Makram thesis presentation

STRUCTRAL, ELECTRICAL AND
THERMOELECTRIC PROPORTIES OF
CrSi2 THIN FILMS

by
Makram Abd El Qader
Candidate for Master of Science in Electrical
Engineering
Department of Electrical and Computer
Engineering Department

 Thermoelectric Materials and Application
Clean source of energy – Power generation upon application of heat gradient

(a) (b)
PN couple used as TEG (a)-Seebeck effect, and TEC (b)-
Peltier effect
Dept. of Electrical and Computer Engineering
2 University of Nevada, Las Vegas

 Thermoelectric phenomena and coefficients
 In Solid state thermoelectric devices
Diffusion Principle in materials Mobile charge carriers Thermal gradient
Charge build up (e-) & (h+) Electrostatic potential (voltage) Seebeck effect-
thermoelectric generation (TEG)

 The efficiency of power generation in thermoelectric devices is
determined by its dimensionless figure of merit (ZT):

ZT=α2σ/κ

α is the Seebeck coefficient µV/K,
σ is the electrical conductivity Ωm, and
κ is the thermal conductivity W/m-K.

 The thermoelectric performance can also be evaluated by the
power factor
P=α2/ρ
ρ is the resistivity Ωm

 Background and Thermoelectric Phenomena
1. Approximately 90% of the world’s electricity is produced by heat energy
as a result of burning fossil fuel

2. Production plants typically operate at 30-40 per cent efficiency, loosing
around 15 terawatts of power in the form of heat to the environment.

3. Waste Heat sources are found almost in every process and electronic
devices (Residential heating, automotive exhaust, and industrial
processes

4. Thermoelectric power generators can convert some of this waste heat
into useful power

5. Thermoelectric devices are potential power source due to their direct
conversion of thermal gradients into electric current.

6. Electronic devices, International space station and Satellites, Automobile
companies, Power plants

Transition
 Potential thermoelectric materials metal silicides
CrSi2, FeSi2, CoSi2,…. …

 Characteristics of Silicides
 Partially filled d- orbital's-
Seebeck value much higher
 High melting point and
chemical stability at high
temperatures
 Relatively low thermal
conductivity values

 Materials with highest figure of merit A good thermoelectric material should
BiT2 and SbTe hold the highest ZT values of 3 have low electrical resistivity, low thermal
conductivity, and a large Seebeck
coefficient.

The issue
The electrical and thermal properties of a material are determined by the same crystal
and electronic structure
Usually:
They cannot be controlled independently. The challenge is to find ways to decouple the
electrical and thermal properties
keys:
Study thermoelectric materials in Thin Film form. This may cause a change in the
material thermal and electrical properties
2-D dimensions
Precise controlled composition
Easy to create defects-doping, process conditions..
Scalable for small/large devices
Theoretical studies predict better enhanced ZT with low dimensional structures.
Motivation
Study the structural, electrical, and thermoelectric properties of CrSi2 thin films to better
enhance the ZT.


Literature data on CrSi2

Physical property Value
Energy gap indirect band gap 2.7eV
Carrier type P type 4×109 cm3
Bulk electrical resistivity at (RT) 0.9 mΩcm
Bulk Seebeck coefficient at (RT) 96µV/K
Bulk thermal conductivity at (RT) 10W/mK
Thin film crystallization temperature 300˚C
Crystal structure Hexagonal structure
Space group P6222
Lattice parameters a= b= 4.4220Å, c=6.351Å
Structural, thermal, and electrical properties of bulk CrSi2 are well studied.”
“Structural, thermal, and electrical properties of CrSi2”, by T. Dasgupta, J. Etourneau,.”
 electrical and structural properties of ( 50nm) thin film of sputtered CrSi2” Electrical and
structural properties of thin films of sputtered CrSi2”, by S.F. Gong a, X.-H. Li a..”
Electrical, structural, and transport properties of CrSi2/ Si (111)

Outline
PART 1
Thin Film Preparation - Experiments on Thin Film samples
 Thin film processing
Energy Dispersive X-ray diffraction (EDAX)
 X-ray Diffraction (XRD)
 Four probe point resistivity measurement
Seebeck coefficient measurement
Power factor measurement
Results and discussion drawn on thin film samples
PART 2
Design and assembly of three gun sputtering system
 Design motivation
 Design methodology
Results and discussion drawn from system pump down
 Final conclusions and Future work


Thin film processing
 quartz glass substrates( κ=1.38W/mK, R=1018Ωm) were prepared by:
Aquasonic deionized water bath, methyl alcohol, dried out with nitrogen
gas, and heated.
 1µm and 0.1µm CrSi2 thin films were prepared by RF sputtering
Process condition Value
Base pressure (torr) 1.2×10-7
Ar gas pressure (mtorr) 1

RF power supplied (W) 200

Target substrate 3
distance (inch)
Pre- sputtering time 10
(min)
Deposition time (min) 7 min for 0.1µm, 37min
for 1µm

Sputtering Process chamber


Thin film thickness measurement

Surface Profiler Veeco Dektak 6M Stylus Profilometer
The obtained thin films have a step profile
similar to the one show below
Deposition Time
S.No Thickness (µm) Deposited CrSi2 material
(min) Glass substrate
Step Profile

1. 5 0.08

2 10 0.12

3 30 0.75

4 45 1.2

5 60 1.4


Thin film annealing

 In order to find out the effect of temperature, the thin film samples were
annealed under argon gas (Ar) ambience.

 Annealing Temperature (T) = 300˚C, 400 ˚C, 500 ˚C, 600 ˚C

 Argon gas Pressure (P) = 695 torr

 Duration time (t) = 60 and 120 minutes


Thin film - Compositional Analysis

 The compositions of processed thin filmsamples were verified by
performing Energy Dispersive X-ray Analysis (EDAX).
 JOEL JSM – 5600 Scanning Electron Microscope, Energy = 15keV

Thin film samples with 0.1µm thickness have
shown an atomic composition of Cr=37.64%
and Si=62.36%.
 Thin films samples with 1µm thickness have
shown an atomic composition of Cr=39.27%
and Si=60.73%

The obtained results show that the discrepancy between the
compositions of the target material and thin films are less than
5%.


Thin film microstructure images-Scanning Electron Microscope (SEM)

0.1 µm thin film as sputtered 0.1 µm thin film after annealing at 300˚C

1 µm thin film as sputtered 1 µm thin film after annealing at 300˚C
14 Dept. of Electrical and Computer Engineering
University of Nevada, Las Vegas

Thin film – Structural Analysis

X-ray diffraction pattern were taken using a Bruker-AXS D8 Vario Advance
using a Johansson-type primary monochromator with Cu kα1 emission

λ=1.54063Å
Incident beam angle θ= 5˚
Reflected angle 2θ=10-90˚

The Rietveld structure refinement allows peaks
fitting by calculating the structure factors for
each lattice plane by applying :
pseudo-Voigt type profile functions (Thompson-
Cox-Hastings)
 fundamental parameter approach.


Thin film structural analysis
The obtained results from the Rietveld refinement for all samples regarding their
X-ray diffraction refinement values for CrSi2 1µm thin films
Sample ID R-Bragg Scaling Factor Refined cell
Refinement parameters, a and c
Residual (<< 5%) (Å)
 Bragg residuals:
indicates the difference CriS2 as-sputtered NA NA NA
between the calculated and 1.103 4.449, 6.293
CriS2 300C 1h 0.000209
measured intensities 4.4331, 6.317
Scaling factor: gives an CriS2 400C 1h 1.292 0.000235
indication about amount of 4.4152, 6.3359
CriS2 500C 1h 1.705 0.0001637
the phase in the material 4.443, 6.244
The refined lattice CriS2 600C 1h 1.309 0.000280
parameters 1.249 4.445, 6.285
CriS2 300C 2h 0.000258
4.4289, 6.304
CriS2 400c 2h 1.353 0.000299
1.891 4.4127, 6.3382
CriS2 500c 2h 0.0002248
4.4304, 6.2981
CriS2 600C 2h 1.388 0.0002750


Thin film structural analysis
X-ray diffraction refinement values for CrSi2 0.1 µm thin films
Sample ID R-Bragg Refinement Scaling Factor Refined cell
Residual (<< 5%) parameters, a and c
(Å)
NA NA
CriS2 as-sputtered NA
0.646 4.438, 6.280
CriS2 300C 1h 0.000219
4.439, 6.253
CriS2 400C 1h 0.814 0.000265
4.425, 6.262
CriS2 500C 1h 0.625 0.000264
4.435, 6.272
CriS2 600C 1h 0.538 0.000452
0.512 4.420, 6.286
CriS2 300C 2h 0.000193
CriS2 400c 2h 0.602 0.000263 4.433, 6.260
0.581 4.423, 6.265
CriS2 500c 2h 0.000225
0.691 4.439, 6.271
CriS2 600C 2h 0.000234
R-Bragg Refinement Residual much less than 5%, thus fit is excellent. lattice parameters
obtained for various thin films are in the within the expected values for CrSi2.

Thin film structural analysis- diffraction patterns

1µm thin film- 1 hour annealing time- 300˚C- 1µm thin film- 2 hour annealing time-
600˚C 300˚C- 600˚C

Crystallization of the hexagonal modification The diffraction pattern for 1 hr. is dominated
of CrSi2 was observed at 300˚C by the (111) and (112) peak intensities, and for
Crystallization became better at higher 2 hr. is dominated by the (111),(112), and (003)
annealing temperatures. peak intensities.

Thin film structural analysis- diffraction patterns
0.1µm thin film- 1 hour annealing time- 0.1µm thin film-2 hour annealing time-
300˚C- 600˚C 300˚C- 600˚C

Crystallization of the hexagonal modification
There is no change in the peak intensities
of CrSi2 was observed at 300˚C
between 1 hr. and in the 2 hr. annealed samples
Crystallization became better at higher
This indicates that 0.1µm CrSi2 thin films are
annealing temperatures.
fully crystallized at 1 hr.

Seebeck coefficient measurement
Seebeck voltages of 1µm and 0.1µm thin films were measured for various annealing
temperatures in the range of 100˚C-600˚C for two different annealing times, 1hr and 2 hr.
A Seebeck voltage measurement device was designed and built to measure the Seebeck
coefficient of the CrSi2 films at room temperature
The estimated accuracy of the seebeck coefficient measured was ±5%, and was verified by
measuring the Seebeck coefficient of Ni samples in both bulk and thin Film form with
known Seebeck coefficient values

Seebeck coefficient measurement apparatus at 20˚C ΔT

Seebeck coefficient results
90
Seebeck coeffcient

80
70
60
1µm thin film
(µV/K)

50
40 Seebeck coefficient (µV/K)-1hr
30 annealing
20 Seebeck coefficient (µV/K)-2 hr
10 annealing
0
0 200 400 600 800

Annealing temperatures(C˚)
70
Seebeck coeffcient

60 0.1µm thin film
50
40
(µV/K)

30 Seebeck coefficient (µV/K)-1hr
20 annealing
10 Seebeck coefficient (µV/K)-2 hr
0 annealing
0 100 200 300 400 500 600 700



1
d2

Seebeck coefficient discussion
Seebeck coefficients in general increase with the annealing temperature for both
thicknesses and annealing times up to 400oC. This behavior is directly related to the
better crystallinity of the thin films at higher annealing temperatures.

In the temperature range of 400 to 500oC, all plots show a sudden change in Seebeck
coefficient

Seebeck coefficient saturates at around 60µV/K for 0.1 µm thin films

.For 1 µm thin films annealed for 1 hr. the Seebeck coefficient shows a plateau
between 400 and 500oC and then increases and reaches 81µV/K close to the reported
bulk value of 96µV/K, whereas the 2 hr. annealed thin film shows a decrease

This difference behavior of the 1 µm thin films can be related to the degradation of the
thin film microstructurally with the creation of voids and cracks at higher annealing
temperature and longer annealing times.


Thin Film resistivity measurement
Resistivity of 1µm and 0.1µm thin films for various annealing temperatures in the range
300oC-600oC for two different annealing time, 1hr and 2hr.

Four probe point resistance measurement apparatus (ASU-Newman Group) was used
at room temperature

Thin film resistivity values were calculated using

with
t is the thin film thickness
s is the spacing between the probes


Thin Film resistivity results
1

0.9

0.8
Resistivity (mΩ-cm)

0.1µm thin film
0.7

0.6

0.5 Resistivity (mΩ-cm)- 1
0.4
hr annealing
0.3
Resistivity (mΩ-cm)- 2
hr annealing
0.2

0.1

0
0 100 200 300 400 500 600 700



Thin Film resistivity results

Resistivity of 1µm thin films couldn’t be measured due to their high resistance values
which exceeded the limitation of the measurement system

It is estimated that 1µm thin films have a resistance value larger than 1MΩ. Based this
estimate, the resistivities of the annealed 1µm thin films were calculated to be larger than
0.000453 MΩ-cm, while the as deposited show to have resistivity of 1.197mΩ-cm.

For both annealing times, 1hr. and 2hr., 0.1 µm thin films show that the resistivity
increases with annealing temperature till 300oC and reaches a value of 0.9 mΩ-cm,
which is close to the reported bulk value and then decreases till 400o C and then saturates

The increase in resistivity is consistent with the film become more crystalline with
temperature. Decrease of resistivity beyond 400oC cannot be explained. This needs to be
investigated further.


Thermoelectric power factor measurement

The thermoelectric power factors, P, of 0.1µm thin films was calculated and
plotted for various annealing temperatures in the range of 300˚C-600˚C for two
different annealing times, 1hr. and 2 hrs.

The thermoelectric power factor, P for 1µm thin films could not be calculated as
resistivity, which is necessary for the calculation could not be measured due to the
limitation instrument.

The calculations of the power factor were done using the following equation:

P=α2/ρ (W/K2 m)
where
α is the Seebeck coefficient
ρ is the resistivity


Thermoelectric power factor results
1.20E-03
Power Factor ( W/K2

1.00E-03 0.1µm thin film
8.00E-04
power factor 0.1 1hour
m)

6.00E-04
annealed
4.00E-04
power factor 0.1 2hour
2.00E-04
annealed
0.00E+00
0 100 200 300 400 500 600 700

Annealing temperature C˚
Thermoelectric power factor increases with annealing temperature from 300oC to
400oC and saturates at about 0.9 x 10-3 W/(K2.m) beyond 400oC for 0.1µm thin films
annealed for 2 hrs
0.1µm thin films annealed for 1 hr, thermoelectric power factor increases with
annealing temperature from 300oC to 500oC and saturates at about 1.1 x 10-3 W/(K2.m)
beyond 500oC
This behavior can be attributed to increase in crystallinity in the higher annealing
temperature range.

 PART 1-Results and discussion

Seebeck coefficient and resistivity increases linearly, between 100˚C to 300˚C
this correlates well with the observation of increased crystallinity of the deposited
thin films.

The difference measured Seebeck coefficients between 0.1 µm and 1 µm thin
films annealed in this temperatures range is very minimal.

The resistivity results show a marked difference with 0.1 µm exhibiting
measurable values in the range of 0.2 to 0.9 mΩ-cm, and 1 µm thin films have
resitivities larger than 0.000453 MΩ-cm

This difference is related to the drastic difference in the mictrostructure between
the two thicknesses. Annealed 1 µm thin films exhibit a large density of pores,
where as 0.1 µm thin films exhibit a smooth texture.

Both 0.1 and 1 µm thin films show a transition in Seebeck coefficient between
300oC and 400oC


 PART 1- Results and discussion
0.1 µm thin film showing a plateau beyond the transition temperature and 1 µm thin film
showing a plateau for about 100 C range and then increasing further for shorter anneal times
and a peak at the transition temperature for longer anneals.

Degradation of properties for 1 µm thin films with longer duration of anneal may be related
to degradation of the thin films microstructurally. In other words, cracks and voids may
cause the degradation.

0.1 µm thin films show a peak in resistivity around 300oC

Decrease of resitivities beyond 300˚C anneal is unclear

1 µm thin films have resistivity larger than the limits of the instrument. Such high
resistance may be a result of porosity observed in the annealed films.

Thermoelectric power factors for 0.1 µm thin films with respect to annealing temperatures
show a behavior similar to that of Seebeck coefficients, increasing with temperature and
reaching a plateau value of 1.0 x 10-3 W/(K2 m) at around 400o C to 450o C

Results and discussion

 Due to highly resistive nature of 1 µm thin films, the thermoelectric power factor for
these films has an upper estimate of 6.403×10-6 W/(K2 m)

 These results suggest that annealed 400˚C thin films of thicknesses in the range of
0.1µm are more suitable for device applications when glass substrates are employed.


PART 2-Design of Three Gun Sputtering System

Investigate ternary
and higher order
thermoelectric alloys

limitation of the Better control over
current sputtering process conditions ( gas
system in the solid input, heat, rotation,
state fabrication vacuum level, etc….)
laboratory at
UNLV.
Design
motivation


Design of Three Gun Sputtering System

Heating 99% pure High vacuum
capability for films level ( 10-9
substrate oxide sale)
remove

multiple target
materials / DC,
Deposition RF power
yield
monitoring
Design
considerations Ion beam
etching and
Precise inert cleaning
gas control capability


Three gun sputtering system building blocks:

Oil sealed rotary mechanical pump (MP)
Molecular drag pump (MDP)
Turbo-molecular pump
CTI Cryogenic pump
Vacuum process chamber
Convectron gauge
Ionization gauge
Capacitance manometer gauge
Mass flow controller
Crystal thickness monitor (QCM)
Substrate table- heat and rotation
Residual gas analyzer (RGA)
Sputter sources
Ion gun
Gate valves
Water chiller


Solid works design

A drawing of the stainless steel 6 way A schematic diagram showing top
cross chamber flange-housing for sputter guns and
shutters

A schematic diagram of top flange with sputter schematic diagram of the three sputter sources-guns used
sources and shutters installed

Solid works design

A schematic diagram illustrating the focus of the three guns
to the location of the substrate

A drawing of the of the deposition chamber

System assembly

A photograph showing the three gun sputtering system


photograph showing an inside look of the chamber


A symbol representation of the 3 gun sputtering system


Three gun sputtering system results
Residual gas analyzer results

The quadrupole gas analyzer spectra's are plots of versus partial pressure

Quadrupole gas analyzer spectrum after initial pump down

It is observed from above spectra that when the system was turned on for the first
time, high Nitrogen (N) at of 28 and Oxygen (O2) of 32 peaks, were observed
making the vacuum level to stay in 10-05 Torr scale.


Before (Yellow) and after (Green) RGA spectrum showing effect of reducing the
foreline pressure of the turbopump by adding a molecular drag pump

It was observed from the green RGA spectrum that the vacuum level in the chamber
gets much better (10-7 torr) after solving the problem of compression ratio by installing
the molecular drag pump between the turbopump and mechanical pump


Quadrupole gas analyzer spectrum of ratio versus partial pressure-
At the present

The system pumped overnight to the mid 10-09 Torr range, leaving the water
peak of 18 as the major one as expected


PART 2- Results and discussion

 In order a deposit ternary and higher order alloys, a three gun sputtering
system was designed, built and tested for its level of vacuum levels and
cleanliness.

 The tests showed that the three-gun sputtering system is of vacuum levels of
10-9 torr and shows extremely low level of impurities and is ready for future
sputtering works in this area.


Conclusion
CrSi2 films of two different thicknesses were prepared by rf sputtering.

As deposited and annealed (300˚C to 600˚C) were characterized for their structural,
electrical, and thermoelectric transport properties

As-sputtered CrSi2 film is amorphous at room temperature and crystallizes around
300˚C independent of thickness.

The Seebeck voltage of the1µm films increase sharply with annealing temperatures
and reaches a value of 81µV/K, which close to that of bulk CrSi2, and 62µV/K for
0.1µm films

These results suggest that annealed thin films of thicknesses in the range of 0.1µm
round 400˚C are more suitable for device applications when glass substrates are
employed.


Recommendation and Future work

Based on our experience with CrSi2 deposition and characterization, and also the design
and assembly of the three gun sputtering system, the following issues are recommended
for future investigation:

Investigation of the structural behavior of the 1µm CrSi2 thin films at annealing
temperatures greater than 300C. In other words, identify the reasons for the film to crack
with annealing.

Study of the electrical and thermoelectric properties as a function of thin film
composition before and after annealing.

Measurement of the thermal conductivity of all deposited thin films before and after
annealing, to allow us calculate the thermoelectric figure of merit ZT.

Use of the designed three gun sputtering system to better sputter CrSi2 thin films.


Acknowledgment Committee members:
Dr. Rama Venkat
Dr. Ravhi Kumar
Dr. Thomas Hartmann
Dr. Nathan Newman

Group members:
Stan Goldfarb
Dr.Paolo Ginobbi
Brandon blackstone
Nirup Bandaru
Jorge Reynaga
Eric Knight
Mike Shappie

Friends and family:
I would like to thank my parents, my
family, and my freinds for their great
support. I would like to thank my brothers
Charbel Azzi and Charles Azzi on their
great support too.


I would like to thank the following companies on their support for making the design of the 3
gun sputtering system possible:

Engineering college-Electrical and computer engineering Department
College of sciences- Physics Dept- High pressure center
UNLV Graduate College
Ron Powell; Novellus
Steve Schwartz and Steven Michaud; Brooks Automation
Dan Watt
John Brooks and Tom Bogdan; MDC;
Fred Van der Linde Chris Malocsay; Semicore
Craig Hall; Ferrofluidics Paul Becker; Fil-Tech
Dave Mahoney; Rigaku Neil Peacock and Dick Jacobs; MKS
Richard Osburn NCCAVS Doug Schatz; Advanced Energy
Ralph Brogan; Pumps International Mark Bernick; Angstrom Sciences
Mike Ackeret; Transfer Engineering Don Sarrach; Plasmaterials
Neal Ely; Las Positas College
Todd Johnson and Harry Grover; MeiVac
Larry Lu; CLuLab
Will Hale; AJA International
Mark Bernick; Angstrom Sciences

THANK YOU ALL


Makram thesis presentation

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Viewers also liked

Viewers also liked (20)

Similar to Makram thesis presentation

Similar to Makram thesis presentation (20)

Recently uploaded

Recently uploaded (20)

Makram thesis presentation