Study of Moisture-Curable Hybrid NIPUs Based on Glycerol with Various Diamines: Emergent Advantages of PDMS Diamines.

A sol/gel curing method is used in this work to synthesize hybrid partially bio-based polyhydroxyurethanes (PHUs) from dicarbonates derived from glycerol and various diamines. The method consists of end-capping the PHU prepolymers with moisture-sensitive groups, so sealants and adhesives can be produced from partially sustainable hybrid PHUs (HPHUs), similar to their preparation from end-capped conventional polyurethanes. Diglycerol dicarbonate (DGC) is synthesized and polymerized with different diamines of various chain lengths, and the resulting structural and thermal properties of the PHUs are qualitatively and quantitively characterized. This characterization led to two potential candidates: PHU 4, made of DGC and a poly(propylene glycol) diamine, and PHU 10, prepared from DGC and a poly(dimethylsiloxane) diamine. These polymers, with respective relative number-average molecular weights of 3200 and 7400 g/mol, are end-capped and left to cure under ambient lab conditions (22 °C and 40-50% humidity), and the curing processes are monitored rheologically. Notably, moisture curing does not require any catalyst. The chemical stability of the resulting hybrid PHUs (HPHUs) 4 and 10 in pure water is investigated to check the viability of applying them under outdoor conditions. Only HPHU 10 is found to be resistant to water and shows hydrophobicity with a contact angle of 109°. Tensile tests are conducted on HPHU 10 samples cured under lab conditions for a week and others cured for another week while being immersed in water. The mechanical properties, tensile strength and elongation at break, improve with the samples cured in water, indicating the high-water repellency of HPHU 10.

Introduction

Polyurethanes (PUs) have become one of the most widely used industrial polymers since they were first discovered by Bayer et al. in 1937.[50−3] More specifically, PU sealants are known to be cheap, having long shelf- and pot-lives, resisting aging, having good antiflammability properties, and a wide range of applications in indoor and outdoor conditions as well as in porous and nonporous materials.[3] As any conventional PU, they are synthesized from the polyaddition of diisocyanates with diols, and the resulting PU prepolymers are cured using moisture-curable agents to get the final sealants.[1−3] As the −NCO end-groups in the prepolymer chain are very reactive, it is common to end-cap the oligomers with moisture-curable silane-terminated groups to produce self-cross-linkable PUs in the presence of moisture.[4−10]

However, many environmental and health issues are associated with the synthesis of conventional PUs, as the diisocyanates are produced from phosgene, which is known to be a lethal gas.[2,11−13] Besides, conventional PUs are too moisture-sensitive and hence chemically unstable.[11] Thus, new routes have been developed for nonisocyanate PU (NIPU) synthesis[14] with the step-growth polyaddition of five-membered cyclic dicarbonates with diamines being investigated the most because they do not form lower molecular weight byproducts and their precursors are abundant.[2,11−18] These NIPUs are also known as poly(hydroxyurethane)s (PHUs) because hydroxyl pendant groups are present in the resulting polymer backbone. As summarized by Maisonneuve et al., cyclic carbonates can be synthesized via different ways leading to five-membered, six-membered, and seven-membered mono-/polycyclic carbonates.[16] More specifically, the route involving the chemical fixation of carbon dioxide (CO2) into epoxy (or oxirane) groups received most researchers’ attention, seeing its importance from a green chemistry perspective.[15,19−22] With the many commercially available diamines, various cyclic dicarbonate/diamine couples were studied, and their respective PHUs showed similar mechanical properties, chemical resistance, and thermal stability as well as lower permeability compared to conventional PUs. PHUs were also investigated for the preparation of sealants by Figovsky et al., who synthesized acrylic-modified petrochemical PHUs that cure under ultraviolet radiation to give hybrid PHU (HPHU) sealants with high performance.[23,24]

While HPHUs have not been extensively studied, there exist different methods to synthesize them.[2,25] For the synthesis of HPHU sealants and adhesives, the sol/gel or moisture-curing method is considered to be the best. Hence, this work uses the sol/gel curing method, as shown in Scheme , to synthesize new partially bio-based HPHUs by end-capping the PHU prepolymers with moisture-curable agents. These agents can be either silane-terminated amines, such as N-(2-aminoethyl)-3-aminopropyltrimethoxysilane (termed DAMO), when the end-group is a cyclic carbonate or silane-terminated epoxides, such as [3-(2,3-epoxypropoxy)-propyl]-trimethoxysilane (termed GLYMO), when the end-group is an amine. Once this is done, the resulting end-capped prepolymers would be ready to self-link through a moisture-curing mechanism, leading to HPHUs, similar to end-capped conventional PU sealants and adhesives.

Scheme 1

Moisture-Curing Method Based on End-Capping the Carbonate and Diamine End-Groups of PHU Prepolymers (Proposed in This Work)

Similar studies were conducted by Gomez-Lopez et al., who recently synthesized moisture-curable PHUs from resorcinol and poly(propylene glycol) (PPG) dicarbonates, with DAMO as the moisture-curable agent. However, in order to obtain faster curing of their samples, they had to increase the temperature to 100 °C and add acetic acid as a catalyst,[26] which were the issues overcome in the present study. Moisture-curable HPHUs were achieved under ambient conditions and without the use of a catalyst. Additionally, Decostanzi et al. reacted DAMO with poly(propylene oxide) dicarbonate as well as carbonate propyltrimethoxysilane with Priamine 1075, a bio-based diamine, or aminopropyl-terminated poly(dimethylsiloxane) (PDMS) with Mn = 2500 g/mol, and they observed a lower swelling index within their PPG-based HPHUs, compared to the other materials, when immersed in THF for 24 h.[27] Sol/gel HPHUs were also investigated by Rossi de Aguiar et al. using PDMS-based cyclic dicarbonate with isophorone diamine and DAMO. The resulting hybrid polymer coatings had high thermal stability.[28]

This work, on the other hand, uses the chemistry described in Scheme to make partially bio-based HPHUs. Diglycerol is used as a precursor of cyclic dicarbonates of this work. Among the bio-renewable sources studied in the last decade,[15,23,29−32] researchers have shown interest in glycerol as it can be obtained from biomass wastes via hydrolysis or methanolysis of triglycerides and from biodiesel processes. Its worldwide annual production has increased over 200% between the years 2000 and 2010, and it was predicted that by the year 2020, its production would be 6 times more than the demand,[23,33] which makes it an abundant raw material for the preparation of new sustainable HPHU sealants and adhesives. Many studies have characterized diglycerol dicarbonate (DGC)-based PHUs prepared from diamines of different chain lengths.[34−41] As for the diamines, bio-based and petrochemical ones are available, and this work uses both with more attention given to poly(dimethylsiloxane) diamines expected to give PHUs with high value-added end properties.[28]

In this work, a different chemistry path is used to afford PDMS or PPG/DGC moisture-curable HPHU sealants and adhesives, compared to previously cited studies. DGC is first reacted with diamines of different chain lengths, including the bio-derived ones, PPG, and PDMS diamines, after which a screening of the resulting PHUs depending on their molecular weight, viscosity, texture, color, and thermal properties is conducted to select the potential DGC/diamine candidates for end-capping, as presented in Scheme . The curing kinetics, under ambient conditions and without the use of a catalyst, are studied to identify the gel time needed to form the HPHU films. Interaction of the cured polymers with water is also studied to ensure that they are chemically stable and exhibit resistance to water.

Results and Discussion

DGC Reaction and Kinetics

The evolution of the reaction between DIG and DMC was tracked to understand the mechanism underlying it and hopefully optimize the synthesis. It is considered to be a complex reaction as it is reversible and has to go through three intermediates before obtaining DGC. Scheme gives the overall reaction scheme, and Figure summarizes the evolution of the reactants’ consumption and products’ formation during 24 h.

Scheme 2

Synthesis of DGC from the DIG and DMC Transesterification Reaction

Figure 1

Kinetic study tracking the evolution of the reaction between DIG and DMC to get DGC, the renewable monomer of this study.

As expected and shown in the FTIR spectra of Figure , the intensity of the OH band decreases with the consumption of DIG, while the primary C=O stretch at 1780 cm–1 increases with the formation of DGC (see Figure S1). These changes are observed until 7 h of the reaction were reached, after which the FTIR spectra quasioverlapped. The only difference between the spectra taken at 7 and 24 h is the increase in the secondary C=O stretch at 1750 cm–1, which can be either that of the intermediates which form before obtaining DGC or that of the byproducts produced throughout the reaction as the DIG sample has originally up to 20 wt % of impurities. Also, the fact that the OH band intensity stopped decreasing after 7 h can be attributed to one of the following factors; the impurities in the original DIG batch, the reaction reaching equilibrium, or both.

To complement the kinetic study, two DGC synthesis reactions were allowed to proceed for 8 and 24 h, and both gave similar DGC yields of 51 and 47%, respectively. Additionally, the residual product from the latter, after the evaporation of DMC and methanol, was brown leading to the addition of more EtAc compared to the former, and the purification was given more time to obtain the final product. Perhaps the increase in the reaction byproducts and/or DGC intermediates, during the 24 h reaction, could have caused this darker brown color, compared to the 8 h reaction. Finally, a white DGC powder was obtained at the end of both reactions with the 1H NMR and FTIR spectra given in the Supporting Information as Figures S1 and S2.

Characterization and Screening of PHU Prepolymers

The polyaddition between DGC and diamines is presented in Scheme . Seeing that DGC has a melting point at around 65 °C,[34,35] it was possible to conduct the polyadditions at moderate temperatures and in bulk, while obtaining high-monomer conversions. Besides and as mentioned in previous studies, no catalyst was needed to synthesize the PHU prepolymers.[34,35]

Scheme 3

Polyaddition Reaction for DGC-Based PHU Synthesis Including the Diamines Investigated in This Work

The physical properties of the synthesized PHU prepolymers were analyzed qualitatively and quantitatively, and they are summarized in Table . PHUs 1 and 2 can be considered fully bio-based prepolymers, as 1,10-diaminodecane can be obtained from bio-based sources and Priamine 1074 is bio-sourced.[42] However, these diamines are low-/medium-chain length diamines, resulting in brittle and rubbery polymers at room temperature, which means that PHUs 1 and 2 were not further investigated in the end-capping process.

Table 1

Physical Properties of the Synthesized PHU Prepolymers of This Work Derived from DGC and Various Amines Based on a 1/1 Molar Equivalent Formulation

prepolymer	diamine, temperature (°C)/time (h)	Mn (g/mol)a	D̵ = Mw/Mna	Color	physical state
PHU 1	1,10-diaminododecane, 80/2	10,500	1.7	light yellow	brittle
PHU 2	Priamine 1074, 80/16	12,700	2.0	Yellow	rubbery
PHU 3	Jeffamine D-2000, 100/24	2700	2.0	clear orange	liquid
PHU 4	Jeffamine D-2000, 120/24	3200	2.8	dark red	liquid
PHU 5	Jeffamine D-2000, 140/18	3300	2.1	dark red	liquid
PHU 6	Jeffamine D-2000, 140/48	2000	1.9	dark brown	liquid
PHU 7	PDMS-2.5k-(NH2)2, 80/24	26,300	1.8	cloudy white	sticky liquid
PHU 8	PDMS-2.5k-(NH2)2, 100/24	33,100	2.0	cloudy white	sticky liquid
PHU 9	PDMS-2.5k-(NH2)2, 120/24	21,500	3.3	clear yellow	sticky liquid
PHU 10	PDMS-5k-(NH2)2, 80/16	7400	2.5	clear colorless	liquid
PHU 11	PDMS-5k-(NH2)2, 80/48	31,300	1.8	clear colorless	sticky liquid
PHU 12	PDMS-5k-(NH2)2, 100/24	35,500	2.3	clear yellow	sticky liquid
PHU 13	PDMS-5k-(NH2)2, 120/24	28,100	3.9	cloudy yellow	rubbery

Molecular weight distributions were estimated from GPC with a THF (only PHU 1 required DMF) eluent at 40 °C and the average molecular weights were reported relative to poly(styrene) standards.

In contrast, PHUs 3 through 6 were made of Jeffamine D-2000, a PPG derivative with telechelic amine functionality, which is relevant as PPG with telechelic hydroxy groups is used as the diol in the synthesis of many conventional PU sealants and adhesives. Their molecular weights can be easily tailored to a few hydroxyurethane units, and they are liquid-like at room temperature, making them good candidates for subsequent end-capping reactions. A kinetic study conducted at 140 °C to study the evolution of the molecular weight of DGC-Jeffamine D-2000-based polymers showed a decrease in average molecular weights when running the reaction beyond 18 h, as shown in Figures and S3. This is equivalent to a DP decrease from, approximately, 3 to 2. This unit cleavage can be due to the reaction temperature that was used or it can be attributed to intermolecular side reactions (transurethanization or urea formation), as observed in previous studies by Besse et al.[43] and Maisonneuve et al.[44] Seeing that 140 °C is causing this phenomenon to occur, PHUs 3 and 4 were synthesized at 100 and 120 °C for 24 h to investigate the effect of temperature on the end product. The formulation leading to PHU 4, whose 1H NMR spectrum is presented in Figure , proved to be the most desirable as its M value is in the maximum range observed in Figure , and the product was liquid-like at room temperature.

Figure 2

Molecular weight evolution of PHU 6 (DGC + Jeffamine D-2000) polymerization at 140 °C for 48 h [based on polystyrene (PS) standards].

Figure 3

1H NMR spectrum of PHU 4 (DGC + Jeffamine D-2000) with all protons labeled) in CDCl3 at 25 °C. Protons denoted by c belong to the diamine backbone (R1), whereas protons denoted by e belong to the open cyclic carbonate ring and what belonged to DGC. The rest of the peaks are predicted.

As for the PHUs made from DGC and PDMS-(NH2)2 diamines, chain scission might have occurred at 120 °C if comparing the molecular weights and dispersities (D̵) of PHU 9 with PHUs 7 and 8 as well as of PHU 13 with PHUs 11 and 12. This might be due to the activation of side reactions at 120 °C that led to lower average molecular weights, as discussed in the previous case. Optimistically, PDMS–(NH2)2 had a positive impact on the color as most of the prepolymers were white and clear. Unlike the reaction of DGC with Jeffamine D-2000, polyadditions leading to PHUs 7 to 13 have to be tailored as most of them gave high molecular weights and highly viscous materials.

The polyaddition leading to PHU 11 was the easiest to control as the reaction was the slowest at 80 °C, resulting in high molecular weights after 24 h as shown in Figure . At 16 h, the molecular weight was around 6300 g/mol, which is equivalent to two hydroxyurethane linkages as PDMS–(NH2)2 with Mn = 5000 g/mol has a relative number-average molecular weight of 2800 g/mol (DP ≈ 2). Hence, PHU 10 (1H NMR spectrum shown in Figure ) was synthesized to obtain a liquid-like and clear prepolymer with Mn = 7400 g/mol, making it a potential sealant or adhesive candidate.

Figure 4

Molecular weight evolution of PHU 11 polymerization at 80 °C for 48 h (measured by SEC relative to PS standards).

Figure 5

1H NMR spectrum of PHU 10 [DGC + PDMS-5k-(NH2)2] in CDCl3 at 25 °C. Protons denoted by c belong to the diamine backbone (R1), whereas protons denoted by e belong to the open cyclic carbonate ring and what belonged to DGC. The rest of the peaks are predicted.

Besides checking for the prepolymers’ physical properties, it is important to assess their thermal properties, which are summarized in Table . PHUs 4 and 10 have both very low Tg s, as they are both made of long-chain length diamines which provide their chains with more flexibility, and hence a lower Tg compared to PHU 2 which was synthesized using a medium-chain length diamine. The Tg of PHU 10 was not measured because of instrument limitation that did not allow cooling less than −90 °C; however, this parameter was measured by Bossion et al. for a similar system, and the value is included in Table .[37] Moreover, PHUs 4 and 10 show high thermal stability compared to other DGC-based PHUs studied in the literature.[34,35] In fact, the chemical structure of both PDMS-5k-(NH2)2 and Jeffamine D-2000 has ether bonds known to increase the thermal stability of PHUs.[45] Finally, and in addition to the properties discussed previously, seeing that both PHUs 4 and 10 exhibit viscosities allowing them to flow easily under ambient conditions, it is confirmed at this stage that these prepolymers can be considered for the next step.

Table 2

Thermal Properties of Selected PHU Prepolymers as Potential Sealant Candidates (Data of PHU 2 Are Also Included Although the Polymer Is Not Considered for End-Capping)a

prepolymer	Tg (°C)	Td,onset (°C)	Td,10% (°C)	η (Pa s)b
PHU 2	–9	200	232
PHU 4	–60	205	300	11.5
PHU 10	–120[37]	250	369	21.5

The formulations of the polymers are given in Table .

Viscosities are measured at room temperature (∼22 °C).

End-Capping of Prepolymers

To proceed with this step, it is important to characterize the end-groups of the prepolymers as it determines the functionality of the capping agent required. The FTIR spectra of both PHUs 4 and 10 are given in Figures and 7. The second C=O stretch appearing at around 1800 cm–1 proves the presence of carbonate end-groups, whereas its absence implies that the prepolymers have amine end-groups. In fact, any nonreacted DGC, whose melting point is at 65 °C, precipitated at the bottom of the reactor after bringing the reaction mixture to room temperature, so the C=O stretch appearing at 1800 cm–1 in Figure had to be that of the carbonate end-groups. As a result, PHUs 4 and 10 were found to possess carbonate and amine end-groups, respectively, as mentioned in Table . The necessary information to conduct these reactions is given in Table (more information regarding the estimation of the absolute molecular weights of the PHUs in this work is given in Section III-A of the Supporting Information), and the different trials with PHUs 4 and 10 are summarized in Tables and 5, respectively.

Figure 6

Labeled FTIR spectrum of PHU 4 (C=O stretch of carbonate end-groups at 1800 cm–1).

Figure 7

Labeled FTIR spectrum of PHU 10 (absence of C=O stretch of carbonate end-groups).

Table 3

Prepolymer Information Necessary for Preparing End-Capped PHUs (HPHUs)

prepolymer	end-group	compatible end-capper	number of hydroxyurethane linkages	estimated molecular weight (g/mol)
PHU 4	carbonate	DAMO	6	6872
PHU 10	amine	GLYMO	2	10,218

Table 4

End-Capping Experiments of PHU 4

trial	end-capper	ncapper/npolymer	temperature (°C), time (h)	curing information
PHU 4–1	DAMO	3	120, 5	curing in the reactor, resulting in brittle material
PHU 4–2	GLYMO	3	120, 5	no curing even after 1 week
PHU 4–3	DAMO	2	120, 2	curing started in the reactor and continued under moisture, resulting in a smooth film
PHU 4–4	DAMO	2	120, 1 h 15 min	curing after 24 h under moisture, resulting in a smooth film
PHU 4–5	DAMO	4	120, 1 h 15 min	curing after 24 h under moisture, resulting in a brittle film
PHU 4–6	DAMO	6	120, 1 h 15 min	curing after 24 h under moisture, resulting in a brittle film
PHU 4–7	DAMO	2	22, 14	no curing even after 1 week

Table 5

End-Capping Experiments of PHU 10

trial	end-capper	ncapper/npolymer	temperature (°C), time (h)	curing information
PHU 10–1	GLYMO	6	80, 4	curing in the reactor, resulting in a brittle material
PHU 10–2	DAMO	6	80, 4	no curing even after 1 week
PHU 10–3	GLYMO	6	80, 2	curing after 24 h under moisture, resulting in a brittle film
PHU 10–4	GLYMO	2	80, 2	curing after 24 h under moisture, resulting in a smooth film
PHU 10–5	GLYMO	3	22, 14	curing after 24 h under moisture, resulting in a brittle film
PHU 10–6	GLYMO	2	22, 14	curing after 24 h under moisture, resulting in a smooth film
PHU 10–7	GLYMO	1	22, 14	partial curing after 1 week, resulting in a softer film

As the PHU 4 prepolymer chain has carbonate end-groups, the end-capping reaction with GLYMO did not work (PHU 4–2 in Table ), and no curing occurred with moisture when exposed to air afterward. On the other hand, the end-capping with DAMO (PHU 4–1, PHU 4–3, and PHU 4–6 in Table ) led to moisture-sensitive HPHUs, which are proven by the disappearance of the C=O stretch at 1800 cm–1, as shown in Figure . The DAMO content was varied (PHU 4–3 to PHU 4–6), and a capper equivalent molar ratio of 2 seems to form the best film, as the film is smooth and not brittle. Figure presents different cases discussed previously. All the films formed using PHU 4-DAMO had a sticky surface, meaning that HPHU 4 could be possibly used as a sealant–adhesive. The curing kinetics of this polymer were rheologically followed for 24 h under ambient lab conditions (22 °C and 40–50% humidity), as shown in Figures , S6, and S7. A gel time of 7.5 h was measured, which proves the qualitative observations discussed previously.

Figure 8

Comparison of FTIR spectra [PHU 4–1 (black) and PHU 4–2 (red)] between successful and unsuccessful end-capping reactions of PHU-4 with DAMO (red) and GLYMO (black). Disappearance of the 1800 cm–1 C=O stretch observed with the DAMO end-capper compared to the GLYMO one.

Figure 9

PHU 4 end-capping reaction summary: PHU 4–1 with excess DAMO resulting in brittle films, PHU 4–2 with excess GLYMO resulting in noncuring materials, and PHU 4–4 resulting in smooth HPHU 4 films.

Figure 10

Curing kinetics of end-capped PHU 4 by following the storage (G′) and loss (G″) moduli (Pa) at a frequency of 1 Hz and a strain of 1% for 20 h. Measurements were done at 22 °C and 40–50% humidity.

Nevertheless, the end-capping reaction of PHU 10 with GLYMO was performed at lower temperatures and proved to work under ambient conditions, as shown elsewhere.[46] As the polyaddition of PHU 10 is very sensitive to temperature (shown previously in Section 2.2 and Figure ), it is convenient to have the end-capping reaction performed at room temperature to inhibit the increase of the prepolymer molecular weight at higher temperatures such as 80 °C. A curing kinetics study on the end-capped PHU 10 proved that polymeric films start forming after 3 h when exposed to moisture under ambient conditions, as shown in Figures , S8 and S9. Similar to PHU 4, increasing the end-capper stoichiometric content more than 2-fold resulted in brittle HPHU 10 films (PHU 10–1, PHU 10–3, and PHU 10–5 in Table ), whereas a lower amount led to soft films as cross-linking was incomplete (PHU 10–7 in Table ). An end-capper stoichiometric content twice that of the polymer end groups gave smooth films with a nonsticky surface, making HPHU-10 materials viable as potential sealant coatings. End-capping PHU 10 was also tried with DAMO, but no curing occurred after 1 week, proving further that the end-groups are amines and not carbonates (PHU 10–2 in Table ). Figure summarizes the different HPHU 10 materials formed under the conditions discussed earlier herein.

Figure 11

Curing kinetics of end-capped PHU 10 by following the storage (G′) and loss (G″) moduli (Pa) at a frequency of 1 Hz and a strain of 1% for 7 h. Measurements were done at 22 °C and 40–50% humidity.

Figure 12

PHU 10 end-capping reaction summary: PHU 10–1 with excess GLYMO resulting in brittle films, PHU 10–2 with excess DAMO resulting in noncuring materials, and PHU 10–4 resulting in smooth HPHU 10 films.

HPHUs–Water Interaction and Chemical Stability

Swelling experiments in purified H2O were conducted on films of both HPHUs 4 and 10 to check for possible interactions between the synthesized materials and water. The data for these experiments are summarized in Tables S3 and S4. HPHU 4 was found to swell in the presence of water with the equilibrium water content (EWC) (%) = 62.97 ± 3.54, equilibrium water absorption (EWA) (%) = 38.62 ± 1.35, and GC (%) = 66.70 ± 0.97, making these materials sensitive to water, and thus could be problematic in outdoor applications. In fact, the PPG comonomer is polar and is expected to be water-swellable.[47] Therefore, these materials are limited to indoor applications only, unless the resin is blended with other additives and reagents. Conversely, HPHU 10 did not swell in the presence of water, and the weights of the samples were the same after 1 week. Sessile drop tests on these films showed that they are hydrophobic with an average contact angle with Milli-Q water of 109° ± 2.1 (Table S5), which is expected as PDMS is known to be hydrophobic with a contact angle around 140°.[48]

Moreover, after drying pure H2O and collecting the residuals (if any), 1H NMR analysis in D2O showed materials being leached from the HPHU 4 samples when they were immersed in water, while no leaching occurred from the HPHU 10 films (Figures S10 and S11, respectively). As a result, it can be implied that HPHU 10 films are suitable for both indoor and outdoor applications, as they are resistant to water. They also exhibited resistance to the acid medium they were immersed in as HPHU 10 samples’ weights remained unchanged during that degradation study (Table S6).

Additionally, it was possible to cut dog-bone shapes from the HPHU 10 films formed previously because they were firm and easier to handle than the HPHU 4 ones. The tensile test results for the samples cured under lab conditions for 1 week only and those cured in pure H2O for another week afterward are presented in Table (refer to Tables S7 and S8 and Figures S12 and S13, where the results of the tests are summarized). These results suggest that the latter were exposed to more “moisture” than the former, making their cross-linking easier. In fact, they exhibited higher EB % and σmax, which prove that these samples achieved a higher cross-linking density than the samples cured under lab conditions only, also confirmed from the dynamic mechanical thermal analysis (DMTA) test results from which the cross-linking density, νe, was calculated at 30 °C and given in Table (refer to Figure S14, where the full storage modulus, G′, variation with temperature is combined for all samples). Lower E was measured for those samples, but that is expected as E and EB % follow opposite trends. However, the HPHU 10 films herein are still weak and soft, which is not a problem as the film is made from the base resin only. Finally, the results in Table also underscore the water resistance of HPHU 10 as the mechanical properties improved when the samples were immersed in water prior to being dried and tested.

Table 6

Comparison of Tensile and Rheological Tests Results for HPHU 10 Samples Cured under Lab Conditions for 1 Week and HPHU 10 Samples Cured for 1 Week under Lab Conditions and Another Week in Pure H2O

Samples	E (MPa)	EB %	σmax (MPa)	νe,30 °C (mol/m3)
cured under lab conditions for 1 week	1.2 ± 0.2	54 ± 11	0.30 ± 0.07	142 ± 55
cured under lab conditions for 1 week and in pure H2O for another week	0.75 ± 0.25	100 ± 8	0.42 ± 0.04	311 ± 11

Conclusions

The sol/gel curing method was taken advantage of in this work, and new hybrid partially bio-based PHUs were developed by end-capping the prepolymers with moisture-curable groups, such as epoxy or amino silanes, depending on the type of end-groups in the prepolymer chains. This type of curing would be crucial in developing new PHU sealants and adhesives, as the conventional PU counterparts are moisture-sensitive and cross-link when in contact with air under ambient conditions. A bio-based dicarbonate, DGC based on glycerol, was targeted in this work. DGC was synthesized and polymerized with diamines of different chain lengths; however, only the long-chain diamines proved to be suitable for this work, as the resulting prepolymers were liquid-like at room temperature. PHU 4, based on DGC and PPG diamine (Jeffamine D-2000), and PHU 10, prepared from DGC and PDMS diamine (PDMS-5k-(NH2)2), were found to be good candidates for end-capping. The end-capping reactions to place the appropriate silane moisture-curable groups were dictated by the nature of terminal groups and were effective under ambient conditions without the use of catalyst, marking one of the first times that hybrid moisture-curable HPHUs are possible under such conditions. The curing kinetics of some HPHUs were studied and typical gel times of 3 and 7.5 h were measured at 22 °C and 40–50% humidity under lab conditions. The steps followed to make the HPHUs of this work are summarized in Figure .

Figure 13

Summary of the method followed to synthesize moisture-curable HPHU sealants and adhesives.

Degradation and swelling studies in H2O were then conducted to ensure that the new films are resistant to water when used in outdoor applications. HPHU 4 films were swollen when immersed in H2O for a week, whereas HPHU 10 films were water-resistant as weight loss was negligible. Furthermore, the 1H NMR spectrum of the residual solvent HPHU 10 study indicated no significant leaching. The latter films were hydrophobic, exhibiting a water contact angle of 109°. The mechanical properties of HPHU 10 improved (EB % and σmax) after the samples were immersed in H2O as the curing process was more effective in that medium compared to that with the samples exposed to ambient conditions only. This proves again the water repellency of the PDMS-based derived HPHUs.

Materials and Methods

Materials

Diglycerol (DIG, ≥80% α,α, impurities consist of mono-, α,β-di-, β,β-di, and triglycerol) was obtained from Tokyo Chemical Industry (TCI). Dimethyl carbonate (DMC, ≥99%, anhydrous) and sodium methoxide (SOM, 95%, anhydrous powder) were purchased from Sigma-Aldrich and Acros, respectively. Ethyl acetate (EthAc, certified grade), tetrahydrofuran (THF, HPCL grade), and dimethyl formamide (DMF, HPLC grade) were purchased from Fischer Chemical. Water purified by a reverse osmosis process (pure H2O) was provided by the McGill Chemical Engineering Department. The diamines used in this work are 1,10-diaminodecane (Aldrich), Jeffamine D-2000, or PPG bis(2-aminopropyl ether) with Mn = 2000 g/mol (Aldrich), and PDMS–(NH2)2 or aminopropyl-terminated polydimethylsiloxane with Mn = 2500 g/mol (PDMS-2.5k-(NH2)2) (Aldrich), PDMS-(NH2)2 with Mn = 5000 g/mol (PDMS-5k-(NH2)2) from Gelest, and Priamine 1074 from Croda. Deuterated dimethyl sulfoxide (DMSO-d6) and deuterated chloroform (CDCl3) were provided from Sigma-Aldrich. Deuterium oxide (D2O, 99.9% D) was purchased from Cambridge Isotope Laboratories, whereas GLYMO or [3-(2,3-epoxypropoxy)-propyl]-trimethoxysilane, DAMO or N-(2-aminoethyl)-3-aminopropyltrimethoxysilane were purchased from Sigma-Aldrich. An acid buffer solution of pH = 3 at 20–25 °C was purchased from Fischer. All the chemicals were used as received.

Experimental Methods

Kinetic Study of DGC Reaction

DIG (6.05 g, 36.4 mmol) and SOM (184 mg, molar ratio of DIG to SOM of 10) were introduced into a 50 mL round-bottomed flask having a stir bar in it. The reactor was set on the stirrer in an oil bath controlled by a thermostat equipped with a temperature sensor and the temperature was adjusted to 75 °C. Meanwhile, the reactor mixture was purged with nitrogen (N2) for 15 min, after which 30 mL of DMC (364 mmol) was added to the mixture, and the reaction was allowed to proceed for 24 h. Aliquots were collected throughout the reaction at 30 min, 1, 3, 5, 7, 9, and 24 h. The collected samples were allowed to cool down to room temperature before drying them in air for 2 days to ensure that DMC and methanol had totally evaporated.

DGC Synthesis and Purification

DIG (30.0 g, 181 mmol) and SOM (970 mg, molar ratio of DIG to SOM of 10) were introduced into a 250 mL round-bottomed flask having a stir bar in it. 152 mL of DMC (1810 mmol) was then added before purging the reaction mixture with N2 for 15 min. The reactor was set on the stirrer in an oil bath controlled by a thermostat equipped with a temperature sensor, and the temperature was adjusted to 75 °C. The reaction was allowed to proceed for 8 h after which the colorless mixture turned light brown. The mixture was allowed to cool down to room temperature, and it was filtered to remove SOM (the catalyst). Unreacted DMC along with the formed methanol was evaporated, and the light brown residue was soaked for 2 h in 100 mL of pure H2O in an Erlenmeyer flask in which a stir bar was set to rotate slowly. After 2 h, the light brown-colored was transferred to water, and a beige solid precipitated. The residue was collected via filtration and dried overnight in a vacuum oven at room temperature. Then, the brown H2O solution was purged with air for 2 h, after which more beige residues precipitated, then filtered, and dried under vacuum like the previous sample. The process of purging the brown H2O solution with air and residue filtration was repeated three times until no residues were observed to precipitate from H2O solution anymore. The collected product was then soaked in 50 mL of EthAc for 30 min to remove the rest of the impurities coming from the byproducts originally found in the DIG sample. The resulting white solids were then filtered and dried overnight in the vacuum oven at room temperature. At the end of the process, 20.0 g of DGC was weighed corresponding to 51% yield based on the total amount of DIG originally loaded. The same procedure was repeated, but in that case the reaction was allowed to proceed for 24 h. The final DGC yield achieved in the latter case was 47%.

Polyaddition of DGC with Diamines (1/1 Molar Equivalent with AHEW and CEW)

2.0 g (9.17 mmol) of DGC was reacted in separate 25 mL round-bottomed flasks with 1.6 g (9.29 mmol) of 1,10-diaminodecane (PHU 1) or 4.9 g (9.18 mmol) of Priamine 1074 (PHU 2). The reactions were allowed to react at 80 °C until no stirring could be achieved in the reactor anymore (around 2 h for the former and 16 h for the latter). 1.0 g (4.59 mmol) of DGC was reacted in separate 50 mL round-bottomed flasks with 9.2 g (4.60 mmol) of Jeffamine D-2000, and the mixtures were allowed to react at 100 °C (PHU 3) and 120 °C (PHU 4) for 24 h and at 140 °C for 18 h (PHU 5) and 48 h (PHU 6). 0.50 g (2.29 mmol) of DGC was reacted in separate 50 mL round-bottomed flasks with 5.8 g (2.32 mmol) of PDMS-2.5k-(NH2)2; the polyadditions were allowed to proceed for 24 h at 80 °C (PHU 7), 100 °C (PHU 8), and 120 °C (PHU 9), then the same amount of DGC was reacted with 11.3 g (2.26 mmol) of PDMS-5k-(NH2)2, and the mixtures were left for 16 h (PHU 10) and 48 h (PHU 11) at 80 °C and 24 h at 100 °C (PHU 12) and 120 °C (PHU 13). All the batches were purged with N2 for 15 min prior to heating them in the oil bath to the specified temperatures, and every reactor was equipped with a high-viscosity stir bar, allowing good mixing of the monomers throughout the polyaddition reaction. Table summarizes the PHU formulations described herein. The amine hydrogen equivalent weight (AHEW) is calculated by dividing the diamine molecular weight by the number of amine hydrogens. For the diamines used herein, the AHEWs are as follows: 43 g/equiv for 1,10-diaminodecane, 134 g/equiv for Priamine 1074, 500 g/equiv for Jeffamine D2000, 625 g/equiv for PDMS-2.5k-(NH2)2, and 1250 g/equiv for PDMS-5k-(NH2)2. For DGC, the carbonate equivalent weight (CEW) is calculated to be 109 g/equiv after dividing the DGC molecular weight by the number of cyclic carbonates in the molecules.

Tailoring PHU Molecular Weights for Sealant and Adhesive Applications

Polymerizations leading to PHU 6 and PHU 11 were kinetically studied during which aliquots from the reactor were taken to track the progress of the PHU molecular weights. Aliquots at 1, 2, 3, 5, 10, 15, 18, 21, 24, 43, and 48 h were taken from PHU 6 polymerization batch, whereas aliquots were taken from PHU 10 batch every 2 h up to 16 h, then every 2 h between 24 and 32 h, and at 48 h. The evolution of the reaction was monitored by estimating the number-average degree of polymerization (DP) using Carothers equations for linear polymers in stoichiometric ratios[49]Here, M is the relative average-number molecular weight measured from gel permeation chromatography (GPC) (discussed in the next section), M0 is the relative average-number molecular weight of diamines in question (the diamines have a much higher molecular weight than DGC), and p is the extent of the reaction that could be found by recovering the unreacted DGC that precipitates in the bottom of the reaction medium.

End-Capping of PHU Prepolymers

PHU 4 and PHU 10 were selected for end-capping. An estimate of their absolute molecular weights was made based on their relative molecular weights and those of their respective diamines found from the GPC technique discussed below. From the estimated absolute molecular weights, the amount of end-cappers to be used, which is ideally set to 2 molar equiv per mol of polymer, was calculated. 2–3 g of polymers was mixed with 1, 2, 3, 4, and 6 molar equivalents of GLYMO or DAMO depending on the end-groups of the prepolymer chains. The end-capping reactions with PHU 4 were run at 120 °C from 75 min to 5 h, and those with PHU 10 were conducted at room temperature (22 °C) and 80 °C from 2 to 14 h. The polymer was purged with N2 for 15 min prior to adding the end-capper.

Characterization Methods

Proton Nuclear Magnetic Resonance Spectroscopy

Solution-phase NMR spectra were recorded on a Bruker 500 MHz instrument (16 scans) at ambient temperature. DGC was dissolved in DMSO-d6, whereas the PHU prepolymers were dissolved in CDCl3 for analysis.

FTIR spectroscopy

FTIR measurements were carried out on a PerkinElmer instrument equipped with a single-bounce diamond-attenuated transmission reflectance for solids and a zinc selenide (ZnSe) holder for liquids. 32 scans were recorded for each sample over the range 4000–500 cm–1 with a normal resolution of 4 cm–1. DGC, PHU prepolymers, and HPHU structures were recorded on the machine. The collected samples from DGC reaction kinetic studies were also analyzed on this machine by tracking the disappearance of the hydroxyl (OH) broad band in the 3200 cm–1 range and the appearance of the carbonyl (C=O) stretch in the 1700–1800 cm–1 range.

Gel Permeation Chromatography

The number-average molecular weight (Mn) and dispersity (D̵ = Mw/Mn) of prepolymer samples were measured using this technique on a Waters Breeze instrument with HPLC-grade THF as an eluent at a flow rate of 0.3 mL/min. Only PHU 1 was measured using HPLC-grade DMF as the eluent because it was not soluble in HPLC-grade THF. The GPC has three Waters Styragel HR columns (HR1 with a molecular weight measurement range of 102 to 5 × 103 g/mol, HR2 with a molecular weight measurement range of 5 × 102 to 2 × 104 g/mol, and HR4 with a molecular weight measurement range of 5 × 103 to 6 × 105 g/mol), a guard column, and a refractive index (RI 2414) detector. The columns were heated to 40 °C during analysis. The molecular weights were determined relative to PS calibration standards from Varian Inc. (ranging from 682 to 2,520,000 g/mol). The reported molecular weights were all relative to the PS standards and not adjusted with Mark–Houwink parameters. The PHU prepolymers’ molecular weights were analyzed using this instrument in addition to the samples collected from the polyaddition kinetic studies of PHUs 6 and 10.

Thermogravimetric Analysis

TGA was performed on a Q500 from TA Instruments system. The thermal degradation of the synthesized PHU prepolymers was measured at a heating rate of 10 °C/min over the temperature range of 25–600 °C under a nitrogen atmosphere. The onset degradation temperature (Td,onset) and the 10% degradation temperature (Td,10%) were calculated using this method.

Differential Scanning Calorimetry

Differential scanning calorimetry (DSC) was performed using a Q2000 TA Instruments calorimeter using standard hermetic aluminum pans, calibrated with indium and nitrogen as purge gases. The samples were analyzed at a heating rate of 10 °C/min over a temperature range of −80 to 120 °C under a nitrogen atmosphere. Glass-transition temperatures (Tg s) were calculated from the second heating ramp.

Rheology

Viscosities of PHUs 4 and 10 were measured on an Anton Paar Instruments rheometer (MCR 302) operated in the parallel plate steady shear mode (gap 1 mm). The samples were placed symmetrically in the center of the plate. The shear stress was measured at different shear rates ranging from 0.01 to 10 s–1 in 19 measurements for 3 min. All the measurements were conducted under ambient conditions.

The curing kinetics of the end-capped PHUs was carried out on the same instrument equipped with two parallel plate geometries at a frequency of 1 Hz and a strain of 1%, and the measurements were carried out at room temperature (around 22 °C) in a humidity range of 40–50%. The evolutions of storage modulus (G′), loss modulus (G″), damping factor (tan δ = G″/G′), and complex viscosity (η*) were monitored as a function of time for 20 and 7 h with PHUs 4 and 10, respectively. End-capped PHU 4 was allowed to cool down to room temperature before proceeding with these measurements.

Water Swelling

Water swelling of PHUs 4 and 10 (in the form of films) was evaluated using water content and absorption measurements at room temperature. Film samples with dimensions of 0.5 cm length, 0.5 cm width, and 0.1 cm thickness were immersed in 7 mL of pure H2O at room temperature. The water uptake was measured every 24 h for 1 week from immersing the samples. The EWC and EWA were estimated using eq and eq , respectively. Afterwards, the samples were dried in a vacuum oven overnight at 40 °C, and the gel content was measured using eq Here, Ws is the weight of the swollen sample, Wd is the weight of the dried sample, Wi is the initial weight, and Wf is the final weight of the dried sample.

HPHU Degradation in Water

The samples from the water swelling study were used herein. The water was dried under an air flow overnight, and the residuals, if any, were dissolved in D2O to be analyzed using 1H NMR. The degradation of HPHU 10 was also assessed in water with the pH adjusted to 3. Three samples of the films were taken and weighed over a week to check for any mass loss.

Water Contact Angle Measurements

Water contact-angle measurements were performed on an OCA 150 apparatus (DataPhysics Instruments GmbH) in the sessile drop configuration by the deposition of a 10 μL droplet of Milli-Q water at the rate of 2.0 μL/s with a 0.5 mL GASTIGHT #1750 syringe. The mean contact angle value on the HPHU 10 film was determined using SCA20 software from five repeated measurements conducted on different locations of the film.

Tensile Testing

Tensile properties were determined at ambient temperature using an EZ Test (Shimadzu) tensile machine at a speed of 20 mm/min with a load capacity of 10,000 N. Young’s modulus (E), tensile strength (σmax), and elongation at break (EB %) were estimated by the average of at least three repeated film samples of cured HPHU 10 films. Five samples were tested after 1 week of curing under ambient conditions and three others were tested after curing under ambient conditions for 1 week, and then in pure H2O for a second week. Free-standing dog-bone-shaped HPHU-10 samples were prepared using Teflon molds with the following dimensions: length of 50 mm, width of 3 mm, thickness of 1 mm, and gauge length of 25 mm. The tensile properties were monitored on WinAGS Lite software.

Dynamic Mechanical Thermal Analysis

This test was performed on an Anton Paar Instruments rheometer (MCR 302) using rectangular HPHU 10 films prepared using Teflon molds with the following dimensions: length of 50 mm (length cut to 45 mm when running the tests), width of 10 mm, thickness of 1 mm, and gauge length of 36 mm. Samples were loaded in tension, and a temperature ramp was performed from 25 to 120 °C at a rate of 5 °C/min, with an oscillation strain of 0.5% and a frequency of 1 Hz, and the evolutions of storage modulus (G′), loss modulus (G″), and damping factor (tan δ = G″/G′) were monitored for each sample. Three samples were tested after 1 week of curing under ambient conditions and three others were tested after curing under ambient conditions for 1 week, and then in pure H2O for a second week. The considered temperature range in this test corresponded to the rubbery plateau of HPHU 10 films. Hence, this test allowed the calculation of the cross-link density (νe) given by the equation below[49]Here, νe is the cross-link density in mol/m3, Gα′ is the rubbery plateau modulus at time α in Pa, R is the gas constant given in this case as 8.314 J/mol K, and Tα is the temperature at time α in K.