Investigation of radio frequency plasma for the growth of diamond like carbon films

Investigation of radio frequency plasma for the growth of diamond like carbon filmsIshpal, Sushil Kumar, Neeraj Dwivedi, and C. M. S. Rauthan Citation: Physics of Plasmas 19, 033515 (2012); doi: 10.1063/1.3694855 View online: http://dx.doi.org/10.1063/1.3694855 View Table of Contents: http://scitation.aip.org/content/aip/journal/pop/19/3?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Graphene diamond-like carbon films heterostructure Appl. Phys. Lett. 106, 102108 (2015); 10.1063/1.4914495 Photoconductivity and characterization of nitrogen incorporated hydrogenated amorphous carbon thin films J. Appl. Phys. 112, 113706 (2012); 10.1063/1.4768286 Influence of flow rate on different properties of diamond-like nanocomposite thin films grown by PECVD AIP Advances 2, 022132 (2012); 10.1063/1.4721654 Characterization of diamond-like nanocomposite thin films grown by plasma enhanced chemical vapordeposition J. Appl. Phys. 107, 124320 (2010); 10.1063/1.3415548 Structure and properties of ZrN doped diamondlike carbon films prepared by pulsed bias arc ion plating J. Vac. Sci. Technol. A 27, 1360 (2009); 10.1116/1.3248274

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

Investigation of radio frequency plasma for the growth of diamond likecarbon films

Ishpal, Sushil Kumar,a) Neeraj Dwivedi, and C. M. S. RauthanPhysics of Energy Harvesting Division, National Physical Laboratory (CSIR), Dr. K. S. Krishnan Road,New Delhi 110012, India

(Received 13 November 2011; accepted 28 February 2012; published online 29 March 2012)

The radio frequency has been used to generate plasma of argon, acetylene gases, and their mixture

should be replaced by mixture in a plasma enhanced chemical vapor deposition system. The generated

plasma discharge has been characterized by an impedance analyzer (VI probe) for the evaluation of

various electrical parameters of the plasma discharge such as rf-voltage, rf-current, phase, impedance,

and actual power consumed by the plasma discharge. These plasma parameters have been analyzed as

a function of self-bias, which are found to depend on applied power, pressure, and reactor geometry of

the system. Subsequently, same plasma conditions were used for the deposition of diamond like

carbon (DLC) films. The argon plasma has lowest impedance (16.02 X) value and highest average

electron density (2.77� 1010 cm�3) value at �150 V self-bias. X-ray photoelectron spectroscopy

(XPS) and Raman spectroscopy measurements have been performed on the prepared DLC films for

the evaluation of the chemical bonding. XPS studies have been used for the evaluation of sp3 and sp2

contents. The film deposited at �150 V self-bias has the highest values of sp3 content (60.97 at. %),

band gap, nanohardness, elastic modulus, plastic index parameter, and elastic recovery, and the lowest

value of sp2 content (27.27 at. %) among the films chosen for the present investigation. These DLC

films properties were found to be well correlating with the evaluated plasma parameters. VC 2012American Institute of Physics. [http://dx.doi.org/10.1063/1.3694855]

I. INTRODUCTION

In last few decades, plasma based techniques such as

plasma assisted etching, plasma enhanced chemical vapor

deposition (PECVD), plasma polymerization, and the plasma

induced surface modification etc., are extensively used for

material synthesis and processing.

The synthesized materials by these techniques have a

wide range of applicability in device fabrication for automo-

bile, aerospace, bio medical industries, microelectronics

components, solar cells, surface hardening by plasma nitrid-

ing, etc.1–3 Among the above mentioned plasma based proc-

esses,4,5 the RF discharge have found extreme importance in

fundamental research as well as applications due to increas-

ing demand of large area processing of highly uniform and

good quality films of desired properties. This has motivated

the researchers for precise investigations of capacitively

coupled plasma discharge parameters. In this regard, inde-

pendent control of flow of active neutrals, ionic species with

specific energies, and electron density is become

important.6–14 In these types of processes, the properties of

synthesized materials strongly depend upon the ion energy

and electron density, which are controlled by the dissipated

power in the plasma. In the capacitive coupled RF plasma,

the applied power is consumed by two processes, power dis-

sipated by ions in the sheath and by electrons collision as

stochastic heating in bulk plasma. When discharge current is

low, then the most of the power is consumed by the electrons

in the bulk of the plasma, and the delivered (dissipated)

power shows almost linear dependency to the discharge cur-

rent (P¼Pbulk¼Vb.I). However, when discharge current is

large, then most of the power is dissipated for the accelera-

tion of the ions in the sheath region and the delivered power

to the plasma shows almost a dependency on the square of

the discharge current (P¼ Psheath¼Rs.I2). Thus, with the

increase of RF current, the power characteristic changes

from linear to square dependence of RF current which indi-

cate the electron power dissipation mode to ion power dissi-

pation mode transformation.15–20

Beneking15 first suggested about the power dissipation

transition mode with the discharge current and driving fre-

quency and later experimentally confirmed by Godyak

et al.16,17 and You et al.19,20 through electrical measurements.

Godyak et al.16 and You et al.18 have also shown the effect of

gas pressure and transverse magnetic field on the power dissi-

pation transition mode, respectively. In the capacitively

coupled RF plasma CVD system, the process parameters such

as gas flow rate, gas pressure, applied power, frequency, self-

bias and system geometry, etc. play vital role in deciding the

properties of the deposited thin films. Thus, it becomes neces-

sary to optimize these process parameters for achieving the

desired properties of the films. However, complete understand-

ing of the processes occurring inside the plasma during deposi-

tion of the thin film is very complex. Thus, it becomes

essential to have precise control and monitoring of the plasma

parameters, separately. For the proper selection of these pa-

rameters and repeatability of process conditions for the growth

of quality materials, diagnosis of the plasma characteristics

and determination of plasma parameters is critical.15–30 In this

a)Author to whom correspondence should addressed. Electronic mail: skumar@

nplindia.org. Telephone: 0091-11-45608650. Fax: 0091-11-45609310.

1070-664X/2012/19(3)/033515/14/$30.00 VC 2012 American Institute of Physics19, 033515-1

PHYSICS OF PLASMAS 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

context, several plasma diagnostic tools such as Langmuir

probe, optical emission spectroscopy (OES), mass spectrome-

try, impedance probe analyzer, etc., have been used by several

research groups.1–5,15–30 However, it is difficult for a single

diagnostic tool to give complete information about the physi-

cal and chemical processes of the plasma discharge. Imped-

ance probe analyzer has shown its usefulness in plasma

diagnostic by monitoring the RF voltage and current and the

phase difference between them, which finally leads to estimate

the plasma impedance for power dissipation. Impedance probe

analyzer is a newly developed ex-situ diagnostic technique and

could be better than the in-situ diagnostic technique due to

imperturbation of plasma as generally done by in-situ tech-

nique. Another main setback with the in-situ technique is the

deposition of the material which can produce wrong results.

But, the main problem with using the ex-situ techniques for

the measurement of the plasma parameters is stray impedance

in the electrical circuit. Since, in case of ex-situ diagnostic

techniques, the point of measurement remain outside the reac-

tor, which results stray impedance between the point of mea-

surement and the actual electrode. In order to minimize the

effect of stray impedance, several methods have been purposed

in the past by different research groups4,5,8,31–33 depending

upon the impedance analyzer. Sobolewski31–33 has also used

the similar type of impedance probe analyzer for the plasma

diagnostic and proposed various electrical equivalent circuit

modals to reduce the stray impedance. He has also studied the

effect of substrate impedance on the ion energy.34 But, the

most suitable to our system and commonly used method for

the impedance probe analyzer is the plasma real time equiva-

lent circuit model purposed by Spiliopoulos et al.4 and Hargis

et al.,35 where the equivalent circuit component were eval-

uated by measuring the plasma parameters (impedance) with

or without discharge. In the present investigation, we have

adopted the same approach and the impedance of the empty

cell without discharge was found to be �5–7 X. Recently, we

have used an impedance probe analyzer (VI probe) for the

characterization of silane plasma for the growth of nanocrys-

talline silicon films,30,36 and the same probe analyzer has been

used in the present investigation for the growth of diamond

like carbon (DLC) films.

The main objective of the present investigation is to

characterize acetylene gas discharge plasma parameters

using VI probe for the deposition of diamond like carbon

coatings. Plasma parameters were estimated as a function of

self-bias in the power dissipation mode for argon and acety-

lene plasmas separately and then their mixture. The gaseous

mixture of argon and acetylene was used under same plasma

conditions for the deposition of DLC films. The deposited

films were characterized for electrical, optical, and chemical

bonding and correlated with the plasma parameters.

II. EXPERIMENTAL DETAILS

An asymmetric capacitively coupled radio frequency

plasma enhanced chemical vapor deposition (RF-PECVD)

system has been used for plasma diagnostic by VI probe and

deposition of DLC films. The power electrode surrounded by

a grounded shield separated by 6 mm ceramic insulator was

connected to the RF power source through the matching net-

work operated at a conventional frequency of 13.56 MHz,

while the other electrode was grounded. The power electrode

(cathode) of diameter 19.9 cm is much smaller than the

ground electrode (anode) separated by a distance of 9.2 cm,

includes the wall of the chamber having diameter �30 cm

and height �25 cm. The RF electrode was connected to L

type matching network. The brief descriptive view of the

system used for the present study is shown in Fig. 1.

The VI probe used for the plasma diagnostic is a MKS,

USA make (M/s ENI. INC) VI probe 350. The probe was

mounted on the power electrode as close as possible after the

matching network. Electrical characteristic measurements

FIG. 1. (a) Schematic representation of PECVD system connected with VI probe. (b) Equivalent circuit model of plasma generated by asymmetric capaci-

tively coupled RF-PECVD system.

033515-2 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

such as voltage, current, impedance, phase difference, and

delivered power to the plasma were performed using this VI

probe. The voltage and current signals were recorded by

MKS analysis through the software controlled by computer

and subsequently, other plasma parameters were derived.

These experiments were carried out at various self-biases

ranging from �50 to �300 V at a constant pressure of

3.5� 10�2 Torr of argon, acetylene, and their mixtures.

Under the same plasma conditions, diamond like car-

bon films were deposited in the mentioned RF-PECVD sys-

tem on highly cleaned Si wafers and 7059 glass substrates.

The deposited films were characterized for bonding struc-

tures using x-ray photoelectron spectroscopy (XPS) etc.

The XPS measurements were carried out on the 1� 1 cm2

Si (100) substrates using Perkin-Elmer (model no. 1257)

instrument. Mg Ka (1253.6 eV) monochromatic x-ray radia-

tions were used for these investigations. The XPS general

survey scan were acquired using a 100 eV pass energy at a

step of 1.0 eV and XPS C1s core level spectra were

acquired at 0.03 eV step with a pass energy of 60 eV. A me-

tallic clamp was used to eliminate the charging effect from

the film surface during XPS measurement. The electrical

conductivity of DLC films was measured on the glass sub-

strate at co-planer structure using Keithley 610 C solid-state

electrometer. The nano indentation measurements have

been performed for hardness and the young modulus of

elasticity on the prepared samples using IBIS nonoindenta-

tion (Fisher-Cripps Laboratories Pvt. Ltd, Australia).

Raman spectroscopic measurements have been performed

on the prepared samples by Reinshaw InVia Reflex micron

Raman spectrometer. Air cooled argon laser of wavelength

�514.5 nm was used as a source.

III. RESULTS AND DISCUSSION

A. Theoretical aspect of power dissipation in RFplasma as a function of RF-current

Various research groups have performed theoretical and

experimental studies of the plasma parameters as a function

of external parameters such as gas pressure, magnetic field,

driving frequency, RF current, and RF power.15–24,31–33 On

the basis of theoretical calculations, Beneking15 have derived

empirical formulas for the estimation of plasma parameters

when plasma is in power dissipation mode. They suggested

that the plasma impedance is a very critical deciding parame-

ter for the behavior of plasma characteristic. The total im-

pedance of the plasma is the sum of the bulk resistance (Rb)

and sheath impedance (resistive component (Rs) and reactive

impedance (X) of sheath). The determination of resistive

component is more important for the precise determination

of dissipated power in the plasma, because the power con-

sumed in the plasma is directly proportional to the resistive

component of the plasma impedance. The total resistive

component of plasma impedance is the sum of resistive com-

ponent (Rs) of sheath impedance and bulk resistance (Rb)

i.e., R¼Rs þ Rb.

Thus, the total power consumed in the plasma is the sum

of the power consumed by the ions in sheath and power con-

sumed by the electron in the bulk of the plasma which is

given by an expression as

Ptotal ¼ Rs:I2 þ Rb:I

2 ¼ Rs:I2 þ Vb:I;

where Rs is a function of the rf current, pressure, and applied

frequency, whereas Vb an ohmic part of the rf-voltage across

the bulk plasma which is independent of discharge current

and the applied frequency and depends on collisional and

stochastic electron heating. The sheath resistance (Rs)

account for the power dissipation in the sheath for the ion

acceleration through sheath, while the bulk plasma resistance

(Rb) account for the power dissipation for electron heating in

the bulk plasma.

These resistances can be expressed as19,20

Rb ¼ const:� Imðm � �1Þ; (1)

Rs ¼ const:� I1=2

x5=2¼ ð

ffiffiffi2pÞ5=2 5

3e0A

� �3=22k

3ffiffiffiffiffiPg

p ! ffiffi

Ip

x5=2

� �;

(2)

where A is an area of the power electrode, e0 is a permittivity

in vacuum, k is mobility constant, Pg is the gas pressure,

and ƒ (x� 2pƒ) is the applied frequency. The expression

given in the Eq. (2) is valid for unmagnetized plasma. How-

ever, for magnetized plasma, the effective gas pressure is

given as

Peff ¼ Pg þ n:B2

Pg

; (3)

where B is magnetic field (G) and n¼ 0.83 (mTorr2/G2) is a

constant

Thus, the total delivered power for unmagnified plasma

can be written as

Ptotal ¼ ðffiffiffi2pÞ5=2 5

3e0A

� �3=22k

3ffiffiffiffiffiPg

p !

I

x

� �5=2

þ Vb:I (4)

and

Psheath

PBulk

� �¼ const:� I3=2ffiffiffiffiffi

Pg

px5=2

: (5)

It has been seen from the Eqs. (1) to (5) that the plasma

power is a function of RF current, effective gas pressure,

magnetic field, applied frequency, and system geometry. The

bulk resistance is inversely proportional to RF current, and

sheath resistance is directly proportional to square root of RF

current. Thus, for a low value of RF current at constant gas

pressure and frequency in unmagnetized plasma for fixed

defined system geometry, the sheath resistance has a low

value, while the bulk resistance has a large value. Thus, for

low value of RF current, most of the power is dissipated by

electron in the bulk of the plasma for collisional and stochas-

tic heating. Subsequently, plasma resistance decreases

almost linearly, while the delivered power increases linearly

033515-3 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

with RF current. This results in an increase of ionization and

dissociation of gaseous molecules and hence, electrons

(plasma) density increases.

There is a decrease of bulk resistance and increase of

sheath resistance with the increase of RF current. Under

these conditions, power dissipation takes place inside the

sheath for ions acceleration and delivered power show an I2

dependency. Thus, with increase of RF current, the power

dissipation transition takes place from electron to ion power

dissipation mode, since the sheath resistance is inversely pro-

portional to the applied frequency, gas pressure, and mag-

netic field. Thus, with the increase of any one of these, the

sheath resistance decreases and the bulk resistance increases,

therefore most of the power is dissipated in the bulk of

plasma and power dissipation mode transition takes place

from ion to electron power dissipation mode. With increas-

ing of RF current, the power dissipation in sheath increases,

while increasing of the pressure and frequency, the bulk

power dissipation increases. You et al.18–20 have shown the

experimental verification of power dissipation transition

mode with different RF frequency, magnetic field, and gas

pressure for argon gas plasma. On the other hand, Lee

et al.23,24 showed the power dissipation mode for the inert

gas such as argon, neon, krypton and for reactive silane gas

for solar cell application.

B. Theoretical aspect of power dissipation in RFplasma as a function of self-bias

Power dissipation in RF plasma as a function of RF cur-

rent, frequency, magnetic field, and gas pressure is well dis-

cussed in the literature.18–24 But, still there is one more

uncovered deposition parameter i.e., self-bias, which

strongly controls the properties of thin films during deposi-

tion. The properties of diamond like carbon thin films depos-

ited by PECVD technique depends upon various process

parameters such as RF power, RF frequency, gas pressure,

type of gas or gaseous mixture, self-bias and system geome-

try, etc. Out of these, self-bias (Vsb) is a critical parameter,

which decides the ultimate properties of the deposited thin

films particularly during the ion assisted deposition. As the

self-bias Vsb¼ a 1A2, Vsb¼b P

P1=2g

then Vsb¼ k P

A2P1=2g

and

Vsb¼Ei. Where a, b, and k are the proportionality constants,

and A, P, Pg, and Ei are the area of the electrode, RF power,

gas pressure, and ion energy, respectively. Thus, the self-

bias depends upon the system geometry, gas pressure, RF

power, and type of the gas, and all these are directly related

to the ion energy, the most critical parameter. Self-bias gov-

erns the ion energy (Ei) to precursor radical used for the dep-

osition of DLC films. However, ion energy (Ei) is not

directly measurable in RF-PECVD system, usually the dis-

charge power and hydrocarbon pressure (Pg) are being used

to control the self-bias. For the deposition of DLC films of

desired properties, these external parameters need to be con-

trolled with a high precision separately. It becomes easier if

one need to control lesser number of parameters to obtained

desire properties of the films. Thus, just by monitoring the

self-bias, the properties of the deposited thin film can be con-

trolled in the PECVD process.1,36–42 Thus, the above equa-

tions in terms of self-bias can be rewritten as

Ptotal ¼ ðffiffiffi2pÞ5=2 5

ffiffiffiffiffiPg

p3e0

!3=22k

3

ffiffiffiffiffiA7

k5

s0@

1A Vsb

Vx

� �5=2

þ Vb:A2Vsb

ffiffiffiffiffiPg

pkV

!(6)

and

Psheath

PBulk

� �¼ const:� ðAVsbÞ3=2

V3=2x5=2: (7)

In the present study, we have used fixed gas pressure

(3.5� 10�2 Torr) and constant RF frequency (13.56 MHz) in

a fixed defined PECVD system geometry and monitor the

plasma parameters as a function of self-bias, for which A,

Pg, and x are constant. Thus, Vsb¼ f P¼ f VI, symbols have

their usual meaning.

Therefore, self-bias (Vsb) is proportional to RF current

(I) and RF voltage (V). If we replace RF current (I) by Vsb

V in

Eqs. (1), (2), and (5), then plasma parameters have the simi-

lar behavior with self-bias as that with RF current and equa-

tions can be written as

Rb ¼ const:� V

Vsb; (8)

Rs ¼ const:�ffiffiffiffiffiffiffiVsb

V

r; (9)

Psheath

PBulk

� �¼ const:� Vsb

V

� �3=2

: (10)

It is revealed from Eqs. (8) to (10) that for a low value of self-

bias the bulk resistance has a higher value than the sheath re-

sistance and most of the power is consumed in the bulk

region. With the increase of self-bias, the bulk resistance

decreases while the sheath resistance increases which results

in an increase of dissipation of power in the sheath region.

Then, the power dissipation transition takes place from elec-

trons power dissipation in bulk region for stochastic electron

heating to ions power dissipation in sheath region for ion

acceleration. In the present study, the power dissipation transi-

tion mode for argon, acetylene, and their mixture as a function

of self-bias has been shown, and same conditions has been

used for the growth of diamond like carbon coatings.

C. Experimental verification of power dissipationmode as a function of self-bias for diamond likecarbon coatings

The plasma discharge was characterized for RF current,

RF voltage, and the phase difference between these two and

plasma impedance etc. with respect to the self-bias at a con-

stant pressure of 3.5� 10�2 Torr of argon, acetylene,and

their mixture (argon 30% and acetylene 70%) by VI probe.

The schematic diagram of RF-PECVD system connected

033515-4 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

with VI probe is shown in the Figure 1(a). Various research

groups3,42–45 have proposed different equivalent circuit mod-

els depending upon their system geometry for the characteri-

zation of plasma by the impedance analyzer. In the present

investigation, the equivalent circuit model (Figure 1(b)) pro-

posed by Kohler et al.3 has been used as it is most suited for

asymmetric RF-PECVD system geometry. The behavior of

electrons and ions into the sheath and bulk region of the

plasma decides the properties of the plasma. The bulk

plasma is characterized by electron motion and represented

by simple resistance (Rb), while the sheath formation is a

combine effect of ion motion in the sheath, electron conduc-

tion through sheath, and capacitive component of the sheath.

The sheath is represented by a parallel combination of resist-

ance (Rsp), capacitance (Csp), and diode (Dsp) for ion motion,

displacement current, and electron conduction, respectively,

at the power electrode. The combination of resistance (Rsp)

and diode (Dsp) resistance gives the real part of the sheath re-

sistance (Rs), while the capacitive resistance provides the

imaginary part (X) of sheath impedance. A similar combina-

tion of resistance (Rsg), capacitance (Csg), and diode (Dsg)

can also used for sheath at grounded electrode (shown in

Fig. 1(b)).

Figure 2 shows the variation of rms value of RF current

and RF voltage as a function of self-bias in the range of

�50V to �300 V for the plasma discharge of argon, acety-

lene, and their gaseous mixture separately at a constant pres-

sure of 3.5� 10�2 Torr. It is evident from the figure that the

RF current and RF voltage show linear behavior with self-

bias, as RF current and RF voltage are directly perportional

to self-bias for constant gas pressure and frequency for sys-

tem having fixed geometry. Thus, the self-bias have similar

effect on the dissipated power as that of RF current i.e.,

whether power dissipated by electrons in bulk or power dis-

sipated by ions in sheath. It is revealed from the figure that

the argon has highest value of rms current and acetylene has

a lowest value, while the value of rms current for their mix-

ture lies between argon and acetylene. However, RF voltage

has a lowest value for argon and highest value for acetylene.

The total current flow through the discharge is the sum of

electron conduction current (Ie), ion conduction current (Ii),

and the displacement current (Id).

Thus, the total current density is given by the expression

as

J ¼ Je þ Ji þ Jd ¼ eniui þ eneue þ xe0E;

where E is the RF electric field, ni and ne are the ion and

electron density, respectively, ui and ue are the ion and elec-

tron drift velocity, e is the electronic charge, and eo is the

permittivity of free space (vacuum). The current flow in the

bulk plasma is mainly due to electrons, while the displace-

ment current is due to ions flow in the sheath region.30,46 The

increase of current with self-bias may be the result of ions

acceleration in the sheath region and ion current contribution

increases in the total current. This is because of the transition

of the electron power dissipation mode to ion power dissipa-

tion mode with the increase of self-bias.

Figure 3 shows the dependence of phase angle (/) with

the self-bias of various plasma discharges argon, acetylene,

and their mixture. It is evident from the figure that the phase

difference between rms voltage and rms current first decreases

(less negative) and then increases (more negative) with the

increase of self-bias. This indicates the transition of resistive

to capacitive discharge regime with self-bias takes place and

plasma become ohmic to capacitive. These results are well

corroborated with reported literature’s results.4,29,47,48

The plasma impedance is another important parameter

which plays a vital role in deciding the properties of the de-

posited thin films. The understanding of plasma impedance

and its role in the characteristic of plasma is very important.

The complex impedance Z is given by expression Z¼R þ jX

having an absolute value of jZj ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiR2 þ X2p

. The value of

plasma impedance is evaluated from the rms voltage, rms cur-

rent, and phase difference between them. The absolute value

FIG. 2. Variation of rms value of rf current and rf voltage with self-bias for

the plasma of argon, acetylene, and their mixture with self-bias.

FIG. 3. Variation of phase difference between RF current and RF voltage

for the plasma of argon, acetylene, and their mixture with self-bias.

033515-5 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

of impedance is given by the expression jZj ¼ VI , while the

resistive (real) component of the plasma impedance is given

by R¼ jZj cos(/) and the reactive (imaginary) component is

given by X¼ jZj sin(/).

As discussed earlier, the total impedance of the plasma

is the sum of the bulk resistance (Rb) and sheath impedance

(resistive component (Rs) and reactive impedance (X)),

while the total resistive component of plasma impedance is

the sum of resistive component (Rs) of sheath impedance

and bulk resistance (Rb) i.e., R¼Rs þ Rb.

It evident from the Figure 4(a) that the absolute value of

plasma impedance initially decreases upto �150 V self-bias

and then increases beyond �150 V self-bias. This may be

because of dependency of the plasma impedance on self-bias

similar to RF current. As for low value of RF current, the

plasma resistance shown inversely depends on rms value of

RF current (Rb¼ const.� Im m��1); while for higher value

of RF current, the plasma resistance shows a directly depend-

ence on rms current (Rs¼ const.�ffiffiIp

). This means there is a

change of electron power dissipation to ion power dissipa-

tion transition mode with the increases of self-bias. Most of

the power consumed by ions in the sheath region at higher

values of self-bias. It can also be seen that argon have lowest

value of plasma impedance and acetylene has highest value,

while the value of impedance for the mixture of argon and

acetylene lies between argon and acetylene. It is also

revealed from the figure that the value of reactive component

(X¼ 1jxC) of plasma impedance decreases, while the resistive

component (R) increases with self-bias. This could be due to

dependence of phase difference between rms current and rms

voltage on cosine and sine functions, respectively. The resis-

tive component of plasma impedance for argon has highest

value, while acetylene has lowest value. This indicates that

maximum power dissipation in the bulk takes place for argon

and lowest for acetylene. As a result, argon gas easily disso-

ciate as compared to acetylene. Figure 4(b) shows the varia-

tion of bulk and sheath resistance as a function of self-bias.

It is revealed from the figure that the bulk resistance

decreases, while the sheath resistance increases with the

increase of self-bias. It indicates that the power consumption

in the bulk region decreases and power consumption in

sheath region increases and power dissipation in bulk to

power dissipation in sheath region for ions acceleration takes

place. Since the plasma impedance has a lowest value in the

range of �100 to �200 V of self-bias, thus, good quality of

DLC films can be deposited in this moderate range of bias.

At higher self-bias, the bombardment of the accelerated ions

takes place on the depositing film surface as due to dissipa-

tion of most of the power in the sheath. The DLC deposited

films in such condition may not be having excellent diamond

like properties. Thus, one not only need to worry about

appropriate selection of self-bias but also about the imped-

ance of plasma for getting DLC films of desire properties.

The observed results from the experimental study of plasma

impedance are well corroborated with the literature in this

area of study.49

Figure 5 shows the variation of delivered power to the

plasma and ratio of power dissipated in sheath to power dis-

sipated in bulk with self-bias for argon, acetylene, and their

mixture. The delivered power to the plasma is consumed by

the two processes, power dissipated by ions in the sheath and

by electron for stochastic heating in the bulk plasma. It is

revealed from the figure that the delivered power increases

with the increases of self-bias and has highest value for ar-

gon and lowest for acetylene. It also indicates that maximum

part of the forward power consumed for the argon plasma

discharge. This behavior of delivered power is an indication

of some sort of ohmic nature of power consumption of the

plasma and given by VI cos(/). However, the delivered

power shows non linear behavior with self-bias, which may

be due to transition of electron to ion power dissipation

mode of plasma. The non linear behavior of delivered power

is also confirmed by other groups.44,50 Since the delivered

power is the sum of the power consumed in the sheath region

FIG. 4. Variation of plasma impedances

(a) absolute value of impedance, resis-

tive component, and reactive compo-

nents; (b) sheath resistance and bulk

resistance for various gases with self-

bias.

033515-6 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

and bulk region. As the ratio of power consumed in sheath to

power consumed in bulk (Psheath/Pbulk) also increases with

self-bias, this confirms the transition of power dissipation

from bulk to sheath. It is elucidated from the Figure 5 that

the delivered power and Psheath/Pbulk have highest value for

argon and lowest value for acetylene, while the delivered

power value for gaseous mixture of argon and acetylene lies

between these gases.

Another important factor to characterize plasma is the

electron density. Since for a collissionless plasma the elec-

tron have an energy of the form Ek¼e2E2

0

2Mex2, where e is the

electronic charge, Me mass of the electron, E0 is the applied

electric field, and x is the driving (applying) frequency. The

energy of the electron is of the order of 11.2 eV for an elec-

tric field of 10 V/cm, then while the ionization energy of the

argon ion is 15.8 eV. Thus, for collissionless plasma, one

needs to apply a stronger electric field. But, when the transi-

tion takes place from collissionless plasma to collisional

plasma at some what higher pressure at a certain applied rf

power, then the electrons may make elastic collision with the

atoms or gas molecules and continuously gain energy from

the applied power. After some successive collisions, they

may have sufficient energy for the ionization of gaseous mol-

ecules through inelastic collision. The power consumed by

an electron is P¼ e2E20

2Me

��m

�2mþx2

�and the effective electric field

increases and has a form of Eeff¼e2E2

0

2Me

��m

�2mþx2

�, where, �m is

the momentum transfer collision frequency for the collisions

between electron and gaseous molecules or atoms. If there

are ne electrons present in the unit volume of gaseous

plasma, then the power consumed by the electron is written

as P¼ nee2E20

2Me

��m

�2mþx2

�and for a volume (U) of gaseous plasma,

the total power consumed by the electron is given as

P¼ neUe2E20

2Me

��m

�2mþx2

�. The driving frequency does not contrib-

ute much, for the frequency �107 Hz and collisional momen-

tum transfer frequency �109 Torr�1 s�1. Thus, the power

consumed by the electron can be written as P¼ neUe2E20

2Me�mand

the average electron density can be written as ne¼ PMe�m

Ue2E20

.

The average electron density1,4,29,30,51 can also be written as

ne¼ 2�mdMe

jZjAe2 , as P¼VI, E0¼ Vd and U¼A.d, V

I ¼ jZj, where A

is the power electrode area and d is inter-electrode distance.

Figure 6 shows the variation in the average electron den-

sity of the plasma with the self-bias for argon, acetylene, and

their mixture. The average electron density for all three

plasma chosen for the present investigation first increases

with the self-bias and then decreases. As average electron

density is directly perportional to the delivered power and

inversely proportional to plasma impedance. Thus, it shows

a reverse trend of variation of plasma impedance with self-

bias. It has been observed that the average electron density

have a highest value and lowest impedance value in the

range of �100 to �200 V self-bias. Also in this range of

self-bias electron to ion transition takes place i.e., the ions

have a moderate energy which may be optimum range for

the deposition of good quality DLC films. Thus, this trend of

average electron density confirms the similar power dissipa-

tion mode transitions as observed in the case of plasma im-

pedance variation with self-bias. The average electron

density for the argon has highest average electron density

and acetylene has lowest average electron density value,

while for their mixture, the average electron density lies

between them as similar to observed for other plasma proper-

ties in the present investigation and also estimated from theo-

retical studies.

D. Correlation of experimental plasma parameterswith the reported literature

It is known that higher electron density is necessary for

higher deposition rate and optimum ion energy is also neces-

sary for the deposition of good quality films. The ion energy

FIG. 5. Variation of delivered power to the plasma and sheath power to

bulk power ratio for various gases with self-bias.

FIG. 6. Variation of average electron density for different gases with

self-bias.

033515-7 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

plays a crucial role in thin film deposition, and Sun et al.52

have suggested that 30-300 eV ion energy is beneficial for

good quality DLC films. While the several research

groups53–59 have observed that the DLC films deposited at

�100 V self-bias or with 100 eV ion energy (as 1 eV corre-

sponds to 1 V) have the more diamond like properties. They

suggested that the properties of the DLC films deteriorate

above this energy values (100 eV) and transformation of

diamond-like (highly sp3 hybridized) to graphite-like (having

high sp2 content) properties takes place due to ionic bom-

bardment which favors sputtering/etching or bond breaking

process at the film surface. They have also suggested that the

dangling bonds are created by the ionic bombardment by

breaking weaker bonds. For the C-C bonds, �607 kJ/mol

energy is required to break the bonds; while for C-H bonds,

337.2 kJ/mol energy is required, in DLC films. Thus, it is

easy to break a C-H bond as compared to C-C bond and

these breaking of C-H bonds create more dangling bonds.

When these dangling bonds recombine with the adjacent

dangling bonds, they transform the sp3 hybridized carbon

atoms into sp2 hybridized carbon atoms. This results into

graphite like properties of the deposited films. Thus, the ion

energy is one of the most important parameter for the proper-

ties of the DLC films.15–24

The electrical properties of the DLC films are governed

by sp2 hybridized carbon atoms, while the mechanical prop-

erties are depend upon the sp3 hybridization of the carbon

atoms. Enormous research has been carried out by various

research groups52–64 on the effect of self-bias on diamond

like hard coatings. It has been observed that with the increase

of self-bias, various properties of diamond like carbon films

like hardness, elastic modulus, stress, sp3 content etc.

increases and roughness value decreases upto a critical value

of self-bias and after that the films properties deteriorates.

Some research groups52,61–64 have observed that the �100 to

�200 V self-bias is beneficial for the good quality DLC

films, and most of the properties of DLC films have a point

of inflection at �150 V of self-bias. DLC films deposited in

this range of self bias have shown high values of hardness,

stress, elastic modulus, refractive index and sp3 content and

lower values of rms roughness, friction coefficient, wearing

and hydrogen content, and good conductivity values. Films

deposited above �200 V self-bias shows sharp increases of

conductivity, whereas the mechanical properties deteriorate

due to increase of sp2 content. Above �200 V self-bias, the

highly energetic ion bombardment takes place on the grow-

ing film surface due to dissipation of most of the power in

the sheath region which used for the acceleration of the ions.

As a result, inter-diffusion of hydrogen and de-link/linking

of bonds like phenomenon may takes place which transform

the sp3 hybridized site into sp2 hybridized sites. Thus, opti-

mization of the plasma parameters is necessary, for the DLC

films deposition having desire properties.

E. Correlation of experimental plasma parameterswith the properties of the deposited DLC films

To correlate the plasma parameters with the properties

of DLC films, XPS measurements have been carried out on

all the DLC films chosen for the present investigation with-

out preliminary argon ion cleaning to avoid the breaking of

bonds on the films surface. Figure 7(a) shows the characteris-

tic XPS spectra of the DLC film deposited at �150 V self-

bias, which confirm that along with main carbon component,

some oxygen also present on the film surface. The observed

position of peaks at 285 and 532 eV have been attributed to

C 1 s and O 1 s, respectively, which are well corroborated

with the reported literature values.66

Figure 7(b) shows the XPS spectra of carbon (C 1 s) of

the DLC films deposited at different self-biases ranging from

�100 to �300 V, which consists of a quite large asymmetric

peak which indicating the presence of carbon atoms in differ-

ent bonding states.67 The peak position of C 1 s first shifted

toward higher binding energy side up to �150 V self-bias

and after �150 V self-bias, the peaks position of the C 1 s

again shifted toward lower binding energy side, which indi-

cating the transformation of sp3 sites into sp2 sites and the

properties of DLC films from diamond-like to graphite-

like.68 This might be because of increasing of the ion energy

beyond the critical energy value which results in sputtering/

etching types of phenomena on the film surface due to con-

sumption of most of the power in the sheath region. Since

with the increases of self-bias, the RF current increases along

with sheath impedance, while the sheath resistance and the

electron density decreases, which indicates the transition of

resistive to capacitive discharge regime. Thus, as we increase

self-bias, the transformation of dissipation of power in bulk

to dissipation of power in sheath takes place, which acceler-

ate the ions in sheath region. When these accelerated ions

strikes on the film surface, they transform the sp3 (diamond

like) bonding into sp2 (graphite like) bonding and film

become more graphitic like. The binding energy peaks posi-

tion of graphite (sp2-hybridize) and polycrystalline diamond

film (sp3 hybridize) are found to be at 284.5 eV (Ref. 65) and

285.3 eV (Ref. 69), respectively.

The evaluation of sp3 contents is critical for deciding the

properties of the DLC films. Therefore, for the evaluation of

sp3 content in DLC film, the C 1 s peak has been deconvo-

luted into three components of a mixture of 2:1 of Gaussian

FIG. 7. (a) XPS general survey scan of DLC film deposited at �150 V self-

bias; (b) C 1 s spectra of DLC films deposited at different self-biases.

033515-8 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

and Lorentzian peaks after subtracting the background from

the spectra by Shirley mode.

Figure 8 shows the deconvoluted spectra of C 1 s peaks

of DLC films deposited at various self-biases. The peaks

were located at 284.4 6 0.1, 285.3 6 0.2, and 287.1 6 0.4 eV

and were assigned to C¼C sp2 hybridized carbon in

graphite-like carbon-carbon bonds, sp3 hybridized carbon in

C-C in diamond like structure and bonding state of C with O,

respectively. The observed peak positions in the present

DLC films match well to the values reported earlier in

literature.70–73

The sp2 and sp3 contents of the DLC films are estimated

by the expressions as, sp2 content¼ A2

A2þA3þAOand sp3

content¼ A3

A2þA3þAO. where A2, A3, and AO are the areas under

the curve having peaks positions at 284.4 6 0.1, 285.3 6 0.2,

and 287.1 6 0.4 eV, respectively, and assigned, as the curves

for sp2 hybridization, sp3 hybridization and bonding between

carbon and oxygen, respectively.

Figure 9 shows the variation of the sp3, sp2 contents, full

width at half maxima (FWHM) of C 1 s peak, and sp3/sp2

ratios as a function of self-bias. The sp3 content in DLC

films, first increases with self-bias up to �150 V and then

decreases beyond �150 V self-bias. In the DLC films, there

are two main source for the formation of carbon sp3 bonding;

ion energy and C-H bonding in the network.74 The ion

energy is a critical parameter in DLC films, because most of

the properties of the DLC films depend upon the ion energy

value, and only in a specific range of ion energy, the DLC

films of good quality can be deposited. Beyond, this critical

energy value the diamond like properties of the films trans-

form in graphite like properties. As discussed earlier, with

increase of self-bias, due to transformation of power dissipa-

tion mode from bulk to sheath, the ion energy increases

beyond the critical energy value. As a result, diamond like

sp3 hybridization change into graphite like sp2 hybridization.

There may be some possibility of sputtering/etching from the

film surface which deteriorates the film properties. Thus, the

diamonds like characteristics of DLC films are found to

depend strongly on ion energy during deposition process. As

we increase the self-bias from �100 to �150 V, the ion

energy increases but as the plasma impedance have the low-

est value and highest value of the bulk resistance. As a result,

most of the power is dissipated by the electrons in to the

bulk of the plasma. As a result of which plasma have highest

electron density value and moderate ion energy value in this

region of self-bias. It is well accepted that the high electron

density and lower plasma impedance with moderate ion

energy is necessary for the good quality films. Thus, this

region of self-bias (�100 to �150 V) have high sp3 content

and excellent film properties. The DLC films have highest

value of sp3 content �60.97% at �150 V and lowest value

of �45.78% at �100 V self-bias. As we further increase the

self-bias beyond �150 V, the ion energy increases beyond

critical energy value due to transformation of dissipation of

power from bulk to sheath.

Along with this, the electron density value decreases

and the plasma impedance value increases, which transform

the discharge from resistive to capacitive. Because of this,

most of the power is dissipated by the sheath for the acceler-

ation of the ions in the sheath region. When these accelerated

ions bombarded the films surface, they break the bonds from

the surface of the films and creates the dangling bonds on the

film surface, and when these dangling bonds recombines

with the neighboring dangling bonds, the C-C sp3 bonding

transform into C¼C sp2 bondings. This results in decrease of

sp3 content and increase of sp2 content. This transformation

of sp3 bondings into sp2 bondings governs the graphite-like

properties to the films. The results obtained in the present

investigation with the self-bias are well corroborated with

the reported literature.56 The value of the FWHM and the

sp3/sp2 ratios of the DLC films have the similar trend as that

of sp3 content with self-bias. The sp3 and sp2 contents of

DLC films are found to strongly depend on FWHM value. It

has been reported in literature75 that the sp3 content vary

with FWHM value. Higher FWHM value means higher sp3

FIG. 8. Deconvoluted XPS spectra of C1s peaks deposited at different self-

biases ranging from �100 to �300 V.

FIG. 9. Variation of sp3, sp2 contents, FWHM of C 1 s and sp3/sp2 with

self-bias.

033515-9 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

content. The sp2 content in DLC films have just reverse trend

than that of FWHM value and sp3 content with self-bias.

The sp3 content is the key parameter of the DLC films,

but the clustering of sp2 phase in DLC films is the another

crucial parameter because DLC films with same sp3 content

and hydrogen concentration can have different electrical, op-

tical, and mechanical properties depending upon the cluster-

ing of the sp2 phase. Like XPS, Raman spectroscopy is also

a non-destructive technique used for the evaluation of the

sp3 content and sp2 clustering in carbon films. However,

Raman spectroscopy is provides only the qualitative infor-

mation about the sp3 hybridization and carbon clustering in

sp2 phase. Figure 10(a) elucidates the Raman spectra taken

in the wave number ranges from 1000 to 2000 cm�1 of the

highly pure (99.999%) graphite (for sp2 hybridization) and

DLC films deposited by RF-PECVD system at various self-

biases. It has been observed from the figure that the Raman

spectrum of the pure (99.999%) graphite has two sharp bands

located at �1570 and 1350 cm�1, while the Raman spectra

of the deposited DLC films have a main major band around

�1570 cm�1 with pronounced shoulder around �1360 cm�1.

The band located in graphite at 1350 cm�1 was diminished

in DLC films and only a shoulder was observed in the DLC

films. These bands at 1570 and 1350 cm�1 were assigned as

G-band for zone center phonons of symmetric E2g C-C

stretching mode and D-band for K-point phonons of A1g

symmetry, due to bond angle disorder in graphite like micro-

domains affected by sp3 bonds, respectively. The diminish-

ing of D band in DLC films is due to enhancement of

ordering of carbon atom with sp3 hybridization.

Since, the sp3 bonded carbon coatings are optically

transparent for visible- Raman spectroscopy and sp2 bonds

are 50–230 times more sensitive to the visible photons

(2.2 eV) preferentially excite the p states.69,76 Thus, the sp2

bonds easily detectable by Raman spectroscopy and indi-

rectly sp3 bonding. To evaluate quantitative results, the

Raman spectra were deconvoluted into two Gaussian peaks

associated with their microstructures and the G band at

�1570 cm�1 due to disordered graphite and D band at

�1360 cm�1 and small domain size graphite regions were

separated. For the evaluation of the amount of sp3 and sp2

contents present in the films, the intensity ratio of the peaks

corresponds to D-band to G-band (ID/IG) has been taken and

found to be �0.525 at �150 V self-bias.

Figure 10(b) shows the variation of G-peak position,

G-FWHM, and ID/IG with self-bias from graphite to DLC

films. It has been observed that the peak position of the G-

band shifted towards lower frequency side with self-bias

from graphite (1570 cm�1) upto �150 V (1542 cm�1) and

shifted towards higher frequency side beyond �150 V self-

bias. It has been further indicated that the FWHM value of

G-band increase from graphite to DLC films. The FWHM

value of G-band increases from graphite (�1570.56 cm�1) to

�150 V (�1548.23 cm�1) self-bias and decreases beyond

�150 V self-bias. Champi and Marques77 and Jiang et al.78

suggested that shifting G-band peak position towards lower

wavenumber side and increase of FWHM value of G-band

favors the formation of the sp3 bonded carbon. Also, the in-

tensity ratios (ID/IG) are found to be decreases from graphite

to �150 V self-bias and increase beyond �150 V. It has

been demonstrated in the literature77–81 that decrease of ID/

IG ratio corresponds to increase of sp3 content. Thus, the film

deposited at �150 V has the highest value of sp3 content.

Gou et al.64 also suggested that higher the ID/IG ratio, higher

the size of the sp2 carbon cluster. Thus, the pure graphite has

a highest sp2 carbon clustering and the DLC film deposited

at �150 V self-bias has a lowest value of sp2 clustering.

Thus, the results obtained from shifting of G-band peak

position and FWHM value and ID/IG ratio suggested that the

DLC film deposited at �150 V self-bias has highest value of

sp3 content and lowest value of carbon clustering. The results

obtained from the XPS studies and Raman spectroscopic

investigations are well corroborated.

Figure 11 shows the variation of electrical conductivity

of the DLC films as a function temperature with self-bias.

The conductivity of the DLC films increases with tempera-

ture as well as self-bias. The increase of electrical conductiv-

ity of the DLC films with temperature indicates the

semiconducting behavior of the samples. The electrical con-

ductivity of the DLC film decreases with increase of self-

bias upto �150 V and increases beyond �150 V self-bias, as

the electrical properties of the DLC films govern by the sp2

hybridization. Thus, it decreases with sp2 hybridization uptoFIG. 10. (a) Deconvoluted Raman spectra of the DLC films and (b) varia-

tion of G-peak position, G-peak FWHM, and ID/IG with self-bias.

033515-10 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

�150 V self-bias and increases again with sp2 hybridization

beyond �150 V self-bias due to graphitization of the DLC

films at higher ion energy value (self-bias) which causes the

bond breaking or sputtering type of phenomena on the film

surface. There could also be decrease of hydrogen content in

these films due to increase of energetic ionic deposition with

the increase of self-bias beyond �150 V. These results are

well corroborated with the results suggested by Yoon et al.56

and discussed earlier that lesser energy is required to break

C-H bonds as compared to C-C bond. Thus, by increasing

the ion energy or the self-bias, the hydrogen is ejected from

the C-H bonds and the hydrogen content decreases with the

self-bias in the carbon films. As a result of which the films

transform from polymer like (having high hydrogen content

�70% and long chains of C-H bonds) to diamond like films

having lesser hydrogen content (<30%). At a higher ion

energy or self-bias, the properties of the carbon films trans-

form from diamond like to graphite like. Yoon et al.56

observed that with increase of self-bias from �25 to �150

V, the carbon film changes from polymer like to diamond

like and �150 to �250 V diamond like to graphite like.

High resolution nano-indentation technique was used to

analyze the nano-mechanical properties of DLC films at vari-

ous loads 5–15 mN. Figure 12(a) illustrates the load versus

displacement curves at indentation loads of 5-15 mN on

DLC films grown at different self-biases ranging from �100

to �300 V. It is evident from the figure that the DLC film de-

posited at �150 V exhibited minimum penetration depth.

However, the DLC films deposited at lower or higher self-

bias to �150 V have higher penetration depth and increases

continuously with self-bias beyond �150 V. It has also been

found that there is a very small difference in the penetration

depth for the films deposited at self-biases from �100 to

�200 V but the DLC film deposited at �300 have a signifi-

cant enhancement in penetration depth due to higher graphitic

character of the film as confirmed by XPS and Raman analy-

sis. Penetration depth in all the DLC films was found to

increase with increasing the load from 5 to 15 mN which sup-

ports the Hook’s force-displacement law. There is no pop in

(cracking of film) observed at this load in the deposited film.

To avoid the substrate effect in the measurement, care has

been taken that indenter not penetrate more than 10% to 25%

FIG. 11. Variation of electrical conductivity of DLC films with self-bias.

FIG. 12. Variation of the optical and mechanical parameters of the DLC

films deposited at various self-biases from �100 to �300 V; (a) Load versus

displacement curve, (b) Variation of optical band gap (Eg), hardness (H),

elastic modulus (E), plastic index parameter (H/E), and %ER with self-bias,

and (c) Variation of H (in inset H vs depth/film thickness) and E (in inset E

vs depth/film thickness) with penetration depth.

033515-11 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

of the total film thickness. If the indenter penetrates more than

10% to 25%, then substrate effect may occur and then the

hardness could be the composite hardness. The load versus

displacement curves were further employed to estimate the

nanohardness (H), elastic modulus (E), plastic index parame-

ter (H/E), and elastic recovery (%ER). Figure 12(b) shows the

variation of H, E, H/E, and %ER at 5-15 mN load and optical

band evaluated from UV-vis spectroscopy with self-bias. The

value of all these mechanical and optical parameters slightly

increased with increasing the self bias from �100 to �150 V

self-bias and decreases continuously beyond �150 V self-

bias. The values of H at 5 mN load for DLC films deposited

were found to be 25.5, 28.2, 23, and 15.1 GPa at �100, �150,

�200, and �300 V self-bias, respectively, while the value of

E at 5 mN load was found to be in the range of 200 (at �150

V) to 280 GPa (at �300 V). As the optical and mechanical

properties were governed by the sp3 hybridization which is

depends upon the ion energy or self-bias. Erdemir and Don-

net82 and Singh et al.83 have suggested that the moderate ion

energy or self-bias (�150 V) is required for complete decom-

position of hydrocarbon precursor for the synthesis of good

quality DLC films. They have suggested that lower self-bias

favors in the formation of polymer like, while higher self-bias

is favorable for enhanced sp2 disorder which results in graph-

ite like DLC films. Thus, with self-bias, transition has been

taken place from polymer-like to diamond-like to graphite-

like of the DLC films. Dwivedi et al.84–86 and Ito et al.87 were

deposited harder DLC films in the range of �100 to �150 V

self-bias.

The plastic index parameter (H/E) and the %ER factor

(shown in Figure 12(b)) are other important parameters

which state the mechanical properties of the DLC films. The

plastic index parameter differentiate between the elastic and

the elastic-plastic behavior of the DLC films. For protective

coating on magnetic hard disk or wear resistant coatings, the

value of H/E must be very high. It is illustrated from the fig-

ure that the value of H/E initially slightly increases with

increase of self-bias from �100 to �150 V at all loads

except 15 mN and significantly decreased beyond �150 V.

The decrease of H/E beyond �150 V revealed that most of

work (load) is dissipated in plastic deformation due to the

initiation of sp2 bonding and a large plastic strain was

expected. The DLC film deposited at �150 V has highest

value of H/E (0.1), while the sample deposited at �300 V

self-bias have a lowest value (0.075) measured at 5 mN. The

value of H/E also decreases with applied load (from 5 to

15 mN) due to influence of substrate effect.

The quality of the films were also evaluated with %ER

and shown in Figure 12(b) at different applied load

(5-15 mN). Higher the elastic recovery factor value higher be

the hardness value of the films and more useful the coatings

for the tribological applications. The elastic recovery of de-

posited films can be evaluated by the expression

%ER¼ dmax�dres

dmax� 100, where dmax and dres are the displace-

ment at the maximum load and residual displacement after

load removal, respectively. %ER has a maximum value

(83.1 at 5 mN) at �150 V self-bias due to high sp3 content

(diamond-like bondings) and lower values both side of it due

to high sp2 content (graphite-like bondings). Thus, all the

mechanical parameters H, E, H/E, and %ER have similar

trend as that of sp3 content with self-bias and all the parame-

ters are increases with sp3 content. Also the mechanical pa-

rameters have similar behavior with as that of electron

density with self-bias and a reversal trend with the plasma

impedance.

It has also been observed from the Figure 12(b) that the

H and E of the DLC films not only depend upon the self-bias

but also on the indentation load and decrease with indenta-

tion load due to substrate effect. To evaluate the effect of

penetration depth on the value of H and E, variation of H

and E was discussed in Figure 12(c). The variation of H and

E with depth/film thickness is given in insets of Figure 12(c),

in respective manner. Depending upon films and the sub-

strates natures, two models, soft film on hard substrate, and

hard film on soft substrate were proposed to analyze the

hardness of films with depth. Saha and Nix88 performed se-

ries of experiments in these types of configurations (hard

films on soft substrates and soft films on hard substrates).

They observed for hard film on soft substrate, the hardness

of composite substrate/film decrease with increasing penetra-

tion depth. On the other hand for soft film on hard substrate,

the hardness of composite substrate/film increases with

increasing the penetration depth. In the present work, the

DLC films were deposited on Si wafer and used for nanoin-

dentation measurement and found that the H value of the

samples decreases with penetration depth, which is indica-

tive of hard film on soft substrate. At lower penetration

depth, the value of H for the deposited DLC films was signif-

icantly higher with negligible substrate effect and the values

of H depends only on self-bias. The value of H decreases

with penetration depth and becomes almost linear, indicating

the substrate effect at higher load. Also the value of E

degraded with depth. Thus, to evaluate the exact mechanical

parameters of the prepared samples, lower possible load

must be applied to avoid the substrate effect. The optical

(band gap) and mechanical properties are found to be

strongly depend on self-bias, sp3 content, electron density,

and plasma impedance. All these parameters are increases

with self-bias, sp3 content, and electron density, and

decreases with plasma impedance. The behavior of electri-

cal, mechanical, and optical properties of DLC films

observed by us are well corroborated plasma parameters and

with the reported literature.89–91

IV. CONCLUSIONS

The effect of self-bias on the plasma parameters of ar-

gon, acetylene, and their mixture discharge has been investi-

gated. All the plasma parameters evaluated from the

experimental studies for argon, acetylene, and their mixture

are well corroborated with the theoretical and experimental

studies of plasma parameter carried out by various research

groups.

This investigation on the variation of plasma parameters

with self-bias provides the information about the conditions

under which DLC films of desired properties can be depos-

ited precisely. Thus, the knowledge of plasma parameters

plays a vital role in the selection of process parameters for

033515-12 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

the deposition of the desired properties of DLC films. The

plasma parameters evaluated in the present study and the

properties of the DLC films observed suggest that the range

of self-bias �100 to �200 V is critical for the deposition of

DLC films. The plasma parameters evaluated in the present

study are well matched with the theoretically suggested pa-

rameters. For the realization of the plasma parameter and the

validation of the VI probe measurements, the DLC films

were deposited under similar deposition conditions and

found that the properties of the DLC films were well corro-

borated with the plasma parameters. For the lower value of

plasma impedance and higher electron density value, the de-

posited films have better mechanical (nanohardness, elastic

modulus, plastic index parameter, and elastic recovery)

properties.

ACKNOWLEDGMENTS

The authors are grateful to the Director, National Physi-

cal Laboratory, New Delhi, (India) for his kind support to

publish this paper. The authors wish to thanks Dr. Govind

for providing XPS data. We acknowledge CSIR, Govt. of

India for sponsoring network project NWP-0027 and for

their financial supports.

1A. Grill, Cold Plasma Material Fabrication: From Fundamentals toApplications (IEEE, NJ, USA, 1993).

2H. Aguas, R. Martins, and E. Fortunato, Vacuum 56, 31 (2000).3K. Kohler, J. W. Coburn, D. E. Horne, E. Kay, and J. H. Keller, J. Appl.

Phys. 57, 59 (1985).4N. Spiliopoulos, D. Mataras, and D. E. Rapakoulias, J. Vac. Sci. Technol.

A 14(5), 2757 (1996).5S. Dine, J. Jolly, and J. Guillon, Ecole Polytechnique, France 188–91.6H. Curtins, N. Wyrsch, M. Favre, and A. V. Shah, Plasma Chem. Plasma

Process. 7, 267 (1987).7M. Surendra and D. B. Graves, Appl. Phys. Lett. 59, 2091 (1991).8M. Heintze, R. Zedlitz, and G. H. Bauer, J. Phys. D 26, 1781 (1993).9T. Kitajima, Y. Takeo, N. Nakano, and T. Makabe, J. Appl. Phys. 84, 5928

(1998).10E. Amanatides and D. Mataras, J. Appl. Phys. 89, 1556 (2001).11M. Fukawa, S. Suzuki, L. Guo, M. Kondo, and A. Matsuda, Sol. Energy

Mater. Sol. Cells 66, 217 (2001).12H. H. Goto, H. D. Lowe, and T. Ohmi, J. Vac. Sci. Technol. A 10, 3048

(1992).13H. C. Kim and V. I. Manousiouthakis, J. Vac. Sci. Technol. A 16, 2162

(1998).14T. Kitajima, Y. Takeo, Z. L. Petrovic, and T. Makabe, Appl. Phys. Lett.

77, 489 (2000).15C. Beneking, J. Appl. Phys. 68, 1461 (1990).16V. A. Godyak, R. B. Piejak, and B. M. Alexandrovich, IEEE Trans.

Plasma Sci. 19, 660 (1991).17V. A. Godyak, R. B. Piejak, and B. M. Alexandrovich, J. Appl. Phys. 69,

3455 (1991).18S. J. You, C. W. Chung, K. H. Bai, and H. Y. Chang, Appl. Phys. Lett. 81,

14 (2002).19S. J. You, H. C. Kim, C. W. Chung, H. Y. Chang, and J. K. Lee, J. Appl.

Phys. 94, 7422 (2003).20S. J. You, S. K. Ahn, and H. Y. Chang, Surf. Coat. Technol. 193, 81

(2005).21F. Schneider, Z. Angew. Phys. 6, 839 (1949).22V. A. Godyak and N. Sternberg, Phys. Rev. A 42, 2299 (1990).23Y. S. Lee, J. H. In, S. K. Ahn, S. H. Seo, H. Y. Chang, D. J. You, S. W.

Ahn, and H. M. Lee, Curr. Appl. Phys. 10, S234 (2010).24Y. S. Lee, H. S. Lee, and H. Y. Chang, Thin Solid Films 518, 6882 (2010).25R. A. Gattscho and T. A. Miller, Pure Appl. Chem. 56(2), 189 (1984).26A. J. Miranda and C. J. Spanos, J. Vac. Sci. Technol. A 14, 1888 (1996).27G. Viera, J. Costa, F. J. Compte, E. Garcia-Sanz, J. L. Andujar, and

E. Bertran, Vacuum 53, 1 (1999).

28E. Amanatides and D. Mataras, Diamond Relat. Mater. 14, 292 (2005).29E. Amanatides, A. Hammed, E. Katsia, and D. Mataras, J. Appl. Phys. 97,

073303–1 (2005).30A. Parashar, S. Kumar, P. N. Dixit, J. Gope, C. M. S. Rauthan, and S. A.

Hashmi, Sol. Energy Mater. Sol. Cells 92, 1194 (2008).31M. A. Sobolewski, J. Vac. Sci. Technol. A 10, 3550 (1992).32M. A. Sobolewski, J. Appl. Phys. 90, 2660 (2001).33M. A. Sobolewski, Phys. Rev. E 59, 1059 (1999).34M. A. Sobolewski, J. Appl. Phys. 100, 063310 (2006).35P. J. Hargis, Jr., K. E. Greenberg, P. A. Millar, J. B. Gerardo, J. R.

Torczynski, M. E. Riley, G. A. Hebner, J. R. Roberts, J. K. Olthoff, J. R.

Whetstone, R. J. Van Brunt, M. A. Sobolewski, H. M. Anderson, M. P.

Splichal, J. L. Mock, P. Bletzinger, A. Garscadden, R. A. Gottscho, G.

Selwyn, M. Dalvie, J. E. Heidenreich, Jeffery W. Butterbaugh, M. L.

Brake, M. L. Passow, J. Pender, A. Lujan, M. E. Elta, D. B. Graves, H. H.

Sawin, M. J. Kushner, J. T. Verdeyen, R. Horwath, and T. R. Turner, Rev.

Sci. Instrum. 65(1), 140 (1994).36A. Parashar, S. Kumar, J. Gope, C. M. S. Rauthan, S. A. Hashmi, and P.

N. Dixit, J. Non-Cryst. Solids 356, 1774 (2010).37T. Catherine, in Diamond and Diamond like Film and Coating NATO-ASI

(Plenum, New York, 1991), Vol. 266, p. 193.38J. Robertson, Mater. Sci. Eng. R 37, 129 (2002).39M. A. Liebermana and A. J. Lichtenberg, Principle of Plasma Discharges

and Materials Processing (Wiley, New York, 1994).40P. Koidl, C. Wagner, B. Dischler, J. Wagner, and M. Ramsteiner, Mater.

Sci. Forum 52, 41 (1990).41A. Bubenger, B. Dischler, D. Brandt, and P. Koidl, J. Appl. Phys. 54, 4590

(1983).42F. Schneider, Z. Angew. Phys. 6, 456 (1954).43H. R. Koenig and L. I. Maissel, IBM J. Res. Dev. 14, 168 (1970).44J. H. Keller and W. B. Pennebaker, IBM J. Res. Develop. 23, 3

(1979).45M. A. Sobolewski, IEEE Trans. Plasma Sci. 23, 1006 (1995).46E. Amanatides, D. Mataras, and D. E. Rapakoulias, J. Appl. Phys. 90(11),

5799 (2001).47N. Spiliopoulos, D. Mataras, and D. E. Rapakoulias, J. Electrochem. Soc.

144, 634 (1997).48E. Amanatides, D. Mataras, and D. E. Rapakoulias, J. Vac. Sci. Technol.

A 20, 68 (2002).49D. A. Ariskin, I. V. Schweigert, A. L. Alexandrov, A. Bogaerts, and F. M.

Peeters, J. Appl. Phys. 105, 063305–1 (2009).50L. Marques, J. Jolly, and L. L. Alves, J. Appl. Phys. 102, 063305–1

(2007).51Reaction Under Plasma Condition, edited by M. Venugopal (Wiley, New

York, 1971).52Z. Sun, C. H. Lin, Y. L. Lee, J. R. Shi, B. K. Tay, and X. Shi, Thin Solid

Films 377–378, 198 (2000).53J. Filik, P. W. May, S. R. J. Pearce, R. K. Wild, and K. R. Hallam,

Diamond Relat. Mater. 12, 974 (2003).54J. J. Cuomo, J. P. Doyle, J. Bruley, and J. C. Liu, Appl. Phys. Lett. 58, 466

(1991).55J. J. Cuomo, D. L. Pappas, J. Bruley, J. P. Doyle, and K. L. Saenger,

J. Appl. Phys. 70, 1706 (1991).56S. F. Yoon, K. H. Tan, R. J. Ahn, and Q. F. Huang, J. Appl. Phys. 89, 4830

(2001).57R. M. Dey, S. B. Singh, A. Biswas, R. B. Tokas, N. Chand, S. Venkatesh-

waran, D. Bhattacharya, N. K. Sahoo, S. W. Gosavi, S. K. Kulkarni, and

D. S. Patil, Curr. Appl. Phys. 8, 6 (2008).58R. Sharma, O. S. Panwar, S. Kumar, D. Sarangi, A. Goullet, P. N. Dixit,

and R. Bhattacharyya, Appl. Surf. Sci. 220, 313 (2003).59K. H. Lai, C. Y. Chan, M. K. Fung, I. Bello, C. S. Lee, and S. T. Lee,

Diamond Relat. Mater. 10, 1862 (2001).60R. C. Weast, Handbook of Chemistry and Physics (CRC, Boca Raton, FL,

1983), p. F-180.61Z. Lin, S. B. Lv, Z. J. Yu, M. Li, T.Y. Lin, D. C. Ba, C. K. Choi, and I. S.

Lee, Surf. Coat. Technol. 202, 5386 (2008).62S. B. Singh, A. K. Poswal, M. Pandey, R. B. Tokas, N. Chand, S. Dhara,

B. Sundaravel, K. G. M. Nair, N. K. Sahoo, and D. S. Patil, Surf. Coat.

Technol. 203, 986 (2009).63D. H. Lee, X. M. He, K. C. Walter, M. Nastasi, J. R. Tesmer, M. Tuszewski,

and D. R. Tallant, Appl. Phys. Lett. 73, 2423 (1998).64W. Gou, G. Li, X. Chu, and B. Zhong, Surf. Coat. Technol. 201, 5043

(2007).65G. Capote and F. L. Freire, Jr, Mater. Sci. Eng. B 112, 101 (2004).

033515-13 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18

66G. E. Mullenberg, Handbook of X-ray Photoelectron Spectroscopy (Perkin

Elmer Corporation, MN USA, 1979).67E. Ech-chamikh, A. Essafti, Y. Ijdhiyaou, and M. Azizan, Sol. Energy

Mater. Sol. Cell 90, 1420 (2006).68T. Ono, Y. Suda, M. Akazawa, Y. Sakai, and K. Suzuki, Jpn. J. Appl.

Phys. 40, 4651 (2002).69K. Y. Amamoto, F. Kokai, Y. Koga, and S. Fujiwara, New Diamond Front.

Carbon Technol. 9, 289 (1999).70P. Merel, M. Tabbal, M. Chaker, S. Moisa, and J. Margot, Appl. Surf. Sci.

136, 105 (1998).71N. Paik, Surf. Coat. Technol. 200, 2170 (2005).72C. K. Park, S. M. Chang, H. S. Uhm, S. H. Seo, and J. S. Park, Thin Solid

Films 420–421, 235 (2002).73Y. Yamauchi, Y. Sasai, S. I. Kondo, and M. Kuzuya, Thin Solid Films

518, 3492 (2010).74J. Robertson, in Proceeding of 1st International Specialist Meeting on

Amorphous Carbon (SMAC-97), edited by S. R. P. Silva et al. (World Sci-

entific, Singapore, 1998), p. 32.75O. S. Panwar, Y. Aparna, S. M. Shivaprashad, M. A. Khan, B. S. Satyanar-

ayana, and R. Bhattacharyya, Appl. Surf. Sci. 221, 392 (2004).76O. S. Panwar, M. A. Khan, M. Kumar, S. M. Shivaprasad, B. S. Satyanar-

ayana, P. N. Dixit, R. Bhattacharya, and M. Y. Khan, Thin Solid Films

516, 2331 (2008).77A. Champi and F. C. Marques, Thin Solid Films 501, 362 (2006).

78H. F. Jiang, X. B. Tian, S. Q. Yang, R. K. Y. Fu, and P. K. Chu, J. Vac.

Sci. Technol. A 26, 1149 (2008).79J. X. Gou, Z. Sun, B. K. Tay, and X. W. Sun, Appl. Surf. Sci. 214, 351

(2003).80L. Y. Huang, K. W. Xu, J. Lu, B. Guelorget, and H. Chen, Diamond Relat.

Mater. 10, 1448 (2001).81C. A. Taylor, M. F. Wayne, and W. K. S. Chiu, Thin Solid Films 429, 190

(2003).82A. Erdemir and C. Donnet, J. Phys. D Appl. Phys. 39, R311 (2006).83S. Singh, M. Pandey, R. Kishore, N. Chand, S. Dash, A. Tyagi, and D.

Patil, Plasma Processes Polym. 5, 853 (2008).84N. Dwivedi, S. Kumar, C. M. S. Rauthan, and O. S. Panwar, Appl. Phys. A

102, 225 (2010).85N. Dwivedi, S. Kumar, C. M. S. Rauthan, and O. S. Panwar, Plasma Proc-

esses Polym. 8, 100 (2011).86N. Dwivedi, S. Kumar, and H. K. Malik, Appl. Surf. Sci. 257, 9953

(2011).87H. Ito, K. Kanda, and H. Saitoh, Diamond Relat. Mater. 17, 688 (2008).88R. Saha and W.D. Nix, Acta Mater. 50, 23 (2002).89K. W. Whang and H. S. Tae, Thin Solid Films 204, 49 (1991).90P. Gupta, “Synthesis, structure and properties of nanolayered DLC/DLC

films,” M.S. thesis 1-100, Mechanical Engineering Department, Louisiana

State University, 2003.91H. J. Ryu, S. H. Kim, and S. H. Hong, Mater. Sci. Eng. A 277, 57 (2000).

033515-14 Ishpal et al. Phys. Plasmas 19, 033515 (2012)

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:

14.139.45.241 On: Sat, 25 Jul 2015 06:35:18