Silyl Ether as a Robust and Thermally Stable Dynamic Covalent

Here we introduce silyl ether linkage as a novel dynamic covalent motif for dynamic ... (26, 27) Additionally, a previous study showed that the rate o...
0 downloads 0 Views 1MB Size
Download PDF
Communication Cite This: J. Am. Chem. Soc. 2017, 139, 14881-14884

pubs.acs.org/JACS

Silyl Ether as a Robust and Thermally Stable Dynamic Covalent Motif for Malleable Polymer Design Yoshio Nishimura, Jaeyoon Chung, Hurik Muradyan, and Zhibin Guan* Department of Chemistry, University of California, Irvine, California 92697, United States

Downloaded via UNIV OF TOLEDO on June 29, 2018 at 16:47:58 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ingly, silicon-oxygen bond exchange, the dynamic chemistry which gives glass itself its fluidity at elevated temperatures and hence “vitrimer” its namehas not yet been employed in a vitrimer system. We were intrigued by the possibility of a silyl ether-based dynamic materials for its versatility in both material and chemical properties. The scope of material properties of Si−Obased materials is vast, ranging from polydimethylsiloxane (PDMS), among the softest of rubbers, to inorganic glass, an extremely strong and hard material. One early study about chemical stress−relaxation of PDMS elastomers was reported in the 1950s.25 Recently, self-healing materials using anionic siloxane exchange were reported.26,27 Additionally, a previous study showed that the rate of alkoxysilane hydrolysis varies over a wide window based on a neighboring amino-group effect.28 Encouraged by our previous finding that variability in molecular exchange kinetics could directly translate to variability in bulk dynamic properties,22 we envisioned the potential of silyl ether to deliver similar versatility in dynamic and mechanical properties to networks through minimal variation in molecular structure. Furthermore, the high chemical and thermal stability of silyl ether linkages adds additional appeal as a candidate for dynamic materials chemistry. Herein we report the first example of a silyl ether exchange-based vitrimer (Figure 1b).

ABSTRACT: Here we introduce silyl ether linkage as a novel dynamic covalent motif for dynamic material design. Through introduction of a neighboring amino moiety, we show that the silyl ether exchange rate can be accelerated by almost three orders of magnitude. By incorporating such silyl ether linkages into covalently cross-linked polymer networks, we demonstrate dynamic covalent network polymers displaying both malleability and reprocessability. The malleability of the networks is studied by monitoring stress relaxation at varying temperature, and their topology freezing temperatures are determined. The tunable dynamic properties coupled with the high thermal stability and reprocessability of silyl ether-based networks open doors to many potential applications for this family of materials.

T

he recent trend in materials chemistry toward bottom-up materials design based on small-molecule dynamics has allowed researchers to introduce remarkable macroscopic responses to bulk materials such as self-healing, shape-memory, and malleability. In particular, dynamic exchangeable cross-links have been incorporated into covalent networks to introduce plasticity and make them adaptive and malleable.1,2 Such networks were initially introduced in the literature as covalent adaptable networks (CANs),3 and the concept has recently been extended to organic network materials called vitrimers with glass-like flow properties.4 These materials combine mechanical robustness and creep resistance of thermosets at service conditions, while displaying the plasticity and/or reprocessability of thermoplastics upon thermal triggers. Importantly, the associative CANs retain their cross-linking density at all times due to their associative exchange mechanisms, and thus resist degradation from solvents. Several dynamic covalent motifs have thus far been incorporated into polymer networks to effect this property. The first reported associative CANs were based on photomediated radical addition−fragmentation chain transfer reactions by using allyl sulfide moieties,5,6 which was further applied to other CANs by using different radical generators with trithiocarbonates.7−9 More recently, transition-metal-catalyzed trans-esterification was successfully used in the design of several vitrimer systems.4,10−16 Other dynamic covalent motifs reported thus far for such applications include olefin cross metathesis,17,18 trans-amination of vinylogous urethanes,19 trans-carbamoylation,20 triazolium trans-alkylation,21 boronic ester exchange,22,23 and poly(alkylurea−urethane).24 Surpris© 2017 American Chemical Society

Figure 1. Design concept of tuning silyl ether exchange dynamics for vitrimer design. (a) Design of bis-alkoxysilanes cross-linkers with and without neighboring amino groups. (b) Dynamic exchange of silyl ether cross-linkers affords dynamic materials. Received: August 18, 2017 Published: October 9, 2017 14881

DOI: 10.1021/jacs.7b08826 J. Am. Chem. Soc. 2017, 139, 14881−14884

Communication

Journal of the American Chemical Society Specifically, we cross-linked a styrene-based copolymer with bis-alkoxysilanes through pendant hydroxyl groups functionalized on the styrene monomer units. Networks were prepared with two cross-linkers of variable dynamics, one with rateaccelerating amino neighboring groups, and the other without (Figure 1a). Gratifyingly, we found a dramatic difference in the temperature-dependent stress−relaxation behavior of the two materials. To our knowledge this demonstrates the first example of utilizing intramolecular catalytic control of bond exchange kinetics to influence the thermal responsive properties of vitrimers. Notably, the remarkable high thermal stability of silyl ether should be beneficial for many practical applications. Based on a previous report showing the effects of amino neighboring groups on the hydrolysis rates of alkoxysilanes,28 we first set out to investigate the differences in silyl ether exchange reactions in model alkoxysilanes 1 (with neighboring amino groups in the γ positions) and 2 (no neighboring groups) (Scheme 1). The small-molecule model reactions were

Table 1. Summary of Thermal Properties and Activation Energies of Silyl Ether Exchange activation energy (kJ mol−1) sample PS-Bis-γNH PS-Bis-C10 a

Tv (°C)a

Tg (°C)b

stress relaxation analysisc

small-molecule studyd

47

125

81

35e

117

123

174 12

65f b

Temperature at which a viscosity of 10 Pa·s is reached. Measured by DMA. cCalculated by the Arrhenius plots of the mean relaxation lifetime τ* (Figure S10). dCalculated by the Arrhenius plots of the rate constants k (Figure S5). eActivation energy of the reaction of 1. f Activation energy of the reaction of 2.

Scheme 2. Preparation of Cross-Linked Networks

Scheme 1. Small-Molecule Model Study of Silyl Ether Exchange Kinetics

performed through trans-methoxylation of CD3 OD to trimethoxysilanes 1 and 2 with an excess of CD3OD, and their rates monitored through disappearance of methoxysilane resonances on 1H NMR. Previous studies on the hydrolysis reactions on alkoxysilanes28 showed that these reactions proceed via a pentacoordinate intermediate, and can be described in terms of pseudo-first-order kinetics when an excess of water is present. By following a similar kinetic treatment, the reaction rate constants for trimethoxysilanes 1 and 2 were calculated and are summarized in Table S2 (details in Supporting Information, including Figures S3 and S4). Consistent with literature,28 compound 1 exhibited almost three orders of magnitude faster rate for CD3OD exchange than compound 2 due to internal catalysis by the neighboring amino group (Figures S3−S5) on the basis of their rate constants. The activation energies of exchange for these two species were calculated to be 35 and 65 kJ/mol, respectively (Table 1, see also Figure S5). Having demonstrated the tunable exchange dynamics in model small molecules, next we aimed to transfer this concept to bulk materials. For the polymer backbone, we synthesized a styrene-based copolymer with hydroxyl group side chains (Scheme 2). Styrene-based monomer 3 with a pendant hydroxyl group was prepared according to literature,29 which was copolymerized with styrene by free radical polymerization to afford poly(styrene-co-styrene-OH) 4. The molar ratio of monomer 3 to styrene in copolymer 4 was determined by 1H NMR to be 9 mol% (Figure S1). On the basis of our small-molecule model study, we chose to use Bis-γ-NH and Bis-C10 (compounds 5 and 6, Scheme 2) as the fast- and slow-exchanging cross-linkers, respectively. Fast exchanging cross-linker 5 is a bis-alkoxysilane cross-linker with

secondary amino groups at γ positions to each respective silicon. Slow exchanging cross-linker 6 has no functional groups between the two silicon atoms. Cross-linked network PS-Bis-γNH was prepared through the reaction between copolymer 4 and cross-linker 5 (0.75 mol% cross-linker with respect to combined monomers) in 1,2-dichlorobenzene. Each crosslinker has six reactive functionalities, resulting in a maximal consumption of 4.5% of the hydroxyl groups on the polystyrene backbone (i.e., half of the total hydroxyl groups). The remaining half of the free hydroxyls will allow associative exchange with the sily ether linkages for the desired vitrimer function. The reaction was first allowed to proceed at 80 °C for 12 h, after which the solvent was removed at 120 °C under reduced pressure (0.03 mmHg) for 24 h. The PS-Bis-C10 network was prepared through the reaction of copolymer 4 with cross-linker 6 (0.75 mol% cross-linker with respect to combined monomers) in the same condition as PS-Bis-γ-NH (Scheme 2). These cured networks were insoluble in 1,2,4trichlorobenzene at 150 °C for 24 h, indicating a robustly cross14882

DOI: 10.1021/jacs.7b08826 J. Am. Chem. Soc. 2017, 139, 14881−14884

Communication

Journal of the American Chemical Society

characterized by polymer chain motion, and follows a very rapid drop in viscosity over a small temperature range, commonly modeled by the Williams−Landel−Ferry (WLF) equation.30 In contrast, Tv arises from molecular exchange kinetics, and gradual, logarithmic decay of viscosity results over a wide temperature range exhibited by the classic Arrhenius model of molecular kinetics. This means that in a Tv ≪ Tg vitrimer, in a narrow temperature range above the Tg, rapid decay of viscosity around Tg predicted by the WLF model should dominate, while at temperatures sufficiently above Tg, the gradual Arrhenial decay of viscosity characteristic of vitrimers should dominate.2 This behavior is indeed seen in PS-Bis-γ-NH network. Observe in Figure 3 as the rapid decay of viscosity (ln η) of the

linked network (Tables S3 and S4). The combination of DMTA data (Figure S9), calculated cross-linking density (see Supporting Information, page S-9), and FT-IR spectra (Figure S12) support that PS-Bis-γ-NH and PS-Bis-C10 have comparable network structures. Next, the cross-linked polymers PS-Bis-γ-NH and PS-BisC10 were subjected to stress relaxation analyses at elevated temperatures (150, 160, 170,and 180 °C) to investigate their flow properties (Figure 2). Consistent with the small-molecule

Figure 3. Temperature dependence of viscosity for PS-Bis-γ-NH network

Figure 2. Stress relaxation tests of (a) PS-Bis-γ-NH network and (b) PS-Bis-C10 network.

material at temperatures ranging from 125 to 140 °C (close to Tg, WLF behavior) gives way to the steady, linearly logarithmic decline of viscosity above 150 °C (Tv dominating Tg with Arrhenius behavior). This type of behavior was proposed in a mini-review by Du Prez and co-workers,2 but to the best of our knowledge this is the first time that this behavior has been experimentally verified. The significance of this result cannot be understated as it provides direct support to both the theory of topology freezing through molecular kinetic arrest of vitrimers, as well as the mechanism of fluidity of the specific system described in this paper. Gratifyingly, the activation energies toward flow for the two networks demonstrate the expected trend. As shown in Table 1, the activation energy of PS-Bis-γ-NH is less than half of that for PS-Bis-C10, reflecting the same trend that was shown for small-molecule exchange kinetics for compounds 1 and 2. However, the absolute values obtained for the polymer networks are substantially higher than those obtained in small-molecule exchange studies. We attribute this difference to two factors, variability in the steric bulk between the exchanging alkoxy groups and diffusion of polymer chains. Namely, in the small-molecule model study, the exchanging alkoxy groups are very small methoxy and deuterated methoxy groups. On the contrary, in bulk polymer samples, the exchanging alkoxy groups on the networks are much more sterically encumbered (Scheme 2). In addition, diffusion of molecules is much slower in bulk polymers. This factor may also affect the difference. Finally, we performed uniaxial tensile extension tests of our samples to gauge the mechanical strength of the materials, as well as to confirm reprocessability of our networks. First, we prepared dogbone-shaped samples of PS-Bis-γ-NH and PSBis-C10, and collected stress−strain profiles of the virgin

exchange kinetics (Table S2), we found PS-Bis-γ-NH released stress much faster than PS-Bis-C10 at all temperatures. For example, the characteristic relaxation time τ* of PS-Bis-γ-NH at 180 °C is 260 s, while τ* of PS-Bis-C10 at 180 °C is 779 s. These results show good qualitative correlation with the smallmolecule kinetics in Table S2, supporting our hypothesis.22 Furthermore, as shown in Figure S10, the temperature dependence of the relaxation time for both samples is in good agreement with the Arrhenius equation, in alignment with vitrimeric properties arising from associative dynamic exchange. Thanks to the chemical stability of silyl ether linkage, the networks exhibit very high thermal stability. Thermogravimetric analysis (TGA) shows that there was negligible thermal degradation for PS-Bis-γ-NH and PS-Bis-C10 at temperatures below 300 and 350 °C, respectively (see Figure S6). The onset of thermal degradation of our current system is ∼100 °C higher than previously reported vitrimers, where degradation onsets have been reported between 200 °C (triazolium transalkylation)21 and 263 °C (trans-carbamoylation).20 Practically, this allows for a wider temperature range for real-life material processing and applications. A defining characteristic of vitrimers is the topology-freezing transition temperature (Tv), the temperature at which the network topology of the material is frozen through kinetic trapping of its internal dynamic chemistry. Following reported protocols,4 Tv of PS-Bis-γ-NH and PS-Bis-C10 are calculated to be 47 and 117 °C, respectively (Table 1; for details of calculation, see Supporting Information, page S-10). The Tv of PS-Bis-γ-NH provides a special case scenario of Tv (47 °C) ≪ Tg (125 °C) rarely seen in vitrimer literature,19 as most vitrimers exhibit Tv > Tg. This property gives rise to unique behavior in the PS-Bis-γ-NH network. The Tg is 14883

DOI: 10.1021/jacs.7b08826 J. Am. Chem. Soc. 2017, 139, 14881−14884

Communication

Journal of the American Chemical Society Notes

samples. After failure of material, each sample was cut into small pieces and then hot-pressed at 160 °C, 6 h for PS-Bis-γNH or at 190 °C, 6 h for PS-Bis-C10. Because of higher Tv of PS-Bis-C10 than that of PS-Bis-γ-NH, the higher temperature is needed to reprocess samples of PS-Bis-C10. The polymer networks were fully reformed. The tensile tests indicate that the mechanical properties of both PS-Bis-γ-NH and PS-Bis-C10 were recovered after the reprocessing conditions (Figure 4, Table S4). FT-IR (Figure S12) and isothermal TGA data (Figure S7) confirm that there was no significant degradation occurred during reprocessing the samples.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the U.S. Department of Energy, Division of Materials Sciences, under Award DE-FG0204ER46162.

■

(1) Bowman, C. N.; Kloxin, C. J. Angew. Chem., Int. Ed. 2012, 51, 4272−4274. (2) Denissen, W.; Winne, J. M.; Du Prez, F. E. Chem. Sci. 2016, 7, 30−38. (3) Kloxin, C. J.; Scott, T. F.; Adzima, B. J.; Bowman, C. N. Macromolecules 2010, 43, 2643−2653. (4) Montarnal, D.; Capelot, M.; Tournilhac, F.; Leibler, L. Science 2011, 334, 965−968. (5) Scott, T. F.; Schneider, A. D.; Cook, W. D.; Bowman, C. N. Science 2005, 308, 1615−1617. (6) Kloxin, C. J.; Scott, T. F.; Park, H. Y.; Bowman, C. N. Adv. Mater. 2011, 23, 1977−1981. (7) Nicolaÿ, R.; Kamada, J.; Van Wassen, A.; Matyjaszewski, K. Macromolecules 2010, 43, 4355−4361. (8) Amamoto, Y.; Kamada, J.; Otsuka, H.; Takahara, A.; Matyjaszewski, K. Angew. Chem., Int. Ed. 2011, 50, 1660−1663. (9) Amamoto, Y.; Otsuka, H.; Takahara, A.; Matyjaszewski, K. Adv. Mater. 2012, 24, 3975−3980. (10) Capelot, M.; Unterlass, M. M.; Tournilhac, F.; Leibler, L. ACS Macro Lett. 2012, 1, 789−792. (11) Pei, Z.; Yang, Y.; Chen, Q.; Terentjev, E. M.; Wei, Y.; Ji, Y. Nat. Mater. 2014, 13, 36−41. (12) Yang, Y.; Pei, Z.; Zhang, X.; Tao, L.; Wei, Y.; Ji, Y. Chem. Sci. 2014, 5, 3486−3492. (13) Brutman, J. P.; Delgado, P. A.; Hillmyer, M. A. ACS Macro Lett. 2014, 3, 607−610. (14) Yu, K.; Taynton, P.; Zhang, W.; Dunn, M. L.; Qi, H. J. RSC Adv. 2014, 4, 10108−10117. (15) Pei, Z.; Yang, Y.; Chen, Q.; Wei, Y.; Ji, Y. Adv. Mater. 2016, 28, 156−160. (16) Chabert, E.; Vial, J.; Cauchois, J.-P.; Mihaluta, M.; Tournilhac, F. Soft Matter 2016, 12, 4838−4845. (17) Lu, Y.-X.; Tournilhac, F.; Leibler, L.; Guan, Z. J. Am. Chem. Soc. 2012, 134, 8424−8427. (18) Lu, Y.-X.; Guan, Z. J. Am. Chem. Soc. 2012, 134, 14226−14231. (19) Denissen, W.; Rivero, G.; Nicolay, R.; Leibler, L.; Winne, J. M.; Du Prez, F. E. Adv. Funct. Mater. 2015, 25, 2451−2457. (20) Fortman, D. J.; Brutman, J. P.; Cramer, C. J.; Hillmyer, M. A.; Dichtel, W. R. J. Am. Chem. Soc. 2015, 137, 14019−14022. (21) Obadia, M. M.; Mudraboyina, B. P.; Serghei, A.; Montarnal, D.; Drockenmuller, E. J. Am. Chem. Soc. 2015, 137, 6078−6083. (22) Cromwell, O.; Chung, J.; Guan, Z. J. Am. Chem. Soc. 2015, 137, 6492−6495. (23) Röttger, M.; Domenech, T.; van der Weegen, R.; Breuillac, A.; Nicolaÿ, R.; Leibler, L. Science 2017, 356, 62−65. (24) Zhang, L.; Rowan, S. J. Macromolecules 2017, 50, 5051−5060. (25) Osthoff, R. C.; Bueche, A. M.; Grubb, W. T. J. Am. Chem. Soc. 1954, 76, 4659−4663. (26) Schmolke, W.; Perner, N.; Seiffert, S. Macromolecules 2015, 48, 8781−8788. (27) Zheng, P.; McCarthy, T. J. J. Am. Chem. Soc. 2012, 134, 2024− 2027. (28) Berkefeld, A.; Fonseca Guerra, C.; Bertermann, R.; Troegel, D.; Daiß, J. O.; Stohrer, J.; Bickelhaupt, F. M.; Tacke, R. Organometallics 2014, 33, 2721−2737. (29) Pasini, D.; Filippini, M.; Pianetti, I.; Pregnolato, M. Adv. Synth. Catal. 2007, 349, 971−978. (30) Williams, M. L.; Landel, R. F.; Ferry, J. D. J. Am. Chem. Soc. 1955, 77, 3701−3707.

Figure 4. Tensile tests and reprocessability tests for (a) PS-Bis-γ-NH and (b) PS-Bis-C10. The top figures are tensile tests of original samples (blue solid lines) and reprocessed samples (red dash lines). The bottom images show cut and reprocessed samples.

In conclusion, we have successfully introduced a robust and thermally stable dynamic covalent chemistry, silyl ether exchange, for malleable polymer design. Notably, this is thus far the only example of using internal catalysis to control the vitrimer topology freezing temperature (Tv). Furthermore, this is the first experimental observation of a transition from WLF behavior to Arrhenius behavior for the temperature dependence of the viscosity for the networks because of a special case scenario of Tv (47 °C) ≪ Tg (125 °C), providing direct support to both the theory of topology freezing through molecular kinetic arrest of vitrimers, as well as the mechanism of fluidity of the system described in this paper. Finally, the tunable exchange dynamics coupled with the high thermal stability and reprocessability of silyl ether-based polymers should allow for broad applications for this family of materials. We are currently investigating the use of silyl ether exchange for various other dynamic material designs.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b08826. Experimental details, including sample preparation and characterization, Figures S1−S12, and Tables S1−S5 (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Zhibin Guan: 0000-0003-1370-1511 14884

DOI: 10.1021/jacs.7b08826 J. Am. Chem. Soc. 2017, 139, 14881−14884