Citation
MSM Photodetectors on Germanium Substrates with SWNT Film Electrodes

Material Information

Title:
MSM Photodetectors on Germanium Substrates with SWNT Film Electrodes
Series Title:
Journal of Undergraduate Research
Creator:
Knauer, Keith
Ural, Ant ( Mentor )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Language:
English

Subjects

Genre:
serial ( sobekcm )

Notes

Abstract:
In this paper, Schottky-like behavior of single-walled carbon nanotube (SWNT) film contacts on Ge is demonstrated by the fabrication and electrical characterization of metal-semiconductor-metal (MSM) photodetectors with SWNT film electrodes. In addition, the effect of device geometry on the dark current is studied by scaling the active area, finger width, and finger spacing of the MSM photodetectors. Finally, the dark current of the SWNT film devices is compared to the dark current of metal control samples, demonstrating that the use of SWNT film electrodes results in a significant reduction in dark current. These experimental results give insight into the electrical properties of the SWNT film-Ge interface while also demonstrating the use of SWNT film as a transparent and conductive material for electronic applications.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.

Downloads

This item has the following downloads:


Full Text



MSM Photodetectors on Germanium Substrates with SWNT Film

Electrodes


Keith Knauer,


College of Engineering, University of Florida


In this paper, Schottky-like behavior of single-walled carbon nanotube (SWNT) film contacts on Ge is demonstrated by the
fabrication and electrical characterization of metal-semiconductor-metal (MSM) photodetectors with SWNT film electrodes. In
addition, the effect of device geometry on the dark current is studied by scaling the active area, finger width, and finger spacing of
the MSM photodetectors. Finally, the dark current of the SWNT film devices is compared to the dark current of metal control
samples, demonstrating that the use of SWNT film electrodes results in a significant reduction in dark current. These experimental
results give insight into the electrical properties of the SWNT film-Ge interface while also demonstrating the use of SWNT film as
a transparent and conductive material for electronic applications.


I. INTRODUCTION


A. Carbon Nanotubes
Carbon nanotubes were discovered in 1991 by Sumio
Iijima at the NEC Tsukuba Research Laboratory using
high resolution transmission electron microscopy to study
the soot from fullerene synthesis by arc discharge [1].
Shown in Fig. 1, a carbon nanotube is a single graphite
plane, known as graphene, that has been rolled up into a
hollow cylinder. During formation, larger cylinders can
encapsulate tubes with smaller diameters and form a
multi-walled carbon nanotube (MWNT). However,
single-walled carbon nanotubes (SWNT) are of more
immediate electrical interest due to their outstanding
electrical and physical properties. They exhibit diameters
of 1 to 10 nm and can exceed several micrometers in
length [2]. Individual SWNTs have been used in
applications such as transistors and sensors, but their
widespread use has been hampered by a lack of control
over their diameter, location, chirality, and direction [3],
[4].
B. Single-Walled Carbon Nanotube Films
SWNT films consist of layers of interwoven SWNTs
forming a three-dimensional mesh. Due to averaging
effects, these films possess uniform and predictable
electronic properties independent of the varying
characteristics of the constituent nanotubes. SWNT films
can be created by vacuum filtration, deposited on
arbitrary substrates, and efficiently patterned, while
possessing low resistivity (on the order of 10-4 Q cm),
high mobility, and high transparency to visible and
infrared wavelengths. Thus, they are promising
candidates to be used as transparent, conductive
electrodes for various optoelectonic applications [5]-[9].
Fig. 2 contains an AFM image of a 20 nm thick SWNT
film patterned by photolithography with subsequent 02
plasma etching. Enabling electronic and optoelectronic


applications of SWNT films motivates the
characterization of SWNT film-semiconductor junctions
with various substrates. In this effort, it has been recently
shown that SWNT films make a Schottky contact on Si
and GaAs, but the nature of SWNT film on Ge remains
largely unexamined [10], [11].









C






Fig. 1 A SWNT can be thought of as a sheet of graphene
rolled into a hollow cylinder along a chiral vector, C. [1].


(b)


0

0 1 2 3 4
Distance (urn)


Fig. 2. (a) AFM image of SWNT film illustrating patterning by
02 plasma etching. The smooth surface between the film is the
SiO2 substrate where the film has been completely removed. (b)
Height of SWNT film shown by cross-section [9].


.**


1 ->
'A


4*
�wg


University of Florida I Journal of Undergraduate Research I Volume 10, Issue 2 | Spring 2010





KEITH KNAUER


C. Metal-Semiconductor-Metal Photodetectors
In this work, the SWNT film-Ge junction is studied
by the fabrication and electrical characterization of metal-
semiconductor-metal (MSM) photodetectors with SWNT
film electrodes on Ge substrates. Photodetector research
has broad implications due to the widespread use of
photodetectors in modem technologies such as high-
speed optical communication systems. An MSM
photodetector converts optical energy into an electrical
current and consists of two back-to-back Schottky
contacts on a semiconductor substrate. Illustrated in Fig.
3, MSM devices possess two metal contacts with
interdigitated finger electrodes. By applying a negative
bias to one contact and a positive bias to the other, a
depletion region and electric field are established. When
photons are incident on the semiconductor, electron-hole
pairs are generated and the electric field sweeps the
carriers out of the depletion region, giving rise to the
photocurrent. Increasing the bias can increase the speed
and collection efficiency of the device, but it can also
sharply increase thermionic emission and tunneling at the
Schottky contacts collectively referred to as the dark
current. In order to minimize power dissipation and
maximize sensitivity, the dark current must be minimized.
MSM photodetectors can be monolithically fabricated for
use in optoelectronic applications and have many
advantages over other types of photodetectors, including
low cost, simplicity of operation, and ease of fabrication
[12], [13]. A major limitation of MSM photodetectors is
their low responsivity due to light being reflected from
the surfaces of their metal electrodes, as depicted in Fig.
4 [13]. This motivates the use of transparent conductive
SWNT film as the electrode material to provide more
active area to incident light.
In this paper, MSM photodetectors with SWNT
film electrodes on Ge substrates are fabricated and
electrically characterized. A comparison of the dark
current of the SWNT film devices and metal control
samples is conducted, revealing that the use of SWNT
film electrodes reduces the dark current significantly.
Lastly, the dependence of the dark current on device
geometry is studied by varying device active area, finger
width, and finger spacing. These experimental results
give insight into the electrical properties of the SWNT
film-Ge interface while also demonstrating the use of
SWNT film as a transparent and conductive material for
electronic applications.

II. EXPERIMENTAL PROCEDURE

A multilayered mask and a Karl Suss MA-6 contact
mask aligner were used to photolithographically pattern
and fabricate the MSM devices. The process steps are
shown in Fig. 5. Following surface cleansing with
solvents, a 100 nm thick isolation layer of SiO2 was
deposited by plasma-enhanced chemical vapor deposition
(PECVD) in an STS 310PC PECVD. Active areas of


light ,


==


Fig. 3. MSM photodetector consisting of two contacts with
interdigitated finger electrodes. Light incident on the
semiconductor surface causes a photocurrent to flow between
the biased contacts. Adapted from [13].


I Incident


Reflected








Fig. 4. Incident light upon metal electrodes is reflected, leading
to decreased responsivity. Transparent conductive SWNT film
electrodes may allow some of this reflected light to be
transmitted to the substrate and contribute to the photocurrent
[13].


SiO2.


SWNT film
\


Ti/Au \


Fig. 5 [(a)-(d)] Fabrication steps used for MSM photodetectors
with SWNT film electrodes as seen through the cross section of
a single finger electrode. (a) Ge substrate prepared by cleaning
with solvents. (b) Isolation Si02 layer deposited by PECVD.
(c) Active area windows wet etched to surface. (d) SWNT film
prepared by vacuum filtration deposited across substrate. (e)
Interdigitated fingers etched in SWNT film by 02 plasma
etching. (f) Ti/Au contacts deposited by e-beam evaporation
with subsequent lift-off.


University of Florida I Journal of Undergraduate Research I Volume 10, Issue 2 I Spring 2010





MSM PHOTODETECTORS ON GERMANIUM SUBSTRATES WITH SWNT FILM ELECTRODES


various dimensions were then patterned using S1813
positive photoresist and subsequently etched to the
substrate by wet etching in 6:1 BOE. A SWNT film of
approximately 50 nm that was prepared by vacuum
filtration was then deposited. Following the film
deposition, interdigitated finger electrodes were patterned
by photolithography with S1813. The electrode fingers
were then etched by inductively coupled plasma etching
on a Unaxis Shuttlelock RIE-ICP using an 02 etching
chemistry. The etch parameters included 100W on the
substrate RF supply, 300W on the ICP RF supply, a
chamber pressure of 45 mtorr, and an 02 flow rate of 20
sem. In addition, a helium flow rate of 10 seem was used
to cool the substrate. After resist removal, the metal
contact pads were patterned using a double resist
photolithography step with S1813 and LOR3B. Here
LOR3B was used to facilitate the final metal lift-off.
Metal contact stacks of 30nm titanium and 70nm gold
were deposited using an Edwards e-beam evaporator.
Lift-off was then performed using PG photoresist
remover. Control samples with metal finger electrodes of
30nm Ti and 70nm gold were also fabricated for
comparison using identical process steps.
Fig. 6 shows optical microscope images of a
completed MSM photodetector with SWNT film
electrodes and a control sample of identical dimensions.
Various devices with differing active area length (AL),
active area width (AW), finger spacing (S), and finger
width (W) were fabricated to determine the dependence
of the dark current on device geometry. Furthermore,
four-point structures were also fabricated to determine the
SWNT film resistance.


III. RESULTS & DISCUSSION


A. Film Resistivity

First, the resistivity of the SWNT film was
determined by four-point probe measurements taken
using a parameter analyzer at room temperature. An
optical microscope image of a four-point probe structure
fabricated on the SWNT film sample with length L = 600
Ipm, width W = 20 Ipm, and thickness t = 50 nm is shown
in Fig.7(a). In taking these measurements, a current is
forced through the outer contacts and the voltage is
measured across the inner contacts. The resitivity, p, is
determined by the formula
L
R = p

where R is the resistance taken from the slope of I-V
curves [Fig. 7(b)]. The I-V curve of this device is shown
in Fig. 7(c), from which it is determined that the film
resistivity is approximately 4.7 x 10'3 0 cm.


3


A
W


. 0
r-

|-40

-80


Fig. 6 [(a)-(b)] Optical images of completed MSM devices. (a)
Photodetector with SWNT film electrodes. (b) Control sample
with metal electrodes for comparison.


-2.0 -1.0 0.0 1.0 2.0

Voltage (V)


Fig. 7 [(a)-(c)] Extraction of SWNT film resistivity (a) Optical
image of a four-point probe structure with SWNT film
electrodes. (b) Illustration of film dimensions. (c) I-V curve of
the four-point probe structure.


University of Florida I Journal of Undergraduate Research I Volume 10, Issue 2 I Spring 2010





KEITH KNAUER


B. Dark Current
Next, the dark current of the SWNT film and control
samples were measured at room temperature. Fig. 8(a)
shows the I-V curve for an MSM photodetector with
SWNT film electrodes, with AL=AW=200 pm and
W=S=15 lim. The I-V curve for the corresponding metal
control sample is shown in Fig 8(b). Schottky-like
behavior can be readily observed, but the curve does not
saturate at large voltages as would be expected due to the
back-to-back Schottky diode making up the MSM
photodetectors. More analysis is required to determine
the cause of this behavior, but it is evident from the
symmetry of these curves that the SWNT film acts as a
uniform material. In addition, Fig. 9 shows a comparison
of the dark current of these two devices on a log scale,
showing that the dark current of the MSM device with
SWNT film electrodes is nearly two orders of magnitude
lower than that of the control sample.
Finally, the effect of device scaling on dark current at
room temperature was studied. The dark current I-V
curves for MSM devices with W=S=20 pm and
decreasing AL=AW are displayed in Fig. 10 (a). The dark
current is shown to uniformly decrease with decreasing
AL=AW. This is expected as reducing the active length
and the number of fingers will lead to a lower dark
current. Fig. 10 (b) shows the room temperature dark
current I-V curves of MSM devices with AL=AW=300
pm and increasing S=W. The dark current decreases
uniformly with increasing S=W. This is because
increasing the finger spacing and width decreases the
total number of fingers, which decreases the dark current.
1.5
1.0 a)
S0.5
E
0.0
|-0.5
S-1.0
-1.5
-2.0 -1.0 0.0 1.0 2.0
Voltage (V)


1.E-02
1.E-03
1.E-04
1.E-05
1.E-06
1.E-07
1.E-08
1.E-09


S-I --- CNT Film
S...... Control


1.0
Voltage (V)


Fig. 9 Dark current comparison on a log scale plot showing the
use of SWNT film electrodes results in a dark current reduction
of nearly two orders of magnitude.


a)
....... AL= AW =400pm
- - AL=AW=300pm
----AL=AW=200pm ..

- .." s

. . ^... -* ,- ,' . . . .1

0.0 0.5 1.0 1.5 2.0
Voltage (V)


80 b)
S...... S=W =1Om10
60 - * -S=W=20im
----S=W=504m


Voltage (V)


- 0.0

S-20.0


Fig. 8 [(a)-(b)]
AL=AW=200pm
electrodes (b) m
evident.


b)


Fig. 10 [(a)-(b)] I-V curves showing the effects of device
scaling on the dark current. (a) Scaling of AL=AW when
W=S=20 pm (b) Scaling S=W when AL=AW=300 pm.



IV. CONCLUSIONS


.U MSM photodetectors with SWNT film electrodes on
-2.0 -1.0 0.0 1.0 2.0 Ge substrates have been fabricated and characterized.
Voltage (V) Schottky-like behavior is observed from room

Scurves of MSM photodetectors having temperature dark current measurements, but more
I-V curves of MSM photodetectors having . .
and W=S=15pm with (a) SWNT film analysis is required to fully understand the underlying
letal electrodes. Schottky-like behaviour is physical causes of this behavior. Insight could be gained
from future studies focusing on extracting the Schottky
University of Florida I Journal of Undergraduate Research I Volume 10, Issue 2 I Spring 2010


-14






MSM PHOTODETECTORS ON GERMANIUM SUBSTRATES WITH SWNT FILM ELECTRODES


barrier height of the SWNT film-Ge junction from dark
current measurements taken as a function of temperature,
as well as determination of the carrier type by measuring
the dark current of fingerless MSM structures with
asymmetric SWNT film contact areas [10]. Moreover, it
has been demonstrated that MSM photodetectors with
SWNT film electrodes have significantly lower dark
current than metal control samples and that their dark
current scales rationally with device geometry. A
comparison of the photocurrent and responsivity of the
MSM devices with SWNT film electrodes versus metal
control samples would be of interest in fully assessing the
advantages of using SWNT film electrodes for MSM
photodetectors on Ge substrates. These experimental
results give insight into the electrical properties of the
SWNT film-Ge interface while also demonstrating the
use of SWNT film as a transparent and conductive
material for electronic applications.

ACKNOWLEDGMENTS
The author would like to thank Dr. Ant Ural for
supervising this research project and providing invaluable
guidance. This work would also not have been possible
without the training from and discussions with the
members of the Nanotechnology Group at the University
of Florida, including Jason Johnson, Ashkan Behnam,
Yongho Choi, Karthik Vishwanathan, and Nischal
Radhakrishna. This work was completed through UF's
University Scholars Program.


REFERENCES
[1] A. Loiseau, et al., Ed., Understanding Carbon Nanotubes: From Basics
to Apphcations (Lecture Notes in Physics 677). Berlin, Germany:
Springer, 2006.
[2] R. Waser, Ed., Nanoelectronics and Information Theory: Advanced
Electronic Materials and Novel Devices, 2nd ed., Weinheim, Germany:
Wiley-VCH, 2005.
[3] A. Javey, J. Guo, Q. Wang, M. Lundstrom, and H. Dai, "Ballistic carbon
nanotube field-effect transistors," Nature, vol. 424, no. 6949, pp. 654-
657, Aug. 2003.
[4] K. Besteman, J. Lee, F. Wiertz, H. Heering, and C. Dekker, "Enzyme-
coated carbon nanotubes as single-molecule biosensors," Nano Letters,
vol. 3, no. 6, pp. 727-730, Jun. 2003.
[5] Z. Wu, et al. "Transparent, conductive carbon nanotube films," Science,
vol. 305, no. 5688, pp. 1273-1276, Aug. 2004.
[6] E.S. Snow, J.P. Novak, P. M. Campbell, and D. Park, "Random
networks of carbon nanotubes as an electronic material," Apphed
F: ..,. . Letters, vol. 82, no. 13, pp. 2145-2147, Mar. 2003.
[7] Y. Zhou, L. Hu, and G. Grfiner, "A method of printing carbon nanotube
thin films,"Apphedi -: ., .. Letters, vol. 88, 123109, Mar. 2006.
[8] E. Bekyarova, et al. "Electronic properties of single-walled carbon
nanotube networks," Journal of the American Chemical Society, vol.
127, no. 16, pp. 5990-5995. Apr. 2005.
[9] A. Behnam, et al., , "Nanolithographic patterning of transparent,
conductive single-walled carbon nanotube films by inductively coupled
plasma reactive ion etching," Journal of Vacuum Science & Technology
B, vol. 25, no. 2, 348-354, Mar/Apr. 2007.
[10] A. Behnam, et al., "Metal-semiconductor-metal photodetectors based on
single-walled carbon nanotube film-GaAs Schottky contacts," Journal
ofApphed ';*.., ... vol. 103, 114315, Jun. 2008.
[11] A. Behnam, et al., "Experimental characterization of single-walled
carbon nanotube film-Si Schottky contacts using metal-semiconductor-
metal structures," Apphedi : .. - Letters. vol. 92, 243116, Jun. 2008.
[12] K. Brennan, J. Haralson, J. Parks, and A. Salem, "Review of reliability
issues of metal-semiconductor-metal and avalanche photodiode
photonic detectors," Microelectronics Rehability, vol. 39, no. 12, pp.
1873-1883, Dec. 1999.
[13] L. Noriega, "Fabrication of novel metal-semiconductor-metal
photodetectors by nanolithographic patterning of single-walled carbon
nanotube films," M.S. thesis, University of Florida, Gainesville, 2006.


University of Florida I Journal of Undergraduate Research I Volume 10, Issue 2 I Spring 2010