Effect of molecular weight of polyurethane toughening agent on adhesive strength and rheological characteristics of automotive structural adhesives

Author’s Accepted Manuscript Effect of molecular weight of polyurethane toughening agent on adhesive strength and rheological characteristics of automotive structural adhesives Daeyeon Kim, Dong Geun Lee, Jin Chul Kim, Choong Sun Lim, Nam Sik Kong, Jin Hong Kim, Hyun Wook Jung, Seung Man Noh, Young Il Park

PII: DOI: Reference:

www.elsevier.com/locate/ijadhadh

S0143-7496(16)30249-4 http://dx.doi.org/10.1016/j.ijadhadh.2016.12.006 JAAD1939

To appear in: International Journal of Adhesion and Adhesives Received date: 28 May 2016 Accepted date: 6 December 2016 Cite this article as: Daeyeon Kim, Dong Geun Lee, Jin Chul Kim, Choong Sun Lim, Nam Sik Kong, Jin Hong Kim, Hyun Wook Jung, Seung Man Noh and Young Il Park, Effect of molecular weight of polyurethane toughening agent on adhesive strength and rheological characteristics of automotive structural a d he s i ve s , International Journal of Adhesion and Adhesives, http://dx.doi.org/10.1016/j.ijadhadh.2016.12.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Effect of molecular weight of polyurethane toughening agent on adhesive strength and rheological characteristics of automotive structural adhesives Daeyeon Kima1, Dong Geun Leeb1, Jin Chul Kima, Choong Sun Lima, Nam Sik Konga, Jin Hong Kima, Hyun Wook Jungb* Seung Man Noha*, Young Il Parka* a

Research Center for Green Fine Chemicals, Korea Research Institute of Chemical Technology, Ulsan, 44412, Republic of Korea

b

Department of Chemical and Biological Engineering, Korea University, Seoul 02841, Republic of Korea

[email protected] [email protected] [email protected] *

Corresponding authors: Dr. Y.I. Park, Dr. S.M. Noh, Prof. H.W. Jung,

Abstract Rheological and adhesive properties of automotive structural adhesives including polyurethane (PU) toughening agents have been investigated by adjusting the molar ratio of polytetrahydrofuran (polyTHF) and hexamethylene diisocyanate (HMDI) for the control of molecular weight of a PU prepolymer. Thixotropic behavior, crosslinking characteristics, and lap-shear and T-peel adhesion strengths for various adhesives were significantly affected by single or binary-mixed PUs with different molecular weights. Thixotropic hysteresis loop of shear viscosity along with shear rate of an adhesive became larger with increasing PU molecular weight, exhibiting favorable flowability, however, the adhesion strengths were not satisfactory for injection and coating applications. We found that both rheological and adhesive properties of adhesives could be interestingly tuned by combining PU components with different molecular weights.

1

These authors contributed equally to this work.

1

Keywords: polyurethane, structural adhesive, rheology, adhesion strength, thixotropic, lapshear strength

1. Introduction The concept of automotive weight lightening is becoming increasingly important under the current situation in which environmentally-friendly automotive technologies should be incorporated into cars to ensure high energy savings along with low CO2 emission. This has led to the rapid development of light-weight technologies for a large variety of automotive applications, in particular, to achieve novel structural adhesives that can replace welding and riveting processes. Most structural adhesives exhibit unique features such as excellent joint durability and outstanding corrosion resistance. These positive effects are the result of increased homogeneity in the distribution of stress over the entire automotive surface and better absorption of stress loads in comparison to welding and riveting processes. Thus, structural adhesive applications can be effectively implemented in the automotive assembly line for the enhancement of mechanical and production characteristics of light-weight car bodies [1-3]. Formulations for automotive adhesives usually involve epoxy resins with special chemical structures, which are crosslinked along with intensive bonding strength on a metal surface, providing outstanding mechanical properties and superb chemical resistance [4]. A critical drawback of such epoxy resins is an intrinsic weakness in static-dynamic strength owing to the brittleness of Bisphenol A epoxy resin, leading to a deterioration of adhesive properties due to the insufficient elasticity of the epoxy. To explore complementary resins that are more capable of elastic bonding and toughening than epoxy resins, a novel class of polyurethanes has been designed, which is very favorable for flexible bonding and attaining superior environmental resistance by controlling the ratio between soft and hard segments [5]. Therefore many researchers have investigated the role of polyurethane (PU) resin as a toughening agent [6-8]. To fully consider the superb functionality of PU in adhesives, it will be important to scrutinize further how the molecular weight of PU as a toughening agent tunes rheological, mechanical, and adhesive properties of epoxy-based adhesive systems.

2

The rheological characteristics of structural adhesives depend considerably on the shear viscosity profiles under various external shear stresses due to the movement changes of their polymeric chains, leading to thixotropic behavior. Thixotropic patterns from shear viscosity data represent the ability of an adhesive to hold its intrinsic shape and are mainly affected by the differences in polymeric chain lengths within adhesives. The thixotropic index, defined as the thixotropic loop area within a given shear rate range [9, 10], of an adhesive with a longer chain length typically shows a relatively high value compared to that of an adhesive with a shorter chain length. In particular, controlling the rheological thixotropic loop of an adhesive under very high-shear rate conditions (at the injection and coating stages) is very important in automotive production lines. In other words, viscoelastic properties of adhesives, e.g., elastic and viscous moduli under the frequency sweep test and compliance under the creep test, should be systematically scrutinized, because these factors reflect the conditions for adhesiveapplication processes. Although adhesives with high solid contents possess some intriguing advantages, there are critical issues that must be overcome before they can be utilized for various applications. One of the most important problems is the meticulous rheological control of structural adhesives, which determines their quality and process applicability in an automotive production line. The viscosities of adhesives can be functions of time as well as shear rate. Previous studies have shown that adhesives should be carefully designed on the basis of rheological information such that their performance remains superior throughout the handling stages from storage to the high-pressurized injection application and curing steps. For example, Hicks et al. [11] described the possible degradation of adhesive properties at the dispensing stage in the case of a large hysteresis loop pattern in the shear viscosity data. Banea et al. [12] developed the inductive heating method for easy bonding of adhesive joints by adding new thermally expandable particles. Baron et al. [13] investigated the rheological and adhesive features of pressure-sensitive adhesives with PU graft copolymers. JaureguiBeloqui et al. [14] reported the effect of silica particles with different surface areas on the rheological properties of PU adhesives. The desirable rheological features generally enable adhesives to have a high enough viscosity during storage to minimize settling phenomena. Under moderate agitation and circulation conditions, their viscosity needs to be reduced for easy flowability. At the high-pressured application step, the viscosity should be very low so that the adhesive can be applied efficiently to the automotive body. As soon as the adhesive is coated on the body, its viscosity should build up quickly to prevent running or sagging. Note 3

that the viscosity level must be adequately controlled for adhesive leveling in this application step. In addition, steady-state viscosity profiles, creep flow, and thixotropic behaviors are directly related to the intrinsic polymer-based adhesive properties in the process. In this study, new synthetic PUs with different molecular weights were prepared by manipulating the isocyanate/polytetrahydrofuran (polyTHF) molar ratio for use as toughening agents in automotive structural adhesives and incorporated with epoxy-based resins to compare rheological properties and adhesive strengths for applications under various conditions. Furthermore, the idea that various properties of adhesives could be tuned by adjusting the mixing ratio of PUs with different molecular weights was substantiated.

2. Experimental 2.1. Materials 2.1.1. Synthesis of polyurethanes PolyTHF (Mn = 2,000 g/mol), hexamethylene diisocyanate (HMDI), trimethylolpropane (TMP), and 2-allylphenol (2-AP, 98%) were supplied by Sigma-Aldrich and utilized to synthesize PU pre-polymers as toughening agents. Dibutyltindilaurylmercaptide (DBTDL) (SND-2800, 95%, Gelest) was added as a polymerization catalyst. PolyTHF and TMP as a chain extender were dehydrated over molecular sieves at 50oC for 48 h and pre-heated in a 500 mL reactor at approximately 90oC for 30 min under N2 until all materials had melted. To synthesize isocyanate-functionalized PU pre-polymers, HMDI and DBTDL were subsequently charged into a reactor. The molecular weight distribution of PU can be adjusted via the molar ratio between polyTHF and HMDI [7, 15-18]. This isocyanate-terminated prepolymer was capped with 2-AP as a blocking agent until the portion of isocyanate was less than 4 % according to the Fourier transform infrared (FT-IR) spectrum at 2271 cm-1. The synthesized PUs with different molecular weights and molecular weight distributions were then prepared, using a polyTHF/HMDI molar ratio of 1:1.6 (PU16K), 1:2.0 (PU10K), and 1:2.4 (PU8K) as listed in Table 1, and were used without further purification. Note that their molecular weights were measured by gel permeation chromatography (GPC, Agilent Tech 1260) using a liquid chromatograph equipped with a series of Agilent PLgel 5m Mixed-D columns, and a 1260 Iso pump at room temperature. Tetrahydrofuran (THF) was used as an eluent at room temperature, with a flow rate 1 ml/min, and a pump pressure of 80 MPa. Chemical structures of the materials and the procedure to synthesize PUs are depicted in 4

Scheme 1. For instance, characterizations by FT-IR and NMR for the synthesized PU8K are identified in Fig. 1.

2.1.2. Preparation of the epoxy-based structural adhesives Three different PUs were incorporated to formulate the structural epoxy-based adhesives (Table 2). Epikote 828 (standard diglycidyl ether of bisphenol A with an epoxy equivalent weight of 187 g/eq, Hexion, USA) (100 phr) was used as an epoxy resin and dicyandiamide (DICY, Air-Products USA) (15 phr) was added as a curing agent. Same contents of filler mixtures containing CaCO3, CaO, and SiO2 were included into all adhesive samples for the basic performance as adhesive. Epikote 828 and synthesized single unmixed PU or mixed PUs (35 phr) were subsequently added to a planetary mixer and blended at a 20 rpm stirring speed at 70oC for 10 min. The filler mixtures were then added to the composition and mixed at 40 rpm at 90oC for 15 min under vacuum conditions. Lastly, DICY was added and mixed at the same temperature and stirring speed for 30 min in a vacuum.

2.2. Characterization methods 2.2.1. Rheological characterization Various tests for measuring the rheological properties of adhesives were performed, i.e., a thixotropic pattern test, a small amplitude oscillatory shear (SAOS) test, and a chemorheological test based on thermal curing [19,20]. An MCR-301 rheometer (Anton Paar, Austria) equipped with a CTD 450 heat chamber was employed for these tests in the 25mm parallel-plate mode with a 0.5mm gap at room temperature except during the thermal curing test. To observe the thixotropic hysteresis loop, shear viscosity data of adhesive samples were measured [9-11,14]; e.g., the shear rate was increased from 0.1 to 100 s-1 in the first-stage and then decreased from 100 to 0.1 s-1 in the second-stage for adhesives with a single PU. Thixotropic indices of the adhesives were evaluated from viscosity data within a given shear rate range. Viscoelastic properties of the adhesive samples, i.e., elastic (or storage, G’) and viscous (or loss, G’’) moduli, were compared in the SAOS test under the 0.01-100 s-1 frequency range. Finally, in the chemo-rheological thermal curing test for the initiation and further development of crosslinking of adhesives [19,20], real-time elastic and viscous moduli of the adhesive samples were recorded by increasing the measuring temperature from 70oC to 170oC and 200oC at a heating rate of 3oC/min in the SAOS mode with a frequency of 5Hz and 1% strain amplitude under linear viscoelasticity conditions. 5

2.2.2. Adhesive strength measurement To compare the adhesive properties of adhesives featuring PUs synthesized by the different polyTHF/HMDI molar ratios, lap-shear and T-peel strengths were evaluated on a cold-rolled steel substrate (SPRC 440, POSCO), according to ASTM D 1002-05 and ASTM D 1876-01, respectively. Typical specimens for the lap-shear and T-peel strength tests of the adhesives are portrayed in Fig. 2. Prior to bonding the substrate was rinsed thoroughly with toluene in order to eliminate dust and oily constituents on its surface. Specimens for two tests were produced by attaching two substrates using adhesives and keeping them at 170oC for 25 min. Note that the crosslinking network within adhesives in this study was typically formed through the polycondensation reaction between the epoxy functional groups and the amine crosslinking agents near 170oC. The adhesive strength properties were measured in a universal testing machine (Model 6982, Instron) with crosshead speeds of 1.3 mm/min (lapshear test) and 254 mm/min (T-peel test).

3. Results and discussion 3.1. Results from adhesives with single PU component The properties of the epoxy-based adhesives comprising various molecular weight PU components (PU8K, PU10K and PU16K) are compared in the following sections with emphasis on thixotropic and viscoelastic characteristics prior to cure, crosslinking characteristics during the thermal cure process and mechanical strength following cure.

3.1.1. Flowability and thixotropic rheological properties Figure 3 shows the thixotropic rheological behaviors of the single component PU adhesives A_PU8K, A_PU10K, and A_PU16K adhesives (formulations shown in Table 2a), such as hysteresis loops in shear viscosity versus shear rate diagrams, at temperatures of 25 o

C, 35 oC, and 45 oC. The general thixotropic nature of an adhesive was briefly addressed in

the Introduction. The change in shear viscosity data during the first shear-rate-increasing and the second shear-rate-decreasing stages was not significant for the A_PU8K and A_PU10K samples with the relatively low-Mw PU. However, the shear viscosity level changed dramatically for A_PU16K with the relatively high-Mw PU under the same measuring conditions. Hysteresis loops for these adhesives, showing disentanglement-entanglement processes, might be noticeably affected by the entangled chain length and polydispersity 6

index (PDI) of adhesives [9]. For instance, the pre-history effect of the A_PU8K adhesive on the applied stress in the second stage is relatively lower than those of others owing to the short entangled chain and its easy recovery, resulting in the narrower hysteresis loop. The following equation is used to define the degree of thixotropy of adhesives [9],

TI  100 

(U D) , U

(1)

where TI represents the thixotropic index, and U and D are the areas under the up-curve and down-curve, respectively, for an adhesive in the log-log plots of Fig. 3. The thixotropic indices at 25oC were 13.7 for A_PU16K, 8.5 for A_PU10K, and 8.5 for A_PU8K, implying that increasing PU molecular weight in an adhesive enhanced the thixotropic effect. Note that the difference between shear viscosities at the same shear rate decreased with increasing measuring temperature for the three adhesives (Figs. 3b and 3c), especially with the A_PU8K and A_PU10K samples having relatively lower Mw-PU in comparison with the A_PU16K. Thixotropic indices at 35 oC and 45 oC were 13.3 and 13.0 respectively for the A_PU16K adhesive, 7.0 and 6.7 respectively for the A_PU10K adhesive, and 4.7 and 4.0 respectively for the A_PU8K adhesive. Note that the thixotropic features of the adhesives at the intermediate (35oC) were more similar to those at 45oC than at 25oC, especially with the A_PU8K and A_PU10K samples. Elastic (G’) and viscous (G’’) moduli for the three adhesives measured at 25oC under SAOS mode are displayed in Fig. 4. It was found that both G’ and G’’ increased with increasing molecular weight of the PUs and G’ became higher than G’’ for A_PU16K with high-Mw PU at a high-frequency regime.

3.1.2. Crosslinking characteristics of adhesives The formation of crosslinked networks within the A_PU8K, A_PU10K, and A_PU16K adhesives during the thermal curing process was interpreted from rheological properties such as the storage modulus (G’) and tan (G’’/G’) under SAOS mode after gradually increasing the environmental temperature from 70oC to 170oC and 200oC (Fig. 5). The time for the initial growth of the storage modulus decreased with increasing PU molecular weight. Moreover, tan values during the thermal curing test decreased with increasing curing time in the early curing stage, implying that the elastic modulus level became more dominant than the viscous modulus (The viscous portion of adhesives, not presented here, was 7

considerably reduced with increasing temperature in that stage.). It is worth mentioning that there are peaks of tan for three adhesives around the initial growth of the elastic modulus linked with the extensive crosslinking of samples. Note that the level of tan for an adhesive with a low-Mw PU is generally lower than that for one with a high-Mw PU during the curing process.

3.1.3. Adhesive properties of adhesives The effect of PU molecular weight on adhesive properties such as lap-shear and T-peel strengths of epoxy-based adhesives was examined using SPRC 440 substrates, as shown in Table 3a and Fig. 6. Maximum lap-shear and T-peel strengths increased as the molecular weight of PU within an adhesive decreased (i.e., from PU16K to PU8K). Furthermore, an A_PU8K adhesive based specimen underwent substantial extension during the lap-shear test in comparison to joints comprising the other adhesive types. The peak load values (in Fig. 6a) and deformation load curves (in Figs. 6a and 6b) obtained through two tests represent the different adhesive bonding levels of three adhesives directly. The bonding strength of the A_PU8K sample was considerably higher than those of A_PU10K and A_PU16K according to both the lap-shear and T-peel tests. These results imply that adhesives including low-Mw PU can provide stronger adhesion strength. It seems that low-Mw PU can induce more hydrogen bonding than high-Mw PU case [21]. Fig. 7 shows failure surfaces following lap shear and T-peel testing of SPRC 440 bonded joints. The failure modes that can be closely related to adhesion strength are typically categorized as cohesive within adhesive and interfacial. For the A_PU8K sample, failure occurred cohesively within the adhesive, whereas the A_PU16K sample showed interfacial failure between the substrate and adhesive. The failure pattern of the A_PU10K sample was intermediate between the cohesive and interfacial failure modes. Note that failure modes on the fractured substrate by different adhesives are more clearly demonstrated by the T-peel test than a lap-shear test. It was confirmed that the mechanical adhesion between the adhesive and substrate, depending on the different molecular weights of the PUs, was directly linked with different cohesive and interfacial failure modes. Compared with the other samples, A_PU8K exhibited considerably higher bonding strength. However, the flowability and thixotropic features of this adhesive showed a narrower hysteresis loop of shear viscosity curves. Hence these features were considered to be unsatisfactory for injection and coating applications. In contrast, the A_PU16K adhesive

8

showed the opposite characteristics. Thus, the properties of adhesives should be appropriately tuned in order to attain synergetic effects on the rheological and adhesive properties by combining PU components with different molecular weights.

3.2. Comprehensive results of adhesives with hybrid PU components As one exemplary strategy to satisfy both adhesive strength and rheological flowability demands for structural adhesives, we scrutinized the effects of binary mixtures of PU8K and PU16K adhesives on the rheological and adhesive properties. The detailed formulations for adhesives with binary PU components are listed in Table 2b. Shear viscosity profiles along with shear rates for selected adhesives at 25oC are shown in Fig. 8a. As the portion of PU16K increases under a constant PU content in an adhesive, the size of the hysteresis loop of shear viscosity becomes wider; i.e., gives favorable operating conditions in the injection and coating line. Note that the shear viscosity level in the first shear-rate-increasing stage until 100 s-1 is quite similar for all adhesives considered here; however, it decreases noticeably in the second stage with increasing portion of PU16K. Figure 8b displays the crosslinking characteristics of adhesives with binary PU components during the thermal curing process. We found that the curing patterns for these samples were similar to those of A_PU16K, albeit with the different portions of PU16K in the hybrid adhesives. Figure 9 and Table 3b show deformation loads for hybrid adhesives by means of lapshear and T-peel tests. As the portion of PU16K in a mixed adhesive increased gradually, the lap-shear and T-peel strengths decreased slightly. Compared to the case of the A_PU16K sample in Fig. 6, yield loads and maximum deformation displacements against external stress from two tests were greatly enhanced with an increasing portion of PU8K in an adhesive, indicating that a hybrid combination of PUs with different molecular weights plays an important role in achieving good adhesive strength properties as well as desirable flowability. In particular, improved fracture images for the A_PU8K05 adhesive with a mixture of PU16K and PU8K in a ratio of 5:5 were clearly observed (Fig. 10), proving that physical modification of PU molecular weights can improve the integrity of the interface between adhesive and substrate.

4. Conclusions

9

Various properties of adhesives with PU toughening agents have been characterized, focusing on the role of PU molecular weight in designing desirable structural adhesives. Using three PU components with different molecular weights which were controlled by the molar ratio of polyTHF and HMDI, rheological properties of adhesives with single and binary-mixed PUs were assessed for flowability, thixotropic hysteresis behavior, and crosslinking characteristics. A thixotropic loop of shear viscosity along with shear rate became favorably larger as the PU molecular weight increased. In addition, the initial growth time for crosslinking of an adhesive effectively decreased with increasing PU molecular weight under thermal curing conditions. However, the high PU molecular weight undesirably reduced lap-shear and T-peel strength values. It was found that binary mixed PU components in an adhesive could contribute to a better trade-off between rheological flowability and adhesion strength for high-performance structural adhesives.

Acknowledgements This study was supported by the Korea Ministry of Environment (MOE) as "The Chemical Accident Prevention Technology Development Project."

References [1] Lutz A, Structural Bonding of Lightweight Cars, Dow Automotive Systems. 2015; March [2] Lutz A, Symietz D, The same structural strength in spite of thinner sheets. Adhesion Adhesives & Sealants 2009;6(1):14-8. [3] Buonocore GG, Schiavo L, Attianese I, Borriello A, Hyperbranched polymers as modifiers of epoxy adhesives. Compos. Part B: Eng. 2013;53:187-192. [4] Dhevi DM, Jaisankar SN, Pathak M, Effect of new hyperbranched polyester of varying generations on toughening of epoxy resin through interpenetrating polymer networks using urethane linkages. Eur. Polym. J. 2013;49:3561-72. [5] Liu Y, Zhang W, Zhou H, Mechanical properties of epoxy resin/hydroxyl‐terminated polyester blends: effect of two‐phase structure. Polym. Int. 2005;54:1408-15. [6] Bhuniya S, Adhikari B, Toughening of epoxy resins by hydroxy‐terminated, silicon‐ modified polyurethane oligomers. J. Appl. Polym. Sci. 2003;90:1497–1506.

10

[7] Jang JY, Jhon YK, Cheong IW, Kim JH, Effect of process variables on molecular weight and mechanical properties of water-based polyurethane dispersion. Colloid. Surface. A. 2002;196:135–43. [8] Harada M, Ohya T, Iida K, Hayashi H, Hirano K, Fukuda H, Increased impact strength of biodegradable poly (lactic acid)/poly (butylene succinate) blend composites by using isocyanate as a reactive processing agent. J. Appl. Polym. Sci. 2007;106:1813-20. [9] Benchabane A, Bekkour K, Rheological properties of carboxymethyl cellulose (CMC) solutions. Colloid. Polym. Sci. 2008;286:1173-80. [10] Barnes HA, Thixotropy - a review. J. Non-Newton. Fluid Mech. 1997;70:1-33. [11] Hicks CR, Carlson BE, Mallick PK, Rheological study of automotive adhesives: Influence of storage time, temperature and shear rate on viscosity at dispensing. Int. J. Adhes. Adhes. 2015;63:108-16. [12] Banea MD, da Silva LFM, Carvas RJC, Debonding on command of adhesive joints for the automotive industry. Int. J. Adhes. Adhes. 2015;59:14-20. [13] Baron A, Rodriguez-Hernandez J, Ibarboure E, Derail C, Papon E, Adhesives based on polyurethane graft multiblock copolymers: Tack, rheology and first morphological analyses. Int. J. Adhes. Adhes. 2009;29:1-8. [14] Jauregui-Beloqui B, Fernandez-Garcia JC, Orgiles-Barcelo AC, Mahiques-Bujanda MM, Martin-Martinez JM, Rheological properties of thermoplastic polyurethane adhesive solutions containing fumed silicas of different surface areas. Int. J. Adhes. Adhes. 1999;19:321-8. [15] Lutz A, Schneider D, Patent WO 2008157571 2008, A2. [16] Heintz A. M., Duffy D. J., Suen W., Chu W., Paul C. W., Shaw L. Hsu, Effects of Reaction Temperature on the Formation of Polyurethane Prepolymer Structures. Macromolecules 2003; 36: 2695-2704. [17] Tian C., Zhou Q., Cao L., Su Z., Chen X., Effect of polyurethane molecular weight on the properties of polyurethane–poly(butyl methacrylate) hybrid latex prepared by miniemulsion polymerization, J. Appl. Polym. Sci. 2012; 124: 5229-5235. [18] Hepburn C., Polyurethane Elastomers (2nd Ed.), Elsevier, Essex; 1992. [19] Hwang JW, Kim KN, Lee GS, Nam JH, Noh SM, Jung HW, Rheology and curing characteristics of dual-curable automotive clearcoats using thermal radical initiator derived from O-imino-isourea and photo-initiator. Prog. Org. Coat. 2013;76;11;16661673.

11

[20] Hwang JW, Kim KN, Noh SM, Jung HW, The effect of thermal radical initiator derived from O-imino-isourea on thermal curing characteristics and properties of automotive clearcoats. J. Coat. Technol. Res. 2015;12;1;177-186. [21] Rahman MM, Hasneen A, Kim HD, Lee WK, Preparation and properties of polydimethylsiloxane

(PDMS)/polytetramethyleneadipate

glycol

(PTAd)‐based

waterborne polyurethane adhesives: Effect of PDMS molecular weight and content. J. Appl. Polym. Sci. 2012;125: 88–96.

12

Scheme 1

PolyTHF (Mw 2,000)

HMDI

Scheme 1. Synthetic scheme of polyurethane pre-polymer.

13

Fig. 1

Figure 1. (a) FT-IR and (b) NMR characterizations of synthesized PU8K.

14

Fig. 2

Figure 2. Specimens for adhesion tests: (a) Lap-shear test mode and (b) T-peel test mode.

15

Fig. 3

Figure 3. Thixotropic curves for adhesives with single PU component (PU8K, PU10K, and PU16K) at (a) 25 oC, (b) 35 oC, and (b) 45 oC.

16

Fig. 4

Figure 4. Elastic and viscous moduli of adhesives with single PU component (PU8K, PU10K, and PU16K) at 25 oC.

17

Fig. 5

Figure 5. Evolution of storage moduli and tan of adhesives during thermal curing process from 70oC to (a) 170oC and (b) 200oC.

18

Fig. 6

Figure 6. Comparison of deformation loads or strengths of adhesives (A_PU8K, A_PU10K, and A_PU16K) through (a) lap-shear and (b) T-peel tests on SPRC440 substrates.

19

Fig. 7.

Figure 7. Failure surfaces of A_PU8K, A_PU10K, and A_PU16K adhesives on SPRC 440 substrate after lap-shear and T-peel tests.

20

Fig. 8

Figure 8. (a) Thixotropic curves for adhesives with binary PU components at 25 oC and (b) storage moduli and tan of adhesives during thermal curing process from 50oC to 200oC with 5oC/min heating rate.

21

Fig. 9

Figure 9. Comparison of deformation loads or strengths of hybrid adhesives (A_PU8K09, A_PU8K07, and A_PU8K05) through (a) lap-shear and (b) T-peel tests on SPRC440 substrates.

22

Fig. 10

Figure 10. Failure surfaces of A_PU8K, A_PU8K09, A_PU8K07, and A_PU8K05 adhesives on SPRC 440 substrate after lap-shear and T-peel tests.

23

Table 1. Molecular weights of polyurethanes produced by different molar ratios between PolyTHF and HMDI (PolyTHF : HMDI).

24

PU8K (1 : 2.4)

PU10K (1 : 2.0)

PU16K (1 : 1.6)

Mn

8,700

10,700

16,900

Mw

17,200

21,900

38,400

PDI

1.97

2.04

2.27

Table 2. Formulation of structural adhesives with (a) single PU (PU8K, PU10K, PU16K) and (b) binary PU components (PU8K and PU16K).

(a) A_PU8K

A_PU10K

A_PU16K

Epikote 828 (phr)

100

100

100

PU (phr)

35

35

35

DICY (phr)

15

15

15

Filler mixtures (phr)

30

30

30

(b)

25

A_PU8K09

A_PU8K07

A_PU8K05

Epikote 828 (phr)

100

100

100

PU16K/PU8K (phr) (Weight ratio)

3.5 / 31.5 (1 : 9)

10.5 / 24.5 (3 : 7)

17.5 / 17.5 (5 : 5)

DICY (phr)

15

15

15

Filler mixtures (phr)

30

30

30

Table 3. Adhesion strengths of adhesives (a) with single PU component (PU8K, PU10K, and PU16K) and (b) with binary PU components of PU8K and PU16K via lap-shear and Tpeel tests on SPRC440 substrate.

(a) Adhesive

Lap-shear strength (MPa)

T-peel strength (N/25mm)

A_PU8K

32.6

361.5

A_PU10K

28.1

296.0

A_PU16K

24.9

229.1

(b)

26

Adhesive

Lap-shear strength (MPa)

T-peel strength (N/25mm)

A_PU8K09

31.8

351.9

A_PU8K07

30.6

337.1

A_PU8K05

29.4

308.3