Preparation of Thin Films by a Bipolar Pulsed-DC Magnetron Sputtering System Using Ca3Co4O9 and CaMnO3 Targets

doi:10.4236/msa.2012.39094

Paper Menu >>

Journal Menu >>

Materials Sciences and Applications, 2012, 3, 645-649

http://dx.doi.org/10.4236/msa.2012.39094 Published Online September 2012 (http://www.SciRP.org/journal/msa)

645

Preparation of Thin Films by a Bipolar Pulsed-DC

Magnetron Sputtering System Using Ca3Co4O9 and

CaMnO3 Targets

Weerasak Somkhunthot1, Nuwat Pimpabute1, Tosawat Seetawan2

1Program of Physics and Science Center, Faculty of Science and Technology, Loei Rajabhat University, Loei, Thailand; 2Thermoelectrics

Research Center, Faculty of Science and Technology, Sakon Nakhon Rajabhat University, Sakon Nakhon, Thailand.

Email: [email protected]

Received June 7th, 2012; revised July 2nd, 2012; accepted August 5th, 2012

ABSTRACT

The thin films were deposited on the glass substrates by an asymmetric bipolar pulsed-dc magnetron sputtering system

using the Ca3Co4O9 (p-type) and CaMnO3 (n-type) targets of 60 mm diameter and 2.5 mm thickness. The targets were

prepared from powder precursors, which obtained by a solid state reaction. Optical emissions from plasmas during

sputter depositions of films were detected using a high resolution spectrometer. Thickness of thin film was estimated by

Tolansky’s Fizeau fringe method and ellipsometic measurement. Crystal structures were studied from X-ray diffraction.

The thermoelectric properties were assessed from Seebeck coefficient and electrical resistivity measurements at room

temperature. The power factors were calculated. It was found that the optical emission spectrums showed that the Ca,

Mn, Co and O atoms were sputtered from the targets onto glass substrates. As-deposited Ca-Co-O and Ca-Mn-O films

thickness values were 0.435 m and 0.449 m, respectively. The X-ray diffraction patterns clearly showed amorphous

nature of the as-deposited films. Determining thermoelectric properties of Ca-Co-O film gave Seebeck coefficient of

0.146 mV/K, electrical resistivity of 0.473 cm, and power factor of 4.531 µW/mK at room temperature. Ca-Mn-O

film baring a high resistance was not the experimental determination of thermoelectric properties.

Keywords: Thermoelectric Thin Film; Ca3Co4O9; CaMnO3; Bipolar Pulsed-DC Magnetron Sputtering System

1. Introduction

In the past, the metal coating in form of thin films to im-

prove the quality of material was done by electroplating

process which is often also called electro-deposition. The

disadvantage of electroplating was harmful to the environ-

ment. Subsequently, the vacuum depositions were devel-

oped in chemical vapor deposition (CVD) and physical va-

por deposition (PVD). CVD is a technique whereby gase-

ous reactants can be deposited onto a substrate. However,

often dangerous by-products are removed by gas flow. PVD

is a clean coating technology that involves evaporation and

deposition of a material. Material vaporizes are removed

from a source by physical processes such as evaporation

sputtering and it is transported in the form of a vapor atomic

beam through a vacuum to the substrate. Magnetron sput-

tering is one of PVD methods, which are widely used in

thin film technology. The various types of magnetron sput-

tering technique are direct current (DC), alternating current

(AC), radio frequency (RF), and pulsed-dc [1]. Pulsed-dc

magnetron sputtering is one of the latest developments of

sputtering technology for thin films deposition, which has

many advantages over others. Namely, it is versatile and

provides the ability to deposit thin films of oxide com-

pounds at high deposition rate and to eliminate arcing

problems of poisoned targets [2]. This is interested to

apply the deposition technology. It may be possible to

customize the deposition conditions so that the thin films

of highly preferred orientation can be grown.

In this work, the depositions of thin films have been car-

ried out by a bipolar pulsed-dc magnetron sputtering sys-

tem using the Ca3Co4O9 and CaMnO3 targets, which were

made from powder precursors obtained from the solid state

reaction (SSR) route. Optical emissions from plasmas dur-

ing sputter deposition of thin films were measured using

a high resolution spectrometer. Crystal structures of the as-

deposited films were studied from X-ray diffraction (XRD).

The thickness of thin films and thermoelectric properties

were investigated.

2. Experimentation

The preparation of thin films by a pulsed-dc magnetron

sputtering system is shown in Figure 1 [3]. The details of

Preparation of Thin Films by a Bipolar Pulsed-DC Magnetron Sputtering System Using Ca3Co4O9 and CaMnO3 Targets

646



off

Time

Cathode

Anode

100 High Voltage Probe

Oscilloscope

Bipolar

Pulsed-DC

Power

Supply

+V

V

+







off

t

Vacuum

Chamber

Magnetron

Sputtering

Gun, Target

Argon (Ar)

Substrate

Figure 1. Experimental setup of a bipolar pulsed-dc magnetron

sputtering system.

deposition conditions, plasma and characterizations, and

thermoelectric properties measurements are given below.

2.1. Deposition Conditions

The sputtering targets were the p-Ca3Co4O9 and n-CaMnO3

pellets of 60 mm diameter, 2.5 mm thickness, 3.218 and

2.862 g/cm3 densities, respectively. The glass slide substr-

ates of 1.0 mm thick in dimension 25.0  50.0 mm2 were

used. The substrates were placed at a distance of 5.0 cm

above the targets and no additional heating was applied.

To generate the pulsed-dc plasma and initiate the thin film

deposition, the vacuum chamber was pumped down to a

base pressure of 2.00 N/m2 and flushed with high purity

argon (Ar 99.999%) gas flow rate of 15.0 ± 0.1 sccm the

total working pressures was 5.33 N/m2 for sputtering from

the Ca3Co4O9 and CaMnO3 targets. The pulse off time was

kept constant at 14 s (off and off

t). The reverse posi-

tive and cathode negative pulse widths of the power supply

were fixed at 10 s (on ) and 20 s (on

t), respectively.

These values of timings give the corresponding pulse

frequency of 17.24 kHz. The anode positive power was

set to be the same current-voltage of 20 mA and 100 V.

The cathode negative current fixed at 120 mA, the output

voltage were about 260 - 280 V with deposition time of

60 minutes. Here are the optimal conditions for the depo-

sition, which are summarized in Table 1.

t





t

2.2. Plasma and Characterizations

Optical emissions from plasma during sputter deposition

of films were observed in the wavelength range of 360 -

800 nm using a high resolution spectrometer (the getSpec-

2048 spectrometer, Sentronic GmbH) as shown in Figure

2. The spectral lines were indexed to the ASD data infor-

mation of National Institute of Standards and Technology

[4]. Crystal structures of as-deposited films were investi-

gated by X-ray diffractometer (PW3043 Philips X-ray dif-

fractometer of the Netherlands) at room temperature using

CuKα radiation,



= 0.15406 nm. Each film was meas-

ured in the 2-theta angle range of 10˚  2θ  70˚ with scan-

ning rate of 0.02˚/s. Thin film thickness can be estimated

from the optical interference using Tolansky’s Fizeau fringe

method which is now accepted [5]. The thickness (t) of the

Table 1. Deposition conditions of thin films.

Ar Flow RateFrequency

Positive Pulse Negative Pulse

 sccm kHz mA V mA  - V

Plasma

High Resolution

Spectrometer

Window

DetectorSubstrat e

Target

Vacuum Chamber

Figure 2. Observation of optical emission from plasma during

sputter deposition.

film is given by Equation (1) [6],







 





 



  (1)

where x is the displacement of fringes at step, x is the

distance between consecutive fringes, and



is the wave-

length of monochromatic light. The experimental arrange-

ment and fringe pattern is shown in Fi gure 3 . The film thic-

kness was measured by the Ellipsometer (Model L 115 S

300, Gaertner Scientific Corporation, USA) for com-

parison of the calculated values of thickness.

2.3. Thermoelectric Properties Measurements

The measurement of thermoelectric properties at room tem-

perature in air by the Keithley instruments included the

charge carrier, Seebeck coefficient, and electrical resistiv-

ity. The experimental setups can be elucidated as follows.

Firstly, the charge carrier and Seebeck coefficient were

determined by hot probe method [7,8] as shown in Fig-

ure 4. The hot and cold junctions between across two ends

of a film were connected to the digital voltmeter (Keithley

617 Programmable electrometer). The temperatures TH

and TC were sensed using the type K thermocouples, which

were connected to the digital thermometers (7563 Digital

thermometer, Yokogawa). The silicone thermal insulator

pads were placed between junctions and thermocouples.

The resistor of 10 W 5  was used to heat at hot junction

by applying currents to a resistor placed on hot side. See-

beck coefficient (S) was measured by the relation between

thermoelectric voltage (V) and temperature difference

(T). The S is defined as [8]:



 (2)

Secondly, the electrical resistivity was measured by four-

point probe method, which can be conveniently determined

by the Van der Pauw resistivity measurement technique

[9] as shown in Figure 5. All contacts were made by silver

paste, which showed ohmic characteristics over a wide

range of currents. The current-voltage characteristics for

Preparation of Thin Films by a Bipolar Pulsed-DC Magnetron Sputtering System Using Ca3Co4O9 and CaMnO3 Targets 647

Glass Slide Substrate

x

Monochromatic Light

Eye Piece Fringe Pattern

Partial Reflector

Film Thickness t x



Sample Film

Figure 3. Experimental arrangement for Tolansky’s Fizeau

fringe method.

Film

Silver Paste

Silicone Pad

Thermocouple

Wires

Glass Slide Substrate

Silver Paste

Silicone Pad

Digital

Thermo met ers

Figure 4. Side view (vertical cross section) of charge carrier

and Seebeck coefficient measurement.

Glass Slide Substrate

Film

a b

c d

Figure 5. Thre e-dime nsional top view of elec tric al re sistiv ity

measurement.

measurement of resistivity (



) are measured. The



can

be estimated from [9]

ab,dc bc,adab,dc

c,ad

ln2 2

RR R

tFR

















(3)

where t is the film thickness, Rab,dc = dc ab

VI, Rbc,ad =

ad bc

VI, and F is correction function which can be

calculated from







ab,dc bc,adab,dc bc,adln2RR RRF

arccosh



exp ln22F



Finally, the thermoelectric efficiency can be examined

from power factor (P), which was calculated from the S

and



in Equation (4) [7,10].

3. Results and Discussion

The results and discussion of plasma, characterizations, and

thermoelectric properties are given below.

3.1. Plasma and Characterizations

The cathode voltage waveforms, photographs of stable

glow discharges, and optical emission spectrums of the bi-

polar pulsed-dc magnetron argon discharge during the sput-

tering of the Ca3Co4O9 and CaMnO3 targets are shown in

Figure 6. These results indicated the good pulsed-dc plas-

ma characteristics. The spectral lines were indexed to the

ASD data information of NIST [4]. It was found that the

emission lines of Ar (422.57, 656.34, 696.66, 706.83,

738.55, 750.50, 751.57, 763.63, 772.51, 794.94 and 797.25

nm), Ca (370.26, 393.36 and 396.85 nm), Co (361.46 and

399.43 nm), and Mn (361.06 and 403.11 nm) were promi-

nent features. The emission line of O (777.41 nm) was de-

tected, but it is not intense due to the strong line of this

species is not in measured range. The optical emission

spectrums showed that Ca, Co, Mn, and O atoms were

sputtered from the targets. Hence, it can be expected that

400 500 600700 800

(b) CaMnO3

Ar: 696.66 nm

Mn: 403.11 nm

O: 777.41 nm

Ar: 794.94, 797.25 nm

Ar: 772.51 nm

Ar I: 763.63 nm

Ar: 750.50, 751.57 nm

Ar: 738.55 nm

Ar: 706.83 nm

Ar: 656.34 nm

Ar: 422.57 nm

Ca: 393.36, 396.85 nm

Ca: 370.26 nm

Mn:361.06 nm

Optical Emission Intensity

Wavelength (nm)

20 mA, +100±5 V

120 mA, 275±5 V

Waveform

Photograph

400 500 600700 800

(a) Ca3Co4O9

Ar: 696.53 nm

Co: 399.43 nm

O: 777.41 nm

Ar: 794.94, 797.25 nm

Ar: 772.51 nmAr I: 763.63 nm

Ar: 750.50, 751.57 nm

Ar: 738.55 nm

Ar: 706.83 nm

Ar: 656.41 nm

Ar: 422.57 nm

Ca: 393.36, 396.85 nm

Ca: 370.26 nm

Co: 361.46 nm

Optical Emission Intensity

Wavelength (nm)

Waveform

20 mA, +100±5 V

120 mA, 265±5 V

Photograph

Figure 6. Waveforms, photographs, and spectral lines dur-

ing the sputtering of (a) Ca3Co4O9 and (b) CaMnO4.

Preparation of Thin Films by a Bipolar Pulsed-DC Magnetron Sputtering System Using Ca3Co4O9 and CaMnO3 Targets

648

the deposited films will contain these atomic species.

From this point onward, the deposited films will be re-

ferred to as Ca-Co-O and Ca-Mn-O containing. The

as-deposited films of 1.60  1.60 cm2 and XRD patterns

are shown in Figure 7. From this figure clearly indicated

amorphous nature of the as-deposited films. This was

expected since the substrates were not heated during the

deposition process. Therefore, the kinetic energy of

atomic species at the substrate surface was not enough to

promote the growth of a crystal. Each film thickness was

initially estimated using optical interference method (yel-

low sodium light,



= 589.3 nm) and was obtained from

ellipsometic measurement (red laser light,



= 632.8 nm).

The results are given in Table 2. The results of meas-

urement gave the thickness of 435.31 nm and 449.35 nm

for as-deposited Ca-Co-O and Ca-Mn-O films, respec-

tively.

3.2. Thermoelectric Properties

The results of investigations on thermoelectric properties

of Ca-Co-O and Ca-Mn-O films such as the types of charge

carrier, Seebeck coefficient, electrical resistivity, and power

factor are presented and discussed.

Firstly, the types of charge carriers were determined by

hot probe method (see Figure 4). The result of measure-

ment on Ca-Co-O film, the cold junction showed higher

voltage than the hot junction, indicating that the holes con-

duction dominated transport property (p-type). The meas-

urement results of relation between Seebeck emf (V) and

temperature difference (T) are shown in Figure 8. Ca-

Co-O film indicated linear dependence between V and

10 20 30 40 50 60 70

Intensity (counts)

2 (degree)

(a) As-deposted Ca-Co-O Film

(a) As-deposted Ca-Mn-O Film

(b) As-deposited Ca-Mn-O Film

(a) As-deposited Ca-Co-O Film

(b)As-deposited Ca-Mn-O Film

(a)As-deposited Ca-Co-O Film

Figure 7. As-deposited Ca-Co-O and Ca-Mn-O films of 1.60 

1.60 cm2 and XRD patterns.

Table 2. Thickness of the as-deposited Ca-Co-O and Ca- M n- O

films.

Samples Calculation

(nm)

Measurement

(nm)

Refractive

Index

Ca3Co4O9

CaMnO3

~441.98

~452.50

435.31  3.37

449.35  1.27

2.22  0.01

2.06  0.01

024681012 14 16 18 20

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

(b) Ca-Mn-O Film: S = 0

(a) Ca-Co-O Film: S = 146.35 9.69 V/K





Seebeck emf (mV)

Temperature Difference (K)

Figure 8. Seebeck emf of thin films as a function of tempera-

ture difference (a) Ca-C o-O film and (b) C a-Mn-O film.

T, the S of 146.35 V/K is obtained. Ca-Mn-O film bar-

ing a high resistance was not the experimental determi-

nation of charge carrier and Seebeck coefficient.

Secondly, the current-voltage characteristics were ob-

tained by the Van der Pauw four-probe measurement (see

Figure 5). The experimental result of Ca-Co-O film is

shown in Figure 9. The plot exhibited good ohmic I-V

characteristics. The



value obtained from this I-V plot, it

is 0.473 cm. For the experimental measurement of Ca-

Mn-O film could not be determined.

Finally, the power factor was calculated from S and



in Equation: 2



. The result of Ca-Co-O film gave

value of 4.53 µW/mK.

4. Conclusion

The preparation of Ca-Co-O and Ca-Mn-O thin films

using a bipolar pulsed-dc magnetron sputtering system

were successfully deposited on glass substrates from the

Ca3Co4O9 and CaMnO3 targets, respectively. The XRD

patterns clearly indicated amorphous nature of the as-depo-

sited films. Determining thermoelectric properties of Ca-

Co-O film showed the low Seebeck coefficient and high

electrical resistivity, which leaded to a low power factor.

Ca-Mn-O film baring a high resistance was not an experi-

ment. The post deposition annealing and doped metals have

been expected candidates for good thermoelectric prop-

erties. This will be further investigated.

5. Acknowledgements

This work was financially supported by the Research and

Development Institute, Loei Rajabhat University (LRU).

The Thermoelectric Research Center (TRC) and Sakon

Nakhon Rajabhat University (SNRU) are acknowledged

for the preparation of sputtering targets. Asst. Prof. Dr.

Thanusit Burinprakhon, the Physics Department, Faculty of

Preparation of Thin Films by a Bipolar Pulsed-DC Magnetron Sputtering System Using Ca3Co4O9 and CaMnO3 Targets

649

-0.3-0.2-0.10.0 0.1 0.2 0.3

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8





Rab,cd = 2.187 0.021 k

dc (V)

Iab (mA)

(a)

-0.3-0.2-0.10.0 0.1 0.2 0.3

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8





Rbc,ad = 1.983 0.007 k

bc (V)

Iad (mA)

(b)

(a) (b)

-0.3-0.2-0.10.0 0.1 0.2 0.3

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8





Rcd,ab = 2.472 0.009 k

ab (V)

Idc (mA)

(c)

-0.3-0.2-0.10.00.1 0.2 0.3

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8





Rad,bc = 3.037 0.146 k

ad (V)

Ibc (mA)

(d)

Figure 9. Plot of the current-voltage characteristics for the electrical resistivity measurement of Ca-Co-O film (a) Iab-Vdc; (b)

Iad-Vbc; (c) Idc-Vab and (d) Ibc-Vad.

[4] J. Curry, “NIST Atomic Spectra Database,” 2012.

http://www.nist.gov/pml/data/asd.cfm

Science, Khon Kaen University (KKU) is gratefully ac-

knowledged for kind help with preparation of thin films by

an asymmetric bipolar pulsed-dc magnetron sputtering sys-

tem and thermoelectric characterizations.

[5] S. Tolansky, “An Introduction to Interferometry,” Kong-

mans, Green & Co. Ltd., London, 1955.

[6] K. Jayachandran, “Electrical, Optical and Structural Studies

in Bismuth, Antimony, Bismuth Oxide and Antimony

Oxide Thin Films,” Ph.D. Thesis, Mahatma Gandhi Uni-

versity, Kerala, 1997.

REFERENCES

[1] D. A. Glocker, S. I. Shah and W. D. Westwood, “Hand-

book of Thin Film Process Technology,” Institute of

Physics Publication, Philadelphia, 1995.

[7] B. V. Zeghbroeck, “Principles of Semiconductor Devices,”

2011. http://ecee.colorado.edu/~bart/book/

[2] J. Sellers, “Asymmetric Bipolar Pulsed DC: The Enabling

Technology for Reactive PVD,” Surface and Coatings

Technology, Vol. 98, No. 1-3, 1998, pp. 1245-1250.

doi:10.1016/S0257-8972(97)00403-9

[8] G. S. Nolas, J. Sharp and H. J. Goldsmid, “Thermoelec-

trics: Basic Principles and New Materials Developments,”

Springer-Verlag, Berling, 2001.

[9] L. J. van der Pauw, “A Method of Measuring the Resistivity

and Hall Coefficient on Lamellae of Arbitrary Shape,”

Philips Technical Review, Vol. 20, No. 8, 1958, pp. 220-224.

[3] W. Somkhunthot, T. Burinprakhon, I. Thomas, V. Amorn-

kitbamrung and T. Seetawan, “Bipolar Pulsed-DC Power

Supply for Magnetron Sputtering and Thin Films Synthe-

sis,” Elektrika: Journal of Electrical Engineering, Vol. 9,

No. 2, 2007, pp. 20-26.

[10] D. M. Rowe, “CRC Handbook of Thermoelectrics,” CRC

Press, Boca Raton, 1995. doi:10.1201/9781420049718