11 pages

Ellipsometry and energy characterization of the electron impact polymerization in the range 0–20 eV

of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Ellipsometry and energy characterization of the electron impact polymerization in the range 0–20 eV
  Ellipsometry and energy characterization of the electron impactpolymerization in the range 0 – 20 eV  V.I. Zyn Department of Physics, Mathematics and Computer Science, Samara State Academy of Social Sciences and Humanities, 65/67, Maxim Gorkiy St., Samara443099, Russia H I G H L I G H T S   Obtaining spectra of activated poly-merization using ellipsometry andKelvin probe.   Identi 󿬁 ed: two resonant and onenon-resonant mechanisms of theactivation.   The resonances are due to the actionof the dissociative electron attach-ment.   Kinetics of transient processes inadsorbed layer under 20 eV pulsedelectron beam. G R A P H I C A L A B S T R A C T < 3 eV3 eV < eU < 10 eV eU > 10 eV E L E C T R O N SS U B S T R A T EF I L M a r t i c l e i n f o  Article history: Received 11 June 2015Received in revised form10 January 2016Accepted 13 January 2016Available online 15 January 2016 Keywords: Electron impact polymerizationEllipsometryKelvin's probeLow energy spectraDissociative electron attachmentKinetic curves a b s t r a c t The electron impact polymerization of adsorbed vapors of a hydrocarbon vacuum oil with molecular mass450 Da (C 32 H 66 ) has been studied in-situ in the range 0 – 20 eV using ellipsometry and a servo system withthe Kelvin's vibrating probe. This allowed registering at the same time the two energy-dependent char-acteristics (spectra) of the process: the  󿬁 lm growth rate and the electrical potential of the irradiated sur-face. The 󿬁 rst spectrum has two resonance maximanear 2.5 and 9.5 eV while the surface potential has onlyone weak extremum near 9.5 eV. The  󿬁 rst growth rate peak at 2.5 eV was connected with a creation of radicals through a resonant process of the dissociative electron attachment and beginning polymerization.The peaks at 9.5 eV in both the spectra mean accelerating polymerization and decreasing surface chargeowing to simultaneous birth of highly active radicals and free electrons. The single resonant processcontrolling both the processes simultaneously is the dissociative attachment of an electron to an anti-bonding molecular orbital, almost the same as at the 2.5 eV but differing by deeper decomposition of thetransient anion, among the products of which are now not the radicals only but also free electrons. Thekinetic curves obtained in pulsed regimes of the electron bombardment were qualitatively identical fordifferent precursors and were used for calculations of cross sections of these processes. &  2016 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/radphyschem Radiation Physics and Chemistry http://dx.doi.org/10.1016/j.radphyschem.2016.01.0200969-806X/ &  2016 Elsevier Ltd. All rights reserved. E-mail address:  zyn37@mail.ruRadiation Physics and Chemistry 122 (2016) 82 – 92  1. Introduction The electron impact is an important part of any plasma che-mical reaction including plasma polymerization. On the otherhand, the phenomenon is used also as a standalone process infabrication of the polymer  󿬁 lms and nanostructures.  “ Controllingthe outcome of reactions is a central issue of chemical research.Physical tools can achieve this if they are able to precisely dis-sociate speci 󿬁 c bonds of a molecule … . An electron beam is aphysical tool that is capable of preparing molecules in reactivestates or, at low electron energies, of initiating highly selectivebond dissociation. ”  (Böhler et al., 2013) A lot of initial monomers,formation regimes and properties of the  󿬁 lms were studied,spheres and methods of their optimal applications were de-termined. ''Despite this, the mechanisms of   󿬁 lm growth are largelyunknown, and current models are based on purely chemical ar-guments.'' (Michelmore et al., 2013) The  󿬁 rst stage of this in manyaspects unknown mechanism is a transformation of the initiallypassive gas into the active species capable of initiating numerousreactions of a synthesis up to a formation of the solid polymermacroparticles and continuous surface  󿬁 lm coatings. The electronimpact plays leading role in this activationwhich can occur both ina plasma volume and at various solid surfaces. The energies of themost of plasma electrons lie within the region 0 – 20 eV. Moreover,there are a lot of electrons below 8 – 10 eV which cannot knockelectrons out of the molecules and so create positive ions but arequite capable of transforming them into chemically active particlesthrough the processes of excitation, dissociation and electron at-tachment. These effects including the latest one were mentionedas a mechanisms of creation of ions and radicals in plasma etchingsystems. (Rueger et al.,1997; Jayaraman et al.,1999) They will also be referred to in our further discussion. Electron collision data formolecular modeling can be found in (Morgan, 2000).Researches of the electron impact in plasma chemistry aredistributed in several directions: (1) surface polymerization in-itiated by relatively high energy electrons, more than 100 eV.Physical aspects of the problem, methods of the researches andeven the language in use resemble those inherent in the radiationchemistry (Weiss et al., 2001; Kim and Huh, 2012; Razanau, 2011; Trey et al., 2010; Feng et al., 2003; Lavalle 2004; Berejka, 2004). (2) Modi 󿬁 cation of   󿬁 nished polymeric materials using the highenergy electron-beam techniques (Bao et al., 2004; Yang, 2009; Shnitov, 2009; Hong et al., 2008; Khan et al., 2008; Nathawat et al., 2007; Egerton et al., 2004). (3) Use of low energy electrons for the surface modi 󿬁 cation. This energy region coincides with main en-ergetic levels of molecules what makes their electronic excitationresonant and very ef  󿬁 cient. Besides, a new channel of generationof active particles is opening - the dissociative electron attachment(DEA) which may have in some cases very large cross sections( 4 10  16 cm 2 ) and is well known to ef  󿬁 ciently break molecularbonds well below the bond energy, even at zero energy if theelectron af  󿬁 nity of the fragment is large enough. The DEA is in nocase a rarity. Vice versa, it is quite common phenomenon formolecular systems with free low-energy electrons (Suzuki, 2011;Stoffels et al., 1998; Bazin et al., 2010, 2009; Zappa et al., 2006; Abouaf and Dunet, 2006; Massey and Sanche, 2013). It is known that the electron irradiation causes numerous reactions in exposedmolecular system. Low energy of the electrons allows carrying outprocesses of modi 󿬁 cation in rather soft conditions and makingchanges in the surface properties reversible (Suzuki, 2011) Theinteractions of molecules with slow electrons are important asbasic processes in plasma modi 󿬁 cation of solid surfaces, in parti-cular those used in the semiconductor industry, medicine andbiotechnology. One of the most important discoveries turned outto be the leading role of the low energy electrons in resonant andselective activating transmutations in the living tissues andtumors, subjected to the ionizing irradiation. Many fundamentalaspects of such interactions were investigated in numerous workssince 1970s and have been summed up in the review article(Massey and Sanche, 2013). In this period there have been devel-oped numerous new ef  󿬁 cient experimental and theoretical ap-proaches, concepts and mechanisms, discovered and studied neweffects and processes, including syntheses, polymerization, mod-i 󿬁 cation and degradation of the matter induced by the low energyelectron impact. There is information concerning the role of sec-ondary and photo-emitted electrons in devices of the vacuumultraviolet light, where the main part of the electrons has energies0 – 20 eV and causes unwanted reactions and contaminations onsurfaces of mirrors with multi-layer interference coatings (Yak-shinskiy et al., 2007). (4) The electron beam can produce polymeraerosols, either through a decomposition of existing volumetricpolymer, or synthesizing particles from the monomer molecules ina gas phase (Weiss et al., 2001; Khan et al., 2008; Singelyn, 1981). (5) Electron beam lithography is now an intensely evolving kind of a precise nanotechnology (Kretz, 2006; Chen et al., 2005; Hu, 2004). (6) Using pulsed electron beams allows to achieve higheref  󿬁 ciency and control in any of the cases mentioned above (Fenget al., 2003; Weiss et al., 2005; Richter et al., 2006). (7) Researches of the mechanisms of the electron-beam polymerization are non-numerous and insuf  󿬁 cient for providing their complete compre-hension, especially in the region of the low electron energies(Böhler et al., 2013; Weiss et al., 2001; Ghosh and Palmese, 2005). Reviews of a lot of fundamental aspects and technological ap-plications of electron beams can be found in references (Masseyand Sanche, 2013; Industrial Electron Beam Processing, 2010; In- dustrial Radiation Processing With Electron Beams and X-rays,2011).The main objects and purposes of the research were:  󿬁 rstly,development of the techniques and obtaining the results of theresearch of the surface polymerization induced by the electronbombardment with the energy 0 – 20 eV, and secondly, formulationof the polymerization mechanism, being in agreement with theobtained experimental data. Saying more precisely, the electron – molecular collisions provide not whole long process of the poly-merization but do only its  󿬁 rst stage - the activation of the poly-merization, i.e. the process of stimulated transition of a gas from achemically passive to chemically active state. The range of theenergy was chosen for reasons of its coincidence with the energiesof excitation, dissociation and ionization of the molecules, whichin general are just the activation processes. This allows carryingout the activation with minimal excitation of additional unwantedprocesses.The used experimental procedure includes obtaining twofunctions of electrons energy (spectra): the  󿬁 lm thickness growthrate and the potential of the  󿬁 lm's surface being charged by theimpinging electrons. In fact, these spectral functions are kineticcharacteristics of the polymerization process, both having beenswept on the electrons' energy. There is a dif  󿬁 culty in maintainingrequired energy of the bombardment caused by the uncontrolledcharging of the  󿬁 lm's surface by the impinging electrons. Theproblem is especially serious in the case of low energies of elec-trons (Ohya, 2014). The task was solved with the help of an au-tomatic system with the vibrating probe tracking the surface po-tential and compensating its changes. The thickness of the  󿬁 lmwas being measured also in-situ with use of the ellipsometry. Newexperimental data allow achieving new theoretical comprehen-sion of the mechanism of the polymerization. 2. Methods and experimental scheme Usually, the most useful information about the mechanism of any induced reaction is contained in its spectrum, which is V.I. Zyn / Radiation Physics and Chemistry 122 (2016) 82 – 92  83  understood as a dependence of the reaction rate on the energy of the impact. In our case such a spectrum is the dependence of thepolymerization rate on the energy of the electrons bombardingsurface of the growing polymer  󿬁 lm. The speci 󿬁 c rate, i.e. the ratecoef  󿬁 cient of the polymerization can be expressed as an incrementof the thickness per unit of a current density at a given voltagebetween the cathode and the  󿬁 lm's surface, in Å A  1 cm 2 с  1 or innm A  1 cm 2 с  1 . Comparatively low rates of the polymerizationinduced by the electron bombardment, especially in the region of small energies, requires use of highly sensitive methods of mea-suring the  󿬁 lm's thickness increments. As the most adequatetechnique the ellipsometry was accepted (Azzam and Bashara,1977). Important advantages of this method are its non-destruc-tive character, the lack of impact on the object, high sensitivity andaccuracy of measuring the thickness ( o 0.1 Å). So the main ex-perimental problem was formulated as in-situ measuring growthrate of a polymer  󿬁 lm's thickness under electron bombardmentwith controlled energy within the range 0 – 20 eV which has to bedetermined with an acceptable accuracy (for instance, no morethan 0.3 eV) at each  󿬁 xed energy level. Substantial uncertainty inmeasuring energy is being caused by the charging of the polymersurface, which makes the potential of the surface inde 󿬁 nite anddiffering from the voltage between the cathode and the substrate.The problem was solved by a stabilization of the surface potentialusing the tracking system with dynamic condenser known also asthe Kelvin probe. Fundamentals and numerous cases of using theKelvin probe can be found in Zisman (1932), Bass et al. (1998) and Artamonov et al. (1974). The combination of simultaneous in-situmeasurements of the  󿬁 lm's thickness and its surface chargingproved very useful for interpretation of data obtained.General scheme of the setup is presented in Fig. 1. Let us ex-plain brie 󿬂 y functioning of the main units of this device.  2.1. Ellipsometry Principles and applications of the ellipsometry are describedwith comprehensive completeness in many publications includingmonographs, as, for instance (Azzam and Bashara, 1977). In ourexperiments the thickness of the polymer  󿬁 lm did not exceed30 nm. An accuracy of the thickness measurements was 0.1 nm atthe angle of incidence 45 ° .  2.2. Measurement and stabilization of a surface potential Charging surface underan electron 󿬂 ow distorts the energy of theincident electrons and  󿬁 nal spectral characteristics of the interaction.In order to take it into account the method was used of measuringthese deviations through an energy shift of the obtained spectrum(Bass et al., 1998; Bass and Sanche, 2003). But such approach is not deprived of some uncertainty because the error itself is an unknownfunction of energies of the incident electrons. This puts forward in-stead of the problem of the measurement of the surface potential itsstabilization during the irradiation. The problem of measurement of charging the surface under the electron bombardment and main-taining the surface potential at zero level can be solved with the helpof tracking system with dynamic condenser as a sensor. A condenseris called as dynamic if one of its plates vibrates so that the distancebetween the plates varies, for instance, according to the sine law. If the  󿬁 eld inside the capacitor is non-zero then the mechanical vi-brations of the plate will cause oscillations of the electric values -capacityof the condenserandchargeontheplates.Oscillationsof thecharge are equivalent to alternate current in the circuit shown inFig. 2,  I  ¼ d Q  /d t  ¼ (d C  /d t  )V  с . The current will be absent only if theelectric 󿬁 eld intensity (slope of the graph of potential in Fig. 2) is zeroin the region of the vibratingelectrode. This is achieved byapplying acompensating bias V  c  to the  󿬁 xed plate 1 (substrate). In the momentof the compensation the potentials of the  󿬁 lm surface and vibratingplate are both equal to zero. As seen in the diagram of potentials inFig. 2, the compensating bias  V  c  exactly equals to the voltage acrossthe 󿬁 lm. This enables obtaining volt-ampere characteristic of the 󿬁 lmright at the time of its growing under the electron beam.For the measurements and stabilization of the  󿬁 lm's surfacepotential the tracking system with 100% negative feedback hasbeen assembled with the sensor described above (Artamonov Fig. 1.  Scheme of the setup. 1  –  light source, 2  –  diaphragm, 3  –  lens, 4  –  󿬁 lter  λ ¼ 546 nm, 5  –  polarizer, 6  –  windows, 7  –  vibrating electrode (probe), 8  –  sub-strate, 9  –  electron gun, 10  –  plate  λ /4, 11  –  analyzer, 12  –  photomuliplier, 13  – galvanometer, 14  –  vacuum chamber. dL = L 0 + a·sinωt 23 VV=0 456897 Film V s V comp 1 Fig. 2.  Scheme of a dynamic condenser and compensation of the potential differ-ence between the vibrating plate (probe) and charged  󿬁 lm's surface by the appli-cation of the counter-bias to the substrate. 1  –  󿬁 xed plate of the condenser, 2  – vibrating probing plate, 3  –  amplitude of vibrations, 4  –  galvanometer, 5  –  powersource, 6  –  potential distribution formed by the charged surface of a dielectric  󿬁 lm,7  –  potential distribution in the mode of compensation, 8 - current through anuncompensated condenser, 9  –  zero current through the compensated condenser. V.I. Zyn / Radiation Physics and Chemistry 122 (2016) 82 – 92 84  et al., 1974). An alternate current through the condenser was usedto automatically control the compensating bias and lead the un-balance to a zero. Fixed electrode functioned as a substrate for the 󿬁 lm growing under the electron bombardment. The vibratingprobe was a tungsten wire 0.2 mm in diameter shielded from theelectron beam. The probe was located near the  󿬁 xed electrode andwas being set in motion by the electromagnetic vibrator. The planeof vibration makes the angle 45 °  to the plane of the  󿬁 xedelectrode.  2.3. Precursors A lot of research works on the electron impact polymerizationhave been carried out with vapors of vacuum oils as precursors,having in mind the role of this phenomenon in electron beamdevices with oil pumping (Christy, 1964; Hill, 1965; White, 1963; Woodman, 1965) Those substances have proved especially con-venient for studies of surface polymerization. They have a suitableworking vapor pressure which provides the characteristic time of adsorption on the surface of the order up to tens of seconds and atthe same time leaves vacuum deep enough for normal collisionlesspassage of the electron beams and prevention of any volumetricreactions. Some chemical uncertainty of the substances can beconsidered not too important as long as we are interested only ingeneral laws of the polymerization process or develop an experi-mental technique but we do not have the aim of achieving the bestor any preset chemical properties of the product. This makes thetransient processes of the adsorption – desorption-polymerizationat the surface relatively easy to observe by means of thicknessmeasurements. In this work we used the two oils  –  the hydro-carbons with molecular mass 450 Da (C 32 H 66 ) and the silicon va-cuum oil-ethylpolysiloxane with molecular mass 700 Da. Someexperiments were carried out with use of acetylene. The kineticlaws of the stimulated desorption – adsorption processes observedwere qualitatively identical in all the cases independently of thechemical nature of the substance. Though they naturally havesome quantitative differences. 3. Experimental results  3.1. Kinetics of transient adsorption – desorption processes Experimental researches of the kinetics of the surface processeswere carried out using the setup described above (Fig.1). Usage of several precursors required several vacuum and electrical regimes;in fact, every monomer needed its own mode. In the case of hy-drocarbons the required pressure was 2 – 3  μ Torr while acetyleneand polysiloxane required its partial pressure 1 – 2 orders of mag-nitude higher to provide the rate of the polymerization of givensubstance essentially higher than the rate of the polymerization of the residual hydrocarbon background. Energy of the electrons inthe beam was 0 – 20 eV. The electron current was about 0.1 – 0.3  μ Afor hydrocarbons and order of 10  μ A for other substances - ethyl-polysiloxane and acetylene. It gave electron  󿬂 ows within the range10 12 – 10 14 cm  2 c  1 .The polymerization was implemented at the glass substratewith deposited gold coating. To clean the surface the substrate wasannealed in vacuum 5  m Torr for several hours at a temperature200 – 250  ° С . Then the elements of elliptical polarization of re- 󿬂 ected light were measured and optical constants of the cleansurface '' n '' and '' k '' were calculated.A glass plate with deposited gold layer was an anode target forthe electrons and at the same time a substrate for the adsorptionof the precursors and growing polymer  󿬁 lm. With the help of theellipsometer the changes in the thickness of the organic  󿬁 lmgrown on the substrate were determined. The procedure of themeasurement was as follows. After pumping out the system andletting a precursor in all the electrodes are supplied with voltage,the potential-stabilizing system is being enabled and the  󿬁 rstpreliminary reading of the ellipsometer is being taken. Then theelectron gun opens and the target is subject to the electron irra-diation; after this the ellipsometer reading is taken once again.Similar steps are repeating until a full kinetic curve has formed.The experimental results are shown in Figs. 3 – 5. They are quali-tatively consistent with the common considerations about the surfacephenomena caused in the adsorption layer by the pulsed electronbombardment, and they will be used further in the Discussion part asa base for a formation of the quantitative theoretical model. As it isseen, all the three researched substances demonstrated similar tran-sient kinetics (Figs. 3 – 5). At the beginning of the bombardment the t   , min0.40.2012345-0.2-0.4∆ d   , nm Fig. 3.  Changing thickness of the  󿬁 lm after switching-on the electron beam. Ob-tained by the ellipsometry, accuracy 7 0.5 Å, hydrocarbons,  p ¼ 10  6 Torr, electronimpact 20 eV,  j ¼ 0.1  μ A cm  2 . 0123451.0 0 . 5123∆ d   , nm Fig. 4.  Saturation of the adsorption centers after stopping bombardment. Ellipso-metry, accuracy 0.5 Å, hydrocarbons,  p ¼ 10  5 Torr, electron impact 20 eV, currents:1 – 0.1  μ A, 2 – 0.06  μ A, 3 – 0.03  μ A. V.I. Zyn / Radiation Physics and Chemistry 122 (2016) 82 – 92  85  󿬁 lm thickness decreases sharply and 20 – 30s later it achieves mini-mum. The thickness here is 3.5Å less than the initial thickness of thelayer of adsorbed molecules. After passing this point the thicknessincreases 6Å relative to its minimal value. Since this moment a sta-tionary regime is establishedwhen the thicknesscontinuestoincreaseslowly and uniformly due to the continuing polymerization only.Switching-off electron beam makes the thickness to increase 10Åowing to saturation of free active centers of adsorption on the surface.Estimations made on the basis of the experimental results in accordwith the model described belowgave the following values of the crosssections of molecular attachment '' a '' and inelastic collisions '' q '': hy-drocarbons  –  a ¼ 0.1  10  17 cm 2 ,  q ¼ 1.1  10  15 cm 2 ; ethylpolysiloxane –  a ¼ 0.2  10  17 cm 2 ,  q ¼ 1.6  10  15 cm 2 .  3.2. Energy spectrum of the polymerization In research of a mechanism of any molecular process the mostimportant role belongs to energy or spectral dependencies of theprocess rates. In given case such a spectrum is energy distributionof the growth rate of the polymer  󿬁 lm. Obtaining of this was im-plemented with the help of the method described earlier in thisarticle. The residual vapor of the hydrocarbon vacuum oil was usedas a precursor. The substance is not a monomer in usual sense butit is easily polymerizable and has some other advantages. Al-though the installation was equipped with a system of inlet, butwhen using hydrocarbon vacuum oil as a working liquid, its usewas not required, and the vapor pressure were being establishedspontaneously during the pump working. Besides, the size of themolecules, tens of Angstroms, is quite enough for providing ob-servable rates of the  󿬁 lm growth even at low pressures 1 – 10  m Torr.One of the aims of this part is a demonstration of capabilities of presented approach for the providing necessary experimentalmaterial for research of the internal mechanism of this complexand in many aspects unclear process, which is typical for electronbeam devices. The energy of the electron bombardment was beingchanged by steps 0.5 eV each. Time of exposition at every step wasdetermined mainly by the current density corresponding to theaccelerating voltage, i.e. by the volt-ampere characteristic of thedevice. In practice, the expositions which ensured observable in-crease of the  󿬁 lm thickness were 10 – 20 min if the current densitywas several  m A cm  2 . The parameters of the elliptical polarizationwere measured before and after the exposition and they wereused for calculation of the  󿬁 lm thickness increase under theelectron impact during this given exposition. A relation of thisincrement to duration of the exposition is usually considered as arate of thickness growth. But in our opinion, more precise char-acteristic of the process is the speci 󿬁 c rate of the polymerization,i.e. the relation of the thickness increase to both the quantity of the electrons bombarding the target during this exposition and toits duration. This speci 󿬁 c rate of the  󿬁 lm growth was determinedas the thickness increment in Angstroms for a minute while thecurrent density is 1  m A cm  2 ,  R ¼ [( Δ d /(  j  t  )], Å cm 2 m A  1 min  1 .This rate of the  󿬁 lm growth coincides with the rate of the poly-merization only in the condition of dynamic equilibrium when therates of adsorption and desorption are balanced. In transient re-gimes these rates commonly do not coincide and one measuresthe rate of the  󿬁 lm growth only but not the rate of the poly-merization (Fig. 5). The measurements carried out within themethod described gave the results depicted in Fig. 6. On thebackground of common increase in the rate of the polymerizationthe function shows two large peaks witnessing about existence of resonant excitations of the molecules in the surface layer as theresult of the electron impact. Because this function  R  is propor-tional to the polymerization cross section  q , it was recalculatedinto corresponding units, cm 2 , which are presented on the rightaxis in Fig. 6.  3.3. Estimation of the cross section of polymerization Therelationshipbetweenapolymercrosssectionanda 󿬁 lmgrowthrate in stationary regime is given by formula (15). This connection hasthe simplest form in the case of small electron  󿬂 ow: the  󿬁 lm growthrate is proportional to the cross section of polymerization  q p ,  R ¼ q p  ν  e  F  , where  F   is number of monomer molecules adsorbed at 1cm 2 of asurface,  ν e  –  󿬂 ow of the electrons impinging the surface. The electroncurrents used in this work were small enough,  1 m A. Speci 󿬁 c  󿬁 lmgrowth rates also had corresponding order of magnitude,  1Åcm 2 / m А min. Ef  󿬁 ciency of the electrons as activators of the polymerizationis expressed as  K  ¼ n / ν  e , where  n  –  number of the molecules activatedby the electrons per1s on1cm 2 , ν  e - the electron 󿬂 ow, ν e ¼  j / e ,where  j –  the current density through the target,  e  –  electric charge of anelectron. While  j ¼ 1 m Acm -2 ,  ν e ¼ 0.625  10 13 cm  2 s  1 . The  󿬂 ow of molecules toward the substrate  ν þ ¼ n  u /4, where a molecular t   , min021120 2 ∆ d   , nm 468 10 3 Fig. 5.  Changes of the  󿬁 lm thickness after switching-on the electron beam. Ellip-sometry, accuracy 0.5 Å, electron impact 20 eV. 1  –  acetylene,  p ¼ 5  10  5 Torr,  j ¼ 7.5  μ A cm  2 . 2  –  ethylpoysiloxane,  p ¼ 2.10  4 Torr,  j ¼ 30  μ A cm  2 . 3-ethylpoysi-loxane,  p ¼ 4.10  4 Torr,  j ¼ 18  μ A cm  2 . р 02  4 681012141621086420Energy of electrons, eVσ(Е)·10 16 , cm 2 R, Å·cm 2 /(mcA·min) Fig. 6.  Rate (left scale) and cross section (right scale) of the polymerization asfunction of the energy of the electron bombardment. Current density is 1  m A/cm 2 inall the cases. V.I. Zyn / Radiation Physics and Chemistry 122 (2016) 82 – 92 86
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks