Highly Tunable Thiol–Ene Networks via Dual Thiol Addition

Article pubs.acs.org/Macromolecules

Highly Tunable Thiol−Ene Networks via Dual Thiol Addition Olivia. D. McNair, Bradley J. Sparks, Andrew P. Janisse, Davis P. Brent, Derek L. Patton, and Daniel A. Savin* School of Polymers and High Performance Materials, University of Southern Mississippi, 118 College Drive #5050, Hattiesburg, Mississippi 39406, United States ABSTRACT: Throughout the past decade, investigations of thick thermoset thiol−ene networks (TENs) have become increasingly prominent in the literature due to facile, quantitative synthesis giving rise to unique network characteristics, specifically high mechanical energy damping. This article reports the synthesis and thermomechanical properties of ternary thiol−thiol−ene systems that exhibit tunable glass transitions that maintain high, narrow tan δ values in the glass transition region. We begin with a base network of a trifunctional thiol and a trifunctional ene and then systematically substitute the trifunctional thiol with a series of difunctional thiols while maintaining stoichiometric balance between total thiol and ene content. The resultant ternary networks exhibit glass transition temperatures that follow the Fox equation. In contrast to other ternary thiol−ene networks, we observe minimal broadening of the glass transition region, which implies that we can retain the energy-absorbing capabilities of the thiol−ene system. This approach has high potential as a simple tool for scientists and researchers to tune Tgs for select networks without detrimentally affecting other physical properties.

■

δ versus temperature curve. If the testing temperature is only a few degrees outside this range, impact absorption decreased significantly and material failure sometimes occurred. Although there is a delicate balance between high impact energy absorption and narrow Tg, the use of TENs has been established commercially throughout the coatings industries. The key factor contributing to its commercial success is the economical production through a rapid, low-energy polymerization process. To expand this material base into new applications, researchers are investigating thick thermoset thiol−ene networks. The first generation of materials target a number of noncoatings applications such as in biomedical devices, composites, and dental restoratives.10−16 In addition to the aforementioned uses, the role of TENs in personal and sport protective equipment is rapidly gaining attention because of proven performance in energy damping applications such as pendulum tests and even bulletproof testing.17−21 In previous studies, Tg and other measured thermal and thermomechanical properties were investigated with respect to monomer structure of both thiol and ene components as well as cross-link density. As expected, higher monomer functionality leads to higher Tg networks and vice versa.22 As Tg is an important factor for protective equipment due to its correlation with damping behavior, it is critical to gain a synthetic handle to control the Tg of the network while maintaining the narrow transition region without sacrificing the integrity of the network. A common approach to manipulate thiol−ene network properties is by incorporating a third component such as an

INTRODUCTION Thiol−ene networks (TENs) are developed from facile photopolymerization synthesis and demonstrate excellent energy damping characteristics at the glass transition temperature due to their uniquely uniform cross-link densities. Synthesized via a “click” reaction between thiol and ene functionalities, thiols readily react with pendent vinyl groups or enes via a radical-stepgrowth reaction mechanism.1−3 UV-generated radicals abstract labile thiol protons, generating thiyl radicals which sequentially add across a carbon−carbon double bond producing carboncentered radicals. In a step-growth step, the carbon-centered radical abstracts a proton from a second thiol, resulting in antiMarkovnikov addition thioether linkage. The nature of this mechanism is the foundation for the very unique mechanical properties of TENs.4,5 For example, in contrast to other radical polymerization processes, significant molecular weight growth and gelation are delayed until the end of the polymerization process, yielding low stress and highly uniform networks that exhibit a narrow glass transition region.6,7 The glass transition temperature (Tg) is a critical polymer property, the nature of which often determines which applications a polymer is suitable for. For example, Gould and co-workers found a direct relationship between mechanical energy damping of TENs and their respective glass transition temperatures (determined by tan δ versus temperature curves from dynamic mechanical analysis.)8,9 Maximum energy absorption occurred when the temperature of testing was in the vicinity to the Tg. In the study, when the Tgs of the TENs were within a few degrees of the testing temperature, up to 90% of input mechanical energy was absorbed by the polymer, with the maximum absorption taking place at the maximum of the tan © 2013 American Chemical Society

Received: April 10, 2013 Revised: June 19, 2013 Published: July 2, 2013 5614

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

acrylate or a urethane. Three-component, or ternary, TENs have been synthesized in order to enhance polymer versatility while maintaining the desirable properties of thiol−ene components. In general, such three-component systems demonstrated hybrid network properties.23−30 For example, networks containing acrylate monomers had broader Tgs and were brittle, while those with added isocyanates were tougher. Acrylates are commonly polymerized via UV-initiated polymerization marked by an inhibition period, early onset of gelation, and high shrinkage. The copolymerization of acrylates with alkenes yields networks with increased homogeneity, reduced oxygen inhibition, reduced shrinkage, and lower stress. One drawback of this approach is incomplete network formation which occurs due to additional cross-polymerization of enes (intended to react with thiols) and acrylate monomers as well as homopolymerization of acrylates. In other words, acrylate monomers react with thiol monomers and with themselves, consequently leading to incomplete network formation and phase-separated networks.23 Nonetheless, it has been shown that there is a clear association between acrylate content and Tg, with Tg increasing as acrylate content increases. Unfortunately, the breadth of the glass transition region and maximum tan δ values were greatly compromised.23,24 Utilizing urethanes as a third component has been achieved both by (1) synthesis of novel urethane-based monomers and (2) one pot dual and simultaneous curing reactions.31,32 The former method is somewhat challenging due to mixing issues with high viscosity urethane-based monomers, while the latter is still under investigation. Thiourethane or urethane linkages are desirable for enhancing mechanical properties of TENs. Tg also varies greatly with respect to isocyanate content; however, these methods complicate what could be a simple UV polymerization. Herein we report the synthesis and mechanical properties of ternary thiol−thiol−ene networks. We begin with a base network of a trifunctional thiol and a trifunctional ene and then systematically replace the trifunctional thiol with a series of difunctional thiols while maintaining stoichiometric balance between total thiol and ene content. Such ternary systems, to the best of our knowledge, have not been studied. In this paper we will explore network properties of TENs composed of two thiol monomers reacting simultaneously with a vinyl monomer via traditional UV photopolymerization. Specifically, we will study how the structure of the second thiol component will alter the glass transition temperature, the width of the transition region, and maximum tan δ values. In addition, we present other thermomechanical properties such storage modulus, loss modulus, and cross-link density.

■

Figure 1. Molecular structures of (a) glycol di(3-mercaptopropionate), (b) trimethylolpropane tris(3-mercaptopropionate), (c) propylene glycol bis(3-mercaptopropionate) (MW ≈ 800 g/mol), (d) propylene glycol bis(3-mercaptopropionate) (MW ≈ 2200 g/mol), (e) 1,3,5triallyl-1,3,5-triazine-2,4,6(1H,3H,5H)-trione, and (f) 2,2-dimethoxy-2phenylacetophenone. total thiol (−SH) and ene (−CC) was 1:1 for all mixtures. The mixtures were poured into silicon molds and passed under a UV Fusion line EPIQ 6000 fitted with a D bulb, an energy output of 453 mJ/cm2, and belt speed of 19 ft/min a total of 10 times. Following UV irradiation, samples were thermally postcured at 90 °C for 24 h to ensure fully cured networks. Kinetic Analysis. Representative kinetic analyses were conducted using real-time FTIR (RT-FTIR) spectroscopy for binary formulations of thiol/ene (100:100 mol %) and ternary formulations of thiol/thiol/ ene (50:50:100 mol % of functional groups, stiochiometrically balanced with respect to total thiol and ene functional groups). Studies were carried out using a Nicolet 8700 FTIR spectrometer having a KBr beam splitter and MCT/A detector. An OmicCure Exfo 1000 Series external light source provided filtered UV light in the range of 320−500 by way of a guided optical cable. Spectra were collected at an approximate rate of 1 s−1 over a range of 650−4000 wavenumbers (cm−1). Following a 10 s rest period, the sample was irradiated with a UV light with intensity of 25 mW/cm2 while conversion of both thiol and ene functional groups was monitored. The peak corresponding to the S−H stretch was found to shift slightly with each different formulation (TMPT−TATAT, 2572 cm−1; GDMP−TATAT, 2573 cm−1; PPGMP800−TATAT, 2569 cm−1; TMPT−PPGMP2200−TATAT, 2575 cm−1). The CC stretch from the ene component remained nearly unchanged at 3084 cm−1. The total time of experiments was 320 s, and all experiments were carried out under N2 at ambient temperature. Thermal and Mechanical Testing. Thermomechanical properties were measured via dynamic mechanical analysis (DMA) using a TA Instruments Q800 DMA. Rectangular specimens with dimensions (ca. 6 ± 0.5 mm × 5 ± 0.05 mm × 0.5 ± 0.05 mm) (length × width × thickness) were deformed in tension mode at a frequency of 1 Hz in air. The thermal history consisted of cooling to −40 °C followed by a 2 min isothermal step and then heating to 100 °C at 5 °C/min while a strain of 0.05% was applied. Storage modulus, loss modulus, and tan δ were measured concurrently as a function of increasing temperature. Thermal properties were determined via differential scanning calorimetry (DSC) using a TA Instruments Q2000 DSC. Thermal transitions were measured using samples (6 ± 1 mg) placed into standard aluminum pans against a blank aluminum pan under nitrogen gas with a 50 mL/min flow rate. The samples were initially heated at 10 °C/min to 100 °C for 5 min to erase thermal history. The samples were then cooled to −70 °C at 10 °C/min, held, and heated at the same rate to a final

EXPERIMENTAL SECTION

Materials. Glycol di(3-mercaptopropionate) (GDMP) and propylene glycol bis(3-mercaptopropionate) of two molecular weights, approximately 800 and 2200 g/mol (PPGMP800 and PPGMP2200, respectively) were graciously donated by Bruno Bock Thio-ChemicalsS. The photoinitiator 2,2-dimethoxy-2-phenylacetophenone (DMPA) was purchased from Sigma-Aldrich. Trifunctional vinyl and thiol compounds, 1,3,5-triallyl-1,3,5-triazine-2,4,6(1H,3H,5H)-trione (TATAT) and trimethylolpropane tris(3-mercaptopropionate) (TMPT), were also purchased from Sigma-Aldrich. All thiols, enes, photoinitiator, and catalysts were used without further purification. Structures for all materials used in this study are shown in Figure 1. Sample Preparation. Dual thiol networks were synthesized by dissolving 0.5 wt % DMPA (compared with total weight) in thiol or thiol mixtures, followed by addition of TATAT (the ene component) and mechanical mixing. Caution was taken to protect the mixtures from any sources of light during mixing. The stoichiometry of functional groups for 5615

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

Table 1. Summary of Thermal and Thermomechanical Properties of Dual Thiol Networks formulation 3T/GDMP/TATAT 100−0−100 80−20−100 60−40−100 40−60−100 20−80−100 0−100−100 3T/PPGMP800/TATAT 100−0−100 80−20−100 60−40−100 40−60−100 20−80−100 0−100−100 3T/PPGMP2200/TATAT 100−0−100 80−20−100 60−40−100 40−60−100

Tg,tan δ (°C) 59 49 41 32 31 23 59 23 3 −9 −16 −23 59 (−47), 58 (−31), 55 (−32)

fwhm (°C)

E′rubbera (MPa)

35 28 23 15 10 5

12 14 13 14 13 17

17.4 13.1 13.6 13.1 9.1 6.0

35 −2 −26 −35 −39 −41

12 21 24 22 19 17

35

12 26 22 19

Tg,DSC (°C)

E″max (°C)

Tdeg (°C)

47 41 31 25 21 13

366 369 364 366 365 362

17.4 11.9 6.9 4.6 3.44 2.2

47 5 −13 −23 −34 −36

366 358 351 348 347 347

17.1 2.5 1.1 0.52

47 −40 −41 −41

366 358 349 343

a

Rubbery storage modulus for GDMP and PPGMP800 systems measured at Tg + 40 K. Rubbery storage modulus for PPGMP2200 measured at 90 °C.

Figure 2. RT-FTIR of thiol and ene conversion for unmodified, stoichimetrically balanced, 100:100 mol %, formulations consisting of (a) TMPT− TATAT, (b) GDMP−TATAT, (c) PPGMP800−TATAT, and (d) PPGMP2200−TATAT in N2 and 0.5 wt % DMPA (light intensity = 25 mW/cm2). (Note: thiol absorbance was below detectable limits for the PPGMP2200−TATAT formulation; therefore, only ene (−CC) conversion of TATAT monomer is shown for that system.) temperature of 140 °C. The glass transition temperature was measured from the second heating cycle. Thermogravimetric analysis (TGA) was performed with a TA Instruments Q500 TGA. Samples with mass 10 ± 1 mg were placed in a platinum pan, equilibrated at 40 °C, and then heated at a rate of 10 °C/min

to 800 °C. The polymer degradation temperature was recorded as the peak value in the first derivative of weight loss vs temperature curve. All samples were tested under a nitrogen purge with a flow rate of 40 mL/min. All instruments for thermal and thermomechanical analysis were calibrated prior to conducting tests on samples. 5616

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

Figure 3. RT-FTIR of thiol and ene conversion for networks from stoichimetrically balanced mixtures of thiols and ene formulation (50:50:100 mol % thiol:thiol:ene by functional groups) consisting of (a) TMPT−GDMP−TATAT, (b) TMPT−PPGMP800−TATAT, and (c) TMPT−PPGMP2200− TATAT in N2 and 0.5 wt % DMPA (light intensity = 25 mW/cm2).

■

RESULTS AND DISCUSSION The ultimate goal of this research was to develop TENs with tunable Tgs, by way of adding a third thiol component, without compromising desirable TEN characteristics such as rapid kinetics of formation and high, narrow tan δ values in the glass transition region. We begin with a base network of a trifunctional thiol (TMPT) and a trifunctional ene (TATAT) and then systematically replace the trifunctional thiol with different difunctional thiols (GDMP, PPGMP800, PPGMP2200) while maintaining a stoichiometric balance between total thiol and ene content. We present and discuss kinetic analysis and thermomechanical analysis below. The thermomechanical data are compiled in Table 1. Comparison of Kinetics. Real-time FTIR (RT-FTIR) was used to monitor the kinetics and conversion of thiol and ene moieties in the formation of TENs. Figure 2 shows the conversion as a function of time for binary systems of TATAT and the four thiols used in these studies. In general, the conversion vs time comparing thiol and ene groups was consistent. In all cases there was negligible conversion until the sample was illuminated with UV radiation, upon which all samples reached maximum conversion after ca. 120 s. It should be noted the thiol peak for PPGMP2200/TATAT formulation had very low intensity due to the low −SH concentration; therefore, kinetic analysis was performed based solely on the ene conversion. Overall, conversion for the neat binary networks based on TMPT, GDMP, PPGMP800, or PPGMP2200 (reacting 1:1 stoichiometrically with TATAT)

were 86%, 98%, 100%, and 98%, respectively. For most systems, near-quantitative conversions were achieved with the exception of the TMPT−TATAT native matrix. Both TMPT and TATAT are trifunctional, forming higher cross-link density networks. It is not uncommon for networks from higher functional group containing monomers to fail to reach complete network cure due to proximity of reacting groups at high conversions.22 For ternary dual-thiol TENs, thiol content was manipulated by gradually decreasing TMPT content and increasing a second, difunctional thiol content, while maintaining stoichiometrically balanced systems. Kinetic profiles of ternary TENs are shown in Figure 3, where the observation of a single −SH peak represents the sum of the two distinct thiols. For each system, half of total thiol content was from TMPT, while the other half was GDMP, PPGMP800, or PPGMP2200. An equimolar number of ene functional groups from TATAT was added to make stoichiometrically balanced systems. It was observed that the rate of the thiol−ene reaction was still remarkably fast; however, all of the maximum conversions drop slightly for the ternary TENs. The largest change in conversion (98−91%) occurred for the GDMP/TMPT/TATAT system, and the other two ternary TENs only showed small changes in conversion. Presumably, the lower conversion observed in the TMPT/TATAT neat binary TEN above means that the incorporation of TMPT, and the corresponding increase in crosslink density, result in a lower overall conversion in ternary networks. Nonetheless, the formation of ternary TENs yields high conversion with rapid kinetics upon illumination. 5617

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

Highly Predictable Glass Transition Temperature of Ternary TENs. One of the key advantages of TENs is their high mechanical energy damping (characterized by a high mechanical loss, i.e., tan δ) over a narrow temperature range (characterized in terms of the width of the glass transition region) due to high network uniformity.6 The challenge has been to design formulations whereby the glass transition temperature can be tuned while retaining the high mechanical loss. Figure 4 presents tan δ results from DMA for stoichiometrically balanced ternary systems containing (a) GDMP/TMPT/

the Tg. As expected, results similar to DMA were found from DSC scans. Figure 5 presents DSC thermograms for stoichio-

Figure 5. DSC thermograms for ternary TENs containing stoichiometric mixtures of thiols (TMPT+ additive) where additives were (A) GDMP and (B) PPGMP800. The mol % of thiol additive was varied as such: (a) 0, (b) 20, (c) 40, (d) 60, (e) 80, and (f) 100 mol %. The glass transition temperature was measured from the second heating scan at 10 °C/min.

metrically balanced ternary systems containing (a) GDMP/ TMPT/TATAT and (b) PPGMP800/TMPT/TATAT as a function of temperature at varying thiol composition. The native TMPT/TATAT binary matrix exhibits a Tg of 35 °C, and the incorporation of the second thiol (GDMP or PPGMP800) decreases the Tg to 5 or −41 °C, respectively. In ternary thiol/thiol/ene networks, the highest Tgs are observed in the networks with the highest cross-link density (TMPT/TATAT), as expected. Replacing the trifunctional thiol TMPT with a difunctional thiol (GDMP or PPGMP800) reduces the Tg in concert with the average cross-link density. To a first approximation, using thiol mixtures in the TENs is effectively like incorporating a statistical copolymer into the network. The relationship between the glass transition temperature and the composition of a second additive has been investigated for many systems, including miscible polymer blends and statistical copolymers, and has been found to follow the Fox equation33,34 w w 1 = 1 + 2 Tg Tg,1 Tg,2

Figure 4. Tan δ vs temperature curves for stoichiometrically balanced ternary thiol ene networks consisting of a base thiol (TMPT) and a second thiol additive A (GDMP) and B (PPGMP800). Symbols represent mol % of each thiol additive as follows: ■, 0%; ●, 20 mol %; ▲, 40 mol %; ▼, 60 mol %; ◆, 80 mol %; ◀, 100 mol %.

TATAT and (b) PPGMP800/TMPT/TATAT as a function of temperature with varying thiol composition. The composition in this case corresponds to the mol % of thiol groups coming from GDMP or PPGMP800 in the ternary formulation. The concentration of thiol groups from TMPT is then adjusted to maintain stoichiometric balance with TATAT ene groups. In both systems, the tan δmax shifts to decreasing temperatures with decreasing TMPT content. For example, the binary network containing TMPT/TATAT has a tan δmax at 59 °C. As the TMPT component is replaced with GDMP, the tan δmax steadily decreases to 23 °C, observed for the binary system of GDMP/ TATAT. Similarly, replacing the TMPT with PPGMP800 steadily decreases the tan δmax to −23 °C, consistent with the PPGMP800/TATAT binary system. TENs containing PPGMP2200 exhibited phase separation, as we discuss later. The same formulations tested by DMA were also investigated by DSC. After a brief annealing of the systems to erase thermal history, a second temperature ramp was conducted to investigate

where Tg, Tg,1, and Tg,2 correspond to the blend, component 1, and component 2, respectively, and w1 and w2 are the weight fractions of each component. In this treatment, we are assuming 5618

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

ternary systems will remain narrow (even with similar reactivity), the transition regions for the systems studied here are considerably narrower than those shown in thiol/ene/acrylate ternary systems.35 It has been shown in pendulum impact tests that the mechanical energy absorbed is a function of temperature and is directly related to the tan δ.8,9 In the development of energy-absorbing materials, it is most beneficial to have a material that exhibits a high, narrow tan δ centered at the use temperature. Ternary thiol/thiol/ene systems here demonstrate tunability in terms of the location of the tan δmax with little sacrifice in the intensity of loss or the width of the glass transition region. Other Thermomechanical Properties and Thermal Stability. As expected, the rubbery storage modulus was directly related to the cross-link density of the networks, such that networks with higher cross-link density had higher moduli. For example, in all the systems studied, the cross-link density decreased with increasing mol percentage of the dual thiol additive, with a corresponding decrease in rubbery modulus. The decrease in rubber modulus was more pronounced with the higher molecular weight dual thiol additives PPGMP800 and PPGMP2200. Polymer degradation temperature directly relates to thermal stability of the networks. Thermogravimetric analysis provides insight into the effects of monomer structure and polymer composition on the ultimate material decomposition. Figure 7

that the thiol/thiol/ene mixture produces a statistically random network. As mentioned above, the Fox equation generally applies to polymer blends and statistical copolymers which we clearly do not have here in either case; however, neglecting the cross-links, the resulting systems are very similar to a statistical copolymer system. Similar monomer reactivity and fast kinetics are two reasons we hypothesize the system can be treated as a statistical copolymer in terms of modeling Tg vs network composition. The fit of the data to the Fox equation for ternary GDMP and PPGMP800 systems is shown in Figure 6. We find excellent

Figure 7. Polymer degradation temperature with respect to mol % of GDMP (■), PPGMP800 (●), and PPGMP2200 (▲). The degradation temperature is measured from peak of the derivative curve of weight loss % vs temperature. Samples were heated 10 °C/min to 800 °C under a N2 purge. The lines are not fits; rather, they are meant to guide the eye.

Figure 6. Glass transition temperatures with respect to increasing wt % of either GDMP or PPGMP800. Graph A represents experimental data fit to a line while graph B represents the same data fit to the Fox equation. Points represent experimental data, and lines represent models.

shows how thermal stability decreases with increasing mol % of dual thiol additives GDMP, PPGMP800, and PPGMP 2200. While GDMP decreases in a somewhat linear fashion, PPGMP800 and PPGMP2200-containing ternary systems appear to decrease in thermal stability more rapidly. This is likely due to the decreased cross-link density of these networks in conjunction with the high volume fraction of the higher molecular weight thiol when compared to GDMP. The native TMPT−TATAT network has a degradation temperature of 366°C, and other neat matrices, GDMP−TATAT, PPGMP800−TATAT, and PPGMP2200− TATAT, show much lower Tdeg values of 362, 347, and 343 °C. To ensure residual monomer or depolymerization was not the cause for the extreme differences, thermal stability of pure, unreacted, monomers was measured. The degradation temperatures of GDMP, PPGMP800, PPGMP2200, and TATAT were 216, 333, 341, and 195 °C, respectively, all lower than any network. This suggests the cross-link density and monomer composition greatly

agreement between the experimental Tg and that predicted by the Fox equation. As such, this provides a powerful mechanism to tune Tg simply by varying composition of the ternary network. As TEN formation is a step-growth process, high molecular weights are not obtained until late in the reaction. As such, TENs typically form highly uniform networks, in terms of cross-link density, that possess narrow glass transition regions. This can be quantified by the full width at half-maximum (fwhm) for the tan δ in the transition region. For both ternary systems discussed above, there is a slight broadening of the transition region at intermediate network compositions. For example, the fwhm for the native matrix of PPGMP800 and TATAT is 12 °C; this increases to 24 °C at 40/60 mol % PPGMP800/TMPT. This broadening was less pronounced for TMPT−GDMP−TATAT TENs, only increasing the fwhm by 5 °C (see Table 1). While it is not expected that the distribution of cross-link densities in the 5619

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

reactive thiol component. The Tgs were found to be highly tunable in a predictable manner based on the weight fraction of the second thiol component. These results are promising for the aforementioned goal of tuning the Tg with simple synthetic modifications. Kinetic profiles are relatively unaffected by addition of this third component for most systems, which increases promise of utilizing simple modifications to the formulation of thick thermoset TENs. The resultant ternary networks exhibit glass transition temperatures that follow the Fox equation. In contrast to other ternary thiol−ene networks, we see minimal broadening of the glass transition region, which implies that we can retain the energy absorbing capabilities of the thiol−ene system. This approach has high potential as simple tool for scientists and researchers in other fields to tune Tgs for select networks without detrimentally affecting other physical properties. In the future we plan to investigate other systems with higher glass transitions temperatures as they will likely have more applicability in real world applications for thick thermoset TENs.

affects thermal stability, potentially as longer chain monomers have a greater volume fraction in the networks. Evidence of Phase Separation in Ternary TENs Containing High-MW PPGMP2200. Although we intended to solely investigate thermal properties of ternary networks, we carefully selected monomers with similar functionalities to eliminate complexities from demixing interactions. Nonetheless, upon UV irradiation, phase separation was immediately evident (based on optical clarity of the TEN) for ternary TENs containing PGMP2200 at intermediate compositions (Figure 8). It is expected

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]; Ph 601-266-5395; Fax 601-266-5504 (D.A.S.). Notes

The authors declare no competing financial interest.

Figure 8. Image of stoichiometrically balanced, UV-cured, dual TENs with vary mol % of PPGMP2200: (a) 0, (b) 20, and (c) 40 mol %. PPGMP2200 was added to TMPT and cured with TATAT using 0.5 wt % DMPA photoinitiator.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge funding from the Office of Naval Research (Award N00014-07-1-1057). O.D.M. and B.J.S. were supported by the U.S. Department of Education GAANN Fellowship Program (Award #P200A090066). We thank Bruno Bock Thio-Chemicals-S for graciously donating thiol monomers.

that phase-separated systems should exhibit two distinct glass transition temperatures. DMA analysis of TMPT:PPGMP2200:TATAT networks indeed show two transitions at temperatures that roughly correspond to PPGMP2200:TATAT (ca. −30 °C) and TMPT:TATAT (ca. 57 °C) (Figure 9). While it may be possible

■

REFERENCES

(1) Dondoni, A. Angew. Chem., Int. Ed. 2008, 47 (47), 8995−8997. (2) Hoyle, C. E.; Bowman, C. N. Angew. Chem., Int. Ed. 2010, 49 (9), 1540−1573. (3) Hoyle, C. E.; Lowe, A. B.; Bowman, C. N. Chem. Soc. Rev. 2010, 39 (4), 1355−1387. (4) Cramer, N. B.; Bowman, C. N. J. Polym. Sci., Part A: Polym. Chem. 2001, 39 (19), 3311−3319. (5) Cramer, N. B.; Davies, T.; O’Brien, A. K.; Bowman, C. N. Macromolecules 2003, 36 (12), 4631−4636. (6) Hoyle, C. E.; Lee, T. Y.; Roper, T. J. Polym. Sci., Part A: Polym. Chem. 2004, 42 (21), 5301−5338. (7) Okay, O.; Reddy, S. K.; Bowman, C. N. Macromolecules 2005, 38 (10), 4501−4511. (8) Gould, T. E.; Piland, S. G.; Shin, J.; Hoyle, C. E.; Nazarenko, S. Dent. Mater. 2009, 25 (6), 771−780. (9) Gould, T. E.; Piland, S. G.; Shin, J.; McNair, O.; Hoyle, C. E.; Nazarenko, S. Dent. Mater. 2009, 25 (12), 1593−1602. (10) Yang, Z.; Wicks, D. A.; Hoyle, C. E.; Pu, H.; Yuan, J.; Wan, D.; Liu, Y. Polymer 2009, 50 (7), 1717−1722. (11) Cakmaki, E.; Mulazim, Y.; Kahraman, M. V.; Apohan, N. K. React. Funct. Polym. 2010, 71 (1), 36−41. (12) Zeng, Z.; Guan, S.; Zhang, H.; Zhang, L.; Liu, B.; Jiang, Z. e-Polym. 2010, 090, 1−9. (13) Black, M.; Rawlins, J. W. Eur. Polym. J. 2009, 45 (5), 1433−1441. (14) Gush, D. P.; Ketley, A. D. Chem. Coat. Conf., 2nd, Tech. Pap. 1978, 35−53. (15) Morgan, C. R.; Magnotta, F.; Ketley, A. D. J. Polym. Sci., Polym. Chem. Ed. 1977, 15 (3), 627−645. (16) Mathias, E.; Ketley, A. D.; Morgan, C. R.; Gush, D. P. In Solventless Radiation Curable Conductive Inks; Radiation Curing, Conference

Figure 9. Tan δ as a function of temperature for select dual thiol networks containing PPGMP2200 and TMPT with TATAT mixed 1:1, of −SH to (−CC) functional groups with 0.5 wt % DMPA in the following ratios (TMPT:PPGMP2200:TTT): (a) 50:0:50, (b) 40:10:50, (c) 30:20:50, and (d) 20:30:50.

that there would also be phase separation in TENs containing PPGMP800, this was not observed via optical clarity or phase imaging using tapping-mode AFM (images not shown.)

■

CONCLUSIONS We have investigated thermal and thermomechanical properties of novel ternary TENs whereby the third component is a second 5620

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621

Macromolecules

Article

Proceedings, Atlanta, GA, 1984; Association for Finishing Processes of SME: Atlanta, GA, 1984; pp 13.1−13.22. (17) Chen, Z.; Chisholm, B. J.; Patani, R.; Wu, J. F.; Fernando, S.; Jogodzinski, K.; Webster, D. C. J. Coat. Technol. Res. 2010, 1−11. (18) Stoye, D.; Freitag, W. Resins for Coatings: Chemistry, Properties, and Applications; Hanser Publishers: Munich, 1996. (19) Hoyle, C. E.; Gould, T.; Piland, S.; Wei, H.; Phillips, B.; Nazarenko, S.; Askim, F.; Cole, M. Radtech Rep. 2006, 12−17. (20) Kade, M. J.; Burke, D. J.; Hawker, C. J. J. Polym. Sci., Part A: Polym. Chem. 2010, 48 (4), 743−750. (21) Cramer, N. B.; Couch, C. L.; Schreck, K. M.; Boulden, J. E.; Wydra, R.; Stansbury, J. W.; Bowman, C. N. Dent. Mater. 2010, 26 (8), 799−806. (22) Li, Q.; Zhou, H.; Hoyle, C. E. Polymer 2009, 50 (10), 2237−2245. (23) Senyurt, A. F.; Wei, H.; Phillips, B.; Cole, M.; Nazarenko, S.; Hoyle, C. E.; Piland, S. G.; Gould, T. E. Macromolecules 2006, 39 (19), 6315−6317. (24) Senyurt, A. F.; Wei, H.; Hoyle, C. E.; Piland, S. G.; Gould, T. E. Macromolecules 2007, 40 (14), 4901−4909. (25) Wei, H.; Senyurt, A. F.; Jonsson, S.; Hoyle, C. E. J. Polym. Sci., Part A: Polym. Chem. 2007, 45 (5), 822−829. (26) Lee, T. Y.; Carioscia, J.; Smith, Z.; Bowman, C. N. Macromolecules 2007, 40 (5), 1473−1479. (27) Lee, T. Y.; Smith, Z.; Reddy, S. K.; Cramer, N. B.; Bowman, C. N. Macromolecules 2007, 40 (5), 1466−1472. (28) Lee, T. A. I. Y.; Cramer, N. B.; Hoyle, C. E.; Stansbury, J. W.; Bowman, C. N. J. Polym. Sci., Part A: Polym. Chem. 2009, 47 (10), 2509− 2517. (29) Senyurt, A. F.; Hoyle, C. E.; Wei, H.; Piland, S. G.; Gould, T. F. Macromolecules 2007, 40, 3174−3182. (30) Li, Q.; Zhou, H.; Wicks, D. A.; Hoyle, C. E. J. Polym. Sci., Part A: Polym. Chem. 2007, 45 (22), 5103−5111. (31) Shin, J.; Matsushima, H.; Comer, C. M.; Bowman, C. N.; Hoyle, C. E. Chem. Mater. 2010, 22 (8), 2616−2625. (32) Senyurt, A. F.; Hoyle, C. E.; Wei, H.; Piland, S. G.; Gould, T. E. Macromolecules 2007, 40 (9), 3174−3182. (33) Heimenz, P. C.; Lodge, T. P. Polymer Chemistry, 2nd ed.; CRC Press: Boca Raton, FL, 2007. (34) Lodge, T. P.; McLeish, T. C. B. Macromolecules 2000, 33 (14), 5278−5284. (35) Senyurt, A. F.; Wei, H. Y.; Hoyle, C. E.; Piland, S. G.; Gould, T. E. Macromolecules 2007, 40 (14), 4901−4909.

5621

dx.doi.org/10.1021/ma400748h | Macromolecules 2013, 46, 5614−5621