AUTOMATION OF CHANNELING EXPERIMENT
FOR LATTICE STRAIN MEASUREMENTS
USING HIGH ENERGY ION BEAMS
S.V.S. Nageswara Rao1, D.K. Avasthi2, E.T. Subramanyam2, Kundan Singh2,
G.B.V.S. Lakshmi1, S.A. Khan2, Azher M. Siddiqui2, A. Tripathi2,
S.K. Srivastava2, Sarvesh Kumar3, T. Srinivasan4, Umesh Tiwari4,
S.K. Mehta4, R. Muralidharan4, R.K. Jain4 and Anand P. Pathak1∗
1
School of Physics, University of Hyderabad, Central University (PO), Hyderabad 500 046, India.
2
Nuclear Science Centre, Post Box No. 10502, Aruna Asaf Ali Marg, New Delhi 110 067, India
3
Department of Physics, R.B.S. College, Agra 282 002, India
4
Solid State Physics Laboratory, Timarpur, Delhi 110 054, India
40 MeV Si channeling studies have been performed on the strained In0.1Ga0.9As layer grown on GaAs
substrate using Molecular Beam Epitaxy (MBE). Three samples with different layer thickness have been
investigated in this study. Channeling experiment has been fully automated so as to minimize the radiation
damage. Suitable software and hardware have been developed to control the precision goniometer using the
CAMAC (Computer automation and control) based data acquisition system. In low energy He channeling,
strain measurements are often misled by the beam steering effects caused due to broad critical angles if the
strain is very low. Such effects can be minimized by increasing the probing beam energy as the channeling
critical angle is inversely proportional to the square root of the incident energy. Hence small angular
misalignments can also be resolved in the high energy channeling experiments. Heavy ions are chosen so as to
have reasonably high scattering cross-section and also to avoid the nuclear reactions.
the interface are of great interest. RBS/C has been
employed to measure such strains by several
researchers because it is a direct and depth
resolved method.
In RBS/C, one can find the axis direction
by experimentally plotting the back scattered yield
versus the sample tilt angle (called angular scan).
Strain measurements are based on the
determination of the direction of the off-normal
axis of strained layer, which is shifted with
respect to that of substrate. This small shift ( )
between epilayer and the substrate gives a
measure of strain ( /sin cos ). Hence the
accuracy of the strain measurement by
INTRODUCTION
Rutherford backscattering spectroscopy
and channeling (RBS/C) is a well established
technique [1-4] to determine the nature and the
concentration of crystal imperfections. Ion
channeling can in fact be used to locate the
defects in a crystal lattice. Recently it has widely
been used to measure the lattice strains present in
technologically important structures like strained
layer superlattices (SLS) [5-8]. SLS have been of
unique interest due to their ability to tune the
bandgap, which depends on the strain at the
interface. Therefore the strain measurements at
∗
Corresponding author. Fax: +91-40-23010181 / 23010227 / 23010120, email: [email protected]
CP680, Application of Accelerators in Research and Industry: 17th Int'l. Conference, edited by J. L. Duggan and I. L. Morgan
© 2003 American Institute of Physics 0-7354-0149-7/03/$20.00
94
RBS/channeling depends on sharpness of the
angular scan. The FWHM of this curve is directly
related to the channeling critical angle which has
one over square root relation to the incident
energy. Hence the angular scans will be sharp in
high energy channeling thereby the strain
resolution and the sensitivity of the method are
improved [6-8]. If the energy is too low then
FIGURE 1(a):
program.
DEVELOPMENT AND OPERATION
OF THE FACILITY
An automated high energy channeling
facility has been developed at Nuclear Science
Center, New Delhi, using the existing goniometer
(HV). A sample holder has been made using G10
with spring ball locking system. The goniometer
sample ladder has six degrees of freedom of which
three are translational and the other three are
rotational (R1, R2 & tilt). The axis of rotation of
R1 (360o, with step 0.018o) is in the vertical plane
and is perpendicular to scattering plane while that
of R2 (200o, with step 0.0125o) is parallel to beam
direction. The axis of rotation of tilt (7o, with step
0.001o) lies in the scattering plane and is
perpendicular to beam directions. All these
rotations can be controlled remotely using the
stepper motor controls provided with the
goniometer. Suitable software and hardware have
been developed to control this precision
goniometer using the CAMAC based data
acquisition system.
Compatible software routines are written
in C++ to integrate the existing data acquisition
system (Freedom [10]) with the goniometer
control. A CAMAC module has been designed
and installed to instruct the stepper motor controls
through data acquisition system. Any or all of the
rotors can be initialized to their lower limits using
the “Initialize” command. At a given time only
one rotor can be operated while the other two can
be initialized if required. Every time, the active
rotor will first be initialized to its lower limit and
then it will reach to the specified initial value.
This method is used to avoid the possible backlash
and also to keep the reference of rotor position
during the power failure. Once initialized, the
rotor will reach the final value in a given number
of steps. Required spectra will be collected at each
angle (step) for a given amount of time. At every
interval the incident current will also be recorded
using a CAMAC based Scalar module which is
normally inputted by a digital current integrator.
The program will also draw the online normalized
angular scan corresponding to a given energy
window. However the normalization is only
optional. The program is user friendly and all
inputs can be fed through popup windows. All the
data can be saved event by event in list mode
configuration. Offline routines have also been
developed to obtain the angular scans from
scans obtained using automation
the results are mislead by steering effect due to the
broad critical angles [6,9]. In high energy case,
heavy ions can be used for a reasonably high
scattering cross-section and also to avoid the
possible nuclear reactions. However there are two
major difficulties with high energy heavy ion
channeling. First of all the alignment becomes
difficult due to small critical angle. The other
difficulty is to minimize the radiation damage.
One can address to both of these problems up to a
considerable extent by using an automated
program for aligning the sample and also for
collecting necessary spectra on an integrated
control and data acquisition system. Radiation
damage can be minimized by keeping low fluence.
A large area position sensitive telescopic detector
has also been developed and installed on the
goniometer chamber at NSC, New Delhi. The use
of this detector in conjunction with ERDA
(Elastic Recoil Detection Analysis) will improve
the measurement quality because of the higher
recoil cross-sections and Z - identification.
Presently the system is ready for such an
experiment but the present experiment is
performed with small usual SSBDs (Surface
barrier detector) only. Irradiation time has been
greatly reduced by using the currently developed
automated system and also by using a separate
detector at forward angles for alignment process.
Following sections describe the development,
operation and results obtained using this facility
in detail.
95
different energy windows depending on the
requirements.
strained In0.1Ga0.9As layer grown on GaAs
substrate at SSPL, Delhi using Molecular Beam
Epitaxy (MBE). Three samples with different
layer thickness (100 Ao, 250 Ao & 400 Ao) have
been investigated in this study.
FIGURE 1(b): A polar chart obtained to locate the
axis.
FIGURE 2 a(left) & b(right) : Angular scans obtained
to measure the angular misalignments a) in 250 Ao
sample and b) in 400 Ao sample.
RESULTS AND CONCLUSION
The “automated high energy channeling
facility” has been tested and found to be working
consistently. The facility has been checked three
times in three different runs. Here we present
some of the results obtained using this facility.
Fig.2 shows the angular scans obtained from
different samples. Reasonably good reduction is
obtained from the scans that are performed on
epilayers. They are either fitted by Gaussian or
second order polynomial functions [6]. The strain
value ( t = 0.4%) obtained from Fig. 2a i.e for the
250 Ao sample turns out to be less when compared
to that of the expected value (~ 0.9%). This may
be caused due to irradiation effect during the
alignment process. The ion irradiation causes ion
beam mixing and will reduce the strain in the
layer [8].
Hence for the other samples we took angular
scans on the fresh spot just by shifting the beam
position by 1 or 2mm vertically (or horizontally)
after axis is found. Very small but fine search is
sufficient to remap the axis if it is missed due to
the above mentioned linear motion. Fig. 2b shows
the angular scans obtained on a fresh spot
(unirradiated portion) of 400Ao sample. The
strain value obtained from the measured angular
misalignment is still less (0.56%) when compared
to that of the expected value but is comparable
(~7%) with earlier results [6]. Fig. 2c shows the
angular scan obtained from the fresh spot of
FIGURE 1(c): Angular scan obtained from substrate of
100Ao sample.
We have followed the standard polar
chart method [1] to locate the required axis. Fig.
1a shows some scans recorded while searching
for planar dips as a part of alignment process. Fig.
1b shows a polar chart used for mapping the axis
and Fig. 1c shows a sample angular scan obtained
from the substrate of the one of the samples.
EXPERIMENTAL
40 MeV Si ions delivered from the
15MV Pelletron accelerator of NSC, New Delhi
have been used as incident beam. A pair of double
slights separated by roughly around 4m was
employed to define the incident angle and also to
get the fine and parallel beam. Two SSB
Detectors were placed at 60o and 110o for
alignment and data collection respectively.
Forward angle was used for alignment process so
as to reduce the irradiation time while back angle
was used to improve the energy resolution.
Channeling studies have been performed on the
96
100Ao sample. Considerable asymmetry is
observed in this angular scan. The yield goes
down on the lower angle side when compared to
that of the higher angle side. However the
angular
scan measured in the same energy
window of the same sample but on the irradiated
spot dose not show any such asymmetry. This
indicates that this asymmetry is directly related to
the strain present in the layer. The effect is
reduced due to the decrease in strain in the
vicinity of the irradiated portion. Such
asymmetric scans are observed in highly strained
SixGe1-x/Ge multi layers [11]. Here the layers are
very thin (mono layers) when compared to the
channeling wave length of the probing beam.
These asymmetric angular scans are analyzed
using Monte-Carlo simulations. The atomic rows
of the tilted (strained) layers look like interstitial
impurities. The strain values obtained
ACKNOWLEDGEMENTS
SVSNR thanks CSIR for SRF. GBVSL thanks
DRDO for the financial support through a
research project. APP & DKA thank DRDO for
supporting this work through a research project.
REFERENCES
1.
Chu W.K, Mayer J.W, Nicolet M.A,
Backscattering spectrometery, Acad. Press,
New York, 1978.
2. Feldman L. C., Mayer J. W. and Picraux S.
T., Materials Analysis by Ion Channeling,
New York, Acad. Press, 1982.
3. Gemmell D. S, Rev. Mod. Phys. 46, 129
(1974).
4. Pathak A. P., Radiation Effects and defects,
30, 193, (1976).
5. Semiconductors and semimetels vol. 33, ed.
Persall et al, Acad. Press, New york (1991).
6. Siddiqui A.M, Nageswara Rao S.V.S, Pathak
A.P, Kulkarni V. N,.Kesav Murthy R, Eric
Williams, Daryush Ila, Claudiu Muntele,
Chandrasekaran K.S and Arora B. M., J.
Appl. Phys., 90, 2824 (2001).
7. Pathak A.P, Nageswara Rao S.V.S, Siddiqui
A.M, Nucl. Inst. and Meth. B, 161-163, 487
(2000).
8. Pathak A.P, Nageswara Rao S.V.S, Siddiqui
A.M, Lakshmi G.B.V.S, Srivastava S.K,
Ghosh S, Bhattacharya D, Avasthi D.K,
Goswami D.K., Satyam P.V, Dev B.N and
Turos A, Nucl. Insrt. Meth. B, 193, 319
(2002).
9. Shin Hashimoto, Feng Y.Q. and Gibson
W.M, Nucl. Inst. and Meth. B, 13, 45 (1986).
10. FREEDOM, The linux based data acquisition
system at NSC, Ajit Kumar, B.P. Et. al.,
SANAI-97 held at BARC on 5th Feb. (1997).
11. Feldman L.C, Bevk J, Davidson B.A,
Gossmann H. J , and Mannaerts J.P., Phys.
Rev. Lett., 59(6), 664, 1987.
FIGURE 2 c(left) & d(right) : Angular scans obtained
to measure the angular misalignments in 100 Ao sample
c) on irradiated spot. d) on fresh spot (unirradiated)
from the angular misalignments using the simple
formula given in this paper are not valid in such
case. These results suggest that the present scans
are also to be analyzed using Monte-Carlo
simulations. This asymmetry is a good tool to
diagnose the exact value of stain. This will
improve the resolution of the measurement.
Moreover the capability of the technique can be
enhanced using the large solid angle position
telescopic detectors in ERDA mode. Such studies
will be taken up as a continuation of this work.
97