Controlled degradation of disulfide-based epoxy thermosets for extreme environments

Accepted Manuscript Controlled Degradation of Disulfide-Based Epoxy Thermosets for Extreme Environments Leah M. Johnson, Eric Ledet, Nicolas D. Huffman, Stephanie L. Swarner, Sarah D. Shepherd, Phillip G. Durham, Ginger D. Rothrock PII:

S0032-3861(15)00248-7

DOI:

10.1016/j.polymer.2015.03.020

Reference:

JPOL 17692

To appear in:

Polymer

Received Date: 19 January 2015 Revised Date:

6 March 2015

Accepted Date: 7 March 2015

Please cite this article as: Johnson LM, Ledet E, Huffman ND, Swarner SL, Shepherd SD, Durham PG, Rothrock GD, Controlled Degradation of Disulfide-Based Epoxy Thermosets for Extreme Environments, Polymer (2015), doi: 10.1016/j.polymer.2015.03.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Graphical Abstract

ACCEPTED MANUSCRIPT

Controlled Degradation of Disulfide-Based Epoxy

RI PT

Thermosets for Extreme Environments

Leah M. Johnson, Eric Ledet, Nicolas D. Huffman, Stephanie L. Swarner, Sarah D. Shepherd, Phillip G. Durham, Ginger D. Rothrock

M AN U

KEYWORDS: thermoset, disulfide, degradable polymer

SC

Emerging Technologies, Research Triangle Institute (RTI) International Research Triangle Park, NC, 27709, USA. Email: [email protected]

ABSTRACT

The burgeoning field of smart materials for oil exploration and production (E&P) demands robust polymers that remain stable in extreme conditions, but readily respond to specific chemical cues. Here, disulfide-based epoxy polymers are designed to withstand harsh, simulated

TE D

oil reservoir conditions while simultaneously retaining the ability to degrade from thiol compounds. Degradable epoxy thermosets are prepared by using Bisphenol-F diglycidyl ether (BFDGE) in combination with blends of p,p’-diaminodicyclohexylmethane (PACM) and 4aminophenyl disulfide (4APDS). These polymer systems retain their mechanical properties after

EP

exposure to extreme conditions (e.g., two weeks, 69 MPa, 100˚C, pH 12), but degrade in the presence of 2-mercaptoethanol (2-ME). Parameters that influence polymer degradation are

AC C

detailed including temperature, concentration of disulfide groups, and monomer stoichiometry. Formulations that entirely dissolve after exposure to 2-ME are utilized further for release of prototype tracers. Disulfide-based epoxy thermosets provide a unique platform for applications in harsh environments requiring robust materials capable of on-demand degradation.

ACCEPTED MANUSCRIPT

INTRODUCTION The design and utilization of innovative polymeric materials is the mainstay for solving numerous challenges in the petroleum industry. Polymers have been used to eliminate insoluble

RI PT

inorganic salt deposition on surfaces of production conduits (i.e., scale formation),[1] to prevent excess water waste during oil recovery (i.e., water shutoff),[2, 3] and to moderate bacterial growth to prevent accumulation of deleterious metabolites (e.g., hydrogen sulfide).[4] Recently, an innovative research direction involves the design of stimuli-responsive polymeric materials for oil E&P that are capable of characterizing specific reservoir properties and delivering

SC

chemical agents. For instance, these new materials include carbon black nano-reporters for hydrocarbon detection,[5] polymer functionalized magnetic nanoparticles for characterization of

M AN U

subsurface topography,[6] and stimuli responsive polymer microcapsules for chemical delivery.[7, 8] Many of these systems integrate specific chemical cues within the material that respond to environmental stimuli and perform pre-programmed functions, such as delivering chemical cargo or tracking migration history through rock formations.[9-13] The new stimuli-responsive materials used in oilfield applications must function under

TE D

the demanding environments of subsurface terrains where extreme temperatures, pressures, and salinity are commonplace. Although conditions vary widely depending on geography and depth, reservoirs contain both oily and aqueous regions that often exceed 35 MPa and temperatures of 100°C and beyond.[14] The requirement of materials to sustain structural integrity under these

EP

harsh conditions necessitates the design of material with unique properties that differ from many current stimuli-responsive materials used in mild environments, such as biological systems.[15] To support smart materials for energy applications, there exists a need for continued

AC C

advancement of materials capable of enduring these extreme reservoir conditions while responding to a specific chemical cues to perform an appropriate function. One material class with great benefit for extreme environments is the epoxy

thermosetting resins. The utility of epoxy thermosets spans various applications that require robust materials, including aerospace operations[16], microelectronics[17], automotive manufacturing[18], and biomedical device fabrication[19, 20]. The petroleum industry has also employed epoxy-based materials to coat pipelines[21] and to reinforce drilling wells.[22, 23] This overall versatility originates from the beneficial properties inherent to many epoxy systems,

2

ACCEPTED MANUSCRIPT

including thermal stability, excellent adhesion, and mechanical strength. Moreover, the minimal ease during handling and low volatiles during processing is attractive for needs in manufacturing scale-up and industrial operations. Epoxy materials can exhibit high monomer conversion resulting in tightly cross-linked networks with resilience to many solvents and capable of

RI PT

withstanding demanding environments.

The design of many stimuli-responsive thermosetting materials, such as epoxy systems, often requires the incorporation of specific cleavable crosslinks. For epoxy thermosets, the

SC

removal and degradation of the material often proves challenging owing to the inherent strength and stability of cured epoxy resins. For example, the removal of cured epoxy resins from substrates typically involves non-specific approaches such as solvent treatments (e.g., methylene

M AN U

chloride, toluene, sulfuric acid), photolytic bond scission[24] or applied mechanical force. An alternative approach for degrading cured epoxy resins involves a ‘command-destruct’ method,[25] wherein an extraneous trigger selectively degrades the material. To achieve controllable degradation profiles, the material is often prepared with monomers containing chemical functional groups that degrade in response to a chemical or thermal trigger. For example, monomers containing acetal and ketal functional groups were used to prepare epoxy

TE D

thermosets capable of degrading in the presence of acidic solvents.[25, 26] Additional degradable epoxy systems have incorporated Diels-Alder adducts within the network that decrosslink at elevated temperatures via cycloreversion of maleimide and furan side groups.[27-30] Other systems have incorporated sulfite,[31] carbamate,[32] carbonate,[33] tertiary esters,[34] or

EP

phosphate[35, 36] groups into the epoxy polymer to facilitate degradation reactions. Another beneficial chemical approach for controlling the breakdown of the epoxy

AC C

structure involves thiol-disulfide interchange chemistry,[37] where disulfide groups integral to the epoxy thermoset cleave specifically in response to a thiol compound. Pioneering work by Tesoro et al. demonstrated the ability to incorporate disulfide degradable crosslinks into an epoxy network for recovery and reuse of epoxy materials.[38-40] Under specific crosslinking densities and sample surface areas, disulfide-containing epoxy thermosets readily degrade in the presence of thiol compounds.[39, 40] To date, however, the ability to guide degradation kinetics of the epoxy thermoset by controlling the disulfide content has not been investigated. Moreover, evaluating the durability of this material after exposure to challenging environmental conditions

3

ACCEPTED MANUSCRIPT

(e.g., high pressure, salinity, and temperature) is essential for use as stimuli-responsive materials for E&P technologies or alternative demanding applications. Continued advancement of technologies for E&P requires further development of smart

RI PT

materials capable of withstanding harsh conditions while simultaneously maintaining function and sensitivity to specific, programmed cues. In this study, we demonstrate that disulfidecontaining epoxy materials may support this need and hold value for the emerging area of smart materials for energy applications. Herein, we show the ability to control the degradation of epoxy

SC

thermosets, which are otherwise stable in harsh, simulated reservoir conditions, by using a specific thiol trigger. This report details the parameters that dictate the degradation kinetics including temperature, concentration of disulfide groups, and monomer stoichiometry. We

M AN U

anticipate that incorporating disulfide moieties integral to an epoxy resin will benefit various industries that require robust materials capable of on-demand degradation, recycling applications, and self-healing capabilities. EXPERIMENTAL

Sample Preparation and Thiol Exposure Studies

TE D

Samples were prepared by combining the BFDGE epoxy monomer (epoxide equivalent weight of 169 (Momentive)) with appropriate amine monomer (PACM with amine hydrogen equivalent weight of 52.5 (Air Products) or 4APDS with amine hydrogen equivalent weight of 62.1 (TCI)). Samples contained either a stoichiometric ratio of monomers (i.e., 1:1 molar ratio of amine

EP

hydrogens: epoxide groups) or an off-stoichiometric ratio of monomers (i.e., 1.5:1 molar ratio of amine hydrogens:epoxide groups). Blends of 4APDS and PACM, in combination with BFDGE,

AC C

were prepared according to weight percent (e.g., 100:0, 75:25, 50:50, 25:75, 0:100 weight percent of 4APDS:PACM). If 4APDS was required for a formulation, it was dissolved with BFDGE using heat and subsequently cooled before added the PACM monomer. After the combined monomers were thoroughly dissolved, the samples were subsequently degassed by vacuum. All material in this manuscript was used as received without further purification. DMA bars were prepared by pouring the pre-mixed monomers into in-house fabricated aluminum molds containing DMA bar slots with dimensions of 35 mm × 12.7 mm × 2.6 mm. Before pouring the resin, the aluminum molds were coated with FreKote 700-NC release agent.

4

ACCEPTED MANUSCRIPT

All samples were cured at 80°C for 60 minutes followed by 150°C for another 120 minutes. For the solvent exposure studies at ambient pressure, cured epoxy sample bars (same dimensions as the DMA bar slots) were placed into 20 mL of 100% of 2-ME (Aldrich) and incubated at 25°C or 80°C for a designated time. The weight of the specimens were acquired before (W0) and after

RI PT

immersion into the solvent (W). After removal from the solvent, the surfaces of the samples were

was calculated using the formula: %W∆ = [(W-W0)/W0] X 100

High Pressure High Temperature (HPHT) Studies

SC

dried prior to weighing by gently wiping the surface. The percentage of weight change (%W∆)

M AN U

HPHT experiments were performed using a Fan Model 275 consistometer. Epoxy sample bars were immersed either into a retarded cement slurry of pH=12.1 (4% cement retarder) or 100% refined mineral petroleum oil (Ibex Chemicals, Inc.) and loaded into the HPHT chamber. The samples were maintained at 69 MPa and 100°C for either one or two weeks, as indicated. Samples were rinsed with water or acetone and patted dry after removal from the HPHT.

TE D

Characterization

Dynamic Mechanical Analysis (DMA). Experiments were performed using a DMA instrument equipped with a single cantilever clamp (Q 800 TA Instruments). DMA tests were performed with a single frequency of 1 Hz and amplitude of 5 µm. The temperature ramp occurred from

EP

30˚C to 180˚C at 2°C/minute. All DMA experiments were repeated at least three times for each condition. The glass transition temperature was taken from the maximum of the peak in the α-

AC C

transition region of the tan delta curve. The molecular weight between crosslinks (Mc) was determined using the following equation: Mc = 3ρRT/E

where ρ=density of the polymer, R=gas constant, T=absolute temperature where the modulus is acquired, and E=storage modulus in the rubbery region (i.e., from 30°C to 35°C above the Tg).[40] Although certain assumptions require further understanding,[41] the relative Mc values between different formulations offer insight into this current system.

5

ACCEPTED MANUSCRIPT

Differential Scanning Calorimetry (DSC). DSC measurements were performed using a Q200 DSC instrument (TA Instruments). The uncured epoxy resin was placed in a hermetic aluminum pan containing a puncture to enable escape of volatile components. DSC scans were performed by ramping the temperature at 10°C/min. The heat of reaction (∆Hrxn) for the cure was

RI PT

determined by integrating the area under the DSC curve for each formulation and the residual cure (∆Hres) was determined by performing dynamic DSC on cured polymer formulations.[42] Thermogravimetric analysis (TGA). TGA measurements were performed using a Q50 instrument (TA Instruments). Experiments were performed within a nitrogen gas (N2) atmosphere at a ramp

SC

rate of 5°C/min from 25°C to 400°C.

UV-Vis Spectroscopy. UV-VIS spectra were acquired using a Cary Series UV-VIS-NIR

M AN U

spectrophotometer (Agilent Technologies) with a quartz cuvette. All samples were blanked against 100% 2-ME. For the UV-VIS studies, the appropriate epoxy samples were immersed in 20 mL of 100% 2-ME and incubated at 80˚C. A 100µL aliquot of the solution was removed from the vial every hour and diluted (1/1,600) in 100% 2-ME before evaluating with UV-VIS. Tracer Release Studies. Core-shell macro-particles were prepared in two steps. First, the cores

TE D

were prepared by combining stearic acid (Aldrich) with approximately 0.025 weight % of rhodamine-B (Aldrich) and co-dissolving at 100°C. The hot solution was immediately poured into in-house fabricated silicone molds containing half-cylindrical indentations (3 millimeter diameter). After cooling, the cores were released from the flexible silicone mold and used for the

EP

subsequent step. Core-shell particles were assembled by first pouring the epoxy resin into inhouse fabricated aluminum molds containing half-cylindrical indentations (4 millimeter

AC C

diameter) and subsequently placing half cylindrical cores within the epoxy resin. Particles were then cured at 80°C for one hour and then 150°C for two hours. For the particles comprising 4APDS, the samples were first incubated at 55°C overnight to facilitate hardening of the epoxy material below the melting point of stearic acid. The macro-particles were immersed in either brine (8 wt % CaCl2, 2 wt % NaCl) or 100% 2-ME at 80°C for 10 hours. The fluorescence (555 excitation/580 emission) was then evaluated using a BioTek Synergy Mx plate reader.

6

ACCEPTED MANUSCRIPT

RESULTS AND DISCUSSION Effect of Disulfide Content and Temperature on the Degradation of Epoxy Thermosets Epoxy thermosets that contain disulfide groups degrade in the presence of thiol

RI PT

compounds under certain conditions, as shown by Tesoro et al.[40] To date, however, the ability to tune degradation properties of epoxy thermosets via disulfide content has not been investigated, which is important when using this system for controlled release applications. Here, we characterized the mechanical properties and degradation profiles of materials prepared using

SC

conventional monomers (BFDGE, PACM) and a disulfide monomer (4APDS) (Figure 1). To ensure that the disulfide bond of 4APDS does not affect polymerization, an analogue compound devoid of primary amines (diphenyl disulfide (DPDS)) was combined with BFDGE and shows

M AN U

no exothermic cure profile in DSC analysis (Figure 1, Supplemental 1). Conversely, the combination of BFDGE with PACM or 4APDS reacted, exhibiting minimal residual cure for all

EP

TE D

formulations (Supplemental 2).

AC C

Figure 1. Chemical structures of compounds used in this study. To control the quantity of disulfide groups within the specimens, a series of polymer samples were prepared using blends of amine monomers (i.e., ratios of PACM:4APDS at 100:0, 75:25, 50:50, 25:75, 0:100), while maintaining stoichiometry between total amine hydrogens and epoxide chemical groups. Without exposure to thiol, the polymers prepared with disulfide monomers exhibit comparable mechanical properties to systems without disulfide groups. For example, the systems containing 100% 4APDS or 100% PACM both exhibit tan delta curves with a narrow damping peak of the α-transition, indicating a highly crosslinked network (Figure 2). All formulations evaluated here show similar mechanical properties with glass transition

7

ACCEPTED MANUSCRIPT

temperatures (Tg) greater than 130˚C and storage modulus values greater than 2300 MPa (at 35˚C) (Table 1). The onset of decomposition (TD) decreases slightly with increasing content of disulfide moieties, likely due to the thermal lability of the disulfide bond (Table 1, Supplemental 3).[40] For example, the TD decreases from 334.6 ±0.2 ˚C to 292.8 ±0.7 ˚C for

RI PT

100% PACM and 100% 4APDS, respectively. Taken as a whole, these results demonstrate the capacity to prepare stable thermoset epoxy materials with tunable quantities of disulfide moieties

TE D

M AN U

SC

that can be used as degradation points within the polymer.

EP

Figure 2. Representative DMA profiles showing the (i) storage modulus and (ii) tan delta curves of formulations comprising a stoichiometric ratio of BFDGE with (A) 100% PACM or (B) 100% 4APDS. These profiles show that the inclusion of disulfide moieties do not significantly alter the mechanical properties of the epoxy polymer.

AC C

Table 1. Properties of polymer samples containing disulfide groups without exposure to thiol.* Three samples were tested for each formulation. Storage PACM: Mc Modulus (MPa) Tg (˚C) TD (˚C) 4APDS at 35˚C 100:0

2432 (±42)

139.7 (±0.4)

334.6 (±0.2)

412 (±17)

75:25

2464 (±141)

133.3 (±2.0)

322.8 (±1.2)

615 (±15)

50:50

2503 (±49)

135.0 (±1.5)

319.4 (±0.4)

621 (±29)

25:75

2424 (±218)

139.3 (±1.6)

302.5 (±0.5)

661 (±28)

0:100 2548 (±128) 143.7 (±1.5) 292.8 (±0.7) 621 (±23) *Prepared with stoichiometric ratios of total amine hydrogen to epoxide groups (BFDGE).

8

ACCEPTED MANUSCRIPT

We next evaluated the effect of the disulfide content on the degradation properties of the polymers by immersing samples into 2-ME at 80˚C for different time periods. After exposure to 2-ME, the mechanical properties of the epoxy samples varied in a manner that correlated with the quantity of disulfide within the polymer (Figure 3). For example, after three hours in 2-ME,

RI PT

polymer samples comprising a 75:25 ratio of PACM:4APDS exhibited a storage modulus of 840 ±63 MPa (at 35˚C), whereas epoxy samples containing a 100:0 ratio of PACM:4APDS exhibited a storage modulus of 2095 ±81 MPa (at 35˚C). Furthermore, Figure 3 shows that the storage modulus decreased with longer durations of immersion in 2-ME. For instance, the storage

SC

modulus of samples containing a 75:25 ratio of PACM:4APDS decreased over an order of magnitude from 1313 ±67 MPa to 103 ±15 MP, after immersion in 2-ME for one hour and 24

M AN U

hours, respectively. Of important note, these DMA studies capture the mechanical properties of partially degraded, wet samples with inhomogeneous structures that have an intact, stiff inner portion and a degraded outer portion resultant of thiol degradation. This report of storage modulus, therefore, is measuring the response of the disulfide-material to the thiol solvent (i.e., surface erosion) rather than providing a measure of a homogenous material. Nevertheless, the decrease in the magnitude of the storage modulus at temperatures below the Tg demonstrates the

TE D

response of the material to the thiol chemistry, wherein a reduction in the crosslink density likely

AC C

EP

occurs due to the scission of disulfide bonds within the polymer by 2-ME.

9

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Figure 3. Storage modulus (measured at 35˚C) for polymers immersed in 100% 2-ME at 80°C for differing times. Polymer samples comprised BFDGE and (●) 0:100, (■) 25:75, (▲) 50:50, and (▼) 75:25, and (♦) 100:0 of PACM:4APDS. All samples were prepared with a stoichiometric ratio of total amine hydrogens to epoxide groups. Experiments were performed in triplicate. In general, the higher content of 4APDS within the cured polymer resulted in higher

TE D

weight gain and swelling of the sample when exposed to 2-ME for identical times (Figure 4A). For example, samples containing 100% 4APDS gained 91 ± 4 % of the initial weight after 16 hours in 2-ME, whereas samples containing 50% 4APDS gained 46 ± 3 % of the initial weight after 16 hours. The correlation between disulfide content and weight gain supports the thiol-

EP

based degradation mechanism, wherein the cleavage of disulfide groups within the polymer network creates diffusion routes for additional solvent absorption (Supplemental 4). Notably, the sample comprising 100% 4APDS shows similar weight gain as compared to 25:75

AC C

(PACM:4APDS), which suggests that a threshold disulfide content within the polymer is reached that enables adequate penetration of the thiol compound. This concept is further supported by the mechanical properties in Figure 3, where the 100% 4APDS samples exhibit slightly higher storage modulus values as compared to 25:75 PACM:4APDS samples after exposure to 2-ME. This trend in degradation behavior requires further investigation to fully understand the resultant material properties at these disulfide contents in this system. As demonstrated in the images in Figure 4B, the degradation proceeds via surface erosion showing an evident boarder of eroded polymer (light yellow color) encompassing the

10

ACCEPTED MANUSCRIPT

intact inner portion. As shown in Figure 4B (time=0), the cured polymer is dark, but becomes a lighter yellow color when degraded by 2-ME. Samples devoid of 4APDS absorbed minimal amount of 2-ME solvent, showing 15 ± 1% weight gain after 24 hours in 2-ME, likely from slight plasticizing by the solvent. Moreover, control studies that exposed 100% 4APDS samples

RI PT

to 1-propanol absorbed minimal quantity of solvent (i.e., less than 1%), supporting further the specific thiol based cleavage mechanism. The samples comprising BFDGE and 100% 4APDS ruptured after an 18 hour immersion in 2-ME, ultimately splitting into multiple fragments, as

TE D

M AN U

SC

shown Figure 4B.

AC C

EP

Figure 4. Weight gain of disulfide-based polymers exposed to 100% 2-ME. (A) Percent weight change for polymers immersed in 100% 2-ME for differing times at 80°C. Polymer samples comprised BFDGE and (●) 0:100, (■) 25:75, (▲) 50:50, and (▼) 75:25, and (♦) 100:0 of PACM:4APDS. All samples were prepared with a stoichiometric ratio of total amine hydrogens to epoxide groups. (B) Digital camera images showing the progressive degradation of a sample comprising BFDGE and 0:100 of PACM:4APDS after immersion in 100% 2-ME at 80°C for different times. The scale bar represents approximately 12.75 mm. Experiments were performed in triplicate. Although thiol-disulfide interchange occurs at room temperature,[37] we anticipated that

higher temperatures may be required to achieve the cleavage of disulfide embedded within a tightly crosslinked polymer. Here, we studied the effect of temperature on our disulfide-based epoxy polymers by immersing samples in 100% 2-ME at differing temperatures and subsequently analyzing via DMA. Our results indicated that the temperature used during the exposure to thiol greatly influenced the resultant mechanical properties and behavior of the 11

ACCEPTED MANUSCRIPT

material. For example, regardless of disulfide content, the epoxy material remained relatively unaffected after immersion in 100% 2-ME at ambient room temperature for one month (Figure 5B, Supplemental 5) and exhibited similar mechanical properties compared to samples without exposure to thiol (Figure 5A). Conversely, after immersion in 100% ME at 80°C for only three

RI PT

hours, the storage modulus rapidly decreased in a manner that correlated with the quantity of disulfide in the polymer, as shown in Figure 5C. Although polymer without disulfide moieties (i.e., 100% PACM) showed a slight decrease in storage modulus and an appearance of an additional tan delta peak, likely correlating with the slight uptake of solvent, which often occurs

SC

in thermoset systems.[43, 44] These results suggest that increases in temperature facilitate solvent diffusivity through the polymer. Although 80°C is below the Tg of the polymer samples,

M AN U

this increase in temperature may be sufficient to enable diffusion of the 2-ME through the polymer. Noticeably, the drop in storage modulus and increase in weight occurred more for the samples containing 75% APDS, as compared to samples containing 100% APDS. This suggests that the content of disulfide is adequate at the 75% APDS formulation to enable degradation and

AC C

EP

TE D

imbibing of 2-ME. Additional studies would be needed to fully understand this occurrence.

Figure 5. The storage modulus as a function of temperature for disulfide-based polymer samples. Polymer samples comprised BFDGE and (i) 100:0, (ii) 75:25, (iii) 50:50, (iv) 25:75, and (v) 0:100 of PACM:4APDS. Samples were evaluated after (A) no exposure to thiol, (B) exposure to 100% 2-ME at ambient room temperature for four weeks or (C) exposure to 100% 2-ME at 80°C for three hours. All samples were prepared with a stoichiometric ratio of total amine hydrogens to epoxide groups. Experiments were performed in triplicate. This section demonstrates the capacity to design epoxy systems that contain tunable quantities of disulfide groups that specifically respond to thiol compounds. These results show

12

ACCEPTED MANUSCRIPT

that the temperature and quantity of disulfide groups within the polymer can be used to control the structural breakdown of the epoxy thermoset material. For E&P applications, temperature gradients exist within subsurface formations due to geothermal variations. However, 80˚C is an average temperature representing approximately 3000 meters in typical reservoirs.[14] In the

RI PT

next section, we explore further the crosslinking density of the material as a means for complete dissolution of the disulfide-containing epoxy via cleavage by 2-ME. Tuning Stoichiometric Ratios to Control Polymer Degradation

SC

Polymeric dissolution is valuable for various applications including plastics recycling, tissue engineering, membrane science, and microlithography.[45] Likewise, chemicals used in many E&P applications also rely upon controlled dissolution of polymers to selectively target

M AN U

viscosities necessary for delivery and recovery from the reservoir.[4] Here, we investigated the dissolution of the disulfide-based epoxy polymers by thiol to understand the potential role of this material for new smart material applications. The disulfide-containing polymers studied in the previous section, which contain a stoichiometric ratio of epoxide groups in BFDGE to total amine groups in the curative blends (i.e., 4APDS, PACM), ruptured into multiple solid fragments after immersion in 100% 2-ME at 80˚C. The rate of the polymer rupture depended on the content

TE D

of disulfide, with samples containing higher disulfide content rupturing much faster. Attempts to completely liquefy stoichiometric polymers (i.e., stoichiometric ratio of BFDGE + 100% APDS) within 100% 2-ME at 80°C for one week were unsuccessful. This finding was in agreement with reports by Tosoro et al., which indicated that a threshold crosslinking density and greater surface

EP

area of the disulfide-containing epoxy polymer was required for complete solubility by a tri-nbutyl phosphine reducing agent.[39, 40] As shown in Table 1, the molecular weight between

AC C

crosslinks (Mc) values for the stoichiometric disulfide polymers ranged from 615 ±15 to 661 ±28, indicating that these crosslinking densities are inadequate for polymer dissolution under our experimental conditions.

We reasoned that a decrease in the crosslinking density would facilitate complete

degradation and dissolution of the disulfide-based epoxy polymers by thiol. By lowering the distance between crosslinks within the networks, permeation by a thiol compound would increase and facilitate thiol-disulfide interchange reactions. To evaluate the effect of monomer stoichiometry on thiol-mediated degradation profiles, off-stoichiometric samples containing an

13

ACCEPTED MANUSCRIPT

excess of amine (i.e., 1.5:1, amine hydrogens to epoxy groups) were prepared. Table 2 shows that using an off-stoichiometric ratio of functional groups results in higher Mc values, as expected. Moreover, the off-stoichiometric samples also exhibit a slightly lower Tg, further

RI PT

supporting the loosening of the polymeric network.

SC

Table 2. Properties of polymer samples containing disulfide groups without exposure to thiol.* Three samples were tested for each condition. Storage PACM: Modulus (MPa) Tg (˚C) TD (˚C) Mc 4APDS at 35˚C 2263 (±22)

108.9 (±0.7)

334.6 (±0.2)

75:25

2692 (±95)

108.4 (±1.1)

326.0 (±1.0)

849 (±42)

50:50

2886 (±39)

111.2 (±1.2)

299.4 (±0.5)

968 (±37)

25:75

2520 (±128)

114.7 (±1.2)

291.5 (±0.5)

1225 (±38)

M AN U

100:0

1239 (±55)

TE D

0:100 2488 (±111) 115.9 (±1.0) 268.7 (±0.6) 1373 (±11) *Prepared with off-stoichiometric ratios (1.5:1) of total amine hydrogens to epoxide (BFDGE) groups.

Next, we evaluated the effects of immersing the off-stoichiometric samples in thiol for different time periods. In correlation with the on-stoichiometric samples (Figure 4), a higher

EP

content of 4APDS within the cured polymer resulted in higher weight gain and swelling of the sample (Figure 6A). However, the weight gain of the off-stoichiometric samples occurred faster

AC C

than the on-stoichiometric samples. For example, after two hours in 100% 2-ME, polymer samples comprising 75% 4APDS (75:25, 4APDS:PACM) gained 17 ±1 weight% for onstoichiometric and 66 ±1 weight% for off-stoichiometric ratios. Notably, the off-stoichiometric formulation comprising 100% 4-APDS completely dissolved in 2-ME at 80°C after approximately 8.5 to 9 hours and easily ruptured upon handling after approximately after 5.5 hours (Figure 6B). During the dissolution of the off-stoichiometric 100% 4APDS sample in 2ME the thiol solution turned yellow, which resembled the color of the 4APDS monomer. The dissolution kinetics of the sample was evaluated by monitoring the increase in absorbance at 310 nm over time of this solution (Figure 7). Figure 7A demonstrates the importance of the

14

ACCEPTED MANUSCRIPT

crosslinking density on the dissolution kinetics, where on-stoichiometric samples, that exhibit lower Mc values, do not show appreciable dissolution. The digital camera images in Figure 7B (taken from a video in Supplemental 6) show the entire dissolution of an off-stoichiometric epoxy sample containing 100% 4APDS after approximately 7 hours under accelerated

TE D

M AN U

SC

100°C followed by 3 hours at 80°C during imaging).

RI PT

dissolution conditions to accommodate video acquisition (i.e., ~4 hour accelerated degradation at

AC C

EP

Figure 6. (A) Percent weight change for polymers prepared with an off-stoichiometric ratio of total amine hydrogens to epoxide groups in BFDGE (1.5:1). All samples were immersed in 100% 2-ME at 80°C for differing times. Polymer samples comprised BFDGE and (●) 0:100, (■) 25:75, (▲) 50:50, and (▼) 75:25, and (♦) 100:0 of PACM:4APDS. (B) Digital camera images showing the degradation of off-stoichiometric samples comprising BFDGE and 100% 4APDS after immersion in 100% 2-ME at 80°C for different times. The scale bar represents approximately 12.75 mm. Experiments were performed in triplicate.

15

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 7. (A) UV-Vis analysis of solutions containing samples of BFDGE and 4APDS with stoichiometry of (●) 1:1 (■) 1.5:1 of total amine to epoxy groups. The polymers were immersed in 100% 2-ME at 80°C for different time periods. (B) Representative digital camera images of off-stoichiometric 4APDS samples during immersion in 100% 2-ME at 80°C. To accelerate video acquisition, the sample was pre-conditioned at approximately 100°C for four hours prior imaging at 80°C. The disulfide-containing polymer entirely dissolved in 2-ME, producing a yellow colored.

TE D

Effect of Harsh Environments on Disulfide-Based Epoxy Polymers Materials used for oilfield applications must exhibit robust properties capable of withstanding extreme conditions including high temperatures and high pressures. Here, studies

EP

were performed to evaluate the mechanical properties of disulfide-based epoxy polymers after exposure to conditions often encountered in subterranean environments. The precise environmental conditions experienced by a material during E&P applications depend upon the

AC C

particular reservoir, which may exhibit differing characteristics based on location, age and formation type (e.g., carbonate). Here, to simulate conditions typical during well transit and subsurface delivery, on-stoichiometric samples were immersed in either mineral oil or a retarded cement slurry of pH 12 and placed within a consistometer at 100˚C and 69 MPa for two weeks. As shown in Table 3, the storage modulus (at 35°C) of all epoxy polymers remains similar with and without exposure to simulated reservoir conditions. For instance, the storage moduli remain above 2000 MPa and Tg values (taken as the second tan delta peak) remain at or above 135˚C for all disulfide-based polymer formulations. Additionally, off-stoichiometric samples also

16

ACCEPTED MANUSCRIPT

maintained their mechanical integrity after exposure to HPHT conditions. For example, polymers prepared with an off-stoichiometric ratio of total amine hydrogen to BFDGE (1.5:1) showed a storage modulus (at 35°C) of 2688 ±212 for 100% 4APDS and 2379 ±43 for 100% PACM samples after immersion in mineral oil at 100°C and 69 MPa for one week. As expected, the

RI PT

height of the tan delta peak decreased in all systems after exposure to HPHT conditions, which may correspond to slight plasticization from water or oil (Figure 8).[43] Moreover, the appearance of an additional tan delta peak was apparent, which is common in many thermosetting materials after exposure to solvents and may result either from drying during the

SC

DMA scan or from a differential plasticization effect within the polymer network.[44] Despite the alterations in the tan delta curves, the materials retained their overall structural integrity

to harsh environmental conditions.

M AN U

demonstrating that these epoxy polymers, regardless of disulfide content, can withstand exposure

Table 3. Mechanical properties of on-stoichiometric polymer samples* containing disulfide groups with and without exposure to HPHT conditions** Three samples were tested for each condition. Storage Modulus (MPa) at 35˚C

Tg (˚C)

HPHT #2

2453 (±27)

2447 (±55)

No Exposure 139.7 (±0.4)

HPHT #1

100:0

No Exposure 2432 (± 42)

136.8 (±0.5)

136.6 (±1.6)

75:25

2464 (± 141)

2452 (±130)

2407 (±53)

133.3 (±2.0)

137.6 (±0.6)

139.1 (±0.3)

50:50

2503 (± 49)

2482 (±43)

2587 (±62)

135.0 (±1.5)

139.8 (±0.4)

139.3 (±0.03)

25:75

2424 (± 218)

2512 (±178)

2476 (±17)

139.3 (±1.6)

140.8 (±0.6)

140.5 (±0.5)

0:100

2548 (± 128)

2449 (± 135) 143.7 (±1.5)

140.6 (±0.2)

143.2 (±0.3)

HPHT #2

TE D

HPHT #1

EP

PACM: 4APDS

2463 (±145)

AC C

*Prepared with a stoichiometric ratio of total amine hydrogen to epoxide groups (BFDGE) **HPHT condition #1: two weeks at 100˚C, 69 MPa, cement slurry at pH =12 HPHT condition #2: two weeks at 100˚C, 69 MPa, mineral oil

17

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 8. Representative tan delta curves of cured samples comprising BFDGE and onstoichiometric ratios of (A) 100% PACM or (B) 100% 4APDS. The tan delta curves show samples (i) without exposure to HPHT conditions or after exposure to 100˚C, 69 MPa for two weeks in either (ii) mineral oil or (iii) retarded slurry cement, pH 12. Importantly, the exposure of disulfide-based epoxy polymers to these extreme conditions

TE D

does not hinder the inherent capability of the material to subsequently degrade in the presence of thiol compounds. For example, even after exposure to mineral oil at 100˚C and 69 MPa for one week, samples retain the capacity to degrade in the presence of 2-ME (Supplemental 7). This suggests that the harsh environmental conditions tested here do not alter the chemical

EP

functionality of the disulfide groups within the polymer, which may enable controlled degradation of materials in-situ within the harsh conditions. To further explore the utility of this

AC C

concept, systems were designed to release model tracers (i.e., rhodamine-B) resultant of polymer degradation. To this end, macro-scale core shell particles containing rhodamine-B cores and different off-stoichiometric epoxy shells (i.e., 4APDS or PACM) were prepared (Figure 9) and subsequently immersed in either brine (8wt% CaCl2, 2wt% NaCl) or 100% 2-ME at 80°C for 10 hours. As expected, all tested macro-particles remained intact when exposed to brine. When exposed to 2-ME, all 4APDS macro-particles entirely dissolved, whereas the PACM macroparticles did not dissolve. In cases where particles had cores, the 4APDS samples released the rhodamine-B tracer during the shell dissolution in 2-ME (relative fluorescence units (RFU) of 20462 ±1479), but did not release the tracer when exposed to brine (RFU of 355 ±559). As

18

ACCEPTED MANUSCRIPT

previously discussed, the 2-ME plasticized the PACM after the prolonged exposure to 2-ME at 80°C, which resulted in a swelling and cracking of the thin PACM shell (approximately 1mm diameter) and the release of the rhodamine-B (RFU of 63085 ±9196). As shown in Figure 9,

M AN U

SC

RI PT

however, the PACM macro-particles did not dissolve.

EP

TE D

Figure 9. Digital camera images of macro-particles exposed to different conditions. The samples comprise (i) solid 4APDS particles (no core), (ii) particles with a 4APDS shell and rhodamine-B/stearic acid core, (iii) solid PACM particles (no core), and (iv) particles with a PACM shell and rhodamine-B/stearic acid core. Samples were immersed in either 100% 2ME or brine (8wt% CaCl2, 2wt% NaCl) for 10 hours at 80°C. The particles were prepared with a 1.5:1 stoichiometric ratio of total amine hydrogens to epoxide groups from BFDGE.

AC C

This section demonstrates that disulfide-based epoxy polymers hold potential as a stimuli-responsive, controlled release system for applications in harsh environments. The polymers are mechanically intact after exposure to HPHT conditions, but degrade when exposed to thiol compounds. Notably, these overall studies evaluated the degradation behavior of polymers in response to pure thiol (i.e., 100% 2-ME), so future investigations are required to understand further the effects of lowered thiol concentrations on degradation kinetics. Evaluating the degradation behavior from decreased thiol concentrations may prove important, given that high thiol concentrations (e.g., as here with 100% 2-ME) are unlikely in many real-world, natural environments. Nevertheless, these studies provide important insight into the mechanical

19

ACCEPTED MANUSCRIPT

properties and degradation kinetics of the disulfide-based thermosets, which can be leveraged for future research endeavors targeting applications benefiting from robust materials capable on ondemand degradation.

RI PT

CONCLUSIONS

Advanced polymeric materials offer innovative solutions to E&P challenges, such as providing new ways to monitor the reservoir (e.g., micro/nano sensors), to redirect waterflood routes (e.g., smart fluids), and to supplement current drilling and completion activities. To

SC

achieve these solutions, many materials must remain stable until a pre-selected time, whereupon the system breaks down from a specific trigger. In this study, we show that the inclusion of disulfide functional groups within epoxy thermoset material satisfied many of these requirements

M AN U

by enabling a mechanically durable material within harsh reservoir-type environments that exclusively responds to a thiol compound (i.e., 2-ME). In this study, we showed that the degradation profiles of disulfide-containing epoxy thermosets, in contact with 100% 2-ME, are precisely controlled by tuning key parameters including the temperature, the stoichiometry of monomers, and quantity of disulfide groups. In particular, we show that the stoichiometry of

TE D

monomers greatly influences the capability of complete solubilization of the epoxy thermosets. These studies show further that disulfide-containing epoxy materials are stable within harsh conditions often encountered within a reservoir environment. For example, epoxy material show near identical mechanical properties with and without exposure to harsh conditions for two

EP

weeks (69 MPa, 100°C in oil or cement slurry at pH =12), but maintain the ability to respond to a controlled chemical trigger and degrade on demand.

These studies offer insight into the

mechanical properties and degradation behavior of the disulfide-based epoxy thermosets, which

AC C

may benefit applications benefit applications requiring the precise placement and subsequent removal or degradation of resilient, durable materials. Acknowledgements

The authors are grateful to the Advanced Energy Consortium (AEC) and member companies (Shell, Petrobras, BG Group, Schlumberger, BP, Total, Repsol, and Statoil) for their support and encouragement in this research. We thank all RTI support staff, particularly Richard Daw for his assistance with DSC.

20

ACCEPTED MANUSCRIPT

References

6.

7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17.

18. 19. 20. 21. 22.

21

RI PT

SC

5.

M AN U

4.

TE D

3.

EP

2.

Verraest D, Peters J, van Bekkum H, and van Rosmalen G. Journal of the American Oil Chemists’ Society 1996;73(1):55-62. Lahalih SM and Ghloum EF. Journal of Applied Polymer Science 2007;104(4):20762087. Zaitoun A, Tabary R, Rousseau D, Pichery TR, Nouyoux S, Mallo P, and Braun O. Using Microgels to Shut Off Water in a Gas Storage Well. International Symposium on Oilfield Chemistry. Houston, Texas: Society of Petroleum Engineers, 2007. Kelland MA. Production Chemicals for the Oil and Gas Industry. Boca Raton, FL: CCR Press, 2009. Hwang C-C, Wang L, Lu W, Ruan G, Kini GC, Xiang C, Samuel ELG, Shi W, Kan AT, Wong MS, Tomson MB, and Tour JM. Energy & Environmental Science 2012;5(8):8304-8309. Bagaria HG, Xue Z, Neilson BM, Worthen AJ, Yoon KY, Nayak S, Cheng V, Lee JH, Bielawski CW, and Johnston KP. ACS Applied Materials & Interfaces 2013;5(8):33293339. Abbaspourrad A, Carroll NJ, Kim S-H, and Weitz DA. Journal of the American Chemical Society 2013;135(20):7744-7750. Abbaspourrad A, Datta SS, and Weitz DA. Langmuir 2013;29(41):12697-12702. Krishnamoorti R. Journal of Petroleum Technology 2006;58(11):24-25. Matteo CP, Candido, R., Fransesca, V. . Am. J. Applied Sci. 2012;9:784-793. Berlin JM, Yu J, Lu W, Walsh EE, Zhang L, Zhang P, Chen W, Kan AT, Wong MS, Tomson MB, and Tour JM. Energy & Environmental Science 2011;4(2):505-509. Park YC, Paulsen J, Nap RJ, Whitaker RD, Mathiyazhagan V, Song Y-Q, Hürlimann M, Szleifer I, and Wong JY. Langmuir 2014;30(3):784-792. ShamsiJazeyi H, Miller CA, Wong MS, Tour JM, and Verduzco R. Journal of Applied Polymer Science 2014;131(15):n/a-n/a. Satter A, Iqbal GM, and Buchwalter JL. Practical Enhanced Reservoir Engineering: Assisted with Simulation Software: PennWell Corporation, 2008. Stuart MAC, Huck WTS, Genzer J, Muller M, Ober C, Stamm M, Sukhorukov GB, Szleifer I, Tsukruk VV, Urban M, Winnik F, Zauscher S, Luzinov I, and Minko S. Nat Mater 2010;9(2):101-113. Moffet ML and Saponara VL. Journal of Applied Polymer Science 2013;130(6):39613971. Kinjo N, Ogata M, Nishi K, Kaneda A, and Dušek K. Epoxy Molding Compounds as Encapsulation Materials for Microelectronic Devices. Speciality Polymers/Polymer Physics, vol. 88: Springer Berlin Heidelberg, 1989. pp. 1-48. Abu Talib AR, Ali A, Badie MA, Azida Che Lah N, and Golestaneh AF. Materials & Design 2010;31(1):514-521. Tingfei X, Jiazhen M, Wenhua T, Xuehui L, Shuhui L, and Baoshu X. Journal of Biomedical Materials Research 1992;26(9):1241-1251. Schmidt CE and Baier JM. Biomaterials 2000;21(22):2215-2231. Guidetti GP, Rigosi GL, and Marzola R. Progress in Organic Coatings 1996;27(1–4):7985. Parker PH, Treadway BR, and Brandt H. Journal of Applied Polymer Science 1967;11(9):1667-1682.

AC C

1.

ACCEPTED MANUSCRIPT

30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45.

22

RI PT

29.

SC

27. 28.

M AN U

26.

TE D

25.

EP

24.

Cestari AR, Vieira EFS, Silva ECS, Alves FJ, and Andrade Jr MAS. Journal of Colloid and Interface Science 2013;392(0):359-368. Mailhot B, Morlat-Thérias S, Ouahioune M, and Gardette J-L. Macromolecular Chemistry and Physics 2005;206(5):575-584. Buchwalter SL and Kosbar LL. Journal of Polymer Science Part A: Polymer Chemistry 1996;34(2):249-260. Hashimoto T, Meiji H, Urushisaki M, Sakaguchi T, Kawabe K, Tsuchida C, and Kondo K. Journal of Polymer Science Part A: Polymer Chemistry 2012;50(17):3674-3681. Bai N, Saito K, and Simon GP. Polymer Chemistry 2013;4(3):724-730. Luo K, Xie T, and Rzayev J. Journal of Polymer Science Part A: Polymer Chemistry 2013;51(23):4992-4997. Marref M, Mignard N, Jegat C, Taha M, Belbachir M, and Meghabar R. Polymer International 2013;62(1):87-98. Tian Q, Yuan YC, Rong MZ, and Zhang MQ. Journal of Materials Chemistry 2009;19(9):1289-1296. Zhao L, Liu Y, Wang Z, Li J, Liu W, and Chen Z. Polymer Degradation and Stability 2013;98(11):2125-2130. Wang L and Wong CP. Journal of Polymer Science Part A: Polymer Chemistry 1999;37(15):2991-3001. Wang L, Li H, and Wong CP. Journal of Polymer Science Part A: Polymer Chemistry 2000;38(20):3771-3782. Okamura H, Toda S, Tsunooka M, and Shirai M. Journal of Polymer Science Part A: Polymer Chemistry 2002;40(17):3055-3062. Liu W, Wang Z, Xiong L, and Zhao L. Polymer 2010;51(21):4776-4783. Liu W, Wang Z, Chen Z, and Zhao L. Polymer Degradation and Stability 2012;97(5):810-815. Singh R and Whitesides GM. Thiol—disulfide interchange. Sulphur-Containing Functional Groups (1993): John Wiley & Sons, Inc., 2010. pp. 633-658. Engelberg PI and Tesoro GC. Polymer Engineering & Science 1990;30(5):303-307. Sastri VR and Tesoro GC. Journal of Applied Polymer Science 1990;39(7):1439-1457. Tesoro GC and Sastri V. Journal of Applied Polymer Science 1990;39(7):1425-1437. Elliott JE, Nie J, and Bowman CN. Polymer 2003;44(2):327-332. Menczel JD, Prime RB, and Gallagher PK. Thermal Analysis of Polymers: John Wiley & Sons, Inc., 2008. pp. 7-317. Nogueira P, Ramírez C, Torres A, Abad MJ, Cano J, López J, López-Bueno I, and Barral L. Journal of Applied Polymer Science 2001;80(1):71-80. Chateauminois A, Chabert B, Soulier JP, and Vincent L. Polymer Composites 1995;16(4):288-296. Miller-Chou BA and Koenig JL. Progress in Polymer Science 2003;28(8):1223-1270.

AC C

23.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

Supplemental 6. Videos showing samples comprising BFDGE and an off-stoichiometric amount of 100% 4-APDS (left) and 100% PACM (right). The samples (approximate dimensions of 35 mm × 13 mm × 2 mm) were submerged in 100% 2-ME at 80˚C. Images were acquired over the course of approximately three hours. To shorten the duration of film, samples were preconditioned by immersion in 100% 2-ME for approximately 4 hours at 100˚C prior to acquiring the video.

ACCEPTED MANUSCRIPT

Highlights for Manuscript Number: POLYMER-15-139 Controlled Degradation of Disulfide-Based Epoxy Thermosets for Extreme Environments Leah M. Johnson, Eric Ledet, Nicolas D. Huffman, Stephanie L. Swarner, Sarah D. Shepherd, Phillip G. Durham, Ginger D. Rothrock

RI PT

SC

M AN U

•

TE D

•

EP

•

Disulfide-based thermosets are prepared using Bisphenol-F diglycidyl ether (BFDGE) in combination with blends of p,p’-diaminodicyclohexylmethane (PACM) and 4aminophenyl disulfide (4APDS). Polymers withstand harsh conditions (e.g., two weeks, 10,000 psi, 100˚C, pH 12) while simultaneously retaining the ability to subsequently degrade from 2-mercaptoethanol. Parameters affecting polymer degradation are detailed, including temperature, concentration of disulfide groups, and monomer stoichiometry. Formulations are described that completely dissolve in 2-mercaptoethanol and are utilized for release of prototype tracers from macro-scale particles.

AC C

•

ACCEPTED MANUSCRIPT

Supplemental Information Controlled Degradation of Disulfide-Based Epoxy Thermosets for Extreme Environments

RI PT

Leah M. Johnson, Eric Ledet, Nicolas D. Huffman, Stephanie L. Swarner, Sarah D. Shepherd, Phillip G. Durham, Ginger D. Rothrock

TE D

M AN U

SC

Emerging Technologies, Research Triangle Institute (RTI) International Research Triangle Park, NC, 27709, USA. Email: [email protected]

AC C

EP

Supplemental 1. Temperature dependent heat flow profiles during the cure of different epoxy formulations. Each curve shows a polymer formulation containing the BFDGE monomer combined with stoichiometric quantity of curative agent: (i) 100:0, (ii) 75:25, (iii) 50:50, (iv) 25:75, (v) 0:100 of PACM:4APDS and (vi) negative control diphenyl disulfide (DPDS) (Aldrich).

ACCEPTED MANUSCRIPT

449.7

0.8

75:25

352.8

0.3

50:50

344.6

0.002

25:75

378.0

0.3

0:100

417.6

0.1

SC

100:0

RI PT

Table S2. DSC analysis of polymer formulations Heat of Residual Heat PACM: Reaction J/g of Reaction 4APDS (∆Hrxn) J/g (∆Hres)

AC C

EP

TE D

M AN U

Supplemental 2. The heat of reaction (∆Hrxn) of the cure was determined by integrating the area under the DSC curve for each formulation and the residual cure (∆Hres) was determined by performing dynamic DSC on cured polymer formulations. The polymer formulations contain the BFDGE monomer combined with a stoichiometric quantity of amine monomer or amine monomer blend.

Supplemental 3. TGA profiles of five different cured epoxy polymer formulations. Each curve shows a formulation containing the BFDGE epoxy monomer cured with a stoichiometric quantity of curative agent blend: (i) 100:0, (ii) 75:25, (iii) 50:50, (iv) 25:75, and (v) 0:100 of PACM:4APDS.

SC

RI PT

ACCEPTED MANUSCRIPT

M AN U

Supplemental 4. Scheme showing thiol-disulfide exchange occurring in the disulfide-based epoxy thermosets. This scheme suggests that the polymer network loosens after the exchange reactions.

TE D

Table S5. Mechanical properties after exposure to 100% 2-ME for one month at ambient temperature PACM: 4APDS

Storage Modulus (MPa) at 35˚C

Tg (˚C)

100:0

2387 (±31)

138.5 (±0.5)

2277 (±28)

135.6 (±0.2)

50:50

1886 (±30)

137.5 (±0.1)

25:75

2043 (±47)

141.3 (±0.4)

0:100

2371 (±102)

143.1 (±0.5)

AC C

EP

75:25

Supplemental 5. Table of mechanical properties of DMA bars after one month immersion in 100% 2-ME at ambient temperature and pressure

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Supplemental 6. Videos showing samples comprising BFDGE and an off-stoichiometric amount of 100% 4-APDS (left) and 100% PACM (right). The samples (approximate dimensions of 35 mm × 13 mm × 2 mm) were submerged in 100% 2-ME at 80˚C. Images were acquired over the course of approximately three hours. To shorten the duration of film, samples were pre-conditioned by immersion in 100% 2-ME for approximately 4 hours at 100˚C prior to acquiring the video.

AC C

Supplemental 7. Digital camera images showing the degradation profiles of samples comprising BFDGE with different amine curatives (i.e., 4APDS or PACM) in 2-ME at 80°C. After removal from HPHT conditions (100°C, 69MPa, 1 week), the samples were subsequently incubated in 100% 2-ME at 80°C for a total of 10 hours. These images demonstrate that disulfide-based epoxy polymers retain the capacity to undergo controlled degradation resulting from thiol compounds even after exposure to harsh HPHT conditions.