Investigation of growth, coverage and effectiveness
of plasma assisted nano-films of fluorocarbon
Pratik P. Joshia,*, Rajasekhar Pulikollua, Steven R. Higginsb,
Xiaoming Hub, S.M. Mukhopadhyaya
aDepartment of Mechanical and Materials Engineering, Wright State University, Dayton, OH 45435, USA
bChemistry Department, Wright State University, Dayton, OH 45435, USA
Received 26 January 2005; received in revised form 15 June 2005; accepted 20 July 2005
Available online 27 September 2005
Plasma-assisted functionalfilmshavesignificant potentialinvariousengineering applications.Theycanbetailoredtoimpart
desired properties by bonding specific molecular groups to the substrate surface. The aim of this investigation was to develop a
fundamental understanding of the atomic level growth, coverage and functional effectiveness of plasma nano-films on flat
surfaces and to explore their application-potential for complex and uneven shaped nano-materials. In this paper, results on
discussed. The film deposition was studied as a function of time on flat single crystal surfaces of silicon, sapphire and graphite,
using microwave plasma. X-ray photoelectron spectroscopy (XPS) was used for detailed study of composition and chemistry of
the substrate and coating atoms, at all stages of deposition. Atomic force microscopy (AFM) was performed in parallel to study
the coverage and growth morphology of these films at each stage. Combined XPS and AFM results indicated complete coverage
of all the substrates at the nanometer scale. It was also shown that these films grew in a layer-by-layer fashion. The nano-films
were also applied to complex and uneven shaped nano-structured and porous materials, such as microcellular porous foam and
nano fibers. It was seen that these nano-films can be a viable approach for effective surface modification of complex or uneven
# 2005 Elsevier B.V. All rights reserved.
Keywords: Growth; Plasma; Nano-film
Plasma thin film deposition techniques are very
useful in many engineering applications. They are
known for their capability of depositing conformal
ultra thin films (several nm) as well as relatively thick
Applied Surface Science 252 (2006) 5676–5686
* Corresponding author. Tel.: +1 937 7755016;
fax: +937 7755009.
E-mail address: firstname.lastname@example.org (P.P. Joshi).
0169-4332/$ – see front matter # 2005 Elsevier B.V. All rights reserved.
films (several mm thick). Applications of plasma films
include improved hydrophobicity, hydrophilicity,
biocompatibility, printability, adhesion, and corrosion
protection. It is agreed upon that plasma deposition
techniques are highly system dependent and influ-
enced by process parameters [1–8]. Optimization of
these parameters and their impact on coating
chemistry has been investigated by many researchers.
Previous publications from this group show that
plasma time and coating chemicals play a crucial role
in the deposition processes [9,10].
Despite the earlier studies performed on thick films
for conventional materials, the fundamental mechan-
isms that explain how functional groups in the plasma
bond with the substrate atoms, begin to nucleate and
eventually grow into complete films are not discussed.
Systematic study ofthe nucleation andgrowth of these
films is needed, to obtain an estimation of the
minimum possible thickness of the coating; that still
provides effective surface modification. This basic
understanding could be extended to effective surface
modification of complex shaped nano-structured
The objective of our research is to characterize,
initial states of nucleation, growth and coverage
of plasma-assisted nano-films. This was accom-
plished by depositing these films as a function of
total deposition time on flat surfaces and analyzing,
ex situ, the change in chemical composition,
morphology and coverage, at the nanometer scale.
The aim of investigation is to discern weather these
nano-films grow as individual islands (three-dimen-
sional growth) or layers (two-dimensional nuclea-
tion growth).In this
characterization of inert fluorocarbon films are
Fluorocarbon films are known for making the
surface inert and thereby make the surface hydro-
phobic or water repellent. This phenomenon is
attributed to the presence of ?CF2/?CF3functional
groups present in these films. The water contact angle
value of such fluorocarbon coated surfaces is around
140–1708 indicating super-hydrophobic surfaces [11–
13]. A commercial Teflon (PTFE) surface has a water
contact angle very close to these values. These films
may have potential application in preventing aggrega-
tion of nanomaterials through surface modification
Plasma-assisted fluorocarbon film deposition was
performed using a microwave plasma reactor (Plas-
maTech Inc., Model No.V-15 G) with multimode
microwave source. The aluminum vacuum chamber
size is 250 mm ? 250 mm ? 250 mm (15.6 l), which
uses a microwave generator magnetron of 850 W
(electrodeless chamber) and the maximum power
available is 600 W. Grade 4.0 octafluoropropane
(C3F8) was used as the coating chemical. The gas
flow rate inside the treatment chamber for all coating
times was kept constant at 50 ml/s.
Flat single crystal substrates of silicon, sapphire
and graphite were used as model surfaces for studying
the chemistry and the morphology of films. Hydro-
philic materials (initial water contact angle 08), which
include flat surfaces as well as inner layers of a multi-
layered stack of porous materials with different
permeation levels were also coated with these
fluorocarbon nanolayers in a wide range of experi-
mental conditions to study the effect on surface
property change using water contact angle analysis.
These results were then correlated with quantitative
XPS data. Fig. 1 shows a typical set-up used for
depositing fluorocarbon films on a batch of model
surfaces. The coating time was varied from 15 s to
10 min in increments of 15 s. These films were also
deposited on carbon foam, a three dimensional
microcellular porous material that may have future
use in lightweight aerospace composites [15–17]. This
complex structure is known to contain layers, flakes,
as well as nano-fibers of graphitic carbon .
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865677
Fig. 1. Microwave plasma chamber set-up showing how model
surfaces are treated in a batch as a function of plasma time.
X-ray photoelectron spectroscopy (XPS) (Kratos-
Axis Ultra System) with monochromatized Al Ka
photons (1486.6 eV) was used for identifying the
chemical composition and bonding states of the top
aperture. The nano-scale surface morphology and
coverage of films were monitored using atomic force
microscopy (AFM) at each stage of deposition. Water
contact angle measurement was used as a direct
measure of effectiveness of the film. Contact angle
results were used in conjunction with XPS to
investigate how thick a film we need in order to
angle > 908).
3. Results and discussion
Fig. 2A shows how the carbon XPS peak changes
evident that the untreated graphite has a single
component carbon peak occurring at a binding energy
(BE) of 284.5 eV, also known as the graphitic carbon
peak. This peak started diminishing as the fluoro-
carbon film deposited on graphite and an additional
peak due to film started showing up at higher BE of
290.3 eV, identified as CF2component. As the coating
time (and therefore thickness) increased, the graphitic
carbon peak diminished and eventually disappeared
whereas the CF2component intensity increased. The
CF2 functional group is responsible for providing
inertness/hydrophobicity to the substrate surface. The
increase in the CF2 component and concomitant
decrease inthegraphitic carbonpeak canbecorrelated
withthe increase inthefluorineonthegraphitesurface
(Fig. 2B). Similar trends of substrate peak attenuation
and silicon substrates (Fig. 3). The attenuation of the
substrate XPS peaks and growth of the coating peaks
as the fluorocarbon film grows on model surfaces can
be explained further as follows.
In XPS, electrons emitted from atoms below the
outer-most layer undergo elastic and inelastic colli-
sions with the outer layer electrons. A decrease in
intensity of as well as energy of electrons occurs
because of these collisions. These lower energy
electrons, also called secondary electrons, form the
background of the XPS spectrum. The sharp peaks
come only from the electrons in the top few layers of
the material that escaped without experiencing any
inelastic collisions. As the coating forms on the
surface, the substrate electrons experience resistance
from the coating material in form of increased
inelastic collisions. As a result the intensity and
energy of the substrate electron decreases. As the film
thickness increases this resistance to the substrate
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865678
Fig. 2. (A) Decay in the substrate carbon peak and subsequent increase in the CF2 peak on graphite surface as the fluorocarbon film grows. (B)
Increase in the fluorine peak on the surface of graphite as the fluorocarbon film grows indicating film thickness increases as deposition time
electrons increases, which is the main reason for the
attenuation of the substrate signal.
This attenuated substrate signal I can be expressed
I ¼ Ioe?x=l
were, Iois the intensity of the substrate signal in the
absence of the film, l the inelastic mean free path of
substrate electrons and x is the average thickness of
the deposited layer with same units as l. When
the coating thickness (x) becomes several times grater
the substrate signal falls below detection level. This
equation can be used to estimate the over-layer thick-
ness (x) ofthe fluorocarbonfilm ifthe XPS intensityof
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865679
Fig. 3. Decrease of substrate peaks in model surfaces with fluorocarbon film growth on (A) sapphire and (B) silicon.
Fig. 4. Estimated over-layer thickness of fluorocarbon film depos-
ited on model surfaces for coating time <1 min. This thickness is
estimated over an analysis area with 110 mm diameter in XPS.
deposited coating clusters. (A) Phase image of silicon surface treated
with fluorocarbon coating for 40 s indicatingdetectable phase change
a substrate peak (that is absent in the film) is measured
before and after the film deposition.
The inelastic mean free path (IMFP) used in above
l ¼ kEm
where, k and m are materials parameters proposed by
Wagner,Davisand Riggs .This equationiswidely
accepted as the closest estimate for a standard solid.
The escape depth values used in this work for the core
electrons of model substrates (Si 2p, Al 2p and C 1s)
are taken from literature [21,22] and are 2.5, 2.0 and
2.7 nm, respectively.
Using the Eq. (1), the ‘‘average’’ film thickness
which complete attenuation of substrate signals is
occurring is in the range of 6–9 nm. Starting from the
initial stages of deposition to all the way to thick over-
layer, deposited film was characterized. AFM was used
to estimate the ‘‘average’’ over-layer thickness.
Fig. 4 shows a graphical representation of
fluorocarbon over-layer thickness as a function of
coating time up to 1 min. A fairly constant rate of film
growth could be seen for each of the model surfaces.
Estimated over layer thickness for 1minute deposition
time it was about 4, 7 and 8 A˚for sapphire, silicon and
graphite respectively. Though all surfaces were coated
simultaneously (together as a batch in plasma
chamber) for specific coating time, the growth rate
was distinctly different. This could be due to various
reasons, such as difference in surface activity and
roughness. These treated surfaces were observed in
AFM for morphology and coverage characterization.
Phase imaging capability of AFM was used to
investigatethegrowthof deposited fluorocarbon films.
The cantilever of the AFM is driven by a piezoelectric
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865680
Fig. 6. AFM phase images taken on fluorocarbon coated silicon surface indicating dispersed coating clusters and increased coverage as coating
device at resonance frequency and phase. Upon
interaction with the surface, this phase tends to
change. The resulting electric phase lag or phase shift
of the cantileveroscillation relative to driving signal is
dependenton thematerialsproperties [23,24].Plotting
this would provide contrast between dissimilar
materials (substrate and coating), which can be used
to observe the growth morphology. Fig. 5 shows an
example of AFM phase image obtained on a silicon
surface coated with fluorocarbon coating for 40 s.
Fig. 6 shows similar images on silicon surface for 30 s
to 1 min coating time. It was evident that the films
were patchy with detectable phase change in the
cantilever vibration over the film when compared to
the substrate. They indicate sub-monolayer coverage
at these deposition times. It was noticed that the film
had formed different size clusters of several hundred
nanometers (lateral dimension). The average height of
these clusters was found to be in the range of 2–3 nm
as indicated by the line profiles taken on the surface
(Figs. 6 and 7). These clusters, which are randomly
dispersed on the substrate surface, tend to grow in
lateral direction at a faster pace than in the vertical
direction. It should be noted that as the coating time
increases from 30 s to 1 min, the substrate coverage
had increased significantly in lateral direction with
almost same vertical height (Figs. 6 and 7 combined).
These individual islands (average height of ?3 nm)
eventually merged into a complete layer between 2
and 3 min of deposition time, as shown below.
Fig. 8 shows a comparison between AFM phase
images (2 mm ? 2 mm area) of 2 and 3 min fluor-
ocarbon coated silicon surfaces. For the coating time
of 2 min, very small patches of the substrate surface
was still exposed, whereas for a 3 min coating time,
complete coverage of the underlying surface was seen
with no detectable phase change (constant phase).
This means that within 2–3 min of deposition time,
fluorocarbon film completely covered the underlying
surface. Fig. 9 shows AFM topographic images of
silicon surface treated for 3 and 5 min with their phase
images, indicating layer-by-layer growth of fluoro-
carbon film at nano-scale with no evident phase
change. AFM data taken on sapphire surface revealed
similar trend as that of silicon (Fig. 10).
XPS results obtained on fluorocarbon treated
samples were evaluated in relation with relevant
AFM images. Fig. 11 show a comparison of the
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865681
Fig. 7. Line profile taken on fluorocarbon treated silicon surfaces
for different coating time indicating average vertical height of the
clusters are in the range of 2–3 nm.
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–5686 5682
Fig. 8. (A) AFM phase image for 2 min indicating patchy coating with evident phase change. (B) AFM phase image for 3 min indicating
complete coverage with no phase change.
Fig.9. AFMtopographicandphaseimagesfor3and5 minindicatinglayer-by-layergrowthwithcompletecoverageforfluorocarboncoatingon
attenuated substrate signals (silicon 2 p in this case)
for coating times between 1 and 5 min.
Following observations were made from compar-
ison of Fig. 11 with phase images obtained for 1, 2, 3
and 5 min of deposition time (refer earlier phase
images). (1) Percentage coverage of substrate surface
increased as the deposition time increased from 1 to
2 min (Figs. 6D, 8A and 11), butdistinct phase change
indicated that some percentage of silicon surface was
still uncovered. The substrate peak in XPS (Si 2p) was
attenuating in accordance with this coverage. (2) No
phase change (constant phase) in AFM image of
silicon coated for 3 min (Figs. 8B and 11) indicated
complete coverage of silicon surface by fluorocarbon
coating. Presence of substrate peak in XPS for this
case indicated that the thickness of deposited layer
was lower or comparable to the escape depth of the
substrate electrons (l = 2.5 nm for silicon). (3) Phase
image of silicon surface coated with 5 minutes
(Figs. 9D and 11), showed surface coverage by more
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–5686 5683
Fig. 11. Attenuation of silicon substrate signal as the coating time
increases from 1 to 5 min. Presence of weak XPS substrate peak for
5 min fluorocarbon coated surface indicates that the coating thick-
ness is comparable to the escape depth of the underlying substrate
electrons (l = 2.5 nm for silicon).
Fig. 10. Fluorocarbon film growth on sapphire surface as a function of coating time.
than one thin layer of fluorocarbons. Aweak substrate
peak presence in XPS indicated the average film
thickness is still comparable to the escape depth
(within 2–3l range, i.e. few nanometers).
It can be seen from the above discussion that the
fluorocarbon film, even when less than 3 nm in
thickness (3 min of coating time), is still completely
covering the substrate, and growing in layer-by-layer
fashion. Fig. 12 shows qualitative comparison of the
XPS survey scans obtained on the model surfaces
treated with fluorocarbon for few minutes in micro-
wave plasma and commercial Teflon. It can be seen
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865684
Fig. 12. XPS survey scans obtained on model surfaces treated for few minutes with fluorocarbon coating show identical chemistry as that of
commercial Teflon surface.
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–56865685
Fig. 13. Contact angle vs. deposition time for one substrate.
Fig. 14. Contact angle vs. F/C ratio of a variety of substrates.
Fig. 15. Effective surface modification of micro-cellular carbon foam surface with nanoscale plasma-assisted fluorocarbon film. (A) SEM and
TEM image of a micro-cellular carbon foam indicating complex surface; (B) water contact angle (77) on untreated carbon foam surface; (C)
water contact angle on flurocarbon treated carbon foam surface indicating hydrophobicity.
that all surfaces look identical in chemical composi-
tion, indicating that deposited fluorocarbon film was
similar in composition to commercial Teflon, but it
was very thin and strongly bonded to the substrate
Fig. 13 shows the variation of contact angle with
the coating time on hydrophilic materials. As
indicated earlier, these include flat surfaces as well
as multilayered stack of porous materials. The contact
angle value (measure of hydrophobicity in this
particular case) increased very fast initially and then
reached a steady state after few minutes of deposition.
In XPS, an increase of about 20% in the surface
fluorine to underlying carbon ratio was found
(Fig. 14), at which contact angle value reached its
saturation. This supports our argument that only few
effective surface modification to any surface.
Fig. 15A shows the microstructure of the micro-
cellular porous carbon foam material. Fig. 15 B and C
show, the water contact angle measured on the surface
of micro-cellular porous foam before and after the
fluorocarbon film deposition. It was observed that
untreated foam had water contact around 778,
whereas, fluorocarbon coated foam had become
hydrophobic with 1408 water contact angle value.
These results are encouraging and indicate that
effective surface modification of complex shaped
nano-structured materials is possible.
Atomic level growth of plasma-assisted fluorocar-
bon films on various surfaces had been investigated. It
was shown that these films were growing in layer-by-
layer fashion and completely covered the underlying
surface even at nano-scale. Combination of XPS,
AFM and water contact angle results showed that even
a few nanometer thin film of fluorocarbon film could
provide effective surface modification to the surface.
useful in surface modification of complex and uneven
shaped nanostructured solids such as nano fibers, and
near net shape cellular foam structures.
 K.L. Mittal, A. Pizzi, Adhesion Promotion Techniques: Tech-
nological Applications, M. Dekker, New York, 1999.
 S. Gaur, G. Vergason, Society of vacuum coaters, in:
Proceedings of the 43rd Annual Technical Conference, vol.
 H. Grunwald, R. Adam, J. Bartella, M. Jung, Surf. Coat.
Technol. 111 (1999) 287–296.
 Aicha Elshabini-Riad, D. Fred, Barlow III, Thin Film Tech-
nology Handbook, McGraw Hill, 1997.
 T.S. Sudharshan, D.G. Bhat, Surface Modification Technolo-
gies II: Proceedings of Second International Conference on
Surface Modification Technologies, 1989.
 H. Yasuda, Plasma Polymerization, Academic press Inc., 1985.
 N. Inagaki, Plasma Surface Modification and Plasma
Polymerization, Technomic publishing Inc., 1996.
 H. Yasuda, T. Hirotsu, J. Polym. Sci., Poly. Chem. Ed. 16
 S.M. Mukhopadhyay, P. Joshi, S. Datta, J. Macdaniel, Appl.
Surf. Sci. 201 (2002) 219–226.
 S.M.Mukhopadhyay, P. Joshi,S. Datta, J.G.Zhao,P.France, J.
Phys. D: Appl. Phys. 35 (2002) 1927–1933.
 S.R. Coulson, I.S. Woodward, J.P.S. Badyal, Chem. Mater. 12
 S.R. Coulson, I.S. Woodward, J.P.S. Badyal, S.A. Brewer, C.
Willis, J. Phys. Chem. B 104 (2000) 8836–8840.
 S.R. Coulson, I.S. Woodward, J.P.S. Badyal, S.A. Brewer, C.
Willis, Langmuir 16 (2000) 6287–6293.
 Pratik Joshi, V. Rajasekhar, Pulikollu, M. Sharmila, Mukho-
padhyay, Surfaces and interfaces in nanostructured materials,
in: TMS Annual Meeting Proceedings, 2004, pp. 153–162.
 J.Klett, R. Hardy, E.Romine,C. Walls,T. Burchell, Carbon38
 K.M. Kearns, US Patents No. 5,868,974, Process of Preparing
Pitch Foams, February (1999).
 Sharmila M. Mukhopadhyay, Rajasekhar V. Pulikollu, Erik
Ripberger, Ajit K. Roy, J.Appl. Phys. 93 (2) (2003) 878–882.
 S.M. Mukhopadhyay, N. Mahadev, P. Joshi, A.K. Roy, K.
Kearns, D. Anderson, Structural investigation of graphitic
foam, J. Appl. Phys. 91 (2002) 5.
 Feldman Leonard, W. Mayer James, Fundamentals of Surface
and Thin Film Analysis, North-Holland, 1986.
 C.D. Wagner, W.M. Riggs, L.E. Davis, J.F. Moulder, Hand-
book of X-ray Photoelectron Spectroscopy,
 M.P. Seah, W.A. Dench, Surf. Interf. Anal. 1 (1979) 2.
 S. Tauma, C.J. Powell, D.R. Penn, Surf. Interf. Anal. 21 (1994)
Hoffmann, Mikrochim. Acta 133 (2000) 331–336.
 I. Schmitz, M. Schreiner, G. Friedbacher, M. Grasserbauer,
App. Surf. Sci. 115 (1997) 190–198.
P.P. Joshi et al./Applied Surface Science 252 (2006) 5676–5686 5686