Persistent photoconductivity on ion irradiated carbon films prepared by plasma assisted chemical vapor deposition and determination of traps in the pseudo-gap region

Article abstract of DOI:10.1063/1.115590

The defects created by ion beam irradiation in disordered conducting carbon films have been studied by the photoconductivity technique. A very complex distribution of traps created mostly by random displacement of carbon atoms by energetic ion beam from its polymeric matrix showed a persistent photoconductivity at low temperature. The decay time constant estimated from the photocurrent is around 15 s at 10 K. From the time constant and the intensity of photocurrent the density of traps and the corresponding activation energies are calculated. This report shows how slow decay of photocurrent can be applied to probe distribution of traps in the amorphous carbon due to ion bombardment in the most general case.

Full text of DOI:10.1063/1.115590

Persistent photoconductivity on ion irradiated carbon films prepared by plasma
assisted chemical vapor deposition and determination of traps in the pseudogap region
Somnath Bhattacharyya, K. S. R. Koteswara Rao, S. V. Subramanyam, and D. Kanjilal
Citation: Applied Physics Letters 68, 2079 (1996); doi: 10.1063/1.115590
View online: http://dx.doi.org/10.1063/1.115590
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/68/15?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Carbon thin films deposited by Capillary Assisted Chemical Vapor Deposition
AIP Conf. Proc. 378, 279 (1996); 10.1063/1.51136
Nitrogenated amorphous carbon films deposited from plasmas of methanolnitrogen mixtures
AIP Conf. Proc. 378, 283 (1996); 10.1063/1.51107
Amorphous and microcrystalline silicon films deposited by hotwire chemical vapor deposition at filament
temperatures between 1500 and 1900°C
J. Appl. Phys. 79, 8748 (1996); 10.1063/1.362501
Raman scattering, laser annealing and pressureoptical studies of ion beam deposited amorphous carbon films
AIP Conf. Proc. 120, 465 (1984); 10.1063/1.34715
Study of gap states in a–Si:H alloys by measurements of photoconductivity and spectral response of MIS solar
cells
AIP Conf. Proc. 73, 227 (1981); 10.1063/1.33035
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:
155.33.148.196 On: Tue, 25 Nov 2014 05:00:45

Persistent photoconductivity on ion irradiated carbon films prepared
by plasma assisted chemical vapor deposition and determination
of traps in the pseudo-gap region
a)
Somnath Bhattacharyya, K. S. R. Koteswara Rao, and S. V. Subramanyam
Department of Physics, Indian Institute of Science, Bangalore 560 012, India
D. Kanjilal
Nuclear Science Centre, New Delhi 110067, India
͑
Received 25 October 1995; accepted for publication 6 February 1996͒
The defects created by ion beam irradiation in disordered conducting carbon films have been studied
by the photoconductivity technique. A very complex distribution of traps created mostly by random
displacement of carbon atoms by energetic ion beam from its polymeric matrix showed a persistent
photoconductivity at low temperature. The decay time constant estimated from the photocurrent is
around 15 s at 10 K. From the time constant and the intensity of photocurrent the density of traps
and the corresponding activation energies are calculated. This report shows how slow decay of
photocurrent can be applied to probe distribution of traps in the amorphous carbon due to ion
bombardment in the most general case. © 1996 American Institute of Physics.
͓
S0003-6951͑96͒03615-7͔
2
Amorphous carbon has a wide variety of structure as
well as electrical properties e.g., diamondlike carbon, graph-
itelike carbon, glassy carbon, plasma deposited soft, and hard
graphite like film in particular at high energy (ϳ10 MeV͒
and at a low doses (1012 ions/cm ). In this letter we report
the formation of defects and their effect on the electronic
structure of carbon films. We have observed persistent pho-
toconductivity at low temperature in those samples which are
irradiated at high energy ͑170 MeV͒ and at a low dosage
2
2
carbon, etc. Depending on the relative concentration of sp
3
and sp hybridization bonds the above features were
1
interpreted. Besides its various applications amorphous car-
2
12
2
bon is a novel material to study the defects and to under-
(10 ions/cm ).
stand the physics of disorder, which can be enhanced by ion
irradiation. The ion implantation study on different types of
carbons at low energy ͑100 keV–1 MeV͒ and at a high dos-
The carbon films were prepared by dc plasma deposition
technique from maleic anhydride as a precursor. The plasma
power was 5 W ͑2 kV electrode potential͒. The gas pressure
of 0.5 Torr and substrate temperature of 550 °C were main-
tained throughout the experiment. The deposited films are
amorphous, brownish black in color and resistive to most of
chemical attack. The typical thickness of the films varies
from 0.1 to 1 ␮m. The typical conductivity of the virgin
samples is ϳ102 S cm and the activation energy at room
temperature is 0.034–0.04 eV. The finite conductivity of the
unirradiated samples at 4.2 K indicates that no gap is present
in the electronic structure of the material.6
1
5
2
age (10 ions/cm ) is attracting the attention of many sci-
3
entists in this direction. The role of ion beam at low-energy
ion implantation is to displace the carbon atoms from its
structure and to introduce disorder in the material. For dia-
3
mondlike carbon films ion implantation transforms sp
Ϫ1
bonds ͑which is thermodynamically less stable͒ to more ther-
2
modynamically stable sp configuration. This enhances the
conductivity of the material to several orders of magnitude.
Sometimes recrystallization takes place in the material to
increase conductivity. On the contrary, in graphite the inci-
dent ion beam dissociates the hexagonal ring structure lead-
ing to the decrease in conductivity. It is known that different
insulating polymers show a drastic increase of conductivity
ϩ13
The virgin films were subjected to ion (Iodine ) beam
of energy 170 MeV, at a dose of 1012 ions/cm and the
substrate temperature is maintained at 300 K. The irradiation
has been carried out at Nuclear Science Centre, New Delhi.
After irradiation, films were subjected to microscopic obser-
vation ͑e.g., XRD,SEM͒. Except for blisters neither clusters
2
4
after ion irradiation. Certainly the change in electronic prop-
erties indicates the deformation of electronic structure in the
material. The change in electronic structure results in the
formation of defect levels in the pseudo-gap region which
leads to the creation of different kinds of trapping centers in
the material ultimately causing changes in conductivity. The
defects are normally characterized by photoluminescence,
7
nor microcrystallinity were observed here. Drastic increase
in resistivity ͑2 to 3 orders of magnitude͒ and change in
activation energy ͑0.08–0.09 eV͒ ͑calculated within the tem-
perature range 300 to 150 K͒ were observed. The depen-
dence of conductivity of the irradiated films with tempera-
ture show an insulator like behavior indicating a gap created
5
Raman spectroscopy, ESR spectroscopy, etc. Photoconduc-
in the electronic structure of the material.8 The virgin
tivity is a good technique to determine defect centers and
distribution of traps in the photosensitive materials both in
crystalline and amorphous state. To the best of our knowl-
edge there is hardly any report on the irradiation effect on
samples are exposed to light at different temperatures but we
could not see any persistent photoconductivity from the
sample.
The irradiated samples are cooled down to 10 K in a
closed cycle refrigerator ͑CTI systems Inc.͒ The samples
were exposed to visible light using an ordinary tungsten
a͒Electronic mail: svs@physics.iisc.ernet.in
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:
Appl. Phys. Lett. 68 (15), 8 April 1996 0003-6951/96/68(15)/2079/3/$10.00 © 1996 American Institute of Physics 2079
55.33.148.196 On: Tue, 25 Nov 2014 05:00:45
1

FIG. 2. Graph shows ln͑normalized photocurrent͒ vs ln͑time͒ to determine
values of ␤ and it shows a distribution of ␤ value. I0 denotes dark current in
the circuit. ᮀ denotes data at temperature of 10 K. ᭺ denotes data at tem-
perature of 20 K.
FIG. 1. Graph shows ln͑normalized photocurrent͒ decay with time͑seconds͒.
Inset shows the experimental data of photocurrent decay with time at 20 K.
lamp. Sufficient voltage ͑ϳ10 V͒ was applied across the film
͑
which is less than the breakdown voltage of the film͒ such
combine the holes. Whether the retrapping is important or
not depends upon the relative rates of thermal emission and
recombination. Considering the strong recombination, i.e.,
retrapping is negligible and the decay of the photocurrent is
limited by thermal emission of trapped electrons. The photo-
that the generated carriers are removed from the active re-
gion before they recombine. The photocurrent is defined as
the difference between the total current in the circuit and the
dark current which was measured by two probe method. The
contact points were covered by black paint to avoid photo-
voltaic effect due to incident light. The typical power of the
Ϫ␤
current is given by I ϳ␶ , where ␤ϭT/T , 0Ͻ␤Ͻ1 and
ph
c
Tc corresponds to the critical temperature of carrier recom-
2
incident light on the film is 20 mW/cm . After sufficient
bination. We have plotted ln(I ) versus ln t which never
ph
time of exposure, i.e., when the current in the circuit gets
saturated, the light is switched off and decay of current is
recorded using Keithley programmable DMM and Electrom-
eter. This isothermal decay of photocurrent was observed
from 10 to 20 K. Above 20 K the photocurrent falls rapidly
and above 50 K the persistency is hardly detected. The ap-
proximate value of photodecay constant at 10 K is 15 s in the
initial part of the decay. It is also observed that the nature of
decay at 20 K is different to that of the 10 K, as shown in
Fig. 1.
shows a single straight line. At 10 K longer decay time is
observed at the beginning of photodecay curve and followed
by a few segments with different values of ␤, indicating that
the distribution of the trap is complex. Such kind of behavior
is also observed at 20 K as plotted in Fig. 2.
In the present material after radiation damage we have
observed an increase of resistivity and the formation of gap
inside the electronic structure of the material. ESR data
shows a distribution of paramagnetic centers whose ‘g’ val-
ues are different from that of the free electron which is con-
sistent with the observation of large number of traps found
from photoconductivity decay. These traps are inside the
band gap and give rise to localized states.
The slow decay of photocurrent ͑I͒ indicates the pres-
ence of traps in solid which can be filled by the carriers
9
generated by the optical source. Under steady state illumi-
ϩ
Ϫ
nation a large number of nonequilibrium carriers are trapped
in the gap states. In the case of transient photoconductivity
which is excited by pulsed light source, the current decay
mechanism is essentially due to the capture of free excess
carriers at the localized states and then recombining with
opposite carriers. But in the case of persistent photoconduc-
tivity the slow decay of photocurrent is mainly governed by
the recombination of nonequilibrium carriers released from
the metastable states.10 The long decay time of the photocur-
rent corresponds to long recombination lifetime of the carri-
ers. If the temperature is not too low, it is likely that after
removal of excitation source, the trapped electrons are ther-
mally released from metastable state to the extended states at
Let C and C be the two types of charged traps that
are generated by irradiation due to structural change. They
will capture electron and hole respectively to produce a neu-
tral trap C0.
ϩ
Ϫ
ϩ
Ϫ
C ϩeϭC , C ϩhϭC , 2C ϭC ϩC .
0
0
0
These two neutral traps (C ) will form potential barrier
0
E. When we shine light the captured electrons and holes get
released from the traps to the conduction band and valence
band, respectively. This leads to the increase in conductivity.
The photocurrent tends to overcome the barrier drift with an
energy Edr , which should be less than E to prevent imme-
diate recombination of carriers. The inhomogeneous distribu-
tion of potential requires the criteria E ϽE to be satisfied
the conduction band mobility edge E before they can re-
c
dr
free
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:
080 Appl. Phys. Lett., Vol. 68, No. 15, 8 April 1996 Bhattacharyya et al.
55.33.148.196 On: Tue, 25 Nov 2014 05:00:45
2
1

In conclusion, we report here the observation of persis-
tent photoconductivity in ion irradiated carbon films. It is
very unlikely to get a proper trap parameter from isothermal
photocurrent decay measurement particularly in this case
where traps are created at random in the material by ion
irradiation. The parameters might be determined directly
from sufficient knowledge on the ion beam and its effect on
the irradiated material. The experiment therefore indicates
that persistent photoconductivity can be manifested in a ma-
terial by creating controlled damage through ion irradiation
and we have given in this letter the rough estimation of re-
combination energies of traps from which order of defect
produced in the material may be estimated.
͑
where Efree corresponds to the energy of free carriers͒. This
criterion is obtained by the formation of highly disordered
state after irradiation. For the present material this residual
photoconductivity occurs at low temperature where conduc-
tion is governed by hopping mechanism. The thermal trans-
fer of electrons to the conduction band is difficult at this
temperature due to the potential barrier. The principal recom-
bination channel may be due to metastable states which can
ensure a sufficiently long time decay of photocurrent. The
degree of inhomogeneity is increased by the radiation dam-
age verified by the XRD data and SEM data. This may give
an idea of the presence of inhomogeneous potentials and
trapping centers inside the conduction and valence bands.
Let E be the recombination energy. Then the photoresponse
relaxation time depends on temperature given by t
ϭt exp(E /kT), where t ϳ10
ing on the nature of sample. For large t, E should be high,
r
Authors would like to thank CSIR and DST ͑India͒ for
financial assistance. The authors are grateful to Professor A.
K. Sreedhar for extremely valuable suggestions.
Ϫ12
Ϫ4
to 10 s, varies depend-
0
r
0
r
i.e., the traps formed should be deep enough, where Er cor-
responds to the energy of carrier recombination.9
The potential barrier due to localized traps is given by
E ϭkT ln(t /t ). The different constants at 20 K are calcu-
1
r
r
0
P. Koidl, Ch. Wild, B. Dischler, J. Wagner, and M. Ramsteiner in Prop-
lated from ln͑photocurrent͒ versus time plot as given in Fig.
. Calculations are done in different segments of time using
the approximate saturation currents and the dark currents
I ), as shown in the inset of Fig. 1. Substituting t ϭ15 s, 34
erties and Characteristics of Amorphous Carbon Films, edited by John J.
Pouch and A. Altovitz ͑Trans Tech Publications, Switzerland, 1990͒, Ma-
terials Science Forum, Vol. 52 & 53, Chap. 1, pp. 41–69.
G. M. Jenkins and K. Kawamura, Polymeric Carbon, 1st ed. ͑Cambridge
University Press, Cambridge, 1976͒.
S. Prawar, R. Khalish, M. Adel, and V. Richter, Appl. Phys. Lett. 49, 1157
͑1986͒.
S. R. Forest, M. L. Kaplan, P. H. Schmidt, T. Venkatesan, and A. J.
Ovinge, Appl. Phys. Lett. 41, 708 ͑1982͒.
John Robertson, Prog. Solid State Chem. 21, 199 ͑1991͒.
A. Sayeed, S. Bhattacharyya, and S. V. Subramanyam, J. Mater. Sci. Eng.
C 5, 231 ͑1995͒.
S. Bhattacharyya, K. S. R. K. Rao, and S. V. Subramanyam, presented at
the Int. Conf. on Mater. Sci. IUMRS-ICA-1995, Seoul, Korea.
S. Bhattacharyya, Ahmed Sayeed, D. Kanjilal, and S. V. Subramanyam,
Vacuum ͑in press͒.
1
2
(
0
r
s, 45 s, etc., we get E ϭ0.056, 0.049, and 0.047 eV with the
3
4
r
activation energy ϳ0.1 eV. From the ratio of t , t , t we
1
2
3
have roughly estimated the potential barrier for few major
localized states. In fact, it is a distribution of traps and the
radiation with high energy produces a random distribution of
localized states. On an average along with the higher activa-
tion energy some localized states are present in between two
bands. We intend to carry out this experiment in a detailed
way with different ion beams and with different higher doses
to find out the relationship between the nature of traps and
the ion dosage.
5
6
7
8
9
0
M. K. Sheinkman and A. Ya Shik, Sov. Phys. Semicond. 10, 128 ͑1976͒.
J. H. Zhou and S. R. Elliot, Phys. Rev. B 46, 12402 ͑1992͒.
1
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:
Appl. Phys. Lett., Vol. 68, No. 15, 8 April 1996 Bhattacharyya et al. 2081
155.33.148.196 On: Tue, 25 Nov 2014 05:00:45