<%BANNER%>

Chemical Alteration of Poly (Tetrafluoroethylene) (TFE Teflon) Induced by Exposure to Hyperthermal Atomic Oxygen and Ult...


PAGE 1

CHEMICAL ALTERATION OF POLY (TETRAFLUOROETHYLENE) (TFE TEFLON ) INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN AND ULTRAVIOLET RADIATION By MICHAEL L. EVERETT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

This thesis is dedicated to Wayne and Kathleen Everett.

PAGE 3

ACKNOWLEDGMENTS I would like to thank Gar B. Hoflund, Jason F. Weaver, Helena Hagelin-Weaver and Oscar D.Crisalle for their guidance, their help, and their support throughout my time at the University of Florida. They have taught me a great deal and I am honored to have worked with them. I would also like to thank Seung-Hoon Oh and Bryan Fitzsimmons for their assistance and friendship. I would finally like to thank my parents, my sister, my entire family and all my friends for their support throughout the years. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES...............................................................................................................v LIST OF FIGURES...........................................................................................................vi ABSTRACT......................................................................................................................vii CHAPTER 1 INTRODUCTION........................................................................................................1 Overview.......................................................................................................................1 Literature Review.........................................................................................................1 O-Atom Source Characteristics....................................................................................3 Ultraviolet Source Characteristics................................................................................6 X-ray Photoelectron Spectroscopy...............................................................................6 2 CHEMICAL ALTERATION OF POLY (TETRAFLUOROETHYLENE) (TFE TEFLON ) INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN.....................................................................................................................9 Experimental.................................................................................................................9 Results and Discussion...............................................................................................10 Summary.....................................................................................................................17 3 CHEMICAL ALTERATION OF POLY (TETRAFLUOROETHYLENE) (TFE TEFLON ) INDUCED BY EXPOSURE TO ULTRAVIOLET RADIATION........27 Experimental...............................................................................................................27 Results and Discussion...............................................................................................28 Summary.....................................................................................................................33 Comparison Between AO and UV Exposure.............................................................34 REFERENCES..................................................................................................................43 BIOGRAPHICAL SKETCH.............................................................................................45 iv

PAGE 5

LIST OF TABLES Table page 2-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%)............24 2-2. Near-surface Compositions of Tefzel after Various Treatments (at%).....................25 2-3. Near-surface Compositions of Tedlar after Various Treatments (at%).....................26 3-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%)............42 v

PAGE 6

LIST OF FIGURES Figure page 1-1. ESD atomic oxygen source..........................................................................................8 2-1. XPS survey spectra obtained from TFE Teflon.........................................................19 2-2. High-resolution XPS obtained from TFE Teflon......................................................20 2-3. XPS survey spectra obtained from TFE Teflon.........................................................21 2-4. XPS survey spectra obtained from TFE Teflon.........................................................22 2-5. XPS survey spectra obtained from TFE Teflon.........................................................23 3-1. XPS survey spectra obtained from TFE Teflon.........................................................36 3-2. High-resolution XPS (A) C 1s and (B) F 1s obtained from TFE Teflon...................37 3-3. XPS survey spectra obtained from TFE Teflon.........................................................38 3-4. Overlays of XPS C 1s spectra....................................................................................39 3-5. XPS survey spectra obtained from UV-exposed TFE Teflon...................................40 3-6. High-resolution XPS C 1s spectra obtained from AO-exposed (dashed line) and UV-exposed (solid line) TFE Teflon...............................................................................41 vi

PAGE 7

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CHEMICAL ALTERATION OF POLY (TETRAFLUOROETHYLENE) (TFE TEFLON ) INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN AND ULTRAVIOLET RADIATION By Michael L. Everett August 2004 Chair: Gar B. Hoflund Major Department: Chemical Engineering In this study the erosion of poly (tetrafluoroethylene) (TFE Teflon) by hyperthermal atomic oxygen (AO) and ultraviolet radiation (UV) (115-400 nm) has been examined using X-ray photoelectron spectroscopy (XPS). The initial F/C atom ratio of 1.66 decreases to 1.15 after a 2-hr exposure to a flux of 2 x 10 15 atoms/cm 2 -s AO with an average kinetic energy of 5 eV. The F/C atom ratio is further reduced to a value of 0.90 after a 25-hr exposure. The high-resolution XPS C 1s data indicate that new chemical states of carbon form as the F is removed, and that the relative amounts of these states depend upon the F content of the near-surface region. The states are most likely due to C bonded only to one F atom and C bonded only to other C atoms. Exposures of the AO-damaged surface to O 2 result in chemisorption of a very small amount of O (~0.8 at%) indicating that large quantities of reactive sites are not formed during the chemical erosion by AO. Further exposure to AO removes this chemisorbed oxygen. After exposing the AO-exposed surface to air for 90 min, both O and H 2 O are chemisorbed, vii

PAGE 8

and the F/C at ratio is reduced to 0.68. Another 46 hrs of AO exposure results in removal of these species and a further decrease in the F/C at ratio to 0.58. The initial F/C atom ratio of 1.98 decreases to 1.65 after a 2-hr exposure to UV radiation. The F/C atom ratio is further reduced to a steady-state value of 1.60 after a 74-hr exposure. The high-resolution XPS C 1s data indicate that new chemical states of carbon form as the F is removed, and that the relative amounts of these states depend upon the F content of the near-surface region. The states are most likely due to C bonded only to one F atom, C bonded only to other C atoms and C which have lost a pair of electrons through emission of F Exposures of the UV-damaged surface to O 2 result in chemisorption of a very small amount of O indicating that large quantities of reactive sites are not formed during the chemical erosion by UV. Further exposure to UV removes this chemisorbed oxygen. Comparison of XPS data indicates that the mechanisms of chemical alteration by UV radiation and hyperthermal atomic oxygen (AO) are different as expected. viii

PAGE 9

CHAPTER 1 INTRODUCTION Overview A novel electron stimulated desorption (ESD) hyperthermal atomic oxygen (AO) source as well as an ultraviolet (UV) radiation lamp were used to induce chemical changes to the surface of a TFE Teflon film. The changes were studied by X-ray photoelectron spectroscopy (XPS). The goal of this study is to examine and attempt to understand the chemical alterations of a TFE Teflon surface when impacted by hyperthermal (~5 eV) AO and UV radiation, as it may occur in low earth orbit (LEO). The average energy and energy spread of the AO from this source are similar to AO in LEO. Also, the UV lamp used covers the range of UV radiation that occurs in LEO, vacuum ultraviolet (VUV) (115-200 nm). There are two types of bonds in TFE Teflon; C-C bonds and C-F bonds with bond strengths of approximately 83 and 111 kcal/mole, respectively. The UV radiation used in this study is energetic enough to break these bonds. Literature Review Polymers are attractive and desirable materials for use in space applications because they are lightweight and typically much easier to process using techniques such as extrusion, casting and injection molding at relatively low temperatures compared to metals and ceramics. They also tend to be more flexible and offer a wide variety of choices from optically transparent to opaque, rubbery to stiff and conducting to 1

PAGE 10

2 insulating. TFE Teflon is one material that is widely used as a thermal blanket for spacecraft flying in low earth orbit (LEO) [ 1 ]. Thermal control is provided by lining the TFE Teflon with aluminum or silver. The TFE Teflon has a high thermal emittance and this system reflects a large fraction of the incident solar energy. However, over the last two decades, it has been well established that polymers undergo severe degradation resulting in reduced spacecraft lifetimes. These materials degrade because spacecraft surfaces are exposed to high fluxes of atomic oxygen (AO), bombardment by lowand high-energy charged particles, thermal cycling and the full spectrum of solar radiation. AO is the main constituent of the atmosphere in LEO. It is formed by dissociation of molecular oxygen by ultraviolet radiation from the sun, resulting in an AO concentration of approximately 10 8 atoms/cm 3 The reverse reaction in which an oxygen molecule forms from AO does not have a high reaction rate because it requires a teratomic collision. The third atom is required to dissipate the energy released by formation of O 2 The actual flux of ~10 14 atoms/cm 2 -s impinging on a spacecraft is high due to orbiting speeds of approximately 8 km/s. At these relative speeds thermal AO collides with a kinetic energy of ~4.5 eV. These highly energetic collisions not only result in surface chemical reactions but can also lead to a pure physical sputtering of the surface atoms in the absence of any chemical changes. Many studies have been conducted in an effort to determine the mechanism of this degradation primarily caused by surface reactions with AO [ 29 ]. However, these studies have all been carried out after exposing these highly reactive surfaces to air prior to analysis, thus introducing artifacts that do not represent the true space environment. Recent studies have shown that exposure to air chemically alters the reactive surfaces formed during AO

PAGE 11

3 exposure [ 1013 ]. It is, therefore, essential that analysis of polymers exposed to AO be carried out in-situ to avoid artifacts induced by air exposure. Several studies have been conducted on the deterioration of fluorinated polymers retrieved from spacecraft subjected to the LEO environment. The outer layer of Teflon-fluorinated ethylene propylene (FEP) multi-layer insulation on the Hubble space telescope (HST) was significantly cracked at the time of the second HST servicing mission, 6.8 years after it was launched into LEO [ 14, 15 ]. Comparatively minor embrittlement and cracking were also observed in the FEP materials retrieved from solar-facing surfaces on the HST at the time of the first servicing mission (3.6 years of exposure). Furthermore, an increased deterioration of fluorinated polymers may result from the synergistic effect of VUV radiation in the presence of AO [ 16 ]. This point remains to be demonstrated. Intuitively low-energy AO would not be expected to break these bonds, and this has been found experimentally [ 4 ]. Gindulyte, Massa, Banks and Miller [ 17 ] have carried out a quantum mechanical study on the erosion of TFE Teflon by AO. They conclude that AO in LEO possesses enough translational energy to erode TFE Teflon by breaking C-C bonds. They did not consider breakage of C-F bonds based on electronegativity arguments. O-Atom Source Characteristics A photograph and schematic diagram illustrating the operational principles of the ESD AO source are shown in figures 1-1a and 1-1b respectively. Ultrahigh-purity molecular oxygen dissociatively adsorbs on the high-pressure (2 Torr) side of a thin metallic Ag-alloy membrane maintained at elevated temperature (~400C) and permeates

PAGE 12

4 through the membrane to the UHV side. There the chemisorbed atoms are struck by a directed flux of primary electrons, which results in ESD of the O atoms forming a continuous flux. The primary electrons are produced by thermionic emission from a coiled hot filament supported around the perimeter of the membrane. An electron reflector (lens assembly) surrounds the filament. It produces a potential field, which creates a uniform flux of electrons over the membrane surface. These primary electrons have a kinetic energy of 1000 eV and provide two functions: ESD of the O atoms and heating of the membrane surface. Another lens is placed between the reflector and the sample for removal of all charged particles including secondary electrons and O + and O ions produced during the ESD process. Several processes have to function in series at sufficiently high rates for this system to work, including dissociative chemisorption of the molecular gas on the metal surface, permeation of atomic oxygen through the membrane and formation of the neutral flux by ESD. Since these processes occur in series, the slowest one determines the magnitude of the AO flux. The sticking coefficient of O 2 on polycrystalline Ag (step 1) is fairly small (< 0.001) so it is necessary to use a high pressure on the upstream side of the membrane. However, the permeation rate through the membrane is proportional to the reciprocal of the membrane thickness. This means that it is desirable to have a high pressure and a thin membrane, but this can lead to membrane failure. The ESD rate can be increased by increasing the primary electron current to the membrane, but this increases the temperature of the membrane and can result in evaporation of Ag, which is unacceptable.

PAGE 13

5 The AO produced by this source has been shown to be hyperthermal (energies greater than 0.01 to 0.02 eV), but the neutral energy distribution has not been measured. Corallo et al. [ 18 ] have measured the energy distribution of O ions emitted by ESD from a Ag(110) surface and found that this distribution has a maximum of ~5 eV and a full-width at half maximum of 3.6 eV. This ion energy distribution sets an upper bound for the neutral energy distribution because ESD neutrals are generally believed to be less energetic than ESD ions based on models of the ESD process. This point has been discussed often in the ESD literature but not actually demonstrated. Since neutral ESD species are difficult to detect, very few ESD studies of neutral species have appeared in the literature. The neutral atom flux has been detected by using a quadrupole mass spectrometer [ 19, 20 ] in the appearance potential (AP) mode to allow the atoms to be distinguished from residual gases and background gas products formed by collisions of the neutrals with the walls of the UHV system. In this experiment the ion acceleration potential was set at 0.0 V. Calibration studies have demonstrated that the ions entering the quadrupole section must have a minimum kinetic energy of 2.0 eV to reach the detector so the ESD neutrals detected have a minimum energy of 2.0 eV. Therefore, the hyperthermal AO produced by this ESD source have energies greater than 2 eV but possibly less than the ion energy distribution. Furthermore, these mass spectrometric experiments have shown that the AO-to-O + ratio is about 10 8 and that the O + -to-O ratio is about 100. Several approaches have been used to measure the magnitude of the hyperthermal AO flux and reasonable agreement was obtained between the various methods. The flux from the ESD AO source is approximately 2x10 15 atoms/cm 2 -s. One of the most reliable

PAGE 14

6 methods for flux determination is the measurement of a ZrO 2 film growth rate [ 21 ]. A Zr flux was generated by e-beam evaporation and the flux was calibrated using a quartz-crystal monitor. Based on the facts that stoichiometric ZrO 2 was produced and that no O 2 is present in the AO flux, the AO flux was calculated. By doubling the Zr flux, stoichiometric ZrO was grown [ 22 ]. The AO flux has also been determined by measuring the chemisorption rate of AO on polycrystalline Au using ion scattering spectroscopy [ 23 ]. The flux determined using this method is in excellent agreement with that determined using the oxide growth rate method. Ultraviolet Source Characteristics The UV source used in this study was a Hamamatsu water-cooled deuterium lamp (model L1835). It had an MgF 2 window allowing UV light with wavelengths ranging from 115 to 400 nm to pass through to the sample in a 2.5 mm diameter aperture. The aperture was approximately 15 cm from the sample and the lamp operated at a power of 150 W. X-ray Photoelectron Spectroscopy X-ray photoelectron spectroscopy (XPS), also known as electron spectroscopy for chemical analysis (ESCA), is a photoemission technique used to determine the composition and chemical state distribution of solid surfaces. It is performed by irradiating the surface with soft x-rays in vacuo and analyzing the emitted electrons by energy. An electrostatic charged particle energy analyzer is used to obtain the spectral peaks generated from the kinetic energies of the emitted electrons. The corresponding binding energies specific to each individual element are then calculated from the following equation: E b = h E k +

PAGE 15

7 where E b is the binding energy (BE) in the solid, E k is the kinetic energy of the emitted electron and is the work function difference between the sample and the detector material assuming there is no electrical charging at the sample surface.

PAGE 16

8 Figure 1-1. ESD atomic oxygen source (a) Photograph of ESD AO source. (b) Schematic illustrating operational principles of ESD AO source.

PAGE 17

CHAPTER 2 CHEMICAL ALTERATION OF POLY (TETRAFLUOROETHYLENE) (TFE TEFLON ) INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN One of the main constituents in LEO is hyperthermal AO, which is known to degrade polymers. In this study XPS was used to characterize the surface of a TFE Teflon film before and after exposures of varying lengths to AO, as well as O 2 and air. The goal of the study was to determine how hyperthermal AO affects the TFE Teflon surface so that a better understanding of its space survivability can be obtained. Experimental An as-received E.I. du Pont Nemours & Co., Inc. TFE Teflon film was wiped with methanol and inserted into the UHV chamber (base pressure <1.0x10 -10 Torr). XPS measurements were performed using a double-pass cylindrical mirror analyzer (DPCMA) (PHI Model 25-270AR). XPS survey spectra were taken in the retarding mode with a pass energy of 50 eV, and high-resolution XPS spectra were taken with a pass energy of 25 eV using Mg K X-rays (PHI Model 04-151 X-ray source). Data collection was accomplished using a computer interfaced, digital pulse counting circuit [ 24 ] followed by smoothing with digital-filtering techniques [ 25 ]. The sample was tilted 30 degrees off the axis of the DPCMA, and the DPCMA accepted electrons emitted into a cone 42.6 6 degrees off the DPCMA axis. XPS spectra were first obtained from the as-entered, solvent-cleaned sample. The sample was then transferred into an adjoining UHV chamber that houses the ESD AO source via a magnetically coupled rotary/linear manipulator. There the surface was 9

PAGE 18

10 exposed to a hyperthermal AO flux and re-examined after various exposure times. The sample was not exposed to air after the AO exposures and before collecting XPS data. However, after the AO exposures the surface was exposed to 150 Torr O 2 in another adjoining chamber or to air to determine how this would affect the AO-exposed surfaces. The order of the TFE Teflon exposures to AO was as follows: 2 hrs, 22 hrs (24 hrs total), 20 min O 2 at 150 Torr, 1 hr (25 hrs total), 3 hrs (28 hrs total), 90min air (75C, 45% relative humidity), 46 hrs (74 hrs total). The approximate normal distance between the sample face and source in this study was 15 cm, at which distance the flux was about 2.0 x 10 15 atoms/cm 2 -s for the instrument settings used. The sample was maintained at room temperature during the AO exposures with a temperature increase to about 50C due to exposure to the X-ray source during XPS data collection. Results and Discussion XPS survey spectra obtained from the TFE Teflon film before and after a 2-hr exposure are shown in figures 2-1a and 2-1b respectively. Spectrum a is identical to that shown in the polymer XPS handbook by Beamson and Briggs [ 26 ] in that the C 1s, F 1s, F Auger (KLL) and F 2s peaks are present. An O 1s peak, which would appear near a binding energy (BE) of 530 eV, is not present. Estimates of the near-surface compositions have been made from the peak areas in the high-resolution spectra using published atomic sensitivity factors [ 27 ] with the assumption of a homogeneous surface region. XPS probes the near-surface region of the sample and yields a weighted average composition with the atomic layers near the surface being weighted more heavily since the photoemitted electrons from these layers have a lower probability of scattering inelastically. The sampling depth is ~4-6 nm, and ~10% of the signal originates from the

PAGE 19

11 outermost atomic layer [ 28 ]. This near-surface region is nonhomogeneous because the AO reacts most strongly with the outermost few atomic layers. Therefore, the region that reacts to the greatest extent with AO also makes the largest contribution to the XPS signal. This fact implies that XPS is an excellent technique for studying AO erosion of spacecraft materials. Even though the distribution functions involving the depth of chemical reactions in the near-surface region and the XPS determination of the weighted average composition of the near-surface region are complex, the compositional values provide a trend which is indicative of the chemical alterations occurring during AO exposure. The compositions determined using the homogeneous assumption are shown in table 2-1 before and after various exposures to AO and O 2 or air. The F/C atom ratio obtained from the as-entered sample is 1.66 which is lower than the stoichiometric value of 2.0. This is probably due to inaccuracies in the sensitivity factors used. After a 2-hr AO exposure (survey spectrum shown in figure 2-1b ), the F/C atom ratio is decreased from 1.66 to 1.15: i.e., over 30% of the F is removed from the near-surface region by this short exposure. The F features do not appear to change shapes or positions in the survey spectra obtained before and after the 2-hr AO exposure. However, the shape of the C 1s peak is significantly altered. This is more apparent in the high-resolution spectra shown in figure 2-2 The C 1s and F 1s features obtained from the as-entered TFE Teflon are shown in figure 2-2Aa and 2-2Ca respectively and in figure 2-2Ab and 2-2Cb for the surface exposed to AO for 2 hrs. The C 1s feature obtained from the as-entered TFE Teflon consists of a single, narrow peak with a BE of 292.5 eV. The C 1s feature obtained after the 2-hr AO exposure is quite broad and complex consisting of contributions from at least

PAGE 20

12 three different species with BEs of 292.5, 290.0 and 288.0 eV. Assigning the peaks to specific species is quite difficult for several reasons. Defining localized species which yield features at specific BEs is quite difficult. For example, species are present in both TFE Teflon and Tefzel ||FC||HCH F (ethylene tetrafluoroethylene, ETFE) but the C 1s BEs are 292.48 and 290.9 eV respectively. Also, species are present in Tefzel and Tedlar (Poly (vinyl fluoride), PVF) with BEs of 286.44 and 285.74 eV respectively. XPS BEs result from the chemical environment around the given element. This fact implies that the chemical environment is larger than the species defined in TFE Teflon, Tefzel and Tedlar. Since there are distinct peaks in these spectra, there are distinct chemical environments, but defining the nature of this chemical environment is not possible. Furthermore, the hyperthermal AO-exposed surface is a damaged surface in that the composition is altered to a great extent, and the structure is most likely altered as discussed below. The new features which appear in the C 1s spectrum after the 2-hr AO exposure have lower BEs indicating that the chemical environments responsible for these features are F depleted. This assertion is consistent with the compositional data shown in table 2-1 The BE of the F 1s peak ( figure 2-2Cb ) remains unchanged, but the peak is slightly broadened to the high-BE side. This indicates that the F remaining in the near-surface region more strongly attracts electrons. An XPS survey spectrum obtained from the TFE Teflon surface after a 24-hr AO exposure is shown in figure 2-3a The increased AO exposure results in a decrease in the F/C atom ratio from 1.15 to 0.90 ( table 2-1 ), and the C 1s peak shape is changed

PAGE 21

13 significantly from that after the 2-hr AO exposure ( figure 2-1b ). This is more apparent in the C 1s spectrum shown in figure 2-2Ac The feature with a BE of 288.0 eV is now most prominent and the feature due to TFE Teflon at a BE of 292.5 eV is the least prominent. These results demonstrate that hyperthermal AO exposure destroys the TFE Teflon structure by removing F from the near-surface region. The two states at BEs of 288.0 and 290.0 eV are due to F-depleted regions. The presence of lower concentrations of F implies that electron density on the C atoms is not decreased to such a large extent resulting in increased C 1s BEs. The feature at 288.0 eV may be due to regions where all of the F has been removed. The corresponding F 1s feature is shown in figure 2-2Cc This feature is shifted from 689.3 to 690.0 eV. The increased BE indicates that the remaining F is able to attract and bind electrons more strongly from the damaged matrix as the F concentration is decreased. Next the 24-hr, AO-exposed TFE Teflon surface was exposed to O 2 for 20 min at room temperature and 150 Torr. The XPS survey spectrum obtained from this surface is shown in figure 2-3b There appears to be negligible differences in the sizes and shapes of the C and F features before and after the O 2 exposure in these survey spectra, but a small peak due to O is apparent at a BE of approximately 535 eV. The fact that this surface dissociatively chemisorbs molecular oxygen implies that reactive sites are present on the AO-exposed TFE Teflon surface. The amount of O 2 that chemisorbs provides a measure of the concentration of reactive sites in the near-surface region. Previous studies have shown that AO-exposed Kapton [ 10 ], Tedlar [ 29 ] and Tefzel [ 30 ] chemisorb very large amounts of molecular oxygen indicating that these surfaces contain a high concentration of reactive sites. These reactive sites are probably present because cross

PAGE 22

14 bonding cannot occur due to geometrical constraints. The amount of molecular oxygen chemisorbed on AO-exposed TFE Teflon is very small indicating that the concentration of reactive species is quite small. The data in table 2-1 indicates that much of the F initially present is removed by the AO exposure. Since very little molecular oxygen chemisorbs, the broken bonds formed by AO exposure must cross link in TFE Teflon. Although the peak shapes are similar before and after the O 2 exposure in the survey spectra, quite significant differences are apparent in the high-resolution spectra ( figure 2-2Ad and 2-2Cd ). Specifically, the feature due to TFE Teflon at 292.5 eV is increased in intensity relative to the F-depleted state at 288.0 eV. Since the AO reacts most strongly with TFE Teflon right at the surface, the AO-exposed surface is most likely a layered structure with the most F-depleted state at the surface (288.0 eV), the next most F-depleted state just beneath (290.0 eV) and the TFE Teflon structure (292.5 eV) beneath these two layers. A possibility is that the O 2 reacts with the F-depleted layer at the surface to form CO or CO 2 This removes part of the top layer making the underlying two layers more prominent in the XPS spectra. This also explains why the oxygen content of the O 2 -exposed surface is so low (< 1 at%). The outermost layer is a damaged layer with many vacancies and defects. This would make it quite reactive chemically. The high-resolution O 1s feature is shown in figure 2-2Bd It is narrow indicating that only one chemical state of oxygen is present. The fact that this peak is present demonstrates that the O 2 bond can be broken at the AO-exposed TFE Teflon surface under the O 2 -exposure conditions used. The corresponding F 1s feature is shown in figure 2-2Cd It is shifted toward the BE of F in TFE Teflon and is broadened. This

PAGE 23

15 indicates that this feature is composed of contributions from F in the middle F-depleted layer and F in TFE Teflon. The O 2 -exposed surface was then exposed to AO for 1 hr (total AO exposure of 25 hrs), and the resulting XPS survey spectrum is shown in figure 2-4a The O 1s peak is not present indicating that AO removes chemisorbed O from the O 2 -exposed, AO-exposed TFE Teflon surface. The compositional data in table 2-1 indicate that the near-surface O concentration decreases from 0.8 to 0.1 at% by the AO exposure. The mechanism of O removal may be physical sputtering or chemical reaction to form O 2 or CO 2 The F/C atom ratio is again decreased from 0.895 to 0.816. This is less than one-half of the F initially present. More information is provided by the high-resolution spectra shown in figure 2-2 In the C 1s spectrum ( figure 2-2Ae ), the F-depleted near-surface region again yields the most prominent feature as before the O 2 exposure ( figure 2-2Ac ). The peak due to the TFE Teflon structure is reduced in magnitude but not to that before the O 2 exposure. The O 1s feature shown in figure 2-2Be is essentially at the noise level indicating that the chemisorbed O is removed by the AO exposure. The F 1s feature ( figure 2-2Ce ) is not changed significantly by the 1-hr AO exposure although it may have shifted slightly back toward higher BEs. Another 3-hr exposure to AO results only in small changes which are not apparent in the survey spectrum shown in figure 2-4b According to the data in table 2-1 the F/C atom ratio remains unchanged in the near-surface region but the last small amount of chemisorbed O is removed. Changes in the high-resolution C 1s and F 1s features shown in figures 2-2Af and 2-2Cf respectively are also small. The F 1s peak is shifted a few tenths of an eV toward higher BEs which is characteristic of larger AO exposures.

PAGE 24

16 The surface was then exposed to air for 90 min at room temperature. This treatment is quite different than an exposure to research purity O 2 because air contains water, hydrocarbons, alcohols and CO 2 which can adsorb at a surface. The XPS survey spectra obtained after this treatment is shown in figure 2-5a A small O 1s peak is apparent which is larger than the O 1s peak obtained after the O 2 exposure ( figure 2-3b ). This is consistent with the compositional data in table 2-1 that gives an oxygen concentration of 1.9 at%. The O 2 exposure does not alter the F/C atom ratio while the air exposure lowers the F/C atom ratio from 0.82 to 0.68 due to both an increase in the C concentration and a decrease in the F concentration. This result is most likely due to several processes. Reactive C near the surface would certainly react with oxygen in the air to form CO or CO 2 This process removes C from the surface. Hydrocarbons and possibly alcohols in air adsorb on the surface. This enhances the intensity of the C signal and decreases the intensity of the F signal. These assertions are consistent with the high-resolution C 1s, O 1s and F 1s features shown in figure 2-2Ag, 2-2Bg and 2-2Cg respectively. In the C 1s spectrum, the feature due to hydrocarbons at 288.0 eV is predominant while the peak due to TFE Teflon at 292.5 eV is fairly small and broad. The shape of the O 1s feature is quite different than that obtained after the 20-min O 2 exposure ( figure 2-2Bd ) in that it contains a shoulder at a BE about 2 eV higher than the predominant feature. This shoulder is assigned to adsorbed water from the air. If this contribution is subtracted from the O 1s feature, the magnitudes of the chemisorbed O are similar for the O 2 exposure and the air exposure. The F 1s peak is shifted to the BE characteristic of an AO-exposed surface. This is due to the F in the middle layer which makes a large contribution to the F 1s signal.

PAGE 25

17 The final treatment was an exposure to the AO flux for another 46 hrs (74 hrs total). This exposure is long enough to represent a steady state in which the surface continually erodes away at a constant rate but no further changes are observed in the surface chemistry. The survey spectrum obtained after this AO exposure is shown in figure 2-5b The O 1s feature is not present indicating that AO removes surface O as found after the O 2 exposure. The F/C atom ratio ( table 2-1 ) is decreased to a value of 0.58. The series of exposures used in this study results in removal of approximately two-thirds of the F in the near-surface region most likely by a physical sputtering mechanism. This is the first step in the AO-erosion of TFE Teflon. An F-depleted C layer is left at the surface which erodes more slowly. The corresponding high-resolution C 1s and F 1s spectra are shown in figure 2-2Ah and 2-2Ch respectively. These spectra are most similar to those obtained after the 24-hr AO exposure. The outermost F-depleted C layer is predominant, and the top two F-depleted layers are so thick that the TFE Teflon peak at 292.5 eV is quite small. Based on mean-free-path arguments the F-depleted layer is probably about 3 nm thick. The F 1s peak has a BE of 690.0 eV which is characteristic of an AO-exposed TFE Teflon surface. Hence, the TFE Teflon structure makes only a small contribution to this surface. Summary In this study the chemical alterations at a TFE Teflon surface caused by exposure to hyperthermal AO have been studied using XPS. A 2-hr exposure to AO results in a decrease in the F/C atom ratio from 1.66 to 1.15 and formation of two new carbon chemical states assigned as C bonded to only one F and C bonded to other C. Another 22 hrs of AO exposure results in a further decrease of the F/C atom ratio to 0.90 and an increase in the concentration of the two new carbon states with the C-bondingto-C state

PAGE 26

18 predominating. Exposure of this AO-exposed surface to O 2 results in a small amount (0.8 at%) of chemisorbed O due to bonding at reactive sites. The fact that so little O 2 chemisorbs indicates that the surface carbon bonds to other carbon which is consistent with the formation of the new C chemical states. The chemisorbed O apparently is present in only one chemical state. Further exposure of this surface to AO for 3 hrs results in removal of the chemisorbed oxygen and a further reduction in the F/C atom ratio to 0.82. Then this AO-exposed surface was exposed to air. Both O 2 and H 2 O are chemisorbed in small quantities (1.9 total O at%). Another 46 hrs of AO exposure results in removal of the chemisorbed O and water and a further reduction in the F/C atom ratio to 0.58.

PAGE 27

19 Figure 2-1. XPS survey spectra obtained from TFE Teflon (a) as entered and (b) after a 2-hr exposure to AO.

PAGE 28

20 Figure 2-2. High-resolution XPS obtained from TFE Teflon (A) C 1s, (B) O 1s, and (C) F 1s (a) as entered, (b) after exposure to AO for 2 hrs, (c) after exposure to AO for a total of 24 hrs, (d) after exposure to O2 at room temperature for 20 min, (e) after exposure to AO for a total of 25 hrs, (f) after exposure to AO for a total of 28 hrs, (g) after exposure to air for 90 min at room temperature and (h) after exposure to AO for a total of 74 hrs.

PAGE 29

21 Figure 2-3. XPS survey spectra obtained from TFE Teflon (a) after a 24-hr exposure to AO and (b) after a 20-min exposure to O2 at room temperature and 150 Torr.

PAGE 30

22 Figure 2-4. XPS survey spectra obtained from TFE Teflon (a) after a 25-hr exposure to AO (1-hr AO exposure after the O2 exposure), and (b) after 28 hrs total AO exposure.

PAGE 31

23 Figure 2-5. XPS survey spectra obtained from TFE Teflon (a) after a 90-min exposure to air at room temperature and (b) after a total of 74 hrs exposure to AO.

PAGE 32

24 Table 2-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%) F O C F/C O/C as entered 62.4 0.0 37.6 1.66 0.0 AO 2 hrs 53.6 0.0 46.4 1.15 0.0 AO 24 hrs 47.4 0.0 52.6 0.90 0.0 O 2 20 min 46.8 0.8 52.4 0.90 0.015 AO 25 hrs 44.9 0.1 55.0 0.81 0.002 AO 28 hrs 45.2 0.0 54.8 0.82 0.0 Air 90 min 39.8 1.9 58.3 0.68 0.03 AO 74 hrs 36.6 0.0 63.4 0.58 0.0

PAGE 33

25 Table 2-2. Near-surface Compositions of Tefzel after Various Treatments (at%) F O C F/C O/C as entered 41.5 2.3 56.2 0.74 0.04 AO 2 hrs 14.5 0.7 84.8 0.17 0.01 AO 24 hrs 1.9 3.5 94.7 0.02 0.04 O 2 20 min 3.7 7.2 89.1 0.04 0.08 AO 25 hrs 1.7 6.8 91.5 0.02 0.07 AO 28 hrs 1.2 4.9 93.9 0.01 0.05

PAGE 34

26 Table 2-3. Near-surface Compositions of Tedlar after Various Treatments (at%) F O C F/C O/C as entered 29.0 7.2 63.7 0.45 0.11 AO 2 hrs 1.7 4.0 94.2 0.018 0.04 AO 24 hrs 2.0 6.3 91.7 0.022 0.071 O 2 20 min 1.1 10.4 88.5 0.012 0.12 AO 25 hrs 1.3 7.0 91.7 0.014 0.08 AO 28 hrs 1.7 6.7 91.6 0.018 0.07

PAGE 35

CHAPTER 3 CHEMICAL ALTERATION OF POLY (TETRAFLUOROETHYLENE) (TFE TEFLON ) INDUCED BY EXPOSURE TO ULTRAVIOLET RADIATION In LEO objects are subjected to the full spectrum of solar radiation. One such range of light is VUV (115-200 nm), which is known to degrade polymers. In this study XPS was used to characterize the surface of a TFE Teflon film before and after exposures of varying lengths to UV radiation (115-400 nm), as well as O 2 and air. The goal of the study was to determine how hyperthermal UV affects the TFE Teflon surface so that a better understanding of its space survivability can be obtained. Experimental An as-received E.I. du Pont Nemours & Co., Inc. TFE Teflon film was wiped with methanol and inserted into the UHV chamber (base pressure <1.0x10 -10 Torr). XPS measurements were performed using a double-pass cylindrical mirror analyzer (DPCMA) (PHI Model 25-270AR). XPS survey spectra were taken in the retarding mode with a pass energy of 50 eV, and high-resolution XPS spectra were taken with a pass energy of 25 eV using Mg K X-rays (PHI Model 04-151 X-ray source). Data collection was accomplished using a computer interfaced, digital pulse counting circuit [ 24 ] followed by smoothing with digital-filtering techniques [ 25 ]. The sample was tilted 30 degrees off the axis of the DPCMA, and the DPCMA accepted electrons emitted into a cone 42.6 6 degrees off the DPCMA axis. XPS spectra were first obtained from the as-entered, solvent-cleaned sample. The sample was then transferred into an adjoining UHV chamber that houses the UV source 27

PAGE 36

28 via a magnetically coupled rotary/linear manipulator. There the surface was exposed to UV radiation and re-examined after various exposure times. The samples were not exposed to air after the UV exposures and before collecting XPS data. However, after the UV exposures the surfaces were exposed to O 2 or air to determine how this would affect the UV-exposed surfaces. The order of the TFE Teflon exposures to UV radiation was as follows: 2 hrs, 22 hrs (24 hrs total), 20 min O 2 at 150 Torr, 1 hr (25 hrs total), 3 hrs (28 hrs total), 90min air (82C, 45% relative humidity), 46 hrs (74 hrs total). The approximate normal distance between the sample face and source in this study was 15 cm. The sample was maintained at room temperature during the UV exposures with a temperature increase to about 50C due to exposure to the X-ray source during XPS data collection. Results and Discussion XPS survey spectra obtained from the TFE Teflon film before and after a 2-hr UV exposure are shown in figure 3-1a and 3-1b respectively. Spectrum a is identical to that shown in the polymer XPS handbook by Beamson and Briggs [ 26 ] in that the C 1s, F 1s, F Auger (KLL) and F 2s peaks are present. An O 1s peak, which would appear near a binding energy (BE) of 530 eV, is not present. Estimates of the near-surface compositions have been made from the peak areas in the high-resolution spectra using published atomic sensitivity factors [ 27 ] with the assumption of a homogeneous surface region. XPS probes the near-surface region of the sample and yields a weighted average composition with the atomic layers near the surface being weighted more heavily since the photoemitted electrons from these layers have a lower probability of scattering inelastically. The sampling depth is ~4-6 nm, and ~10% of the signal originates from the outermost atomic layer [ 28 ]. This near-surface region is nonhomogeneous because the

PAGE 37

29 UV radiation reacts most strongly with the outermost few atomic layers. Therefore, the region that reacts to the greatest extent with UV radiation also makes the largest contribution to the XPS signal. This fact implies that XPS is an excellent technique for studying erosion of spacecraft materials by UV radiation and atomic oxygen. Even though the distribution functions involving the depth of chemical reactions in the near-surface region and the XPS determination of the weighted average composition of the near-surface region are complex, the compositional values provide a trend which is indicative of the chemical alterations occurring during UV exposure. The compositions determined using the homogeneous assumption are shown in table 3-1 before and after various exposures to UV radiation and O 2 or air. The F/C atom ratio obtained from the as-entered sample is 1.98, which is very close to the stoichiometric value of 2.0. After a 2-hr UV exposure (survey spectrum shown in figure 3-1b ), the F/C atom ratio is decreased from 1.98 to 1.65: i.e., about 17% of the F is removed from the near-surface region by this short exposure. The F features do not appear to change shapes or positions in the survey spectra obtained before and after the 2-hr UV exposure. However, the shape of the C 1s peak is significantly altered in figure 3-1b This is more apparent in the high-resolution spectra shown in figure 3-2 The C 1s and F 1s features obtained from the as-entered TFE Teflon are shown in figure 3-2Aa and 3-2Ba respectively and in figure 3-2Ab and 3-2Bb for the surface exposed to UV for 2 hrs. The C 1s feature obtained from the as-entered TFE Teflon consists of a single, narrow peak with a BE of 292.5 eV. The C 1s feature obtained after the 2-hr UV exposure is quite broad and complex consisting of contributions from at least five different species with BEs of 294.3, 292.5, 288.0 and 284.2 eV. The feature at 284.2 eV is due to a small amount of

PAGE 38

30 hydrocarbon contamination on the as-entered surface. Another small and ill-defined peak appears to be present at 290.0 eV. Assigning the peaks to specific species is again quite difficult for the reasons discussed in Chapter 2 In addition the UV-exposed surface is a damaged surface in that the composition is altered to a great extent, and the structure is most likely altered in the manner discussed below. The peak at 294.3 eV has a higher BE than carbon in the TFE Teflon environment indicating that electron-deficient species are formed under UV irradiation. This may occur by emission of F species leaving carbon species, which have lost an electron pair. The nuclei of these species then attract the C 1s electrons more strongly resulting in a larger C 1s BE. The other two states with BEs of 290.0 and 288.0 may be carbons, which have lost one and two fluorines respectively. An XPS survey spectrum obtained from the TFE Teflon surface after a 24-hr UV exposure is shown in figure 3-3a The increased AO exposure results in a decrease in the F/C atom ratio from 1.65 to 1.58 ( table 3-1 ), and the C 1s peak shape is changed significantly from that after the 2-hr UV exposure ( figure 3-1b ). This is more apparent in the C 1s spectrum shown in figure 3-2Ac The features with BEs of 294.3, 290.0 and 288.0 eV are now more prominent relative to the TFE Teflon feature at 292.5 eV. These results demonstrate that UV exposure destroys the TFE Teflon structure by removing F from the near-surface region. The two states at BEs of 288.0 and 290.0 eV are due to F-depleted regions. The presence of lower concentrations of F implies that electron density on the C atoms is not decreased to such a large extent by withdrawal toward the F resulting in increased C 1s BEs. The corresponding F 1s feature is shown in figure 3-2Bc It remains unchanged except for a slight reduction in size.

PAGE 39

31 Next the 24-hr, UV-exposed TFE Teflon surface was exposed to O 2 for 20 min at room temperature and 150 Torr. The XPS survey spectrum obtained from this surface is shown in figure 3-3b There appears to be negligible differences in the sizes and shapes of the C and F features before and after the O 2 exposure in these survey spectra, but a very small peak due to O is apparent at a BE of approximately 535 eV. The fact that this surface dissociatively chemisorbs molecular oxygen implies that reactive sites are present on the UV-exposed TFE Teflon surface. The amount of O 2 that chemisorbs provides a measure of the concentration of reactive sites in the near-surface region. Previous studies have shown that AO-exposed Kapton [ 10 ], Tedlar [ 29 ] and Tefzel [ 30 ] chemisorb very large amounts of molecular oxygen indicating that these surfaces contain a high concentration of reactive sites. These reactive sites are probably present because cross bonding cannot occur due to geometrical constraints. The amount of molecular oxygen chemisorbed on UV-exposed TFE Teflon is very small indicating that the concentration of reactive species is quite small. The data in table 3-1 indicates that much of the F initially present is removed by the UV exposure. Since very little molecular oxygen chemisorbs, the broken bonds formed by UV exposure must cross link in TFE Teflon. Although the peak shapes are similar before and after the O 2 exposure in the survey spectra, quite significant differences are apparent in the high-resolution C 1s spectrum ( figure 3-2Ad ), and these are accentuated in figure 3-4a where the C 1s spectrum taken after the 24-hr UV exposure is shown as a solid line and that taken after the 20-min O 2 exposure is shown as the dashed line. Specifically, the feature at 288.0 eV is decreased in intensity while the features at 290.0 and 294.3 eV are increased in intensity. This suggests that the O 2 chemisorbs at the C sites, which have lost two fluorines. After

PAGE 40

32 another 1 hr and then 3 hrs of UV exposure ( figures 3-2Ae and 3-2Af respectively), the 288.0 eV feature increases in size as do the 290.0 and 294.3 eV features. This is clearly observed in figure 3-4b where the C 1s spectrum obtained after the O 2 exposure (dashed line) and after the 28-hr UV exposure (solid line). The O 1s peak is not present in the survey spectrum (not shown) indicating that UV removes chemisorbed O from the O 2 -exposed, UV-exposed TFE Teflon surface. The mechanism of both F and O removal is photon stimulated desorption (PSD) which breaks the chemisorption bonds and creates an antibonding potential [ 31 ]. The F/C atom ratios are similar after the O 2 exposure and the 1 hr UV exposure but drops to 1.43 after the 28-hr UV exposure. This is a large decrease, which may be due to the O 2 treatment. The surface was then exposed to air for 90 min at room temperature. This treatment is quite different than an exposure to research purity O 2 because air contains water, hydrocarbons, alcohols and CO 2 which can adsorb at a surface. The XPS survey spectra obtained after this treatment is shown in figure 3-5a A very small O 1s peak is apparent as is a very small N 1s peak. The high-resolution C 1s and F 1s spectra are shown in figure 3-2Ag and 3-2Bg respectively, and a detailed comparison of the C 1s spectra obtained after the 28-hr UV exposure (solid line) and 90-min air exposure are shown in figure 3-4c The feature at 295.4 eV is reduced in intensity suggesting that the F and e -depleted site is the location for chemisorption of O 2 and N 2 The F/C ratio is decreased from 1.43 to 1.32 by this treatment. The final treatment was an exposure to the UV flux for another 46 hrs (74 hrs total). This exposure is long enough to represent a steady state in which the surface continually erodes away at a constant rate but no further changes are observed in the

PAGE 41

33 surface chemistry. The survey spectrum obtained after this AO exposure is shown in figure 3-5b The O 1s and N 1s features are not present indicating that UV removes surface O and N. The F/C atom ratio ( table 3-1 ) is increased from a value of 1.32 to 1.60. This indicates that a value of 1.60 is the steady-state value and is the same value obtained after the 24-hr UV exposure. The corresponding high-resolution C 1s and F 1s spectra are shown in figure 3-2Ah and 3-2Bh respectively. Compared to the C 1s spectrum obtained after the 24-hr AO exposure, the F-depleted species with BEs of 288.2, 289.8 and 295.6 eV are more pronounced relative to the C 1s feature due to TFE Teflon at 292.6 eV. This is quite apparent from the comparison of the C 1s spectra obtained after the 90-min air exposures (dashed line) and 74-hr UV exposure (solid line) is shown in figure 3-4d Summary In this study the chemical alterations at a TFE Teflon surface caused by exposure to UV radiation (115-400 nm) have been studied using XPS. A 2-hr exposure to UV radiation results in a decrease in the F/C atom ratio from 1.98 to 1.65 and formation of three new carbon chemical states assigned as C bonded to only one F, C bonded to other C and C which have lost a pair of electrons by emission of F Another 22 hrs of UV exposure results in a further decrease of the F/C atom ratio to 1.58 and an increase in the concentration of the three new carbon states with the electron-depleted C state predominating. Exposure of this UV-exposed surface to O 2 results in chemisorption of a very small amount of O due to bonding at reactive sites. The fact that so little O 2 chemisorbs indicates that the surface carbon bonds to other carbon, which is consistent with the formation of the new C chemical states. Further exposure of this surface to UV radiation for 3 hrs results in removal of the chemisorbed oxygen and a further reduction

PAGE 42

34 in the F/C atom ratio to 1.43. Then this UV-exposed surface was exposed to air. Both O 2 and N 2 are chemisorbed in very small quantities. Another 46 hrs of UV exposure results in removal of the chemisorbed O and N and attainment of a steady-state F/C atom ratio of 1.60. Comparison Between AO and UV Exposure Both UV and AO induce chemical alterations at a TFE Teflon surface, which result in erosion. The AO exposure results in formation of C states bonded to only one F and to other carbons just as the UV exposure. The results are quite different indicating that the chemical alterations result from different mechanisms. A comparison of the high-resolution C 1s spectra obtained from UV-exposed TFE Teflon (solid lines) and AO-exposed TFE Teflon (dashed lines) is shown in figure 3-6 The spectra obtained from the two as-entered samples ( figure 3-6a ) are identical as expected, but the AOand UV-exposed TFE Teflon spectra are all quite different ( figure 3-6b to h ). With increasing AO exposure the feature due to TFE Teflon at a BE of 292.5 eV is diminished. It is quite small after the 74-hr exposure. This is consistent with the compositional data, which indicates that the F/C atom ratio decreases to one-third of its initial value while UV exposure only results in a decrease to four-fifths of its initial value. Hence, the peak due to TFE Teflon remains as the predominant feature after the 74-hr UV exposure ( figure 3-6h ). Another significant difference is that a very large peak forms at 294.5 eV due to emission of F during UV exposure while this feature is only a very small shoulder for AO-exposed TFE Teflon. The two features at BEs of 288.3 and 288.9 eV are present in the spectra obtained from UV-exposed TFE Teflon, but they are smaller than those from AO-exposed TFE Teflon. AO-exposed TFE Teflon also chemisorbs larger amounts of O 2 than UV-exposed TFE Teflon and does not chemisorb N 2 These vary significant

PAGE 43

35 differences indicate that the mechanisms that result in chemical alteration of the TFE Teflon surface are different for UV radiation and hyperthermal AO as expected.

PAGE 44

36 Figure 3-1. XPS survey spectra obtained from TFE Teflon (a) as entered and (b) after a 2-hr exposure to UV radiation.

PAGE 45

37 Figure 3-2. High-resolution XPS (A) C 1s and (B) F 1s obtained from TFE Teflon (a) as entered, (b) after exposure to UV for 2 hrs, (c) after exposure to UV for a total of 24 hrs, (d) after exposure to O2 at room temperature and 150 Torr for 20 min, (e) after exposure to UV for a total of 25 hrs, (f) after exposure to UV for a total of 28 hrs, (g) after exposure to ambient air for 90 min at and (h) after exposure to UV for a total of 74 hrs.

PAGE 46

38 Figure 3-3. XPS survey spectra obtained from TFE Teflon (a) after a 24-hr exposure to UV radiation and (b) after a 20-min exposure to O2 at room temperature and 150 Torr.

PAGE 47

39 Figure 3-4. Overlays of XPS C 1s spectra obtained (a) after the 24-hr UV exposure (solid line) and the O2 exposure (dashed line), (b) after the O2 exposure (dashed line) and 28-hr UV exposure (solid line), (c) after the air exposure (dashed line) and the 28-hr UV exposure (solid line) and (d) after the air exposure (dashed line) and the 74-hr UV exposure (solid line).

PAGE 48

40 Figure 3-5. XPS survey spectra obtained from UV-exposed TFE Teflon (a) after a 90-min exposure to air at room temperature and (b) after a total of 74 hrs exposure to UV (46 hrs of UV exposure after the air exposure).

PAGE 49

41 Figure 3-6. High-resolution XPS C 1s spectra obtained from AO-exposed (dashed line) and UV-exposed (solid line) TFE Teflon (a) as entered, (b) after exposure for 2 hrs, (c) after exposure for a total of 24 hrs, (d) after exposure to O 2 at room temperature and 150 Torr for 20 min, (e) after exposure for a total of 25 hrs, (f) after exposure for a total of 28 hrs, (g) after exposure to air for 90 min at room temperature and (h) after exposure for a total of 74 hrs.

PAGE 50

42 Table 3-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%) F C F/C as entered 66.5 33.5 1.98 UV 2 hrs 62.3 37.7 1.65 UV 24 hrs 61.2 38.8 1.58 O 2 20 min 60.7 39.3 1.54 UV 25 hrs 61.0 39.0 1.56 UV 28 hrs 58.9 41.1 1.43 Air 90 min 56.9 43.1 1.32 UV 74 hrs 61.6 38.4 1.60

PAGE 51

REFERENCES 1. B.A. Banks, The Use of Fluoropolymers in Space Applications. Modern Fluoropolymers, John Wiley & Sons, New York 1997. 2. L.J. Leger and J.T. Visentine, J. Spacecraft Rockets, 23, 505 (1986). 3. S.L. Koontz, L.J. Leger, J.T. Visentine, D.E. Hunton, J.B. Cross and C.L. Hakes, J. Spacecraft Rockets, 32, 483 (1995). 4. R.C. Tennyson, Can. J. Phys, 69, 1190 (1991). 5. S.L. Koontz, L.J. Leger, S.L. Rickman, C.L. Hakes, D.T. Bui, D.E. Hunton and J.B. Cross, J. Spacecraft Rockets, 32, 475 (1995). 6. M. R. Reddy, J. Mater. Sci., 30, 281 (1995). 7. S. Packirisamy, D. Schwam and M.H. Litt, J. Mater. Sci., 30, 308 (1995). 8. M.R. Reddy, N. Srinivasamurthy and B.L. Agrawal, Eur. Space Agency J., 16, 193 (1992). 9. K. K. de Groh, and B. A. Banks, J. Spacecraft Rockets, 31, 656 (1994). 10. E. Grossman, Y. Lifshitz, J.T. Wolan, C.K. Mount and G.B. Hoflund, J. Spacecraft Rockets, 36, 75 (1999). 11. R.I. Gonzalez, S.H. Phillips and G.B. Hoflund, J. Spacecraft Rockets, 37, 463 (2000). 12. S.H. Phillips, G.B. Hoflund and R.I. Gonzalez, Proc. 45th International SAMPE Symposium, 45, 1921 (2000). 13. G. B. Hoflund, R. I. Gonzalez, S. H. Phillips, J. Adhesion Sci. Technol., 15, 1199 (2001). 14. K.K. DeGroh, J.R. Gaier, M.P. Espe, D.R. Cato, J.K. Sutter, and D.A. Scheiman, High Perform. Polym., 12, 83 (2000). 15. J.A. Dever, K.K. DeGroh, B.A. Banks, J.A. Townsend, J. L. Barth,, S. Thomson, T. Gregory, W. Savage, High Perform. Polym., 12, 125 (2000). 43

PAGE 52

44 16. F.A. Rasoul, D.J.T. Hill, J.S. Forsythe, J.H. O'Donnell, G.A. George, P.J. Pomery, P.R. Young, and J.W. Connell, J. Appl. Polym. Sci., 58, 1857 (1995). 17. A. Gindulyte, L. Massa, B.A. Banks and S.K.R. Miller, J. Phys. Chem. A, 106, 5463 (2002). 18. G. R. Corallo, G.B. Hoflund and R.A. Outlaw, Surf. Interface Anal., 12, 185 (1988). 19. M.R. Davidson, G.B. Hoflund, and R.A. Outlaw, Surf. Sci., 281, 111 (1993). 20. R.A. Outlaw, G.B. Hoflund, and G.R. Corallo, Appl. Surf. Sci., 28, 235 (1987). 21. E. Wisotzki, A.G. Balogh, H. Horst, J.T. Wolan, G.B. Hoflund, J. Vac. Sci. Technol. A, 17, 14 (1999). 22. E. Wisotzki, H. Hahn, G.B. Hoflund, in Trends and New Applications of Thin Films, Horst Hoffman, Trans Tech Publications LTD, Switzerland, 1998, 181, pp. 277-278. 23. M. R. Davidson, G. B. Hoflund, R. A. Outlaw, J Vac. Sci. Technol. A, 11, 264 (1993). 24. R.E. Gilbert, D.F. Cox and G.B. Hoflund, Rev. Sci. Instrum., 53, 1281 (1982). 25. A. Savitzky and and M.J.E. Golay, Anal. Chem., 36, 1627 (1964). 26. G. Beamson and D. Briggs (Eds.), High Resolution XPS of Organic Polymers: The Scienta ESCA300 Database, pp. 54, 226-236, Wiley, Chichester (1992). 27. C.D. Wagner, W.M., Riggs, L.E Davis, J.F. Moulder and G.E. Muilenberg (Eds.), Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer Corp., Eden Prairie, MN, (1979). 28. G.B. Hoflund in: Handbook of Surface and Interface Analysis: Methods in Problem Solving, J.C. Rivire and S. Myhra (Eds.), pp. 57-158 Marcel Dekker, New York (1998). 29. G.B. Hoflund and M.L. Everett, In press. 30. M.L. Everett and G.B. Hoflund, In press. 31. Hoflund, G. B., Scanning Elect. Microsc., 4, 1391 (1985). TFE Teflon FEP Teflon Tefzel Tedlar and Kapton are registered trademark names by E.I. du Pont Nemours & Co., Inc.

PAGE 53

BIOGRAPHICAL SKETCH Michael L. Everett was born on February 23, 1978, in Springfield, Massachusetts, to Wayne R. Everett and Kathleen S. Everett. He graduated from Springfield Central High School in 1996. That fall Michael began his college education at Worcester Polytechnic Institute in Worcester, Massachusetts. In May 2000 he received his Bachelor of Science in chemical engineering. Michael then entered the Graduate School at the University of Florida in Gainesville, Florida, to further his education in chemical engineering. As a graduate student he began working under Dr. Gar B. Hoflund, performing research in surface science. 45


Permanent Link: http://ufdc.ufl.edu/UFE0006920/00001

Material Information

Title: Chemical Alteration of Poly (Tetrafluoroethylene) (TFE Teflon) Induced by Exposure to Hyperthermal Atomic Oxygen and Ultraviolet Radiation
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006920:00001

Permanent Link: http://ufdc.ufl.edu/UFE0006920/00001

Material Information

Title: Chemical Alteration of Poly (Tetrafluoroethylene) (TFE Teflon) Induced by Exposure to Hyperthermal Atomic Oxygen and Ultraviolet Radiation
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006920:00001


This item has the following downloads:


Full Text











CHEMICAL ALTERATION OF POLY TETRAFLUOROETHYLENEE) (TFE
TEFLON") INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN
AND ULTRAVIOLET RADIATION














By

MICHAEL L. EVERETT


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004


































This thesis is dedicated to Wayne and Kathleen Everett.
















ACKNOWLEDGMENTS

I would like to thank Gar B. Hoflund, Jason F. Weaver, Helena Hagelin-Weaver

and Oscar D.Crisalle for their guidance, their help, and their support throughout my time

at the University of Florida. They have taught me a great deal and I am honored to have

worked with them. I would also like to thank Seung-Hoon Oh and Bryan Fitzsimmons

for their assistance and friendship. I would finally like to thank my parents, my sister,

my entire family and all my friends for their support throughout the years.


















TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .........__.. ..... .__. .............._ iii..

LIST OF TABLES ........._.___..... .__. ...............v....


LI ST OF FIGURE S .............. .................... vi


AB STRAC T ................ .............. vii


CHAPTER

1 INTRODUCTION ................. ...............1.......... ......


Overview ................. ...............1.................
Literature Review .............. .. ...............1....
O-Atom Source Characteristics .............. ...............3.....
Ultraviolet Source Characteristics .............. ...............6.....
X-ray Photoelectron Spectroscopy .............. ...............6.....


2 CHEMICAL ALTERATION OF POLY TETRAFLUOROETHYLENEE) (TFE
TEFLON") INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC
OXYGEN .............. ...............9.....


Experimental ................. ...............9.................
Results and Discussion ................ ...............10........... ....
Summary ................. ...............17.................


3 CHEMICAL ALTERATION OF POLY TETRAFLUOROETHYLENEE) (TFE
TEFLON") INDUCED BY EXPO SURE TO ULTRAVIOLET RADIATION ........27


Experimental ............ ..... .._ ...............27...
Results and Discussion .............. ...............28....
Summary ............... ............__ ... ......__ .........33
Comparison Between AO and UV Exposure .............. ...............34....

REFERENCES .............. ...............43....


BIOGRAPHICAL SKETCH .............. ...............45....

















LIST OF TABLES


Table pg

2-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%) ............24

2-2. Near-surface Compositions of Tefzel after Various Treatments (at%) ................... ..25

2-3. Near-surface Compositions of Tedlar after Various Treatments (at%) ................... ..26

3-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%) ............42


















LIST OF FIGURES

Figure pg

1-1. ESD atomic oxygen source............... ...............8.

2-1. XPS survey spectra obtained from TFE Teflon............... ...............19.

2-2. High-resolution XPS obtained from TFE Teflon .............. ...............20....

2-3. XPS survey spectra obtained from TFE Teflon............... ...............21.

2-4. XPS survey spectra obtained from TFE Teflon............... ...............22.

2-5. XPS survey spectra obtained from TFE Teflon............... ...............23.

3-1. XPS survey spectra obtained from TFE Teflon............... ...............36.

3-2. High-resolution XPS (A) C 1s and (B) F 1s obtained from TFE Teflon...................37

3-3. XPS survey spectra obtained from TFE Teflon............... ...............38.

3 -4. Overlays of XPS C 1s spectra. ........... ......__ ...............3

3-5. XPS survey spectra obtained from UV-exposed TFE Teflon .............. ..................40

3-6. High-resolution XPS C 1s spectra obtained from AO-exposed (dashed line) and UV-
exposed (solid line) TFE Teflon............... ...............41.
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

CHEMICAL ALTERATION OF POLY TETRAFLUOROETHYLENEE) (TFE
TEFLON") INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN
AND ULTRAVIOLET RADIATION

By

Michael L. Everett

August 2004

Chair: Gar B. Hoflund
Major Department: Chemical Engineering

In this study the erosion of poly tetrafluoroethylenee) (TFE Teflon) by

hyperthermal atomic oxygen (AO) and ultraviolet radiation (UV) (115-400 nm) has been

examined using X-ray photoelectron spectroscopy (XPS). The initial F/C atom ratio of

1.66 decreases to 1.15 after a 2-hr exposure to a flux of 2 x 1015 atoms/cm2-S AO with an

average kinetic energy of 5 eV. The F/C atom ratio is further reduced to a value of 0.90

after a 25-hr exposure. The high-resolution XPS C 1s data indicate that new chemical

states of carbon form as the F is removed, and that the relative amounts of these states

depend upon the F content of the near-surface region. The states are most likely due to C

bonded only to one F atom and C bonded only to other C atoms. Exposures of the AO-

damaged surface to 02 TOSult in chemisorption of a very small amount of O (~0.8 at%)

indicating that large quantities of reactive sites are not formed during the chemical

erosion by AO. Further exposure to AO removes this chemisorbed oxygen. After

exposing the AO-exposed surface to air for 90 min, both O and H20 are chemisorbed,









and the F/C at ratio is reduced to 0.68. Another 46 hrs of AO exposure results in removal

of these species and a further decrease in the F/C at ratio to 0.58.

The initial F/C atom ratio of 1.98 decreases to 1.65 after a 2-hr exposure to UV

radiation. The F/C atom ratio is further reduced to a steady-state value of 1.60 after a 74-

hr exposure. The high-resolution XPS C 1s data indicate that new chemical states of

carbon form as the F is removed, and that the relative amounts of these states depend

upon the F content of the near-surface region. The states are most likely due to C bonded

only to one F atom, C bonded only to other C atoms and C which have lost a pair of

electrons through emission of F~. Exposures of the UV-damaged surface to 02 TOSult in

chemisorption of a very small amount of O indicating that large quantities of reactive

sites are not formed during the chemical erosion by UV. Further exposure to UV

removes this chemisorbed oxygen. Comparison of XPS data indicates that the

mechanisms of chemical alteration by UV radiation and hyperthermal atomic oxygen

(AO) are different as expected.















CHAPTER 1
INTTRODUCTION


Overview

A novel electron stimulated desorption (ESD) hyperthermal atomic oxygen (AO)

source as well as an ultraviolet (UV) radiation lamp were used to induce chemical

changes to the surface of a TFE Teflon" film. The changes were studied by X-ray

photoelectron spectroscopy (XPS). The goal of this study is to examine and attempt to

understand the chemical alterations of a TFE Teflon surface when impacted by

hyperthermal (~5 eV) AO and UV radiation, as it may occur in low earth orbit (LEO).

The average energy and energy spread of the AO from this source are similar to AO in

LEO. Also, the UV lamp used covers the range of UV radiation that occurs in LEO,

vacuum ultraviolet (VUV) (115-200 nm). There are two types of bonds in TFE Teflon;

C-C bonds and C-F bonds with bond strengths of approximately 83 and 111 kcal/mole,

respectively. The UV radiation used in this study is energetic enough to break these

bonds.

Literature Review

Polymers are attractive and desirable materials for use in space applications

because they are lightweight and typically much easier to process using techniques such

as extrusion, casting and injection molding at relatively low temperatures compared to

metals and ceramics. They also tend to be more flexible and offer a wide variety of

choices from optically transparent to opaque, rubbery to stiff and conducting to









insulating. TFE Teflon is one material that is widely used as a thermal blanket for

spacecraft flying in low earth orbit (LEO) [1]. Thermal control is provided by lining the

TFE Teflon with aluminum or silver. The TFE Teflon has a high thermal emittance and

this system reflects a large fraction of the incident solar energy.

However, over the last two decades, it has been well established that polymers

undergo severe degradation resulting in reduced spacecraft lifetimes. These materials

degrade because spacecraft surfaces are exposed to high fluxes of atomic oxygen (AO),

bombardment by low- and high-energy charged particles, thermal cycling and the full

spectrum of solar radiation. AO is the main constituent of the atmosphere in LEO. It is

formed by dissociation of molecular oxygen by ultraviolet radiation from the sun,

resulting in an AO concentration of approximately 10s atoms/cm3. The reverse reaction

in which an oxygen molecule forms from AO does not have a high reaction rate because

it requires a teratomic collision. The third atom is required to dissipate the energy

released by formation of Oz. The actual flux of ~1014 atoms/cm2-S impinging on a

spacecraft is high due to orbiting speeds of approximately 8 km/s. At these relative

speeds thermal AO collides with a kinetic energy of ~4.5 eV. These highly energetic

collisions not only result in surface chemical reactions but can also lead to a pure

physical sputtering of the surface atoms in the absence of any chemical changes. Many

studies have been conducted in an effort to determine the mechanism of this degradation

primarily caused by surface reactions with AO [2-9]. However, these studies have all

been carried out after exposing these highly reactive surfaces to air prior to analysis, thus

introducing artifacts that do not represent the true space environment. Recent studies

have shown that exposure to air chemically alters the reactive surfaces formed during AO










exposure [10-13]. It is, therefore, essential that analysis of polymers exposed to AO be

carried out in-situ to avoid artifacts induced by air exposure.

Several studies have been conducted on the deterioration of fluorinated polymers

retrieved from spacecraft subjected to the LEO environment. The outer layer of Teflon-

fluorinated ethylene propylene" (FEP) multi-layer insulation on the Hubble space

telescope (HST) was significantly cracked at the time of the second HST servicing

mission, 6.8 years after it was launched into LEO [14,15]. Comparatively minor

embrittlement and cracking were also observed in the FEP materials retrieved from solar-

facing surfaces on the HST at the time of the first servicing mission (3.6 years of

exposure). Furthermore, an increased deterioration of fluorinated polymers may result

from the synergistic effect of VUV radiation in the presence of AO [16]. This point

remains to be demonstrated.

Intuitively low-energy AO would not be expected to break these bonds, and this

has been found experimentally [4]. Gindulyte, Massa, Banks and Miller [17] have

carried out a quantum mechanical study on the erosion of TFE Teflon by AO. They

conclude that AO in LEO possesses enough translational energy to erode TFE Teflon by

breaking C-C bonds. They did not consider breakage of C-F bonds based on

electronegativity arguments.

O-Atom Source Characteristics

A photograph and schematic diagram illustrating the operational principles of the

ESD AO source are shown in figures 1-la and 1-1b, respectively. Ultrahigh-purity

molecular oxygen dissociatively adsorbs on the high-pressure (2 Torr) side of a thin

metallic Ag-alloy membrane maintained at elevated temperature (~4000C) and permeates









through the membrane to the UHV side. There the chemisorbed atoms are struck by a

directed flux of primary electrons, which results in ESD of the O atoms forming a

continuous flux. The primary electrons are produced by thermionic emission from a

coiled hot filament supported around the perimeter of the membrane. An electron

reflector (lens assembly) surrounds the filament. It produces a potential field, which

creates a uniform flux of electrons over the membrane surface. These primary electrons

have a kinetic energy of 1000 eV and provide two functions: ESD of the O atoms and

heating of the membrane surface. Another lens is placed between the reflector and the

sample for removal of all charged particles including secondary electrons and O+ and O-

ions produced during the ESD process.

Several processes have to function in series at sufficiently high rates for this

system to work, including dissociative chemisorption of the molecular gas on the metal

surface, permeation of atomic oxygen through the membrane and formation of the neutral

flux by ESD. Since these processes occur in series, the slowest one determines the

magnitude of the AO flux. The sticking coefficient of 02 On polycrystalline Ag (step 1)

is fairly small (< 0.001) so it is necessary to use a high pressure on the upstream side of

the membrane. However, the permeation rate through the membrane is proportional to

the reciprocal of the membrane thickness. This means that it is desirable to have a high

pressure and a thin membrane, but this can lead to membrane failure. The ESD rate can

be increased by increasing the primary electron current to the membrane, but this

increases the temperature of the membrane and can result in evaporation of Ag, which is

unacceptable.









The AO produced by this source has been shown to be hyperthermal (energies

greater than 0.01 to 0.02 eV), but the neutral energy distribution has not been measured.

Corallo et al. [18] have measured the energy distribution of O ions emitted by ESD from

a Ag(110) surface and found that this distribution has a maximum of ~5 eV and a full-

width at half maximum of 3.6 eV. This ion energy distribution sets an upper bound for

the neutral energy distribution because ESD neutrals are generally believed to be less

energetic than ESD ions based on models of the ESD process. This point has been

discussed often in the ESD literature but not actually demonstrated. Since neutral ESD

species are difficult to detect, very few ESD studies of neutral species have appeared in

the literature. The neutral atom flux has been detected by using a quadrupole mass

spectrometer [19, 20] in the appearance potential (AP) mode to allow the atoms to be

distinguished from residual gases and background gas products formed by collisions of

the neutrals with the walls of the UHV system. In this experiment the ion acceleration

potential was set at 0.0 V. Calibration studies have demonstrated that the ions entering

the quadrupole section must have a minimum kinetic energy of 2.0 eV to reach the

detector so the ESD neutrals detected have a minimum energy of 2.0 eV. Therefore, the

hyperthermal AO produced by this ESD source have energies greater than 2 eV but

possibly less than the ion energy distribution. Furthermore, these mass spectrometric

experiments have shown that the AO-to-O+ ratio is about 10s and that the O -to-O- ratio

is about 100.

Several approaches have been used to measure the magnitude of the hyperthermal

AO flux and reasonable agreement was obtained between the various methods. The flux

from the ESD AO source is approximately 2x1015 atoms/cm2-S. One of the most reliable









methods for flux determination is the measurement of a ZrO2 film growth rate [21]. A Zr

flux was generated by e-beam evaporation and the flux was calibrated using a quartz-

crystal monitor. Based on the facts that stoichiometric ZrO2 WAS produced and that no 02

is present in the AO flux, the AO flux was calculated. By doubling the Zr flux,

stoichiometric ZrO was grown [22]. The AO flux has also been determined by

measuring the chemisorption rate of AO on polycrystalline Au using ion scattering

spectroscopy [23]. The flux determined using this method is in excellent agreement with

that determined using the oxide growth rate method.

Ultraviolet Source Characteristics

The UV source used in this study was a Hamamatsu water-cooled deuterium lamp

(model L1835). It had an MgF2 window allowing UV light with wavelengths ranging

from 115 to 400 nm to pass through to the sample in a 2.5 mm diameter aperture. The

aperture was approximately 15 cm from the sample and the lamp operated at a power of

150 W.

X-ray Photoelectron Spectroscopy

X-ray photoelectron spectroscopy (XPS), also known as electron spectroscopy for

chemical analysis (ESCA), is a photoemission technique used to determine the

composition and chemical state distribution of solid surfaces. It is performed by

irradiating the surface with soft x-rays in vacuo and analyzing the emitted electrons by

energy. An electrostatic charged particle energy analyzer is used to obtain the spectral

peaks generated from the kinetic energies of the emitted electrons. The corresponding

binding energies specific to each individual element are then calculated from the

following equation:


Eb = hv Ek +8









where Eb is the binding energy (BE) in the solid, Ek is the kinetic energy of the emitted

electron and AQ is the work function difference between the sample and the detector

material assuming there is no electrical charging at the sample surface.























(br) conflat flange
with thnruhole conflat flange
to uhy system


electrical
| ~--- feed-thru
mini flange




cooling rodl
O, supply tub~e


O-atom flux

10"atoms/cmas .


rleflector Ag alloy
membra ne


Figure 1-1. ESD atomic oxygen source (a) Photograph of ESD AO source. (b)
Schematic illustrating operational principles of ESD AO source.














CHAPTER 2
CHEMICAL ALTERATION OF POLY TETRAFLUOROETHYLENEE) (TFE
TEFLON") INDUCED BY EXPOSURE TO HYPERTHERMAL ATOMIC OXYGEN

One of the main constituents in LEO is hyperthermal AO, which is known to

degrade polymers. In this study XPS was used to characterize the surface of a TFE

Teflon film before and after exposures of varying lengths to AO, as well as 02 and air.

The goal of the study was to determine how hyperthermal AO affects the TFE Teflon

surface so that a better understanding of its space survivability can be obtained.

Experimental

An as-received E.I. du Pont Nemours & Co., Inc. TFE Teflon film was wiped

with methanol and inserted into the UHV chamber (base pressure <1.0x10-10 Torr). XPS

measurements were performed using a double-pass cylindrical mirror analyzer (DPCMA)

(PHI Model 25-270AR). XPS survey spectra were taken in the retarding mode with a

pass energy of 50 eV, and high-resolution XPS spectra were taken with a pass energy of

25 eV using Mg Ka X-rays (PHI Model 04-151 X-ray source). Data collection was

accomplished using a computer interfaced, digital pulse counting circuit [24] followed by

smoothing with digital-filtering techniques [25]. The sample was tilted 30 degrees off the

axis of the DPCMA, and the DPCMA accepted electrons emitted into a cone 42.6 + 6

degrees off the DPCMA axis.

XPS spectra were first obtained from the as-entered, solvent-cleaned sample. The

sample was then transferred into an adjoining UHV chamber that houses the ESD AO

source via a magnetically coupled rotary/linear manipulator. There the surface was










exposed to a hyperthermal AO flux and re-examined after various exposure times. The

sample was not exposed to air after the AO exposures and before collecting XPS data.

However, after the AO exposures the surface was exposed to 150 Torr Oz in anOther

adj oining chamber or to air to determine how this would affect the AO-exposed surfaces.

The order of the TFE Teflon exposures to AO was as follows: 2 hrs, 22 hrs (24 hrs total),

20 min Oz at 150 Torr, 1 hr (25 hrs total), 3 hrs (28 hrs total), 90min air (750C, 45%

relative humidity), 46 hrs (74 hrs total).

The approximate normal distance between the sample face and source in this study

was 15 cm, at which distance the flux was about 2.0 x 1015 atoms/cm2-S for the

instrument settings used. The sample was maintained at room temperature during the AO

exposures with a temperature increase to about 500C due to exposure to the X-ray source

during XPS data collection.

Results and Discussion

XPS survey spectra obtained from the TFE Teflon film before and after a 2-hr

exposure are shown in figures 2-la and 2-1b respectively. Spectrum a is identical to that

shown in the polymer XPS handbook by Beamson and Briggs [26] in that the C 1s, F 1s,

F Auger (KLL) and F 2s peaks are present. An O 1s peak, which would appear near a

binding energy (BE) of 530 eV, is not present. Estimates of the near-surface

compositions have been made from the peak areas in the high-resolution spectra using

published atomic sensitivity factors [27] with the assumption of a homogeneous surface

region. XPS probes the near-surface region of the sample and yields a weighted average

composition with the atomic layers near the surface being weighted more heavily since

the photoemitted electrons from these layers have a lower probability of scattering

inelastically. The sampling depth is ~4-6 nm, and ~10% of the signal originates from the










outermost atomic layer [28]. This near-surface region is nonhomogeneous because the

AO reacts most strongly with the outermost few atomic layers. Therefore, the region that

reacts to the greatest extent with AO also makes the largest contribution to the XPS

signal. This fact implies that XPS is an excellent technique for studying AO erosion of

spacecraft materials. Even though the distribution functions involving the depth of

chemical reactions in the near-surface region and the XPS determination of the weighted

average composition of the near-surface region are complex, the compositional values

provide a trend which is indicative of the chemical alterations occurring during AO

exposure. The compositions determined using the homogeneous assumption are shown

in table 2-1 before and after various exposures to AO and 02 Or air. The F/C atom ratio

obtained from the as-entered sample is 1.66 which is lower than the stoichiometric value

of 2.0. This is probably due to inaccuracies in the sensitivity factors used. After a 2-hr

AO exposure (survey spectrum shown in figure 2-1b), the F/C atom ratio is decreased

from 1.66 to 1.15: i.e., over 30% of the F is removed from the near-surface region by this

short exposure.

The F features do not appear to change shapes or positions in the survey spectra

obtained before and after the 2-hr AO exposure. However, the shape of the C 1s peak is

significantly altered. This is more apparent in the high-resolution spectra shown in figure

2-2. The C 1s and F 1s features obtained from the as-entered TFE Teflon are shown in

figure 2-2Aa and 2-2Ca respectively and in figure 2-2Ab and 2-2Cb for the surface

exposed to AO for 2 hrs. The C 1s feature obtained from the as-entered TFE Teflon

consists of a single, narrow peak with a BE of 292.5 eV. The C 1s feature obtained after

the 2-hr AO exposure is quite broad and complex consisting of contributions from at least









three different species with BEs of 292.5, 290.0 and 288.0 eV. Assigning the peaks to

specific species is quite difficult for several reasons. Defining localized species which



yield features at specific BEs is quite difficult. For example, F -C -F species



are present in both TFE Teflon and Tefzel" (ethylene tetrafluoroethylene, ETFE) but the


C 1s BEs are 292.48 and 290.9 eV respectively. Also, H -C -H species are



present in Tefzel and Tedlar" (Poly (vinyl fluoride), PVF) with BEs of 286.44 and 285.74

eV respectively. XPS BEs result from the chemical environment around the given

element. This fact implies that the chemical environment is larger than the species

defined in TFE Teflon, Tefzel and Tedlar. Since there are distinct peaks in these spectra,

there are distinct chemical environments, but defining the nature of this chemical

environment is not possible. Furthermore, the hyperthermal AO-exposed surface is a

damaged surface in that the composition is altered to a great extent, and the structure is

most likely altered as discussed below. The new features which appear in the C 1s

spectrum after the 2-hr AO exposure have lower BEs indicating that the chemical

environments responsible for these features are F depleted. This assertion is consistent

with the compositional data shown in table 2-1. The BE of the F 1s peak (figure 2-2Cb)

remains unchanged, but the peak is slightly broadened to the high-BE side. This

indicates that the F remaining in the near-surface region more strongly attracts electrons.

An XPS survey spectrum obtained from the TFE Teflon surface after a 24-hr AO

exposure is shown in figure 2-3a. The increased AO exposure results in a decrease in the

F/C atom ratio from 1.15 to 0.90 (table 2-1), and the C 1s peak shape is changed










significantly from that after the 2-hr AO exposure (figure 2-1b). This is more apparent in

the C 1s spectrum shown in figure 2-2Ac. The feature with a BE of 288.0 eV is now

most prominent and the feature due to TFE Teflon at a BE of 292.5 eV is the least

prominent. These results demonstrate that hyperthermal AO exposure destroys the TFE

Teflon structure by removing F from the near-surface region. The two states at BEs of

288.0 and 290.0 eV are due to F-depleted regions. The presence of lower concentrations

of F implies that electron density on the C atoms is not decreased to such a large extent

resulting in increased C 1s BEs. The feature at 288.0 eV may be due to regions where all

of the F has been removed. The corresponding F 1s feature is shown in Eigure 2-2Cc.

This feature is shifted from 689.3 to 690.0 eV. The increased BE indicates that the

remaining F is able to attract and bind electrons more strongly from the damaged matrix

as the F concentration is decreased.

Next the 24-hr, AO-exposed TFE Teflon surface was exposed to 02 for 20 min at

room temperature and 150 Torr. The XPS survey spectrum obtained from this surface is

shown in Eigure 2-3b. There appears to be negligible differences in the sizes and shapes

of the C and F features before and after the Oz exposure in these survey spectra, but a

small peak due to O is apparent at a BE of approximately 535 eV. The fact that this

surface dissociatively chemisorbs molecular oxygen implies that reactive sites are present

on the AO-exposed TFE Teflon surface. The amount of 02 that chemisorbs provides a

measure of the concentration of reactive sites in the near-surface region. Previous studies

have shown that AO-exposed Kapton" [10], Tedlar [29] and Tefzel [30] chemisorb very

large amounts of molecular oxygen indicating that these surfaces contain a high

concentration of reactive sites. These reactive sites are probably present because cross









bonding cannot occur due to geometrical constraints. The amount of molecular oxygen

chemisorbed on AO-exposed TFE Teflon is very small indicating that the concentration

of reactive species is quite small. The data in table 2-1 indicates that much of the F

initially present is removed by the AO exposure. Since very little molecular oxygen

chemisorbs, the broken bonds formed by AO exposure must cross link in TFE Teflon.

Although the peak shapes are similar before and after the Oz exposure in the

survey spectra, quite significant differences are apparent in the high-resolution spectra

(figure 2-2Ad and 2-2Cd). Specifically, the feature due to TFE Teflon at 292.5 eV is

increased in intensity relative to the F-depleted state at 288.0 eV. Since the AO reacts

most strongly with TFE Teflon right at the surface, the AO-exposed surface is most likely

a layered structure with the most F-depleted state at the surface (288.0 eV), the next most

F-depleted state just beneath (290.0 eV) and the TFE Teflon structure (292.5 eV) beneath

these two layers. A possibility is that the Oz reacts with the F-depleted layer at the

surface to form CO or CO2. This removes part of the top layer making the underlying

two layers more prominent in the XPS spectra. This also explains why the oxygen

content of the O2-exposed surface is so low (< 1 at%). The outermost layer is a damaged

layer with many vacancies and defects. This would make it quite reactive chemically.

The high-resolution O 1s feature is shown in figure 2-2Bd. It is narrow indicating that

only one chemical state of oxygen is present. The fact that this peak is present

demonstrates that the Ol bond can be broken at the AO-exposed TFE Teflon surface

under the O2-exposure conditions used. The corresponding F 1s feature is shown in

figure 2-2Cd. It is shifted toward the BE of F in TFE Teflon and is broadened. This









indicates that this feature is composed of contributions from F in the middle F-depleted

layer and F in TFE Teflon.

The O2-exposed surface was then exposed to AO for 1 hr (total AO exposure of

25 hrs), and the resulting XPS survey spectrum is shown in figure 2-4a. The O 1s peak is

not present indicating that AO removes chemisorbed O from the O2-exposed, AO-

exposed TFE Teflon surface. The compositional data in table 2-1 indicate that the near-

surface O concentration decreases from 0.8 to 0.1 at% by the AO exposure. The

mechanism of O removal may be physical sputtering or chemical reaction to form 02 Of

CO2. The F/C atom ratio is again decreased from 0.895 to 0.816. This is less than one-

half of the F initially present. More information is provided by the high-resolution

spectra shown in figure 2-2. In the C 1s spectrum (figure 2-2Ae), the F-depleted near-

surface region again yields the most prominent feature as before the Oz exposure (figure

2-2Ac). The peak due to the TFE Teflon structure is reduced in magnitude but not to that

before the Oz exposure. The O 1s feature shown in figure 2-2Be is essentially at the

noise level indicating that the chemisorbed O is removed by the AO exposure. The F 1s

feature (figure 2-2Ce) is not changed significantly by the 1-hr AO exposure although it

may have shifted slightly back toward higher BEs.

Another 3-hr exposure to AO results only in small changes which are not apparent

in the survey spectrum shown in figure 2-4b. According to the data in table 2-1, the F/C

atom ratio remains unchanged in the near-surface region but the last small amount of

chemisorbed O is removed. Changes in the high-resolution C 1s and F 1s features shown

in figures 2-2Af and 2-2Cf respectively are also small. The F 1s peak is shifted a few

tenths of an eV toward higher BEs which is characteristic of larger AO exposures.









The surface was then exposed to air for 90 min at room temperature. This

treatment is quite different than an exposure to research purity 02 because air contains

water, hydrocarbons, alcohols and CO2 which can adsorb at a surface. The XPS survey

spectra obtained after this treatment is shown in figure 2-5a. A small O 1s peak is

apparent which is larger than the O 1s peak obtained after the Oz exposure (figure 2-3b).

This is consistent with the compositional data in table 2-1 that gives an oxygen

concentration of 1.9 at%. The Oz exposure does not alter the F/C atom ratio while the air

exposure lowers the F/C atom ratio from 0.82 to 0.68 due to both an increase in the C

concentration and a decrease in the F concentration. This result is most likely due to

several processes. Reactive C near the surface would certainly react with oxygen in the

air to form CO or CO2. This process removes C from the surface. Hydrocarbons and

possibly alcohols in air adsorb on the surface. This enhances the intensity of the C signal

and decreases the intensity of the F signal. These assertions are consistent with the high-

resolution C 1s, O 1s and F 1s features shown in figure 2-2Ag, 2-2Bg and 2-2Cg

respectively. In the C 1s spectrum, the feature due to hydrocarbons at 288.0 eV is

predominant while the peak due to TFE Teflon at 292.5 eV is fairly small and broad. The

shape of the O 1s feature is quite different than that obtained after the 20-min Oz

exposure (figure 2-2Bd) in that it contains a shoulder at a BE about 2 eV higher than the

predominant feature. This shoulder is assigned to adsorbed water from the air. If this

contribution is subtracted from the O 1s feature, the magnitudes of the chemisorbed O are

similar for the Ol exposure and the air exposure. The F 1s peak is shifted to the BE

characteristic of an AO-exposed surface. This is due to the F in the middle layer which

makes a large contribution to the F 1s signal.









The Einal treatment was an exposure to the AO flux for another 46 hrs (74 hrs

total). This exposure is long enough to represent a steady state in which the surface

continually erodes away at a constant rate but no further changes are observed in the

surface chemistry. The survey spectrum obtained after this AO exposure is shown in

Eigure 2-5b. The O 1s feature is not present indicating that AO removes surface O as

found after the Oz exposure. The F/C atom ratio (table 2-1) is decreased to a value of

0.58. The series of exposures used in this study results in removal of approximately two-

thirds of the F in the near-surface region most likely by a physical sputtering mechanism.

This is the first step in the AO-erosion of TFE Teflon. An F-depleted C layer is left at the

surface which erodes more slowly. The corresponding high-resolution C 1s and F 1s

spectra are shown in Eigure 2-2Ah and 2-2Ch respectively. These spectra are most

similar to those obtained after the 24-hr AO exposure. The outermost F-depleted C layer

is predominant, and the top two F-depleted layers are so thick that the TFE Teflon peak at

292.5 eV is quite small. Based on mean-free-path arguments the F-depleted layer is

probably about 3 nm thick. The F 1s peak has a BE of 690.0 eV which is characteristic

of an AO-exposed TFE Teflon surface. Hence, the TFE Teflon structure makes only a

small contribution to this surface.

Summary

In this study the chemical alterations at a TFE Teflon surface caused by exposure to

hyperthermal AO have been studied using XPS. A 2-hr exposure to AO results in a

decrease in the F/C atom ratio from 1.66 to 1.15 and formation of two new carbon

chemical states assigned as C bonded to only one F and C bonded to other C. Another 22

hrs of AO exposure results in a further decrease of the F/C atom ratio to 0.90 and an

increase in the concentration of the two new carbon states with the C-bonding-to-C state










predominating. Exposure of this AO-exposed surface to 02 TOSults in a small amount (0.8

at%) of chemisorbed O due to bonding at reactive sites. The fact that so little Oz

chemisorbs indicates that the surface carbon bonds to other carbon which is consistent

with the formation of the new C chemical states. The chemisorbed O apparently is

present in only one chemical state. Further exposure of this surface to AO for 3 hrs

results in removal of the chemisorbed oxygen and a further reduction in the F/C atom

ratio to 0.82. Then this AO-exposed surface was exposed to air. Both 02 and H20 are

chemisorbed in small quantities (1.9 total O at%). Another 46 hrs of AO exposure results

in removal of the chemisorbed O and water and a further reduction in the F/C atom ratio

to 0.58.









11111 111111lll""" I'""l "'" "I"" "I"'"I"""'"" '""" "


lunaml11111 un111Inumulmn111111h11nual1111mn1 1lmond11111 mn11l


4-F 1s


m
c,

"1

i~t
k
c,

L/
W


(a)


(b)


800


600


400


200


Binding Energy (eV)


Figure 2-1. XPS survey spectra obtained from TFE Teflon (a) as entered and (b) after a
2-hr exposure to AO.



















~(e)

~II (e)
S(d)



(c)

(b)
(d) kr I(b)


(a) (a)

295 290 285 540 535 530 695 690 68i5
Binding Energy (eV)

Figure 2-2. High-resolution XPS obtained from TFE Teflon (A) C 1s, (B) O 1s, and (C)
F 1s (a) as entered, (b) after exposure to AO for 2 hrs, (c) after exposure to
AO for a total of 24 hrs, (d) after exposure to 02 at room temperature for 20
min, (e) after exposure to AO for a total of 25 hrs, (f) after exposure to AO for
a total of 28 hrs, (g) after exposure to air for 90 min at room temperature and
(h) after exposure to AO for a total of 74 hrs.









l""""'1111111l""""'ll" "'l" "'l" "'"" "'"" "'"" "'


b.,nn1111nnn1 111h111.111 1l11nn1n11Innn n11lnnnn11l1111nunin111 .111nI


F 1s


m
c,



k
c,

L/
W


n ~(a) F


,x 5
o


(b)


800


600


400


200


Binding Energy (eV)


Figure 2-3. XPS survey spectra obtained from TFE Teflon (a) after a 24-hr exposure to
AO and (b) after a 20-min exposure to 02 at room temperature and 150 Torr.










l""""'1111111l""""'ll" "'l" "'l" "'"" "'"" "'"" "'


b.,nn1111nnn1 111h111.111 1l11nn1n11Innn n11lnnnn11l1111nunin111 .111nI


m
c,

"1

i~t
k
c,


L/

W


U m
IN
n (a) Ftl


(b)


800


600


400


200


Binding Energy (eV)



Figure 2-4. XPS survey spectra obtained from TFE Teflon (a) after a 25-hr exposure to
AO (1-hr AO exposure after the 02 exposure), and (b) after 28 hrs total AO
exposure.









l" """I """"'l""""111 1 111 I II II l"" 111 I ""1I""" "" "I "'""" 1 11 1I '"'" '" iI I


-F 1s









O


m
c,



-r



F~1
Z


n ~(a) e


(b)


800


600


400


200


Binding Energy (eV)


Figure 2-5. XPS survey spectra obtained from TFE Teflon (a) after a 90-min exposure to
air at room temperature and (b) after a total of 74 hrs exposure to AO.


Immmlanuminanonhannelanonalunmulumnahnesml111 11111111111







24


Table 2-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%)

F O C F/C O/C
as entered 62.4 0.0 37.6 1.66 0.0
AO 2 hrs 53.6 0.0 46.4 1.15 0.0
AO 24 hrs 47.4 0.0 52.6 0.90 0.0
02 20 min 46.8 0.8 52.4 0.90 0.015
AO 25 hrs 44.9 0.1 55.0 0.81 0.002
AO 28 hrs 45.2 0.0 54.8 0.82 0.0
Air 90 min 39.8 1.9 58.3 0.68 0.03
AO 74 hrs 36.6 0.0 63.4 0.58 0.0







25


Table 2-2. Near-surface Compositions of Tefzel after Various Treatments (at%)

F O C F/C O/C
as entered 41.5 2.3 56.2 0.74 0.04
AO 2 hrs 14.5 0.7 84.8 0.17 0.01
AO 24 hrs 1.9 3.5 94.7 0.02 0.04
02 20 min 3.7 7.2 89.1 0.04 0.08
AO 25 hrs 1.7 6.8 91.5 0.02 0.07
AO 28 hrs 1.2 4.9 93.9 0.01 0.05







26


Table 2-3. Near-surface Compositions of Tedlar after Various Treatments (at%)

F O C F/C O/C
as entered 29.0 7.2 63.7 0.45 0.11
AO 2 hrs 1.7 4.0 94.2 0.018 0.04
AO 24 hrs 2.0 6.3 91.7 0.022 0.071
02 20 min 1.1 10.4 88.5 0.012 0.12
AO 25 hrs 1.3 7.0 91.7 0.014 0.08
AO 28 hrs 1.7 6.7 91.6 0.018 0.07














CHAPTER 3
CHEMICAL ALTERATION OF POLY TETRAFLUOROETHYLENEE) (TFE
TEFLON") INDUCED BY EXPOSURE TO ULTRAVIOLET RADIATION

In LEO objects are subjected to the full spectrum of solar radiation. One such

range of light is VUV (115-200 nm), which is known to degrade polymers. In this study

XPS was used to characterize the surface of a TFE Teflon film before and after exposures

of varying lengths to UV radiation (115-400 nm), as well as 02 and air. The goal of the

study was to determine how hyperthermal UV affects the TFE Teflon surface so that a

better understanding of its space survivability can be obtained.

Experimental

An as-received E.I. du Pont Nemours & Co., Inc. TFE Teflon film was wiped

with methanol and inserted into the UHV chamber (base pressure <1.0x10-10 Torr). XPS

measurements were performed using a double-pass cylindrical mirror analyzer (DPCMA)

(PHI Model 25-270AR). XPS survey spectra were taken in the retarding mode with a

pass energy of 50 eV, and high-resolution XPS spectra were taken with a pass energy of

25 eV using Mg Ka X-rays (PHI Model 04-151 X-ray source). Data collection was

accomplished using a computer interfaced, digital pulse counting circuit [24] followed by

smoothing with digital-filtering techniques [25]. The sample was tilted 30 degrees off the

axis of the DPCMA, and the DPCMA accepted electrons emitted into a cone 42.6 + 6

degrees off the DPCMA axis.

XPS spectra were first obtained from the as-entered, solvent-cleaned sample. The

sample was then transferred into an adjoining UHV chamber that houses the UV source









via a magnetically coupled rotary/linear manipulator. There the surface was exposed to

UV radiation and re-examined after various exposure times. The samples were not

exposed to air after the UV exposures and before collecting XPS data. However, after

the UV exposures the surfaces were exposed to 02 Or air to determine how this would

affect the UV-exposed surfaces. The order of the TFE Teflon exposures to UV radiation

was as follows: 2 hrs, 22 hrs (24 hrs total), 20 min Oz at 150 Torr, 1 hr (25 hrs total), 3

hrs (28 hrs total), 90min air (820C, 45% relative humidity), 46 hrs (74 hrs total).

The approximate normal distance between the sample face and source in this study

was 15 cm. The sample was maintained at room temperature during the UV exposures

with a temperature increase to about 500C due to exposure to the X-ray source during

XPS data collection.

Results and Discussion

XPS survey spectra obtained from the TFE Teflon film before and after a 2-hr UV

exposure are shown in figure 3-la and 3-1b respectively. Spectrum a is identical to that

shown in the polymer XPS handbook by Beamson and Briggs [26] in that the C 1s, F 1s,

F Auger (KLL) and F 2s peaks are present. An O 1s peak, which would appear near a

binding energy (BE) of 530 eV, is not present. Estimates of the near-surface

compositions have been made from the peak areas in the high-resolution spectra using

published atomic sensitivity factors [27] with the assumption of a homogeneous surface

region. XPS probes the near-surface region of the sample and yields a weighted average

composition with the atomic layers near the surface being weighted more heavily since

the photoemitted electrons from these layers have a lower probability of scattering

inelastically. The sampling depth is ~4-6 nm, and ~10% of the signal originates from the

outermost atomic layer [28]. This near-surface region is nonhomogeneous because the









UV radiation reacts most strongly with the outermost few atomic layers. Therefore, the

region that reacts to the greatest extent with UV radiation also makes the largest

contribution to the XPS signal. This fact implies that XPS is an excellent technique for

studying erosion of spacecraft materials by UV radiation and atomic oxygen. Even

though the distribution functions involving the depth of chemical reactions in the near-

surface region and the XPS determination of the weighted average composition of the

near-surface region are complex, the compositional values provide a trend which is

indicative of the chemical alterations occurring during UV exposure. The compositions

determined using the homogeneous assumption are shown in table 3-1 before and after

various exposures to UV radiation and 02 Or air. The F/C atom ratio obtained from the

as-entered sample is 1.98, which is very close to the stoichiometric value of 2.0.

After a 2-hr UV exposure (survey spectrum shown in figure 3-1b), the F/C atom

ratio is decreased from 1.98 to 1.65: i.e., about 17% of the F is removed from the near-

surface region by this short exposure. The F features do not appear to change shapes or

positions in the survey spectra obtained before and after the 2-hr UV exposure. However,

the shape of the C 1s peak is significantly altered in figure 3-1b. This is more apparent in

the high-resolution spectra shown in figure 3-2. The C 1s and F 1s features obtained

from the as-entered TFE Teflon are shown in figure 3-2Aa and 3-2Ba respectively and in

figure 3-2Ab and 3-2Bb for the surface exposed to UV for 2 hrs. The C 1s feature

obtained from the as-entered TFE Teflon consists of a single, narrow peak with a BE of

292.5 eV. The C 1s feature obtained after the 2-hr UV exposure is quite broad and

complex consisting of contributions from at least five different species with BEs of

294.3, 292.5, 288.0 and 284.2 eV. The feature at 284.2 eV is due to a small amount of










hydrocarbon contamination on the as-entered surface. Another small and ill-defined peak

appears to be present at 290.0 eV. Assigning the peaks to specific species is again quite

difficult for the reasons discussed in Chapter 2. In addition the UV-exposed surface is a

damaged surface in that the composition is altered to a great extent, and the structure is

most likely altered in the manner discussed below. The peak at 294.3 eV has a higher BE

than carbon in the TFE Teflon environment indicating that electron-deficient species are

formed under UV irradiation. This may occur by emission of F~ species leaving carbon

species, which have lost an electron pair. The nuclei of these species then attract the C 1s

electrons more strongly resulting in a larger C 1s BE. The other two states with BEs of

290.0 and 288.0 may be carbons, which have lost one and two fluorines respectively.

An XPS survey spectrum obtained from the TFE Teflon surface after a 24-hr UV

exposure is shown in figure 3-3a. The increased AO exposure results in a decrease in the

F/C atom ratio from 1.65 to 1.58 (table 3-1), and the C 1s peak shape is changed

significantly from that after the 2-hr UV exposure (figure 3-1b). This is more apparent in

the C 1s spectrum shown in figure 3-2Ac. The features with BEs of 294.3, 290.0 and

288.0 eV are now more prominent relative to the TFE Teflon feature at 292.5 eV. These

results demonstrate that UV exposure destroys the TFE Teflon structure by removing F

from the near-surface region. The two states at BEs of 288.0 and 290.0 eV are due to F-

depleted regions. The presence of lower concentrations of F implies that electron density

on the C atoms is not decreased to such a large extent by withdrawal toward the F

resulting in increased C 1s BEs. The corresponding F 1s feature is shown in figure 3-

2Bc. It remains unchanged except for a slight reduction in size.









Next the 24-hr, UV-exposed TFE Teflon surface was exposed to 02 for 20 min at

room temperature and 150 Torr. The XPS survey spectrum obtained from this surface is

shown in figure 3-3b. There appears to be negligible differences in the sizes and shapes

of the C and F features before and after the Oz exposure in these survey spectra, but a

very small peak due to O is apparent at a BE of approximately 535 eV. The fact that this

surface dissociatively chemisorbs molecular oxygen implies that reactive sites are present

on the UV-exposed TFE Teflon surface. The amount of 02 that chemisorbs provides a

measure of the concentration of reactive sites in the near-surface region. Previous studies

have shown that AO-exposed Kapton [10], Tedlar [29] and Tefzel [30] chemisorb very

large amounts of molecular oxygen indicating that these surfaces contain a high

concentration of reactive sites. These reactive sites are probably present because cross

bonding cannot occur due to geometrical constraints. The amount of molecular oxygen

chemisorbed on UV-exposed TFE Teflon is very small indicating that the concentration

of reactive species is quite small. The data in table 3-1 indicates that much of the F

initially present is removed by the UV exposure. Since very little molecular oxygen

chemisorbs, the broken bonds formed by UV exposure must cross link in TFE Teflon.

Although the peak shapes are similar before and after the Oz exposure in the survey

spectra, quite significant differences are apparent in the high-resolution C 1s spectrum

(figure 3-2Ad), and these are accentuated in figure 3-4a where the C 1s spectrum taken

after the 24-hr UV exposure is shown as a solid line and that taken after the 20-min Oz

exposure is shown as the dashed line. Specifically, the feature at 288.0 eV is decreased

in intensity while the features at 290.0 and 294.3 eV are increased in intensity. This

suggests that the Oz chemisorbs at the C sites, which have lost two fluorines. After









another 1 hr and then 3 hrs of UV exposure (figures 3-2Ae and 3-2Af respectively), the

288.0 eV feature increases in size as do the 290.0 and 294.3 eV features. This is clearly

observed in figure 3-4b where the C 1s spectrum obtained after the Oz exposure (dashed

line) and after the 28-hr UV exposure (solid line). The O 1s peak is not present in the

survey spectrum (not shown) indicating that UV removes chemisorbed O from the 02-

exposed, UV-exposed TFE Teflon surface. The mechanism of both F and O removal is

photon stimulated desorption (PSD) which breaks the chemisorption bonds and creates an

antibonding potential [31]. The F/C atom ratios are similar after the Oz exposure and the

1 hr UV exposure but drops to 1.43 after the 28-hr UV exposure. This is a large

decrease, which may be due to the Oz treatment.

The surface was then exposed to air for 90 min at room temperature. This

treatment is quite different than an exposure to research purity 02 because air contains

water, hydrocarbons, alcohols and CO2 which can adsorb at a surface. The XPS survey

spectra obtained after this treatment is shown in figure 3-5a. A very small O 1s peak is

apparent as is a very small N 1s peak. The high-resolution C 1s and F 1s spectra are

shown in figure 3-2Ag and 3-2Bg respectively, and a detailed comparison of the C 1s

spectra obtained after the 28-hr UV exposure (solid line) and 90-min air exposure are

shown in figure 3-4c. The feature at 295.4 eV is reduced in intensity suggesting that the

F- and e -depleted site is the location for chemisorption of 02 and N2. The F/C ratio is

decreased from 1.43 to 1.32 by this treatment.

The final treatment was an exposure to the UV flux for another 46 hrs (74 hrs

total). This exposure is long enough to represent a steady state in which the surface

continually erodes away at a constant rate but no further changes are observed in the









surface chemistry. The survey spectrum obtained after this AO exposure is shown in

Eigure 3-5b. The O 1s and N 1s features are not present indicating that UV removes

surface O and N. The F/C atom ratio (table 3-1) is increased from a value of 1.32 to 1.60.

This indicates that a value of 1.60 is the steady-state value and is the same value obtained

after the 24-hr UV exposure. The corresponding high-resolution C 1s and F 1s spectra

are shown in figure 3-2Ah and 3-2Bh respectively. Compared to the C 1s spectrum

obtained after the 24-hr AO exposure, the F-depleted species with BEs of 288.2, 289.8

and 295.6 eV are more pronounced relative to the C 1s feature due to TFE Teflon at

292.6 eV. This is quite apparent from the comparison of the C 1s spectra obtained after

the 90-min air exposures (dashed line) and 74-hr UV exposure (solid line) is shown in

figure 3-4d.

Summary

In this study the chemical alterations at a TFE Teflon surface caused by exposure

to UV radiation (115-400 nm) have been studied using XPS. A 2-hr exposure to UV

radiation results in a decrease in the F/C atom ratio from 1.98 to 1.65 and formation of

three new carbon chemical states assigned as C bonded to only one F, C bonded to other

C and C which have lost a pair of electrons by emission of F~. Another 22 hrs of UV

exposure results in a further decrease of the F/C atom ratio to 1.58 and an increase in the

concentration of the three new carbon states with the electron-depleted C state

predominating. Exposure of this UV-exposed surface to 02 TOSults in chemisorption of a

very small amount of O due to bonding at reactive sites. The fact that so little Oz

chemisorbs indicates that the surface carbon bonds to other carbon, which is consistent

with the formation of the new C chemical states. Further exposure of this surface to UV

radiation for 3 hrs results in removal of the chemisorbed oxygen and a further reduction









in the F/C atom ratio to 1.43. Then this UV-exposed surface was exposed to air. Both 02

and N2 are chemisorbed in very small quantities. Another 46 hrs of UV exposure results

in removal of the chemisorbed O and N and attainment of a steady-state F/C atom ratio of

1.60.

Comparison Between AO and UV Exposure

Both UV and AO induce chemical alterations at a TFE Teflon surface, which

result in erosion. The AO exposure results in formation of C states bonded to only one F

and to other carbons just as the UV exposure. The results are quite different indicating

that the chemical alterations result from different mechanisms. A comparison of the

high-resolution C 1s spectra obtained from UV-exposed TFE Teflon (solid lines) and

AO-exposed TFE Teflon (dashed lines) is shown in figure 3-6. The spectra obtained

from the two as-entered samples (figure 3-6a) are identical as expected, but the AO- and

UV-exposed TFE Teflon spectra are all quite different (figure 3-6b to h). With

increasing AO exposure the feature due to TFE Teflon at a BE of 292.5 eV is diminished.

It is quite small after the 74-hr exposure. This is consistent with the compositional data,

which indicates that the F/C atom ratio decreases to one-third of its initial value while

UV exposure only results in a decrease to four-fifths of its initial value. Hence, the peak

due to TFE Teflon remains as the predominant feature after the 74-hr UV exposure

(figure 3-6h). Another significant difference is that a very large peak forms at 294.5 eV

due to emission of F~ during UV exposure while this feature is only a very small shoulder

for AO-exposed TFE Teflon. The two features at BEs of 288.3 and 288.9 eV are present

in the spectra obtained from UV-exposed TFE Teflon, but they are smaller than those

from AO-exposed TFE Teflon. AO-exposed TFE Teflon also chemisorbs larger amounts

of 02 than UV-exposed TFE Teflon and does not chemisorb N2. These vary significant






35


differences indicate that the mechanisms that result in chemical alteration of the TFE

Teflon surface are different for UV radiation and hyperthermal AO as expected.









11111 111111lll""" I'""l "'" "I"" "I"'"I"""'"" '""" "


VI1
U


lunaml=1111 un.Inumulmnahanualnmnalmondnmnal1 11111111111 11111


-F 1s


m
c,

"1

i~t
k
c,

L/
W


(a)


(b)


800


600


400


200


Binding Energy (eV)


Figure 3-1. XPS survey spectra obtained from TFE Teflon (a) as entered and (b) after a
2-hr exposure to UV radiation.


11








I li I I I I


(d)
(d)











(a) I(a)

300 295 290 285 695 690 685
Binding Energyr (eV)

Figure 3-2. High-resolution XPS (A) C 1s and (B) F 1s obtained from TFE Teflon (a) as
entered, (b) after exposure to UV for 2 hrs, (c) after exposure to UV for a total
of 24 hrs, (d) after exposure to 02 at room temperature and 150 Torr for 20
min, (e) after exposure to UV for a total of 25 hrs, (f) after exposure to UV for
a total of 28 hrs, (g) after exposure to ambient air for 90 min at and (h) after
exposure to UV for a total of 74 hrs.






l""""'1111111l""""'ll" "'l" "'l" "'"" "'"" "'"" "'


II.........h.... .1111h1111111 1In1111111h11111 111I1111111 11h111111111h1 1111111l
800 700 600 500 400 300 200 100 0
Binding Energy (eV)

Figure 3-3. XPS survey spectra obtained from TFE Teflon (a) after a 24-hr exposure to
UV radiation and (b) after a 20-min exposure to 02 at room temperature and
150 Torr.


4_F 1s


m
c,

k
c,
L/
W


(a) F


-x 20
O 1s


(b)


II









1 1 1 1 | 1 l | l l I l l I
C 1s 'i j


300 295 290 285 280

Binding Energy (eV)


Figure 3-4. Overlays of XPS C 1s spectra obtained (a) after the 24-hr UV exposure (solid
line) and the 02 exposure (dashed line), (b) after the 02 exposure (dashed
line) and 28-hr UV exposure (solid line), (c) after the air exposure (dashed
line) and the 28-hr UV exposure (solid line) and (d) after the air exposure
(dashed line) and the 74-hr UV exposure (solid line).









l""""'l""""'l"""'l""""'""""'l""""'""""'1""""'


I~ ___


-F-]E 1s


O 1s


CR
C,

IJ

ed
k
r~l

ed

W


-x2- 20
N 1s_


~cw*I~J


V1
O


(a) Cr,


(b)
A


800 700 r600


Binding Energy (eV)


Figure 3-5. XPS survey spectra obtained from UV-exposed TFE Teflon (a) after a 90-
min exposure to air at room temperature and (b) after a total of 74 hrs
exposure to UV (46 hrs of UV exposure after the air exposure).


'v~~


-- ---w I-
h........h... ..1111h111111111h111111111In1111 111h111111111hn11 1nl1111111111


500 400 300 200 100



























_~ i (a)
300 29 9 8 8
Bindn Enry(V

Figre 3-6 Hihrslto P s pcr bandfo A -xoe dse ie
an Vepsd(oi ie)TETfo a setre,()atrepsr o
2 r,()atrepsr o oa f2 r,()atrepsr o0 tro
tepeaur nd10 or o 2 in e)ate xpsrefr otlof2 hs
(f) afte exosr for a toa f2 rs g fe xpsr oar o 0mna
room tmperaure an (h) fte exouefrattlo 4hs













F C F/C
as entered 66.5 33.5 1.98
UV 2 hrs 62.3 37.7 1.65
UV 24 hrs 61.2 38.8 1.58
02 20 min 60.7 39.3 1.54
UV 25 hrs 61.0 39.0 1.56
UV 28 hrs 58.9 41.1 1.43
Air 90 min 56.9 43.1 1.32
UV 74 hrs 61.6 38.4 1.60


42


Table 3-1. Near-surface Compositions of TFE Teflon after Various Treatments (at%)
















REFERENCES

1. B.A. Banks, The Use of Fluoropolymers in Space Applications. Modern
Fluoropolymers, John Wiley & Sons, New York 1997.

2. L.J. Leger and J.T. Visentine, J. Spacecraft Rockets, 23, 505 (1986).

3. S.L. Koontz, L.J. Leger, J.T. Visentine, D.E. Hunton, J.B. Cross and C.L. Hakes, J.
Spacecraft Rockets, 32, 483 (1995).

4. R.C. Tennyson, Can. J. Phys, 69, 1190 (1991).

5. S.L. Koontz, L.J. Leger, S.L. Rickman, C.L. Hakes, D.T. Bui, D.E. Hunton and
J.B. Cross, J. Spacecraft Rockets, 32, 475 (1995).

6. M. R. Reddy, J. Mater. Sci., 30, 281 (1995).

7. S. Packirisamy, D. Schwam and M.H. Litt, J. Mater. Sci., 30, 308 (1995).

8. M.R. Reddy, N. Srinivasamurthy and B.L. Agrawal, Eur. Space Agency 1., 16, 193
(1992).

9. K. K. de Groh, and B. A. Banks, J. Spacecraft Rockets, 31, 656 (1994).

10. E. Grossman, Y. Lifshitz, J.T. Wolan, C.K. Mount and G.B. Hoflund, J. Spacecraft
Rockets, 36, 75 (1999).

11. R.I. Gonzalez, S.H. Phillips and G.B. Hoflund, J. Spacecraft Rockets, 37, 463
(2000).

12. S.H. Phillips, G.B. Hoflund and R.I. Gonzalez, Proc. 45th International SAM\~PE
Symposium, 45, 1921 (2000).

13. G. B. Hoflund, R. I. Gonzalez, S. H. Phillips, J. Adhesion Sci. Technol., 15, 1199
(2001).

14. K.K. DeGroh, J.R. Gaier, M.P. Espe, D.R. Cato, J.K. Sutter, and D.A. Scheiman,
High Perform. Polym., 12, 83 (2000).

15. J.A. Dever, K.K. DeGroh, B.A. Banks, J.A. Townsend, J. L. Barth,, S. Thomson, T.
Gregory, W. Savage, High Perform. Polym., 12, 125 (2000).










16. F.A. Rasoul, D.J.T. Hill, J.S. Forsythe, J.H. O'Donnell, G.A. George, P.J. Pomery,
P.R. Young, and J.W. Connell, J. Appl. Polym. Sci., 58, 1857 (1995).

17. A. Gindulyte, L. Massa, B.A. Banks and S.K.R. Miller, J. Phys. Chem. A, 106,
5463 (2002).

18. G. R. Corallo, G.B. Hoflund and R.A. Outlaw, Surf: Interface Anal., 12, 185
(1988).

19. M.R. Davidson, G.B. Hoflund, and R.A. Outlaw, Surf: Sci., 281, 111 (1993).

20. R.A. Outlaw, G.B. Hoflund, and G.R. Corallo, Appl. Surf: Sci., 28, 235 (1987).

21. E. Wisotzki, A.G. Balogh, H. Horst, J.T. Wolan, G.B. Hoflund, J. Vac. Sci.
Technol. A, 17, 14 (1999).

22. E. Wisotzki, H. Hahn, G.B. Hoflund, in Trends and New Applications of Thin
Films, Horst Hoffman, Trans Tech Publications LTD, Switzerland, 1998, 181, pp.
277-278.

23. M. R. Davidson, G. B. Hoflund, R. A. Outlaw, J Vac. Sci. Technol. A, 11, 264
(1993).

24. R.E. Gilbert, D.F. Cox and G.B. Hoflund, Rev. Sci. Instrum., 53, 1281 (1982).

25. A. Savitzky and and M.J.E. Golay, Anal. Chem., 36, 1627 (1964).

26. G. Beam son and D. Briggs (Eds.), High Resolution XPS of Organ2ic Polymers: The
Scienta ESCA300 Database, pp. 54, 226-236, Wiley, Chichester (1992).

27. C.D. Wagner, W.M., Riggs, L.E Davis, J.F. Moulder and G.E. Muilenberg (Eds.),
Handbook ofX-ray Photoelectron Spectroscopy, Perkin-Elmer Corp., Eden Prairie,
MN, (1979).

28. G.B. Hof lund in: Handbook of Surface and Interface Analysis: M~ethods in
Problem Solving, J.C. Riviere and S. Myhra (Eds.), pp. 57-158 Marcel Dekker,
New York (1998).

29. G.B. Hoflund and M.L. Everett, In press.

30. M.L. Everett and G.B. Hoflund, In press.

31. Hoflund, G. B., Scanning Elect. M~icrosc., 4, 1391 (1985).

TFE Teflon FEP Teflon Tefzel", Tedlar" and Kapton" are registered trademark
names by E.I. du Pont Nemours & Co., Inc.
















BIOGRAPHICAL SKETCH

Michael L. Everett was born on February 23, 1978, in Springfield, Massachusetts,

to Wayne R. Everett and Kathleen S. Everett. He graduated from Springfield Central

High School in 1996. That fall Michael began his college education at Worcester

Polytechnic Institute in Worcester, Massachusetts. In May 2000 he received his Bachelor

of Science in chemical engineering. Michael then entered the Graduate School at the

University of Florida in Gainesville, Florida, to further his education in chemical

engineering. As a graduate student he began working under Dr. Gar B. Hoflund,

performing research in surface science.