372
S.I. Patil, K.P. Adhi, P. Misra, and L.M. Kukreja, J. Appl.
373
Phys. 99, 014907 (2006).
374
22. S.-H. Jeong, B.-S. Kim, and B.-T. Lee, Appl. Phys. Lett. 82,
375
2625 (2003).
376
23. I.S. Jeong, J.H. Kim, and S. Im, Appl. Phys. Lett. 83, 2946
377
(2003).
378
24. Y.S. Choi, J.Y. Lee, S. Im, and S.J. Lee, J. Vac. Sci. Technol.
379
B 20, 2384 (2002).
380
25. J.Y. Lee, Y.S. Choi, J.H. Kim, M.O. Park, and S. Im, Thin
381
Solid Films 403“404, 553 (2002).
382
26. P. Chen, X. Ma, and D. Yang, J. Appl. Phys. 101, 053103
383
(2007).
384
27. X.D. Chen, C.C. Ling, S. Fung, and C.D. Beling, Appl. Phys.
385
Lett. 88, 132104 (2006).
386
28. R.F. Pierret, Semiconductor Device Fundamentals (Addison-
387
Wesley Publishing, 1996).
388
29. H.L. Hartnagel, A.L. Dawar, A.K. Jain, and C. Jagadish,
389
Semiconducting Transparent Thin Films (IOP Publishing
390
Ltd, 1995).
391
30. V. Srikant and D.R. Clarke, J. Appl. Phys. 83, 5447 (1998).
392
31. D.C. Look, J.W. Hemsky, and J.R. Sizelove, Phys. Rev. Lett.
393
82, 2552 (1999), and references cited therein.
394
32. G. Milnes and D.L. Feucht, Heterojunctions and Metal-
395
Semiconductor Junctions (New York: Academic, 1972).
396
33. P.Chen, X.Ma, and D.Yang,J.Appl.Phys.102,083103(2007)