Highly Cross-linked Epoxy Coating for Barring Organophosphate Chemical Warfare Agent Permeation.

Chemical warfare agents (CWAs) can be absorbed in polymeric coatings through absorption and permeation, thus presenting a lethal touch and vapor hazards to people. Developing a highly impermeable polymer coating against CWAs, especially against organophosphate CWAs (OPs), is challenging and desirable. Herein, fluorinated epoxy (F-EP) and epoxy (EP) coatings with different cross-link densities were prepared to resist OPs. The effects of the polymer coating structure, including cross-link density, chemical composition and free volume, on the chemical resistance to dimethyl methylphosphonate (DMMP, Soman simulant) were investigated in detail. Meanwhile, the chemical resistance to Soman and VX was examined. The results reveal that the cross-link density is a critical factor in determining the chemical resistance of the coatings. Highly cross-linked EP and F-EP coatings with dense and solid cross-linked networks can fully bar DMMP and OPs permeation during the test time. At low or medium cross-link densities, the EP coating with a lower retention of DMMP exhibited a higher resistance than the F-EP coating due to the lower interaction with DMMP and smaller free-volume holes and lower relative fractional free volume. These results suggest that increasing the cross-link density is a reasonable approach to control the chemical resistance of polymer networks against OPs.

Introduction

Although the usage of chemical warfare agents (CWAs) is explicitly prohibited internationally, the occasional use of CWAs in recent terrorist events provides a reminder that the threat of CWAs still exists.[1−5] CWAs create an immediate threat to creatures upon release, but they also pose a threat via absorption on and permeation into surfaces and could not be decontaminated effectively, creating lethal off-gassing and contact hazards over extended periods of time.[6−10] Polymer coatings in particular, such as those found on vehicles, equipment and buildings, provide a sufficient sink for CWAs uptake.[8−11] So, developing a fully impermeable polymer coating for protection from CWAs is highly desirable.[12−18]

The main reason for the permeation of CWAs into polymer coating can be attributed to the absorption, dissolution and diffusion behaviors of CWAs.[11,19,20] Firstly, like other small molecules, CWAs molecules absorb on the surface of the polymer coating, then gradually permeate into the interior of the polymer due to the thermal movement of the molecules, weakening the interaction force between the macromolecules, and then cause a swelling phenomenon in macroscopic volume expansion.[21−25] In addition, the interaction derived from the physical properties and chemical structure between the CWAs and polymers can also affect the permeation behaviors.[21,26] Organophosphate chemical warfare agents (OPs, nerve agents), such as Soman, VX, and Sarin, are highly toxic organophosphate compounds, which could effectively inhibit acetylcholinesterase (AchE) and lead to nervous system dysfunction. OPs easily spread on the coating surface.[19,27] On the one hand, as a kind of H-bond acceptor, OPs have a strong interaction with many chemical groups such as N–H,[26] the hexafluoroisopropanol (HFIP) group,[28] and other H-bond donors due to H-bond interaction between these groups and phosphate esters.[29] On the other hand, working as a halogen bond (XB) acceptor, the oxygen of the phosphate ester of OPs also has strong interaction with halogens (F, Cl, Br, I) and other XB donors.[30,31] As a result, once they permeate into the polymer coating, it is difficult to decontaminate them. These facts represent a considerable challenge and demand for developing an impermeable polymer coating against OPs. To our knowledge, there are only few reports on protective polymer coating against OPs.[32]

Epoxy resin (EP) is ubiquitously used in various high-performance materials, which is attributed to its ability to form three-dimensional cross-linked networks upon reaction with suitable curing agents.[33] Cross-linking is an effective way to improve the properties of polymers by adjusting their microstructures such as free volume, glass transition temperature, pack density, and the mobility of polymer chains. A properly cross-linked structure determines the properties of the EP polymer such as high mechanical property[34,35] and chemical resistance property.[36,37] For linear polymers, the loosening and untangling of macromolecular chains are caused by swelling, resulting in their being penetrated and dissolved. However, the swelling equilibrium will occur when swelling to a certain extent for cross-linked polymers, and they will not be penetrated and dissolved further.[38,39] The higher the cross-link density, the smaller the degree of swelling when the material reaches the swell equilibrium, and the lower the corresponding permeability. Cross-links are reported to restrict plasticized Brownian motions that assist in fluid uptake behavior.[40]

Herein, fluorinated epoxy (F-EP) and epoxy (EP) coatings with different cross-link densities were prepared by curing a diglycidyl ether of hexafluorobisphenol A and a diglycidyl ether of bisphenol-A with diethylenetriamine (DETA), respectively, and characterized by Fourier transform infrared (FT-IR), NMR, and dynamic mechanical analysis (DMA). The influence of the coating structure on chemical resistance to dimethyl methylphosphonate (DMMP) was investigated in detail by gas chromatography (GC), scanning electron microscopy (SEM), FT-IR, and positron annihilation lifetime spectroscopy (PALS) measurements. The interactions of the F-EP/EP coating surface with DMMP were determined using a quartz crystal microbalance with dissipation monitoring (QCM-D). The protective performance of the prepared polymer coatings against VX and Soman was also examined. The EP and F-EP coatings with a high cross-link density displayed excellent chemical resistance against OPs.

Materials and Methods

Chemicals and Materials

Hexafluorobisphenol A (BPAF, 98.0%), epichlorohydrin (ECH, analytical grade), and diethylenetriamine (DETA, 99.0%) were purchased from Aladdin Reagent Co. Ltd. N,N-Dimethylformamide (DMF, 99.8%) and dimethyl methylphosphonate (DMMP, 98.0%) were obtained from Adamas-β Reagent Co. Ltd. The diglycidyl ether of bisphenol-A type epoxy (DGEBA, E-42) was purchased from Macklin Reagent Co. Ltd. Sodium hydroxide (NaOH, 97.0%) was supplied by Energy Chemical. We bought anhydrous ethanol (EtOH, 99.8%), hydrochloric acid (HCl, 36.0%), acetone (99.5%), and dichloromethane (DCM, 99.5%) from Greagent. Deionized water (H2O) was fabricated in our lab. All of these reagents were directly used as received without further purification.

Synthesis and Characterization of DGEFBA

The diglycidyl ether of hexafluorobisphenol A type epoxy (DGEFBA) was prepared by the condensation polymerization of BPAF and ECH. BPAF and ECH were added into a round-bottomed flask (250 mL) at a molar ratio of 1:10, magnetically stirred, and heated to 55 °C within 10 min to completely dissolve the reaction mixture. After adding a little deionized water, 8 g of NaOH was added in four batches. After refluxing at 95 °C for 5 h, the mixture was filtered and rotary evaporated to remove the excess ECH; the obtained light-yellow viscous liquid was dried at 80 °C in vacuum for 12 h, resulting in DGEFBA.

The epoxy value of DGEFBA was determined according to the Chinese standard GB/T 1677–2008 and by the hydrochloric acid-acetone method using eq .where C (mol/L) is the equivalent concentration of the NaOH-EtOH standard solution, V0 (mL) is the volume of the NaOH-EtOH standard solution consumed in the blank experiments, V1 (mL) represents the volume of the NaOH-EtOH standard solution consumed in the parallel experiments, and m (g) represents the mass of DGEFBA. The epoxy value of DGEFBA was determined to be 0.40 mol/100 g. DGEBA was purchased from Macklin Reagent Co. Ltd, and its epoxy value was 0.41 mol/100 g.

Fabrication of F-EP and EP

F-EP coatings with different cross-link densities were obtained by curing of DGEFBA with different amounts of curing agent (DETA), as summarized in Table . Firstly, DGEFBA was dissolved in DMF (0.2 g/mL), and DETA was added by stirring. The mixture was then poured into a Teflon mold and cured at 70 °C for 12 h, followed by 100 °C for 2 h to achieve complete curing. After naturally cooling down to room temperature, the coating was peeled off, resulting in F-EP. Using the same procedure, EP coatings with different cross-link densities were prepared by curing of DGEBA with DETA, as summarized in Table . The obtained samples were named F-EP90/EP90∼F-EP200/EP200. The number represents the weight of the DETA in the curing reaction.

Table 1

Summary of Mass of DGEFBA and DETA Used for Different Samples

samples	DGEFBA/DGEBA (g)	DETA (mg)
F-EP90/EP90	2	90
F-EP100/EP100	2	100
F-EP120/EP120	2	120
F-EP140/EP140	2	140
F-EP160/EP140	2	160
F-EP200/EP200	2	200

Chemical Resistance of the Coating to DMMP, Soman, and VX

The chemical resistance of the coating to DMMP was evaluated by measuring the amount of DMMP remaining in and penetrating through the coating. The experiments were performed as follows: three small pieces of filter paper were placed under the peeled coating (0.20 ± 0.05 mm), which were used to absorb DMMP penetrating the coating. Then, DMMP (20 mg) was dropped (see as Figure S1) on the coating surface (2 cm × 2 cm) at 25 °C and a relative humidity of 55%. After 3 h, the surface of the coating was wiped lightly with a cotton ball to remove the remaining DMMP on the surface. Then, the coating and the filter paper were soaked in 10 mL of DCM for 24 h, respectively, and ultrasonicated for 10 min. After filtering, the extraction solvent was analyzed by gas chromatography (GC) to determine the amount of DMMP. The retention of DMMP in the coating and penetration of DMMP through the coating were calculated using eqs and 3.(41)where Wt is the total amount of agents (mg) dropped on the coating, Wr is the mass of DMMP (mg) absorbed by the coating, and Wp is the mass of DMMP penetrating through the coating as analyzed by GC. The chemical resistance test of the coating to Soman and VX was conducted using the same procedure, except that the amount of tested Soman and VX dropped on the coating was 1 mg, respectively.

Characterization

Dynamic Mechanical Analysis (DMA)

DMA was performed using a dynamic mechanical analyzer (TA Instrument DMA 850) to measure the storage modulus (Ge) and evaluate the cross-link density of the F-EP and EP coatings at a heating rate of 5 °C/min from −10 to 250 °C, a frequency of 1.0 Hz, and a strain amplitude of 0.1%.

Gas Chromatography (GC)

GC was performed using Shimadzu GC 2010 to identify the DMMP and then quantitatively analyze it using the standard curve constructed by plotting the DMMP concentrations against their peak area. The tests for Soman and VX were conducted using the same procedure.

Positron Annihilation Lifetime Spectroscopy (PALS) Experiments

PALS experiments were performed using a conventional fast-fast coincidence spectrometer with a time resolution of 210 ps (in FWHM) at 25 °C in air at the University of Science and Technology of China (USTC) to investigate the free volume (Vf) of coatings with different cross-link densities. The positron source prepared by sandwiching the 22Na radioactive source (activity of ∼500 kBq) between two Kapton polyimide foils (10 mm × 10 mm × 7.5 μm) was then sandwiched between two identical samples’ stacking of multiple pieces of membranes with a size of 10 mm × 10 mm × 2 mm.

Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D)

QCM-D measurements were performed using QSense Pro (Finland/Biolin Scientific AB). The DGEFBA and DGEBA cured with DETA dissolved in DMF (3 mg/mL) were coated by the drop-coating method (20 μL) on the surface of quartz crystal Au electrode sensors (QSX 301, 4.95 ± 0.05 MHz, 14 mm diameter) purchased from Finland/Biolin Scientific AB. Subsequently, the coated sensors were dried at 70 °C for 12 h, followed by 100 °C for 2 h, resulting in a stable coating, and were then placed in the measurement chamber. The fundamental resonant frequency of the quartz crystal and its overtones (n) were excited by applying an AC voltage. The frequency shifts can be monitored by QCM-D because of the piezoelectric effect of a quartz crystal after adsorbing DMMP.[42] The sensors were tested with DMMP at 25 °C and a flow rate of 100 μL/min, while DMMP adsorption was measured after the establishment of a stable baseline.

Other Characterizations

The 1H and 19F NMR of DGEFBA and DGEBA were carried out using Bruker 500 NMR (AVANCE III HD 500). Fourier transform infrared (FT-IR) spectrometry of the coatings in the scanning range of 4000–650 cm–1 was characterized using Spectrum 100 (Perkin Elmer). Using the sessile drop method, the contact angles (CA) of the coatings with the liquid were determined using a DSA-30 (Kruss Company) optical contact-angle device by injecting 5 μL of DMMP onto the coatings at 25 °C. The surface and cross-sectional SEM images of the membranes before and after being contaminated were taken using a field emission scanning electron microscope (Nova NanoSEM 450).

Results and Discussion

Synthesis of F-EP and EP with Different Cross-link Densities

The cross-linked fluorinated epoxy network (F-EP) was synthesized by two steps as shown in Scheme . The first one is the synthesis of the fluorinated epoxy resin DGEFBA, followed by curing of DGEFBA with the curing agent of an aliphatic triamine (DETA). DETA was chosen due to its relatively low curing temperature and the ability of reaching high cross-link densities. DGEFBA was characterized by 1H NMR (Figure a),19F NMR (Figure b), and FT-IR (Figure c). The 1H NMR peaks of DGEFBA at 2.86 and 2.74 ppm in Figure a were attributed to the two hydrogens of the methylene group on the epoxy group, and the peak at 3.36 ppm was assigned to the methine hydrogen on the epoxy group. The chemical shifts of the two hydrogens on the methylene group adjacent to the epoxy group were at 3.92 and 4.38 ppm, respectively, and the chemical shifts at 7.29 and 7.08 ppm were attributed to the adjacent proton peaks on the two aromatic rings. The 19F NMR peak of −63.76 ppm was attributed to trifluoromethyl (−CF3). The characteristic IR band of DGEFBA at 915 cm–1 was clearly observed as shown in Figure c, which can be attributed to the stretching vibration of the epoxy group, while the peak intensity was significantly weakened after curing. Moreover, an obvious hydroxyl stretching vibration peak appeared at 3400 cm–1 after curing, and the band of C-N between 3300 and 3500 cm–1 coincided with the hydroxyl peak. The above results confirmed that the F-EP was successfully prepared. The EP analogue was prepared by curing of commercial DGEBA with DETA, and was characterized by FT-IR (Figure d). The FT-IR spectrum of EP is almost the same as that of F-EP, except for the −CF3 peak at 1100–1200 cm–1.

Scheme 1

Formation of a Cross-Linked Fluorinated Epoxy Network. (a) Synthesis of DGEFBA from BPAF and ECH; (b) Chemical Reactions between DGEFBA and the Curing Agent DETA

Figure 1

(a) 1H NMR and (b) 19F NMR spectra of DGEFBA; FT-IR spectra of (c) DGEFBA, F-EP and (d) DGEBA, EP.

The cross-link density can be obtained based on the standard rubber elasticity theory, which assumes that the van der Waals and other interactions between molecules have been broken above the glass transition temperature (Tg) and the storage modulus (Ge) is proportional to the number of cross-links present in the network.[43] In this study, the cross-link density was assessed by determining the number average molecular weight between the cross-links (Mc), which is inversely proportional to μ and can be calculated by Ge using eq .[44,45]where ρ (g/cm3) represents the epoxy density and Ge is the rubber storage modulus in the DMA test. Then, the cross-link density (μ) was estimated empirically using eq , in which N (6.02 × 1023) represents Avogadro’s number[44]

Figure a shows the DMA test curves of the F-EP coatings prepared by using six different amounts of the curing agent DETA. As shown in Figure , the glass transition of each sample can be seen in the peak of the tan (delta) plots with temperature, and the polymer transits from the rigid glass state to the flexible rubbery state. The DMA curves gradually decrease from a high storage modulus to the rubber plateau with a low storage modulus (Ge value). With the increase of the cross-link density, the molecular chains are connected into a dense three-dimensional network, which restricts the movement of the molecular chains, resulting in the increase of Ge and Tg. The Ge, Tg, Mc, and μ values calculated from eqs and 5 for the six F-EP coatings are summarized in Table . When the amount of DETA increased from 90 to 200 mg, the cross-link density increased about 12 times from 1.3 × 1020 to 15.7 × 1020 mol/cm3, and the Tg of the coatings increased from 95 to 158 °C, indicating the formation of very dense and solid cross-linking networks. Meanwhile, the Ge, Mc, μ, and Tg values of the EP coatings were determined and calculated by the same method (Figure b and Table ). As shown in Table , when the dosage of DETA for EP is the same as that for F-EP, the cross-link density and Tg of EP are basically the same as those of F-EP. Consequently, based on the above results, the simplified schematic for the three kinds of F-EP and EP cross-link networks with low, medium, and high cross-link densities are shown in Figure .

Figure 2

DMA test curves of (a) F-EP and (b) EP.

Table 2

Ge, Mc, μ and Tg Values of F-EP and EP

samples	Ge (MPa)	Mc (g/mol)	μ × 1020 (mol/cm3)	Tg (°C)
F-EP90/EP90	1.2/1.2	3023/3023	1.3/1.3	95/90
F-EP100/EP100	1.5/1.5	1664/1664	2.4/2.4	103/100
F-EP120/EP120	3.9/3.8	495/505	8.1/8.0	136/133
F-EP140/EP140	7.2/7.0	341/347	11.8/11.6	146/144
F-EP160/EP160	10.0/9.0	293/307	13.7/13.1	155/152
F-EP200/EP200	14.0/13.0	255/263	15.7/15.3	158/153

Figure 3

Simplified schematic of F-EP or EP networks for different cross-link densities. (a)–(c) correspond to low, medium, and high cross-link densities respectively.

Chemical Resistance of F-EP or EP Coatings to DMMP, VX, and Soman

Firstly, the chemical resistance of F-EP or EP coatings to DMMP is qualitatively evaluated. The wettability of DMMP can influence its adsorption on the coating surface. Figure a shows the contact angle (CA) of DMMP on the F-EP and EP coatings. All CAs of DMMP on the F-EP and EP coatings are less than 90°, indicating that DMMP wets and spreads on the F-EP and EP surfaces. Probably due to the CF3 groups in F-EP, the DMMP CAs on F-EP are slightly higher than those on the EP coatings. Figure b,c represents the FT-IR spectra of F-EP120 and EP120 before and after contamination by DMMP, respectively. Before the FT-IR test, the coatings after contamination by DMMP were placed in a 70 °C oven for 12 h to remove the completely residual DMMP. There was no obvious peak change in the spectra, which indicated that no chemical reaction happened between the coatings and DMMP. Figure displays the pictures of the F-EP and EP before and after contamination by DMMP for 3 h taken by a digital camera. There is obvious deformation and swelling for F-EP100 and EP100. However, as the cross-link densities of the F-EP and EP coatings increased, no deformation and swelling were observed for the other F-EP/EP samples. Moreover, the surface and cross-sectional SEM images (Figure S2) of all of the samples revealed that there were no conspicuous destructions such as holes for the coatings after being contaminated by DMMP.

Figure 4

(a) Contact angle (CA) for DMMP on F-EP and EP by injecting 5 μL of DMMP onto the coatings at 25 °C; FT-IR spectra of (b) F-EP120 and (c) EP120 before and after contamination by DMMP.

Figure 5

Pictures of F-EP and EP before and after contamination by DMMP for 3 h taken by a digital camera.

Furthermore, the chemical resistance of F-EP and EP coatings to DMMP was quantitatively evaluated for the DMMP remaining in the coatings and penetrating through the coatings. As shown in Figure a,b, both F-EP and EP demonstrated excellent chemical resistance to DMMP. No DMMP penetrated through the coatings within 3 h. The retention of the DMMP in the coatings gradually decreased to zero when the cross-link density of the coatings increased to 13.7 × 1020 mol/cm3 for F-EP160 and 11.6 × 1020 mol/cm3 for EP140. It means that the F-EP and EP coatings with relatively high cross-link densities are able to completely bar the DMMP during the test time. The dense and solid cross-linked networks fully restrict DMMP from permeating and penetrating the polymer coatings. At low and medium cross-link densities, EP exhibited much lower retention of DMMP than F-EP with the same cross-link density. When the cross-link density was 1.33 × 1020 mol/cm3, the retention of DMMP in F-EP90 was 51.14%, while the DMMP retention in EP90 was only 9.84%. Also, when the retention of DMMP in EP140 with a cross-link density of 11.6 × 1020 mol/cm3 reduced to zero, there was still 5.74% DMMP in the F-EP140. The reasons will be discussed later.

Figure 6

Retention and penetration of DMMP to (a) F-EP and (b) EP with different cross-link densities. Retention and penetration of DMMP to (c) F-EP and (d) EP after being contaminated for different hours at 25 °C and a relative humidity of 55%. The thickness of the coatings was 0.20 ± 0.05 mm.

We further explored the kinetic plots of DMMP permeation in F-EP and EP. As shown in Figure c,d, for F-EP200, EP140, and EP200, no DMMP permeated into the coatings during the test time of 3 h. For F-EP140, DMMP started to permeate into the coating after 2 h. As for F-EP100, 25% of the DMMP permeated into the coating after 0.5 h, and the retention further increased to 30% after 2 h, then quickly rose to 51% after 3 h. The trend of DMMP retention in EP100 with the contaminated time is similar to that of F-EP100, but with a much lower retention value. DMMP can easily ingress into the F-EP and EP coatings at low cross-link densities and then swell the coatings, which further caused the quick increase of the permeability rate with time. However, the highly cross-linked coatings (F-EP200 and EP200) were able to resist the DMMP completely during the test time. The cross-link density of the coatings is thus a key factor that affects the chemical resistance to DMMP.

In order to verify the protective performance of the coating against real OPs, the chemical resistance values of F-EP200 and EP200 to VX and Soman were calculated and are listed in Table . Neither retention nor penetration of VX and Soman was detected by GC after the coatings were contaminated for 24 h, which revealed the excellent protective performance of the highly cross-linked F-EP/EP coatings.

Table 3

Chemical Resistance of F-EP and EP Coating to VX and Soman

	Soman		VX
	retention (%)	penetration (%)	retention (%)	penetration (%)
F-EP200	not detected	not detected	not detected	not detected
EP200	not detected	not detected	not detected	not detected

Free Volume of F-EP and EP

Free volume is obtained from the nanoscaled holes randomly distributed in glassy polymers, and it has an important influence on the impermeability of the polymers.[46] Reducing the free-volume hole size is an approach to control the solvent properties for EP networks.[47] Herein, positron annihilation lifetime spectroscopy (PALS) was performed to evaluate the size of the free-volume holes of the polymer coatings.[48] PALS data was analyzed using the LTv9 program, and all spectra were divided into different lifetime components (τ1< τ2 < τ3). The longest lifetime component τ3 generally corresponds to the pick-off annihilation of ortho-positronium (o-Ps), which provides information about the size of the free-volume holes.[49] The average radius (R) of the holes was estimated by a semi-empirical equation according to a spherical infinite potential well model[50]where ΔR (0.1656 nm) is an empirical parameter of the thickness of the electron layer. The average free volume (Vf) of the holes was calculated byThe relative fractional free volume (fr, %) was calculated empirically using eq ,[48] where the intensity (I3) corresponded to τ3 and reflected the number of Vf holes.Figure a,b shows the τ3 and I3 of low, medium, and high cross-linked F-EP samples, while Vf and fr are displayed in Figure c,d. The Vf of the F-EP samples ranged from 90 to 100 Å3 and slightly decreased with the cross-link density. All of the Vf values of the samples are smaller than the calculated van der waals volumes (vdW volume) of DMMP (106 Å3), Soman (171 Å3), and VX (261 Å3). A lower Vf would create a statistically lower probability of chemical transport. The coating with a higher cross-link density will have a lower Vf and higher Tg and rigidity, which would lead to enhanced resistance against DMMP. However, DMMP still can permeate into the low and medium cross-linked coatings even if the coatings have smaller free-volume hole sizes than the calculated vdW volumes of DMMP, as shown in Figure . This is probably because the Vf obtained by PALS is an average value. There are some big free-volume holes in the coating that cannot bar DMMP permeation. Once DMMP enters the coating, it further swells the coating, enlarges the free-volume holes, and then causes a quick increase of permeability. On the other hand, EP100 exhibited a much smaller Vf and a slightly lower fr than F-EP100, indicating that the macromolecular chains of EP100 without the −CF3 groups are arranged much more closely, which should contribute to the improved chemical resistance of EP100 to DMMP than F-EP100.

Figure 7

(a) o-Ps lifetime, (b) o-Ps intensities, (c) free volume, and (d) relative fractional free volume of the F-EP coatings.

Interaction between DMMP and F-EP/EP

Furthermore, the physical interaction of the F-EP and EP surfaces with DMMP was investigated by QCM-D measurements. By monitoring the change of crystal oscillation frequency (F, Hz), the adsorption characteristics of DMMP on the F-EP and EP surfaces were investigated. According to Sauerbrey equation[51] (eq ), the negative-frequency changes are proportional to the adsorption mass deposited (Δm)where C represents the mass sensitivity constant (C = 17.7 ng·cm–2·Hz–1 for a 5 MHz crystal sensor[52]) and ΔF represents the frequency shift of the nth overtone. As shown in Figure a, a negative F with a sharp drop was observed, implying the mass deposition (Figure b) of DMMP onto the F-EP/EP-coated sensor. Compared to EP, the higher ΔF and Δm of the F-EP-coated sensor at adsorption equilibrium revealed that there existed a stronger physical interaction between the F-EP surface and DMMP. It is known that DMMP is a good halogen bond acceptor due to the chemical portion of −P=O and −P–O– bonds.[26] There may exist a halogen bond[31] interaction between the trifluoromethyl group (−CF3) and phosphate esters, probably resulting in a stronger interaction and higher adsorption between DMMP and F-EP. So, because of the lower free-volume hole size and weaker physical interaction with DMMP, EP has an advantage as a polymer coating material resistant to organophosphate compounds over F-EP.

Figure 8

(a) QCM-D response curves for DMMP adsorption on the F-EP- and EP-coated quartz crystal surface; (b) mass changes of sensors (Δm) and QCM-D frequency shifts (ΔF) for DMMP adsorption on F-EP and EP.

Conclusions

The fundamental understanding of the relationship between the structure and anti-OPs property of an epoxy coating has been preliminarily established by varying the cross-link density of epoxy and fluorinated epoxy analogues. The results demonstrated that the anti-CWAs performances of the coatings have a strong dependence on their cross-link density. Highly cross-linked EP and F-EP coatings with dense and solid cross-linked networks can completely bar the DMMP simulant and OPs permeation during the test time. At low or medium cross-link densities, the DMMP gradually ingresses into the coatings even if the coatings have smaller free volume than the van der waals volume of DMMP. Due to a weaker interaction with DMMP, smaller free-volume holes, and a relatively fractional free volume, the EP coating exhibited a higher resistance to DMMP than the F-EP coating. These results may play a constructive role in the design and preparation of highly protective coatings against OPs.