P

NU

a

AA

KXPSP

1

Sutps

ostwsipi

tpasfft

(

0d

Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408

Contents lists available at ScienceDirect

Journal of Electron Spectroscopy andRelated Phenomena

journa l homepage: www.e lsev ier .com/ locate /e lspec

lasma-modified polymer surfaces: Characterization using XPS

. Vandencasteele ∗, F. Reniers ∗

niversité Libre de Bruxelles, Faculty of Sciences, cp 255, Bld Triomphe, 2-B-1050 Bruxelles, Belgium

r t i c l e i n f o

rticle history:vailable online 13 January 2010

a b s t r a c t

Amongst all the available surface treatments, the plasma techniques have taken a major importance these

eywords:PSlasmaurfaces

last decades, both in fundamental studies and for industrial applications.This paper presents the utility of X-ray photoelectron spectroscopy for the characterization of plasma-

modified polymer surfaces. Due to the originality of the reactivity of the plasmas (generation of radicals),and to the surface-selective character of the plasma treatments, some specific uses of XPS are emphasized,such as peak fitting, derivatization, angle-resolved XPS. We also show several selected examples of typicalXPS analysis of plasma-treated polymers: ageing, biocompatibility, plasma polymerization, adhesion,

uncti

olymers

surface grafting of polar f

. Introduction

In the framework of this special issue of Journal of Electronpectroscopy and Related Phenomena, we detail in this review thetility of X-ray photoelectron spectroscopy for the characteriza-ion of plasma-modified polymer surfaces. This introduction willresent the general scope of plasmas, followed by their use for theurface modification of polymers.

The second section of the paper continues with the specific usesf XPS in this field. First the issue of surface charging and energycale calibration, crucial for most polymers, is described. Second,he XPS characterization using peak fitting (mostly of the C1s peak)ill be presented, as it is the major tool used, especially for the

urface modification of polymers. As in some cases, peak fittings not sufficient to identify the surface functionalities grafted bylasma, the derivatization technique will be presented. Finally, the

nterest of ARXPS for plasma treated polymers will be emphasized.After these topics, more specific to XPS, the paper will focus in its

hird section on some typical applications of XPS for surface-treatedolymers: ageing, plasma polymerization, adhesion, membranes,nd biocompatibility. Although the reader could have the feeling ofome redundancy, we think that this double entry approach (firstocusing on the analytical technique XPS, then on specific examplesor each topic more dedicated to the modified material) will provide

o the scientist the information he is looking for.

∗ Corresponding authors. Tel.: +32 2 6503116.E-mail addresses: nvdcaste@ulb.ac.be (N. Vandencasteele), freniers@ulb.ac.be

F. Reniers).

368-2048/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.elspec.2009.12.003

ons.© 2010 Elsevier B.V. All rights reserved.

2. Section 1: Plasmas

Plasmas, sometimes called the fourth state of matter, have beenknown for a very long time now. The term plasma itself is oftenattributed to Langmuir [1]. Atmospheric plasmas for the genera-tion of ozone are also used for a long time, as the Siemens processwas presented in 1857 [2]. Plasmas are often defined as a partly orfully ionized gas. The energy required for the ionization process canbe provided by various sources (heat, electromagnetic field, light,etc.). For the surface modification of polymers the energy is usuallycoming from an electric field. The electric field accelerates electronswhich collide with atoms or molecules and ionize them, producingnew charged particles (electrons, atomic or molecular ions) that arealso accelerated.

A plasma can be characterized by several parameters such as;its degree of ionization, the energy of the various species it contains(electrons, ions, neutrals and photons), the pressure, the mode ofgeneration of the discharge (AC, DC, RF, HF, etc.), or the density ofthe plasma (often expressed in electrons per cm3).

The degree of ionization (�) represents the ratio of ionizedatoms/molecules over the total amount of particles:

˛ = ne

ne + n0(1)

ne is the electron density (m−3); n0 is the neutral species density(m−3).

Depending on the value of ˛, two major kind of plasmas can bedefined; the weakly ionized one with ˛ values between 10−7 and

10−4 and the strongly ionized plasma where ˛ is close to 1. Plasmasused to modify polymers are usually weakly ionized.

In plasmas, each species have their own energy range. It is there-fore possible to define an electronic temperature (Te), an ionictemperature (Ti) and a neutral temperature (T0). In an equilibrium

troscopy and Related Phenomena 178–179 (2010) 394–408 395

pItmttaimota(mTpihtgtcadma

t

ctmnebds

iosoapafias

nttig

otopgeutsmt

ization (Fig. 2). In this case a precursor is sent into the plasma,the reactive species from the plasma will activate the precursormolecules that will react with each other and polymerize. A verylarge variety of polymers can be created by this technique rangingfrom diamond like carbon [12] to polystyrene [13] and PTFE [14]

N. Vandencasteele, F. Reniers / Journal of Electron Spec

lasma, all the particles have the same range of energy Te ≈ Ti ≈ T0.n non-equilibrium plasma, electrons have a much higher energyhan neutrals and ions Te � Ti ≈ T0. Plasmas used to modify poly-

ers are most of the times non-equilibrium plasma. The energy ofhe neutrals and ions is usually in the range of 300–1000 K, whereashe energy of the electrons can vary from 1000 to 100 000 K, withn average of a few eV (a few 10 000 K). Therefore, the gas phases referred as ‘cold’. This allows to modify the surface of poly-

ers samples without destroying them. The very high reactivityf cold plasma for the surface treatment of polymers originates inhe collisions in the gas phase between the high energy electronsnd the atoms/molecules that create very reactive excited specieswith a high potential energy) that are able to react with even the

ost unreactive samples such as polytetrafluoroethylene (PTFE, oreflon®). Several techniques are used to obtain non-equilibriumlasma. The easiest approach is to operate at low pressure, this min-

mizes the number of collisions between the light (electrons) andeavy (ions and neutrals) particles, and therefore no thermaliza-ion of the plasma can take place. One can also use a high frequencyenerator (kHz to GHz). At those high frequencies the mobility ofhe ions is strongly reduced because of their much higher massompare to the electrons (a proton is 1836 times heavier thann electron). Special plasma geometries, like the dielectric barrierischarge, allow to operate at high (atmospheric) pressure, whileinimizing the number of highly energetic collisions. This leads

lso to cold plasmas.More detailed information about plasma can be obtained from

he following books [3,4].Plasmas are very complex media full of highly reactive parti-

les (radicals, ions, electrons, UV photons), all of these reachinghe polymer surface. As a consequence, due to this complexity, the

echanisms involved in the surface modification of polymer areot yet fully understood. Also, because of the many “reactants”, thend result often contains a wide variety of “products”, which cane for instance different functionalities grafted. It is therefore veryifficult to have a full identification and quantification of all thepecies.

Low pressure and high pressure plasma have been used empir-cally for the treatment of surfaces for many decades, but it isnly since the 80s that the plasma treatment of surfaces reallytarted to gain importance. Fig. 1(a) and (b) shows the resultsf a SciFinder search using the Keywords: “plasma and surface”nd “glow discharge and surface” present in the titles of papersublished. Although the word “plasma” is dangerous (as it referslso to the blood component); a clear trend can be seen in bothgures. This is correlated with the development of commerciallyvailable surface analysis techniques; mostly X-ray photoelectronpectroscopy.

Plasma treatment of polymers results mainly in two mecha-isms: ablation of the polymer and/or grafting of new species onhe surface of the sample. The two reactions are taking place athe same time but depending on the experimental conditions ones more important than the other. The most commonly use is therafting of new functionalities on the sample surfaces.

There are major scientific and technological reasons for the usef plasma for the surface treatment of polymers. The most impor-ant one in terms of industrial applications is certainly the increasef the adhesion properties of the polymer [5–11]. Indeed, mostolymers, despite their very good properties (chemical inertness,ood weight to strength ratio, transparency, barrier properties,tc.), exhibit very poor adhesion properties. Plasma treatments are

sed to increase their surface energy by grafting polar species onheir surfaces. As plasmas modify only the surface, that can beolved, while maintaining unaltered the bulk properties of the poly-ers. Moreover, plasmas are environmentally friendly, contrary to

raditional wet chemistry, and they are fast.

Fig. 1. # of references containing both the concept: “plasma” and “surface” (A)(SciFinder, March 2009) or “glow discharge” and “surface” (B) (SciFinder, May 2009).

Another use of plasma for polymer science is plasma polymer-

Fig. 2. # of references containing the concept “plasma polymer” (SciFinder, May2009).

396 N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408

F ple. (�� surfac

fiadpaesfipo

sipWodthbii(

�

�fc

esa

W

Wt

at

years. Not only XPS instruments became easily commercially avail-able, but the improvement of the technology was also a significantfactor. The development of new detectors (multiple channeltronor channelplates) lead to a spectacular increase in the number ofcounts per s (cps), and therefore to a higher sensitivity and to a

Table 1Some example of techniques used to analyze polymer modified by plasma or plasmapolymers.

Technique Information Drawbacks

WCAa Surface energy, polarand dispersivecomponents

Possible modification of thesurface by water. No chemicalinformation. Mix of polarityand roughness

XPSb Composition,functionalities

Not totally surface specific;X-ray degradation; hydrogennot detected; limited energyresolution

(S)SIMSc Functionalities Fragmentation, complex dataanalysis

SEMd Topographicinformation

Surface must be conducting.No chemical information

AFMe Topographicinformation

No chemical information

ATR FTIRf Functionalities Not surface specific.

ig. 3. (a) Measurement of the contact angle of a liquid drop deposited on a solid samLV: liquid/vapour surface tension). Water droplets deposited on: (b) hydrophobic

lms. This technique allows to deposit polymers on a very largemount of substrates. Those plasma polymers (pp) are used forielectric coating (microelectronic), corrosion protection, biocom-atible surfaces, gas barrier (food packaging). They present severaldvantages over traditional polymers. For instance, these filmsxhibit usually a higher degree of crosslinking, and have thereforeuperior mechanical properties which allow to deposit ultra-thinlms that present good adhesion on various substrates. Plasmaolymer films also exhibit uniform surfaces with a full coveragef the substrate [15–17].

As plasmas modify only the top layers of the sample, thereforeurface specific techniques are required to analyze polymers mod-fied by plasma. Two of the main techniques used to characterizedlasma-modified polymers are XPS and water contact angle (WCA).CA is a “true” surface analysis technique, with a depth analysis

f around 0.5 nm [18]. During water contact angle measurements arop of water is deposited on the sample, then the angle betweenhe drop and the surface is measured (Fig. 3a). Hydrophilic surfacesave low WCA values as the water spread on the sample; hydropho-ic samples exhibit high WCA values as the drop tends to minimize

ts contact surface with the sample (Fig. 3b–d). The surface energys related to the cosine of the WCA by the Young–Laplace equationEq. (2)):

SV − �SL = �LV cos � (2)

SV is the solid/vapour surface tension; �SL is the solid/liquid sur-ace tension; �LV is the liquid/vapour surface tension; � is theontact angle.

Another useful value is the adhesion work, defined by the Dupréquation (Eq. (3)). It is defined as the reversible work required toeparate the interface from the equilibrium state of two phases ton infinite distance:

a = �L + �S − �SL (3)

a is the work of adhesion; �L is the liquid surface tension; �S ishe solid surface tension; �SL is the solid/liquid surface tension.

WCA is not only a relatively low cost (and therefore easily avail-ble) technique, but it also provides an immediate information onhe adhesion properties of the modified surface.

: contact angle, �SV: solid/vapour surface tension, �SL: solid/liquid surface tension,e, (c) hydrophilic surface and (d) ultra-hydrophobic surface.

Other surface analysis techniques can also be used, and are pre-sented in Table 1. Fig. 4 presents a typical experimental strategyto study a plasma-modified polymer. XPS analysis of plasma-modified samples became relatively common only in the 90s withthe increase in the commercially available XPS equipments. Indeed,it is useful to remind that XPS is still a relatively young technique,as Siegbahn was awarded the Nobel prize for it only in 1981. Fig. 5presents a SciFinder Scholar statistic study of the number of hitscontaining both the items “XPS” and “plasma”. From a few papersper year until the mid 80s, it reached around 100 papers at thebeginning of the 90s, to jump to an average of 800 papers these last

a Water contact angle.b X-ray photoelectron spectroscopy.c (Static) secondary ion mass spectrometry.d Secondary electron microscopy.e Atomic force microscopy.f Attenuated total reflectance Fourier transform infrared spectroscopy.

N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408 397

gy to

bwii

3tp

3

efsfAacw

FM

Fig. 4. Typical experimental strate

etter signal/noise ratio. That favored the use of monochromatorshich are often necessary to unambiguously identify functional-

ties grafted onto a polymer surface. A recent review about themprovements in XPS can be found in Ref. [19].

. Section 2: Analytical considerations for the use of XPS inhe characterization of plasma treated polymers andlasma polymers

.1. Charging and energy scale calibration

Most of the time, polymers are non-conductive. Due to the X-rayxcitation mechanism, and to the ejection of the (photo) electronsrom the sample, a positive charge spontaneously appears on theurface. Because of this surface charging, spectra are shifted of a

ew eV towards higher binding energies (lower kinetic energies).lthough in non-monochromatized XPS the X-ray gun also gener-tes some electrons that can reach the surface and that can partlyompensate for the positive charging, this is not the case anymoreith monochromatized XPS. For plasma treated samples, this is

ig. 5. # of reference containing both the concept “XPS” and “plasma” (SciFinder,arch 2009).

study plasma-modified polymers.

dramatic, as one of the main uses of XPS is the identification bypeak fitting of the chemicals functionalities grafted. To compen-sate for this charging effect some authors use an electron flood gun[20–27] or a magnetic lens [28].

If no flood gun is used, the spectra are shifted by the user, withthe software, using internal References Most of the time the inter-nal reference used is the C1s peak of CC/CH bonds coming fromthe polymer itself or the CC signal coming from small contam-inations by adventitious carbon. Depending on the authors, thisreference value is 284.6 eV [21,24,25,29–36] or 285.0 eV [37–50].There is therefore already a 0.4 eV difference between the refer-ences used. Consequently, as some components are separated byless than 0.4 eV it is not easy to have an unambiguous identificationof the functionalities. Although most of the papers in the literatureuse now this approach (energy calibration on the C1s peak), this isnot in perfect agreement with the usually accepted good practicesin analysis: the C1s peak is usually the one which is strongly mod-ified due to the change in chemical environment. On fluorinatedsamples some authors use the F1s signal [51–54] instead of the C1sone. The F1s peak is very symmetrical and can be fit with just onecomponent, even for copolymer having fluorine atoms with differ-ent neighborhoods [52]. The only difference between the differentneighborhoods is the position of the F1s peak (Table 2). Ferrariaet al. [52] propose the following equation (Eq. (4)) to estimate theposition of the F1s peak. This equation was determined from exper-imental data of PVF, PVdF and PTFE. BE stands for binding energy,F/C is the atomic ratio of fluorine over carbon:

( ) ( )

BE(F1s) = −0.6

F

C

2+ 3.32

F

C+ 685.43 (4)

As it can be seen in Fig. 6, the position of the F1s peak in thecopolymer named Viton fits very well with the proposed equation.

Table 2F1s binding energy as a function of the F/C ratio. From Ref. [70].

Polymer F/C ratio F1s position, eV

PVF (–CHF–CH2–) 0.5 686.94PVdF (–CH2–CF2–) 1 688.15PTFE (–CF2–CF2–) 2 689.67Viton 8/5 688.8

398 N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408

FPeF

re

3

ptaptesFts

Table 3Binding energy of various carbon species.

Chemical function C1s B.E., eV Ref.

C–C 284.6 [138]C C 284.7 [70](CH2)n 285.0 [70,138,139]CHF–CH2 285.7 [70]CHF–CH2 285.9 [140]C–N 286.3 [141]–(CF2–CH2)–n 286.4 [70,139]C–O–C 286.45 [139]C–O or C–CF 286.5 [142]C N 286.9 [143]CHF–CNH2F 287.4 [141]CH2–CHF–CH2 287.7 [139]C O or CHF–CH2 287.9 [70,139]C O or CF or N–C O 288.0 [142,143]CHF–CH2 288.1 [140]CF 288.3 [139]CHF–CHF 288.4 [140]O–C O or HFC–CF2 or CF–CF2 289.2 [142]CF 289.8 [145]–(CH2–CF2)–n 290.9 [70]–(CHF–CF2)–n 291.6 [139]CF2 or CFO 292.2 [144]

Fi

ig. 6. F1s binding energy as a function of the atomic ratio F/C for PVF, PVdF andTFE. The squares represent experimental data points. The line corresponds to thempirical equation (Eq. (4)). The full square is the point corresponding to Viton.rom Ref. [52].

According to this, if one would use the F1s peak as an internaleference, its binding energy should be set in accordance with thexpected F/C ratio.

.2. Peak fitting

As one of the major applications of the plasma treatment ofolymers is to chemically modify their surface, the analysis ofhe high-resolution C1s peak (HRC1s) shape with peak fitting ispowerful tool to identify the functionalities grafted on the sam-le. Depending on the chemical environment of the carbon atom,he C1s peak can present high chemical shifts making it relatively

asy to identify its main components. Fig. 7 presents the HRC1spectrum of a modified polyimide sample. The addition of O, N andsubstituent results in strong changes in the C1s envelope due to

he various chemical environments of the carbon atoms (chemicalhifts). The peak fitting can help to identify the grafted components.

ig. 7. C1s spectra from modified polyimides used as liquid crystal alignment layers. Samncreased nearest-neighbor (beta-C) effects as well as the addition of the peak at ∼292 eV

–(CF2–CF2)–n 292.5 [70,139,140]CF3–or CF2–O 294.1 [144]CF3–CF2 294.6 [139]

However, it is not always as straightforward, as various speciespresent chemical shifts very similar to one and other (Table 3).

In some cases the difference in binding energy between twospecies is smaller than the resolution of the apparatus, even witha monochromator. In this case the two species cannot be distin-guished from each other.

3.3. Derivatization

For some components, like C–N and C–O (see Table 3), thebinding energies of the C1s peak are very close, making it almostimpossible to identify unambiguously the components. In suchcases, a chemical derivatization may help to identify specific

ple contains a side-chain which includes both an amide and –CF3. This results in[137].

N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408 399

Fig. 8. Derivatization reactions used to identify hydroxyl, carbonyl

Fig. 9. XPS C1s spectra of polypropylene treated by an oxygen plasma. Top beforeand lower after trifluoroacetic anhydride (TFAA) derivatization. TFAA is use to detect–OH functionalities. From Ref. [63].

and carboxyl groups (a). Or amine groups (b) from Ref. [67].

species. The basic principle is to have a very specific reactionbetween the target functionalities to be studied and a specialmarker molecule. The marker molecule usually contains fluorineatoms because of the very large chemical shift it induces in the C1sbinding energy. More generally the marker must have at least oneelement not present in the original sample. Therefore the targetfunctionality can be detected unambiguously by the new elementpresent in the spectra. Careful calculations, taking into account thestoichiometry of the reaction, may allow to quantify the targetfunctionalities probed. Fig. 8 shows some examples of derivatiza-

tion reactions. After the reaction, new components appear on theHRC1s spectra. Those new components are the signature of the tar-geted functionalities that have reacted with marker molecule. Fig. 9presents the HRC1s spectra of O2 plasma treated polypropylene

Fig. 10. Analysis depth varying with the photoemission angle. The sapling depth ishighlighted in grey, and is also shown with the dotted line for 0◦ angle and with thedashed line for a 70◦ angle.

400 N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408

Fp

st

n[sm[

3

eha

Fsa

Fig. 13. Surface analysis by XPS and water contact angle (WCA) of PTFE samplestreated by a r.f. nitrogen plasma (p(N2) = 5 × 10−2 Torr). Full plot of the cos(WCA) as

More specifically, when it comes to plasma-treated polymers,

ig. 11. Apparent oxygen concentration as a function of the depth of analysis, for aolystyrene sample treated by a 10% O2–90% He r.f. plasma. From Ref. [145].

ample, before and after derivatization of the hydroxyl groups byrifluoroacetic anhydride (TFAA).

We emphasize here that derivatization is a very powerful tech-ique, but the overall analysis procedure is delicate, and not trivial55,56]. Those derivatization reactions are of great interest forurfaces containing both nitrogen and oxygen. The most com-on groups to be derivatized are carboxyl [44,57–62], hydroxyl

57–59,61,63,64], carbonyl [57,58,61] and amine [34,61,64–67].

.4. Angle-resolved XPS (ARXPS)

In-depth information in XPS is usually obtained by means of iontching (“depth profiling”). However, this approach is mostly pro-ibited for polymers, as the information about the chemical bondsnd the chemical environment will be lost by the sputtering [68].

ig. 12. Polystyrene treated by a 10% O2–90% He r.f. plasma: effect of the photoemis-ion angle on the C1s envelope: (a) 0◦ photoemission angle; (b) 75◦ photoemissionngle. Data not corrected for charging. From Ref. [69].

a function of the F surface concentration (bottom scale, full marks), and as a functionof the N concentration (top scale, open marks). The circles, squares, and trianglesrepresent the different plasma powers applied. For one single set of data (for instancethe open circles), the only parameter varied is the time of exposure. From Ref. [87].

During angle-resolved XPS measurements the angle betweenthe sample surface and the entrance of the analyzer is changed, thisallows to analyze different depth. When the photoemission angleis set to 0◦ the sampling depth is maximum, when the angle is setto 90◦ the sampling depth is minimum (Fig. 10).

only the very first atomic layers will be chemically modified by thetreatment. As during XPS measurements, the information is aver-aged out over a depth of 3–6 nm, depending on the attenuation of

Fig. 14. Effect of the r.f. power of H2 plasma treatment on: (a) the [O]/[C] and [F]/[C]ratio of the PTFE surfaces; and (b) the 180◦-peel adhesion strength of the Cu/PTFEassembly. From Ref. [111].

N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408 401

Fig. 15. The XPS wide scan spectra of the delaminated PTFE and Cu surfaces of theCu/pp–GMA–PTFE assemblies: (a) having a 180◦-peel strength of about 0.5 N/cmand prepared in the absence of H2 plasma pre-activation and (b) having a 180◦-peelstrength of about 5 N/cm and prepared in the presence of H2 plasma pre-activation.From Ref. [111].

Fig. 16. A plasma polymerized film examined by XPS. Points are experimentaldpfis

ttbp

Fa

Fig. 18. Surface composition variation as a function of time storage in ambient air.aPS: atactic polystyrene, iPS: isotactic polystyrene From Ref. [20].

3.5. X-ray degradation

One of the big issues in XPS analysis of polymers is sample degra-

ata fitted with seven contributions in the C1s region. A precisely resolved C–Nieak at 283.6 eV permits the identification of direct carbon–nickel bonds at thelm/substrate interface, which account for the good adhesion of the film to theubstrate. From Ref. [113].

he emergent electrons [69], the entire useful in-depth informa-ion can therefore be obtained by ARXPS. It allows to discriminateetween the compositions of the top layers (those modified bylasma) and the averaged out surface. For example during the expo-

ig. 17. Wetting properties of CO2 plasma-treated PTFE foils depending on thembient conditions. From Ref. [114].

Fig. 19. Relative contact angle variation as a function of time storage in ambient air.aPS: atactic polystyrene, iPS: isotactic polystyrene From Ref. [20].

sure of a polystyrene sample to an oxygen/helium plasma, the Oat.% decreases while the sampling depth increases (decreasing pho-toemission angles), Fig. 11. The shape of the HRC1s also changeswith the depth of analysis. The HRC1s peak has more oxygenatedcomponent (C–O, C O. . .) at the surface, showing that the plasmatreatment only affects the top layers of the sample (Fig. 12).

dation. It is indeed well known that some polymers are degraded

Fig. 20. The ARXPS data for the O–CO–O component in the C1s envelope, obtainedin Al (close symbols) or Mg (open symbols) radiation: square = 0◦ , circle = 15◦ , uptriangle = 30◦ , down triangle = 45◦ , diamond = 60◦ , star = 75◦ . From Ref. [71].

402 N. Vandencasteele, F. Reniers / Journal of Electron Spectrosco

Fig. 21. Utility of XPS for the study of plasma polymerization. The upper spectrumshows the C1s peak of the monomer (6FDA) whereas the lower spectrum shows theC1s peak of the resulting polymer. From Ref. [76].

Fig. 22. C1s photoelectron peak for (a) commercial amorphous native polystyrene (PS), (b(10′ , 1 kV, 14 kHz, 2 × 104 Pa), (e) pp-PS on PTFE DBD Ar (10′ , 1 kV, 14 kHz, 2 × 104 Pa) and

py and Related Phenomena 178–179 (2010) 394–408

by X-ray exposure [69–72]. In fact this degradation comes from thesecondary electrons emitted during the X-ray exposure. In mostof the cases this degradation is slow enough not to change thecomposition of the sample during the analysis. However, this canbe a problem while performing ARXPS, often used to characterizeplasma-treated samples. The X-ray exposure time is much longer,as several spectra of the same region have to be recorded at differentangles [73]. The exposure time is therefore multiplied by the num-ber of angles at which the measurements have to be done. To avoidX-ray damage on the samples some authors use a reduced X-raypower (150 W instead of the usual 300 W power) [21,30,52,74–86].

4. Section 3: Applications of XPS for selected“materials–plasma” applications

4.1. Modification of polymers by plasma–adhesion

Fig. 13 [87] shows the correlation between the surface energyand the surface composition for PTFE samples treated by a lowpressure radio frequency (r.f.) N2 plasma. The surface energy is

represented by the cosine of the contact angle (Eq. (2)). It can beseen that an increase in the nitrogen content and a decrease in thefluorine content are correlated to an increase in the surface energy.

XPS is used to quantify the amount of polar species graftedand their nature. One of the most common and efficient treat-

) commercial crosslinked PS, (c) native PTFE (substrate), (d) pp-PS on PTFE DBD He(f) pp-PS on PTFE Atom flow (10′ , 50 W) [13].

trosco

ma[aitoprbrt[

asato

teaFCritP

itaposa

futbmmmao(

4

itmprhraeMctw

monomer in the plasma, and therefore provide information aboutthe mechanism of reaction. Indeed, if the ratios of the peaks of theelements are the same in the pp-film and in the monomer one canconclude that there was not much degradation of the monomerfunctionalities by the plasma.

N. Vandencasteele, F. Reniers / Journal of Electron Spec

ent to increase the surface energy of polymer and therefore theirdhesion properties is the exposure to oxygen containing plasma6,12,63,75,88–93]. Other gases are also used CO2, N2, NH3, air, H2nd Ar treatment [12,61,94–104]. Most of them result in an increasen the oxygen content. This oxygen originates either directly fromhe plasma itself or is introduced through post plasma reactionr reactions of the grafted species with oxygen or water vapourresent in the atmosphere [8,105]. The sometimes long lifetime ofadicals generated by the plasma but present at the surface or in theulk of the polymer is still an issue for industrial applications. Thoseadicals can react with the environment and cause unwanted post-reatment reactions resulting in a degradation of the final product106,107].

Adhesion of metal on polymers with low-k dielectric constant isn important matter for the microelectronics industry. The Cu/PTFEystem is of particular importance because of the excellent thermalnd dielectric properties of PTFE as well as the low resistance ofhe copper [108–112] making it the new system of choice insteadf aluminum for integrated circuits [26].

As polymers tend to have a low surface energy, their adhesiono metals is weak. As already shown, plasma treatments are veryfficient to increase the surface energy of polymers, therefore theyre used to increase the adhesion between metal and polymers.ig. 14 [111] presents the increase in adhesion strength between theu/PTFE assembly after plasma treatments (H2 plasma treatment,.f. power = 80 W, p = 100 Pa, H2 flow rate = 20 sccm and t = 20 s). Thisncrease in the adhesion strength is correlated with the increase inhe [O]/[C] and with the decrease in the [F]/[C] atomic ratio on theTFE surface.

The study of both the polymer and the metal surface after delam-nation gives information about the type of bond between thewo surfaces. If metal is still detected on the metallic surface, thedhesion was not good; the metal is totally delaminated from theolymer. On the other hand if the surface analysis of the metal sidenly shows polymer signal it means that adhesion was good and theample have delaminated by cohesive failure. Fig. 15 [111] showsn example of both cases.

The adhesion between metals and polymers is also importantor micro- and nano-technologies. Ultra-thin fluorinated films aresed as anti-adhesive layer in the replication of micro- and nanos-ructures by hot embossing, where a controlled interface is desiredetween the hard master and the thermoplastic [113]. The hardaster is the mould used to replicate the nanostructure on the ther-oplastic material. The fluorinated films were deposited on the Niaster by r.f. plasma sputtering of a PTFE target. HRC1s spectrum

llows the author to have information on the adhesion of the filmn the metallic master. The C–Ni bond detected in the HRC1s peakFig. 16), accounts for the good adhesion of the film to the substrate.

.2. Study of the ageing of plasma-modified polymers by XPS:

Plasma treatments are often used to increase the hydrophilic-ty (increase of the surface energy) of the sample. However, dueo the rotation and the movement of the polymer chains, this

odification is not permanent. This is referred to as the ageinghenomenon. When the samples are stored in a hydrophobic envi-onment such as air, a hydrophobic recovery occurs, which is notappening when the samples are stored in a hydrophilic envi-onment like water. Fig. 17 shows the contact angle results for

PTFE surface treated by a CO2 plasma (MW–ECR, microwave

lectron cyclotron resonance, p = 2 × 10−3 mbar, P = 300 W, Pulsedode Duty Cycle = 50%, treatment time = 4 min). Storage in water

auses a significant improvement in wettability. However, whenhe sample is removed from the water, a fast hydrophobic recoveryas observed [114].

py and Related Phenomena 178–179 (2010) 394–408 403

This hydrophobic recovery can be explained by a change inthe surface composition of the samples during the ageing. Whenthe samples are stored in air, the polar functionalities present atthe surface (–OH, –COOH. . .), slowly disappear in the bulk of thesample. Fig. 18 [20] presents the O/C ratio of isotactic and atacticpolystyrene sample treated by an oxygen plasma (p(O2) = 4 mbar,P = 6 W, t = 60 s Duty Cycle = 1%). A few days after the treatmentthe O/C ratio at the surface is already decreasing [20,71,72] as thepolar functionalities are buried in the bulk of the polymer, toodeep to be detected by XPS. This change in the surface composi-tion is very well correlated with the change of the surface energy(Figs. 18 and 19).

These changes in surface composition are also correlated tochanges in the envelope of the HRC1s peak. Fig. 20 [71] presentsthe % of the C1s envelope attributed to O–CO–O component for apolystyrene sample treated by a 10% oxygen–90% helium plasma.The plasma parameters were: r.f. power = 15 W at 13.56 MHz,p = 13.3 Pa, gas flow = 0.169 Pa m3 s−1). The high energy component,coming from polar functionalities (e.g. C–OH, C O, COOH. . .) slowlydisappears with time. This example also underlines the power ofARXPS to study the ageing of plasma-treated polymer surfaces.

4.3. XPS to study plasma polymerized films

XPS is also used to characterize plasma polymerized films (pp-film). It is one of the few techniques that give access to the atomiccomposition of the films. Indeed, one of the advantages of XPS (overSIMS for example) is the relatively easy and robust atomic quantifi-cation. This can also give information about the degradation of the

Fig. 23. Comparison of valence band (VB) spectra of pulsed plasma polymerized PSand commercial PS. From Ref. [116].

404 N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408

F tylene[

tp4pdcg

stmpdaSci

ig. 24. Comparison of VB spectra from pulsed plasma polymerized ethylene, ace116].

The HRC1s peak could also be used to study fragmenta-ion and/or damage of the monomer during the polymerizationrocess. Fig. 21 shows the HRC1s spectra of the monomer,,4′-hexafluoroisopropylidene diphthalic anhydride (6FDA) beforelasma polymerization and of the plasma polymer film. The strongecrease of the (CO)–O– component (±288 eV) in the pp-6FDA filmlearly evidences damage to the monomer with the lost of CO/CO2roups from the anhydride ring [76].

For aromatic monomers, the �–�* shake-up can be used totudy if the aromatic structure of the monomer was preserved inhe polymerized film. Fig. 22 compares the HRC1s spectra for com-

ercially available polystyrene (PS) (Fig. 22a and b) and plasmaolymerized PS (pp-PS). The pp-PS where synthesized with a

ielectric barrier discharge (DBD) in He (Fig. 22d) or Ar (Fig. 22e)nd with a commercially available r.f. plasma torch (Atomflo 250 DurfX Technologies) (Fig. 22f). It can be seen that the aromaticity isonserved for both films synthesized by DBD as the �–�* shake-ups present in the spectra. On the other hand, the film prepared with

and butadiene with commercial poly(butadiene) and poly(propylene). From Ref.

the atmospheric plasma torch has lost his aromaticity as no �–�*shake-up is detected in the spectrum anymore [13].

Touzin et al. [115] studied the stability of plasma polymerizedfluorocarbon on stainless steel substrates. The survey spectra nevershow any metallic element, even after ageing. Therefore they con-cluded that there is no delamination or cracks of the film even afterageing.

Some authors use the valence band (VB) analysis to studytheir plasma polymerized films [15,116]. Although usually, XPSdeals with core electrons (C1s, N1s, O1s, etc.), there is alsosignificant information at low binding energies representativefrom electrons originating from the valence band of the sample.Those authors compare the VB spectra of plasma polymerized

polystyrene (pp-PS), polyethylene (pp-PE), poly(acetylene) andpoly(butadiene) with those of commercial PS, PE, polypropylene(PP) and poly(butadiene) [116]. For the pp-PS the VB spectraare very similar for both samples (Fig. 23). Comparison betweencommercial poly(butadiene) and pp ethylene, acetylene and buta-

N. Vandencasteele, F. Reniers / Journal of Electron Spectroscopy and Related Phenomena 178–179 (2010) 394–408 405

Table 4XPS results of untreated, plasma treated and aged PSf, PES and PE membranes. From Ref. [121].

Material XPS (at.%)

Carbon Oxygen Sulfur Fluorine Nitrogen

Stoichiometric PSf 84.3 12.0 3.7Stoichiometric PES 75 18.8 6.3Stoichiometric PE 100Untreated PSf (both side) 84.5 ± 0.5 11.8 ± 0.5 3.7 ± 0.1Treated PSf (open)a 74.4 ± 0.2 22.1 ± 0.5 2.6 ± 0.1 0.7 ± 0.2Treated PSf (tight)a,b 73.7 ± 0.6 21.5 ± 0.1 2.6 ± 0.1 0.6 ± 0.1Untreated PES (both sides)c 75.5 ± 0.9 17.1 ± 0.6 6.8 ± 0.4Treated PES (both sides)a 67.9 ± 1.1 25.2 ± 0.5 4.7 ± 0.2 1.4 ± 0.9Untreated PE (both sides) 100Treated PE (A)c,d 72.3 ± 3.5 23.5 ± 1.7 0.8 ± 0.2Treated PE (B)c,d 84.4 ± 2.3 15.4 ± 2.1

ent.

dhpce

tidtirc

4

fitsahtasinb[tecsatspPstccm

mab

tance to protein adsorption. XPS is used to detect the presence orthe absence of the protein on the surface of the sample. Usuallythe N1s signal coming from the protein is used to detect it. Someauthors use plasma treatment to decrease the adsorption of pro-

a Trace impurities (<2%) of atomic silicon also detected.b Trace impurities (<2%) of atomic sodium also detected.c Trace impurities (<2%) of atomic chlorine also detected.d Side A is facing the coil and side B is the downstream side during plasma treatm

iene reveal a good correspondence of the overall shape. Howeverigher densities of state at around 16 eV in the pp spectra (like inolypropylene) suggest that the pp-films are more branched androsslinked (Fig. 24) than the traditional polymers. This is usuallyxpected for plasma polymers.

However, most of the time, XPS analysis alone is not sufficiento obtain a correct chemical characterization of plasma polymer-zed films. This requires a multitechnique approach [117]. Indeeduring the plasma polymerization a lot of fragmentation fromhe monomer could happen, which result in very complex chem-stry. Amongst the other techniques, Tof-SSIMS, nuclear magneticesonance and infrared spectroscopy are the most powerful andomplementary to XPS.

.4. XPS for plasma-modified membranes

Membranes are used in a wide variety of applications: ultra-ltration [118], fuel cells [119], and gas separation [120]. Usuallyhey are made of porous thermoplastic polymers, such as poly-ulfone and fluoropolymers, because of their good mechanicalnd chemical resistance. As the majority of those polymers areydrophobic, before they can be used to filtrate aqueous solutionshey must be rendered hydrophilic by the addition of a wettinggent and/or a surface chemical modification [121]. As alreadyhown plasmas are very efficient for increasing the hydrophilic-ty of polymers. They involve clean and rapid reactions that graftew functionalities and crosslink the surface, without altering theulk properties for which the polymer was originally selected122]. However, for the specific case of membranes it is impor-ant that the modification penetrates through the membrane soven the inside of the pores are treated. XPS has been used toheck for this. If the concentration of the grafted groups is theame on both sides of the membrane, the one facing the plasmand the other one, it can be concluded that the plasma has effec-ively penetrated through the entire membrane [121]. Table 4hows results of polysulfone (PSf), polyethersulfone (PES) andolyethylene (PE) membranes treated by plasma. For both PSf andES membrane the oxygen concentration is the same on eachide of the membrane indicating a penetration of the plasmahrough the membrane. However for the PE membrane the oxygenoncentration is lower on the side not facing the plasma, indi-ating that in this case the plasma has not fully penetrated the

embrane.In the plasma treatment of membranes, acrylic acid is often a

olecule of choice [122–130], as this allows to graft carboxyliccid groups on the membrane. Indeed, those groups are suitable foriotechnological applications such as covalent binding of enzymes

[123]. XPS is used to determine the concentration of carboxylicgroups grafted on the sample (Fig. 25). Derivatization reactions(described in Section 3.3) can be used to have a better estimationof the carboxylic group content.

4.5. XPS for the detection of biofouling properties ofplasma-treated polymers

Plasma treatments are also used to obtain protein resistant sur-faces (i.e. surfaces which do not adsorb proteins). Polyethyleneglycol [131] surfaces are amongst the most used but other exist,such as ultra-hydrophobic PTFE [86] which also present good resis-

Fig. 25. High-resolution XPS spectra of the C1s peak of an untreated polyurethane(PU) membrane, acrylic acid (AA) plasma-treated PUs at several plasma treatmenttimes and synthesized poly(acrylic acid). Plasma power: 5 W. The C O and COOgroups are shown in different spectra for better presentation. From Ref. [122].

406 N. Vandencasteele, F. Reniers / Journal of Electron Spectrosco

Fig. 26. XPS spectra of PTFE samples submitted to various O2 plasma. On the left:samples that have not been exposed to the protein (BSA). On the right: samples thathave been exposed to the BSA. Component 5, 6 and 7 are coming from the BSA. ThepaC

tcat

surface can now be achieved.

Ftt

eak assignments are: (1) CF3 (294 eV), (2) CF2 (292 eV), (3) CF (290 eV), (4) C–CFnd/or C–Ox (287.4 eV), (5) CC (284.6 eV), 6 N–C O (288.2 eV) and (7) C–O and/or–N (286 eV). From Ref. [86].

ein on membranes. The adsorption of proteins on the membraneauses a strong decrease of the permeate flux [118,132]. Wang etl. [132] used the N1s signal from the peptide bond of the pro-ein to detect the adsorption of protein onto their plasma-treated

ig. 27. Left: mechanism of the oligopeptide grafting by radiofrequency glow discharge. Rreatment (100 W, 100 mTorr, 1 min, Ar/O2 mixture); (c) AA-grafted; (d) after bNH2PEG cime is 3 h. From Ref. [136].

py and Related Phenomena 178–179 (2010) 394–408

membranes. The presence of the protein can also be seen on theHRC1s peak shape. Fig. 26 [86] shows the HRC1s spectra of PTFEsamples treated by various O2 plasma before and after exposure toa protein (bovine serum albumin, BSA) solution. Samples that donot resist the protein adsorption exhibit new components comingfrom the protein itself. Those new components are C–C at 285.0 eV,C–N or C–O between 285.7 and 286.5 eV and N–C O at 288.2 eV[133].

Plasma treatments are used to immobilize biomolecules on sur-faces. This is of great importance in tissue engineering research[134,135]. The following example shows the plasma treatment(radio frequency glow discharge) of expended PTFE (ePTFE) witholigopeptides, known for their role in mediating the adhesion ofcells to the extracellular matrix [136]. As shown in Fig. 27 the XPSmeasurements allow to follow the grafting of the species on thesample.

5. Conclusions

X-ray photoelectron spectroscopy is, without any doubt, one ofthe most useful tools to characterize plasma-treated polymers.

The control of the plasma modification of surfaces is closelyrelated to the development of surface analytical techniques, ableto characterize the resulting surface.

XPS is an essential tool to probe the physical and chemicalproperties of the plasma-treated polymers: adhesion, biocompati-bility, ageing are today studied using XPS, in conjunction with othertechniques. XPS is mostly used to probe for the identification ofchemical groups grafted, through the compositional analysis andthe peak fitting (mostly of the C1s peak). With the help of derivati-zation and ARXPS, a decent characterization of the plasma-treated

Finally, plasma polymerization, which is taking more and moreimportance since a few years, also needs XPS to probe for the con-servation of the monomer chemistry and to better understand thepolymerization mechanism.

ight: XPS C1s signal for each modification step. (a) pristine ePTFE; (b) after plasmaoupling; (e) Peptide Arg-Gly-Asp-Cys (RGDV) immobilized ePTFE. The AA grafting

trosco

A

Gbkgf“P2

R

N. Vandencasteele, F. Reniers / Journal of Electron Spec

cknowledgements

The authors would like to thank, J. Hubert, C. Barroo, A. dehellinck and S. Vranckx for their useful help in collecting theibliographic data. The advice of Prof. D.H. Fairbrother (Johns Hop-ins University, Baltimore, US) for the derivatization reactions wasreatly appreciated This work was supported by the financial helprom the Belgian Federal Government through the IAP projectPhysical Chemistry of Plasma Surface Interactions (PSI)” project6/08 and from the Belgian National Fund for Research FNRS (grant.4543.04).

eferences

[1] I. Langmuir, Proceedings of the National Academy of Sciences of the UnitedStates of America 14 (1928) 627–637.

[2] W. Siemens, Poggendorff’a Ann. Phys. Chem. 102 (1857) 66.[3] H.V. Boenig, Fundamentals of Plasma Chemistry and Technology, Technomic

Publishing Company, Lancaster, 1988.[4] B. Chapman, Glow Discharge Processes Sputtering and Plasma Etching, Wiley

Interscience, New York, 1980.[5] Plasma surface modification of polymer for improved adhesion: a critical

review, in: M. Strobel, C.S. Lyons, K.L. Mittal (Eds.), Plasma surface Modifi-cation of Polymers: Relevance to Adhesion, VSP, Utrecht, 1994.

[6] R. Foerch, J. Izawa, G. Spears, Journal of Adhesion Science Technology 5 (1991)549–564.

[7] D. Hegemann, H. Brunner, C. Oehr, Nuclear Instruments and Methods inPhysics Research Section B-Beam Interactions with Materials and Atoms 208(2003) 281–286.

[8] S. O’Kell, et al., Surface and Interface Analysis 23 (1995) 319–327.[9] F. Poncin-Epaillard, G. Legeay, Annales de Chimie Science des Materiaux 28

(2003) 55–66.[10] N. Vandencasteele, H. Fairbrother, F. Reniers, Plasma Processes and Polymers

2 (2005) 493–500.[11] N. Vandencasteele, D. Merche, F. Reniers, Surface and Interface Analysis 38

(2006) 526–530.[12] K. Ozeki, K.K. Hirakuri, Applied Surface Science 254 (2008) 1614–1621.[13] D. Merche, et al., IEEE Transactions on Plasma Science (2009),

doi:10.1109/TPS.2009.2014165.[14] Reniers, F., N. Vandencasteele, O. Burry, Procédé pour déposer une couche flu-

orée à partir d’un monomère précurseur, O.E.D. Brevets, Editor. 2008.[15] J.F. Friedrich, et al., Surface and Coatings Technology 142–144 (2001)

460–467.[16] J.L. Delattre, R. d’Agostino, F. Fracassi, Applied Surface Science 252 (2006)

3912–3919.[17] T.R. Gengenbach, H.J. Griesser, Polymer 40 (1999) 5079–5094.[18] C.D. Bain, G.M. Whitesides, Journal of the American Chemical Society 110

(1988) 5897–5898.[19] F. Reniers, C. Tewell, Journal of Electron Spectroscopy and Related Phenomena

142 (2005) 1–25.[20] J. Larrieu, et al., Surface and Coatings Technology 200 (2005) 2310–2316.[21] M. Atreya, et al., Polymer Degradation and Stability 65 (1999) 287–296.[22] M. Dhayal, M.R. Alexander, J.W. Bradley, Applied Surface Science 252 (2006)

7957–7963.[23] L.J. Gerenser, et al., Surface and Interface Analysis 29 (2000) 12–22.[24] S. Vasquez-Borucki, C.A. Achete, W. Jacob, Surface and Coatings Technology

138 (2001) 256–263.[25] D. Wolany, et al., Surface and Interface Analysis 27 (1999) 609–617.[26] K.J. Kim, et al., Thin Solid Films 398–399 (2001) 657–662.[27] C.E. Moffitt, et al., Plasma and Polymers 6 (2001) 193–209.[28] A. Walker, S. Page, Charged Particle Energy Analyzers, Kratos Analytical Ltd.,

UK, 1994.[29] H. Ardelean, et al., Applied Surface Science 243 (2005) 304–318.[30] M. Atreya, et al., Polymer Degradation and Stability 63 (1999) 53–59.[31] H. Bubert, et al., Spectrochimica Acta Part B 57 (2002) 1601–1610.[32] C.-H. Dai, et al., Polymer 47 (2006) 8583–8594.[33] A.K. Mishra, et al., Progress in Organic Coatings 55 (2006) 231–243.[34] S.L. Ren, S.R. Yang, Y.P. Zhao, Applied Surface Science 227 (2004) 293–299.[35] C.-H. Wen, M.-J. Chang, G.-H. Hsiue, Applied Surface Science 252 (2006)

3799–3805.[36] L. Zhang, et al., Surface and Interface Analysis 28 (1999) 16–19.[37] C. Aspagaus, A. Rossi, P. Rudolf von Rohr, Applied Surface Science 252 (2005)

1581–1595.[38] G. Borcia, C.A. Anderson, N.M.D. Brown, Applied Surface Science 225 (2004)

186–197.[39] G. Borcia, C.A. Anderson, N.M.D. Brown, Applied Surface Science 221 (2004)

203–214.[40] C.M. Cepeda-Jiménez, R. Torregrosa-Macià, Surface and Coatings Technology

174–175 (2003) 94–99.[41] M. Collaud Coen, et al., Applied Surface Science 207 (2003) 276–286.[42] N.-Y. Cui, et al., Applied Surface Science 252 (2007) 3865–3871.[43] N. De Geyter, et al., Surface and Coatings Technology 201 (2007) 7066–7075.

py and Related Phenomena 178–179 (2010) 394–408 407

[44] S. Kröpke, Y.S. Akishev, A. Holländer, Surface and Coatings Technology142–144 (2001) 512–516.

[45] C. Liu, N.M.D. Brown, B.J. Meenan, Surface Science 575 (2005) 273–286.[46] R. Morent, et al., Surface and Interface Analysis 40 (2007) 597–600.[47] D. Nicolas-Debarnot, et al., Surface and Coatings Technology 200 (2006)

4257–4265.[48] M. Strobel, et al., Plasma and Polymers 8 (2003) 61–95.[49] D.J. Upadhyay, et al., Colloids and Surfaces A: Physicochemical and Engineer-

ing Aspects 248 (2004) 47–56.[50] C. Vilani, et al., Applied Surface Science 254 (2007) 131–134.[51] J. Carpentier, G. Grundmeier, Surface and Coatings Technology 192 (2005)

189–198.[52] A.M. Ferraria, J.D. Lopes da Silva, A.M. Botelho do Rego, Polymer 44 (2003)

7241–7249.[53] I. Sugimoto, et al., Polymer 41 (2000) 511–522.[54] V. Yanev, et al., Surface Science 566–568 (2004) 1229–1233.[55] A. Chilkoti, B.D. Ratner, Surface and Interface Analysis 17 (1991) 567–574.[56] A. Chilkoti, B.D. Ratner, D. Briggs, Chemistry of Materials 3 (1991) 51–61.[57] L.A. Langley, D.E. Villanueva, D.H. Fairbrother, Chemistry of Materials 18

(2006) 169–178.[58] R.P. Popat, I. Sutherland, E. Sheng, Journal of Material Chemistry 5 (1995)

713–717.[59] I. Sutherland, et al., Journal of Material Chemistry 4 (1994) 683–687.[60] M.R. Alexander, P.V. Wright, B.D. Ratner, Surface and Interface Analysis 24

(1996) 217–220.[61] N. Inagaki, et al., Journal of Applied Polymer Science 46 (1990) 399–413.[62] M.R. Alexander, T.M. Duc, Polymer 40 (1999) 5479–5488.[63] J.F. Friedrich, et al., Surface and Coatings Technology 74–75 (1995) 664–669.[64] G. Kühn, et al., Surface and Coatings Technology 142–144 (2001)

494–500.[65] A. Holländer, S. Kröpke, F. Pippig, Surface and Interface Analysis 40 (2008)

379–385.[66] S. Lucas, et al., Nuclear Instrument and Methods in Physics Research B 266

(2008) 2494–2497.[67] J. Kim, et al., Applied Surface Science 253 (2007) 4112–4118.[68] P.J. Cumpson, Journal of Electron Spectroscopy and Related Phenomena 73

(1995) 25–52.[69] R.W. Paynter, Surface and Interface Analysis 33 (2002) 14–22.[70] G. Beamson, D. Briggs, High Resolution XPS of Organic Polymers. The Scienta

ESCA300 Database, John Wiley & Sons, Chichester, 1992.[71] M.-C. Tremblay, R.W. Paynter, Surface and Interface Analysis 35 (2003)

502–514.[72] R.W. Paynter, H. Benalia, Journal of Electron Spectroscopy and Related Phe-

nomena 136 (2004) 209–220.[73] R.W. Paynter, Surface and Interface Analysis 33 (2002) 862–868.[74] E. Durand, et al., Plasma and Polymers 7 (2002) 311–325.[75] A. Vesel, M. Mozetic, A. Zalar, Vacuum 82 (2008) 248–251.[76] G. Maggioni, et al., Surface and Coatings Technology 200 (2005) 481–485.[77] K. Kin Chee Ho, A.F. Lee, A. Bismarck, Carbon 45 (2007) 775–784.[78] D.S. Bodas, A.B. Mandale, S.A. Gangal, Applied Surface Science 245 (2005)

202–207.[79] C. Jama, et al., Surface and Interface Analysis 27 (1999) 653–658.[80] A. Raffaele-Addamo, et al., Applied Surface Science 252 (2006) 2265–2275.[81] G. Xu, et al., Colloids and Surfaces B: Biointerfaces 19 (2000) 249–256.[82] D.J. Wilson, R.L. Williams, R.C. Pond, Surface and Interface Analysis 31 (2001)

385–396.[83] H. Krump, et al., Applied Surface Science 252 (2006) 4264–4278.[84] W. Prissabaroon, et al., Thin Solid Films 477 (2005) 131–139.[85] Q.F. Wei, et al., Applied Surface Science 245 (2005) 16–20.[86] N. Vandencasteele, et al., Plasma Processes and Polymers 5 (2008) 661–671.[87] N. Vandencasteele, F. Reniers, Surface and Interface Analysis 36 (2004)

1027–1031.[88] R.W. Paynter, Surface and Interface Analysis 26 (1998) 674–681.[89] N. Shahidzadeh-Ahmadi, et al., Colloids and Surfaces A: Physicochemical and

Engineering Aspects 105 (1995) 277–289.[90] K.S. Kim, et al., Polymer 44 (2003) 6287–6295.[91] M.O.H. Cioffi, H.J.C. Voorwald, R.P. Mota, Materials Characterization 50 (2003)

209–215.[92] M. Morra, E. Occhiello, F. Garbassi, Langmuir 5 (1989) 872–876.[93] T. Belmonte, et al., Surface and Coatings Technology 200 (2005) 26–30.[94] I. Banik, et al., Polymer 44 (2003) 1163–1170.[95] F. Poncin-Epaillard, B. Chevet, J.-C. Brosse, Macromolecular Chemistry 192

(1991) 1589–1599.[96] R. Foerch, Les vide, les couches minces 246 (1989) 213–215.[97] M. Bryjak, et al., European Polymer Journal 38 (2002) 717–726.[98] J.E. Klemberg-Sapieha, et al., Journal of Vacuum Science Technology 9 (1991)

2975–2981.[99] P. Favia, M.V. Stendardo, R. d’Agostino, Plasma and Polymers 1 (1996) 91–112.

[100] N.-Y. Cui, N.M.D. Brown, Applied Surface Science 189 (2002) 31–38.[101] C. Riccardi, et al., Applied Surface Science 211 (2003) 386–397.

[102] D.T. Clark, A. Dilks, Journal of Polymer Science 16 (1978) 911–936.[103] V. Stelmashuk, et al., Vacuum 77 (2005) 131–137.[104] J.C. Caro, et al., European Polymer Journal 35 (1999) 1149–1152.[105] A.J. Wagner, D.H. Fairbrother, F. Reniers, Plasmas and Polymers 8 (2003)

119–134.[106] B. Finke, K. Schroder, A. Ohl, Plasma Processes and Polymers 5 (2008) 386–396.

4 trosco

[

[[[[[[[[[

[[[[[

[[

[[[[

08 N. Vandencasteele, F. Reniers / Journal of Electron Spec

107] N. Inagaki, Plasma Surface Modification and Plasma Polymerization CRC(1996).

108] K.S. Kim, et al., Thin Solid Films 377–378 (2000) 122–128.109] S. Wu, et al., Polymer 40 (1999) 6955–6964.110] Y.Q. Zhu, et al., Applied Surface Science 225 (2004) 144–155.111] X.P. Zou, et al., Polymer 42 (2001) 6409–6418.112] M.S. Silverstein, L. Sandrin, E. Sacher, Polymer 42 (2001) 4299–4307.113] R.W. Jaszewski, et al., Applied Surface Science 143 (1999) 301–308.114] U. König, et al., Colloids and Surfaces B: Biointerfaces 25 (2002) 313–324.115] M. Touzin, et al., Surface and Coatings Technology 202 (2008) 4884–4891.116] I. Retzko, et al., Journal of Electron Spectroscopy and Related Phenomena 121

(2001) 111–129.117] W. Unger, et al., Surface and Interface Analysis 38 (2006) 522–525.118] K.S. Kim, et al., Journal of Membrane Science 199 (2002) 135–145.119] D. Ramdutt, et al., Journal of Power Sources 165 (2007) 41–48.120] E. Finot, et al., Applied Surface Science 187 (2002) 326–338.121] M.L. Steen, A.C. Jordan, E.R. Fisher, Journal of Membrane Science 204 (2002)

341–357.122] D.E. Weibel, et al., Journal of Membrane Science 293 (2007) 124–132.

123] H. Nedelmann, et al., Surface and Coatings Technology 116–119 (1999)

973–980.124] Z.Y. Cheng, S.H. Teoh, Biomaterials 25 (2004) 1991–2001.125] S.D. Lee, et al., Biomaterials 17 (1996) 1599–1608.126] M. Ulbricht, G. Belfort, Journal of Membrane Science 111 (1996) 193–215.127] D.S. Wavhal, E.R. Fisher, Journal of Membrane Science 209 (2002) 255–269.

py and Related Phenomena 178–179 (2010) 394–408

[128] D.S. Wavhal, E.R. Fisher, Langmuir 19 (2003) 79–85.[129] T. Yamaguchi, S. Nakao, S. Kimura, Macromolecules 24 (1991) 5522–5527.[130] H.Y. Yu, et al., Water Research 42 (2008) 4341–4347.[131] G.P. Lopez, et al., Journal of Biomedical Materials Research 26 (1992)

415–439.[132] P. Wang, et al., Journal of Membrane Science 195 (2002) 103–114.[133] C.D. Tidwell, et al., Surface and Interface Analysis 31 (2001) 724–733.[134] J.-L. Dewez, et al., Biomaterials 20 (1999) 547–559.[135] I. Djordjevic, L.G. Britcher, S. Kumar, Applied Surface Science 254 (2008)

1929–1935.[136] C. Baquey, et al., Nuclear Instrument and Methods in Physics Research B 151

(1999) 255–262.[137] J.E. Fulghum, Journal of Electron Spectroscopy and Related Phenomena 100

(1999) 331–355.[138] F. Garbassi, M. Morra, E. Occhiello, Polymer Surfaces from Physics to Technol-

ogy, John Wiley & Sons, 1998.[139] C. Girardeaux, et al., Applied Surface Science 96–98 (1996) 586–590.[140] G. Nanse, et al., Carbon 35 (1997) 175–194.[141] S.D. Pringle, V.S. Joss, C. Jones, Surface and Interface Analysis 24 (1996)

821–829.[142] D.J. Wilson, et al., Surface and Interface Analysis 30 (2000) 36–39.[143] F.M. Petrat, et al., Surface and Interface Analysis 21 (1994) 274–282.[144] R.Q. Liang, et al., Surface and Coatings Technology 131 (2000) 294–299.[145] R.W. Paynter, M. Ménard, Journal of Electron Spectroscopy and Related Phe-

nomena 151 (2006) 14–18.