Novel perfluoropolyalkylethers monomers: synthesis and photo-induced cationic polymerization

Graphical abstract Several difunctional oligomers were synthesized by functionalizing perfluoropolyalkylether (PFPAE) chains with different vinyl ethers and epoxides end-groups. Due to their innate synthetic challenges and demanding purification protocols, the PFPAE derivatives were obtained in low yield and with an average functionality lower than 2. However, the functionalized PFPAE oligomers were successful in being used in photo-induced cationic polymerization processes, obtaining transparent and soft films. The influences of the fluorinated chains, and various end-groups on the photopolymerization process were investigated, as well their chemical stability, thermal degradation, and surface properties. Supplementary Information The online version contains supplementary material available at 10.1007/s00396-021-04838-1.


Introduction
Cationic photo-induced polymerization of reactive monomers, such as epoxides and vinyl ethers, is a process that has an established history [1,2]. At the moment, it is commonly used in several industrial fields such as inks, coatings, adhesives, and microelectronic devices thanks to the significant benefits of the process (lack of solvents and of oxygen inhibition, low toxicity, fast, and room temperature operations) and to the important advantages shown by the obtained cured materials (low shrinkage, good mechanical, and adhesion properties).
Having established industrial relevance, cationic photoinduced polymerization utility can be further refined by gaining photocontrol over chain growth [3], using a variety of catalysts and chain-transfer agents [4]. Spatiotemporal control is easily obtained by light and allows for precise regulation of polymer structure, function, and architectural design. Recently, radical photo-controlled (co)polymerization of fluoro-olefins have been reported [5], allowing smooth transformation of CTFE at ambient pressure and room temperature.
For these reasons, PFPAEs have previously been explored containing functionalization with acrylic [23,24], methacrylic [10,[25][26][27][28], or maleimidic [29,30] end-groups to be used in radical photopolymerization. It is well known that this process suffers from oxygen inhibition; this drawback is even more severe for fluorinated products, as they show a high oxygen solubility. Synthesizing fluorinated monomers with functionalities suitable for cationic photoprocesses is therefore attractive. However, up to now few structure are available and have been investigated only as comonomers [31][32][33][34][35][36][37][38]. The synthesis of functional PFPAE derivatives is a demanding target to achieve: PFPAEs are extremely non-polar substances, poor nucleophile, and insoluble in almost all the organic solvents, except for few fluorinated solvents [29,31,32].
In this work, several difunctional oligomers were synthesized by functionalizing PFPAEs chains with vinyl ethers and epoxides end-groups: their general structure is RG-S-PFPAE-S-RG where RG represents the reactive function and S is a non-fluorinated alkyl spacer.
The functionalized PFPAE oligomers were previously used in copolymerization with hydrogenated vinylethers and epoxides: here, after describing their synthesis, the molecular structures and main physico-chemical properties by multinuclear NMR, their reactivity in photo-induced cationic polymerization was studied, showing the influence of the fluorinated chain, and of the different length of the hydrogenated spacers by means of FTIR spectroscopy and photo DSC. The thermal and the surface properties of the obtained polymers are also presented.

Synthesis of the fluorinated vinyl ethers and epoxides
Materials Oligo(PFPAE) diacyl fluoride was synthesized via the anionic ring-opening polymerization of hexafluoropropoxide (HFPO, generously supplied by Chemours™, USA), following the procedure described elsewhere [39][40][41][42]. Different batches of the synthesized oligo(PFPAE) diacyl fluoride were mixed to be used for the syntheses of vinyl ether derivatives. Oligo(PFPAE) dicarboxylic acid, obtained by hydrolysis of the acyl fluoride, was used for the synthesis of the epoxy derivatives.

Monomer characterization
The structure of the synthesized reagents and PFPAE monomers were determined by NMR spectroscopy at room temperature (25°C). NMR spectra were recorded on a Bruker AVANCE III 400 MHz spectrometer instruments using deuterated chloroform-d, dichloromethane-d 2 , and benzene-d 6 capillaries as internal references for the fluorinated products. The experimental conditions were accomplished by using MestReNova 12.00 operating at 400.13 ( 1 H), 376.46 ( 19 F), 100.62 ( 13 C) MHz. The letters s, d, t, q, quint, sext, and spt stand for singlet, doublet, triplet, quartet, quintuplet, sextet, and septuplet, respectively.
GC-MS spectra were obtained by using an Agilent Technologies 6890N gas-chromatographer (GC), coupled with an Agilent Technologies 7638B series injector. An Agilent Technologies 5975B inert mass spectrometer (MS) was employed with 70 eV electron impact (EI) as the mode of ionization. The GC was equipped with a Zebron ZB-5ms column, 30 m × 0.18 mm id, 0.18 mm df. The detector and the injector temperatures were 200°C and 280°C, respectively. The temperature program started from 50°C with a 2 min hold then the heating rate was 25°C/min until reaching 250°C and holding at 250°C for 2 min. Total pressure was 108 kPa, total flow was 25.9 mL/min, column flow 0.74 mL min −1 , purge flow 3 mL min −1 , linear velocity 38.2 cm s −1 , and a split injection of 30:1. The samples were dil uted in methoxyperfluorobutane (3M's Novec™ HFE-7100) in a GC vial. The written fragments correspond to at least 10% of the highest signal.

General synthesis of PFPAE-vinyl ethers
A solution of PFPAE diacyl fluoride (1 equiv., M n = 4450 g mol −1 ) in 5 mL of dried PFB was added to a 100 mL 3-necked round-bottomed flask and cooled at 0°C. A mixture of triethylamine (2.5 equiv.), DMAP (6 wt.%), and alkyl glycol vinyl ether (5 equiv.) in 25 mL of dry 40:60 DCM:PFB solution was added dropwise to the PFPAE diacyl fluoride/PFB mixture, at 0°C, under continuous stirring. After 30 min, the ice bath was removed, and the reaction mixture was let to stir at room temperature during 24 h. The reaction progress was monitored by thin layer chromatography (TLC): powdered iodine was used as TLC stain and a 5:95 EtOAc/pentane solution as eluent. The conversion of the diacyl fluoride into the corresponding vinyl ether was checked by the appearance of an orange-brown spot on light yellow background, located between the baseline of the TLC plate and the eluent front). The reaction was completed after 24 h, as suggested by gas chromatography-mass spectrometry (GC-MS). Then, the reaction mixture was purified by flash chromatography (5:95 EtOAc/pentane solution used as eluent). After vacuum purging, a transparent oil was obtained as pure product (S1 in the Supporting Information).

Synthesis
, from − 85.48 to − 79.10 (CF 3 and CF 2 of repeat unit). The molecular weight of the product was calculated from the 19 F-NMR spectra, as in S15 of the Supporting Information.

General synthesis of the oxirane precursors
The bromo-alkene was added to a solution of mCPBA in DCM (15 mL). The solution was allowed to stir overnight (Scheme 1). After mixing, the reaction mixture was cool down at 0°C: the mCPBA/benzoic acid precipitate was filtered through a short column, and the solvent was removed under vacuum (×3). The mixture was then washed with 10% aqueous solution of Na 2 SO 4 (×3), with a saturated solution of NaHCO 3 (×3), and with a saturated aqueous solution of LiCl. The crude was dried over Na 2 SO 4 and the solvent traces were then removed under vacuum, to afford the desired known compound as a colorless liquid.

General synthesis of PFPAE-oxiranes
Silver oxide (2 equiv.) in 40:60 acetonitrile:PFB solution (25 mL) was added to a 125 mL 3-necked round-bottomed flask, and placed under N 2 purge without expose the Ag 2 O to the light. PFPAE dicarboxylic acid (1 equiv., M n = 3131 g mol −1 ) was slowly introduced at RT to the 3 necked round-bottomed under continuous stirring. The bromo-alkyl-oxirane (3 equiv.) was added dropwise at RT to the silver carboxylate mixture under continuous stirring. The reaction was complete after 48 h, as suggested by the formation of an AgBr whiteish precipitate and by GC-MS. Then, the reaction mixture was purified by flash chromatography; pure hexane was used first as eluent to remove fluorinated undesired byproduct as HFPO II hydrogen end cap (R F -H). Then, the polarity of the eluent was slowly increased up to 40:60 diethyl ether:hexane. Ceric ammonium molybdate was used as TLC stain. The conversion of the silver carboxylate salt into the corresponding epoxide was checked by the appearance, after a gently heating, of a dark blue spot on light blue background, located between the baseline of the TLC plate and the eluent front. After vacuum purging, a transparent oil was obtained as pure product (S19 in the Supporting Information). PFPAE dicarboxylic acid (M n = 3131 g mol −1 , 1 mmol, 1.57 mL), silver oxide (2 equiv., 2 mmol, 0.462 g), epibromohydrin (3 equiv., 3 mmol, 0.26 mL). Isolated yield: 13%. 1 H-NMR (400 MHz, MHz, benzene-d 6 capillary, 25°C, S20 in the Supporting Information): ). Impurities 6.28-6.15 ppm attributable to the HFPO II hydrogen end cap.

Synthesis of polymers by photoinduced polymerization
To PFPAE monomers was added the 2 wt% of photoinitiator, which is 50 wt% of triphenylsulfonium hexafluorophosphate salt in propylene carbonate, purchased from Sigma Aldrich (Italy). The UV-sensitive mixtures were coated onto a glass substrate, using a wire-wound applicator. The glass substrate used was a microscope slide (pre-cleaned/ready-to-use) purchased from Thermo-Scientific, used as received without any surface treatment.
The samples were irradiated by means of a lamp Dymax ECE 5000, predominately producing UV-A light (440-360 nm) and some amount of UV-B light (320-280 nm) [44], using a light intensity of 150 mW cm −2 for 5 min. Samples with different thickness, going from 100 to 300 μm, were prepared. Light intensity was measured with a UV Power Puck II digital radiometer.
After irradiation, the samples were stored for at least 48 h at room temperature before evaluating their properties.

Polymer characterization
The photopolymerization reaction was monitored by Fourier transform-infrared (FT-IR) analysis and by differential scanning calorimetry (DSC).
Real-time FT-IR spectroscopy analyses were performed using a Nicolet™ iS50 FT-IR spectrometer (Thermo Fisher Scientific). Simultaneously with the FT-IR scan acquisition (i.e., 1 scan/spectrum), thin films (i.e., about 10 μm on a Si wafer as substrate) of the reactive monomeric mixtures were irradiated for 10 min with a UV Hamamatsu LC8 lamp, provided of an optical fiber, having an intensity equal to 100 mW cm −2 . Light intensity was measured with a UV Power Puck II digital radiometer.
Polymerization conversion was followed by monitoring the decrease in the absorbance of the vinyl ether groups at 1620 cm −1 or of the epoxy groups in the region 900-920 cm −1 as a function of irradiation time. The conversion (χ) is given by Eq. 1: where A t is the area of the peak of interest at the end of the irradiation, while A 0 is the area of the peak of interest before the irradiation. FT-IR spectra were also recorded in attenuated total reflectance mode (ATR), employing a Nicolet™ Smart iTX accessory equipped with an ATR diamond crystal, to check the final monomer-to-polymer conversion at the polymer surface after dark-curing (i.e., 48 h at room temperature).
Photo-DSC analyses were performed by means of a Mettler-Toledo DSC1 STARe System, appropriately modified with a Hamamatsu LC8 lamp with two beams, one for the sample side and the other for the reference side, as reported in previous works [45,46]. To ensure equal illumination conditions throughout the sample volume, about 5 mg of each formulation were cured in open aluminum pans. The sample space was flushed with nitrogen. Two scans were performed on each sample to subtract the thermal effect of UV irradiation from the photocuring experiment, each one consisting of 4 min of temperature conditioning, 20 min of irradiation, and 4 min more without UV light. The measurements were carried out at 25 mW cm −2 of intensity and at room temperature, the increase of temperature due to the irradiation of the lamp being less than 1°C at the end of the reaction. The reaction was considered completed when it was no longer possible to detect a change in the heat flux. The total heat of reaction was then calculated by the integration of the area under the exothermic peak [47]. The conversion (χ) over time was calculated by Eq. 2: where dh/dt is the heat flow released by the sample during the photocuring and Δh theor is estimated considering the heat of reaction of 60 kJ mol −1 for the polymerization of the epoxy monomers [48], and the heat of reaction of 84 kJ mol −1 for the polymerization of the vinyl-ether monomers [49]. The insoluble fraction of the cross-linked samples (gel content) was evaluated using the weight loss of the network after a 24-h extraction by 1:1 DCM:PFB solution (expressed in weight fraction) at room temperature (ASTM D2765-16) [50]. The cross-linked fraction was then calculated by Eq. 3: where W t is the weight of dry sample after extraction, while W 0 is the weight of the original sample.
The thermal stability of the polymers was determined by thermogravimetric analysis (TGA) using a Mettler Toledo TGA/SDTA-851 instrument. Approximately 10 mg of the sample were placed in an alumina crucible and heated from room temperature to 800°C under inert atmosphere (N 2 , 60 mL min −1 ), with a heating rate of 10°C min −1 .
Differential scanning calorimetry (DSC) thermograms were recorded using a Mettler-Toledo DSC1 STARe System in the temperatures ranging from − 100 to 50°C using a heat/cool/heat method at a heating and cooling scanning rate of 10°C min −1 , under nitrogen flux. The glass transition temperature (T g ) was determined using the midpoint of the heat capacity jump on the second heating cycle thermogram.
Static contact angle measurements were performed with a FTA 1000C instrument, equipped with a video camera and image analyzer, at room temperature with the sessile drop technique. The probe liquids were water and hexadecane, whose surface tension is 72.1 mN m −1 and 28.1 mN m −1 , respectively. The surface energy was calculated by the Owens-Wendt's geometric mean method [51]. Three to five measurements were performed on each sample, placing the liquid drops in different parts of the sample surface: the mean value and the error were determined.
Dynamic measurements were done by increasing the drop volume in the wetting process (advancing contact angle) and then decreasing it in the de-wetting phase (receding contact angle). The syringe needle remained in the drop during the whole process. In the first wetting step a 3 μL drop was created on the solid surface and then slowly increased in volume. In the second step, the surface was de-wetted and the drop size reduced, the whole cycle was repeated 3 times at 1 μL s −1 , with a delay time of 1 s between each cycle. The contact angle hysteresis (CAH) was calculated as the difference between the measured advancing and receding contact angles.
For the sliding angle measurements [52], the same FTA 1000C apparatus was employed, equipped with a tilting stage. After placing a 20 μL liquid drop on the test surface, the film was tilted at 0.5°s −1 . During the tilting stage, dynamic contact angles were measured, as well as the difference (Δα) between the advancing contact angle and the receding contact angle assessed on the sliding drop. The sliding angle was determined as the angle of inclination at which the drop starts to move and slides across the surface.

Synthesis of the fluorinated monomers
The best synthetic pathways that allowed to obtain PFPAE-based vinyl ethers and epoxides are those described in Scheme 2.
First, the synthesis of telechelic PFPAE oligomers was conducted by anionic ring-opening polymerization of hexafluoropropylene oxide similarly reported by Ding et al. and Hill et al. [39,53]. An alkoxide was prepared for allowing initiation: in our case, CsF in tetraglyme reacted with octafluoroadipoylfluoride; then, the monomer, HFPO, was added for propagation. By chain transfer, termination occurred and the telechelic PFPAE diacylfluoride was obtained. This product was used to prepare vinyl ethers, while it was oxidized to dicarboxylic acid for the subsequent preparation of epoxides [41,42]. T h e i n i t i a t i o n r e a c t i o n b y m e a n s o f t h e octafluorodipoylfluoride is, however, accompanied by an initiation reaction due to HFPO itself, which is followed by a propagation leading to a monofunctional product (Scheme 3).
Therefore, a monofunctional growing chain propagates producing monofunctional products and competes with the telechelic polymerization. Thus, the final product is a mixture of the structures depicted in Scheme 4.
The functionalization reactions to obtain vinyl ether-ended or oxirane-ended oligomers were conducted on the mixtures, as the separation of the monofunctional from the difunctional PFPAEs was unsuccessful, although numerous chromatography experiments were performed, trying to find the best eluents and TLC stains (S1 and S19 in the Supporting Information).
The synthesis of functional PFPAE derivatives, starting from the acyl fluoride or the carboxylic acid, was challenging as known from previous works [29,[54][55][56][57]. One of the main issues was related to the extreme non-polarity of the PFPAE chains [10,20,56]; indeed, the PFPAE solubility in solvents commonly used in organic chemistry is quite low. Moreover, even when the solubility of PFPAEs was optimized, the PFPAE starting oligomers reactivity was reduced due to the presence of the longfluorinated chain. Therefore, the PFPAE vinyl ethers and epoxides were obtained with a low yield (~15% of isolated product), but with good purity (see 1 H-NMR and 13 C-NMR spectra in the Supporting Information). As well as the precursors, the obtained functional monomers (epoxides and vinyl ethers) contain both monofunctional and telechelic oligomers ( Table 1). The telechelic oligomer molar percentage, as estimated by 19 F-NMR spectra (see the Supporting Information for the calculation), is reported in Table 2. The molecular weight and the degree of polymerization of the products were also calculated by 1 9 F-NMR spectra and are reported in Table 2 , distinguishing the values of the monofunctional oligomer (M n1 ) and those of the difunctional oligomer (M n2 ). The functional group density δ FG and the average functionality f of the PFPAE-based derivatives were evaluated, considering the experimental ratio between mono and difunctional oligomers: as reported in Table 2, the values are very close.

Photoinduced polymerization
After the addition of a common cationic photoinitiator, the PFPAE derivatives were casted on glass slides and irradiated in air. The kinetics of the photopolymerization process was monitored in real time by FT-IR and photo-DSC.
In Fig. 1, the conversion curves by real-time FT-IR of the vinyl ether monomers (Fig. 1a) and of the epoxides (Fig. 1b) are reported. The conversion was evaluated during the irradiation of the sample with UV light by monitoring the decrease of the intensity of the band of the double bond (1620 cm −1 ) and of the band of the oxirane ring (900-920 cm −1 ), for the polymers with vinyl ether and epoxide reactive groups, respectively. The curves follow typical cationic photoinduced polymerization trends exhibited by vinyl ethers and epoxides [58][59][60][61] with the kinetics regimes and the complex mechanisms occurring during photopolymerization reactions discussed in literature [60,[62][63][64][65][66].
The fluorinated vinyl ethers (inset of Fig. 1a) react faster than the epoxides polymers (inset of Fig. 1b), as it can be noticed by the higher conversion rate in the first 10 s of irradiation; in about 180 s, their conversion is nearly 60-70% (Fig. 1a), while for PFPAE epoxides is lower than 50% in the same time scale (Fig. 1b). At the end of the irradiation (i.e., 600 s), the final conversion of the fluorinated vinyl ethers is higher than 80% (Fig. 1a), while the fluoroepoxides reach a maximum conversion of approximately 65% (Fig. 1b).
Comparing the oligomers having the same photoreactive groups, it is shown in the insets of Fig. 1  Scheme 3 Reaction mechanism leading to a monofunctional product polymerization rate follows the order PFPAE-EGVE < PFPAE-BGVE < PFPAE-DEGVE (Fig. 1a) and PFPAE-MO < PFPAE-EO < PFPAE-PO (Fig. 1b). Interestingly, there is a different reactivity, although the functionality and the functional density of the systems are very similar (see Table 2). Therefore, one could suppose that the reactivity depends on the structure of the oligomers, in particular on the length of the hydrogenated spacer, which corresponds to the number of methylene units between the fluorinated chains and the reactive end-groups. The PFPAEs functionalized with longer hydrogenated spacer show a faster initial polymerization rate, also reaching a higher maximum degree of conversion. The effect of the spacer on reactivity has been observed and studied, for both living cationic [37,64,67] and radical [68,69] polymerizations. A long-fluorinated chain induces an electron withdrawing effect, which decreases the monomer reactivity [70][71][72]. Increasing the number of methylene units between the fluorinated segment and the vinyl ether group or epoxy group, one can modulate the polymerization rate.
The polymerization kinetics of the PFPAE polymers was also studied employing an alternative, but accurate and reproducible [26,47] method: the photo-DSC. In Fig. 2a, b, the first derivative of the conversion curve, which represents the reaction rate, is plotted as a function of the conversion (photo-DSC conversion curves are reported in S36 of the Supporting Information).
Also in this case, considering both the initial and the maximum conversion rate, the PFPAE vinyl ethers (inset of Fig. 2a, S37 in the Supporting Information) are demonstrated to react faster than the fluorinated epoxides (inset of Fig. 2b, S37 in the Supporting Information).
Furthermore, the conversion reached by the vinyl ethers (around 80%, Fig. 2a) is higher than the final degree of conversion showed by the epoxides (ca. 60%, Fig. 2b).
For both the fluorinated vinyl ethers and the epoxides, the photo-DSC analyses confirmed that the different number of methylene units between the fluorinated segment and the reactive end-group has a non-negligible influence on the initial polymerization rate. In general, functionalized PFPAEs having shorter hydrogenated spacers show slower polymerization rates and lower conversions (Fig. 2a, b), as already observed in the real-time FT-IR kinetics plots (Fig. 1a, b).
Furthermore, considering the gel point, taken as the maximum of the conversion rate, it can be observed that the transition from a polymerizing system to a network slowing the reaction and changing its kinetic regime occurs for lower degrees of conversion for the PFPAEs functionalized with shorter hydrogenated spacers (Fig. 2a, b). The conversion degrees obtained by different techniques are collected in Table 3. The comparison of the data from real time FT-IR and photo-DSC analysis, evaluated at the same UV dose (i.e., 25 J cm −2 ), are in good agreement. Then, the samples were stored for 48 h to allow the "dark curing" reaction to occur [60,73,74], i.e., further monomers conversion after the end of the irradiation, and the photopolymerization conversions were estimated by FT-IR ATR (Table 3, S38 in the Supporting I n f o r m a t i o n ) . D a t a s h o w t h a t t h e c a t i o n i c photopolymerization spontaneously continues in the dark leading to higher conversion values, i.e., almost complete conversion for the fluoro-vinyl ethers, and higher than 70% for the PFPAE-epoxides. The residual insoluble fraction (gel content, Table 3) of the photopolymerized networks, determined by solvent extraction, is in good agreement with the trend of the conversion data collected by FT-IR ATR. The values are rather high (i.e., always > 80%), showing polymers are cross-linked networks, and higher values are found for the polymers obtained from the PFPAE derivatives with longer hydrogenated spacers. However, the photocured films (colored but transparent) were soft, and more similar to highly viscous liquids rather than solid films due to the high content of PFPAE chains, which are very mobile [32,75,76] with a low T g , at the same time the hydrogenated phase is rubbery at room temperature (as it will be discuss later).

Photopolymers characterization
The thermal resistance of the photo-cured polymers was assessed by TGA: the derivative weight loss (DTG) curves, recorded in inert atmosphere, are shown in Fig. 3. The DTG curves of the PFPAE polymers display two distinct regions of weight loss, implying that two main steps of degradation occur in those samples.
The first step, between 150 and 400°C, is responsible for the scission of the perfluoropolyalkylether main chains [69] for both the PFPAE vinyl ethers and epoxides (T 1 region, Fig.  3a, b). The PFPAE thermal degradation may be triggered by the presence of Lewis acids such as AlF 3 [77] or Al 2 O 3 [78] and can result in a complete loss of the fluorinated segments above 180°C. Moreover, the thermal behavior could be adversely affected by residues of the superacids generated during the photoinitiation of the cationic polymerization and still present in the network.
The maximum weight loss of the first degradation step (ca. 30% by weight for each polymer) is higher for the functional PFPAE derivatives with a shorter hydrogenated spacer.
During the second weight loss, which occurs between 450 and 600°C (T 2 region, Fig. 3a, b), the C-C and C-O bonds of the vinyl ethers and epoxides end-groups are further destroyed due to the high temperatures [69]. The maximum temperature of this second weight loss stage corresponds to ca. the 90 wt.% of degradation (~530°C), and it is almost the same for all the Table 2 Degrees of polymerization (DP n ), molecular weights of monofunctional (M n1 ), and difunctional (M n2 ) oligomers, functional group density (δ FG ), and average functionalities f of the products

Monomer
Telechelic monomer molar percentage (%) DP n M n1 (g mol −1 ) M n2 (g mol −1 ) δ FG (mol g -1 ) f  polymers. The final residue is zero, indicating that in N 2 a complete degradation of material occurs above 600°C. As suggested by the crosscheck of the degradation temperature data (collected in S39 of the Supporting Information) with the photopolymerization kinetics data (Table 2), it seems that, in general, as expected, with a lowering of the conversion, the photocured polymers exhibit a poorer thermal resistance.
The phase transitions of the fluorinated homopolymers can be observed in the DSC curves reported in Fig. 4. The photopolymers show two different glass transitions: the phase transition occurring at lower temperature (T g1 ≅− 65°C) can be attributed to the PFPAE moieties, while the glass transition that takes place at higher temperature (T g2 = 17-23°C) can be assigned to the alkyl spacing groups of the functionalized fluoromonomers or to a interphase region, mainly containing hydrogenated segments [79,80]. These results indicate biphasicity of the PFPAE polymers at molecular level, since their structure contains a fluorinated chain linked to alkyl spacers bearing hydrogenated end-groups [81]. However, films are transparent at a visual inspection. As reported before, the photocured films are rubbery at room temperature.
Wettability of the UV-cured PFPAE polymers was evaluated by contact angle measurement (Table 4) using water (polar solvent) and hexadecane (apolar solvent). The water static contact angle of the photopolymerized PFPAEs was higher than 85°, while a contact angle of about 70°was measured when hexadecane was used as measuring liquid. These contact angle values are commonly exhibited by highly fluorinated materials [31,32,38]. They are similar to those shown by photocrosslinked copolymers containing less than 5 wt% of a fluorinated monomer with the same PFPAE chain structure [31,32]. Neither the length of the hydrogenated spacer nor  As expected on the basis of the wettabilities, the surface energy (γ) of the PFPAE polymers is very low (Table 4), and especially the polar component, γ p , that is lower than 10 mN m −1 , is the most responsible of the drop of γ. Low surface energy can guarantee a difficult wettability of most surfaces and repellency toward most liquids: anti-graffiti and antistaining properties, for example, can be achieved [31].
Further information on the surface properties of the UVcured polymers can be obtained by the contact angle hysteresis (CAH), i.e., the difference between advancing and receding contact angle (S40 in the Supporting Information). Hysteresis is mainly caused by surface roughness and/or heterogeneity and commonly occurs in materials containing at the surface both hydrophobic and hydrophilic groups [27,82].
PFPAE polymers show hysteresis values that are generally very low (Table 4), independent from the monomer structure (namely the hydrogenated spacer between the fluorinated chain and the reactive end-group), and, therefore, related to the main PFPAE chain which is the same in all monomers. These low hysteresis values suggest that the photocured PFPAE polymers can be exploited as anti-staining and selfcleaning materials [31,83].
For this reason, the PFPAE polymers surface behavior was also tested by sliding angle measurements. The polymer film surface, on which a water drop is deposited, is tilted, and a threshold sliding angle can be defined as the tilting angle at incipient droplet motion [84]. Figure 5 shows a graph of Δα (i.e., the difference between the advancing contact angle and the receding contact angle measured on the sliding drop during the tilting stage) versus the tilting angle on a UV-cured  . 5a) and PFPAE-EO (Fig. 5b). As it can be observed, the threshold sliding angle is comprised between 12°and 15°. Similar results were found for most of the films prepared: these values are comparable to those previously reported in similar surfaces containing perfluoropolyalkylether chains [31], demonstrating that these polymers can be exploited as self-cleaning materials.

Conclusions
The synthesis of polyfunctional vinyl ether and epoxide monomers containing the same PFPAE chain and different hydrogenated spacers between the perfluoropolyether chain and the photoreactive groups was successfully performed, although the reactions had low yields. The PFPAE derivatives with an average functionality around 1.5 were successfully employed in the photoinduced cationic polymerization obtaining transparent films, as assessed by visual inspection. The interesting photopolymerization behavior, of both the PFPAE vinyl ethers and epoxides, was studied using different techniques, namely real time FT-IR and photo-DSC: crosslinked polymers with high gel content were obtained, while conversion was in the range 60-90% depending on the monomer structure and functionality. In particular, the reactivity of the oligomers depended on the spacer, i.e., the length of the hydrogenated chain connecting the PFPAE and the functional group.
The obtained polymers exhibited biphasicity as two distinct T g s were present, one at low temperature due to the PFPAE blocks, the second slightly below room temperature (25°C) due to the hydrogenated domains. Therefore, the materials are soft. TGA results indicated that the thermal stability of the PFPAE polymers can be adversely affected by the presence of acids in the cured networks. Interestingly, hydrophobic and oleophobic properties were shown, and very low contact angle hysteresis and sliding angle. Based on these results, we can conclude that these materials can be proposed to produce high-performance and self-cleaning coatings. Table 4 Static contact angle (θ), surface energy (γ), divided in its dispersive (γ d ) and polar (γ p ) components, and contact angle hysteresis (CAH), estimated from the difference between the advancing (θ adv ) and the receding (θ rec ) angles, of the UV-cured fluoropolymers