Epoxy Resin Joint by Electron Beam to PET Prior to Assembly

Adhesive Force Improvement of Polyethylene Terephthalate (PET)/Epoxy Resin Joint by Electron Beam to PET Prior to Assembly

Chisato Kubo1,+, Masae Kanda2, Kenshin Miyazaki1,+, Takumi Okada1,+, Michael C. Faudree1 and Yoshitake Nishi1,2

1Graduate School of Engineering, Tokai University, Hiratsuka 259-1292, Japan 2School of Engineering, Tokai University, Hiratsuka 259-1292, Japan

Adhesive 2-layer lamination joints of polyethylene terephthalate (PET)/epoxy resin were prepared without the use of hot press using a new adhesion method of applying homogeneous low energy electron beam irradiation (HLEBI) to the PET prior to assembly by hand pressure. o HLEBI treatment within the range of 0.13 to 0.43 MGy increased the adhesive force of peeling ( Fp) substantially over the untreated. The largest o ¹1 Fp values at optimal dose 0.30 MGy were 12.4, 37.2 and 190 Nm , which were more than 2.2, 2.8 and 9.5 times larger than 5.76, 13.5 and ¹1 20.0 Nm of the untreated at low-, median- and high peeling force accumulative probability, Pp of 0.06, 0.50 and 0.94, respectively. The o statistically lowest Fp for safety design (Fs at Pp = 0) iterated by the 3-parameter Weibull equation was raised from zero for the untreated to 10.7 Nm¹1 for the 0.30 MGy samples indicating increased reliability by the HLEBI. XPS (X-Ray Photoelectron Spectroscopy) observations of the peeled 0.30 MGy HLEBI PET revealed generation of a C-O peak at 286 eV possibly explaining the increased adhesion. Therefore residual epoxy deposition is apparently found to be retained on the PET sheet by inter-matrix fracture of epoxy resin further into the thickness. This can be explained by the adhesion force from crosslinking between PET/epoxy being stronger than the cohesive force of epoxy polymer itself. Since o the experimental data shows the optimum HLEBI dose is about 0.30 MGy, above which at 0.43 MGy the Fp begins to drop, carefulness in optimization is highly recommended when applying in industry to insure safety. [doi:10.2320/matertrans.M2015082]

(Received February 27, 2015; Accepted August 27, 2015; Published October 9, 2015) Keywords: joint, adhesion, electron beam, irradiation, polyethylene terephthalate (PET), epoxy, crosslinking

1. Introduction HLEBI is surface treatment that cuts the chemical bonds at the polymer surface and generates active terminated atoms in Joining polymers has extensive application hence advance- polymers generating strong adhesive bond by crosslinking ments in techniques are always highly sought after. Recent at the PET/epoxy joint interface. Total treatment time of advances for joints include impact resistant joints of pressure HLEBI application is only a few seconds preventing excess sensitive adhesive (PSA) used in smart phones,1) thermo- heating. plastic adhesive containing SiC layers for high frequency However, if radiation is too high it degrades materials for welding of polypropylene (PP)2) and its glass fiber reinforced example, NASDA reports PET films for thermal control polymer (GFRP).3) materials used in space are given tolerance limits certified up Applying surface treatment of low dose of electron beam to 3 MGy, and 1 MGy where high optical transmission is (EB) irradiation on the order of 0.01 to 1 MGy has been expected.13) Moreover, for the PET/epoxy joint when HLEBI gaining momentum as a successful method to adhere over 0.43 MGy is applied the peeling strength starts to polymeric materials without the use of adhesive. Polyimide/ degrade therefore carefulness in optimization is highly polytetrafluoroethylene (PI/PTFE) layered films have been recommended when adjusting dose for strengthening prac- successfully fabricated by Homogeneous low energy electron tical parts for maximum safety. Therefore, the purpose of the beam irradiation (HLEBI);4) while applying 0.50 MGy EB research is reproducibility to obtain the large adhesive force o dose increased adhesive bond strength between PET films of peeling resistance ( Fp) for the PET/Epoxy polymers joint and acrylic adhesive.5) HLEBI has been found to increase by applying 0.30 MGy-HLEBI to the PET joining surface adhesive mechanical properties of polymer-polymer joints for prior to lamination assembly by hand pressure without use of biomedical applications of PDMS (polydimethylsilozane)/ glue or hot press. PTFE,6) PDMS/PP (polypropylene),7) and create strong adhesion in the difficult to bond PTFE/PE (polyethylene).8) 2. Experimental Procedure Increase in adhesion by EB is reported to occur by crosslinking between the two polymer surfaces as in rubbers9) 2.1 Preparation of PET/Epoxy laminated sheets and polyurethane (PU) based adhesives,10) and dangling bond For epoxy preparation, epoxy resin and curing agent (Craft generation: scission of bonds detected by electron spin Resin, NISSIN RESIN Co. Ltd., Japan) were mixed in a 5 : 1 resonance (ESR) creating reactive sites in PDMS/PP.7) In the by weight ratio, followed by applying heat at 350 K for 1950s Charlesby-Pinner derived an expression for crosslink 10 min. Next, the epoxy resin was spread onto a 15 µm thick density as a function of radiation dose dependent on sol nylon 6 supporting film with uniform film thickness (³250 fraction, fracture density of main chain scission per unit dose, +/¹ 10 µm, measured after drying) using a doctor blade and initial number average degree of polymerization.11) One (Elcometer, Belgium). To a thin PET (polyethylene tereph- assumption they made crosslinking is independent of chain thalate) sheet (10 © 40 © 0.50 mm, Teijinμ Tetoronμ Film, scission.11,12) Teijin DuPont Films, Japan) HLEBI was then applied to one side. Following this, the HLEBI-treated side of the PET sheet +Graduate Student, Tokai University was laminated to the epoxy resin with hand pressure, then 1822 C. Kubo et al. dried for 1 day at 296 K below the components’ glass transition temperatures, Tg of 389 and 353 K, of epoxy resin and PET, respectively. Since the PET was HLEBI-treated, the joint is referred to as “PET/Epoxy”. Note although the epoxy resin by itself was heated during preparation, hot press was not used on the PET/Epoxy joint.

2.2 Homogeneous low energy electron beam irradiation (HLEBI) Fig. 1 Schematic diagram of T-Peeling test of the PET/Epoxy polymers The PET sheet was irradiated by using an electron-curtain laminate sheet. processor (Type CB175/15/180L, Energy Science Inc., Woburn, MA, Iwasaki Electric Group Co. Ltd. Tokyo).1423) The specimen was homogeneously irradiated with the Based on eq. (2), the dropped potential values, "VTi and sheet HLEBI with low energy through a titanium thin film "VN2 are estimated from the acceleration potential (170 keV), window attached to a 240 mm diameter vacuum chamber. A the 10 µm thickness (TTi) of the titanium window (density: tungsten filament in a vacuum is used to generate the electron 4540 kg m¹3), and the 35 mm distance between the sample beam at a low energy (acceleration potential, V: kV), of and the window (TN2) in the N2 gas atmosphere (density: 2 ¹3 170 keV and irradiating current density (I,A/m ) of 0.089 µN2 = 1.13 kg m ). A/m2. V ¼ðT =D Þ170 keV Although the sheet electron beam generation is in a Ti Ti thTi ¼ T µ =½ : ð Þ2=3 vacuum, the irradiated sample has been kept under protective Ti Ti 66 7 170 keV nitrogen at atmospheric pressure. The distance between ¼ð105 mÞð4540 kg m3Þ=½66:7 ð170 keVÞ2=3 sample and window is 35 mm. To prevent oxidation, the ¼ 22:2 keV ð3Þ samples are kept in a protective atmosphere of nitrogen gas V ¼ðT =D ÞV with a residual concentration of oxygen below 400 ppm. The N2 N2 thiN2 Ti fl / ¼ T µ =½ : ðV Þ2=3 ow rate of nitrogen gas is 1.5 L s at 0.1 MPa nitrogen gas N2 N2 66 7 Ti pressure. ¼ð35 10 3 mÞð1:13 kg m3Þ The absorbed dose is controlled by the integrated 2=3 irradiation time in each of the samples. Here, absorbed dose =½66:7 ð170 keVÞ ð4Þ is corrected from irradiation dose by using an FWT nylon Since the dropped potential values are 22.2 and 18.2 keV, dosimeter of RCD radiometer film (FWT-60-00: Far West the specimen surface electrical potential, V is 129.6 keV as Technology, Inc. 330-D South Kellogg Goleta, California follows. 93117, USA) with an irradiation reader (FWT-92D: Far West V ¼ 170 keV 22:2 keV 18:2 keV ¼ 129:6 keV ð5Þ Technology, Inc. 330-D South Kellogg Goleta, California 93117, USA). The absorbed dose corresponded to irradiation Given typical density of PET is 1380 kg m¹3, the HLEBI dose is 0.0432 MGy at each irradiation, which is applied for depth into the PET film estimated from eq. (1) is Dth = only a short time (0.23 s) to avoid excessive heating of the 238 µm, more than four times larger than the PET thickness sample; the temperature of the sample surface remains below of 50 µm, hence the HLEBI penetrated through the entire 323 K just after irradiation. The sample in the aluminum plate thickness. holder (0.15 m © 0.15 m) is transported on a conveyor at a speed of 9.56 m/min. The sheet HLEBI is applied inter- 2.3 T-peeling test mittently. Repeated irradiations to both side surfaces of the Composite samples after removing the 15 µm thick nylon samples are used to increase the total irradiation dose. The 6 supporting film were prepared for the T-peeling test to interval between the end of one period of irradiation and the evaluate the influence of HLEBI on the mean adhesive force o start of the next operation is 30 s. When the irradiation of peeling resistance ( Fp), as shown in Fig. 1. Peeling current (I, mA), the conveyor speed (S,m/min) and number adhesive force (Fp) vs. peeling distance (dp) curves were of irradiations (N) are determined, the irradiated dosage is obtained by using a micro-load tensile tester (F-S Master-1K- proportional to the yield value from the irradiation current 2N, IMADA Co. Ltd., Japan) with a strain rate of 10 mm/ 6) ¹1 o (I, mA), the conveyor speed (S,m/min), and number of min. Since the unit of the Fp was Nm , the Fp was used irradiations (N). instead of the adhesive strength, whose units should be Based on the density (µ:kg/m3) and irradiation voltage at Nm¹2. The sample condition of tensile test was as follows: the specimen surface (V: kV), the penetration depth (Dth:m) (1) The vertical length from the peeling contact point to the of HLEBI is expressed by the following equation.24) end of the sample was 5 mm.

5=3 (2) The Fp was determined by using micro-load tensile Dth ¼ 66:7V =µ ð1Þ o tester. The Fp was estimated by the peeling load and Specimen surface electrical potential (V ) was mainly reduced experimental peeling width and length of 10 and 30 mm, going through the Ti window ("VTi) and N2 gas atmosphere respectively. The initial distance before peeling (di) was ("VN2). deﬁned at the start point of peeling force, which corresponds to the start point of the ﬁrst relaxation. The d value is V ¼ 170 keV V V ð2Þ i Ti N2 ³1mm. Adhesive Force Improvement of Polyethylene Terephthalate (PET)/Epoxy Resin Joint by Electron Beam to PET Prior to Assembly 1823

4 1.0

0.30 MGy 0.8 3 Untreated p P / N p L 0.6 2

0.4 Untreated

Peeling Load, Load, Peeling 1 0.04 MGy

Peeling Probability, Probability, Peeling 0.13 MGy 0.2 0.22 MGy 0.30 MGy 0 0.43 MGy 0 10 20 30 40 0 100 101 102 103 Peeling Distance, d /mm p ࠋ -1 Mean Adhesive Force, Fp / Nm Fig. 2 Peeling load (Lp)-peeling distance (dp) curves of PET/Epoxy o laminated sheets before and after 0.30 MGy-HLEBI to PET at Pp of 0.50. Fig. 3 Relationships between Fp and the accumulative probability of peeling force (Pp) of PET/Epoxy laminated sheets untreated and HLEBI- treated to PET.

2.4 X-ray photoelectron spectrometer (XPS) measure- ments 103 X-ray photoelectron spectrometer (XPS: Quantum 2000, P = 0.94 ULVAC Co., JAPAN)6) was used for surface analysis of p -1 Pp = 0.50 peeled 0.30 MGy HLEBI PET and unassembled untreated Pp = 0.06 2 PET. Both PET and Epoxy resin contain elements C, H and / Nm 10 p O. Narrow scans for the C (1s) and O (1s) signals from the F ࠋ PET surfaces were detected by the XPS.

3. Results 101

3.1 Peeling load (Lp)-peeling distance (dp) curve Force, Adesive Figure 2 shows a comparison of Lp (N) vs. peeling distance, dp (mm) curves between HLEBI and untreated 100 PET/Epoxy joint at median accumulative probability of 0 0.1 0.2 0.3 0.4 0.5 = peeling force, Pp 0.50. Although without HLEBI a large Absorbed Dose, D / MGy adhesive load of peeling resistance in the PET/Epoxy could o / not be obtained, by applying HLEBI at 0.30 MGy the peeling Fig. 4 Changes in experimental Fp at each Pp of PET Epoxy laminated sheets at low-, median-, and high-Pp of 0.06, 0.50 and 0.94 against load, Lp is significantly increased (³1.9 N) over the low value absorbed dose to PET. of the untreated (³0.20 N). The 0.30 MGy-HLEBI therefore laminates the epoxy resin with the PET sheets, generating the higher peeling resistance. Fracture was observed to always occur at the interface. The curves are flat, indicating stability 0.30 MGy-HLEBI samples at 12.4, 37.2 and 190 Nm¹1, o and homogeneity throughout the 35 mm peeling distance up respectively. The Fp at low- and median-Pp of 0.06 and 0.50, until fracture. were 12.4 and 37.2 Nm¹1, respectively which were more than 2.2 and 2.8 times larger than 5.76 and 13.5 Nm¹1 before o o 3.2 Adhesive force of peeling ( Fp) as a function of treatment. All Fp values of PET/Epoxy laminated sheets accumulative probability of peeling force (Pp) with 0.30 and 0.43 MGy apparently exceed all corresponding Figure 3 plots the relationships between adhesive force of values of untreated samples. Thus, adhesion of PET/Epoxy o peeling ( Fp) at each accumulative probability of peeling laminated sheets with 0.30 and 0.43 MGy-HLEBI deems force (Pp) of the PET/Epoxy laminated sheets for the effective. untreated and HLEBI-treated showing applying 0.30 MGy o HLEBI gives the highest Fp values, particularly above 4. Discussion Pp > 0.2. Most notably, at high-Pp of 0.94 the 0.30 MGy o HLEBI raised the Fp significantly, 850% from 20 to 4.1 The statistically lowest adhesive force ¹1 o 190 Nm .AtPp = 0.85 the Fp was raised 584% from 19 In order to obtain the statistically lowest peeling stress for ¹1 o to 130 Nm . safety design, the lowest Fp value at Pp = 0(Fs) is assumed o Figure 4 shows the maximum Fp for low-, median-, and to be attained from the adaptable relationship of the 3- high-Pp of 0.06, 0.50 and 0.94 against HLEBI occurs in the parameter Weibull equation iterating to the high correlation 1824 C. Kubo et al.

1.00 15 -1 F 0.95

/ Nm 10 p F ࠋ

0.90 Untreated 5 Pp = 0.06 0.04 MGy Pp = 0 lation Coefficient, Coefficient, lation 0.13 MGy 0.85 0.22 MGy Force, Adesive Corre 0.30 MGy 0.43 MGy 0 0.80 0 0.1 0.2 0.3 0.4 0.5 0 2 4 6 8 10 12 14 EB Irradiation Dose, D / MGy

e -1 = o = Lowest Adhesive Force, Fs / Nm Fig. 7 Changes in the lowest adhesive force ( : Fs Fp at Pp 0) calculated for PET/Epoxy laminated sheets against HLEBI dose to PET, ﬁ o Fig. 5 Changes in correlation coef cient (F ) of eq. (1) against the together with experimental Fp at low-Pp of 0.06. o e potential Fs value ( Fs) for PET/Epoxy laminated sheets at each absorbed dose of HLEBI to PET. 2500 Untreated before assembly 2 Untreated after assembly 0.30 MGy before assembly 0.30 MGy after assembly

2000 C-H 1

)] 1500 p P

0 C-O c / s 1000 -1 O – C = O ln[ - ln (1 - 500 -2

0 -3 295 290 285 280 275 270 265 -2 0 2 4 6 ࠋ -1 Binding Energy (eV) ln[( Fp - Fs)] / ln Nm Fig. 8 XPS signals of carbon (C (1s)) of peeled surface of PET sheets o Fig. 6 Liner relationships between ln( Fp ¹ Fs) and ln[¹ln(1 ¹ Pp)] for untreated and 0.30 MGy HLEBI of PET/epoxy lamination, together with PET/Epoxy laminated sheets at each absorbed dose of HLEBI to PET. single PET sheets untreated and 0.3 MGy-irradiated before assembly. coefficient (F). The Pp depends on the risk of rupture Fs values of the PET/Epoxy laminated sheets over that of the o 6,2531) ([ Fp ¹ Fs]/FIII). untreated. The 0.30 MGy-HLEBI apparently enhances the Fs ¹1 ¹1 P ¼ ½ð½oF F =F ÞmðÞ from 0 N·m for the untreated to 10.7 N·m ; as well as at p 1 exp p s III 6 o low Pp of 0.06 (the lowest experimental Fp) from 5.8 for the o ¹1 The FIII value is the Fp value, when the term ln[¹ln(1 ¹ untreated to 12.4 N·m . Consequently, HLEBI enhances the o Pp)] is zero. When Pp = 0, the required Fp value to evaluate safety level (reliability) of PET/Epoxy laminated sheets. This new structural materials is defined as the Fs. In predicting indicates HLEBI induced adhesion can be applied to practical the Fs, coefficient (m) and constant (FIII) are the key articles with sterilization without volatilization, when the parameters. Figure 5 plots the iteration to obtain the highest adhesive force of peeling resistivity is less than 10.7 N·m¹1. correlation coefficient (F) with respect to the potential o e adhesive force of peeling Fs value ( Fs) estimated from the 4.2 X-ray photoelectron spectrometry (XPS) of PET logarithmic form. surface Figure 6 illustrates the linear relationships between Figures 8 and 9 show fracture surface analysis by X-ray o ln( Fp ¹ Fs) and ln[¹ln(1 ¹ Pp)]. The values of FIII and m photoelectron spectrometry (XPS) of carbon (C (1s)) and are determined by the least-squares best fit method. The m oxygen (O (1s)) signals of the peeled surface of PET sheets value is estimated by the slope of the relationship when untreated and 0.30 MGy HLEBI to PET of PET/epoxy e Fs = Fs. laminated, together with single PET sheets untreated and Figure 7 shows Fs is always lower than the experimental 0.3 MGy-irradiated before assembly. Results indicate the o Fp value. The HLEBI from 0.13 to 0.43 MGy improves the HLEBI acts to produce adhesion in the PET/epoxy Adhesive Force Improvement of Polyethylene Terephthalate (PET)/Epoxy Resin Joint by Electron Beam to PET Prior to Assembly 1825

2000 untreated PET sheet after lamination (Fat broken line in C O Fig. 8), although the intensities of C-H and O-C=O peaks of

O- laminated sheets doesn’t change by HLEBI. It is explained

C- that HLEBI activates the PET surface. The active PET attracts O = C - the oxygen atoms from atmospheric molecules with increas- Untreated before assembly Untreated after assembly ing oxygen content. On the contrary, the active PET easily 0.30 MGy before assembly 0.30 MGy after assembly adheres the Epoxy with decreasing the oxygen content at 1000 PET surface. Based on the both reaction, the adhesive force c / s was increased. Therefore residual epoxy deposition is apparently found to be retained on the PET sheet by inter- matrix fracture of epoxy resin further into the thickness. This can be explained by the adhesion force from cross-linking between PET/epoxy being stronger than the cohesive force of epoxy polymer itself. 0 In Fig. 9 the XPS narrow scan of oxygen (O (1s)) of the 545 540 535 530 525 520 515 peeled 0.30 MGy sample shows peaks at 532 and 533.5 eV Binding Energy (eV) corresponding with the O (1s) in O=C-O and C-O-C groups. In order to calibrate the results in detail, XPS signals of O Fig. 9 XPS signals of oxygen (O(1s)) of peeled surface of PET sheets = untreated and 0.30 MGy HLEBI of PET/epoxy laminated, together with (1s) in O C-O and C-O-C have been obtained for PET with single PET sheets untreated and 0.3 MGy-irradiated before assembly. and without HLEBI before assembly (Narrow solid and Broken lines in Fig. 9). The lowest intensity of O (1s) in O=C-O and C-O-C signals for untreated PET after lamination joint where fracture generally occurred near the lamination is gotten. Since 0.30 MGy-HLEBI remarkably peeled PET-epoxy interface. enhances the intensity, it remarkably enhances the oxygen In Fig. 8 the XPS narrow scan of carbon (C (1s)) of the concentration at PET interface. Since HLEBI activates the peeled 0.30 MGy sample shows peaks at ³284, 286 and terminated polymer atoms, oxygen atoms from atmospheric 288 eV corresponding with the C (1s) in C-H, O-C=O and C- molecules are attracted to surface, and then contaminate the O groups. Since the deceasing carbon ratio of PET to Epoxy polymers. is remarkably smaller than that of the O (1s), the change in C The XPS signals of O (1s) in O=C-O and C-O-C of (1s) peak intensity is smaller than that of O (1s) as predicted. untreated PET sheet before assembly (Narrow broken line in In order to calibrate the results in detail, XPS signals of C Fig. 9) are higher than those of untreated PET sheet after (1s) have been obtained for PET with and without HLEBI lamination (Fat broken line in Fig. 9). It shows that the (Solid and Broken lines in Fig. 8) before assembly (narrow lamination simply increase the oxygen concentration because lines in Fig. 8). The highest intensity of C-H signal for of oxygen contamination effects from atmospheric mole- untreated PET before assembly is gotten. Although cules, as shown in Fig. 8. The contamination probably 0.30 MGy-HLEBI slightly decreases the C-H intensity, it attributes to weak adhesion force of the PET/Epoxy remarkably enhances the C-O and O-C=O intensities. Since lamination untreated, as shown in Figs. 3 and 4. HLEBI activates the terminated polymer atoms, oxygen The oxygen atoms exist as O=C-O before assembly, atoms from atmospheric molecules are attracted to surface, whereas they exist as O=C-O and C-O-C after lamination. and then contaminate the polymers, resulting in increasing The peak intensity value (oxygen concentration) of O (1s) in oxygen concentration at PET interface. C-O-C is equal to that in C=O-C after lamination. Each atomic ratio of elements of carbon, hydrogen and Increase in the peak intensity value (oxygen concentration) oxygen of PET ((C10H8O4)n) and Epoxy ((C18H23O2)n)is of O (1s) in C-O-C induced by the lamination from 263 to (C : 0.455, H : 0.364, O : 0.181) and (C : 0.418, H : 0.535, 1477 c/s is two times larger than that of O=C-O from 846 O : 0.047), respectively. Each atomic ratio of carbon and to 1493 c/s. The oxygen contamination from atmospheric oxygen of PET is 1.09 and 3.85 times higher than those of molecules mainly converts to the oxygen atoms in C-O-C. Epoxy, respectively, whereas the ratio of hydrogen of Epoxy The peak intensity of O (1s) in O=C-O and C-O-C of is 3.85 times higher than that of PET. 0.30 MGy-HLEBI PET sheet after lamination (Fat solid line The XPS signals of C-O and O-C=O of untreated PET in Fig. 9) is slightly lower than that of untreated PET sheet sheet before assembly (narrow broken line in Fig. 8) are after lamination (Fat broken line in Fig. 9). Since atomic ratio higher than those of untreated PET sheet after lamination (Fat of oxygen (O : 0.181) of PET ((C10H8O4)n) is 3.85 times broken line in Fig. 8), whereas the lamination decreases the higher than that (O : 0.047) of Epoxy ((C18H23O2)n), the intensity of C-H signal. Since the lamination increases the reduction of peak of O (1s) in O=C-O and C-O-C in the concentrations of carbon and oxygen and decreases the HLEBI sample possibly indicates epoxy adheres to the PET. hydrogen concentration because of oxygen contamination Therefore, residual epoxy deposition is apparently found to from atmospheric molecules, it probably attributes to weak be retained on the PET sheet by inter-matrix fracture of epoxy adhesion force of the PET/Epoxy lamination untreated, as resin further into the thickness. This can be explained by shown in Figs. 3 and 4. the adhesion force from crosslinking between PET/epoxy The intensity of C-O peak of 0.30 MGy-HLEBI PET sheet being stronger than the cohesive force of epoxy polymer after lamination (Fat solid line in Fig. 8) is higher than that of itself. 1826 C. Kubo et al.

On the other hand, the peak intensity of O (1s) in O=C-O REFERENCES and C-O-C of 0.30 MGy-HLEBI PET sheet after lamination (Fat solid line in Fig. 9) is higher than that of 0.30 MGy- 1) H. Hayashida, T. Sugaya, S. Kuramoto, C. Sato, A. Mihara and T. HLEBI PET sheet before assembly (Narrow solid line in Onuma: Int. J. Adhes. Adhes. 56 (2015) 61 72. 2) M. Sano, H. Oguma, M. Sekine and C. Sato: Int. J. Adhes. Adhes. 47 Fig. 9). The concentration of oxygen is determined by both (2013) 5762. atmospheric molecules and adhering Epoxy to PET. 3) M. Sano, H. Oguma, M. Sekine and C. Sato: Int. J. Adhes. Adhes. 54 (2014) 124130. 5. Conclusions 4) A. Oshima, H. Nagai, F. Muto, T. Miura and M. Washio: J. Polym. Sci. Tech. 19 (2006) 123127. / 5) M. Zenkiewicz: Int. J. Adhes. Adhes. 24 (2004) 259 262. A polyethylene terephthalate (PET) Epoxy joint was 6) Y. Nishi, M. Uyama, H. Kawazu, H. Takei, K. Iwata, H. Kudoh and K. rapidly fabricated for bio-adaptable applications without the Mitsubayashi: Mater. Trans. 53 (2012) 16571664. use of glue or hot press by applying homogeneous low energy 7) Y. Nishi, H. Kawazu, H. Takei, K. Iwata, H. Kudoh and K. electron beam irradiation (HLEBI) surface treatment to the Mitsubayashi: Mater. Trans. 52 (2011) 19431948. PET prior to laminating to the epoxy resin with hand pressure. 8) C. Kubo, T. Okada, M. Uyama, M. Kanda and Y. Nishi: Mater. Trans. 55 (2014) 17421749. (1) By applying 0.30 MGy HLEBI the adhesive force 9) W. Smitthipong, M. Nardin, J. Schultz and K. Suchiva: Int. J. Adhes. o of peeling ( Fp) was increased substantially over the Adhes. 27 (2007) 352357. o untreated. The largest Fp values at optimal dose 0.30 10) A. K. Singh, D. S. Mehra, U. K. Niyogi, S. Subharwal, J. Swiderska, Z. MGy were 12.4 and 37.2 Nm¹1, which were more than Czech and R. K. Khandal: Int. J. Adhes. Adhes. 41 (2013) 7379. 2.2 and 2.8 times larger than 5.76 and 13.5 Nm¹1 before 11) A. Charlesby and S. H. Pinner: Proc. Roy. Soc. A 249 (1959) 367 386. 12) P. J. Flory: J. Am. Chem. Soc. 69 (1947) 3035. treatment at low- and median-accumulative probability 13) National Space Development Agency of Japan (NASDA) Contract of peeling force, Pp of 0.06 and 0.50, respectively. Report: JAXA Repository, Osaka Nuclear Science Association, Study (2) Moreover, at high-Pp of 0.94 the 0.30 MGy HLEBI on radiation effects to materials, Part 4, On evaluation technique of o ¹1 radhard characteristics, Report Number: NASDA-CNT-990013. raised the Fp signiﬁcantly, 850% from 20 to 190 Nm = o % 14) K. Oguri, N. Iwataka, A. Tonegawa, Y. Hirose, K. Takayama and Y. while at Pp 0.85 the Fp was raised 584 from 19 to ¹1 Nishi: J. Mater. Res. 16 (2001) 553 557. 130 Nm . 15) A. Mizutani and Y. Nishi: Mater. Trans. 44 (2003) 18571860. o (3) The statistically lowest Fp for safety design (Fs at 16) K. Sonoda, Y. Kaneda, I. K. Nakazaki, Z. Enomoto and K. Murayama: Pp = 0) iterated by the 3-parameter Weibull equation Proc. Symp. of Space Sciences and Technology, (Utyu kagaku gizyutu was raised from 0 Nm¹1 for the untreated to 10.7 Nm¹1 renngou kouennkai kouennsyu, JST No. S0277A) 28 (1984) pp. 14 for the 0.30 MGy samples indicating increased reli- 15. 17) K. Oguri, K. Fujita, M. Takahashi, Y. Omori, A. Tonegawa, N. Honda, ability over the untreated. M. Ochi, K. Takayama and Y. Nishi: J. Mater. Res. 13 (1998) 3368 (4) Although the ESR peaks of the HLEBI-treated PET 3371. were low intensity, applying the HLEBI to the PET 18) K. Oguri, N. Iwatani, H. Izumi, A. Tonegawa, K. Takayama and Y. made it possible to raise the peel force of the laminated Nishi: Proc. 2nd Japan-France Seminar on Intelligent Materials and PET/Epoxy. Structures, (University of Louis Pasteur Strasbourg, France) (1998) pp. 142144. (5) The adhesion mechanism is HLEBI generally cuts the 19) Y. Nishi, A. Mizutani and N. Uchida: J. Thermoplastic Compos. Mater. atomic bonding at weak chemical bonding sites and 17 (2004) 289302. dangling bonds are formed at the terminated atoms with 20) Y. Nishi, T. Toriyama, K. Oguri, A. Tonegawa and K. Takayama: low dissociation energies in the PET at methylene and J. Mater. Res. 16 (2001) 16321635. carboxylate groups. The dangling bonds were probably 21) Y. Nishi, A. Mizutani, A. Kimura, T. Toriyama, K. Oguri and A. Tonegawa: J. Mater. Sci. 38 (2003) 8992. in the form of free radicals bonding the PET and epoxy 22) K. Inoue, K. Iwata, T. Morishita, A. Tonegawa, M. Silvia and Y. Nishi: by cross-linking. Int. J. Appl. Electromag. Mech. 23 (2003) 251256. (6) Larger HLEBI dose of more than 0.43 MGy decreased 23) Y. Nishi, K. Inoue and M. Salvia: Mater. Trans. 47 (2006) 28462851. the oF as general radiation damage. Therefore careful- 24) M. Kanda, K. Yuse, B. Guiffard, L. Lebrun, Y. Nishi and D. Guyomar: p ness to optimize the HLEBI dose for practical Mater. Trans. 53 (2012) 1806 1809. 25) H. Takei, M. Salvia, A. Vautrin, A. Tonegawa and Y. Nishi: Mater. applications is highly recommended. Trans. 52 (2011) 734739. 26) K. Iwata and Y. Nishi: Mater. Trans. 49 (2008) 20582062. Acknowledgements 27) Y. Nishi, H. Kobayashi and M. Salvia: Mater. Trans. 48 (2007) 1924 1927. The authors wish to thank Prof. Akira Tonegawa of Tokai 28) N. Tsuchikura, M. C. Faudree and Y. Nishi: Mater. Trans. 54 (2013) 371379. University for his useful help. Our sincere gratitude also goes 29) H. Takei, K. Iwata, M. Salvia, A. Vautrin and Y. Nishi: Mater. Trans. 51 to Eye Electron Beam Co., Ltd. (Gyoda, Saitama, Japan) for (2010) 22592265. their support with this work. This work was partly supported 30) W. Weibull: Ingeniörs Vetenskaps Akademien,nr.153 (Generalstabens by the Japan Science and Technology Agency (JST) A-STEP litograﬁska anstalts förlag, Stockholm, 1939) pp. 1622. Program. 31) Yu. P. Yampolskii: Russian Chem. Rev. 76 (2007) 59 78.