Impact of fixed charge on metal-insulator-semiconductor barrier height reduction

Article abstract of DOI:10.1063/1.3669414

Recently, the insertion of ultrathin insulators to form metal-insulator-semiconductor (MIS) contacts has been used extensively to reduce the Schottky barrier height and to shift the Fermi level pinning. In this paper, we investigate the physical non-idealities of the ultrathin insulator in Al/Al₂O₃/n-GaAs MIS through stoichiometry, density, and bandgap measurements. These structural non-idealities electrically manifest as bulk and interface fixed charges that are found to contribute to the observed barrier height reduction. The effect of fixed charge has not been considered before, and when combined with the previously reported interface dipoles, it provides a more thorough understanding of the MIS contacts.

Full text of DOI:10.1063/1.3669414

Impact of fixed charge on metal-insulator-semiconductor barrier height reduction
Jenny Hu, Aneesh Nainani, Yun Sun, Krishna C. Saraswat, and H.-S. Philip Wong
Citation: Applied Physics Letters 99, 252104 (2011); doi: 10.1063/1.3669414
View online: http://dx.doi.org/10.1063/1.3669414
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/99/25?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Antimony and bismuth passivations of InP and characterizations of InP metal-insulator-semiconductor structures
fabricated by plasma oxidation of Sb- and Bi-passivated InP
J. Vac. Sci. Technol. A 21, 226 (2003); 10.1116/1.1531135
Spin-polarized light-emitting diode using metal/insulator/semiconductor structures
Appl. Phys. Lett. 81, 694 (2002); 10.1063/1.1496493
Passivation of GaAs metal–insulator–semiconductor structures by ( NH 4 ) 2 S x and by evaporation of SiO 2
J. Vac. Sci. Technol. A 20, 1154 (2002); 10.1116/1.1463078
Characteristics of an indium antimonide metal-insulator-semiconductor structure prepared by remote plasma
enhanced chemical vapor deposition
J. Appl. Phys. 82, 3917 (1997); 10.1063/1.365697
On the inversion in GaAs metal-insulator-semiconductor heterostructures
Appl. Phys. Lett. 70, 228 (1997); 10.1063/1.118374
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:
149.150.51.237 On: Tue, 16 Sep 2014 19:12:01

APPLIED PHYSICS LETTERS 99, 252104 (2011)
Impact of fixed charge on metal-insulator-semiconductor barrier
height reduction
Jenny Hu,1,a) Aneesh Nainani,1 Yun Sun,2 Krishna C. Saraswat,1 and H.-S. Philip Wong1
1Department of Electrical Engineering, Paul G. Allen Center for Integrated Systems, Stanford University,
Stanford, California 94305, USA
2Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, Menlo Park,
California 94025, USA
(Received 31 August 2011; accepted 22 November 2011; published online 20 December 2011)
Recently, the insertion of ultrathin insulators to form metal-insulator-semiconductor (MIS)
contacts has been used extensively to reduce the Schottky barrier height and to shift the Fermi level
pinning. In this paper, we investigate the physical non-idealities of the ultrathin insulator in
Al/Al2O3/n-GaAs MIS through stoichiometry, density, and bandgap measurements. These structural
non-idealities electrically manifest as bulk and interface fixed charges that are found to contribute
to the observed barrier height reduction. The effect of fixed charge has not been considered before,
and when combined with the previously reported interface dipoles, it provides a more thorough
C
V
understanding of the MIS contacts. 2011 American Institute of Physics. [doi:10.1063/1.3669414]
Recently, the insertion of ultrathin insulators in between
the metal and semiconductor interfaces has been experimen-
tally found to significantly reduce the Schottky barrier height
and contact resistance in Si,1,2 Ge,3–7 GaAs,8 InGaAs,8,9 and
GaSb (Ref. 10) semiconductors (Fig. 1(a)). These effective
barrier height (UB,eff) reductions in the metal-insulator-semi-
conductor (MIS) contacts were achieved using a variety of
high-j dielectrics SiN,1,4,8,9 AlOx,5,7,8 GeOx,5 Ge3N4,6 and
TiO2 (Refs. 3 and 10) and appeared to be independent of the
deposition method. However, these reported results have
been from as-deposited samples on lightly doped substrates
where the effect of thermal annealing has not been studied.
Furthermore, there is not a definitive consensus in the
reported literature on understanding of the underlying physi-
cal mechanisms behind the observed barrier height reduc-
tions. Two commonly cited explanations are (1) Fermi-level
depinning by using the insulators to reduce the penetration
of metal induced gap states (MIGS) (Fig. 1(a)) and (2)
Fermi-level shifting through the formation of electronic
dipoles at the insulator/semiconductor interfaces (Fig. 1(b)).
In efforts to elucidate the UB,eff reduction, the MIS contact
has been modeled and compared with reported experimental
results.11,12 Lin et al. modeled the mechanism as a combina-
tion of Fermi level depinning and the presence of an interfa-
cial dipole due to polar/non-polar surfaces,11 while Wager
and Robertson used the MIGS theory to model two dipoles
at both the metal/insulator and insulator/semiconductor inter-
faces.12 These explanations provide the basis for an initial
understanding of the experimental observations. But the
dielectrics in these theoretical works were assumed to be
ideal, crystalline structures. In reality, these dielectrics are
non-ideal, amorphous materials that can be non-
stoichiometric, with a significant amount of defects and
impurities. In this work, we consider the non-idealities of the
dielectric properties in Al/Al2O3/n-GaAs MIS contacts, and
through thermal annealing, demonstrate that the presence of
bulk fixed charge in the dielectric can partly explain the
effective barrier height reduction in MIS contacts. The effect
of fixed charge has not been considered before, and in com-
bination with the existence of interface dipoles, it provides a
more thorough understanding of the MIS contacts.
The design of experiments to study the dielectric proper-
ties included three types of samples: (1) Bare Al2O3 on
GaAs to measure the film stoichiometry, density, and band
offsets, (2) Pt/Al2O3/n-InGaAs metal-oxide-semiconductor
capacitors (MOSCAPs) to extract fixed charge (QF), and (3)
Al/Al2O3/n-GaAs MIS contacts to evaluate the contact re-
sistance (RC) and the effective barrier height. The bare
Al2O3 on GaAs samples were fabricated similarly to the MIS
contacts, with the only difference being the thickness of the
oxide and absence of a metal electrode. To begin, the lightly
doped (2 ꢀ 1016 cmꢁ3) epitaxially grown n-GaAs substrates
underwent organics degrease, HCl native oxide removal, and
(NH4)2S sulfur passivation. The samples were then immedi-
ately transferred to the atomic layer deposition (ALD) cham-
ber for Al2O3 deposition at 250-300 ꢂC using H2O and
trimethylaluminum (TMA), with TMA as the starting pulse
to minimize native oxide formation.13 For the MIS contacts,
after ALD, the electrodes were defined by Al metal evapora-
tion through shadow masks. The MOSCAPs were fabricated
on epitaxially grown 40 nm n-In0.53Ga0.47As (1 ꢀ 1018 cmꢁ3
)
on nþ-InP substrates. The samples were organically
degreased and pre-cleaned with NH4OH immediately prior
to ALD Al2O3 deposition of 50-100 cycles at 300 ꢂC. Pt top
electrodes were then e-beam evaporated through shadow
masks, followed by deposition of Ti/Pt backside ohmic
contacts.
In evaluating the bare Al2O3 samples through x-ray pho-
toelectron spectroscopy (XPS), the Al2O3 was found to be
consistently non-stoichiometric throughout the film. Instead
of the ideal crystalline O to Al ratio of 1.5, the ratio was
measured to be 1.82 with 35.4% Al 2p and 64.6% O 1s
atomic concentrations. The film density was measured to be
a)Author to whom correspondence should be addressed. Electronic mail:
jennyhu@stanford.edu.
C
V
0003-6951/2011/99(25)/252104/4/$30.00
99, 252104-1
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:
149.150.51.237 On: Tue, 16 Sep 2014 19:12:01

252104-2
Hu et al.
Appl. Phys. Lett. 99, 252104 (2011)
VB GaAs
VB GaAs
(d)
hhυν == 112200eeVV (e)
(a)
(b)
(f)
(c)
Tunneling
States
∆E
Tunneling
Current
∆EC
Barrier Height
Dominates
Tunneling
Resistance
1.8 CeV
VB Al O /GaAs
Al2p
Al2p
VB Al2O3 /GaAs
2
3
Φ
ΦB,eff
Φ
ΦB
,eff
B,eff
B,eff
(RSB
)
Dominates (RT)
Eg
Eg
6.4eV
EG
EG
+-
Blocked
Deep States
1.4 eV
Metal
Metal
Optimal
Thickness
Semi-
conductor
∆E
∆EVV
6.4eV
3.2eV
3.2 eV
Semiconductor
-2
0
2
4
70 75 80 85
Al2O3 GaAs
Al2O3 GaAs
Insulator
Binding Energy (eV)
Insulator
Insulator Thickness
FIG. 1. (Color online) (a) RC in MIS contacts. There exists an optimal thickness for minimal RC, arising from the tradeoff between a reduced barrier (RSB) and
an increased tunneling resistance (RT). (b) Illustration of a reduction in the barrier height by minimizing the penetration of metal induced gap states. (c) Band
diagram showing a shift in the Fermi level through the formation of an electronic dielectric dipole at the insulator/semiconductor interface. (d) Al 2p peak
SRPES spectrum of 10 nm Al2O3 taken using 120 eV photon energy. The energy loss spectrum indicates a bandgap of 6.4 eV. (e) Valence band spectra of
Al2O3/GaAs and the GaAs substrate, illustrating a DEV of 3.2 eV. (f) Summary of the band offset values.
3.04 g/cm3 by x-ray reflectivity (XRR), significantly lower
than the 3.9 g/cm3 of bulk crystalline14 Al2O3. Synchrotron
radiation photoemission spectroscopy (SRPES) was then
used to determine the band alignment of Al2O3/GaAs (Figs.
1(d)–1(f)). The SRPES experiments were performed at
beamline 8-1 of the Stanford Synchrotron Radiation Light-
source, which provides a tunable range of monochromatic
photons up to 160 eV, yielding high surface sensitivity with
ferent electrical behavior of MIS contacts. These non-
idealities can be related through the presence of defects that
leave the film oxygen rich and electrically behave as fixed
charge. Since it is not possible to extract QF in ultrathin MIS,
MOSCAPs with thicker dielectric are used as a vehicle to
extract these values.
To investigate the electrical non-idealities, such as the
fixed charge present at the insulator-semiconductor interface
and inside the bulk of the film, capacitance-voltage (C-V)
characteristics of as-deposited Pt/Al2O3/n-InGaAs MOS-
CAPs with 50, 75, and 100 cycles of ALD Al2O3 were meas-
ured (Fig. 2). The film was found to have a significant
amount of positive bulk fixed charge (QF ¼ 1.9 ꢀ 1019 cmꢁ3),
as indicated by the oxide thickness dependent flatband volt-
age (VFB) shift. Forming gas ambient (FGA) annealing was
reported17 to passivate the dangling bonds responsible for QF
in ALD Al2O3. These results were verified through our sam-
ples, where after a FGA anneal at 300 ꢂC for 30 min, the
fixed charges were minimized (QF ¼ 4.1 ꢀ 1018 cmꢁ3), and
there was an alignment of the VFB across all thicknesses.
˚
a minimum electron escape depth of 5 A. A valence band
offset of DEV ¼ 3.2 eV was extracted from the difference in
the valence band (VB) spectra of Al2O3/GaAs and GaAs
samples after aligning the Ga 3d reference peaks. From the
Al 2p energy loss spectrum, the Al2O3 band gap was meas-
ured to be 6.4 eV, and the Al2O3/GaAs conduction band off-
set was calculated to be DEC ¼ 1.8 eV. As with the
stoichiometry and density, the measured bandgap is lower
than the ideal crystalline value15 of 8.7 eV. Though the
bandgap is less than the ideal value, it is comparable to those
measured by other researchers16 for similarly deposited ALD
Al2O3.
These structural non-idealities of the film: oxygen rich
stoichiometry, low density, and smaller bandgap indicate
that the material properties are difficult to model, whereby
use of the ideal crystalline properties may result in very dif-
103
FGA Annealed
102
101
100
10-1
(a) As-Deposited
(b) After FGA Annealing
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
t = 4.7 nm
t = 4.7 nm
t = 6.7 nm
t = 6.7 nm
t = 9.2 nm
t = 9.2 nm
f = 1 MHz
f = 1 MHz
10-2
As Deposited
10-3
0.0
0.5
1.0
1.5
2.0
Al2O3 Thickness (nm)
Ideal V
Ideal VFFBB
FIG. 3. (Color online) Al/Al2O3/n-GaAs MIS RC versus insulator thickness
(tINS) measured using the contact end resistance method (see Ref. 18), where
RC is expressed as a ratio taken relative to the Schottky diode case. As de-
posited, the MIS contact has an optimal tINS for minimal RC, which implies
a reduction in the GaAs UB,eff. However, after removing the positive fixed
charge by annealing in FGA at 300 ꢂC for 15 min, there is no longer a
reduced barrier height, and the tunneling resistance branch dominates. We
note that the Al did not appear to spike through the ultrathin Al2O3, because
if it did, the contacts should have a Schottky behavior independent of the
Al2O3 thickness.
-3 -2 -1
0
1
2
-3 -2 -1
0
1
2
3
Voltage (V)
FIG. 2. (Color online) (a) C-V for Pt/Al2O3/n-InGaAs with 50, 70, and 100
cycles of ALD. As deposited, there is a VFB shift with thickness, indicating
the presence of significant oxide charge. (b) After an FGA anneal at 300 ꢂC
for 30 min, the fixed charge is minimized and there is VFB alignment across
all oxide thicknesses.
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:
149.150.51.237 On: Tue, 16 Sep 2014 19:12:01

252104-3
Hu et al.
Appl. Phys. Lett. 99, 252104 (2011)
charge.19 In comparing the 250 ꢂC and 300 ꢂC samples (Fig.
4), the 250 ꢂC film with higher QF achieved an overall lower
RC. This further supports the idea of QF modulating the bar-
rier height on the semiconductor side. However, it is unlikely
that all the observed MIS contact behaviors reported in liter-
ature are due to positive/negative bulk or interface fixed
charge, since materials can have a range of defects and
impurities. The overall mechanism of MIS contacts can be a
combination of fixed charge and dipoles, dependent upon the
specific dielectric material used.
Band diagrams are used to illustrate the overall effect of
dipoles and fixed charge on Schottky (Fig. 5(a)) and MIS
contacts (Fig. 5(b)). The current reported MIS interface
dipole models2,8 can explain the observed reduction in UB,eff
by modeling the dipoles as changes in the vacuum level (Fig.
5(c)), where in the case of extremely large dipole magni-
tudes, the surface will be in accumulation (Fig. 5(d)). How-
ever, this same effect on UB,eff can also be achieved using
positive QF to increase the amount of potential that is
dropped across the insulator to reduce the semiconductor
depletion charge (Fig. 5(e)). In the case of extremely large
amount of positive QF, electrons can be induced to the sur-
face, bringing the semiconductor into accumulation (Fig.
5(f)). In this case, the tunneling distance through the semi-
conductor is greatly reduced from the Schottky case particu-
larly for low doped substrates, further increasing the current
and reducing RC. From the known non-idealities in the
dielectric, the Al/Al2O3/GaAs MIS behavior discussed in
this work appears to be due to the presence of QF. However,
the overall effect of dipoles and fixed charge can be addi-
tively or subtractively combined, depending on the direction
of the dipole and the sign of QF, so the MIS behavior can
also be due to a combination of the two (Fig. 5(g)).
FIG. 4. (Color online) Al/Al2O3/n-GaAs RC vs. tINS comparing samples
with ALD deposition temperatures of 250 and 300 ꢂC. Lower temperature
ALD Al2O3 results in a larger DVFB shift with oxide thickness in MOS-
CAPs, which indicates a larger amount of fixed charge (Ref. 19). The lower
RC in the 250 ꢂC sample with more fixed charge further verifies the impor-
tance of positive fixed charge in the UB,eff reduction on n-GaAs.
To understand how the positive fixed charge in the oxide
would affect the MIS contacts, the MIS was also annealed in
FGA to remove QF. RCs were measured before and after
annealing using the contact end resistance method.18 Fig. 3
illustrates how the RC vs. Al2O3 thickness tradeoff changes
with annealing. As deposited, there exists an optimal thick-
8
ness for minimal RC which indicates a reduction in UB,eff
.
After FGA, the MIS contact behavior changed entirely.
There is no longer a decrease in RC which implies that there
is also no reduction of UB,eff in the annealed MIS. The mono-
tonic increase of RC after annealing is due to the tunneling li-
mitation the oxide imposes on the current conduction.8 We
note that the slope of the tunneling resistance limited branch
in the RC trade-off is less steep after annealing, possibly due
to a change in the tunneling barrier either by changes in the
band structure or densification of the film. The presence of
fixed charge appears to be responsible for the observed
reduction in the barrier height using ultrathin insulators.
To verify the effect of QF, MIS contacts were fabricated
using different ALD Al2O3 deposition temperatures, where
lower temperature film results in more positive fixed
In summary, we introduced the concept of fixed charge
being partially or fully responsible for the observed UB,eff
reduction in MIS contacts. In the case of Al/Al2O3/n-GaAs,
it is believed to be due to a large concentration of positive
fixed charge. Evaluation of the film stoichiometry, density,
bandgap, and fixed charge demonstrated that the ALD Al2O3
is unlike the ideal crystalline properties, making the MIS
behavior difficult to model accurately. The combined effect
FIG. 5. (Color online) Schematic band diagrams illustrating how the presence of electronic dipoles and fixed charge can affect UB,eff. (a) A Schottky barrier
with a large depletion width in the semiconductor. (b) In the MIS structure, some of the potential is dropped across the insulator, which reduces the depletion
width and semiconductor barrier height. (c) A small electronic dipole at the insulator/semiconductor interface can cause a change in the vacuum level. (d) A
larger dipole can in fact push the semiconductor surface to be in accumulation. (e) With fixed charge in the oxide, more potential is dropped across the insula-
tor and the barrier height is reduced. (f) A larger amount of fixed charge amplifies the effect and can induce negative charge in the semiconductor. (g) If both
fixed charge and interface dipole are present, then there is a further reduction of the barrier height.
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:
149.150.51.237 On: Tue, 16 Sep 2014 19:12:01

252104-4
Hu et al.
Appl. Phys. Lett. 99, 252104 (2011)
4M. Kobayashi, A. Kinoshita, K. Saraswat, H.-S. P. Wong, and Y. Nishi,
J. Appl. Phys. 105, 023702 (2009).
of fixed charge and electronic dipoles provides a more com-
plete understanding of the MIS contacts. We have also found
that FGA annealing, which is commonly used in CMOS fab-
rication, can drastically increase the RC of MIS contacts.
Further study on highly doped substrates and the allowed
thermal budget are necessary in order to understand how
these contacts could be integrated into CMOS. Though this
study was based on III-V MIS, the results can be extended to
the understanding of Si and Ge MIS contacts.
5T. Nishimura, K. Kita, and A. Toriumi, Appl. Phys. Express 1, 051406
(2008).
6R. R. Lieten, S. Degroote, M. Kuijk, and G. Borghs, Appl. Phys. Lett. 92,
022106 (2008).
7Y. Zhou, M. Ogawa, X. Han, and K. L. Wang, Appl. Phys. Lett. 93,
202105 (2008).
8J. Hu, K. C. Saraswat, and H.-S. P. Wong, J. Appl. Phys. 107, 063712
(2010).
9J. Hu, K. C. Saraswat, and H.-S. P. Wong, Appl. Phys. Lett. 98, 062107
(2011).
10Z. Yuan, A. Nainani, Y. Sun, J.-Y. J. Lin, P. Pianetta, and K. C. Saraswat,
Appl. Phys. Lett. 98, 172106 (2011).
The authors would like to thank Dr. Arturas Vailionis
for his help with the XRR measurements. This work is sup-
ported in part by the Focus Center Research Program
(MSD), Intel Corporation, and member companies of the Ini-
tiative for Nanoscale Materials and Processes (INMP) at
Stanford. J. Hu is additionally supported by the Intel Ph.D.
Fellowship, Stanford Graduate Fellowship, and the National
Defense Science and Engineering Graduate (NDSEG)
Fellowship.
11L. Lin, J. Robertson, and S. J. Clark, Microelectron. Eng. 88, 1461
(2011).
12J. F. Wager and J. Robertson, J. Appl. Phys. 109, 094501 (2011).
13C. L. Hinkle, A. M. Sonnet, E. M. Vogel, S. McDonnell, G. J. Hughes,
M. Milojevic, B. Lee, F. S. Aguirre-Tostado, K. J. Choi, H. C. Kim,
J. Kim, and R. M. Wallace, Appl. Phys. Lett. 92, 071901 (2008).
14A. Mortensen, Concise Encyclopedia of Composite Materials, 2nd ed.
(Elsevier Science, Amsterdam, The Netherlands, 2007).
15B. Ealet, M. H. Elyakhloufi, E. Gillet, and M. Ricci, Thin Solid Films 250,
92 (1994).
16N. V. Nguyen, M. Xu, O. A. Kirillov, P. D. Ye, C. Wang, K. Cheung, and
J. S. Suehle, Appl. Phys. Lett. 96, 052107 (2010).
17B. Shin, J. R. Weber, R. D. Long, P. K. Hurley, C. G. Van de Walle, and
P. C. McIntyre, Appl. Phys. Lett. 96, 152908 (2010).
18D. K. Schroder, Semiconductor Material and Device Characterization, 3rd
ed. (Wiley-Interscience, New Jersey, 2006).
1D. Connelly, C. Faulkner, P. A. Clifton, and D. E. Grupp, Appl. Phys.
Lett. 88, 012105 (2006).
2B. E. Coss, W.-Y. Loh, R. M. Wallace, J. Kim, P. Majhi, and R. Jammy,
Appl. Phys. Lett. 95, 222105 (2009).
3J. -Y. J. Lin, A. M. Roy, A. Nainani, Y. Sun, and K. C. Saraswat, Appl.
Phys. Lett. 98, 092113 (2011).
19G. Thareja, personal communication (2010).
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:
149.150.51.237 On: Tue, 16 Sep 2014 19:12:01