coatings

Article
Effect of Magnetic Field Arrangement of Facing
Targets Sputtering (FTS) System on Controlling
Plasma Confinement
Sangmo Kim and Kyung Hwan Kim *
Department of Electrical engineering, Gachon University, Seongnam-si, Gyeonggi-do 13120, Korea;
singmul0227@gachon.ac.kr
* Correspondence: khkim@gachon.ac.kr




Received: 17 February 2020; Accepted: 25 March 2020; Published: 28 March 2020


Abstract: Conventional sputtering method uses a single cathode with a permanent magnet.
Facing targets sputtering (FTS) methods consists of two cathodes. Because of a unique structure,
FTS can prepare high quality films with low temperature and low plasma damage. During the film
sputtering process, density and confinement of discharged plasma depend on the arrangement of
a permanent magnet in the cathode. In this study, we designed two types of permanent magnet
arrangements in the FTS system and the designed permanent magnet was inserted into two cathodes
in the FTS system. The system was operated in different permanent magnet conditions, and their
discharge voltage and properties of as-grown films were recorded. In the designed FTS, compared to
a conventional magnetron sputtering method, the substrate temperature increased to a value under
80 ◦ C, which is relatively low, even though the films’ sputtering process was completed.

Keywords: facing targets sputtering; magnetic field; plasma

1. Introduction
Films preparation methods are classified as follows: chemical vapor deposition (CVD) and physical
vapor deposition (PVD). Among these methods, sputtering methods with magnetron are widely used
in various electrical applications such as display, solar cell, and sensors because it could prepare films
across a wide area for oxide and non-oxide materials [1–3]. It is crucial to identify methods to control
the sputtering plasma discharged on the target during the film sputtering process. The plasma could be
confined by a magnetic field inserted inside the cathode of the sputtering source. A normal magnetic
field is produced by the Nd-Fe-B magnet. Researchers have attempted to improve performance of
sputtering by designing a new type of permeant magnet inside the cathode. A magnetic field of a
magnetron allows the operation of an intense sputtering discharge at low neutral gas densities [4,5].
The magnetic field strength is a critical parameter in a magnetron design but is often chosen in
practice using empirical methods. Therefore, a model that not only provides insight into magnetron
operation but also practical criteria for designing a magnetron is required. Originally, facing targets
sputtering (FTS) method was proposed for growing magnetic materials such as Co-Cr, Co-Cr-Ta,
and Co-ferrite for perpendicular recording media [6–8].
The FTS system could grow high-density-low-defect films. Compared to conventional magnetron
sputtering (CMS) methods, a sputtering target and substrate surface in the FTS system generate a
sputtering plasma in the space between two sputtering targets. Neutral working gas and high-energy
radical ion particles that originated in the plasma were confined between two targets and move
along with the magnetic field [9–11]. Therefore, the substrate surface can be directly bombarded by
secondary electrons, Ar atoms, and negative ions that obtain high kinetic energy at a cathode sheath

Coatings 2020, 10, 321; doi:10.3390/coatings10040321 www.mdpi.com/journal/coatings

Coatings 2020, 10, 321 2 of 13

generated above the target surface, especially during reactive sputtering with oxygen or nitrogen
gas. As a result, the film was grown with strong internal stress, defects in the grain boundary, and a
low-density columnar structure. Furthermore, the radiation of secondary electrons ejected from
the target surface raise the temperature of the substrate surface. Compared with other magnetron
sputtering techniques, FTS also has many advantages, such as high sputtering efficiency, high target
utilization rate, and deposition of films with low defect states and a smooth surface [12,13].
Sputtering techniques depend on plasma generated around the targets after the collision of
working gas (argon and oxygen) and high energy particles (electron) inside a high vacuum chamber.
The plasma could be controlled by the confirmed magnetic field contributed by a permeant magnet
inside the cathode of the sputtering equipment. Until now, magnetic field dependence on conventional
sputtering systems have been reported in several research studies. Murphy et al. [14] reported the
geometry of a magnetic field in a magnetron sputtering system. Hollerweger et al. [15] reported
the effect on magnetic field strength of films in magnetron sputtering. Meanwhile, Noda et al. [16]
introduced the magnetic field and materials in FTS considered as the most developed off-axis geometry
sputtering by Hoshi et al. [17]. They focused on the unique structure (two targets facing each other)
and dealt with the plasma damage on the film sputtering.
In this study, we designed a new-type magnet arrangement inside the cathode in the FTS system
and inserted the designed magnet into the cathode. We investigated the effect on the magnetic field
generated by a permanent magnet. Moreover, the films were prepared on the glass substrate and their
properties were investigated. We predicted the target erosion in the sputtering and improved it.

2. Experimental Methods and Measurements

All sputtering processes were carried out under the following conditions. Before sputtering,
the base pressure was maintained at the high vacuum of 5.0 × 10−6 Torr using a turbo-molecular pump
and a rotary pump. A DC power supply (MDX Magnetron delta 5kW, Advanced Energy Industries,
Fort Collins, CO, USA) is employed as the sputtering power source.
More detailed films’ sputtering conditions are given in Table 1. To compare various permanent
magnet arrangements inside the sputtering source, Indium Tin Oxide (ITO) targets were used for
the discharge test and films were sputtered on the glass substrate. Soda-lime glass substrates were
prepared after a standard cleaning process with Acetone, Ethanol, and D.I. water at the ultrasonication
for 10 min. We designed two kinds of Open-type FTS (OFTS) and closed-type FTS (CFTS).

Table 1. Comparison of detailed sputtering conditions by the magnet arrangement.

Conditions
Parameters
Open-Type FTS (OFTS) Closed-Type FTS (CFTS)
Target material ITO (In2 O3 : 90%, SnO2 : 10%)
Target size 250 × 100 mm 250 × 50 mm
Base pressure 5 × 10−6 Torr
Working pressure 0.5–5 mTorr
Working gas Ar: 9–10 sccm, O2 : 0–1sccm
DC power (5kW) 500 W
Distance from the target to the substrate
100 mm
(DT-S )
Distance from the target to the target
100 mm
(DT-T)
Film thickness 100 nm

Two types of permanent magnetic arrangements were inserted into two cathodes, as shown
Figure 1. Among common structural characteristics, they have two cathodes with opposite magnetic
poles and plasma is generated between two targets. Then, a substrate is placed at the center of
two targets. The working gas is mixed with argon and oxygen. Although the current FTS has
Coatings 2020, 10, x FOR PEER REVIEW 3 of 13

Two types of permanent magnetic arrangements were inserted into two cathodes, as shown
Figure
Coatings1.2020,
Among
10, 321common
structural characteristics, they have two cathodes with opposite magnetic
3 of 13
poles and plasma is generated between two targets. Then, a substrate is placed at the center of two
targets. The working gas is mixed with argon and oxygen. Although the current FTS has better
better characteristics than those of DC magnetron sputtering, we still need better-deposited thin
characteristics than those of DC magnetron sputtering, we still need better-deposited thin films,
films, which can be used for higher-density magnetic recording media and other applications [17–19].
which can be used for higher-density magnetic recording media and other applications [17–19]. The
The main disadvantages of current FTS techniques is that magnetic fields are not applied perfectly
main disadvantages of current FTS techniques is that magnetic fields are not applied perfectly
perpendicularly to both of the targets, as shown in Figure 1. Therefore, the substrate is exposed to
perpendicularly to both of the targets, as shown in Figure 1. Therefore, the substrate is exposed to
leaked magnetic fields, which may lead to the films on the substrate being damaged by plasma during
leaked magnetic fields, which may lead to the films on the substrate being damaged by plasma during
deposition. To solve this problem, it is necessary to generate uniform magnetic fields by using a unique
deposition. To solve this problem, it is necessary to generate uniform magnetic fields by using a
arrangement of magnets [20].
unique arrangement of magnets [20].

Figure 1. Schematic diagram of facing target sputtering (FTS) system with various magnet arrays.
Figure 1. Schematic diagram of facing target sputtering (FTS) system with various magnet arrays. (a)
(a) and (b) Open-type FTS, (c) and (d) Closed-type FTS.
and (b) Open-type FTS, (c) and (d) Closed-type FTS.
The surface morphology was measured via scanning probes microscopy (SPM, Park Systems
XE-150, surface
The Suwon,morphology
Korea) at was measured
Smart Materialsvia Research
scanning probes
Center microscopy
for IoT in(SPM,GachonParkUniversity.
Systems
XE-150, Suwon, Korea) at Smart Materials Research Center for IoT in Gachon University.
The crystallographic structural properties of the samples were measured via X-ray diffraction (XRD, The
crystallographic structural properties of the samples were measured via X-ray diffraction
RINT2100, Rigaku Corporation, Tokyo, Japan) using Cu-Kα radiation (λ = 0.154056 nm, 40 kV, 40 mA) (XRD,
RINT2100, Rigaku
in the 2-theta scanCorporation, Tokyo,
mode. Electrical Japan) using
properties of the Cu-Kα radiation
films were examined(λ = using
0.154056 nm, 40 kV,
a four-point 40
probe
mA) in the 2-theta scan mode. Electrical properties of the films were examined
(CMT-Series, Chang Min Tech. Corporation, Chatsworth, CA, USA). Magnetic flux density was using a four-point
probe (CMT-Series,
measured Chang
using a gauss Min (MG-3002,
meter Tech. Corporation, Chatsworth,
Lutron Co.) and opticalCA, USA). Magnetic
transmittance flux density
was measured wasa
using
measured using a gauss meter (MG-3002, Lutron Co.) and
UV-Vis spectrometer (LAMBDA750, PerkinElmer, Shelton, CT, USA). optical transmittance was measured using
a UV-Vis spectrometer (LAMBDA750, PerkinElmer, Shelton, CT, USA).
3. Results and Discussion
3. Results and Discussion
3.1. Magnetism Simulation
3.1. Magnetism Simulation
Before the preparation of magnetic arrangements, we performed magnetism simulation using the
BeforeMultiphysics
COMSOL the preparation of magnetic
software to obtainarrangements, we performed
a prediction plasma similar tomagnetism simulation
that generated using
in FTS systems.
the COMSOL Multiphysics
In the initial discharge software to obtain aelectrons
for film sputtering, prediction
wereplasma similar
generated to the
from thatsurface
generated intarget
of the FTS
systems.
(or cathode) using a DC power supply and accelerated toward the anode (substrate or another target).
Along the way to the anode, the electrons collided with other ions, which, thereby, increased the neutral
background gas and the number of electrons. Simultaneously, the charged particles move in the radial
direction of the magnetic field due to a Lorenzo force. Furthermore, the electric field intensity (E)
 Coatings 2020, 10, x FOR PEER REVIEW 4 of 13

In the initial discharge for film sputtering, electrons were generated from the surface of the target
(or cathode) using a DC power supply and accelerated toward the anode (substrate or another target).
Along
Coatings the10,way
2020, 321 to the anode, the electrons collided with other ions, which, thereby, increased the
4 of 13
neutral background gas and the number of electrons. Simultaneously, the charged particles move in
the radial direction of the magnetic field due to a Lorenzo force. Furthermore, the electric field
induces
intensitythe(E)
electrons
inducesto the
accelerate toward
electrons the anodetoward
to accelerate with substantially high substantially
the anode with energy. Accelerated high
high energy.
energy atoms transfer their energy to bonded electrons within ions and neutral atoms.
Accelerated high energy atoms transfer their energy to bonded electrons within ions and neutral Thus, more
atoms
atoms. areThus,
ionized,
moreand cations
atoms areand anions
ionized, andarecations
produced. Moreover,
and anions the atoms and
are produced. ions arethe
Moreover, confined in
atoms and
discharged plasma generated above the target.
ions are confined in discharged plasma generated above the target.
3.2. Film Uniformity and Target Erosion
3.2. Film Uniformity and Target Erosion
The permanent magnet is placed behind the target/Cu back plate and generates a magnetic field
The permanent magnet is placed behind the target/Cu back plate and generates a magnetic field
through the plate. Figure 2 shows the magnet simulation results, and the lines represent the 2D
through the plate. Figure 2 shows the magnet simulation results, and the lines represent the 2D
orientation of the magnetic field that was generated from the permanent magnet. The cross-section of
orientation of the magnetic field that was generated from the permanent magnet. The cross-section
the magnet is separately shown in Figure 2a,b. The parts attached to both ends of the magnet is the
of the magnet is separately shown in Figure 2a,b. The parts attached to both ends of the magnet is the
york (previously referred to as the plate) that is held in position and prevents the magnetic field from
york (previously referred to as the plate) that is held in position and prevents the magnetic field from
permeating into its backside.
permeating into its backside.

Figure 2. 2-D magnetism simulation results of the magnetic field induced as a function of the
Figure 2. 2-D
arrangement magnetismmagnets
of permanent simulation results
obtained of the
using magnetic
COMSOL field induced
Multiphysics as a (a)
software. function of the
Open-type
and (b) Closed-type.
arrangement of permanent magnets obtained using COMSOL Multiphysics software. (a) Open-type
and (b) Closed-type.
In Figure 2a, an open-type exhibits three magnetic field circular loops and the generated plasma
was induced
In Figureto be2a,divided into three
an open-type parts.three
exhibits However,
magnetic a closed-type
field circularhasloops
only and
one the
magnetic fieldplasma
generated loop,
and
was it the plasma
induced candivided
to be be confined in theparts.
into three spaceHowever,
between the targets, as shown
a closed-type has only in one
Figure 1c,d along
magnetic fieldwith
loop,
the magnetic field. Thus, the magnetism simulation results show that the
and it the plasma can be confined in the space between the targets, as shown in Figure 1c,d along plasma can be confined in
the space
with thebetween
magneticthe targets
field. Thus,using the magnetic
the magnetism field generated
simulation resultsbyshowthe that
permanent magnet
the plasma can be(Ne-Fe-B)
confined
setinbehind
the spaceeachbetween
target. the targets using the magnetic field generated by the permanent magnet (Ne-
Fe-B)In Figure
set behind3, theeach
distribution
target. of magnetic flux density as a function of the arrangement of permanent
magnets In in the cathode
Figure 3, the isdistribution
illustrated. The magneticflux
of magnetic flux density wasas ameasured
function at offive
thepoints betweenof
arrangement
the two targets.
permanent The film
magnets inwas
the deposited
cathode isonillustrated.
the glass substrate (200 mm
The magnetic flux× 200
densitymm)wasat a measured
target thickness
at five
ofpoints
100 nm. The Nd-Fe-B magnet employed in this study has a maximum
between the two targets. The film was deposited on the glass substrate (200 mm × 200 mm) surface flux density ofat
approximately
a target thickness 1321G. However,
of 100 nm. The theNd-Fe-B
sputtering source
magnet consists of
employed in athis
target,
study shield
has aring, and housing
maximum surface
(stainless materials). Therefore, the magnet flux density decreased after all parts
flux density of approximately 1321G. However, the sputtering source consists of a target, shield ring, were fully assembled.
Compared
and housing to CFTS, OFTSmaterials).
(stainless exhibits a higher magnet
Therefore, the flux
magnetdensity,
fluxas showndecreased
density in Figure 3a. However,
after all partsfilm
were
uniformity is consistent
fully assembled. across to
Compared allCFTS,
systems (Figure
OFTS 3b). The
exhibits film growth
a higher magnetoffluxthedensity,
sputtered astarget
shownisin related
Figure
to3a.
theHowever,
outer region filmofuniformity
the magnetic field. Even
is consistent uponall
across a 50%
systemsreduction
(Figureof3b). the The
size film
of the target of
growth andthe
magnet, the same films’ uniformity can be obtained on the substrate.
Coatings 2020, 10, x FOR PEER REVIEW 5 of 13

Coatings 2020, 10, 321 5 of 13

sputtered target is related to the outer region of the magnetic field. Even upon a 50% reduction of the
size of the target and magnet, the same films’ uniformity can be obtained on the substrate.

(a) (b)

(c)
Figure 3. Distribution of magnetic
Distribution of magneticflux
fluxdensity
densitybetween
betweentwo
twotargets
targetsasas a function
a function of of
thethe arrangement
arrangement of
of permanent
permanent magnets
magnets in cathode
in cathode (Y: perpendicular
(Y: perpendicular direction).
direction). (a) Open-type
(a) Open-type Facing
Facing Targets Targets
Sputtering
Sputtering
(FTS), (FTS), (b) Closed-type
(b) Closed-type FTS, and (c)FTS, and (c)
magnetic magnetic
flux density flux density
measured onmeasured onofthe
the surface thesurface of the
substrate.
substrate.
Furthermore, during sputtering, a target erosion area is formed along with a magnetic field.
TargetFurthermore,
materials sputtered
during from the target
sputtering, erosionerosion
a target area arearea
displaced
is formedtoward thewith
along substrate and grow
a magnetic on
field.
the substrate
Target materialswithsputtered
a high energy.
from Therefore, a large target
the target erosion area areerosion can achieved
displaced toward the by optimally
substrate designing
and grow
the magnet
on the arrayswith
substrate and controlling
a high energy. the incidence
Therefore,ofaplasma on theerosion
large target target. can achieved by optimally
We observed
designing the magnetthe arrays
conditionandofcontrolling
the ITO film the subsequent
incidence ofto sputtering,
plasma on theand the photograph was
target.
obtained, as shownthe
We observed in Figure 3. Moreover,
condition of the ITOwe measured
film subsequentthe magnetic field strength
to sputtering, and the on the surface
photograph of
was
the substrate
obtained, between
as shown in the two3.targets.
Figure Afterwe
Moreover, themeasured
substratethe wasmagnetic
mountedfieldon the substrate
strength holder
on the in the
surface of
chamber, as shown
the substrate betweenin Figure
the two1,targets.
we measured
After thethesubstrate
magneticwas fieldmounted
density on onthe
thesurface
substrateof the substrate
holder in the
from the outside
chamber, as shown using a Gauss
in Figure meter.
1, we The magnetic
measured field density
the magnetic was approximately
field density on the surface of zero
theregardless
substrate
of thethe
from type of sputtering,
outside as shown
using a Gauss in Figure
meter. 3c. This field
The magnetic means that awas
density magnetic field loopzero
approximately was regardless
generated
around the of
of the type cathode and the
sputtering, anode where
as shown the 3c.
in Figure plasma was confined
This means between field
that a magnetic the two looptargets.
was generated
In the
around the OFTS
cathodesystem,
and thea two-part
anode where erosion was formed
the plasma on the ITO
was confined target.the
between However, in the CFTS
two targets.
system,
In the OFTS system, a two-part erosion was formed on the ITO target. However, insimulation
a one-part erosion was formed. This could also be anticipated from the magnetism the CFTS
results
system,(Figure 2). The
a one-part OFTSwas
erosion system exhibits
formed. Thismagnetic
could alsoflux circuits induced
be anticipated fromby theeach permanent
magnetism magnet.
simulation
In the regions
results (Figure where the loops
2). The OFTSoverlapped, particles
system exhibits with high
magnetic energy
flux exhibited
circuits induced increased
by each bombardment
permanent
with the targets
magnet. compared
In the regions to the
where thebombardment
loops overlapped, in the particles
other regions. Furthermore,
with high we confirmed
energy exhibited that
increased
such sputteringwith
bombardment couldthebetargets
employed to prepare
compared to thefilms regardless of
bombardment inthe
thetarget
other size, as shown
regions. in Figurewe
Furthermore, 4.
Moreover,
confirmed we thatcalculated the usage
such sputtering couldof be
theemployed
sputter targets by using
to prepare filmsthe target erosion
regardless of theratio
targetequation,
size, as
as follows.
shown in Figure 4. Moreover, we calculated the usage of the sputter targets by using the target
erosion ratio equation, as follows.
Target erosion ratio (%) = [(Erosion target area)/(Total target area)] × 100
 Coatings
Coatings 2020,2020,
10, x 10,
FOR x FOR
PEERPEER REVIEW
REVIEW 6 of 613of 13

Coatings 2020, 10, 321 6 of 13

Target
Target erosion
erosion ratioratio
(%) =(%) = [(Erosion
[(Erosion target
target area)/(Total
area)/(Total target
target )] × )]
areaarea 100× 100

Figure 4. Photos of targets after sputtering and cross-sectional profiles of sputter target erosion.
Figure
Figure 4. Photos of targets after sputtering
and and cross-sectional profiles of sputter target erosion.
(a) (a)
(a) 4. Photos
Open-type ofFTS
targets after
and (b) sputtering
closed-type FTS. cross-sectional profiles of sputter target erosion.
Open-type FTS and (b) closed-type
Open-type FTS and (b) closed-type FTS. FTS.
Figure 5 shows the eroded region in the surface of the target after sputtering. Using the equation,
Figure
Figure 5 shows eroded
5 shows the eroded region in surface
the surface of target
the target after sputtering. Using equation,
the equation,
the target erosiontheratio of theregion
OFTS in the
system of the
is determined to beafter
53% sputtering.
and that of Using
CFTS isthe
70%. The target
the target
the target erosion
erosion ratio of the
ratio of decreased OFTS
the OFTS when system
systemthe is determined
is arrangement
determined to to be 53% and that of CFTS is 70%.
The The
material consumption of be
the53% and that
magnets was of CFTS is
changed 70%.
from open-type
target
target material consumption
material consumption decreased
decreased when when the arrangement of the magnets was changed
fromfrom
to closed-type for sputtering using the samethe arrangement
film. It is suggested of the magnets
that enhancingwasthe
changed
flux densities at
open-type
open-type to closed-type
to closed-type for sputtering
for sputtering usingusing the same
the same film. film.
It is Itsuggested
is suggested that enhancing the flux
the target surface enhance the sputtering efficiency by as much as 50%. that enhancing the flux
densities
densities at target
at the the target surface
surface enhance
enhance the sputtering
the sputtering efficiency
efficiency by as bymuch
as much as 50%.
as 50%.

Figure
Figure Photo images
5. Photo
5. images of
of the
the target
target erosion
erosion region
region for
for aa target
target erosion
erosion ratio. (a)
(a) Open-type
Open-type FTS
FTS (OFTS)
(OFTS)
Figure 5. Photo images of the target erosion region for a target erosion ratio.ratio.
(a) Open-type FTS (OFTS)
and (b) Closed-type
andClosed-type
(b) Closed-type FTS (CFTS).
FTS (CFTS).
and (b) FTS (CFTS).
3.3. Properties of Films
3.3. Properties
3.3. Properties of Films
of Films
Figure 6 shows the discharge characteristics with respect to working pressure for sputtering at
Figure 6 shows the discharge characteristics
withwith respect to working pressure for sputtering at
aFigure 6 shows
fixed 500 W usingthe discharge characteristics
the FTS system. The typical respect
dischargetoproperties
working pressure for sputtering
of magnetron at are
sputtering
a fixed
a fixed 500 500 W using
W using the FTS system. The typical discharge propertiesmagnetron
of magnetron sputteringare are
well-known to be the FTS system.
dependent on theThe typical
discharge discharge
voltage and properties
the currentof
of the cathode,sputtering
working pressure,
magnetic flux density, and input power [21,22].
Coatings 2020, 10, x FOR PEER REVIEW 7 of 13

well-known to be dependent on the discharge voltage and the current of the cathode, working
Coatings 2020, 10, 321 7 of 13
pressure, magnetic flux density, and input power [21,22].

(a) (b)
Dischargevoltage
Figure 6. Discharge voltagewith
withrespect
respecttoto working
working pressure
pressure forfor
(a) (a) open-type
open-type FTSFTS (OFTS)
(OFTS) andand
(b)
(b) closed-type
closed-type FTSFTS (CFTS).
(CFTS).

In aa normal
In sputtering system,
normal sputtering system, the
the anode
anode isis perpendicular
perpendicular to to the
the target,
target, such
such asas the
the cathode.
cathode.
However, in
However, the FTS
in the FTS system
system configuration,
configuration, twotwo targets
targets are
are connected
connected in in series
series with
withthe
thepower
powersupply,
supply,
and these act as the cathode. The anode is dark shielding and substrate. In the plasma,
and these act as the cathode. The anode is dark shielding and substrate. In the plasma, the number of the number
of Ar/Ar + atoms/ions and electrons were increased, and film sputtering enabled the growth of the
Ar/Ar + atoms/ions and electrons were increased, and film sputtering enabled the growth of the
substrate at
substrate at aa relatively low working
relatively low pressure (~1
working pressure mTorr). Therefore,
(~1 mTorr). Therefore, itit is
is crucial
crucial for
for the
the film
film in
in the
the
sputtering process. We investigated the relationship between the voltage and working
sputtering process. We investigated the relationship between the voltage and working pressure, and pressure, and the
results
the are summarized
results are summarized in Figure 6. 6.
in Figure
At a constant input power, with aa changing
At a constant input power, with changing working
working pressure
pressure inin the
the chamber,
chamber, the current (I)
the current (I) and
and
voltage (V) are as follows:
voltage (V) are as follows:
P = (4I) × (4V)
P = (△I) × (△V)
where 4I is the variation value in the current, 4V is the variation in the voltage, and P is the input
power.△I
where Regardless of the value
is the variation magnetic arrays
in the used,
current, △V theisvoltage was observed
the variation to decrease
in the voltage, and Pasis operating
the input
pressureRegardless
power. increased.ofThe theinitial voltage
magnetic of CFTS,
arrays used, which exhibits
the voltage wasa observed
higher magnetic
to decreaseflux density than
as operating
OFTS, is approximately
pressure increased. The 620 V (80.1
initial mA).
voltage of CFTS, which exhibits a higher magnetic flux density than
OFTS, The initial voltage 620
is approximately of CFTS was
V (80.1 mA)approximately 315 V, and its initial current decreased from a
maximum of 158.7
The initial voltagemA of asCFTS
the functions of working315
was approximately pressure.
V, and Therefore, in the decreased
its initial current case of thefrom OFTS a
configuration,
maximum the cathode
of 158.7 mA ascurrents were observed
the functions of workingto bepressure.
higher than those in in
Therefore, thethe
casecaseof CFTS
of theunder
OFTSa
similar working
configuration, thepressure
cathodeand input power.
currents It has been
were observed to be reported
higher that
thanthe magnetic
those field of
in the case in CFTS
FTS enables
under
acontrolling the confinement
similar working pressure of plasma
and inputaspower.
well asIta has
lower voltage
been and higher
reported that thecurrent
magnetic at a low
fieldworking
in FTS
pressure,controlling
enables in contrastthe to conventional
confinement of magnetrons
plasma as at lowaspressures.
well a lower voltage and higher current at a low
workingThe pressure,
XRD patterns of the prepared
in contrast ITO films
to conventional are shownat
magnetrons inlow
Figure 7. From the XRD patterns, it can
pressures.
be observed
The XRD that the crystal
patterns of thequality of ITO
prepared ITOwhere
filmsthearefilm is grown
shown in Figureat a 7.
working
From thepressure of 1 mTorr
XRD patterns, it
without substrate heating is almost amorphous even though two kinds
can be observed that the crystal quality of ITO where the film is grown at a working pressure of 1 of magnetic armaments were
installedwithout
mTorr in the sputtering
substrate source.
heating In is the sputtering
almost amorphousprocess, the crystallization
even though two kinds of films of ismagnetic
affected
by the film were
armaments thickness and in
installed substrate temperature
the sputtering source.[23–25]. In conventional
In the sputtering process, magnetron sputtering,
the crystallization of
the kinetic energies of sputtered particles and other energetic particles (Ar and O - ) cause an increase
films is affected by the film thickness and substrate temperature [23–25]. In conventional magnetron
in the substrate
sputtering, temperature
the kinetic energies andof enhance
sputtered the crystallinity
particles of theenergetic
and other films during the surface
particles (Ar andmigration,
O-) cause
as numerous
an increase inparticles
the substratearrivetemperature
at the substrate. Therefore,
and enhance ITO
the films withof
crystallinity a polycrystalline
the films during structure can
the surface
be grown using
migration, magnetron
as numerous sputtering
particles arriveatatroom temperature
the substrate. [24–26].ITO
Therefore, However,
films withShigesato et al. [27]
a polycrystalline
reported that
structure can ITO films with
be grown usinga thickness
magnetron of 170 nm exhibited
sputtering at room filmtemperature
growth with[24–26].
the crystallization
However,
of the (400)
Shigesato et plane
al. [27]and that polycrystalline
reported that ITO films ITO with was grown on
a thickness a glass
of 170 substrate film
nm exhibited at a growth
thickness of
with
approximately
the crystallization 240ofnm. Moreover,
the (400) in contrast
plane and with conventional
that polycrystalline ITO wasmagnetron
grown on asputtering
glass substratesystems,
at a
particles with
thickness a lower kinetic
of approximately 240 energy that impact
nm. Moreover, less plasma
in contrast damage are magnetron
with conventional produced in the FTS
sputtering
 Coatings 2020, 10, x FOR PEER REVIEW 8 of 13

Coatings 2020, 10, 321 8 of 13

systems, particles with a lower kinetic energy that impact less plasma damage are produced in the
FTS system [27,28]. Consequently, the internal damage caused by particle bombardment during the
film growth
system [27,28].isConsequently,
substantially reduced.
the internal damage caused by particle bombardment during the film
growth is substantially reduced.

Figure
Figure 7. X-ray
7. X-ray diffraction
diffraction patterns
patterns of prepared
of prepared films atfilms at a working
a working pressure pressure
of 1 mTorrofwithout
1 mTorr without
substrate
heating. (a) heating.
substrate Open-type(a) FTS (OFTS) FTS
Open-type and (OFTS)
(b) Closed-type FTS (CFTS).FTS (CFTS).
and (b) Closed-type

InInthe
thecase
case ofofthe FTS
the FTS system,
system, phenomena
phenomena asas similar
similar toto those
thosementioned
mentioned above
above were
were observed
observed inin
this study as well. No crystal peak could be observed in the XRD pattern due
this study as well. No crystal peak could be observed in the XRD pattern due to ITO film being 100- to ITO film being 100-nm
thick. Moreover,
nm thick. Moreover,its substrate was located
its substrate outside
was located the high-density
outside the high-density plasma region.
plasma The increase
region. The increasein
substrate temperature
in substrate was relatively
temperature low during
was relatively lowtheduring
sputteringthe process. Theprocess.
sputtering relativelyThelowrelatively
temperature low
could contribute to be the final ITO film being amorphous
temperature could contribute to be the final ITO film being amorphous [29,30]. [29,30].
ToToinvestigate
investigate thethe
effect of temperature
effect of temperature on theonsputtering,
the sputtering,we employed a thermos
we employed label taplabel
a thermos (NiGK tap
Corportion Company, 80–95 ◦ C). For conventional DC sputtering, we used DC magnetron sputtering
(NiGK Corportion Company, 80–95 °C). For conventional DC sputtering, we used DC magnetron
(ITO target size:
sputtering (ITO2target
inch) size:
in our 2 laboratory.
inch) in ourWe observedWe
laboratory. theobserved
thermos thelabelthermos
tape in label
the sputtering
tape in the
condition
sputtering including
condition 3 mTorr with Ar
including atmosphere
3 mTorr with Ar andatmosphere
distance from andthe target to
distance the substrate
from the target(DtoT-Sthe
)
atsubstrate
100 mm. (D The substrate temperature is shown in Figure 8. Before the substrates
T-S) at 100 mm. The substrate temperature is shown in Figure 8. Before the substrates were were loaded onto
the holderonto
loaded in the
thechamber,
holder inthe thetape was attached
chamber, the tapebehind the substrate,
was attached behindandthethe color change
substrate, and the of color
the
tapes
changewasofobserved.
the tapes Figure 8 showsFigure
was observed. the images
8 showsin which reversed
the images dark-black
in which reversed colored spots can
dark-black be
colored
observed after and before sputtering for 1 h. The substrate temperature increased to at least 80 ◦ C even
spots can be observed after and before sputtering for 1 hour. The substrate temperature increased to
though
at leastfilm
80 sputtering
°C even though was successfully completed.
film sputtering In contrast with
was successfully normalIn
completed. DC magnetron
contrast withsputtering,
normal DC
nomagnetron
color change was observed in the tape in the case of all types of FTS
sputtering, no color change was observed in the tape in the case of all types systems. Therefore, it is thought
of FTS
that FTS could
systems. be employed
Therefore, to prepare
it is thought that films at a low
FTS could be temperature
employed to with lowfilms
prepare damage.at a low temperature
withAccording
low damage. to previous results obtained by another group of using magnetron sputtering, ITO
films could be made to exhibit improved electrical properties by adding a small amount of oxygen gas
to the pure argon gas atmosphere during sputtering. We investigated the effect of oxygen gas flow on
the electrical properties of ITO films, and sheet resistance was measured with respect to an increase in
the oxygen amount in the Ar working gas atmosphere, as shown in Figure 9. At oxygen flow rates of
0 to 1.0 sccm, the ITO film (100 nm) exhibited a sheet resistance of 550 ohm/sq in the case of OFTS and
820 ohm/sq in the case of CFTS. With an increasing amount of oxygen gas, sheet resistance of the films
decreased from 820 to 85.53 ohm/sq (CFTS) and from 550 to 67.42 ohm/sq (OFTS).
Coatings 2020, 10, 321 9 of 13
Coatings 2020, 10, x FOR PEER REVIEW 9 of 13

Figure 8. Color changes of the thermo label tape before and after sputtering. (a) Before sputtering, (b)
After sputtering in DC magnetron, (c) After sputtering in OFTS, and (d) After sputtering in CFTS.

According to previous results obtained by another group of using magnetron sputtering, ITO
films could be made to exhibit improved electrical properties by adding a small amount of oxygen
gas to the pure argon gas atmosphere during sputtering. We investigated the effect of oxygen gas
flow on the electrical properties of ITO films, and sheet resistance was measured with respect to an
increase in the oxygen amount in the Ar working gas atmosphere, as shown in Figure 9. At oxygen
flow rates of 0 to 1.0 sccm, the ITO film (100 nm) exhibited a sheet resistance of 550 ohm/sq in the case
of OFTS and 820 ohm/sq in the case of CFTS. With an increasing amount of oxygen gas, sheet
resistance of the films decreased from 820 to 85.53 ohm/sq (CFTS) and from 550 to 67.42 ohm/sq
Figure 8.
(OFTS).
Figure Colorchanges
8. Color changesofofthe
thethermo
thermolabel
labeltape
tape before
before and
and after
after sputtering.
sputtering. (a) (a) Before
Before sputtering,
sputtering, (b)
(b) After
After sputtering
sputtering in DC
in DC magnetron,
magnetron, (c) After
(c) After sputtering
sputtering in OFTS,
in OFTS, andand
(d) (d) After
After sputtering
sputtering in CFTS.
in CFTS.

According to previous results obtained by another group of using magnetron sputtering, ITO
films could be made to exhibit improved electrical properties by adding a small amount of oxygen
gas to the pure argon gas atmosphere during sputtering. We investigated the effect of oxygen gas
flow on the electrical properties of ITO films, and sheet resistance was measured with respect to an
increase in the oxygen amount in the Ar working gas atmosphere, as shown in Figure 9. At oxygen
flow rates of 0 to 1.0 sccm, the ITO film (100 nm) exhibited a sheet resistance of 550 ohm/sq in the case
of OFTS and 820 ohm/sq in the case of CFTS. With an increasing amount of oxygen gas, sheet
resistance of the films decreased from 820 to 85.53 ohm/sq (CFTS) and from 550 to 67.42 ohm/sq
(OFTS).

Figure 9. Electrical
Figure properties
9. Electrical of of
properties prepared
preparedfilms as as
films a function of oxygen
a function gasgas
of oxygen flow. (a)(a)
flow. Open-type FTS
Open-type FTS
(OFTS) andand
(OFTS) (b)(b)
Closed-type FTSFTS
Closed-type (CFTS).
(CFTS).

It is
It known
is known that thethe
that electrical properties
electrical propertiesof of
ITO films
ITO depend
films depend onon
free carriers,
free a strong
carriers, a strong scattering
scattering
center, and oxygen vacancies. The increase in sheet resistance was due to the
center, and oxygen vacancies. The increase in sheet resistance was due to the decrease of the decrease of the carrier
carrier
mobility [31]. The mobility was improved due to an increasing oxygen flow rate
mobility [31]. The mobility was improved due to an increasing oxygen flow rate during sputtering, during sputtering,
even though carrier concentration decreased [32,33]. The enhancement of sheet resistance may be
attributed to the formation of SnO2 in the ITO films, which results in the decrease of oxygen vacancies
in the films, as the electrons that are released due to the substitution enter Sn atoms in the sublattice
and form doubly charged oxygen vacancies [26]. Meng [34] reported that a high insertion of oxygen
gas could damage the grown film and substrate as a result of an increase in the plasma bombardment
energy during
Figure film sputtering
9. Electrical and
properties changes in
of prepared theasfilm
films deposition
a function rate in
of oxygen gasthe conventional
flow. (a) Open-type magnetron
FTS
sputtering
(OFTS)system.
and (b) Closed-type FTS (CFTS).
However, based on XRD patterns (Figure 7) and sheet resistance of ITO films (Figure 9), all ITO
films,Itgrown
is known in the
thatOFTS and CFTS,
the electrical have noofcrystal
properties peakdepend
ITO films and weon confirmed thatathe
free carriers, oxygen
strong gas is
scattering
center, and oxygen vacancies. The increase in sheet resistance was due to the decrease of the carrier
mobility [31]. The mobility was improved due to an increasing oxygen flow rate during sputtering,
 in the films, as the electrons that are released due to the substitution enter Sn atoms in the sublattice
and form doubly charged oxygen vacancies [26]. Meng [34] reported that a high insertion of oxygen
gas could damage the grown film and substrate as a result of an increase in the plasma bombardment
energy during film sputtering and changes in the film deposition rate in the conventional magnetron
sputtering
Coatings system.
2020, 10, 321 10 of 13
However, based on XRD patterns (Figure 7) and sheet resistance of ITO films (Figure 9), all ITO
films, grown in the OFTS and CFTS, have no crystal peak and we confirmed that the oxygen gas is
more
moreeffective
effectivewhenwhenenhancing
enhancingthe theelectrical
electricalproperties
propertiesof ofITO
ITOfilms
filmsusing
usingless
lessthan
than5%5%of ofoxygen
oxygengas gas
in an Ar/O mixture.
in an Ar/O22 mixture.
The
The spectral
spectral transmittance
transmittance of of the
the ITO
ITO films
films was
was obtained
obtained by by scanning
scanning the the films
films from
from 340 nm to
340 nm to
800 nm and
800 nm andisisillustrated
illustratedin inFigure
Figure10.10. All
All ITO
ITO films
films on
on the
the glass
glass substrate
substrate deposited
deposited under
under different
different
oxygen
oxygenflowflowrates
ratesexhibit
exhibittransmittance
transmittance ofof
more
more than 85%
than 85%at the wavelength
at the wavelength of approximately
of approximately 550 nm550
except for thefor
nm except samples prepared
the samples without oxygen
prepared withoutgas. With an
oxygen increasing
gas. With anoxygen flow, oxygen
increasing the transmittance
flow, the
attains a maximum
transmittance attainsvalue of 95% forvalue
a maximum the wavelength
of 95% for of the550 nm. Guptaofet550 nm.
wavelength al. [35] reported
Gupta etthat the
al. [35]
absorption
reported that edge theisabsorption
moved to edgehigheris energies
moved to(referred to as blue
higher energies shift ortoBurstein-Moss
(referred as blue shift or (BM)) as a
Burstein-
result
Moss of the as
(BM)) high charge
a result of carrier
the highconcentration
charge carrierofconcentration
ITO films as arising from as
of ITO films anarising
increase in the
from oxygen
an increase
gas flow.
in the Moreover,
oxygen although
gas flow. the BM
Moreover, phenomenon
although the BM could be one of the
phenomenon reasons
could be one forofthe
theblue shiftfor
reasons of the
the
absorption
blue shift ofedge is the high charge
the absorption edge iscarrier
the highconcentration.
charge carrier Therefore, the charge
concentration. carrierthe
Therefore, concentration
charge carrierof
these samples should
concentration of theseincrease
sampleswith an rise
should in thewith
increase oxygen gas in
an rise flow
therate [36,37].
oxygen gas flow rate [36,37].

(a) (b)
Figure10.
Figure 10. Optical
Optical transmittance
transmittance spectra
spectra of ITO thin films deposited under different oxygen flow
different oxygen flow rates.
(a)Open-type
(a) Open-typeFTS
FTSand
and(b)
(b)Closed-type
Closed-typeFTS.
FTS.

The
The roughness
roughness of ofeach
eachsurface
surfacewaswasobserved
observedusing
usingSPM,
SPM,asasshown
shownin inFigure
Figure11.
11. More
Moredetailed
detailed
values
values of the surface roughness are shown in Table 2. The surface roughness (RMS) of the ITO film
of the surface roughness are shown in Table 2. The surface roughness (RMS) of the ITO film
prepared
preparedvia viaOFTS,
OFTS,fromfrom1.565
1.565to to
2.006 nm,nm,
2.006 is shown in Figure
is shown 11a,b.11a,b.
in Figures The RThe
a (roughness average)
Ra (roughness value
average)
of the film
value of thealso
filmslightly increased
also slightly from 2.151
increased fromto 2.575
2.151 tonm. Using
2.575 AFM, AFM,
nm. Using we confirmed that ITO
we confirmed thatfilms
ITO
have the
films have
Coatings smoothened
thex FOR
2020, 10, surface
smoothened
PEER REVIEW regardless of the magnetic type.
surface regardless of the magnetic type. 11 of 13

Figure 11. Surface morphology of ITO film obtained via (a) open-type FTS and (b) closed-type FTS.
Figure 11. Surface morphology of ITO film obtained via (a) open-type FTS and (b) closed-type FTS.

Table 2. Surface roughness values of prepared ITO films.

Magnetic Type Ra (nm) RMS (nm)

Open-type FTS(1) 2.006 2.575

Coatings 2020, 10, 321 11 of 13

Table 2. Surface roughness values of prepared ITO films.

Magnetic Type Ra (nm) RMS (nm)

Open-type FTS(1) 2.006 2.575
Closed-type FTS(2) 1.565 2.151
Sputtering conditions: (1) Oxygen gas flow = 0.8 sccm and (2) Oxygen gas flow = 0.6 sccm.

4. Conclusions
It is well known that the FTS system possesses an off-axis substrate-target geometry in contrast
with traditional magnetron sputtering with an on-axis geometry in which the substrate faces the target.
We investigated the effect of the magnet arrangement inside the cathode of the FTS system on the
sputtering outcome. Prior to analyzing the magnetic array and performing sputtering, we predicted
plasma confinement on the target surface using the result of a magnetism simulation. We selected
two kinds of sputtering systems known as OFTS and CFTS, and inserted two cathodes in the
systems. In the optimized conditions of sputtering, the obtained films (thickness: 100 nm) exhibited
optical transmittance up to 85% in the visible range and approximately 67 ohm/sq without heating.
Moreover, a maximum target erosion of 70% was observed, and it is considered that this could enhance
the sputtering efficiency by as much as 50% if the target size is reduced to half the original size of
the target. The substrate temperature was relatively low compared to that exhibited by conventional
magnetron sputtering even though film sputtering was successfully achieved.

Author Contributions: S.K.; Writing-review and Original draft preparation. K.H.K. Reviewing and editing.
All authors have read and agreed to the published version of the manuscript.
Funding: Gachon University research fund of 2019 (GCU-2019-0302) and Basic Science Research Capacity
Enhancement Project through Korea Basic Science Institute (National research Facilities and Equipment Center)
grant funded by the Ministry of Education (Grant No. 2019R1A6C1010016).
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Pereira, P.J.S.; Escrivão, M.L.; Teixeira, M.R.; Maneira, M.J.P.; Nunes, Y. Analytical model and measurements
of the target erosion depth profile of balanced and unbalanced planar magnetron cathodes. Plasma Sources
Sci. Technol. 2014, 23. [CrossRef]
2. Hou, Q.; Buckeridge, J.; Lazauskas, T.; Mora-Fonz, D.; Sokol, A.A.; Woodley, S.M.; Catlow, C.R.A.
Defect formation in In2 O3 and SnO2 : A new atomistic approach based on accurate lattice energies. J. Mater.
Chem. C 2018, 6, 12386–12395. [CrossRef]
3. Tien, C.-L.; Lin, H.-Y.; Chang, C.-K.; Tang, C.-J. Effect of Oxygen Flow Rate on the Optical, Electrical,
and Mechanical Properties of DC Sputtering ITO Thin Films. Adv. Condens. Matter Phys. 2018, 2018.
[CrossRef]
4. Shin, H.-S.; Seo, K.-W.; Lee, J.-H.; Kim, H.-K. Effect of Target–Substrate Distance on the Properties of Hetero
Facing Target Sputtered Al-Ga-Zn-O Films. Jpn. J. Appl. Phys. 2013, 52. [CrossRef]
5. Xiao, Y.; Gao, F.; Dong, G.; Guo, T.; Liu, Q.; Ye, D.; Diao, X. Amorphous indium tin oxide films deposited
on flexible substrates by facing target sputtering at room temperature. Thin Solid Films 2014, 556, 155–159.
[CrossRef]
6. Kadokura, S.; Naoe, M.; Nakagawa, S. Nano-size Magnetic Crystallite Formation in Go-Cr Thin Films for
Perpendicular Recording Media. IEEE Trans. Magn. 1998, 34, 1642–1644. [CrossRef]
7. Noda, K.; Kadokura, S.; Naoe, M. Microstructure and noise characteristics of Co-Cr-Ta films on ultraflat
glass substrates for longitudinal recording disks. J. Appl. Phys. 2000, 87, 6343–6345. [CrossRef]
8. Matsushita, N.; Nakagawa, S.; Naoe, M. Preparation of Co ferrite thin films with large perpendicular and
in-plane coercivities by facing targets sputtering. IEEE Trans. Magn. 1992, 28, 3108–3110. [CrossRef]
9. Kim, H.-K.; Kim, S.W.; Lee, K.-S.; Kim, K.H. Direct Al cathode layer sputtering on LiF@Alq3 using facing
target sputtering with a mixture of Ar and Kr. Appl. Phys. Lett. 2006, 88. [CrossRef]
Coatings 2020, 10, 321 12 of 13

10. Fan, Q.H.; Chen, H.Y. Effects of magnetic field on the target erosion in the symmetrical magnets magnetron
sputtering method. Thin Solid Films 1993, 229, 143–145. [CrossRef]
11. Liu, C.-T.; Lai, M.-C.; Hwang, C.-C.; Tu, C.-H.; Liu, L.-Y.; Hsu, Y.-W. Enhancements of Substrate Deposition
Rate and Target Erosion Profile in a DC Magnetron Sputtering System. IEEE Trans. Magn. 2009, 45, 4391–4394.
[CrossRef]
12. Lin, C.; Sun, D.C.; Ming, S.L.; Jiang, E.Y.; Liu, Y.G. Magnetron facing target sputtering system for fabricating
single-crystal films. Thin Solid Films 1996, 279, 49–52. [CrossRef]
13. Shin, H.-S.; Lee, J.-H.; Kwak, J.-S.; Lee, H.-H.; Kim, H.-K. Linear facing target sputtering of the epitaxial
Ga-doped ZnO transparent contact layer on GaN-based light-emitting diodes. J. Phys. D Appl. Phys. 2013,
46, 1–6. [CrossRef]
14. Murphy, M.J.; Cameron, D.C.; Karim, M.Z.; Hashmi, M.S.J. Magnetic fields in magnetron sputtering systems.
Surf. Coat. Technol. 1993, 57, 1–5. [CrossRef]
15. Hollerweger, R.; Holec, D.; Paulitsch, J.; Rachbauer, R.; Polcik, P.; Mayrhofer, P.H. Magnetic field strength
influence on the reactive magnetron sputter deposition of Ta2 O5 . J. Phys. D Appl. Phys. 2013, 46. [CrossRef]
16. Noda, K.; Hirata, T.; Takashi, K.; Naoe, M. Novel facing targets sputtering apparatus with uniform magnetic
field and plasma-free substrates. Vacuum 1998, 51, 687–690. [CrossRef]
17. Hoshi, Y.; Naoe, M.; Yamanka, S. High Rate Deposition of Iron Films by Sputtering from Two Facing Targets.
Jpn. J. Appl. Phys. 1977, 16, 1715–1716. [CrossRef]
18. Kageyama, J.; Yoshimoto, M.; Matsuda, A.; Akao, Y.; Shidojim, E. Numerical simulation of plasma confinement
in DC magnetron sputtering under different magnetic fields and anode structures. Jpn. J. Appl. Phys. 2014,
53. [CrossRef]
19. Guo, T.; Dong, G.; Liu, Q.; Wang, M.; Wang, M.; Gao, F.; Chen, Q.; Yan, H.; Diao, X. Optimization of oxygen and
pressure of ZnO: Al films deposited on PMMA substrates by facing target sputtering. Superlattices Microstruct.
2013, 64, 552–562. [CrossRef]
20. Kim, H.-K.; Moon, J.-M. Sputtering of Aluminum Cathodes on OLEDs Using Linear Facing Target Sputtering
with Ladder-Type Magnet Arrays. J. Electrochem. Soc. 2008, 155, J187–J192. [CrossRef]
21. Liebrerman, M.A.; Lichtenberg, A.J. Principles of Plasma Discharges and Materials Processing, 2nd ed.; John Wiley
& Sons Inc.: Hoboken, NJ, USA, 2005.
22. Westwood, W.D.; Maniv, S.; Scanlon, P.J. The current-voltage characteristic of magnetron sputtering systems.
J. Appl. Phys. 1983, 54, 6841–6846. [CrossRef]
23. Ghorannevis, Z.; Akbarnejad, E.; Ghoranneviss, M. Structural and morphological properties of ITO thin
films grown by magnetron sputtering. J. Theor. Appl. Phys. 2015, 9, 285–290. [CrossRef]
24. Morikawa, H.; Fujita, M. Crystallization and decrease in resistivity on heat treatment of amorphous indium
tin oxide thin films prepared by dc magnetron sputtering. Thin Solid Films 1999, 339, 309–313. [CrossRef]
25. Shigesato, Y.; Panie, D.C. A microstructural study of low resistivity tin-doped indium oxide prepared by dc
magnetron sputtering. Thin Solid Films 1994, 238, 44–50. [CrossRef]
26. Hosono, H.; Ueda, K. Transparent Conductive Oxides. In Springer Handbook of Electronic and Photonic Materials,
2nd ed.; Kasap, S., Capper, P., Eds.; Springer: New York, NY, USA, 2017; pp. 1391–1403.
27. Shigesato, Y.; Koshi-ishi, R.; Kawashima, T.; Ohsako, J. Early stages of ITO deposition on glass or polymer
substrates. Vacuum 2000, 59, 614–621. [CrossRef]
28. Lines, V.; Bivour, M.; Iwata, H.; Ortner, K. Comparison of low damage sputter deposition techniques to
enable the application of very thin a-Si passivation films. AIP Conf. Proc. 2019, 2147. [CrossRef]
29. Song, P.K.; Akao, H.; Kamei, M.; Shigesato, Y.; Yasui, I. Preparation and Crystallization of Tin-doped and
Undoped Amorphous Indium Oxide Films Deposited by Sputtering. Jpn. J. Appl. Phys. 1999, 38, 5224–5226.
[CrossRef]
30. Ahmed, N.M.; Sabah, F.A.; Abdulgafour, H.I.; Alsadig, A.; Sulieman, A.; Alkhoaryef, M. The effect of post
annealing temperature on grain size of indium-tin-oxide for optical and electrical properties improvement.
Results Phys. 2019, 13. [CrossRef]
31. Chen, A.; Zhu, K.; Zhong, H.; Shao, Q.; Ge, G. A new investigation of oxygen flow influence on ITO thin
films by magnetron sputtering. Sol. Energy Mater. Sol. Cells 2014, 120, 157–162. [CrossRef]
32. Li, S.; Qiao, X.; Chen, J. Effects of oxygen flow on the properties of indium tin oxide films. Mater. Chem. Phys.
2006, 98, 144–147. [CrossRef]
Coatings 2020, 10, 321 13 of 13

33. Tahar, R.B.H.; Ban, T.; Ohya, Y.; Takahashi, Y. Tin doped indium oxide thin films: Electrical properties.
J. Appl. Phys. 1998, 83, 2631–2645. [CrossRef]
34. Meng, L.-J.; Maçarico, A.; Martins, R. Study of annealed indium tin oxide films prepared by rf reactive
magnetron sputtering. Vacuum 1995, 6, 673–680. [CrossRef]
35. Gupta, L.; Mansingh, A.; Srivastava, P.K. Band GAP Narrowing and the band structure of tindoped indium
oxide films. Thin Solid Films 1989, 176, 33–44. [CrossRef]
36. Cho, S.H.; Park, J.H.; Lee, S.C.; Cho, W.S.; Lee, J.H.; Yon, H.H.; Song, P.K. Effects of O2 addition on
microstructure and electrical property for ITO films deposited with several kinds of ITO targets. J. Phys.
Chem. Solids 2008, 69, 1334–1337. [CrossRef]
37. Kim, C.E.; Moon, P.; Kim, S.; Myoung, J.-M.; Jang, H.W.; Bang, J.; Yun, I. Effect of carrier concentration on
optical bandgap shift in ZnO: Ga thin films. Thin Solid Films 2010, 518, 6304–6307. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).