- Email: [email protected]
Supercritical CO2 mediated functionalization of highly porous emulsion-derived foams: ScCO2 absorption and epoxidation
Journal of CO₂ Utilization 21 (2017) 336–341
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
Supercritical CO2 mediated functionalization of highly porous emulsionderived foams: ScCO2 absorption and epoxidation Nina Trupeja, Zoran Novaka, Željko Kneza, Christian Slugovcb, Sebastijan Kovačiča,c, a b c
MARK
⁎
University of Maribor, Faculty of Chemistry and Chemical Engineering, Smetanova 17, SI-2000 Maribor, Slovenia Graz University of Technology, Institute for Chemistry and Technology of Materials, NAWI Graz, Stremayrgasse 9, A-8010 Graz, Austria National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia
A R T I C L E I N F O
A B S T R A C T
Keywords: Supercritical CO2 Polymer-supported reagents High internal phase emulsions (HIPEs) Prileschajew epoxidation β-Amino alcohols
Highly macroporous polymers based on dicyclopentadiene (DCPD) have been successfully functionalized via a supercritical CO2 method aimed at minimizing the usage of organic solvents typically necessary for functionalization of polymers. The macroporous morphology of DCPD-based polyHIPEs resists the high pressures and the material exhibits an unusually high uptake of scCO2 at zero swelling. The scCO2 treatment creates additional mesoporosity, facilitating for the functionalisation within the void walls. The use of scCO2 as a solvent in the Prileschajew epoxidation and consecutive epoxide aminolysis of such epoxidized polydicyclopentadiene (pDCPD) monoliths is described. The procedure yields a high degree of functionalization with up to 6 mmol of βamino alcohol derivatives per g of polymer.
1. Introduction Heterogeneous-phase organic synthesis can combine the benefits of solid-phase chemistry with the advantages of solution-phase synthesis [1]. The essential advantage of heterogeneous-phase synthesis is that in just few reaction steps, libraries of molecularly diverse resins can be functionalized (synthesized), wherein excess reagents can be easily removed by extraction and filtration, and purification achieved, by washing [2]. Since heterogeneous-phase synthesis occurs at the interface boundary between the polymeric support and the liquid-phase, it requires longer reaction times to achieve comparable yields as classical solution phase synthesis [3]. Over the last two decades, polymeric supports for heterogeneous-phase synthesis evolved tremendously as morphology dictates the reaction kinetics due to the mass-transfer and diffusion limitations [4,5]. With increasing morphological complexity and decreasing feature sizes of porous morphologies down to the nanoscale, heterogeneous-phase synthesis was found destructive to the polymeric supports, as organic solvents damage the functional surfaces upon cycling due to the liquid’s viscosity and surface tension. Among the morphologies that have been used in heterogeneousphase synthesis, a 3D-interconnected porous morphology of polyHIPEbased resins derived from high internal phase emulsions (HIPEs), makes this polymeric supports highly usable since it provides unobstructed diffusivity for reagents to access resin’s reactive sites (functional
groups) [6]. Major advantage of polyHIPEs (polymerized high internal phase emulsions) [7] is their high porosity (between 74–95 vol.%) and a unique void and window structure that makes them further usable in low-pressure, continuous flow set-ups [8]. Gas-like diffusivity and liquid-like density of scCO2 on the other hand, makes it an excellent choice as reaction medium. High diffusivity enhances mass transfer and reaction kinetics while low viscosity and surface tension of scCO2 allow complete wetting of the polymeric supports without any damage of the surfaces [9]. The use of scCO2 as the solvent in organic and polymer chemistry has been comprehensively explored during the last 30 years, since it can dissolve many organic compounds and sustain inherent environmental, health, and safety (EHS) [10] advantages due to its environmental benignness. Low viscosity of supercritical fluids (0.01–0.003 mPa) as opposed to liquids (0.2–0.3 mPa s), make them as an attractive solvents for chemical synthesis [11]. ScCO2 has been studied as an alternative solvent in polymer synthesis, [12,13] porous polymeric materials production, [14] polymer and bio-polymer processing [15,16], polymer particle synthesis [17], impregnation and dyeing of polymeric materials [18], polymer extraction and purification [19], and heterogeneous chemical modification of polymers [20]. However, the number of examples where scCO2 is used as a solvent in the heterogeneous-phase organic synthesis is as yet fairly small. Combination of polyHIPE-based resins and compressed (supercritical) CO2 (scCO2) as it will be disclosed herein, is a promising
⁎ Corresponding author at: University of Maribor, Faculty of Chemistry and Chemical Engineering, Smetanova 17, SI-2000 Maribor, Slovenia. National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia. E-mail address: [email protected] (S. Kovačič).
http://dx.doi.org/10.1016/j.jcou.2017.07.024 Received 28 April 2017; Received in revised form 25 July 2017; Accepted 27 July 2017 Available online 15 September 2017 2212-9820/ © 2017 Elsevier Ltd. All rights reserved.
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
provided the inlet of CO2 into the autoclave into the middle of the autoclave, so the inlet was as close as possible to the polymer. The outlet nozzles are the macromatic and micromatic nozzles before the capturing flask. The cell is provided with two sapphire windows for visual observation of the interior. The polymer was cut in a shape of a cube with side length of 0.5 cm and put in a sample container and placed in the high pressure measuring cell. The system was heated to the temperature of 40 °C. Firstly, the mass of the polymer in vacuum was measured until equilibrium was reached. This represents the mass of the polymer without absorbed gas (m0). Then the pressure was increased (either to 100 or 200 bar) and the measurements were continued at these conditions till equilibrium was established, which represents the mass of the polymer with absorbed gas (m1). The mass (Δm) difference is the amount of gas that was absorbed in the polymer under pressure and temperature. The measurements were corrected due to the buoyancy effect.
solution to avoid aggressive and hazardous organic solvents in the heterogeneous-phase organic synthesis and to improve the sustainability of the process. In this article we describe a straightforward, environmentally friendly supercritical CO2 technology for the postfunctionalisation of polymeric foams through the Prileschajew epoxidation of pDCPD followed by the aminolysis towards β-amino alcohol derivatives. PDCPD is known as an industrially important cross-linked polymer with high E-modulus, impact strength, high ductility, chemical resistance and high operating temperature. DCPD-based polyHIPEs exhibit very favourable mechanical properties such as high modulus of 330 MPa and sample could tolerate a stress of 12 MPa before breaking [21]. After being oxidized, E-modulus of DCPD-based polyHIPEs significantly increased to extraordinarily high 770 MPa. As a consequence of the metathesis polymerisation, DCPD-based polyHIPEs bears a high degree of unsaturation offering ways for further functionalisation [22]. Accordingly, pDCPD foams ideally fulfil the prerequisites for the preparation of highly functionalised porous materials.
2.4. Prilezhaev epoxidation and aminolysis
2. Experimental
After the polymer was synthesized it was kept in acetone at atemperature below 0 °C. Before the polymer was used for functionalization a surface layer of 2 mm was removed from the polymer (this layer might have already oxidized). Then the polymer was dried at ambient temperature with an evaporator under pressure of 250 mbar for 24 h. Afterwards, 1 g of pDCPD (15.13 mmol of double bonds) was suspended in scCO2 at 40 °C and 100 bar for 24 h in the autoclave of the experimental setup shown in Fig. 1. The mCPBA was dissolved in DCM (concentration of 2 g/mL). The syringe hand pump with capacity of 12 mL was filled with the mCPBA solution and 12 mL or 24 mL of the solution was slowly injected into the autoclave. The solution was injected under pressure (the syringe pump was connected to the autoclave) and the pressure increased from 100 to 120 bar after 24 mL was injected. After 24 h, the autoclave was set at 100 bar and the polymer was cleaned with constant flow of carbon dioxide at 40 °C. Approximately 0.7 kg of CO2 (at flow rate 1 dm3/min) were used for
2.1. Materials The monomer was dicyclopentadiene (DCPD, Aldrich). The surfactant was a commercial triblock copolymer (poly(ethylene glycol)-blockpoly(propylene glycol)-block-poly(ethylene glycol), Pluronic L-121 with a molecular weight of 4400 g/mol, Sigma-Aldrich). The initiator was (H2IMes)(PCy3)Cl2Ru(3-phenyl-indenylid-1-ene) (M2, Umicore’s 2nd generation initiator). The internal phase was deionized water (Milli-Q ultra-pure water). The solvents used included toluene (p.a. Aldrich) and acetone (Sigma–Aldrich) and CO2 (Messer; purity 99.5%). The functionalization of pDCPD was done using 3-chloroperbenzoic acid (mCPBA, Sigma–Aldrich) and ethylenediamine (Sigma–Aldrich). All the materials were used as received. 2.2. Synthesis of DCPD based polymers 2.2.1. DCPD-based polyHIPEs Dicyclopentadien (8 g) and surfactants (Pluronic®L-121; 0.62 g) were placed in a three necked 250 mL flask and the mixture was stirred with an overhead stirrer at 400 rpm. The corresponding amount of deionized water (65 mL) was added drop wise under constant stirring. After addition of water, the water-in-oil high internal phase emulsion was further stirred for 1 h until a uniform emulsion was produced. The initiator M2 (8.13 mg) dissolved in toluene (0.6 mL), was added to the emulsion and the mixture was stirred for further 1 min. Subsequently, the emulsion was transferred into the mold (polystyrene container) and was cured at 80 °C for 4 h. Resulting polymers were purified via Soxhlet extraction with acetone and dried under vacuum until constant weight was obtained. The polyHIPE organogel composition was described using the sample name pDCPD_90, where 90 reflect the HIPE internal phase content (the polyHIPE porosity). 2.2.2. DCPD-based bulk organogels The referenced bulk organogels were polymerized using the same amounts as described above for polyHIPEs, but without the Pluronic®L121 and without the water. The reference bulk organogels are described using the sample name pDCPD_0. 2.3. Supercritical CO2 uptake CO2 uptake in the polymer was determined gravimetrically with a magnetic suspension balance with accuracy of 20 μg in high-pressure cell (NWA GMBh, Lorrach, Germany). A detailed description can be found in the literature [23]. Briefly, the cell is made of stainless steel (AISI 316). The apparatus can be used at pressures up to 300 MPa and temperature of 250 °C and has a volume of 60 cm3. The inlet nozzle
Fig. 1. Schematic representation of the equipment used for the functionalization under supercritical conditions. The set-up consist of syringe hand pump (1), valve (2), high pressure pump (3), CO2 supply (4), pressure (5) and temperature indicator (6), autoclave (7), magnetic stirrer (8), heating system (9), micrometric valve (10), capturing flask (11), gas flow meter (12).
337
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
Fig. 2. Preparation of porous pDCPD monoliths and their subsequent functionalization via epoxidation and aminolysis in scCO2 (circled P means polymer chain; please note that only idealized chemical structures of the functionalized products are shown).
purification of the polymer-foam. β-Amino alcohol was formed through the aminolysis of the epoxidized pDCPD polyHIPEs with ethylenediamine. The syringe hand pump was filled with a solution of ethylenediamine in DCM (concentration 0.1 g/mL) and 12 mL was injected under pressure in the autoclave. The pressure increased for approximately 30 bar. After 24 h the polymer was again purified with a constant flow of CO2 at conditions 40 °C and 100 bar (the amount of used CO2 was approximately 0.7 kg).
3. Results and discussion 3.1. Preparation, swelling and supercritical CO2 absorption Microcellular dicyclopentadien (DCPD)-based polyHIPEs were synthesized as depicted in Fig. 2. A mixture of DCPD (10 vol% according to water) as a monomer and surfactant Pluronic L- 121 (7 vol% according to monomer) as a stabilizer were used as organic (continuous) phase, while the internal (droplet) phase was deionized water (90 vol%) [24]. This formulation was subsequently polymerized via Ring-opening Metathesis Polymerization (ROMP) triggered by the addition of the initiator M2. The densities (ρPH) and polymerization yields (YPH) of asobtained pDCPD_90 were calculated immediately after purification/ drying in order to minimize the effect of oxidation. YPH was found to be 79% and ρPH 0.14 g/cm3. The porous microstructure of pDCPD_90 is depicted in Fig. 3A and B and resembles the typical polyHIPE morphology since it consists of voids with diameter of 9 μm and the walls of these voids further contains the interconnecting pores of 3 μm in diameter. The total porosity of pDCPD_90 (PPH-T) and the reference bulk organogel pDCPD_0 (PbOGT) were calculated from the polyHIPEs’ or bulk organogels densities and densities of the fully dense polymers. Densities of the corresponding reference bulk organogels (ρbOG) and polymers (ρP) were found 1.05 and 1.16 g/cm3, respectively. The PPH-T and PbOGT were found to be 88 and 9%, respectively. The PPH-T of pDCPD_90 was further divided into two subtypes, i.e. the porosity from the pDCPD_90 voids (PPH-V) and the porosity within the pDCPD_90 void walls (PPH-OG) [25]. The PPH-V and PPH-OG were determined to be 87% and 1%, respectively, indicating predominant contribution of the macroporous voids to the overall porosity and the void walls to be almost nonporous. Generally, the polyHIPE foams tend to have rather low BET surface area (SBET) due to the low amount of micro-/mesopores within the void walls. Therefore, the specific surface area (SBET) was determined for the pDCPD_90. pDCPD_90 shows typical type IV isotherm with a steep increase of N2 sorption uptake at p/po ≈ 1 (Fig. 3C). This indicates mainly the presence of macropores with no accessible microporosity (PPH-OG of only 1%) and is reflected in the relatively low SBET of 5 m2/g. Afterwards, the polymer was exposed to the scCO2 and treated at constant flow of CO2 for 24 h in order to test the impact of scCO2 on the polyHIPEs porosity. The specific surface area somewhat raised to 7 m2/g showing still type IV isotherm but with hysteresis loop observed in the pressure range of 0.5–1.0 p/p0. Hence, the mesopore
2.5. Characterization The polymerization yields (YPH) defined as percent conversion were calculated using a mass balance (using Eq. (1)) following Soxhlet extraction that assumes that the Pluronic L-121 was removed. YPH = mP/mM
(1)
where mP is the mass of the polyHIPE after Soxhlet extraction and drying and mM is the mass of the monomer. The equilibrium volume-swelling ratio was measured using Eq. (2) Si = VD/VSW
(2)
Where i is “B” for the bulk organogel and “PH” for the polyHIPE, VD is the initial, dry volume and VSW is the final, swollen volume. The polyHIPE’s (ρPH) and bulk organogels (ρbOG) densities were determined gravimetrically, while polymers (ρP) density was determined by using helium pycnometry. FT-IR spectra were recorded on a Shimadzu IR Affinity-1 spectrometer (KBr pellets). CHNS elemental analyses were done on a Perkin Elmer CHN 2400 analyzer. The surface area and pore size distribution were calculated from nitrogen adsorption isotherms (Micromeritics TriStar II 3020) using the Brunauer, Emmett, and Teller (BET), and Barrett, Joyner and Halenda (BJH) methods. Morphology investigations were done by scanning electron microscopy (SEMs were taken on a JWS-7515, JEOL Ltd. microscope). Micrographs were taken at several magnifications from 2500 to 7000 fold, at 7 mm working distance and at 20 kV acceleration voltage. The samples were broken and mounted on a carbon tab and a thin layer of gold was sputtered onto the samples.
338
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
volume was assessed by the BJH method and the results are shown in Fig. 3(D). The BJH curves of nitrogen desorption show the cumulative mesopore volume to increases for almost of factor of 3 in the case of pDCPD_90 treated with the scCO2, with dominate mesopores of 2.6 nm (8 nm in case of not treated pDCPD_90). Whereas the treatment with the scCO2 was responsible for releasing additional mesopore volume that was unavailable before, the equilibrium volume swelling ratios and absorption in scCO2 of pDCPD_0 and pDCPD_90 cube shaped samples were further investigated. Fig. 4 shows the volume change of the samples treated at 40 °C and different pressures, i.e. at 100 and at 200 bar. The overall volume change, hence equilibrium volume swelling ratios for pDCPD_0 (SB) and pDCPD_90 (SPH), were found insignificant although the absorption of scCO2 in pDCPD is clearly visible as depicted in Fig. 4B and C. The freshly prepared pDCPD_0 and pDCPD_90 cubes of the same size as used in the swelling tests were further tested in the absorption experiments. The samples were placed in a cage within the high-pressure balance. Constant temperature and pressure were applied until the absorption equilibrium was achieved and the total (equilibrium) absorption for pDCPD_90 (APH) and pDCPD_0 (AB) determined. As expected, the APH was substantially higher with up to 8 g/g as AB with up to 0.40 g/g (Table 1). While the neat scCO2 has comparable dissolving properties as the hexane [26], the SB, SPH, AB and APH values in hexane were further determined for the pDCPD_0 and pDCPD_90. SB and SPH were found 1.3 and 2 V/V0, which is somewhat higher than those of the equivalent equilibrium swelling ratios in scCO2 but the APH and AB of 8.6 and 0.30 g/g, respectively, display the same trend as found in scCO2. The amount of the scCO2 absorbed by the pDCPD_90 was unexpected, as the V/V0 in scCO2 showed no swelling. It seems that low viscosity and high diffusivity impart high penetrating capability of scCO2 within the thin walls of pDCPD_90, which might be a reason that pDCPD_90 absorb scCO2 even at zero swelling through the void walls. 3.2. Prileschajew epoxidation and aminolysis in scCO2 m-Chloroperoxybenzoic acid (mCPBA) is a widely used reagent for generating organic epoxides due to its versatile oxidizing power [27]. The OeO bond of mCPBA transfers an oxygen atom to electron-rich substrates (e.g. epoxidation of olefins), while the nucleophilic attack of mCPBA on ketones and aldehydes results in the insertion of an oxygen atom (e.g. Baeyer-Villiger reaction) [28]. Despite many oxidation procedures, the use of peroxy acids still constitutes one of the most useful synthetic procedures for the epoxidation of alkenes [29]. This synthetic procedure, known as the Prileschajew epoxidation [30], was also used for converting the double bonds of ROMP derived polymers into the corresponding epoxide [31]. To prepare epoxidized polymer network (Fig. 1), functionalization of pDCPD_90 with mCPBA was attempted at a range of super critical CO2 conditions (temperatures and pressures), such that the density exceeds the critical density (0.47 g/cm3) and approach those of the conventional liquid solvents (0.6–1.0 g/cm3). It was found that at lower pressures (< 100 bar), the mCPBA was not soluble enough to be effectively transported into the polymer, whereas at higher pressures (100 and 200 bar), the mCPBA was sufficiently soluble in scCO2 and effective functionalization was achieved. Although scCO2 has a zero dipole moment and a very low polarizability, it has a strong quadrupolar moment that allows significant interactions with mCPBA to dissolve it [26]. The conversion of double bonds at different conditions is reported in Table 2. The high density of double bonds along the backbone makes pDCPD very prone to oxidation when exposed to water or air [32,33]. Hence, elemental analysis (EA) was done immediately after purification/drying wherein 4.1% (2.5 mmol/g) of oxygen was found. Fully oxidized pDCPD has an oxygen content of about 36% (22.5 mmol/g) [32]. EA further revealed 28% (17.5 mmol/ g) of oxygen after epoxidation of pDCPD_90 with 1 equiv of mCPBA (hereafter referred to as pDCPD_epoxE) and 17% (11 mmol/g) of oxygen after epoxidation of pDCPD_90 with 0.5 equiv of mCPBA
Fig. 3. Scanning electron micrographs of pDCPD_90 after treatment at 100 bar (A) and 200 bar (B), N2 sorption isotherms measured at 77 K (C) and mesopore volume by the BJH model (D).
339
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
Fig. 4. A–C) pDCPD_0 in a cage within the high-pressure balance during the swelling tests at 40 °C and A) 1 bar, B) 100 bar and C) 200 bar. D–F) pDCPD_90 in a cage within the highpressure balance during the swelling tests at 40 °C and D) 1 bar, E) 100 bar and F) 200 bar. pDCPD_90 were weighted down in order not to float.
Table 1 The equilibrium absorption values. T [°C]
p [bar]
ρ(CO2) [g/cm3]
AB [g/g]
APH [g/g]
40
100 200 100 200
0.657 0.844 0.188 0.489
0.19 0.44 0.01 n.d.
4.33 8.11 1.70 4.34
100
Table 2 Elemental analysis data and degrees of conversion. Samplea
Oxygen [mmol/ g]b
Oxygen [mmol/ g]c
Oxygen [mmol/ g]d
N [%]e
(CH2NH2)2 [mmol/g]f
[%]g
pDCPD_epoxE
15.13 7.56
17.5 11
15 8.5
12 16
4.25 5.85
28 77
pDCPD_epoxH
Fig. 5. FTIR spectra of pDCPD unoxidized (the top), pDCPD oxidized (middle) and epoxidized pDCPD (bottom).
a
epoxE denotes 1 equivalent of mCPBA used to functionalize pDCPD, epoxH denotes 0.5 equivalent of mCPBA used to functionalize pDCPD. b Calculated. c Determined as a balance between 100% and wt% of CHNS from elemental analyses. d Corrected amount taking into account oxygen before epoxidation. e Found from elemental analyses. f Loading of reactive groups. g The degree of conversion calculated from the ratio of the loading of amine nucleophile to corrected amount of oxygen in the polyHIPE material. It was assumed that no additional crosslinking occurred.
stretch vibrations at 1258, 901, and 840 cm−1 respectively. The Csp2eH vibrations at 3050 cm−1 almost completely disappeared, which proves that the mCPBA reacted with most of the olefins in pDCPD (Fig. 5) [34]. The formation of cyclic carbonates upon the reaction of epoxide groups with CO2 under supercritical conditions cannot be ruled out [35]. In a next step the preparation of the β-amino alcohol derivatives from pDCPD_epox polyHIPEs (Fig. 2) was attempted. One of the most practical and widely used routes for the synthesis of these compounds is the direct aminolysis of epoxides at elevated temperature with an excess of amine [36]. To test the reactivity of epoxide ring in scCO2, ethylenediamine was chosen as the nucleophile. The percent of conversions are reported in Table 2. 4.25 or 5.85 mmol/g of amine groups was found after ethylenediamine functionalization that corresponds to 28 or 77% of conversion for pDCPD_epoxE or pDCPD_epoxH polyHIPEs, respectively. Somewhat better conversion in the case of pDCPD_epoxH can be ascribed to the lower yields of the overoxidized products after epoxidation, which results in higher amount of the epoxide rings that reacts readily with the
(hereafter referred to as pDCPD_epoxH). By taking into account the 2.5 mmol/g of the oxygen that is present prior the epoxidation, corrected oxygen amount after epoxidation would be 15 mmol/g or 8.5 mmol/g for pDCPD_epoxE or pDCPD_epoxH, respectively. Whereas 15.13 mmol/g or 7.56 mmol/g of oxygen for pDCPD_epoxE or pDCPD_epoxH, respectively, is expected for the complete conversion, slightly higher amounts of oxygen determined by the EA pointed to the presence of overoxidized products (due to the action of oxygen). The IR spectrum of pDCPD_epox showed the characteristic carbonyl stretch vibration at 1715 cm−1 confirming overoxidation. More importantly, IR spectra of pDCPD_epox displayed the characteristic epoxide ring 340
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
early during process development, Ind. Eng. Chem. Res. 39 (2000) 960–972. [11] P. Munshi, S. Bhaduri, Supercritical CO2: a twenty-first century solvent for the chemical industry, Curr. Sci. 97 (2009) 63–72. [12] J.L. Kendall, D.A. Canelas, J.L. Young, J.M. DeSimone, Polymerizations in supercritical carbon dioxide, Chem. Rev. 99 (1999) 543–563. [13] C. Boyère, C. Jérôme, A. Debuigne, Input of supercritical carbon dioxide to polymer synthesis: an overview, Eur. Polym. J. 61 (2014) 45–63. [14] R. Butler, C.M. Davies, A.I. Cooper, Recent developments in materials synthesis and processing using supercritical CO2, Adv. Mater. 13 (2001) 1111–1114. [15] A.I. Cooper, J.M. DeSimone, Polymer synthesis and characterization in liquid/supercritical carbon dioxide, Curr. Opin. Solid State Mater. Sci. 1 (1996) 761–768. [16] Ž. Knez, E. Markočič, Z. Novak, M. Knez Hrnčič, Processing polymeric biomaterials using supercritical CO2, Chem. Ing. Tech. 83 (2011) 1371–1380. [17] Ž. Knez, Particle formation using supercritical fluids – a short review, Chem. Ind. Chem. Eng. Q. 12 (2006) 141–146. [18] S.K. Liao, P.S. Chang, Literatures on dyeing technique of supercritical fluid carbon dioxide, Am. J. Anal. Chem. 3 (2012) 923–930. [19] E. Ramsey, Q. Sun, Z. Zhang, C. Zhang, W. Gou, Mini-review: green sustainable processes using supercritical fluid carbon dioxide, J. Environ. Sci. 21 (2009) 720–726. [20] L. Dua, J.Y. Kellya, G.W. Roberts, J.M. DeSimone, Fluoropolymer synthesis in supercritical carbon dioxide, J. Supercrit. Fluids 47 (2009) 447–457. [21] S. Kovačič, K. Jeřabek, P. Krajnc, C. Slugovc, Ring opening metathesis polymerisation of emulsion templated dicyclopentadiene giving open porous materials with excellent mechanical, Polym. Chem. 3 (2012) 325–328. [22] (unlabelleda) A.-C. Knall, S. Kovačič, M. Hollauf, D.P. Reishofer, R. Saf, C. Slugovc, Inverse electron demand Diels–Alder (iEDDA) functionalisation of macroporous poly(dicyclopentadiene) foams, Chem. Commun. 7 (2013) 325–7327; (unlabelledb) S. Kovačič, F. Preishuber-Pflügl, D. Pahovnik, E. Žagar, C. Slugovc, Covalent incorporation of the surfactant into high internal phase emulsion templated polymeric foams, Chem. Commun. 51 (2015) 7225–7228. [23] N. Trupej, M. Knez Hrnčič, M. Škerget, Ž. Knez, Solubility and binary diffusion coefficient of argon in polyethylene glycols of different molecular weights, J. Supercrit. Fluids 103 (2015) 10–17. [24] S. Kovačič, N.B. Matsko, K. Jerabek, P. Krajnc, C. Slugovc, On the mechanical properties of HIPE templated macroporous poly(dicyclopentadiene) prepared with low surfactant amounts, J. Mater. Chem. A 1 (2013) 487–490. [25] M. Ovadia, M.S. Silverstein, High porosity responsive hydrogel copolymers from emulsion templating, Polym. Int. 65 (2016) 280–289. [26] T. Delgado-Abad, J. Martinez-Ferrer, A. Caballero, A. Olmos, R. Mello, M. Elena Gonzalez-Nunez, P.J. Perez, G. Asensio, Supercritical carbon dioxide: a promoter of carbon–halogen bond heterolysis, Angew. Chem. Int. Ed. 52 (2013) 13298–13301. [27] S. Cha, J. Hwang, M.G. Choi, S.K. Chang, Dual signaling of m-chloroperbenzoic acid by desulfurization of thiocoumarin, Tetrahedron Lett. 51 (2010) 6663–6665. [28] S. Kamijo, S. Matsumura, M. Inoue, CCl3CN: a crucial promoter of mCPBA-mediated direct ether oxidation, Org. Lett. 12 (2010) 4195–4197. [29] J.O. Edwards, Peroxide Reaction Mechanisms, in: J.O. Edwards (Ed.), Interscience, New York, 1962p. 67. [30] N. Prileschajew, Oxydation ungesättigter Verbindungen mittels organischer Superoxyde, Ber. 42 (1909) 4811–4815. [31] T.J. Boyd, R.R. Schrock, Sulfonation and epoxidation of substituted polynorbornenes and construction of light-emitting devices, Macromolecules 32 (1999) 6608–6618. [32] S. Kovačič, P. Krajnc, C. Slugovc, Inherently reactive polyHIPE material from dicyclopentadiene, Chem. Commun. 46 (2010) 7504–7506. [33] S. Kovačič, F. Preishuber-Pflugl, C. Slugovc, Macroporous polyolefin membranes from dicyclopentadiene high internal, Macromol. Mater Eng. 299 (2014) 843–850. [34] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identification of Organic Compounds, John Wiley & Sons, New York, 1991. [35] M. North, R. Pasquale, C. Young, Synthesis of cyclic carbonates from epoxides and CO2, Green Chem. 12 (2010) 1514–1539. [36] N. Azizi, M.R. Saidi, Highly chemoselective addition of amines to epoxides in water, Org. Lett. 7 (2005) 3649–3651.
amine nucleophile. 4. Conclusions Supercritical CO2 has been shown to provide a simple and effective processing route to generate functional macroporous polymers. Heterogeneous-phase organic synthesis using scCO2 as a solvent is a promising way as products are simply isolated by depressurization, which results immediately in dry, final compounds eliminating postreaction steps like filtration and drying indispensable when liquid organic solvents are used. The absorption of scCO2 in pDCPD polymers was found to be higher at lower temperatures and increased with increasing pressure at zero swelling, while maintaining the characteristic open-porous morphology of emulsion-templated pDCPD foams. Functionalization of pDCPD polyHIPEs in scCO2 was performed through the Prileschajew epoxidation using mCPBA. Exploiting reactivity of epoxides, further functionalization with the amine nucleophile was obtained. A high loading of up to 5.85 mmol/g of β-amino alcohols (11.4 mmol/g of nitrogen) was obtained. Acknowledgements This work was supported via the grant BI-AT/16-17-023. The authors gratefully acknowledge the financial support of the Ministry of Higher Education, Science and Technology of the Republic of Slovenia, and the Slovenian Research Agency (Program No. P2-0145 and P2–0046). References [1] L. Kircher, P. Theato, N. R.Cameron, Functionalization of porous polymers from high-internal-phase emulsions and their applications, in: P. Theato, H.A. Klok (Eds.), Functional Polymers by Post-polymerization Modification: Concepts, Practical Guidelines and Applications. Wiley-VCH, Weinheim, 2012. [2] A. Kirschning, H. Monenschein, R. Wittenberg, Functionalized polymers emerging versatile tools for solution-phase chemistry and automated parallel synthesis, Angew. Chem. Int. Ed. 40 (2001) 650–679. [3] M.V. Fawaz, P.J.H. Scott, Solid-phase organic synthesis, Kirk-Othmer Encyclopedia of Chemical Technology, (2014), pp. 1–27. [4] P. Hodge, D.C. Sherrington, Polymer-Supported Reactions in Organic Synthesis, Wiley, Chichester, UK, 1980. [5] Polymeric Materials in Organic Synthesis and Catalysis, in: M.R. Buchmeiser (Ed.), Wiley-VCH, Weinheim, 2003. [6] D.C. Sherrington, Polymer-supported reagents, catalysts, and sorbents: evolution and exploitation—a personalized view, J. Polym. Sci. Part A 39 (2001) 2364–2377. [7] P.W. Small, D.C. Sherrington, Design and application of a new rigid support for high efficiency continuous-flow peptide synthesis, J. Chem. Soc. Chem. Commun. 21 (1989) 1589–1591. [8] N.R. Cameron, P. Krajnc, M.S. Silverstein, Colloidal templating, in: M.S. Silverstein, N.R. Cameron, M.A. Hillmyer (Eds.), Porous Polymers, John Wiley & Sons, Inc, Hoboken, NJ, USA, 2011. [9] J.A. Darr, M. Poliakoff, New directions in inorganic and metal-organic coordination chemistry in supercritical fluids, Chem. Rev. 99 (1999) 495–542. [10] U. Fischer, K. Hungerbühler, Assessing safety, health, and environmental impact
341
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
Supercritical CO2 mediated functionalization of highly porous emulsionderived foams: ScCO2 absorption and epoxidation Nina Trupeja, Zoran Novaka, Željko Kneza, Christian Slugovcb, Sebastijan Kovačiča,c, a b c
MARK
⁎
University of Maribor, Faculty of Chemistry and Chemical Engineering, Smetanova 17, SI-2000 Maribor, Slovenia Graz University of Technology, Institute for Chemistry and Technology of Materials, NAWI Graz, Stremayrgasse 9, A-8010 Graz, Austria National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia
A R T I C L E I N F O
A B S T R A C T
Keywords: Supercritical CO2 Polymer-supported reagents High internal phase emulsions (HIPEs) Prileschajew epoxidation β-Amino alcohols
Highly macroporous polymers based on dicyclopentadiene (DCPD) have been successfully functionalized via a supercritical CO2 method aimed at minimizing the usage of organic solvents typically necessary for functionalization of polymers. The macroporous morphology of DCPD-based polyHIPEs resists the high pressures and the material exhibits an unusually high uptake of scCO2 at zero swelling. The scCO2 treatment creates additional mesoporosity, facilitating for the functionalisation within the void walls. The use of scCO2 as a solvent in the Prileschajew epoxidation and consecutive epoxide aminolysis of such epoxidized polydicyclopentadiene (pDCPD) monoliths is described. The procedure yields a high degree of functionalization with up to 6 mmol of βamino alcohol derivatives per g of polymer.
1. Introduction Heterogeneous-phase organic synthesis can combine the benefits of solid-phase chemistry with the advantages of solution-phase synthesis [1]. The essential advantage of heterogeneous-phase synthesis is that in just few reaction steps, libraries of molecularly diverse resins can be functionalized (synthesized), wherein excess reagents can be easily removed by extraction and filtration, and purification achieved, by washing [2]. Since heterogeneous-phase synthesis occurs at the interface boundary between the polymeric support and the liquid-phase, it requires longer reaction times to achieve comparable yields as classical solution phase synthesis [3]. Over the last two decades, polymeric supports for heterogeneous-phase synthesis evolved tremendously as morphology dictates the reaction kinetics due to the mass-transfer and diffusion limitations [4,5]. With increasing morphological complexity and decreasing feature sizes of porous morphologies down to the nanoscale, heterogeneous-phase synthesis was found destructive to the polymeric supports, as organic solvents damage the functional surfaces upon cycling due to the liquid’s viscosity and surface tension. Among the morphologies that have been used in heterogeneousphase synthesis, a 3D-interconnected porous morphology of polyHIPEbased resins derived from high internal phase emulsions (HIPEs), makes this polymeric supports highly usable since it provides unobstructed diffusivity for reagents to access resin’s reactive sites (functional
groups) [6]. Major advantage of polyHIPEs (polymerized high internal phase emulsions) [7] is their high porosity (between 74–95 vol.%) and a unique void and window structure that makes them further usable in low-pressure, continuous flow set-ups [8]. Gas-like diffusivity and liquid-like density of scCO2 on the other hand, makes it an excellent choice as reaction medium. High diffusivity enhances mass transfer and reaction kinetics while low viscosity and surface tension of scCO2 allow complete wetting of the polymeric supports without any damage of the surfaces [9]. The use of scCO2 as the solvent in organic and polymer chemistry has been comprehensively explored during the last 30 years, since it can dissolve many organic compounds and sustain inherent environmental, health, and safety (EHS) [10] advantages due to its environmental benignness. Low viscosity of supercritical fluids (0.01–0.003 mPa) as opposed to liquids (0.2–0.3 mPa s), make them as an attractive solvents for chemical synthesis [11]. ScCO2 has been studied as an alternative solvent in polymer synthesis, [12,13] porous polymeric materials production, [14] polymer and bio-polymer processing [15,16], polymer particle synthesis [17], impregnation and dyeing of polymeric materials [18], polymer extraction and purification [19], and heterogeneous chemical modification of polymers [20]. However, the number of examples where scCO2 is used as a solvent in the heterogeneous-phase organic synthesis is as yet fairly small. Combination of polyHIPE-based resins and compressed (supercritical) CO2 (scCO2) as it will be disclosed herein, is a promising
⁎ Corresponding author at: University of Maribor, Faculty of Chemistry and Chemical Engineering, Smetanova 17, SI-2000 Maribor, Slovenia. National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia. E-mail address: [email protected] (S. Kovačič).
http://dx.doi.org/10.1016/j.jcou.2017.07.024 Received 28 April 2017; Received in revised form 25 July 2017; Accepted 27 July 2017 Available online 15 September 2017 2212-9820/ © 2017 Elsevier Ltd. All rights reserved.
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
provided the inlet of CO2 into the autoclave into the middle of the autoclave, so the inlet was as close as possible to the polymer. The outlet nozzles are the macromatic and micromatic nozzles before the capturing flask. The cell is provided with two sapphire windows for visual observation of the interior. The polymer was cut in a shape of a cube with side length of 0.5 cm and put in a sample container and placed in the high pressure measuring cell. The system was heated to the temperature of 40 °C. Firstly, the mass of the polymer in vacuum was measured until equilibrium was reached. This represents the mass of the polymer without absorbed gas (m0). Then the pressure was increased (either to 100 or 200 bar) and the measurements were continued at these conditions till equilibrium was established, which represents the mass of the polymer with absorbed gas (m1). The mass (Δm) difference is the amount of gas that was absorbed in the polymer under pressure and temperature. The measurements were corrected due to the buoyancy effect.
solution to avoid aggressive and hazardous organic solvents in the heterogeneous-phase organic synthesis and to improve the sustainability of the process. In this article we describe a straightforward, environmentally friendly supercritical CO2 technology for the postfunctionalisation of polymeric foams through the Prileschajew epoxidation of pDCPD followed by the aminolysis towards β-amino alcohol derivatives. PDCPD is known as an industrially important cross-linked polymer with high E-modulus, impact strength, high ductility, chemical resistance and high operating temperature. DCPD-based polyHIPEs exhibit very favourable mechanical properties such as high modulus of 330 MPa and sample could tolerate a stress of 12 MPa before breaking [21]. After being oxidized, E-modulus of DCPD-based polyHIPEs significantly increased to extraordinarily high 770 MPa. As a consequence of the metathesis polymerisation, DCPD-based polyHIPEs bears a high degree of unsaturation offering ways for further functionalisation [22]. Accordingly, pDCPD foams ideally fulfil the prerequisites for the preparation of highly functionalised porous materials.
2.4. Prilezhaev epoxidation and aminolysis
2. Experimental
After the polymer was synthesized it was kept in acetone at atemperature below 0 °C. Before the polymer was used for functionalization a surface layer of 2 mm was removed from the polymer (this layer might have already oxidized). Then the polymer was dried at ambient temperature with an evaporator under pressure of 250 mbar for 24 h. Afterwards, 1 g of pDCPD (15.13 mmol of double bonds) was suspended in scCO2 at 40 °C and 100 bar for 24 h in the autoclave of the experimental setup shown in Fig. 1. The mCPBA was dissolved in DCM (concentration of 2 g/mL). The syringe hand pump with capacity of 12 mL was filled with the mCPBA solution and 12 mL or 24 mL of the solution was slowly injected into the autoclave. The solution was injected under pressure (the syringe pump was connected to the autoclave) and the pressure increased from 100 to 120 bar after 24 mL was injected. After 24 h, the autoclave was set at 100 bar and the polymer was cleaned with constant flow of carbon dioxide at 40 °C. Approximately 0.7 kg of CO2 (at flow rate 1 dm3/min) were used for
2.1. Materials The monomer was dicyclopentadiene (DCPD, Aldrich). The surfactant was a commercial triblock copolymer (poly(ethylene glycol)-blockpoly(propylene glycol)-block-poly(ethylene glycol), Pluronic L-121 with a molecular weight of 4400 g/mol, Sigma-Aldrich). The initiator was (H2IMes)(PCy3)Cl2Ru(3-phenyl-indenylid-1-ene) (M2, Umicore’s 2nd generation initiator). The internal phase was deionized water (Milli-Q ultra-pure water). The solvents used included toluene (p.a. Aldrich) and acetone (Sigma–Aldrich) and CO2 (Messer; purity 99.5%). The functionalization of pDCPD was done using 3-chloroperbenzoic acid (mCPBA, Sigma–Aldrich) and ethylenediamine (Sigma–Aldrich). All the materials were used as received. 2.2. Synthesis of DCPD based polymers 2.2.1. DCPD-based polyHIPEs Dicyclopentadien (8 g) and surfactants (Pluronic®L-121; 0.62 g) were placed in a three necked 250 mL flask and the mixture was stirred with an overhead stirrer at 400 rpm. The corresponding amount of deionized water (65 mL) was added drop wise under constant stirring. After addition of water, the water-in-oil high internal phase emulsion was further stirred for 1 h until a uniform emulsion was produced. The initiator M2 (8.13 mg) dissolved in toluene (0.6 mL), was added to the emulsion and the mixture was stirred for further 1 min. Subsequently, the emulsion was transferred into the mold (polystyrene container) and was cured at 80 °C for 4 h. Resulting polymers were purified via Soxhlet extraction with acetone and dried under vacuum until constant weight was obtained. The polyHIPE organogel composition was described using the sample name pDCPD_90, where 90 reflect the HIPE internal phase content (the polyHIPE porosity). 2.2.2. DCPD-based bulk organogels The referenced bulk organogels were polymerized using the same amounts as described above for polyHIPEs, but without the Pluronic®L121 and without the water. The reference bulk organogels are described using the sample name pDCPD_0. 2.3. Supercritical CO2 uptake CO2 uptake in the polymer was determined gravimetrically with a magnetic suspension balance with accuracy of 20 μg in high-pressure cell (NWA GMBh, Lorrach, Germany). A detailed description can be found in the literature [23]. Briefly, the cell is made of stainless steel (AISI 316). The apparatus can be used at pressures up to 300 MPa and temperature of 250 °C and has a volume of 60 cm3. The inlet nozzle
Fig. 1. Schematic representation of the equipment used for the functionalization under supercritical conditions. The set-up consist of syringe hand pump (1), valve (2), high pressure pump (3), CO2 supply (4), pressure (5) and temperature indicator (6), autoclave (7), magnetic stirrer (8), heating system (9), micrometric valve (10), capturing flask (11), gas flow meter (12).
337
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
Fig. 2. Preparation of porous pDCPD monoliths and their subsequent functionalization via epoxidation and aminolysis in scCO2 (circled P means polymer chain; please note that only idealized chemical structures of the functionalized products are shown).
purification of the polymer-foam. β-Amino alcohol was formed through the aminolysis of the epoxidized pDCPD polyHIPEs with ethylenediamine. The syringe hand pump was filled with a solution of ethylenediamine in DCM (concentration 0.1 g/mL) and 12 mL was injected under pressure in the autoclave. The pressure increased for approximately 30 bar. After 24 h the polymer was again purified with a constant flow of CO2 at conditions 40 °C and 100 bar (the amount of used CO2 was approximately 0.7 kg).
3. Results and discussion 3.1. Preparation, swelling and supercritical CO2 absorption Microcellular dicyclopentadien (DCPD)-based polyHIPEs were synthesized as depicted in Fig. 2. A mixture of DCPD (10 vol% according to water) as a monomer and surfactant Pluronic L- 121 (7 vol% according to monomer) as a stabilizer were used as organic (continuous) phase, while the internal (droplet) phase was deionized water (90 vol%) [24]. This formulation was subsequently polymerized via Ring-opening Metathesis Polymerization (ROMP) triggered by the addition of the initiator M2. The densities (ρPH) and polymerization yields (YPH) of asobtained pDCPD_90 were calculated immediately after purification/ drying in order to minimize the effect of oxidation. YPH was found to be 79% and ρPH 0.14 g/cm3. The porous microstructure of pDCPD_90 is depicted in Fig. 3A and B and resembles the typical polyHIPE morphology since it consists of voids with diameter of 9 μm and the walls of these voids further contains the interconnecting pores of 3 μm in diameter. The total porosity of pDCPD_90 (PPH-T) and the reference bulk organogel pDCPD_0 (PbOGT) were calculated from the polyHIPEs’ or bulk organogels densities and densities of the fully dense polymers. Densities of the corresponding reference bulk organogels (ρbOG) and polymers (ρP) were found 1.05 and 1.16 g/cm3, respectively. The PPH-T and PbOGT were found to be 88 and 9%, respectively. The PPH-T of pDCPD_90 was further divided into two subtypes, i.e. the porosity from the pDCPD_90 voids (PPH-V) and the porosity within the pDCPD_90 void walls (PPH-OG) [25]. The PPH-V and PPH-OG were determined to be 87% and 1%, respectively, indicating predominant contribution of the macroporous voids to the overall porosity and the void walls to be almost nonporous. Generally, the polyHIPE foams tend to have rather low BET surface area (SBET) due to the low amount of micro-/mesopores within the void walls. Therefore, the specific surface area (SBET) was determined for the pDCPD_90. pDCPD_90 shows typical type IV isotherm with a steep increase of N2 sorption uptake at p/po ≈ 1 (Fig. 3C). This indicates mainly the presence of macropores with no accessible microporosity (PPH-OG of only 1%) and is reflected in the relatively low SBET of 5 m2/g. Afterwards, the polymer was exposed to the scCO2 and treated at constant flow of CO2 for 24 h in order to test the impact of scCO2 on the polyHIPEs porosity. The specific surface area somewhat raised to 7 m2/g showing still type IV isotherm but with hysteresis loop observed in the pressure range of 0.5–1.0 p/p0. Hence, the mesopore
2.5. Characterization The polymerization yields (YPH) defined as percent conversion were calculated using a mass balance (using Eq. (1)) following Soxhlet extraction that assumes that the Pluronic L-121 was removed. YPH = mP/mM
(1)
where mP is the mass of the polyHIPE after Soxhlet extraction and drying and mM is the mass of the monomer. The equilibrium volume-swelling ratio was measured using Eq. (2) Si = VD/VSW
(2)
Where i is “B” for the bulk organogel and “PH” for the polyHIPE, VD is the initial, dry volume and VSW is the final, swollen volume. The polyHIPE’s (ρPH) and bulk organogels (ρbOG) densities were determined gravimetrically, while polymers (ρP) density was determined by using helium pycnometry. FT-IR spectra were recorded on a Shimadzu IR Affinity-1 spectrometer (KBr pellets). CHNS elemental analyses were done on a Perkin Elmer CHN 2400 analyzer. The surface area and pore size distribution were calculated from nitrogen adsorption isotherms (Micromeritics TriStar II 3020) using the Brunauer, Emmett, and Teller (BET), and Barrett, Joyner and Halenda (BJH) methods. Morphology investigations were done by scanning electron microscopy (SEMs were taken on a JWS-7515, JEOL Ltd. microscope). Micrographs were taken at several magnifications from 2500 to 7000 fold, at 7 mm working distance and at 20 kV acceleration voltage. The samples were broken and mounted on a carbon tab and a thin layer of gold was sputtered onto the samples.
338
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
volume was assessed by the BJH method and the results are shown in Fig. 3(D). The BJH curves of nitrogen desorption show the cumulative mesopore volume to increases for almost of factor of 3 in the case of pDCPD_90 treated with the scCO2, with dominate mesopores of 2.6 nm (8 nm in case of not treated pDCPD_90). Whereas the treatment with the scCO2 was responsible for releasing additional mesopore volume that was unavailable before, the equilibrium volume swelling ratios and absorption in scCO2 of pDCPD_0 and pDCPD_90 cube shaped samples were further investigated. Fig. 4 shows the volume change of the samples treated at 40 °C and different pressures, i.e. at 100 and at 200 bar. The overall volume change, hence equilibrium volume swelling ratios for pDCPD_0 (SB) and pDCPD_90 (SPH), were found insignificant although the absorption of scCO2 in pDCPD is clearly visible as depicted in Fig. 4B and C. The freshly prepared pDCPD_0 and pDCPD_90 cubes of the same size as used in the swelling tests were further tested in the absorption experiments. The samples were placed in a cage within the high-pressure balance. Constant temperature and pressure were applied until the absorption equilibrium was achieved and the total (equilibrium) absorption for pDCPD_90 (APH) and pDCPD_0 (AB) determined. As expected, the APH was substantially higher with up to 8 g/g as AB with up to 0.40 g/g (Table 1). While the neat scCO2 has comparable dissolving properties as the hexane [26], the SB, SPH, AB and APH values in hexane were further determined for the pDCPD_0 and pDCPD_90. SB and SPH were found 1.3 and 2 V/V0, which is somewhat higher than those of the equivalent equilibrium swelling ratios in scCO2 but the APH and AB of 8.6 and 0.30 g/g, respectively, display the same trend as found in scCO2. The amount of the scCO2 absorbed by the pDCPD_90 was unexpected, as the V/V0 in scCO2 showed no swelling. It seems that low viscosity and high diffusivity impart high penetrating capability of scCO2 within the thin walls of pDCPD_90, which might be a reason that pDCPD_90 absorb scCO2 even at zero swelling through the void walls. 3.2. Prileschajew epoxidation and aminolysis in scCO2 m-Chloroperoxybenzoic acid (mCPBA) is a widely used reagent for generating organic epoxides due to its versatile oxidizing power [27]. The OeO bond of mCPBA transfers an oxygen atom to electron-rich substrates (e.g. epoxidation of olefins), while the nucleophilic attack of mCPBA on ketones and aldehydes results in the insertion of an oxygen atom (e.g. Baeyer-Villiger reaction) [28]. Despite many oxidation procedures, the use of peroxy acids still constitutes one of the most useful synthetic procedures for the epoxidation of alkenes [29]. This synthetic procedure, known as the Prileschajew epoxidation [30], was also used for converting the double bonds of ROMP derived polymers into the corresponding epoxide [31]. To prepare epoxidized polymer network (Fig. 1), functionalization of pDCPD_90 with mCPBA was attempted at a range of super critical CO2 conditions (temperatures and pressures), such that the density exceeds the critical density (0.47 g/cm3) and approach those of the conventional liquid solvents (0.6–1.0 g/cm3). It was found that at lower pressures (< 100 bar), the mCPBA was not soluble enough to be effectively transported into the polymer, whereas at higher pressures (100 and 200 bar), the mCPBA was sufficiently soluble in scCO2 and effective functionalization was achieved. Although scCO2 has a zero dipole moment and a very low polarizability, it has a strong quadrupolar moment that allows significant interactions with mCPBA to dissolve it [26]. The conversion of double bonds at different conditions is reported in Table 2. The high density of double bonds along the backbone makes pDCPD very prone to oxidation when exposed to water or air [32,33]. Hence, elemental analysis (EA) was done immediately after purification/drying wherein 4.1% (2.5 mmol/g) of oxygen was found. Fully oxidized pDCPD has an oxygen content of about 36% (22.5 mmol/g) [32]. EA further revealed 28% (17.5 mmol/ g) of oxygen after epoxidation of pDCPD_90 with 1 equiv of mCPBA (hereafter referred to as pDCPD_epoxE) and 17% (11 mmol/g) of oxygen after epoxidation of pDCPD_90 with 0.5 equiv of mCPBA
Fig. 3. Scanning electron micrographs of pDCPD_90 after treatment at 100 bar (A) and 200 bar (B), N2 sorption isotherms measured at 77 K (C) and mesopore volume by the BJH model (D).
339
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
Fig. 4. A–C) pDCPD_0 in a cage within the high-pressure balance during the swelling tests at 40 °C and A) 1 bar, B) 100 bar and C) 200 bar. D–F) pDCPD_90 in a cage within the highpressure balance during the swelling tests at 40 °C and D) 1 bar, E) 100 bar and F) 200 bar. pDCPD_90 were weighted down in order not to float.
Table 1 The equilibrium absorption values. T [°C]
p [bar]
ρ(CO2) [g/cm3]
AB [g/g]
APH [g/g]
40
100 200 100 200
0.657 0.844 0.188 0.489
0.19 0.44 0.01 n.d.
4.33 8.11 1.70 4.34
100
Table 2 Elemental analysis data and degrees of conversion. Samplea
Oxygen [mmol/ g]b
Oxygen [mmol/ g]c
Oxygen [mmol/ g]d
N [%]e
(CH2NH2)2 [mmol/g]f
[%]g
pDCPD_epoxE
15.13 7.56
17.5 11
15 8.5
12 16
4.25 5.85
28 77
pDCPD_epoxH
Fig. 5. FTIR spectra of pDCPD unoxidized (the top), pDCPD oxidized (middle) and epoxidized pDCPD (bottom).
a
epoxE denotes 1 equivalent of mCPBA used to functionalize pDCPD, epoxH denotes 0.5 equivalent of mCPBA used to functionalize pDCPD. b Calculated. c Determined as a balance between 100% and wt% of CHNS from elemental analyses. d Corrected amount taking into account oxygen before epoxidation. e Found from elemental analyses. f Loading of reactive groups. g The degree of conversion calculated from the ratio of the loading of amine nucleophile to corrected amount of oxygen in the polyHIPE material. It was assumed that no additional crosslinking occurred.
stretch vibrations at 1258, 901, and 840 cm−1 respectively. The Csp2eH vibrations at 3050 cm−1 almost completely disappeared, which proves that the mCPBA reacted with most of the olefins in pDCPD (Fig. 5) [34]. The formation of cyclic carbonates upon the reaction of epoxide groups with CO2 under supercritical conditions cannot be ruled out [35]. In a next step the preparation of the β-amino alcohol derivatives from pDCPD_epox polyHIPEs (Fig. 2) was attempted. One of the most practical and widely used routes for the synthesis of these compounds is the direct aminolysis of epoxides at elevated temperature with an excess of amine [36]. To test the reactivity of epoxide ring in scCO2, ethylenediamine was chosen as the nucleophile. The percent of conversions are reported in Table 2. 4.25 or 5.85 mmol/g of amine groups was found after ethylenediamine functionalization that corresponds to 28 or 77% of conversion for pDCPD_epoxE or pDCPD_epoxH polyHIPEs, respectively. Somewhat better conversion in the case of pDCPD_epoxH can be ascribed to the lower yields of the overoxidized products after epoxidation, which results in higher amount of the epoxide rings that reacts readily with the
(hereafter referred to as pDCPD_epoxH). By taking into account the 2.5 mmol/g of the oxygen that is present prior the epoxidation, corrected oxygen amount after epoxidation would be 15 mmol/g or 8.5 mmol/g for pDCPD_epoxE or pDCPD_epoxH, respectively. Whereas 15.13 mmol/g or 7.56 mmol/g of oxygen for pDCPD_epoxE or pDCPD_epoxH, respectively, is expected for the complete conversion, slightly higher amounts of oxygen determined by the EA pointed to the presence of overoxidized products (due to the action of oxygen). The IR spectrum of pDCPD_epox showed the characteristic carbonyl stretch vibration at 1715 cm−1 confirming overoxidation. More importantly, IR spectra of pDCPD_epox displayed the characteristic epoxide ring 340
Journal of CO₂ Utilization 21 (2017) 336–341
N. Trupej et al.
early during process development, Ind. Eng. Chem. Res. 39 (2000) 960–972. [11] P. Munshi, S. Bhaduri, Supercritical CO2: a twenty-first century solvent for the chemical industry, Curr. Sci. 97 (2009) 63–72. [12] J.L. Kendall, D.A. Canelas, J.L. Young, J.M. DeSimone, Polymerizations in supercritical carbon dioxide, Chem. Rev. 99 (1999) 543–563. [13] C. Boyère, C. Jérôme, A. Debuigne, Input of supercritical carbon dioxide to polymer synthesis: an overview, Eur. Polym. J. 61 (2014) 45–63. [14] R. Butler, C.M. Davies, A.I. Cooper, Recent developments in materials synthesis and processing using supercritical CO2, Adv. Mater. 13 (2001) 1111–1114. [15] A.I. Cooper, J.M. DeSimone, Polymer synthesis and characterization in liquid/supercritical carbon dioxide, Curr. Opin. Solid State Mater. Sci. 1 (1996) 761–768. [16] Ž. Knez, E. Markočič, Z. Novak, M. Knez Hrnčič, Processing polymeric biomaterials using supercritical CO2, Chem. Ing. Tech. 83 (2011) 1371–1380. [17] Ž. Knez, Particle formation using supercritical fluids – a short review, Chem. Ind. Chem. Eng. Q. 12 (2006) 141–146. [18] S.K. Liao, P.S. Chang, Literatures on dyeing technique of supercritical fluid carbon dioxide, Am. J. Anal. Chem. 3 (2012) 923–930. [19] E. Ramsey, Q. Sun, Z. Zhang, C. Zhang, W. Gou, Mini-review: green sustainable processes using supercritical fluid carbon dioxide, J. Environ. Sci. 21 (2009) 720–726. [20] L. Dua, J.Y. Kellya, G.W. Roberts, J.M. DeSimone, Fluoropolymer synthesis in supercritical carbon dioxide, J. Supercrit. Fluids 47 (2009) 447–457. [21] S. Kovačič, K. Jeřabek, P. Krajnc, C. Slugovc, Ring opening metathesis polymerisation of emulsion templated dicyclopentadiene giving open porous materials with excellent mechanical, Polym. Chem. 3 (2012) 325–328. [22] (unlabelleda) A.-C. Knall, S. Kovačič, M. Hollauf, D.P. Reishofer, R. Saf, C. Slugovc, Inverse electron demand Diels–Alder (iEDDA) functionalisation of macroporous poly(dicyclopentadiene) foams, Chem. Commun. 7 (2013) 325–7327; (unlabelledb) S. Kovačič, F. Preishuber-Pflügl, D. Pahovnik, E. Žagar, C. Slugovc, Covalent incorporation of the surfactant into high internal phase emulsion templated polymeric foams, Chem. Commun. 51 (2015) 7225–7228. [23] N. Trupej, M. Knez Hrnčič, M. Škerget, Ž. Knez, Solubility and binary diffusion coefficient of argon in polyethylene glycols of different molecular weights, J. Supercrit. Fluids 103 (2015) 10–17. [24] S. Kovačič, N.B. Matsko, K. Jerabek, P. Krajnc, C. Slugovc, On the mechanical properties of HIPE templated macroporous poly(dicyclopentadiene) prepared with low surfactant amounts, J. Mater. Chem. A 1 (2013) 487–490. [25] M. Ovadia, M.S. Silverstein, High porosity responsive hydrogel copolymers from emulsion templating, Polym. Int. 65 (2016) 280–289. [26] T. Delgado-Abad, J. Martinez-Ferrer, A. Caballero, A. Olmos, R. Mello, M. Elena Gonzalez-Nunez, P.J. Perez, G. Asensio, Supercritical carbon dioxide: a promoter of carbon–halogen bond heterolysis, Angew. Chem. Int. Ed. 52 (2013) 13298–13301. [27] S. Cha, J. Hwang, M.G. Choi, S.K. Chang, Dual signaling of m-chloroperbenzoic acid by desulfurization of thiocoumarin, Tetrahedron Lett. 51 (2010) 6663–6665. [28] S. Kamijo, S. Matsumura, M. Inoue, CCl3CN: a crucial promoter of mCPBA-mediated direct ether oxidation, Org. Lett. 12 (2010) 4195–4197. [29] J.O. Edwards, Peroxide Reaction Mechanisms, in: J.O. Edwards (Ed.), Interscience, New York, 1962p. 67. [30] N. Prileschajew, Oxydation ungesättigter Verbindungen mittels organischer Superoxyde, Ber. 42 (1909) 4811–4815. [31] T.J. Boyd, R.R. Schrock, Sulfonation and epoxidation of substituted polynorbornenes and construction of light-emitting devices, Macromolecules 32 (1999) 6608–6618. [32] S. Kovačič, P. Krajnc, C. Slugovc, Inherently reactive polyHIPE material from dicyclopentadiene, Chem. Commun. 46 (2010) 7504–7506. [33] S. Kovačič, F. Preishuber-Pflugl, C. Slugovc, Macroporous polyolefin membranes from dicyclopentadiene high internal, Macromol. Mater Eng. 299 (2014) 843–850. [34] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identification of Organic Compounds, John Wiley & Sons, New York, 1991. [35] M. North, R. Pasquale, C. Young, Synthesis of cyclic carbonates from epoxides and CO2, Green Chem. 12 (2010) 1514–1539. [36] N. Azizi, M.R. Saidi, Highly chemoselective addition of amines to epoxides in water, Org. Lett. 7 (2005) 3649–3651.
amine nucleophile. 4. Conclusions Supercritical CO2 has been shown to provide a simple and effective processing route to generate functional macroporous polymers. Heterogeneous-phase organic synthesis using scCO2 as a solvent is a promising way as products are simply isolated by depressurization, which results immediately in dry, final compounds eliminating postreaction steps like filtration and drying indispensable when liquid organic solvents are used. The absorption of scCO2 in pDCPD polymers was found to be higher at lower temperatures and increased with increasing pressure at zero swelling, while maintaining the characteristic open-porous morphology of emulsion-templated pDCPD foams. Functionalization of pDCPD polyHIPEs in scCO2 was performed through the Prileschajew epoxidation using mCPBA. Exploiting reactivity of epoxides, further functionalization with the amine nucleophile was obtained. A high loading of up to 5.85 mmol/g of β-amino alcohols (11.4 mmol/g of nitrogen) was obtained. Acknowledgements This work was supported via the grant BI-AT/16-17-023. The authors gratefully acknowledge the financial support of the Ministry of Higher Education, Science and Technology of the Republic of Slovenia, and the Slovenian Research Agency (Program No. P2-0145 and P2–0046). References [1] L. Kircher, P. Theato, N. R.Cameron, Functionalization of porous polymers from high-internal-phase emulsions and their applications, in: P. Theato, H.A. Klok (Eds.), Functional Polymers by Post-polymerization Modification: Concepts, Practical Guidelines and Applications. Wiley-VCH, Weinheim, 2012. [2] A. Kirschning, H. Monenschein, R. Wittenberg, Functionalized polymers emerging versatile tools for solution-phase chemistry and automated parallel synthesis, Angew. Chem. Int. Ed. 40 (2001) 650–679. [3] M.V. Fawaz, P.J.H. Scott, Solid-phase organic synthesis, Kirk-Othmer Encyclopedia of Chemical Technology, (2014), pp. 1–27. [4] P. Hodge, D.C. Sherrington, Polymer-Supported Reactions in Organic Synthesis, Wiley, Chichester, UK, 1980. [5] Polymeric Materials in Organic Synthesis and Catalysis, in: M.R. Buchmeiser (Ed.), Wiley-VCH, Weinheim, 2003. [6] D.C. Sherrington, Polymer-supported reagents, catalysts, and sorbents: evolution and exploitation—a personalized view, J. Polym. Sci. Part A 39 (2001) 2364–2377. [7] P.W. Small, D.C. Sherrington, Design and application of a new rigid support for high efficiency continuous-flow peptide synthesis, J. Chem. Soc. Chem. Commun. 21 (1989) 1589–1591. [8] N.R. Cameron, P. Krajnc, M.S. Silverstein, Colloidal templating, in: M.S. Silverstein, N.R. Cameron, M.A. Hillmyer (Eds.), Porous Polymers, John Wiley & Sons, Inc, Hoboken, NJ, USA, 2011. [9] J.A. Darr, M. Poliakoff, New directions in inorganic and metal-organic coordination chemistry in supercritical fluids, Chem. Rev. 99 (1999) 495–542. [10] U. Fischer, K. Hungerbühler, Assessing safety, health, and environmental impact
341