Advertisement

Chemistry Africa

, Volume 2, Issue 2, pp 253–265 | Cite as

Oligoester-Derivatized (Semi-)Interpenetrating Polymer Networks as Nanostructured Precursors to Porous Materials with Tunable Porosity

  • Daniel GrandeEmail author
  • Géraldine Rohman
Original Article

Abstract

Porous polymeric materials with tunable porosity can be engineered from oligoester-derivatized semi-interpenetrating polymer networks (semi-IPNs) or IPNs, respectively composed of either uncrosslinked or crosslinked aliphatic oligoesters entangled in a stiff subnetwork. In this paper, miscellaneous polyester/poly(methyl methacrylate)-based semi-IPN and IPN systems are first prepared as precursors with varying structural parameters, especially the nature [i.e., poly(d,l-lactide), poly(ε-caprolactone)] and the molar mass (i.e., from 560 to 3700 g mol−1) of the oligoester precursor. (Nano)porous networks with defined porosity are then generated through two complementary routes. This original paper discusses the scope and limitations of both approaches and investigates the correlation between the structure and morphology of the generated networks and the porosity of the resulting porous materials. We demonstrate that the choice of the precursors with defined compatibility is of paramount significance in the length scale of phase separation associated with nanostructured networks as well as in the porosity scale of (nano)porous materials derived therefrom. Indeed, we find that the quantitative extraction of uncrosslinked oligoesters from semi-IPNs allows for the elaboration of nanoporous networks with pore diameters lower than 150 nm, provided that a high miscibility between both partners in semi-IPN precursors is attained, i.e. when using the lower molar mass oligoester. Alternately, the total hydrolysis of the polyester subnetwork associated with IPNs offers more versatility, since nanoporous networks can be obtained with a pore size range of 20–150 nm, regardless of the oligoester nature and molar mass in IPN precursors.

Keywords

Aliphatic oligoesters (Semi-)interpenetrating polymer networks Extraction Hydrolysis Porous materials Structure–morphology relationships 

1 Introduction

The design of functional porous polymeric materials has been the subject of widespread interest and intense research, as they are involved in a wide array of applications, e.g. monoliths for chromatographic techniques, separation membranes, interlayer dielectrics, high surface area catalytic supports, as well as size/shape-selective nanoreactors [1, 2, 3, 4, 5, 6, 7, 8, 9]. Besides the classical synthetic strategies that rely on the use of solvents or gases as porogens, original approaches with porogen templates, capable of inducing specific structural pores within the residual structures, have been developed [10, 11]. These template-oriented routes are quite interesting, since a wide array of porous polymers with a well-defined porosity can be designed.

Resorting to (semi-)interpenetrating polymer networks ((semi-)IPNs) as nanostructured precursors to engineer porous crosslinked materials has solely been studied by a few research teams [12, 13, 14, 15, 16, 17, 18, 19]. While semi-IPNs are composed of uncrosslinked sub-chains entrapped in a polymer network, IPNs constitute an intimate combination of two independent sub-networks, at least one of which is synthesized in the immediate presence of the other [20, 21, 22, 23, 24, 25]. Although IPNs do not generally lead to chain interpenetration at the molecular length scale, small domain sizes from tens to a few hundreds of nanometers can be obtained under synthetically controlled experimental conditions. The peculiar interlocking framework confers to IPNs a microphase-separated co-continuous morphology leading to their unique properties. The IPN morphology essentially depends on the compatibility of each partner (i.e., thermodynamic factor) and the kinetics and mechanism associated with the formation of each sub-network (i.e., kinetic parameters). Nevertheless, if the phase morphology may be varied with several parameters, it has been reported that the phase continuity is controlled by the partner volume fractions and a co-continuous morphology is generally obtained with a 50/50 composition [21, 22, 23]. Moreover, when IPNs are synthesized by the sequential method, i.e. when each sub-network is obtained in two consecutive steps, the first sub-network generally constitutes the most continuous phase and its cross-linking density is an important parameter to control the IPN morphology, while the cross-linking density of the second sub-network has no significant effect [26]. Consequently, the order of the sub-network formation is crucial as different phase morphologies and properties may be obtained. When arising from the combination of two partners that exhibit a contrasted degradability under specific conditions, such complex polymer structures are particularly interesting since nanoporous networks can be generated through selective degradation methods [12, 13, 14, 15, 16, 17, 18, 19]. In this regard, IPNs based on a hydrolyzable polyester, such as poly(d,l-lactide) (PLA) or poly(ε-caprolactone) (PCL), and a non-hydrolyzable polymer, such as poly(methyl methacrylate) (PMMA), can be considered as appropriate precursors.

Straightforward routes to porous cross-linked polymeric materials through the utilization of PLA/PMMA semi-IPNs or IPNs as precursors were previously developed [26, 27]. Porous methacrylic networks were readily generated from the quantitative extraction of uncrosslinked PLA oligomers in semi-IPNs or they were engineered from the selective hydrolysis of PLA sub-network in IPNs. The scope and limitations of both systems for the generation of (meso)porous materials were investigated. The effect of the cross-linker nature and the cross-linking density of the PMMA sub-network on the domain sizes in the (semi-)IPN precursors, and the correlation with the pore sizes distributions in the porous materials derived therefrom, were carefully studied. Macro- to mesoporous networks were derived from semi-IPNs and the variation of pore sizes was attributed to the miscibility of both partners in semi-IPN precursors [27]. For IPN systems, the pore sizes were always smaller than 100 nm, regardless of the cross-linker nature and concentration, which showed that the cross-linking density of the PMMA sub-network and the compatibility between both sub-networks had a relatively small impact on phase separation in IPN precursors [28].

In light of the scarcity of studies on porous materials derived from (semi-)IPN systems, we decided to get a better insight into the correlation between the morphology of the (semi-)IPN precursors and the pore sizes of the resulting porous methacrylic networks, and notably on the impact of the molecular features of the first sub-network generated during IPN synthesis (i.e. polyester partner). In the present paper, miscellaneous polyester/PMMA (semi-)IPNs with a 50/50 wt% composition are prepared to ensure the co-continuity of each phase, and the effect of the nature and molar mass of the oligoester precursor is systematically investigated for the first time. In this regard, poly(d,l-lactide) (PLA) or poly(ε-caprolactone) (PCL) oligomers are used with different molar masses. We particularly pay attention to the correlations between the structure and morphology of the (semi-)IPN precursors and the porosity of the resulting porous materials through various physico-chemical analyses, since the control of the precursor miscibility associated with both types of reference systems is of paramount significance for the design of nanoporous materials with defined pore sizes.

2 Experimental

2.1 Materials

Dihydroxy-telechelic PLA oligomers were synthesized by ring-opening polymerization of d,l-lactide initiated by the ethylene glycol/tin(II) octanoate system, according to a literature method [29]. Dihydroxy-telechelic PCL oligomers were purchased from Aldrich. Dibutyltin dilaurate (DBTDL, Fluka) was used as received. 4,4′,4″-triisocyanato-triphenylmethane (Desmodur® RU; 1.25 mol L−1 in dichloromethane solution) was provided by Bayer. Methyl methacrylate (MMA, Aldrich) was dried over CaH2, and distilled under vacuum prior to use. Diurethane dimethacrylate (DUDMA) were purchased from Aldrich and used as received. AIBN (Merck) was purified by recrystallization in methanol.

2.2 Preparation of Precursory Networks

In all experiments, a mold was devised by clamping together two glass plates separated by a 2 mm-thick silicone rubber gasket.

2.2.1 Polyester Single Networks

Polyester single networks were prepared with a (Oligoester + Desmodur® RU)/dichloromethane mass composition of 50/50 wt% and ratios of [NCO]0/[OH]0 and [DBTDL]0/[Oligoester]0 equal to 1.4 and 0.44, respectively. An example of the preparation of a PLA single network is given hereafter. 1 g of PLA (Mn = 1700 g mol−1; 5.9 × 10−4 mol) was dissolved in 0.91 mL of dichloromethane (1.21 g), and subsequently, 0.45 mL of Desmodur® RU (0.21 g; 5.6 × 10−4 mol) and 0.15 mL of DBTDL (2.5 × 10−4 mol) were added. The solution was then poured into the mold and kept at room temperature for 20 h. Thereafter, the mold was heated to 65 °C for 2 h, and then to 110 °C for 2 h (Fig. 1).
Fig. 1

Preparation of polyester-based single networks

2.2.2 PMMA Single Network

An example of the preparation of a single network with a MMA/DUDMA molar composition of 90/10 mol% is given hereafter. A mixture of MMA (0.9 g; 9 × 10−3 mol), DUDMA (0.47 g; 10−3 mol), and AIBN (0.033 g; 0.2 × 10−3 mol, [AIBN]0/([MMA]0 + 2[DUDMA]0) = 0.02) was degassed under vacuum, poured under nitrogen into the mold. Then, the mixture was heated at 65 °C for 2 h, and cured at 110 °C for 2 h (Fig. 2).
Fig. 2

Preparation of PMMA-based single networks

2.2.3 Oligoester/PMMA Semi-IPNs

An example of the preparation of a semi-IPN with a PLA/PMMA mass composition of 50/50 wt% and a MMA/DUDMA molar composition of 90/10 mol% ([AIBN]0/([MMA]0 + 2[DUDMA]0) = 0.02) is given hereafter. 0.50 g of PLA (Mn = 1700 g mol−1; 2.9 × 10−4 mol), 0.33 g of MMA (3.3 × 10−3 mol), 0.17 g of DUDMA (3.6 × 10−4 mol), and 12 mg of AIBN (7.3 × 10−5 mol) were mixed, degassed under vacuum, and poured into the mold under nitrogen. The system was then heated to 65 °C for 2 h, and cured at 110 °C for a 2 h (Fig. 3).
Fig. 3

Synthesis of oligoester/PMMA-based semi-IPNs and design of porous PMMA networks derived therefrom

2.2.4 Polyester/PMMA IPNs

An example of the preparation of an IPN with a PLA/PMMA mass composition of 50/50 wt%, ratios of [NCO]0/[OH]0 and [DBTDL]0/[Oligoester]0 equal to 1.4 and 0.44, respectively, and a MMA/DUDMA molar composition of 90/10 mol% ([AIBN]0/([MMA]0 + 2[DUDMA]0) = 0.02) is given hereafter. 1 g of PLA (Mn = 1700 g mol−1; 5.9 × 10−4 mol), 0.76 g of MMA (7.6 × 10−3 mol), 0.40 g of DUDMA (8.5 × 10−4 mol), and 27.7 mg of AIBN (1.7 × 10−4 mol) were mixed and degassed under vacuum. Then, 0.45 mL of Desmodur® RU (0.21 g; 5.6 × 10−4 mol) and 0.15 mL of DBTDL (2.5 × 10−4 mol) were added under nitrogen, and the solution was poured into the mold and kept at room temperature for 20 h. Finally, the system was heated to 65 °C for 2 h, and cured at 110 °C for 2 h (Fig. 4).
Fig. 4

Synthesis of polyester/PMMA-based IPNs and design of porous methacrylic networks derived therefrom

Other single polyester networks and (semi-)IPNs were synthesized in similar ways by changing the nature and/or the molar mass of the oligoester precursor.

2.3 Extraction of Networks and Formation of Porous Networks from Semi-IPNs

All networks and (semi-)IPNs were Soxhlet extracted with dichloromethane for 24 h at 40 °C. After extraction, the samples were dried under vacuum, weighed, and the sol fractions (mass percentages of extractables) were calculated. The extraction of linear oligoesters from semi-IPNs led to the formation of residual porous PMMA networks (Fig. 3).

2.4 Formation of Porous Networks by Partial Hydrolysis of IPNs

0.2 g of IPNs were immersed at 60 °C in a mixture composed of 4 mL of a methylamine solution (pH 13.6) and 4 mL of ethanol. After 24 h, the residual networks were rinsed with deionized water up to neutral pH, and dried under vacuum. The mass loss (Δm) was assessed as follows (Eq. 1):
$$\Delta m\,({\text{wt}}\% ) = 100 \times (m_{0} - m_{\text{d}} )/m_{0} ,$$
(1)
where m0 and md stand for the initial mass of the samples and their residual mass after vacuum drying, respectively.

The total hydrolysis of the polyester sub-network led to the formation of residual porous PMMA networks (Fig. 4).

2.5 Instrumentation

FTIR spectra were recorded between 4000 and 450 cm−1 by averaging 32 consecutive scans with a resolution of 4 cm−1 on a Bruker Tensor 27 DTGS spectrometer in Attenuated Total Reflection (ATR) mode.

DSC thermograms were recorded with a Perkin Elmer DSC7 calorimeter under nitrogen atmosphere. The analyses were carried out at a heating rate of 20 °C min−1 and the second run was performed after quenching. In order to avoid network degradation during the analysis, PMMA single networks and (semi-)IPNs were scanned twice from − 100 to 200 °C, while polyester single networks and oligoester precursors were scanned from − 100 to 100 °C and from − 100 to 200 °C for the first and second scans, respectively. The Tg ranges were measured in the second run through the determination of the values associated with the intercepts of tangent to midpoint of the specific heat increment with “glassy” (lower limit, Tg,onset) and “viscous” baselines (upper limit, Tg,end), respectively.

1H and 13C spectra of sol fractions and hydrolysis products were recorded at room temperature using a Bruker Avance II spectrometer operating at a resonance frequency of 400 and 100 MHz, respectively. The sample concentration was 10 mg mL−1, and CDCl3 was used as the solvent and internal standard (7.27 ppm).

The size exclusion chromatography (SEC) equipment comprised a Spectra Physics P100 pump, two PLgel 5 μm mixed-C columns (Polymer Laboratories), and a Shodex RI 71 refractive index detector. Tetrahydrofuran (THF) was used as the eluent at a flow rate of 1 mL min−1, and polystyrene standards (Polymer Laboratories) were employed for calibration.

Scanning electron microscopy (SEM) analyses were performed with a LEO 1530 microscope equipped with a high-vacuum (10−10 mmHg) Gemini column. The accelerating tensions ranged from 1 to 5 kV; two types of detectors (InLens and Secondary Electron) were used. Prior to analyses, the samples were cryofractured and coated with a Pd/Au alloy (4 nm) in a Cressington 208 HR sputter-coater.

2.6 Thermoporometry by DSC

The pore size and pore size distribution of the porous materials were determined through thermoporometry based on the melting temperature (Tm) depression of a liquid constrained within the pores [30, 31, 32]. To this purpose, the samples were immersed in ethanol for 2 h, and then placed for 1 h in ethanol/water mixtures of various compositions (70/30, 50/50, 30/70 vol%). After a 2 week immersion in pure water, the melting thermograms of wiped samples were recorded from − 50 to 5 °C at a heating rate of 1 °C min−1.

The pore size distribution was obtained by plotting dV/dR vs. the pore diameter (Dp) evaluated by using Eqs. (2) and (3), respectively [30, 31, 32]:
$${\text{d}}V/{\text{d}}R\,({\text{cm}}^{3} \cdot {\text{nm}}^{ - 1} \cdot {\text{g}}^{ - 1} ) = [({\text{d}}q/{\text{d}}t) \times (T_{\text{m}} - T_{{{\text{m}}0}} )^{2} ]/[32.33 \times \rho \times \upsilon \times m \times \Delta H(T)]$$
(2)
$$D_{\text{p}} ({\text{nm}}) = 2 \times [0.68 - 32.33/(T_{\text{m}} - T_{{{\text{m}}0}} )],$$
(3)
where Tm and Tm0 are the melting temperatures of confined and bulk water, respectively, and dq/dt, ρ, v, m and ΔH (T) are the heat flow recovered by DSC, the water density, the heating rate, the sample mass and the melting enthalpy of water, respectively. ΔH (T) was calculated from Eq. (4) [30, 31, 32]:
$$\Delta H(T)\,({\text{J}} \cdot {\text{g}}^{ - 1} ) = 332 + 1.39 \times (T_{\text{m}} - T_{{{\text{m}}0}} ) + 0.155 \times (T_{\text{m}} - T_{{{\text{m}}0}} )^{2} .$$
(4)

2.7 Determination of Density and Porosity Ratio Values

The samples were immersed in ethanol for 2 h, and then placed for 1 h in ethanol/water mixtures of various compositions (70/30, 50/50, 30/70 vol%). After a 2 week immersion in pure water, the wet mass was measured, and the mass swelling ratio (qw), the pore volume (Vpore), and the apparent density (dapp) were calculated from Eqs. (5), (6) and (7), respectively [33]:
$$q_{w} = m_{w} /m_{d}$$
(5)
$$V_{\text{pore}} ({\text{cm}}^{3} \cdot {\text{g}}^{ - 1} ) = (q_{w} - 1 ) /d_{s}$$
(6)
$$V_{\text{pore}} = 1/d_{\text{app}} - 1/d_{\text{true}} ,$$
(7)
where mw, md, ds and dtrue stand, respectively, for the samples wet mass, their mass after vacuum drying, the solvent density (water), and the true density of the PMMA matrix measured by helium pycnometry at 25 °C using a Micromeritics Accupyc 1330 equipment.
The porosity ratio P was then calculated by using Eq. (8):
$$P = 1 - d_{\text{app}} /d_{\text{true}} .$$
(8)

3 Results and Discussion

3.1 Preparation of Semi-IPN Precursors and Investigation of Phase Separation

Various oligoester/PMMA (50/50 wt%) semi-IPNs were prepared by bulk free-radical copolymerization of MMA and DUDMA with a composition of 90/10 mol%, in the presence of oligoesters of different natures and molar masses (Table 1). The radical polymerization occurred at 65 °C and the cross-linking completion arose during the final curing at 110 °C (Fig. 3), which was confirmed by the absence of the C=C absorption band of MMA and dimethacrylate monomers at 1640 cm−1 in the FTIR spectra. In a previous study concerning the kinetics of network formation, it was shown that “soft” oligoesters behaved like a diluent toward the methacrylic copolymerization process which led to the complete conversion of C=C bonds before the curing process [25]. Moreover, the kinetic process was not greatly affected by the variation of the polyester nature or molar mass (data not shown).
Table 1

Molecular and thermal characteristics of oligoesters

Nature

M n a (g mol−1)

Ð b

T g c (°C)

ΔC p d (J g−1 °C−1)

Tm (°C)

ΔHm (J g−1)

X c e (%)

PLA

570

1.2

− 20

0.57

1700

1.2

10

0.55

3700

1.4

25

0.61

PCL

560

1.7

− 80

0.16

36

46.1

34

2100

2.1

− 70

0.35

65

89.7

66

aMolar mass as determined by 1H NMR

bPolydispersity index as determined by SEC in THF with polystyrene standards for calibration

cTg value as obtained at the midpoint

dΔCp = Cp,v − Cp,g: heat capacity jump at Tg as determined by DSC, where Cp is the heat capacity, the subscripts v and g refer to the “viscous” and “glassy” states, respectively

eCrystallinity degree as calculated with ΔHm0 (100% crystalline PCL) = 135 J g−1. [Kweon HY, Yoo MK, Park IK, Kim TH, Lee HC, Lee HS, Oh JS, Akaike T, Cho CS (2003) Biomaterials 24:801]

For (semi-)IPN-related materials [21, 22], the turbidity τ, which corresponds to the relative attenuation of light by the materials, accounts for the degree of chain interpenetration and allows for the determination of the microdomain size, provided the difference between the refractive indices of both partners is significant (Δn ≥ 0.02). Indeed, microdomain sizes are smaller than about 150 nm when materials are transparent, while the opaque ones possess microdomain sizes higher than 150 nm [34]. Table 2 reports the different visual aspects observed for semi-IPNs before extraction. In the case of PLA-based semi-IPNs, the experimental n D 25 values measured for the PLA oligomer and a PMMA network were quite different (n D 25 PLA = 1.46, n D 25 PMMA = 1.49), and therefore the material turbidity can be used to evaluate the effect of the oligoester molar mass on the phase separation. It is noteworthy that the increase in the PLA molar mass led to an increase in the oligoester domain sizes, as semi-IPNs were transparent when the molar mass of the oligomer was equal to 570 g mol−1 and opaque when the latter was equal to 3700 g mol−1. For PCL-based semi-IPNs, the n D 25 values of PMMA sample and PCL oligomer (n D 25 PCL = 1.49) were quite identical, and it was consequently difficult to take into account the transparency of the samples. Nevertheless, there was clearly a phase separation when the samples were opaque. Consequently, it was possible to conclude that PCL oligomers led to higher oligoester domain sizes compared to PLA oligomers with the same molar mass, as semi-IPNs were opaque with PCL 2100 g mol−1 and translucent for PLA 1700 g mol−1.
Table 2

Visual aspect and DSC analysis of semi-IPNs before and after extraction

Oligoester nature and molar mass

Visual aspect

Semi-IPNs before extraction

Semi-IPNs after extraction

Tg onset (°C)

ΔT g a (°C)

ΔC p b (J g−1 °C−1)

Tg,onset (°C)

ΔT g a (°C)

ΔC p b (J g−1 °C−1)

PLA 570 g mol−1

tr

− 14

25

0.25

116

25

0.24

PLA 1700 g mol−1

tl

25

18

0.23

116

27

0.19

PLA 3700 g mol−1

o

39

13

0.19

106

31

0.23

PCL 560 g mol−1

tl

− 70

40

0.05

112

31

0.17

(Tm = 38 °C, ΔHm = 22.2 J g−1, X c c  = 16%)

PCL 2100 g mol−1

o

− 65

38

0.18

117

25

0.18

(Tm = 63 °C, ΔHm = 40.0 J g−1, X c c  = 30%)

o opaque, tl translucent, tr transparent

aΔTg = Tg,end − Tg,onset: range of temperatures in which the glass transition occurs as determined by DSC

bΔCp = Cp,v − Cp,g: heat capacity jump at Tg as determined by DSC, where Cp is the heat capacity, the subscripts v and g refer to the “viscous” and “glassy” states, respectively

cCrystallinity degree as calculated with ΔHm0 (100% crystalline PCL) = 135 J g−1. [Kweon HY, Yoo MK, Park IK, Kim TH, Lee HC, Lee HS, Oh JS, Akaike T, Cho CS (2003) Biomaterials 24:801]

Before extraction, it is noteworthy that semi-IPN materials exhibit a single glass transition (Table 2) between those of the corresponding oligoester (Table 1) and the PMMA single network precursors (Table 3). Nevertheless, the Tg values were quite low and did not obey the Flory–Fox equation. The transparent aspect of the semi-IPN prepared with the low molar mass PLA (570 g mol−1) supported the hypothesis of the absence of phase separation at a length scale of a few hundreds of nanometers. Moreover, the values of the heat capacity jump at the glass transition (ΔCp) was found to be close to the value obtained for a miscible 50/50 wt% physical blend constituted of PLA oligomer and PMMA prepared under experimental conditions identical to those employed for single network preparation (ΔCp = 0.33 J g−1 °C−1). For higher molar mass PLA, ΔCp decreased and the Tg ranges became narrower, thus putting in evidence an increase in the heterogeneity of the materials. The Tg values were then assumed to be those of PLA-rich domains, while those of PMMA-rich domains were not detectable. This result corroborated the conclusions inferred from the visual observations concerning the increase in phase separation when using oligomers with higher molar masses. In the case of PCL oligomers, the values of ΔCp found for miscible blends were 0.18 and 0.28 J g−1 °C−1 for PCL 560 and 2100 g mol−1, respectively. As a consequence, both PCL-containing semi-INPs exhibited a phase separation at the length scale of a few hundred nanometers. One more time, the phase separation was found higher for PCL-based semi-IPNs compared to PLA ones with the same molar mass as the differences between the values of ΔCp for semi-IPNs and those for miscible blends increased. Finally, a decrease in the enthalpy of melting and in the crystallinity degree of PCL oligomers were observed in semi-IPNs as PMMA hindered the oligoester crystallization [35].
Table 3

DSC analysis of polyester and PMMA single networks

 

Oligoester nature and molar mass

Tg,onset (°C)

ΔT g a (°C)

ΔC p b (J g−1 °C−1)

Polyester single networks

PLA 570 g mol−1

50

30

0.32

PLA 1700 g mol−1

20

25

0.29

PLA 3700 g mol−1

30

15

0.27

PCL 560 g mol−1

− 5

25

0.28

PCL 2100 g mol−1

− 55

10

0.32

(Tm = 50 °C, ΔHm = 4 J g−1)

PMMA single network

118

19

0.16

aΔTg = Tg,end − Tg,onset: range of temperatures in which the glass transition occurs as determined by DSC

bΔCp = Cp,v − Cp,g: heat capacity jump at Tg as determined by DSC, where Cp is the heat capacity, the subscripts v and g refer to the “viscous” and “glassy” states, respectively

In a previous paper [25], we related the dependence of domain size to the polymer–polymer miscibility in semi-IPN precursors, thanks to the calculation of PLA/PMMA interaction parameters (χ) and critical interaction parameters (χcr). We hypothesized that the domain sizes could be tuned by varying structural parameters through the variation of miscibility between the oligoester and the PMMA sub-network. In order to confirm this idea, we also evaluated these parameters in PLA/PMMA and PCL/PMMA semi-IPNs under investigation, and we studied the effect of the variation of the oligoester molar mass.

Taking into account that the PMMA sub-network is cross-linked and that its polymerization degree is equal to infinity, the χ and χcr values of semi-IPNs can be calculated using Eqs. (9) and (10) [36, 37, 38, 39]:
$$\chi = V_{\text{m}} \times \left( {\delta_{1} - \delta_{2} } \right)^{2} /(R \times T)$$
(9)
$$\chi_{\text{cr}} = 1/(2N_{1} ),$$
(10)
where Vm and T are the reference molar volume and the temperature, taken as 100 cm3 mol−1 and 298 K, respectively, R is the gas constant, δ1 is the Hildebrand solubility parameter of the oligoester, δ2 is the Hildebrand solubility parameter of the methacrylic copolymer network, and N1 is the polymerization degree of the oligoester.

δ1 was calculated using the group contribution method based on Van Krevelen’s molar attraction constants [38]: δ1 = 18.56 ± 0.15 MPa1/2 for PLA and δ1 = 17.94 ± 0.15 MPa1/2 for PCL. δ2 was calculated using an additive molar contribution relationship and was found to equal 20.76 ± 0.15 MPa1/2 for a 90/10 mol% MMA/DUDMA network [28]. As a matter of fact, the values of χ were 0.19 ± 0.05 and 0.32 ± 0.07 for PLA/PMMA and PCL/PMMA semi-IPNs, respectively. As the χ value was higher for PCL-based systems, it could be concluded that PCL oligomers led to a lower extent of miscibility with the PMMA sub-network in semi-IPN precursors.

According to Eq. (10), χcr could be calculated for each oligoesters and it is clear that χcr decreased when the molar mass of the oligomers increased. However, it should be stressed that the values obtained from Eq. (10) are valid for systems with only dispersive forces, and they can be incremented when intermolecular interactions arise between the polymeric components of the systems [38]. We previously demonstrated that hydrogen bonding could potentially be established between the urethane functions of DUDMA from the PMMA sub-network and the ester groups of the oligoester [27]. Therefore, the χcr values obtained for 10 mol% DUDMA-based systems with the different oligoesters are given in Table 4. Consequently, these values showed that the miscibility decreased when the oligoester molar mass increased as χcr became smaller. Finally, the limits of miscibility could be predicted from the difference χcr − χ. Indeed, when the value of χcr − χ > 0, the system can be considered to be miscible, while it is immiscible if χcr − χ < 0. Therefore, 10 mol% DUDMA-based semi-IPNs containing PLA 570 g mol−1 would not exhibit phase separation, while PLA 3700 g mol−1 or PCL 2100 g mol−1-based systems should. At least, PLA 1700 and PCL 560 g mol−1-based systems were in the limit of phase separation as the values of χcr − χ were around 0. To conclude, this theoretical approach may well account for the different behaviours towards phase separation in semi-IPNs when varying the oligoester nature or molar mass. Indeed, the calculation of oligoester/PMMA interaction parameters and the comparison between χ and χcr values corroborated the conclusions inferred from our previous results associated with DSC analysis and visual aspects.
Table 4

Variations of χcr with oligoester nature and molar mass for 10 mol% DUDMA-based systems

Oligoester nature and molar mass

χcr

PLA 570 g mol−1

0.32 ± 5.10−3

PLA 1700 g mol−1

0.21 ± 5.10−3

PLA 3700 g mol−1

0.17 ± 5.10−3

PCL 560 g mol−1

0.34 ± 5.10−3

PCL 2100 g mol−1

0.24 ± 5.10−3

3.2 Synthesis of IPN Precursors and Investigation of Microphase Separation

Polyester/poly(methyl methacrylate)-based IPN systems with a 50/50 wt% composition were synthesized by the in situ sequential method which consists in initially mixing all the precursors and then forming successively each sub-network through two non-interfering cross-linking reactions. The first formed sub-network was the polyester one by a polyaddition mechanism at room temperature between dihydroxy-telechelic oligoesters (molecular characteristics described in Table 1) and a pluriisocyanate, i.e. Desmodur® RU. After increasing the temperature to 65 °C, the methacrylic sub-network was created by free-radical copolymerization of MMA and DUDMA with a molar ratio of 90/10. Finally, the system was cured at 110 °C to ensure a near completion of the cross-linking processes (Fig. 4) [28, 40, 41]. Next, the IPN samples were Soxhlet extracted with CH2Cl2 for 24 h, and the amounts of soluble fractions are reported in Table 5. The soluble fractions were always below 14 wt% which fairly matched with the sum of the corresponding polyester and PMMA single networks soluble fractions, except for PLA 3700 g mol−1-based IPNs in which soluble fractions were very high. In the latter case, the SEC analysis of the soluble fractions showed that they were constituted in majority of PLA oligomer precursors (around 85%). DBTDL, PMMA oligomers, and only traces of residual methacrylates were also found by 1H NMR. On the contrary, a high ratio of PMMA with molar mass ranging from 100 to 60,000 g mol−1 was extracted from PLA 570 g mol−1-based IPNs, and a higher content of residual methacrylates was also found. However, the FTIR study of networks after extraction revealed that hydroxyl groups were still present in IPNs synthesized from oligoester with low molar mass which indicated that the cross-linking of polyester sub-network was not complete. The variations of the kinetic process of the polyester sub-network with the oligoester precursor molar mass were not studied as the Lambert–Beer Law was not valid for low molar masses.
Table 5

Soluble fractions associated with single networks and polyester/PMMA (50/50 wt%) (semi-)IPNs after extraction as well as mass loss associated with polyester/PMMA (50/50 wt%) IPNs after hydrolysis

Oligoester nature and molar mass

Soluble fractions after extraction (wt%)

Mass loss (wt%) after IPN hydrolysis

Polyester single networks

PMMA single networks

Semi-IPNs

IPNs

PLA 570 g mol−1

9

53

11

46

PLA 1700 g mol−1

6

51

10

47

PLA 3700 g mol−1

5

48

23

48

PCL 560 g mol−1

11

50

11

48

PCL 2100 g mol−1

12

56

14

49

3

Experimental conditions: [DBTDL]0/[Oligoester]0 = 0.44 − [NCO]0/[OH]0 = 1.4 − MMA/DUDMA molar ratio: 90/10 − [AIBN]0/([MMA]0 + 2[DUDMA]0) = 0.02 − 20 h at room T (only for polyester single networks and IPNs) + 2 h at 65 °C + 2 h at 110 °C

Concerning the PMMA sub-network formation, we previously showed that the “soft” polyester sub-network acted as a diluent towards the methacrylic copolymerization process and prevented the system from reaching the glassy state [28]. However, as the free-radical polymerization proceeded, the Tg of the system increased and the temperature of the reaction medium, i.e. 65 °C, was around the Tg values of IPNs (Table 6). Therefore, the polymerization was considered to take place in a confined medium (into the polyester sub-network matrix) and was going to stop unless the temperature was increased to 110 °C. We also found that the kinetic process was not greatly affected by the variation of the polyester nature or molar mass (data not shown).
Table 6

DSC analysis of IPNs before and after hydrolysis

Oligoester nature and molar mass

IPNs before hydrolysis

IPNs after hydrolysis

Tg,onset (°C)

ΔT g a (°C)

ΔC p b (J g−1 °C−1)

Tg,onset (°C)

ΔT g a (°C)

ΔC p b (J g−1 °C−1)

PLA 570 g mol−1

58

35

0.26

122

23

0.21

PLA 1700 g mol−1

45

46

0.22

128

16

0.15

PLA 3700 g mol−1

45

49

0.22

126

19

0.21

PCL 560 g mol−1

38

32

0.23

121

15

0.19

PCL 2100 g mol−1

− 50

52

0.14

126

20

0.19

aΔTg = Tg,end − Tg,onset: range of temperatures in which the glass transition occurs as determined by DSC

bΔCp = Cp,v − Cp,g: heat capacity jump at Tg as determined by DSC, where Cp is the heat capacity, the subscripts v and g refer to the “viscous” and “glassy” states, respectively

Interestingly, all IPNs were transparent indicating that domain sizes were always smaller than 150 nm as the difference between the refractive indices of both sub-networks was significant (Δn ≥ 0.02). Actually, the experimental n D 25 values measured for the PLA and PCL sub-networks were equal to 1.51 and 1.53, respectively, while that for PMMA sub-network was equal to 1.49. Such visual observations showed that chain interpenetration of both polyester and PMMA sub-networks in the interlocking configuration of IPNs led to a significant decrease in the phase separation compared to the corresponding semi-IPN homologues. This was particularly true for systems with a higher degree of immiscibility like those with PLA 3700 g mol−1 and PCL 2100 g mol−1 as the semi-IPNs synthesized with these oligomers were opaque.

DSC analyses also confirmed the decrease in phase separation in IPNs (Table 6). Indeed, all IPNs displayed single glass transition ranges comprised between the Tg of corresponding polyester sub-networks and that of PMMA single network (see Table 3), which argued in favor of the absence of phase separation at a length scale of a few hundreds of nanometers. No melting temperatures were detected in PCL-based IPNs, as the oligoester cross-linking totally prevented PCL crystallization. Moreover, the values of the heat capacity jump at the glass transition (ΔCp) were always equal or close to the expected ΔCp value of a 50/50 wt% physical blend theoretically constituted of polyester and PMMA networks (ΔCp,theoretical = 0.25 J g−1 °C−1), except for PCL 2100 g mol−1-based IPNs. Nevertheless, the ΔCp values decreased when the oligoester precursor molar mass increased showing that the domain sizes were still higher for these systems, like in the case of semi-IPN systems. This was very clear when using PCL 2100 g mol−1 as the value of ΔCp was equal to only 0.14 J g−1 °C−1. This trend was also supported by the increase in ΔTg with the oligoester precursor molar mass. All these results confirmed that the phase separation in IPNs was mainly governed by the interpenetration created by the cross-linking of both partners, preventing any phase segregation at the nanoscopic scale [28, 40, 41].

3.3 Generation of Porous Networks from Semi-IPNs

Porous methacrylic networks were obtained by a simple extraction procedure in a good solvent of the un-crosslinked oligoesters from oligoester/PMMA (50/50 wt%) semi-IPNs (Fig. 3). In order to avoid the collapse of the residual porous structures, the extraction was carried out at a temperature (40 °C) far below the Tg value of PMMA network [27, 42, 43, 44]. The amounts of soluble fractions are reported in Table 5. The extractable contents were always higher than or equal to 50 wt%, regardless of the oligoester nature and molar mass, demonstrating the quantitative extraction of the oligomers. Furthermore, 1H NMR and SEC analyses of the soluble fractions showed that they were constituted of more than 95% of oligoester precursor. No undesired grafting of PLA or PCL sub-chains onto PMMA sub-networks through transfer reactions was detected in the present systems. The total disappearance of the hydroxyl and carbonyl groups bands of the α,ω-dihydroxy oligoester in the FTIR spectra of the extracted semi-IPNs confirmed the latter assertion. DSC analyses (Table 2) showed that the Tg and ΔCp values of semi-IPNs after extraction matched pretty well those of the corresponding single network (Table 3). This arose from the quantitative extraction of oligoesters. Nevertheless for PLA 3700 g mol−1-based systems, Tg,onset was slightly lower than the value expected, probably due to a non-quantitative extraction as shown also by an extractable content of 48%.

The morphologies of semi-IPNs after extraction were examined by SEM (Fig. 5). The micrographs revealed highly porous structures, thus showing the effective role of oligoesters as template porogens.
Fig. 5

SEM micrographs of porous networks derived from extraction of oligoester/PMMA (50/50 wt%) semi-IPNs: PLA-based systems: a 3700 g mol−1, b 1700 g mol−1, c 560 g mol−1; PCL-based systems: d 2100 g mol−1, e 570 g mol−1

Pore sizes strongly depended on the oligoester nature and molar mass. Indeed, pore diameters were as large as 450 nm for PLA-based systems and 1000 nm for PCL-based-system homologues. Moreover, pore sizes clearly increased when increasing the oligoester molar mass (Table 7). The pore size distributions were also determined through DSC-based thermoporometry. We already used this technique for the determination of pore size distributions in (semi-)IPNs which also gave complementary results compared to SEM analysis, as thermoporometry allows for the measurement of the smaller pore diameters (< 200 nm) [27, 28]. The pore size ranges are reported in Table 7 and pore size distributions are shown in Fig. 6a.
Table 7

Pore diameters of resulting porous networks after extraction of semi-IPNs or hydrolysis of IPNs as determined by SEM and DSC-based thermoporometry

Oligoester nature and molar mass

Pore diameters

Semi-IPNs

IPNs

SEM (nm)

DSC (nm)

DSC (nm)

PLA 570 g mol−1

10–75

40–115

17–75

PLA 1700 g mol−1

25–150

30–140

20–80

PLA 3700 g mol−1

75–450

10–200a

30–150

PCL 560 g mol−1

25–150

50–170

20–65

PCL 2100 g mol−1

150–1000

10–200a

20–95

aDSC-based thermoporometry only allowed for determination of pore size up to 200 nm. For larger pore sizes, the melting peaks of confined and bulk water were not resolved: confined water just behaved as a bulk solvent

Fig. 6

Pore size distribution profiles of porous methacrylic networks as determined by DSC-based thermoporometry for PLA-(full symbols) and PCL-(empty symbols) containing systems with varying oligoester molar masses: a semi-IPNs after extraction, b IPNs after hydrolysis

Overall, the pore sizes obtained by SEM and those determined by thermoporometry were in reasonable agreement. Finally, the dependence of pore sizes in such porous materials mirrored the differences observed in semi-IPN precursors from visual aspects, DSC analysis, and polymer–polymer miscibility inferred from the calculation of oligoester/PMMA interaction parameters. To conclude, the miscibility of both partners in semi-IPN systems is the most essential parameter controlling the pore size distribution after extraction. It is noteworthy that nanoporous structures with pores smaller than 150 nm could be engineered when using a PLA oligomer with a molar mass lower than around 2 000 g mol−1, while a PCL oligomer with the latter molar mass led to a porous structure with a very broad pore size distribution.

Pore volumes of porous networks were obtained through mass swelling ratio in water and led to the determination of apparent density and porosity ratio values (Table 8). Porosity ratios were all around 50 vol% which matched the expected values, taking into account the quantitative extraction of linear oligoesters from semi-IPNs with a 50/50 wt% oligoester/PMMA composition. As also seen in SEM micrographs, this strongly indicated that the porous network structures were constituted of interconnected open pores through which a fluid could circulate. Moreover when wrapped in a polyethylene film, the porous samples did float in a low-density solvent like n-pentane (d20 °C = 0.624), thus confirming apparent density values lower than or equal to values around 0.6.
Table 8

Pore volume (Vpore), apparent density (dapp), and porosity ratio (P) of resulting porous networks after extraction of semi-IPNs or hydrolysis of IPNs

Oligoester nature and molar mass

Semi-IPNs

IPNs

Vpore (cm3 g−1)

d app

P

Vpore (cm3 g−1)

d app

P

PLA 570 g mol−1

0.87

0.59

0.51

0.42

0.80

0.33

PLA 1700 g mol−1

0.89

0.58

0.52

0.55

0.72

0.40

PLA 3700 g mol−1

0.86

0.59

0.51

0.54

0.72

0.40

PCL 560 g mol−1

0.84

0.59

0.50

0.42

0.79

0.34

PCL 2100 g mol−1

0.83

0.60

0.50

0.60

0.69

0.42

3.4 Engineering Nanoporous Networks from IPNs

Porous methacrylic networks were obtained by hydrolysis of the polyester sub-network from polyester/PMMA (50/50 wt%) IPNs (Fig. 4) [28, 40, 41]. The degradation was conducted using a 50/50 vol% mixture of methylamine (pH 13.6) and ethanol based on the contrasted hydrolytic degradability of aliphatic polyesters and PMMA in these conditions. In order to avoid the collapse of the residual porous structures, the hydrolysis was performed at an intermediate temperature (60 °C) between the Tg of polyester single networks and that of PMMA single network. The degradation temperature was also chosen to ensure an efficient degradation of the polyester partner. The role of ethanol was dual: (1) increasing the hydrophilic behaviour of the methacrylic sub-network to allow a good diffusion of the degradation medium into the IPN structure [45, 46], and (2) increasing the diffusion rate of the degradation products outside the network as ethanol is a good solvent of oligoesters. With these conditions, PLA- and PCL-based single networks were completely hydrolyzed after 5 min and 3.5 h, respectively, while no mass loss was detected for 10 mol% DUDMA-based single networks. Table 5 gives the mass losses obtained after hydrolysis of IPNs. Regardless of the oligoester precursor nature or molar mass, mass loss values close to 50 wt% were determined, which demonstrated a nearly quantitative degradation of the oligoester-derivatized sub-network from polyester/PMMA (50/50 wt%) IPNs. Moreover, the total disappearance of the urethane and carbonyl groups bands associated with the polyester sub-network in the FTIR spectra of the hydrolyzed IPNs confirmed the latter assertion and the methacrylic character of the residual structure. Nevertheless, in comparison to the corresponding PMMA single network, the residual methacrylic networks, after IPN hydrolysis, exhibited a low intensity band from 3700 to 2300 cm−1 and a band at 1660 cm−1. This showed the appearance of carboxylic acid groups due to the partial hydrolysis of ester groups of the PMMA sub-network. Furthermore, the Tg,onset values of IPNs after hydrolysis (see Table 6) were higher than that of the corresponding PMMA single network (i.e., 118 °C). As no mass loss was detected for this single network, it was possible to hypothesize that the partial hydrolysis of PMMA sub-network was essentially due to the hydrolysis of ester side groups in the PMMA sub-chains without degrading the main chain and cross-linking points.

The morphologies of typical IPNs after hydrolysis were examined by SEM (Fig. 7). It was difficult to obtain images with a good resolution without degrading the materials during the SEM analysis. Nevertheless, it is noteworthy that the residual networks typically exhibited nanoporous structures with pore sizes smaller than 100 nm. With such materials, the thermoporometry technique was very useful to determine pore sizes and the pore size distributions. The pore size ranges are reported in Table 7, and pore size distributions are shown in Fig. 6b. Pore sizes were always smaller than 150 nm, and this result mirrored the observations made from visual aspects of IPN precursors. As expected, compared to semi-IPN analogues, the pore diameters were dramatically decreased due to the good chain interpenetration of both polyester and PMMA sub-networks in the interlocking framework of IPN precursors. Moreover, as already hypothesized from DSC analysis, the pore size increased with the oligoester precursor molar mass, but no differences were noticed between PLA- and PCL-based IPN precursors. This meant that domain sizes in IPNs were larger when the cross-linking density of the polyester sub-network was smaller, regardless of the miscibility between PLA/PCL and PMMA. This result was in a good agreement with literature reports [26, 47, 48]. As a matter of fact, when synthesizing IPNs by the sequential method, the network first formed constitutes the most continuous phase, and its cross-linking density generally determines the final morphology of the system, while the cross-linking density of the second network has no significant effect.
Fig. 7

SEM micrographs of porous networks derived from partial hydrolysis of polyester/PMMA (50/50 wt%) IPNs:PLA-based systems: a 1700 g mol−1, b 560 g mol−1; PCL-based systems: c 2100 g mol−1

Pore volumes of nanoporous networks were obtained through mass swelling measurements in water and led to the determination of the apparent density and the porosity ratio values (Table 8). Porosity ratios ranged from 33 to 42%, which was lower than expected values for a quantitative hydrolysis of polyester sub-network from polyester/PMMA (50/50 wt%) IPNs. This was probably due to a partial collapse of the resulting porous structures. Porosity ratio increased with the oligoester precursor molar mass, due to larger pore sizes.

4 Conclusions

This contribution highlights the effectiveness and versatility of using oligoester-derivatized semi-IPNs and IPNs as nanostructured precursors to porous networks with tunable porosity. We clearly demonstrated that the choice of the precursors with defined compatibility is of paramount significance in the preparation of porous materials with controlled porosity scale. Miscellaneous PLA- or PCL-containing methacrylic (semi-)IPNs were indeed prepared as model systems from corresponding oligomers with different molar masses, and the pore sizes of related porous materials could be correlated with the polyester domain sizes inferred from the physico-chemical analysis of the (semi-)IPN precursors. Even though the quantitative extraction of uncrosslinked oligoesters from semi-IPNs generally led to the formation of macroporous networks, the semi-IPN approach constituted a straightforward and effective route toward nanoporous networks provided that a high miscibility between both partners in semi-IPN precursors was reached, i.e. when using the lower molar mass oligoester and preferably PLA. Alternatively, the complete hydrolysis of the polyester sub-network from IPNs afforded more versatility, since it was possible to generate nanoporous networks, regardless of the molecular features of the oligoesters. The nanoporosity could be attributed to the good degree of chain interpenetration of both sub-networks in IPN precursors due to their peculiar interlocking framework. Still, pore sizes in nanoporous frameworks increased with increasing oligoester molar masses, as the extent of microphase separation in IPNs was larger when the crosslinking density of polyester sub-network decreased.

Such complementary routes involving semi-IPN and IPN systems may provide a straightforward and facile means to tune the morphology associated with (nano-)porous polymeric materials. Moreover, the functionalization of the pore surface through the initial incorporation of functional monomers in the (semi-)IPNs precursors may allow for the development of a large variety of applications mainly expected in the areas of separation techniques (chromatographic supports, selective membranes) and chemistry in confined media (nanoreactors, catalytic supports).

Notes

Acknowledgements

The “Région Ile-de-France” is gratefully acknowledged for financial support through SESAME projects allowing for the purchase of SEM equipment. The authors are indebted to late Prof. Ph. Guérin, Prof. F. Lauprêtre and Dr. S. Boileau for fruitful discussions in the field of chemistry and physico-chemistry of (semi-)IPNs.

Compliance with Ethical Standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

References

  1. 1.
    Buchmeiser MR (2001) Angew Chem Int Ed 40:3795CrossRefGoogle Scholar
  2. 2.
    Sykora D, Peters EC, Svec F, Fréchet JMJ (2000) Macromol Mater Eng 275:42CrossRefGoogle Scholar
  3. 3.
    Hentze HP, Antonietti M (2002) Rev Mol Biotechnol 90:27CrossRefGoogle Scholar
  4. 4.
    Dauben M, Reichert KH, Huang P, Fock J (1996) Polymer 37:2827CrossRefGoogle Scholar
  5. 5.
    Allender CJ, Richardson C, Woodhouse B, Heard CM, Brain KR (2000) Int J Pharm 195:39CrossRefGoogle Scholar
  6. 6.
    Byrne ME, Oral E, Hilt JZ, Peppas NA (2002) Polym Adv Technol 13:798CrossRefGoogle Scholar
  7. 7.
    Cameron NR (2004) Polymer 46:1439CrossRefGoogle Scholar
  8. 8.
    Poupart R, Benlahoues A, Le Droumaguet B, Grande D (2017) ACS Appl Mater Interfaces 9:31279CrossRefGoogle Scholar
  9. 9.
    Mezhoud S, Paljevac M, Koler A, Le Droumaguet B, Grande D, Krajnc P (2018) React Funct Polym 132:51CrossRefGoogle Scholar
  10. 10.
    Balaji R, Boileau S, Guérin P, Grande D (2004) Polym News 29:205CrossRefGoogle Scholar
  11. 11.
    Wu D, Xu F, Sun B, Fu R, He H, Matyjaszewski K (2012) Chem Rev 112:3959CrossRefGoogle Scholar
  12. 12.
    Kayaman-Apohan N, Baysal BM (2001) Macromol Chem Phys 202:1182CrossRefGoogle Scholar
  13. 13.
    Widmaier JM, Sperling LH (1982) Macromolecules 15:625CrossRefGoogle Scholar
  14. 14.
    Widmaier JM, Sperling LH (1984) Br Polym J 16:46CrossRefGoogle Scholar
  15. 15.
    Du Prez F, Goethals EJ (1995) Macromol Chem Phys 196:903CrossRefGoogle Scholar
  16. 16.
    De Clercq RR, Goethals EJ (1992) Macromolecules 25:1109CrossRefGoogle Scholar
  17. 17.
    Hu J, Pompe G, Schulze U, Pionteck J (1998) Polym Adv Technol 9:746CrossRefGoogle Scholar
  18. 18.
    Hu J, Schulze U, Pionteck J (1999) Polymer 40:5279CrossRefGoogle Scholar
  19. 19.
    Pionteck J, Hu J, Schulze U (2003) J Appl Polym Sci 89:1976CrossRefGoogle Scholar
  20. 20.
    Sperling LH (1981) Interpenetrating polymer networks and related materials. Plenum Press, New YorkCrossRefGoogle Scholar
  21. 21.
    Sperling LH (1988) Comprehensive polymer science, vol 6. Pergamon Press, New York, pp 423–436Google Scholar
  22. 22.
    Sperling LH, Mishra V (1996) Polym Adv Technol 7:197CrossRefGoogle Scholar
  23. 23.
    Kim SC, Sperling LH (1997) IPNs around the world: science and engineering. Wiley, ChichesterGoogle Scholar
  24. 24.
    Athawale VD, Kolekar SL, Raut SS (2003) J Macromol Sci Part C Polym Rev C43:1CrossRefGoogle Scholar
  25. 25.
    Thomas S, Grande D, Cvelbar U, Raju KVSN, Narayan R, Thomas S, Akhina H (2016) Micro- and nano-structured interpenetrating polymer networks: from design to applications. Wiley, HobokenCrossRefGoogle Scholar
  26. 26.
    Dean K, Cook WD (2002) Macromolecules 35:7942CrossRefGoogle Scholar
  27. 27.
    Rohman G, Grande D, Lauprêtre F, Boileau S, Guérin P (2005) Macromolecules 38:7274CrossRefGoogle Scholar
  28. 28.
    Rohman G, Lauprêtre F, Boileau S, Guérin P, Grande D (2007) Polymer 48:7017CrossRefGoogle Scholar
  29. 29.
    Bachari A, Bélorgey G, Hélary G, Sauvet G (1995) Macromol Chem Phys 196:411CrossRefGoogle Scholar
  30. 30.
    Brun M, Lallemand A, Quinson JF, Eyraud C (1977) Thermochim Acta 21:59CrossRefGoogle Scholar
  31. 31.
    Iza M, Woerly S, Danumah C, Kaliaguine S, Bousmina M (2000) Polymer 41:5885CrossRefGoogle Scholar
  32. 32.
    Hay JN, Laity PR (2000) Polymer 41:6171CrossRefGoogle Scholar
  33. 33.
    Rabelo D, Coutinho FMB (1993) Polym Bull 30:725CrossRefGoogle Scholar
  34. 34.
    Okay O (2000) Prog Polym Sci 25:711CrossRefGoogle Scholar
  35. 35.
    Eguiburu JL, Iruin JJ, Fernandez-Berridi MJ, San Roman J (1998) Polymer 39:6891CrossRefGoogle Scholar
  36. 36.
    Klein PG, Ebdon JR, Hourston DJ (1988) Polymer 29:1079CrossRefGoogle Scholar
  37. 37.
    Grulke EA (1999) In: Brandrup J, Immergut EH, Grulke EA (eds) Polymer handbook, 4th edn. Wiley-Interscience, New York, pp 7–679Google Scholar
  38. 38.
    Coleman MM, Serman CJ, Bhagwagar DE, Painter PC (1990) Polymer 31:1187CrossRefGoogle Scholar
  39. 39.
    Van Krevelen DW (1997) Properties of polymers. Elsevier, AmsterdamCrossRefGoogle Scholar
  40. 40.
    Majdoub R, Dirany M, Benzina M, Grande D (2013) Matériaux Tech 101:405CrossRefGoogle Scholar
  41. 41.
    Grande D, Beurroies I, Denoyel R (2010) Macromol Symp 291:168CrossRefGoogle Scholar
  42. 42.
    Grande D, Rohman G, Millot MC (2008) Polym Bull 61:129CrossRefGoogle Scholar
  43. 43.
    Lav TX, Carbonnier B, Guerrouache M, Grande D (2010) Polymer 51:5890CrossRefGoogle Scholar
  44. 44.
    Lav TX, Grande D, Gaillet C, Guerrouache M, Carbonnier B (2012) Macromol Chem Phys 213:64CrossRefGoogle Scholar
  45. 45.
    Yan J, Wang XH, Chen J (2000) J Appl Polym Sci 75:536CrossRefGoogle Scholar
  46. 46.
    Wei J, Bai XY, Yan J (2003) Macromolecules 36:4960CrossRefGoogle Scholar
  47. 47.
    Donatelli AA, Sperling LH, Thomas DA (1976) Macromolecules 9:671CrossRefGoogle Scholar
  48. 48.
    Hourston DJ, Schäfer FU (1996) Polymer 37:3521CrossRefGoogle Scholar

Copyright information

© The Tunisian Chemical Society and Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Institut de Chimie et des Matériaux Paris-Est (ICMPE), UMR 7182 CNRS-Université Paris-Est CréteilThiaisFrance
  2. 2.Laboratoire de Chimie, Structures, Propriétés de Biomatériaux et d’Agents Thérapeutiques (CSPBAT), UMR 7244 CNRS-Université Paris-NordBobigny CedexFrance

Personalised recommendations