вход по аккаунту



код для вставкиСкачать
Received: 4 July 2017
Revised: 23 August 2017
DOI: 10.1002/ppap.201700132
Accepted: 18 September 2017
Energetics of reactions in a dielectric barrier discharge with
argon carrier gas: VI PEG-like coatings
Bernard Nisol1
Sophie Lerouge3
| Sean Watson1 | Anne Meunier2 | David Juncker2 |
| Michael R. Wertheimer1
1 Groupe
des Couches Minces (GCM) and
Department of Engineering Physics,
Polytechnique Montréal, Box 6079, Station
Centre-Ville, Québec H3C 3A7, Canada
2 Micro and Nanobioengineering
Laboratory, Biomedical Engineering
Department, McGill University, Montréal,
Québec, Canada
3 Research Centre, Centre Hospitalier de
l’Université de Montréal (CRCHUM), and
Department of Mechanical Engineering,
École de technologie supérieure (ÉTS),
Montréal, Québec, Canada
Michael R. Wertheimer, Groupe des
Couches Minces (GCM) and Department of
Engineering Physics, Polytechnique
Montréal, Box 6079, Station Centre-Ville,
Montreal, Québec, H3C 3A7, Canada.
Funding information
the Natural Sciences and Engineering
Research Council of Canada (NSERC); the
Fonds de recherche du Québec – Nature et
technologies (FRQNT)
We have studied “PEG-like” plasma-deposited coatings of poly(ethylene glycol),
some of which prevent protein adsorption and cellular adhesion. This enables
inhibition of possible inflammatory reactions or rejection of an implant following
its insertion into living tissue. Our approach, based on electrical measurements in
atmospheric pressure Ar dielectric barrier discharges, enables precise measurements
of Em , the energy absorbed per monomer molecule. Here, we demonstrate
the importance of Em in preparing PEG-like coatings for biomedical
applications, for example by highlighting the great importance of molecular
weight of monoglyme (1G) or
diglyme (2G) monomers, and
by obtaining anti-fouling layers,
“PP-2G,” only with the diglyme. We demonstrate resistance to protein adsorption and
cell adhesion of PP-2G surfaces
prepared with optimized Fd
(and Em ) values.
anti-fouling, argon carrier, atmospheric pressure, dielectric barrier discharge, energetics,
polymerization, mono- and diglyme
Atmospheric-pressure (AP) cold plasma-enhanced chemical vapor deposition (PECVD) of thin coatings destined
for biomedical applications is a vast, ever-growing
field.[1–4] One of the main reasons is that AP plasmas,
like their low-pressure (LP) counterparts,[4] enable fine
surface engineering but also promise more economical,
easier implementation by obviating the need for costly
vacuum installations.[5–7]
Plasma Process Polym. 2017;e1700132.
This work focuses on anti-fouling poly(ethylene glycol)like (PEG-like) coatings, known to inhibit protein adsorption
thanks notably to strong water-PEG interactions[8]; as a result,
they have the potential to prevent adhesion of cells and
bacteria, inflammatory reactions and other undesired “walloff” reactions from the biological environment (eg, collagenous encapsulating response).[9] These particular interactions
with water render conventional PEG—also referred to as
poly(ethylene oxide) (PEO)—completely soluble in water, in
all proportions and for all degrees of polymerization at room
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1 of 10
2 of 10
temperature.[8,10] As a result, PEG cannot be used as a single
material for manufacturing medical devices. Various solutions have been envisaged, all with the aim of introducing
PEG chains either into (a) PEG-containing copolymers,[11,12]
or onto the surface of another material by way of (b)
physisorption,[13,14] or (c) covalent grafting.[15,16] Materials
and surfaces produced in this way nevertheless have certain
drawbacks, namely low strength for (a)[9,17]; for the case of
(b), low durability and high PEG concentrations are hard to
attain,[9] while (c) requires multi-step, complex, and costly
Plasma polymerization (PP) is a sub-field of PECVD
where the precursor gas or vapor is an organic compound; it is
an elegant solution that allows one to deposit thin organic
layers, for example PEG-like ones, with relatively high
degrees of cross-linking and good adhesion to most
substrates. An important additional benefit of PP is the
ability to work with organic precursors that CANNOT
polymerize by conventional chemical routes, such as the fully
saturated compounds used in this present study
(precursor ≠ monomer).
There already exists a body of literature relating to LP
plasma deposition of PP-PEG coatings, some of the most
important being references.[18–22] In the most relevant
example, published in 2005, Johnston et al[22] reported a
systematic LP PP study using various saturated PEG-like
precursors (including oligoglymes, dioxane, and crown
ethers). The resulting PP-PEG coatings were compared in
terms of concentration (%) of “ethylene glycol (EG) units,”
something hereafter referred to as “PEG character.” Those
authors observed that the highest PEG character was achieved
using tetra(ethylene glycol) dimethyl ether (tetraglyme, 4G),
in comparison with the lighter analogous oligoglymes,
namely monoglyme (1G), diglyme (2G), and triglyme
(3G); the authors concluded that increasing the number of
ethylene glycol units in the precursor significantly decreased
the chances for them to be fragmented in the discharge.
Breakage of a C─O bond obviously results in disappearance
of one EG building block from the initial molecule in favor of
other species contributing to film growth, thus leading to a
decrease in the PP's PEG character. In the case of multiple EG
building blocks, the probability of conserving PEG character
is therefore obviously higher. Also, it has been demonstrated
that the non-fouling properties of PP-PEG strongly depended
on maximizing the PEG character.[22,23] Following this logic,
4G was selected in the very first studies reporting successful
AP plasma deposition of PP-PEG coatings.[10,24] However,
due to the low vapor pressure of 4G, entraining a sufficient
amount of its vapor requires heating (up to 100°C), including
the gas lines and reactor walls. Despite such heating,
condensation on conduits, and reactor walls can occur within
seconds after initiating the process, which can detrimentally
affect its control and reproducibility.
Such experimental complications led several research
groups, including the present authors, to prefer lighter
precursors. For example, vinyl-containing compounds such
as tri(ethylene glycol) divinyl ether,[25] or di(ethylene glycol)
vinyl ether[26,27] were found to yield quite high PEG character
in resulting PP. One of the above-cited light oligoglymes,
namely diglyme (2G) was used by Bhatt et al[28] in AP PP
experiments with an argon plasma jet; they reported partiallyreduced cell adhesion (human ovarian carcinoma cell line
[NIH:OVCAR-3]) on their coatings; those coatings possessed
a PEG character of ca. 56%, below the accepted anti-fouling
threshold of ca. 65% for plasma polymers,[24,27] to be clarified
further below.
Some of the present authors have recently shown that
dielectric barrier discharge (DBD) plasmas with Ar as carrier
gas can be used to deposit PP coatings from a wide variety of
monomers, with precise knowledge of the energy per
precursor molecule, Em , absorbed from the plasma's “energy
reservoir.”[29–32] In recent joint communications with D.
Hegemann,[33,34] this approach was even shown to permit
comparisons between AP and LP PP processes, based on a
new “energy conversion efficiency” (ECE) parameter
reminiscent of the “Yasuda parameter—W/FM” concept
from the 1970s and 80s.[35]
In this present research, we adopted our above-cited AP
DBD methodology as a routine tool for depositing PP-PEG
using low molecular weight monoglyme (1G) and diglyme
(2G) as precursors (see Figure 1), and we henceforth refer to
the resulting coatings as PP-1G and PP-2G, respectively. Fine
tuning of Em once again enabled unprecedented process
control, the goal being to achieve anti-fouling properties by
maximizing the coatings’ PEG character.
2.1 | Plasma polymerization process and
material characterization
The DBD plasma reactor consists of two top electrodes
(6 × 18 cm2) and a single large bottom electrode separated by
a 2 mm gap between top and bottom dielectrics, respectively
Low molecular weight glymes, used as PP-PEG
precursors in this study
Macor® ceramic (3.50 ± 0.05 mm thick) and glass
(3.00 ± 0.02 mm) plates, for a total plasma volume of
v = 43.2 cm3. For more details about the complete plasma
reactor system the reader is kindly invited to consult previous
All PP experiments were carried out in cold (T < 35 °C)
AP DBD plasma sustained by audio-frequency power at
constant frequency and applied voltage, f = 20 kHz,
V a ðf Þ = 2.8 kVrms (=8 kVpp, peak-to-peak). Argon (Ar)
was used as the inert carrier gas (99.9 + % purity, Air
Liquide Canada, Ltd., Montreal), controlled by a rotametertype flowmeter (Matheson, model 7642H, tube 605). Two
liquid reagents, 1,2-dimethoxyethane (1G, Sigma-Aldrich,
purity > 99.5%) and di(ethylene glycol) dimethyl ether (2G,
Sigma-Aldrich, purity > 99%) were the selected precursor
molecules. Coating homogeneity was ensured by operating
the AP DBD reactor in “dynamic” mode, the moveable
grounded electrode (+dielectric) platen, on which were
fixed the substrates being displaced back-and-forth at a
15 mm s−1 scan rate.
A glass bubbler (∅ = 30 mm, height = 70 mm) contained
several cm3 of either 1G or 2G. Vapor flows, Fd , in standard
cm3 per minute (sccm), corresponding to ‰ concentrations in
the 10 slm Ar carrier gas stream, were entrained by a smaller,
separately metered parallel flow of Ar through the bubbler. In
each such case, a calibration was first carried out by
measuring steady-state rate of mass change of monomer in
the bubbler with time. The temperatures of the bubbler and
conduits were appropriately controlled: 1G was cooled to
0 °C (ice water bath), while 2G was thermalized at
60 °C (warm water bath), both at standard (atmospheric)
Without repeating details presented earlier,[29,36] measurements of the absorbed energy difference, ΔEg , and the
resulting Em value (in eV/molecule)—which plays a key role
—were carried out. This was possible by using an equivalent
circuit model along with the MATLAB® program for
computing Eg , energy dissipated in the AP plasma during
each applied high-voltage cycle.[29,36] The uncertainty of Em
is dictated by that of Fd, being lowest (<5%) for larger Fd
(≥1 sccm) and highest (up to 15%) for the very smallest Fd
(<1 sccm); data points on the graphs are an average of at least
three separate experiments.
Techniques used for characterizing the PP-PEG coatings
were the following:
1. Spectroscopic ellipsometry (SE) measurements, to evaluate film thickness and deposition rate, were carried out on
coated fragments of single-crystal silicon (c-Si) wafers.
The parameters Ψ and Δ were determined at 45°, 55°, 65°,
and 75° incidence angles using a J. A. Woollam RC2®
instrument operating in the 193-1700 nm wavelength
range, and they were interpreted using CompleteEASE™
3 of 10
software with a Gaussian oscillator-based optical
2. Attenuated total reflectance (ATR) Fourier-transform
infrared (FTIR) spectroscopy measurements were
performed using a Digilab® FTS 7000 spectrometer,
equipped with a UMA 600 microscope and a hemispherical Ge ATR crystal, for operation in the region from 4000
to 700 cm−1. Spectra with a resolution of 4 cm−1, 128 coadded measurements, were acquired for PP-PEG films
deposited on KBr (99 + %, Fisher Scientific IR grade) disc
substrates of ca. 0.5 mm thickness and 12.9 mm diameter.
Deposits of PP films on KBr were d ≥ 200 nm thick.
3. X-ray photoelectron spectroscopy (XPS) analyses were
carried out in a VG ESCALAB 3 MKII spectrometer.
Spectra were acquired using a Mg anode (1253.6 eV)
operating at 300 W. The pass energy was set to 100 eV for
survey spectra (not presented here), and 20 eV for highresolution (HR) C 1s peak shape analysis. The HR C 1s peak
fitting was performed using CasaXPS (CasaSoftware Ltd.),
by considering the lowest number of physically meaningful
components; for the sake of clarity, these were restricted to
the “first neighbor”-induced chemical shift (C─C/C─H,
C─O─R, CO/O─C─O, COOR [R = alkyl or H]).[38] The
FWHM (20 eV pass energy) of these components was
constrained to 1.7 ± 0.1 eV, and binding energies were
charge-referenced by setting the hydrocarbon (C─C/C─H)
component to 285.0 eV.
2.2 | Evaluation of anti-fouling properties
All solutions used for protein adsorption and cell adhesion
assays were prepared with ultrapure water from a Millipore
Milli-Q system (resistivity: 18 MΩ cm) and passed through a
0.2 µm filter. Phosphate buffered saline (PBS 1X, pH = 7.4)
was procured from Fisher Scientific and contained
11.9 × 10−3, 137.0 × 10−3, and 2.7 × 10−3 mol L−1 of phosphates, NaCl and KCl, respectively. Trypsin-EDTA and
fluorescently labeled bovine serum albumin-fluorescein
isothiocyanate (BSA-FITC) were obtained from SigmaAldrich. Triton X–100 and paraformaldehyde (PFA) were
purchased from Fisher Scientific. Dulbecco's modified Eagle
medium (DMEM), fetal bovine serum (FBS) and 4′,6diamidino-2-phenylindole (DAPI) were purchased from Life
Technologies. Antibiotics (penicillin/streptomycin) were
obtained from Invitrogen.
For protein adsorption assays, substrates consisted of high
density polyethylene (HDPE) disks (∅ = 15 mm, thickness = 0.85 mm). Cell adhesion assays were conducted on cSi substrates. In both types of assays, samples were half-masked
during the PP process to obtain a clear, well-defined border
between the PP-PEG coating and the uncoated substrate. All
deposits of PP films on HDPE and c-Si were ca. 200 nm thick.
4 of 10
2.2.1 | Protein adsorption assays
All protein adsorption experiments were performed at room
temperature. Samples were immersed in PBS 1X for 15 min, then
the PBS solution was removed and the samples were incubated
for 120 min in BSA-FITC solution (1.0 mg mL−1). After
incubation, samples were rinsed twice, first with PBS then
with ultrapure water, and finally dried under a dry nitrogen stream.
2.2.2 | Cell adhesion assays
Normal human IMR-90 lung fibroblasts, stably expressing
the fluorescent marker mCherry and kindly provided by
TABLE 1 Identification and numerical values (Fd [in sccm] and Em
[in eV]) of the deposition conditions for all characterized PP-1G and
PP-2G samples
PP sample
Fd (sccm)
Em (eV)
Dr. M. Park (McGill University, Montreal, Canada), were
cultured in DMEM supplemented with 10% FBS and 1%
(v/v) antibiotics, and maintained in 5% CO2 at 37 °C in
25 cm2 flasks (Corning, NY, USA). For all samples, a
poly(dimethylsiloxane) (PDMS) well (I.D. ≈ 5 mm) was
centered on the coating/substrate border, delimiting a ca.
20 mm2 area, half-coated with PP-PEG. Once (80-90%)
confluent, cells were harvested from flasks using trypsin.
Cells in suspension were rinsed by dilution in PBS, then
centrifuged at 4600 rpm for 5 min and finally re-suspended
in culture medium. About 5 × 105 cells mL−1 were seeded in
each well and incubated in culture medium overnight. Cells
were gently rinsed by replacing culture medium with PBS,
then PBS was removed and cells were fixed with PFA (4%)
for 10 min, rinsed again, twice with PBS for 5 min, and
permeabilized with Triton X–100 (0.2%) for 5 min. Cells
were rinsed twice with PBS (5 min) and their nucleus was
stained with DAPI (0.1 µg mL−1) for 5 min. Samples were
rinsed one last time with PBS (5 min) and imaged
immediately thereafter.
FIGURE 2 Plots of (a) Em (average energy absorbed per
molecule in eV, see text) versus precursor vapor flow, Fd . Labels on
the plots identify the conditions selected for this study (see text and
Table 1); (b) Em versus 1=Fd , for monoglyme (1G, blue diamonds)
and diglyme (2G, red triangles). Regions (i), (ii), and (iii) are
identified for 1G (see text). The inset presents numerical values of
slopes (in watts), ðEm Þmax (in eV), and ðFd Þcrit (in sccm) (see text).
The uncertainty of Em is dictated by that of Fd, being lowest (<5%)
for larger Fd (≥1 sccm) and highest (up to 15%) for the very smallest
Fd (<1 sccm); data points are an average of at least three separate
Deposition rate, r, for 1G (blue diamonds) and 2G
(red triangles), plotted as a function of precursor flow rate, Fd
5 of 10
ΔI fluo ¼ ½I PPPEGþBSA ½I PPPEG Cell adhesion was estimated by
qualitatively comparing the cell surface
coverage on c-Si and on PP-PEG
regions. IMR-90 that already expressed
mCherry (red fluorescence) were further
stained with DAPI (blue fluorescence,
nucleus). Multi-layering of cells in some
regions disabled quantitative analysis.
Auto-fluorescence from PP-PEG coatings (green) was very low, and high
exposure times (ca. 4 s) were used to
localize the c-Si/PP-PEG border regions.
ATR FTIR spectra of (a) liquid 1G precursor (top) and PP-1G deposits; and
(b) 2G (top) and PP-2G, all normalized with respect to the area of the CH2/CH3 stretching
band. Labels to the left identify the PP conditions (see Table 1); numbers in brackets: Em
(in eV). Also shown are spectral assignments of the dominant absorption bands
2.2.3 | Fluorescence microscopy
After protein adsorption and cell adhesion assays, samples were
placed upside down on the platform of an inverted microscope
(TE-2000-E, Nikon) connected to a CCD camera (QuantEM
512SC, Photometrics); fluorescence images were recorded with
NIS-Elements Advanced Research software (Nikon) and
analyzed with ImageJ software.[39] Images were collected using
a mercury arc lamp, and 41001 (blue, for FITC), 41004 (green,
for mCherry), and 31000v2 (UV for DAPI) filter cubes (Chroma
Technology Corp.) were used.
Both HDPE and PP-PEG exhibited small auto-fluorescence signal intensities. For protein assays, all images
presented in this work were normalized to a unique color
scale (800-13 500 a.u., from blue to yellow); the low autofluorescence from HDPE (ca. 2% of the color scale) and
PP-PEG regions (ca. 10% of the color scale) did not prevent
qualitative estimation of protein adsorption by direct
comparison of the fluorescence intensity of each region
before and after immersion in BSA-FITC. For quantitative
analyses, HDPE and PP-PEG regions were imaged before and
after BSA-FITC incubation and auto-fluorescence intensities
from both regions were subtracted, respectively. Each image
was acquired with a 10× objective, a gain of one and an
exposure time of 1 s. The mean fluorescence intensity before
(I HDPE or I PPPEG ) and after (I HDPEþBSA or I PPPEGþBSA )
incubation was determined by averaging on five (5) areas and
two (2) replicated experiments. The variation of fluorescence
intensity due to BSA-FITC adsorption (ΔI fluo ) was thus
obtained as follows:
3.1 | Energy measurements
Figure 2 shows plots of Em , (a) versus Fd , the flows of 1G and
2G vapors (in standard cm3 per minute, sccm); (b) versus
1/Fd . For both compounds, three distinct regimes can be noted
(see Figure 2b):
(i) a very narrow region, corresponding to the lowest Fd
values, reveals a sharp rise in Em as Fd is increased and is
referred to as “monomer-lean.”[40] The upper limit of this
region is bounded by a first peak, defined as ðEm Þmax
[Fd = ðFd Þcrit ], which corresponds to the particular
condition where maximum transfer of energy from the
plasma resulted in presumed total fragmentation,
breakage of all covalent bonds in the molecule. Here,
ðEm Þmax values for 1G and 2G were 61 eV and 76.1 eV,
respectively. In agreement with previous work,[32] the
heavier molecule exhibits a significantly higher value of
ðEm Þmax ;
(ii) a “monomer-rich” region spreading over high Fd values,
where Em smoothly decreases with rising Fd .[40]
Figure 2b also shows plots of Em , this time versus
1/Fd , and it reveals quasi-linear behavior in the
monomer-rich region, something that we had already
amply discussed in our earlier reports.[31,32,40] The initial
slopes for 1G and 2G were 8.8 W and 11.1 W,
respectively. These are closely related to the propensity
of a precursor molecule to absorb energy available from
the plasma; a higher value in the case of 2G, heavier than
1G but from the same chemical family (here, ethers), is
also in agreement with previous observations[32];
(iii) a “dip region” separating (i) and (ii) is a transition feature
that was already reported (and interpreted) in our earlier
article dealing with a series of esters.[31] It is suggested
6 of 10
that this region, neighboring ðEm Þmax and in which Em
drops drastically, results from a marked fragmentation of
oxygen-bearing hydrocarbon groups, dominated by
liberation of CO2.
remains low for PP-1G, reaching an asymptotic limit near
8 nm min−1. The presence of a “plateau” in the r(Fd) curve
could be explained by the following: in order to contribute to
film growth, a given precursor molecule must be activated in
the discharge; here, activation via at least one bond breakage
taught us that the upper limit of
Earlier experience
(radical or ion formation) is required because the present
region (iii) (lower limit of [ii], see Figure 2) corresponds to the
precursors, 1G and 2G, are fully saturated. As one increases
one of interest for most PP experiments. For both precursors a
Fd to reach the “deeper” monomer-rich region (Fd > 5 sccm in
the 1G case), probability for the molecules to pass through the
few such conditions were therefore selected, with Em values
ranging upward from ca. 20 eV. These are identified by
discharge without being activated also increases. Since it was
suitable labels in Figure 2a and in Table 1; they once again
felt to stray from the main objectives of this study, we did not
underline the advantages of this approach as a routine tool that
investigate whether or not the plot for 2G would level off, as
enables one to pinpoint a “sweet spot” in the PP process,
one might expect. At Fd = 7 sccm, r values of the two differ
by a factor of ca. 4, signaling particular fragility of the 1G
before any material characterization is undertaken.
molecules in plasma. Deposition rates are seen to be low for
both 1G and 2G close to region (iii) (Fd ∼2 sccm, see Figure
3.2 | Characterization of PP-PEG coatings
2a), namely r < 5 and r < 10 nm min−1, respectively.
Figure 3, a plot of PP deposition rate, r, versus Fd , reveals two
It is noteworthy that r ∼20–30 nm min−1 values observed
distinct behaviors for PP-1G and −2G. While PP-2G shows a
for PP-2G are adequate for many bio-technological applications,
pseudo-linear increase from about r = 9 to ca. 32 nm min−1, r
because thicknesses in the 20-100 nm range usually suffice.
Figure 4 shows ATR FTIR spectra of
(a) PP-1G; and (b) PP-2G, all being
normalized to the area of the CH2/CH3
stretching (str.) band (2800-3000 cm−1).
The upper spectra in (a) and (b)
respectively correspond to the liquid
1G and 2G precursors. They allow one to
identify the characteristic absorption
bands of glymes, and to note that these
are preserved in PP-PEG; they are: C-O
str. (1100-1200 cm−1), CH2/CH3 str.
(2800-3000 cm−1) and the multiple
C─H bending modes in the 12001500 cm−1 range.[10]
Clearly, for PP-1G the C─O str.
band intensity is quite low, even under
“mild” deposition conditions (1G-7,
Em = 18.9 eV), and is much decreased
when Em is raised. This is a characteristic of intense precursor fragmentation,
responsible for decrease in “ethylene
glycol” and increase in “hydrocarbon”
content in the coating.
For PP-2G coatings (Figure 4b), high
relative intensity of the C─O str. band
signals fair conservation of the “ethylene
glycol” content, in particular for 2G-7
(Em = 24.5 eV) and 2G-5 (33.9 eV), for
which precursor fragmentation was
quite low. However, under the harshest
condition, 2G-2 (43.3 eV), carbonyl
FIGURE 5 High-resolution XPS C 1s spectra of PP-1G (left, blue); and PP-2G (right, red),
(CO str. near 1715 cm−1) and hydroxyl
and corresponding peak fittings. Labels to the left identify the PP conditions (see Table 1);
numbers in brackets are corresponding Em values (in eV)
(OH str., 3200-3500 cm−1) groups gain
7 of 10
TABLE 2 XPS C 1s peak fitting values for PP-1G and PP-2G samples
PP sample
74.7 ± 2.1
19.0 ± 1.6
5.3 ± 0.4
1.0 ± 0.1
69.1 ± 0.1
23.2 ± 0.7
6.0 ± 0.7
1.7 ± 0.4
64.3 ± 1.2
27.1 ± 0.3
6.8 ± 0.6
1.8 ± 0.4
63.5 ± 0.3
28.9 ± 0.3
6.6 ± 0.2
1.0 ± 0.2
56 ± 0.1
31.0 ± 0.1
10.8 ± 0.2
2.2 ± 0.1
39.5 ± 0.9
45.8 ± 0.8
11.5 ± 0.3
3.2 ± 0.4
30.0 ± 0.3
53.6 ± 0.1
12.7 ± 0.1
3.7 ± 0.2
26.2 ± 0.1
59.3 ± 0.4
11.3 ± 0.1
3.2 ± 0.6
significance. Surprisingly, these were minimally present in the
case of PP-1G, confirming major differences in the plasmainduced behaviors of the two precursors. It is noteworthy that the
hydroxyl peaking at around 3350 cm−1 remains quite low for all
conditions except 2G-2; consequently, we assign minimal
importance to that condition. For comparison, an FTIR spectrum
of poly(vinyl alcohol), PVA,[41] manifests a gigantic OH
stretching band, which dominates all other features, counter to
the situation observed here. Therefore, we feel comfortable
stating that OH groups play a relatively minor role here, and that
XPS data hereunder are reliable in their interpretation.
Figure 5 shows the XPS C 1s high resolution spectra of
PP-1G (left, blue); and PP-2G (right, red), with corresponding
peak fittings. The calculated area ratios of the fitting
components are listed in Table 2. We remind the reader
that an “ideal” PEG XPS C 1s peak (100% PEG character)
would solely comprise the C─OR component; % C─OR is
thus considered a semi-quantitative marker of the PEG
character, one that may include in this case some small C─OH
In agreement with the corresponding ATR-FTIR
spectra, XPS data confirm that low PEG character
(C─OR ≤ 29%) is associated with high hydrocarbon content
(C─C/C─H = 60-70%) for all of the PP-1G deposition
conditions investigated.
Now, focusing on PP-2G, the PEG character is observed
to have varied significantly, between ca. 31% for the highest
Em = 43.4 eV (2G-2), in region (iii), to the more encouraging
value of ca. 59% for the case of 2G-7 (24.5 eV). Interestingly,
even at high Em , for example 2G-2, diglyme yielded coatings
with higher PEG character than any of those obtained with
monoglyme; this agrees well with ATR-FTIR and deposition
rate data, and it confirms the earlier-mentioned extreme
fragility of 1G in the PP process.
Fluorescence micrographs of partly coated (PP-1G and PP-2G on bottom half) HDPE substrates, after BSA-FITC treatment. All
of the presented images focus in the same orientation, centered on the coating/substrate border. Labels at the top/bottom identify PP conditions
(see Table 1)
8 of 10
On hand of the PP characterization data presented in
preceding sections, we rejected conditions in region (iii) from
those selected for preparing samples evaluated by biological
3.3 | Anti-fouling properties versus PEG
Fluorescence microscopy was used to evaluate anti-fouling
properties of selected PP-1G and PP-2G samples. First, protein
adsorption tests were conducted using fluorescently labeled
bovine serum albumin (BSA-FITC). As shown in previous
work, the fluorescence intensity, I fluo , can be used as a quasidirect indicator of protein adsorption[25]: Figure 6 shows
fluorescence micrographs of partly coated (coating on the
bottom-half) polyethylene (HDPE) substrates after BSA-FITC
treatment. For clarity, emphasis is on the coating/substrate
border region, so as to better highlight differences in the
fluorescence intensity. Qualitatively, one clearly notes two
extremes: 1G-2.3 (low PEG character ≈ 23%) exhibits a
significant increase in I fluo in comparison with bare HDPE; in
the case of 2G-7 (PEG character ≈ 59%), I fluo is greatly reduced,
down to PP-PEG's auto-fluorescence value (see Section 2.2).
Quantitative data forΔI fluo , the difference in fluorescence
intensity before (auto-fluorescence) and after adsorption
tests, is shown in Figure 7a; it confirms the great difference
between PP-1G and PP-2G coatings. PP-1G samples (PEG
character ≤ 29%) adsorbed significant amounts of BSAFITC, comparable to—or even significantly higher
than—HDPE (twice as much in the case of 1G-2.3). All
PP-2G samples tested showed a marked decrease in ΔI fluo ,
near-zero under the “mild” 2G-7 (24.5 eV) deposition
In Figure 7b, ΔI fluo is plotted as a function of PEG
character expressed as XPS C 1s % C-OR. Based on BSA
adsorption assays, we note that a PEG character close to 60%
may suffice for anti-fouling, a value somewhat below the sofar accepted 65% threshold for plasma polymers.[24,27] The
current results are somewhat inferior to those of Nisol et al[10]
(ca. 83% using 4G aerosol; ca. 63% using 4G vapor), Da Ponte
et al[24] (65% using 4G aerosol), but comparable to Bhatt
et al[28] (56% using 2G vapor).
The anti-fouling character of PP-1G and PP-2G samples
was then further assessed with human IMR-90 lung fibroblast
adhesion assays: c-Si substrates, partly coated with PP-PEG
layers, were examined using fluorescence microscopy after
cell adhesion (24 h). All of the images presented in Figure 8
are taken in the same orientation, focused at the coating
(bottom-half)/substrate (top-half) border.
Except for 2G-7 (Figure 8c,d), on all investigated PP-1G
and PP-2G deposits—only two shown here—, both the
coating and substrate surfaces were covered with dense
mono-, or even multi-layers of cells, as seen in Figure 8a,b.
On the contrary, 2G-7 (24.5 eV, PEG char. ≈ 59 eV)
manifested strong anti-fouling character, because almost no
IMR-90 fibroblasts could be seen to adhere (see Figure 8c).
This strongly endorses the protein adsorption assay as a probe
for anti-fouling surface properties and also confirms the
necessity to maximize the coating's PEG character by way of
well-controlled PP processing. The present methodology also
allowed us to achieve a new (lower) minimum threshold value
of PPs’ PEG character, namely ≈ 59% compared with the
hitherto accepted 65% value.[24,27]
FIGURE 7 (a) Variation of fluorescence intensity (ΔI fluo ) after
BSA-FITC adsorption on HDPE substrates, PP-1G and PP-2G
coatings (labeled as in Table 1). The error bars represent the standard
error calculated from all samples of each type. (b) ΔI fluo correlated to
the PEG character, here expressed as % C─OR of the XPS C 1s highresolution spectrum
9 of 10
merits further testing and experimentation,
notably for use with blood and other biofluids that can lead to rapid surface fouling.
The authors are grateful for financial support
from the Natural Sciences and Engineering
Research Council of Canada (NSERC) and
from the Fonds de recherche du Québec—
Nature et technologies (FRQNT) via Plasma
Québec. We thank Yves Leblanc for skilled
technical support. DJ and AM acknowledge
funding from NSERC and CIHR (Canadian
Institutes of Health Research) as part of the
Collaborative Health Research Projects
(CHRP) program. SL wishes to thank the
Canada Research Chairs Program.
Fluorescence micrographs of partly coated (PP-1G and PP-2G, bottomhalves) c-Si substrates, after IMR-90 human lung fibroblast adhesion assays (24 h). All of
the presented images are taken in the same orientation, focused at the coating/substrate
border. (a, b, c) Merged fluorescence [red (cytoplasm); blue (nuclei)] micrographs reveal
the cells. (d) Same area as in (c), acquired in the green channel, reveals the coating (high
exposure time = 4 s). Labels on the top/bottom of the images identify the PP conditions
(see Table 1)
In this present work we have, once again, demonstrated that
methodology exemplified in Figure 2 lends excellent control
and reproducibility over the energetics of AP PECVD (PP)
experiments. They were used here to create PP-PEG coatings
from fully-saturated, low-MW monoglyme (1G), and
diglyme (2G) precursors.
For 2G, a narrow Em range in Figure 2, roughly between
25 and 45 eV/molecule, yielded PP-2G coatings with variable
chemical/biological properties. Low Em resulted in satisfactory retention of the precursor's PEG character (ca. 59%, for
sample 2G-7) and remarkable anti-fouling properties, a
condition that co-applies to both proteins and cells. Now,
unlike that ready ability of 2G to form useful PP coatings, this
was not the case with 1G as precursor: even within the wider
Em range, ca. 19-57 eV/molecule, it was impossible to obtain a
PP-1G deposit with anti-fouling properties, the mildest
condition (1G-7; 19.8 eV) resulting in PEG character <30%.
To conclude, it is believed that the current research
establishes PP-2G as a promising anti-fouling coating that
[1] G. Da Ponte, E. Sardella, F. Fanelli, R.
d’Agostino, P. Favia, Eur. Phys. J.—Appl.
Phys. 2011, 56, 24023 (4 pp).
[2] S. Bhatt, J. Pulpytel, F. Arefi-Khonsari, Surf.
Innov. 2015, 3, 63.
[3] K. S. Siow, L. Britcher, S. Kumar, H. J. Griesser,
Plasma Process. Polym. 2006, 3, 392.
K. S. Siow, S. Kumar, H. J. Griesser, Plasma Process. Polym.
2015, 12, 8.
D. Pappas, J. Vac. Sci. Technol. A 2011, 29, 020801 (17 pp).
F. Massines, C. Sarra-Bournet, F. Fanelli, N. Naudé, N. Gherardi,
Plasma Process. Polym. 2012, 9, 1041.
D. Merche, N. Vandencasteele, F. Reniers, Thin Solid Films 2012,
520, 4219.
J. H. Lee, H. B. Lee, J. D. Andrade, Prog. Polym. Sci. 1995, 20, 1043.
D. G. Castner, B. D. Ratner, Surf. Sci. 2002, 500, 28.
B. Nisol, C. Poleunis, P. Bertrand, F. Reniers, Plasma Process.
Polym. 2010, 7, 715.
D. K. Gilding, A. M. Reed, Polymer 1979, 20, 1454.
I. Vulić, T. Okano, S. W. Kim, J. Feijen, J. Polym. Sci. A Polym.
Chem. 1988, 26, 381.
W. Wasiewski, M. J. Fasco, B. M. Martin, T. C. Detwiler, J. W.
Fenton, Thromb. Res. 1976, 8, 881.
Y. L. Cheng, S. A. Darst, C. R. Robertson, J. Colloid Interface Sci.
1987, 118, 212.
N. P. Desai, J. A. Hubbell, J. Biomed. Mater. Res. 1991, 25, 829.
A. Kishida, K. Mishima, E. Corretge, H. Konishi, Y. Ikada,
Biomaterials 1992, 13, 113.
K. Fujimoto, H. Inoue, Y. Ikada, J. Biomed. Mater. Res. 1993, 27,
10 of 10
[18] M. N. Mar, B. D. Ratner, S. S. Yee, Sens. Actuator B-Chem 1999,
54, 125.
[19] F. Palumbo, P. Favia, M. Vulpio, R. d'Agostino, Plasmas and
Polym. 2001, 6, 163.
[20] M. Shen, M. S. Wagner, D. G. Castner, B. D. Ratner, T. A. Horbett,
Langmuir 2003, 19, 1692.
[21] Y. Wang, E. B. Somers, S. Manolache, F. S. Denes, A. C. L. Wong,
J. Food Sci. 2003, 68, 2772.
[22] E. E. Johnston, J. D. Bryers, B. D. Ratner, Langmuir 2005, 21, 870.
[23] B. Nisol, G. Oldenhove, N. Preyat, D. Monteyne, M. Moser, D.
Perez-Morga, F. Reniers, Surf. Coat. Technol. 2014, 252, 126.
[24] G. Da Ponte, E. Sardella, F. Fanelli, R. d'Agostino, R. Gristina, P.
Favia, Plasma Process. Polym. 2012, 9, 1176.
[25] B. Nisol, A. Meunier, C. Buess-Herman, F. Reniers, Plasma
Process. Polym. 2015, 12, 991.
[26] I. Gordeev, A. Choukourov, M. Šimek, V. Prukner, H. Biederman,
Plasma Process. Polym. 2012, 9, 782.
[27] I. Gordeev, M. Šimek, V. Prukner, A. Artemenko, J. Kousal, D.
Nikitin, A. Choukourov, H. Biederman, Plasma Process. Polym.
2016, 13, 823.
[28] S. Bhatt, J. Pulpytel, S. Mori, M. Mirshahi, F. Arefi-Khonsari,
Plasma Process. Polym. 2014, 11, 24.
[29] B. Nisol, H. Gagnon, S. Lerouge, M. R. Wertheimer, Plasma
Process. Polym. 2016, 13, 366.
[30] B. Nisol, S. Watson, S. Lerouge, M. R. Wertheimer, Plasma
Process. Polym. 2016, 13, 557.
[31] B. Nisol, S. Watson, S. Lerouge, M. R. Wertheimer, Plasma
Process. Polym. 2016, 13, 900.
[32] B. Nisol, S. Watson, S. Lerouge, M. R. Wertheimer, Plasma
Process. Polym. 2017, 14, e1600191 (9 pp.).
[33] D. Hegemann, B. Nisol, S. Watson, M. R. Wertheimer, Plasma
Process. Polym. 2016, 13, 834.
[34] D. Hegemann, B. Nisol, S. Watson, M. R. Wertheimer, Plasma
Chem. Plasma Process. 2017, 37, 257.
[35] H. K. Yasuda, T. Hirotsu, J. Polym. Sci., Polym. Chem. Ed. 1978,
16, 743.
[36] M. Archambault-Caron, H. Gagnon, B. Nisol, K. Piyakis, M. R.
Wertheimer, Plasma Sources Sci. Technol. 2015, 24, 045004
(16 pp).
[37] B. Nisol, S. Watson, S. Lerouge, M. R. Wertheimer, Plasma
Process. Polym. 2016, 13, 965.
[38] B. Nisol, F. Reniers, J. Electron Spectrosc. Relat. Phenom. 2015,
200, 311.
[39] C. A. Schneider, W. S. Rasband, K. W. Eliceiri, Nat. Methods
2012, 9, 671.
[40] S. Watson, B. Nisol, S. Lerouge, M. R. Wertheimer, Langmuir
2015, 31, 10125.
[41] E. F. d. Reis, F. S. Campos, A. P. Lage, R. C. Leite, L. G. Heneine,
W. L. Vasconcelos, Z. I. P. Lobato, H. S. Mansur, Mat. Res. 2006,
9, 185.
How to cite this article: Nisol B, Watson S,
Meunier A, Juncker D, Lerouge S, Wertheimer MR.
Energetics of reactions in a dielectric barrier
discharge with argon carrier gas: VI PEG-like
coatings. Plasma Process Polym. 2017;e1700132,
Без категории
Размер файла
5 191 Кб
201700132, ppap
Пожаловаться на содержимое документа