A new oligo(hexafluoropropylene oxide)-b-oligo(ethylene oxide) diblock surfactant obtained by radical reactions

Abstract

The synthesis and characterization of a new oligo(hexafluoropropylene oxide)-b-oligo(ethylene oxide), oligo(HFPO)-b-oligo(PEG), diblock co-oligomer are presented. First, the model reactions dealing with the radical addition of 1-iodoperfluorohexane (C₆F₁₃I) onto allyl alcohol and allyl-O-PEG-OCH₃ were optimized in terms of the choice of the initiator (azobisisobutyronitrile [AIBN], tert-butylperoxypivalate [TBBPi], and benzoyl peroxide [BPO]) and of the solvent, temperature, and time. Allyl-O-PEG-OCH₃ was obtained from the etherification of ω-hydroxy-PEG with allyl bromide. End-capping of oligo(HFPO) with PEG was successfully achieved by the radical addition of 1-iodoperfluoropropyl-2-oligo(hexafluoropropylene oxide) [oligo(HFPO)-CF(CF₃)CF₂I] onto allyl-O-PEG-OCH₃ using the best conditions of the model reactions. Although TBPPi failed and led to oligo(HFPO)-isobutyl iodide, AIBN and BPO yielded oligo(HFPO)-CH₂CHICH₂-oligo(PEG). The selective reduction of the latter compound led to oligo(HFPO)-b-oligo(PEG) in 77% yield, the surface tension properties of which were compared to those of commercially available ammonium perfluorooctanoate (APFO) and perfluorooctanoic acid (PFOA). Its critical micelle concentration was 0.04 g mol⁻¹. All models, intermediates, and diblock co-oligomers were characterized by ¹H, ¹⁹F, and ¹³C NMR spectroscopy as well as matrix assisted laser desorption ionization (MALDI) and atmospheric pressure solids analysis probe (ASAP) time-of-flight mass spectrometry (TOF-MS).

---

1. Introduction

Polyfluorinated compounds (PFCs) are useful chemicals involved in a wide range of products. Among them, molecules that bear both fluorinated hydrophobic moieties and hydrophilic parts, called “surfactants”,^1–4 are valuable compounds. Surfactants are being used in more than 200 applications^5–10 ranging from the protection of surfaces (textile, paper, carpets, masonry, metal, and leather), as stimulating fluids for oil recovery, fire-fighting foam, skin protection from chemical agents, soil and stain-repellents, plane hydraulic fluids,^1,11 paints, lubricants, electroplating, photographic emulsifiers, pressure sensitive additives, pharmaceuticals, and insecticides, or are involved in cosmetic formulations. Perfluorooctanoic acid (PFOA), for example, is frequently used as a surfactant in the aqueous media (co)polymerization of hydrophobic monomers (e.g. fluorinated monomers^5,12 PFOA and perfluorooctane sulfonic acid (PFOS) are the most used fluorosurfactants). Their surface tensions and critical micelle concentrations are very low.^1,2 They feature both pronounced hydrophobicity and oleophobicity and display high chemical and thermal resistance. They can be synthesized either by electrochemical fluorination (ECF)¹³ or by telomerization.¹⁴ The former process leads to ca. 70% straight chain PFOS along with 30% branched and cyclic isomers.¹⁵ PFOA is being used in fast food packaging (≥300 μg kg⁻¹) owing to its water and oil repellencies.^4b

Telomerization reactions¹⁴ are straightforward and the most representative example is the radical telomerization of tetrafluoroethylene (produced industrially by various chemical industries) that generated telomers containing a hydrophobic perfluorinated C_nF_2n+1 end-group (where n is an even number).^5,14 D'Eon and Mabury¹⁶ reported the fluorotelomer production values from 1970 to 2020. Telomeric alcohols have been widely used for polymers and surface coatings with an estimated annual production of ca. 12 000 t per year in 2004.⁷ In spite of its remarkable properties and wide range of applications, PFOS was listed in Annex B of the Stockholm convention as a persistent organic pollutant in 2009.¹⁷ As research has demonstrated that many of the long-chain PFCs are toxic, persistent, and bioaccumulative, government and regulatory agencies have been working toward agreements and regulations to limit the production of some PFCs.^10,18 These three severe health and environmental issues arise from the high stability of the perfluorinated chains which are not degraded either by enzymes or metabolic processes.¹⁹ In addition, PFCs bioaccumulate in the food chains and have long half-lives in human blood:¹⁶ 3.6, 5.4,^20–22 and 8.5 years for PFOA, PFOS and perfluorohexane sulfonate, respectively.²³ Other severe limitations are the foetal development toxicity, the immunotoxicity, and the effects on thyroidal hormones.²⁴ PFOS itself is quite stable in the environment, with no known natural mechanism of degradation.²⁴ Attempts to degrade PFOA and PFOS were suggested by Parsons et al.²² PFOS remains the predominant PFC found in all living species, tissues, and locations analyzed around the world.²⁵ PFOA and PFOS accumulations in marine mammals (e.g., >1200 ng g⁻¹ in liver) are common all over the Northern hemisphere.^11,16,26 In a recent review, Fromme et al.²⁷ evaluated potential PFC exposures from indoor and outdoor air, house dust, drinking water, and food²⁸ and concluded that median uptakes of PFOS and PFOA were of the order of 2–3 ng kg⁻¹ per day, with food being responsible for greater than 90% of this exposure.²⁹ However, with the wide variety of foods consumed and the difficulty in establishing sensitive analytical methods that accurately measure contaminants, there is still a great deal of uncertainty about the role of food as an exposure route.³⁰ An increasing number of studies have suggested that fish from contaminated water bodies may be the prevailing source of exposure to PFOS and possibly other long-chain perfluorocarbon acids (PFCAs).^26,31 In addition, the consumption of contaminated drinking water¹⁵ has been linked with an increased surfactant amount in blood, as reviewed by Lindstrom et al.¹⁵

Indeed, in 2006, the European Union set out a ban on the use of PFOS in a number of goods. However, PFCs are still produced and released into the environment¹⁶ and have been quoted as the “PCBs of the XXIst century”. In the same time, EPA launched the 2010/2015 PFOA Stewardship Program³² to reduce emissions and residual content of PFOA and long-chain PFCAs by 95% by 2010 with the goal to eliminate long chain PFCs by 2015. While there has been some success with voluntary controls for some PFCs,³² limited incentive for companies to join these voluntary agreements was noted. Hence, a growing interest in the synthesis of short perfluoroalkyl chain surfactants was noted,^33–36 especially since the 3M company developed perfluorobutane sulfonyl compounds.^37–39

Various strategies to synthesize potentially non-bio accumulable alternatives to PFOA have been reported:⁴⁰ (1) chemicals bearing either a CF₃O or (CF₃)₂N end-group,⁴¹ (2) compounds produced from small perfluorinated chains with non-ionic oligoethylene oxide or carbohydrate,^36,42 (3) carboxylate gemini surfactants,^43,44 (4) vinylidene fluoride (VDF) telomers with short 1-iodoperfluoroalkane where methylene groups may act as “weak” degradable points,^45,46 (5) 3,3,3-trifluoropropene (TFP) telomers from either 1-iodoperfluoroalkanes or other chain transfer agents,⁴⁷ (6) VDF and TFP cotelomers,⁴⁸ and (7) compounds derived from oligo(hexafluoropropylene oxide), oligo(HFPO).⁴⁹ Interestingly, these perfluorooligoethers were shown to be nontoxic and nonbioaccumulable⁵⁰ and thus offer further opportunities to synthesize more environmentally friendly fluorinated surfactants. For example, Li et al.⁵¹ reported the synthesis of oligo(HFPO)-CO₂-PEG from the esterification of trimer(HFPO)C(O)F with HO-PEG leading to a surfactant that possesses a critical micelle concentration of 0.6 g L⁻¹. However, the ester group induces limitations in some applications due to their stability and polarity. The objective of this article is thus to overcome this issue. It describes the use of longer oligo(HFPO)-based novel surfactants. Oligo(HFPO) iodide was reacted under radical conditions with ω-unsaturated oligo(ethylene oxide) to yield the desired oligo(HFPO)-based surfactants. The study of their surface tension properties is also presented.

2. Results and discussion

The purpose of this work was to explore conditions that were favorable in converting 1-iodo perfluorooligoalkane ether, specifically oligo(hexafluoropropylene oxide) [oligo(HFPO)] primary iodide, into diblock surfactants using polyethylene glycol (PEG) as the hydrophilic sequence. This oligo(HFPO)-b-oligo(EG) diblock surfactant was achieved by the radical addition of [oligo(HFPO)] primary iodide onto allyl-PEG followed by a selective reduction of the iodine atom (Scheme 1).

Scheme 1 Synthesis of oligo(hexafluoropropylene oxide)-b-polyethylene glycol, oligo(HFPO)-b-PEG diblock surfactant by radical addition of [oligo(HFPO)] primary iodide onto allyl PEG followed by a selective reduction of the iodine atom.

Earlier work had demonstrated that the synthesis of an oligomeric diblock was possible with aromatics and oligo(HFPO)-CF(CF₃)CF₂I.⁵² Other studies reported that perfluoroalkyl iodides can be added onto alkenes in high yields⁵³ under radical conditions, as well as act as efficient chain transfer agents in iodine transfer copolymerization of vinylidene fluoride (VDF) and hexafluoropropylene (HFP).⁵⁴ The radical addition process was considered the best route to form a diblock of perfluoroalkyl or 1-iodo perfluorooligoalkane ether with polyethylene glycol (PEG).⁹ Indeed, other types of linkages such as esters or amides are likely to hydrolyze over time and thus prone to degradation. The synthesis of the target amphiphilic diblock proceeded in two steps: (i) the introduction of an allylic ether on the hydroxyl end-group of monomethyl ether polyethylene glycol via Williamson ether syntheses (Scheme 2), and (ii) the radical addition of the iodide precursor onto the allylic moiety. Depending on the conditions used, a reduction step was achieved.

Scheme 2 Synthesis of allyl oligo(ethylene oxide) allyl-O-PEG-OCH₃ from the etherification of HO-PEG-OCH₃ with allyl bromide.

The yield of allyl-O-PEG-OCH₃ was found to be ca. 65%. The ¹H NMR spectrum (Fig. S1†) displays the expected multiplets assigned to the three ethylenic protons centered at 5.11, 5.20, and 5.85 ppm while the signals assigned to the methylene groups adjacent to oxygen atoms in the oligo(ethylene oxide) chain and characteristic of the CH₃O end-group are located at 3.96 ppm and a singlet is centered at 3.31 ppm, respectively. The ¹³C NMR spectrum (Fig. S2†) confirms its structure from the signals centered at 136 and 116, 70.7, and 59 ppm assigned to both ethylenic carbon atoms, methylenes in ethylene oxide units, and the –OCH₃ end-group, respectively. The synthetic strategy for this work relies on the radical addition of 1-iodoperfluoroalkane or 1-iodoperfluoroalkylether [oligo(HFPO)-CF(CF₃)CF₂I] onto an alkene compound such as the above allyl-O-PEG-OCH₃.

To explore the appropriate and best conditions to prepare oligo(HFPO)-b-oligo(PEG) diblock co-oligomers in high yield using radical initiators, a systematic study was undertaken. This article reports the study of five systems: (1) short 1-iodoperfluoroalkane (C₆F₁₃I) reacting with a model alkene [e.g., allyl alcohol], (2) small perfluorinated alkyl iodides reacting with a polymeric alkene [e.g., allyl-O-PEG-OCH₃], (3) polymeric fluorinated iodides [oligo(HFPO)-CF(CF₃)CF₂I] with a simple alkene [e.g., allyl alcohol], (4) polymeric fluorinated iodides with a polymeric alkene, and (5) small molecule and polymeric fluorinated iodides reacting strictly with three radical initiators: azobisisobutyronitrile [AIBN], tert-butylperoxypivalate [TBBPi], and benzoyl peroxide [BPO], to observe potential by-product formation.

2.1. Radical addition of 1-iodoperfluorohexane (C₆F₁₃I) onto unsaturated alkenes (allyl alcohol and allyl ether, allyl-O-PEG-OCH₃)

2.1.1 Radical addition onto allyl alcohol. The radical addition of 1-iodoperfluoroalkanes onto unsaturated derivatives has been reported in many reviews and articles, involving thermal and photochemical initiations, metallic salts and radical initiators.⁵⁵ The model reaction of 1-iodoperfluorohexane (C₆F₁₃I) with allyl alcohol (a simple functional alkene compound), initiated by azobisisobutyronitrile, AIBN, tert-butylperoxypivalate, TBBPi, or benzoyl peroxide, BPO (Scheme 3), was an important starting point to further explore the use of oligo(HFPO) diblocks.

Scheme 3 Radical reaction of 1-iodoperfluorohexane with allyl alcohol and allyl-O-PEG-OCH₃ (initiated by azobisisobutyronitrile AIBN, tert-butylperoxypivalate, TBBPi, and benzoyl peroxide, BPO).

The reaction temperatures were first adjusted so that the half-lives of the initiators were approximately one hour (Table 1). The reactions were initially monitored by gas chromatography/mass spectrometry (GC/MS) to determine reaction times. After the reaction and removal of traces of unreacted reactants, the total product mixture was characterized by ¹H ¹⁹F, and ¹³C NMR spectroscopy to determine the reaction yield. In all cases, C₆F₁₃I conversion was higher than 93%. The first ¹H NMR spectrum (Fig. S3†) did not display any signals in the ca. 5.5–7.0 ppm range assigned to the ethylenic protons of allyl alcohol (Fig. S4†) but instead showed complex systems at 2.95 and 2.65 ppm and 4.3 ppm attributed to C₆F₁₃CH₂– and –CHI–, respectively. The ¹⁹F-NMR spectrum (Fig. S5†) showed the high field shift of the signal assigned initially to –CF₂I (−59 ppm, Fig. S6†) and then to –CF₂CH₂– (−112 ppm) as a characteristic AB system. All other signals such as the CF₃– centered at −83 ppm, and those in the −123 to −128 ppm range assigned to the (CF₂)_n–, remained relatively unchanged compared to those of the spectrum of C₆F₁₃I. Finally, the ¹³C NMR spectrum (Fig. S7†) exhibited a doublet of triplets, and both singlets were centered at 37.18, 67.8, and 20.4 ppm assigned to the methylene groups in –CF₂CH₂–, –CH₂OH, and the methine –CHI– group, respectively. This ¹³C NMR spectrum did not show any signal assigned to –CF₂I at 110 pm (Fig. S8†) for the potentially remaining C₆F₁₃I and displayed a minimal amount of unreacted allyl alcohol (Fig. S9†). Electron impact mass spectrometry (Fig. S10†) was also useful in detecting the formation of the iodhydrin characterized by the molecular ion at m/z = 504 (M+ =C₆F₁₃CH₂CHICH₂OH), and characteristic fragments at m/z = 377 (M − I) and m/z = 357 (M − I − HF).

Table 1 Model reactions: radical addition of C₆F₁₃I onto allyl compounds

Rxn #	Allyl compound	C6F13I (mmol)	Allyl (mmol)	Initiator (mmol)	Initiator	t (h)	T (°C)	Conv. C6F13I^d (%)	Yield (%) ^19F-NMR
Initiator for reactions 1–6 was only added once at the beginning of the reaction.a Ref. 56.b Ref. 57.c Ref. 9a; AIBN, TBPPi, and BPO stand for azobisisobutyronitrile, tert-butylperoxypivalate, and benzoyl peroxide, respectively.d Assessed from ^19F NMR.
1^a	Allyl alcohol	1.150	1.875	0.152	TBPPI	2	75	>99	>99
2^a^,^b	Allyl alcohol	1.120	1.484	0.023	AIBN	4	90	93	92
3	Allyl alcohol	1.120	1.680	1.008	BPO	8	90	100	0
4	Allyl-O-PEG-OCH3	0.359	0.417	0.217	TBPPI	4	90	84	69
5^c	Allyl-O-PEG-OCH3	0.209	0.417	0.006	AIBN	4	90	51	37
6	Allyl-O-PEG-OCH3	0.093	0.417	0.310	BPO	8	90	100	0

---

However, it is worth noting that conversion of the alkyl iodide does not necessarily equate to product yield. In all cases, complete conversion of the iodide took place within 8 to 9 h. Both TBPPI and AIBN were suitable in yielding the desired iodhydrin adduct. In contrast, benzoyl peroxide did not yield any product (yield = 0%, Table 1).

2.1.2 Radical addition onto allyl-O-PEG-OCH₃. The radical addition of C_nF_2n+1I (n = 6 or 8) onto allyl oligo(ethylene oxide) (with average repeating units of 2, 3 or 4), reported by Koplanik et al.,^9a initiated by AIBN or sodium dithionite, led to the expected iodinated product in 87–96% yield. However, in contrast to our present study, the chain length chosen by these authors was much shorter. Using similar conditions to those of section 2.1.1, for the radical addition of C₆F₁₃I onto an allyl-O-PEG-OCH₃, a striking difference was noted between the three initiators (Table 1). As reported by Koplanik et al.,^9a AIBN led to the iodinated diblock co-oligomers, but only in fair yield (37%) probably because of the long chain length. However, TBPPI enabled the reaction to achieve the highest yield of C₆F₁₃-CH₂CHICH₂-O-PEG-OCH₃. One possible explanation for such a feature is the higher amount of radicals produced (the reactions were carried out at 90 °C). 84% of the converted iodide resulted in the desired product using TBPPI (69% yield) whereas, in the case of AIBN, 51% conversion and 37% yield were observed. In the ¹H-NMR spectrum (Fig. S11†), the multiplet centered at 4.35 ppm indicates the methine –CHI– formed by the radical addition of C₆F₁₃I onto the allyl-O-PEG-OCH₃. The ¹⁹F-NMR spectrum (Fig. S12†) is most useful for demonstrating connectivity. It clearly shows the AB system for the –CF₂CH₂– group centered at around −118 ppm. In the ¹³C-NMR spectrum (Fig. S13†), the difluoromethylenes are hidden in the background noise but the triplet centered at 37.3 ppm and the signal at 14.38 ppm were assigned to the –CF₂CH₂– and –CHI, respectively. On further examination of TBPPI and AIBN initiated-reactions, a substantial difference is noted in the initiator/C₆F₁₃I ratio: a 0.6 ratio for TBPPI and only 0.03 for AIBN. Again, benzoyl peroxide was able to convert C₆F₁₃I but did not lead to the desired product. The GC/MS (Fig. S14A†) shows that BPO converts the C₆F₁₃I into an aromatic adduct. Retention times of 7.9, 8.2, and 8.5 min show that the major products formed are the ortho, meta, and para isomers of C₆F₁₃PhI.⁴³ This can be easily identified by mass spectrometry with m/z fragments at 126, 252, and 522 m/z assigned to PhCF₂+, I–Ph–CF₂+, and the molecular weight of C₆F₁₃PhI fragments, respectively (Fig. S14B†).

2.2. Radical addition of 1-iodoperfluoropropyl-2-oligo(hexafluoropropylene oxide) onto unsaturated alkenes (allyl alcohol and ether)

2.2.1. Radical addition onto allyl alcohol. The analysis of the results of the reaction of polymeric iodides [1-iodo-2-oligo(hexafluoropropylene oxide) perfluoropropane] with allyl alcohol (Scheme 4) showed a considerable increase in time necessary to form the iodhydrin when a similar molar scale as C₆F₁₃I was used. These reactions required five times longer time than in the case of C₆F₁₃I.

Scheme 4 Radical reaction of 1-iodoperfluoropropyl-2-oligo(hexafluoro-propylene oxide) onto allyl alcohol and allyl-O-PEG-OCH₃ (initiated by AIBN, TBPPI, and BPO).

The progress of the reaction was monitored by the decrease of the starting iodide (Fig. S15†) and the formation of the product (Fig. S16†) using gas chromatography/mass spectrometry (GC/MS). The spectrum showed that 38 to 72 hours were necessary to reach the desired yields (>80%). The ¹H NMR spectrum of the iodhydrin based on oligo(HFPO) (Fig. S17 or S18†) exhibited the characteristic signals for –CHI– group at 4.31 and 4.15 ppm resulting from the –CF₂CH_aH_b– at 2.84 and 2.54 ppm, respectively. The respective methylene in –CH₂OH and the –OH are noted at 3.66 and 3.99 ppm, respectively (Fig. S18†). The ¹⁹F-NMR spectra (Fig. 1A & B) show the absence of the characteristic triplet of doublets centered at −58 ppm (Fig. 1A and S19†) assigned to the CF₂I end-group in oligo(HFPO)-CF(CF₃)CF₂I. Instead, they show a high field shifted signal centered at −112 ppm and resulting from a complex AB system containing two stereocenters (Fig. 1B and S20†). The ¹³C-NMR spectrum (Fig. S21†) exhibits the absence of the triplet of doublets at 91.3 ppm, which is normally present in the iodide (Fig. S22†). Typical signals of oligo(HFPO)-CF(CF₃)CF₂ were observed in the 124 ppm to 98 ppm range, while those of iodhydrin can be noted at 66.32, 36.52 and 18 ppm attributed to the –CH₂OH, –CF₂CH₂–, and –CHI groups, respectively.

Fig. 1 ¹⁹F-NMR spectra of 1-iodo-2-oligo(hexafluoropropylene oxide)perfluoropropane(F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂I) (A) and F[CF(CF₃)CF₂O]_8.9CH₂CHICH₂OH (B).

The atmospheric pressure solids analysis probe (ASAP) time-of-flight mass spectrometry (TOF-MS) spectrum in negative ion mode for oligo(HFPO)-CF(CF₃)CF₂I (Fig. S23†) displays one distribution between 2400 and 4200 m/z corresponding to oligo(HFPO)-based polymers as evidenced by the m/z difference between two consecutive oligomeric peaks (Δm/z = 166 Da) which corresponds to the HFPO repeat unit mass. This distribution is attributed to the deprotonation molecular ion of (F[CF(CF₃)CF₂O]_nCF(CF₃)CF₂I + CH₃OH–H)– when methanol is present in the probe. In addition, fragments of F[CF(CF₃)CF₂O]_nCF(CF₃)O– were observed between 184 and 1845 m/z. In negative ion mode, the MALDI-TOF-MS spectrum of oligo(HFPO)-CF(CF₃)CF₂I (Fig. S24†) shows one distribution between 2300 and 4400 m/z. This distribution corresponds to the deprotonation of oligo(HFPO)-I with formic acid (F[CF(CF₃)CF₂O]_n CF(CF₃)CF₂ I + HCOOH–H)–. The repeat units Δm/z = 166 Da confirm the presence of oligo(HFPO). In the ASAP-TOF-MS spectrum of F[CF(CF₃)CF₂O]_n CF(CF₃)CF₂CH₂CHICH₂OH iodhydrin (Fig. S25†) in positive ion mode a main distribution of the M+ = (F[CF(CF₃)CF₂O]_n CF(CF₃)CF₂CH₂CHICH₂)+ radical cation was observed. The loss of the OH group is possible in this ionization mode. A second distribution corresponding to the radical cation M+ with acetonitrile from the probe and a few fragments that may arise from the main compound were also observed. The MALDI-TOF-MS spectrum of F[CF(CF₃)CF₂O]_nCF(CF₃)CF₂CH₂CHICH₂OH iodhydrin (Fig. S26†) highlights the positive ion of F[CF(CF₃)CF₂O]_nCF(CF₃)CF₂CH₂CHICH₂OH. One distribution was detected between 1400 and 3000 m/z. This corresponds to (F[CF(CF₃)CF₂O]_nCFCF₃CF₂CH₂CHICH₂O + Li)⁺ lithium adduct. The repeat units of 166 m/z indicate that oligo(HFPO) was present, whereas several fragmentations of this compound were observed between 400 and 3000 m/z.

The reaction of oligo(HFPO)-CF(CF₃)CF₂I with allyl alcohol carried out on the same scale (0.56 mmol) initiated by TBPPI (0.5 mole ratio to the iodide) was almost twice faster than AIBN (0.6 mole ratio to the iodide). It is believed that the rate of the reaction was increased with a higher amount of formed free radicals. TBPPI produces tBu˙ while t-BuO˙ initially generated also released CH₃˙ at more elevated temperatures.⁵⁸ In contrast, AIBN only produces two (CH₃)₂(CN)C˙. Once again, BPO⁴³ did not form any of the desired iodhydrin although it produces two radicals as well: Ph˙ or PhC(O)O˙ (Table 2).

Table 2 Radical addition of oligo(HFPO)-CF(CF₃)CF₂I onto allyl compounds^a

Rxn #	Allyl compound	HFPO-I (mmol)	Allyl (mmol)	Initiator (mmol)	Initiator	t (h)	T (°C)	Conv. HFPO-I (%)	1 HEC (%)	Yield (%) ^19F-NMR
a Additions beyond the first amount of initiator – (7) 6 additions, (8) 18 additions, (9) 16 additions, (10) 7 additions, (11) 17 additions, (12) 35 additions, (13) 0 addition, (14) 1 addition, (15) 19 additions, (16) 15 additions, (17) 20 additions, (18) 86 additions, (19) 1 addition, (20) 5 additions; calculated using ^19F NMR.
7	Allyl alcohol	0.564	5.279	0.080	TBPPI	48	75	100	2.3	88
8		0.564	16.357	0.284		38	90	94.5	4	80
9		5.640	30.120	7.293		32	75	73	2.5	0
10	Allyl alcohol	1.147	12.162	0.180	AIBN	64	90	100	1.12	86
11		0.564	1.510	0.203		72	90	100	3	86
12		5.640	112.861	4.056		72	90	100	2.5	89
13	Allyl alcohol	0.564	1.119	0.845	BPO	8	90	68	2.7	0
14		0.564	2.256	3.619		8	90	95.5	6.5	0
15	Allyl-O-PEG(450)-OCH3	0.564	0.862	0.230	TBPPI	80	75	96.5	8.1	0
16		5.640	17.239	18.365		64	75	99.0	4	0
17	Allyl-O-PEG(450)-OCH3	0.564	2.586	2.326	AIBN	80	90	100	25	48
18		5.640	23.015	23.997		172	90	100	24	50
19	Allyl-O-PEG(450)-OCH3	0.564	2.761	2.343	BPO	16	90	100	6	71
20		5.640	11.280	23.618		48	90	97.0	19	51

---

The reactions with allyl alcohol are difficult to reproduce due to the biphase nature of the reaction media, the reactants being immiscible, and it is expected that the reaction most likely occurred at the interface of both liquid phases (Table 2, RXNs 9 and 10). Because of such a non-homogeneous medium, a situation may be created in which the radicals react more easily with oligo(HFPO)-CF(CF₃)CF₂I. The abstraction of the iodine atom in oligo(HFPO)-CF(CF₃)CF₂I from the radical initiators leads to the oligo(HFPO)-CF(CF₃)CF₂˙ macroradical that may recombine with itself. But this observation was not noted in any of the MALDI-TOF-MS spectra.⁵⁹ However, this macroradical could trap a radical initiator fragment or add onto an aromatic decomposition product. The product corresponding to the latter hypothesis was detected by ¹⁹F NMR spectroscopy (Fig. 2) as evidenced by the presence of the broad signal centered at −116.3 ppm. This signal is in close agreement with the oligo(HFPO)-CF(CF₃)CF₂-C₆H₅ structure previously reported.⁴³ The difluoromethylene attached to benzene gives a complex multiplet from −110.76 to −113.67 ppm; an upfield shift to −116 ppm is probably with an activated group such as iodine attached to the benzene ring. The ¹H-NMR spectrum (Fig. S27†) does indicate the presence of aromatic compounds in the mixture ranging from 7.5 to 8.5 ppm.

Fig. 2 Comparison of the ¹⁹F-NMR expansions of ¹⁹F NMR spectra the reaction of oligo(HFPO)-CF(CF₃)CF₂I with the initiator (TBPPi, AIBN, BPO) (right column) and initiator with allyl alcohol (left column).

2.2.2. Radical addition onto allyl-O-PEG-OCH₃. The results of the reactions of polymeric 1-iodoperfluoroalkane (oligo(HFPO)-CF(CF₃)CF₂I) with allyl-PEG-OCH₃ (Scheme 4, Table 2) were very surprising. The best yields in the oligo(HFPO)-b-PEG diblock co-oligomer were achieved using benzoyl peroxide (71%), followed by AIBN (48%), and lastly by TBPPI (0%) from the respective [initiator]_o/[oligo(HFPO)-CF(CF₃)CF₂I]_o initial ratios of 6.0, 4.0, and 0.4. All three initiators almost completely converted all the iodide into a product or byproducts (Table 3). Two important differences in these reactions, compared to previous ones, were: (i) the absence of a hydroxyl functional group and (ii) more sterically hindered iodide and allyl derivatives. This combination of structural changes can greatly affect the solubility of the reactants.

Table 3 Conversions of 1-iodoperfluorohexane and oligo(HFPO)-CF(CF₃)CF₂I and the formation of the undesired hydrogen end-capped (HEC) compound for the radical addition of 1-iodoperfluoro(alkoxy)alkanes onto allyl alcohol and allyl-O-PEG-OCH₃. Experimental conditions of runs are detailed in Tables 1 and 2^a

Rxn #	Initiator type	Catalyst/Solvent	1 HEC (%)	Conv. 1 iodide (%)
a 1 HEC = C6F13H or oligo(HFPO)-CF(CF3)CF2H.
1, 4, 7–9, 15–16	TBPPI	None	2.3–8.1	94.5–100
3, 6, 13–14, 19–20	BPO	Cu(OAc)2	2.7–19	68–100
2, 5, 10, 12, 17–18	AIBN	None	1.12–25	100
11	AIBN in CHCl3	None	3	100

---

Fig. 3 compares the ¹H NMR spectra of allyl-PEG-OCH₃ (Fig. 3A) with the resulting iodinated diblock co-oligomer (Fig. 3B or S28†). It is noted that the ethylenic protons initially located at 5.11, 5.20, and 5.85 ppm have disappeared and are replaced by an AB system and two multiplets centered at 2.55 and 3.05 ppm, 3.79, and 4.36 ppm assigned to methylene groups in –CF₂CH_aH_b, –CH₂OH, and –CHI–, respectively.

Fig. 3 ¹H-NMR spectra of (A) CH₂=CH–CH₂O(CH₂CH₂O)_9.5CH₃, allyl-PEG-OCH₃ and (B) F[CF(CF₃)CF₂-O]_8.9CF(CF₃)CF₂-(CH₂CH₂O)_9.5CH₃.

As expected, the ¹⁹F NMR spectrum exhibits the high field shift from −60 ppm (assigned to the CF₂I end-group in the iodinated precursor) to −112 to −117.5 ppm in the diblock co-oligomers (Fig. S29†).

The ¹³C NMR spectrum (Fig. 4 or S30†) also shows the absence of ethylenic carbon atoms at 116 and 136 ppm and the presence of fluorinated groups in the 100–125 ppm range, and of the methylene groups in PEG at 70.54 ppm.

Fig. 4 ¹³C-NMR spectrum of F[C(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CHICH₂O(CH₂CH₂O)_9.5CH₃.

ASAP (Fig. 5 or S31†) and MALDI (Fig. S32†) spectra were critical for determining the coupling of the two oligomers. ASAP was more helpful in determining whether the iodo-diblock co-oligomer based on oligo(HFPO) and PEG was formed. In other experiments, it was observed that ASAP ionization caused the fragmentation of polymers.⁶⁰ The desorption temperature, ranging from 50 to 650 °C, plays an important role since the high temperature may also induce a thermal degradation. The analyses using ASAP ionization of oligo(HPFO) oligomers thus led to molecular weights lower than those assessed by MALDI-TOF-MS. In ASAP ionization, the main distribution was observed between 1090 and 2700 m/z which corresponds to the F[CF(CF₃)CF₂O]_nCF(CF₃)CF₂CH₂CHICH₂O(OCH₂CH₂)₄CH₃+ radical cation. This distribution displays a 166 m/z repeat unit as well as a 44 m/z repeat unit characteristic of HFPO and EG, respectively (Fig. 5). In positive ion mode, the MALDI-TOF-MS spectrum of F[CF(CF₃)CF₂O]_nCF(CF₃)CF₂CH₂CHICH₂O(OCH₂CH₂)_9.5CH₃ exhibits only one distribution between 1500 and 3200 m/z for the adduct (M + Li)⁺ (Fig. S32†).

Fig. 5 Positive mode Atmospheric pressure Solids Analysis Probe (ASAP) time-of-flight mass spectrometry (TOF-MS) spectrum of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CHICH₂O(CH₂CH₂O)_9.5–CH₃ (initiated with BPO). 166 m/z is the repeat unit for HFPO [CF(CF₃)CF₂O] and 44 m/z is the repeat unit for ethylene oxide (CH₂CH₂O).

2.3. Radical reactions of 1-iodoperfluorohexane and 1-iodoperfluoro-propylene-2-oligo(hexafluoropropylene oxide) with various initiators (TBPPi, AIBN, and BPO)

In all radical reactions involving an allyl reagent, an almost complete conversion of the fluorinated iodides was observed [seen in the ¹⁹F-NMR spectra (Fig. 2 and S33–35†)]. However, this did not necessarily imply the formation of the desired product. Thus, it was worth studying the nature of radical initiators to understand their role in the production of undesired side-products. As expected, AIBN did not add onto any of the fluorinated alkyl iodides both in the case of neat addition (RXN 22 and 26, Table 4) or in the presence of sodium dithionite (RXN 28) or when chloroform was used as a solvent (RXN 23 and 27). In the presence of radicals, chloroform normally generates CCl₃˙ or HCCl₂˙, but the ¹³C-NMR spectrum did not show any chemical shifts ranging between 80 and 100 ppm.⁶¹ In contrast, TBPPI and BPO formed side-products with both fluorinated alkyl iodides, C₆F₁₃I and oligo(HFPO)-CF(CF₃)CF₂I. When C₆F₁₃I reacted with TBPPi (RXN 21), the ¹⁹F NMR spectrum (Fig. S33†) shows complex signals ranging between −112.3 and −113.6 ppm assigned to the by-product and a signal at −112 ppm characteristic of –CF₂CF₂I in the remaining C₆F₁₃I. C₄F₉I is known to react with aryl-compounds as evidenced by a ¹⁹F-NMR chemical signal centered at ca. −111 ppm assigned to a –CF₂-aromatic group.⁶² The ¹⁹F-NMR spectrum of RXN 24 (Fig. S33†) displays two signals in this same area. These signals were also present in the case of radical reactions of C₆F₁₃I onto allyl compounds (Fig. S33–34†), while the other two initiators are more efficient in product formation (yield = 37 to 99%).

Table 4 Radical reactions of C₆F₁₃I and oligo(HFPO)-CF(CF₃)CF₂I with only a radical initiator (TBPPi, AIBN, and BPO)

Rxn #	1 iodide	Initiator type	Catalyst/Solvent	1 HEC (%)	Conv. 1 iodide (%)
21	C6F13I	TBPPI	None	0	92
22	C6F13I	AIBN	None	0	0
23	C6F13I	AIBN (CHCl3)	None	0	0
24	C6F13I	BPO	Cu(OAc)2	0	100
25	OligoHFPO-I	TBPPI	None	4.5	95.5
26	OligoHFPO-I	AIBN	None	0	0
27	OligoHFPO-I	AIBN (CHCl3)	None	0	0
28	OligoHFPO-I	AIBN	Na2S2O4	0	0
29	OligoHFPO-I	BPO	Cu(OAc)2	0.5	7.5

---

As mentioned earlier, BPO produces aromatic side-products. The structures of these side-products were observed by GC/MS (Fig. S14†). From the reaction of 1-iodoperfluorohexane with BPO, the prevailing aromatic products were iodobenzene (4.8 min, 77 and 204 m/z), benzoic acid (7.1 min, 77, 105, 122 m/z), ortho, meta, para isomers of perfluorohexyl iodobenzene (7.9, 8.1, and 8.5 min; 126, 253, and 522 m/z, respectively) and the minor products arising from the addition of perfluorohexyl radical onto biphenyl (7.4 and 7.8 min) and benzoic acid (9.77 and 10.2 min). Therefore, it is believed that the dominant side-product in the case of the reaction between BPO and the 1-iodoperfluoroalkanes (Table 4) is an iodobenzene perfluoroadduct.

The same radical reactions were studied in the presence of oligo(HFPO)-CF(CF₃)CFI. Surprisingly, this iodo compound shows a different behavior than that of C₆F₁₃I. It was noted that neither AIBN (RXN 26) nor BPO (RXN 29) led to any new signals in the −112 to −118 ppm range in the ¹⁹F-NMR spectra (Fig. S35†). In contrast, TBPPi converted all of the iodide into the undesired side-product. ¹³C-NMR spectroscopy (APT, Fig. S36†) and ASAP mass spectrometry (Fig. S37†) were instrumental in identifying the main side-product using TBPPi. The side-product arises from the addition of the 1-iodoperfluoroalkane onto 2-methyl-propene (Scheme 5), a decomposition product from TBPPi. This was evidenced in the ¹³C NMR spectrum (Fig. S36†) by the presence of the characteristic triplet (²J_CF = 19 Hz) of doublets (³J_CF = 8 Hz) centered at 47.5 ppm and assigned to –CH₂CF₂–. The other two signals which can be seen at 36.1 and 32.0 ppm are attributed to the methyl groups in iso-butyl and the tertiary carbon atom connected to the iodine atom, respectively. It is known that TBPPi generates two radicals: tert-butoxyl˙ and tert-butyl-CO₂˙.^58,63 The positive mode in ASAP analysis also confirms the addition of oligo(HFPO)-I onto 2-methyl-propene with the main distribution observed between 900 and 1950 m/z corresponding to the (F[CF(CF₃)CF₂O]_nCF(CF₃)CF₂CH₂C(CH₃)₂ + CH₃CN)+ radical cation formed with acetonitrile in the probe.

Scheme 5 Conversion of poly(hexafluoropropylene oxide) primary iodide to undesired side-product using TBPPI as the radical initiator.

This work carried out with the 1-iodoperfluoroalkanes and initiators provides a clear explanation for the results listed in Table 2. Most reactions were biphasic due to the immiscibility of the reactants. During the reaction with allyl alcohol and BPO in acetic acid, a dominating amount of iodobenzene was formed and could be more soluble in oligo(HFPO)-CF(CF₃)CF₂I or C₆F₁₃I than acetic acid. This led to a perfluoroalkyl iodide conversion into an aromatic side-product by reaction of the oligo(HFPO)-CF(CF₃)CF₂I- or C₆F₁₃I-derived radicals onto the iodobenzene. As for reactions with allyl-O-PEG-OCH₃ initiated by BPO, the solubility of the iodobenzene in C₆F₁₃I is higher than that of the allyl-PEG. This difference in solubility may explain the failure of the reactions (yield = 0%). In the reaction involving oligo(HFPO)-CF(CF₃)CF₂I, allyl-O-PEG-OCH₃ may have some partial solubility in the fluorinated ether which reduced the production of iodobenzene and thus minimized side-products. In summary, BPO only formed by-products with C₆F₁₃I, but not with oligo(HFPO)-CF(CF₃)CF₂I (at least not detectable at the scale of the reactions).

In the case of the reactions involving oligo(HFPO)-CF(CF₃)CF₂I and allyl-O-PEG-OCH₃, if TBPPi is used as the initiator, 2-methylpropene was quickly released at 90 °C (Scheme 5) and oligo(HFPO)-CF(CF₃)CF₂I reacted faster with the smaller 2-methylpropene rather than with the bulky and sterically hindered allyl-O-PEG-OCH₃.

2.4. Reduction of oligo(HFPO)-CH₂CHICH₂-oligo(PEG) using tributyltin hydride

Kaplánek et al.^9a reported the formation of ω-perfluorohexyl poly(ethylene oxide) diblock co-oligomer [C₆F₁₃CH₂CHICH₂PEG] by the reduction of the iodine atom in the presence of zinc, NiCl₂·6H₂O in THF–H₂O in 62% yield. The present study used tributyltin hydride which was efficient in previous studies to achieve the desired reduction.^64,65 The successful reduction can be evidenced from the ¹H-NMR spectrum (Fig. S38†) with the high field shift from 4.36 ppm (assigned to –CHI–, Fig. S28†) to 1.89 ppm characteristic of the central –CH₂– group. The ¹⁹F-NMR spectrum (Fig. S39†) is similar to that of the iodide-containing diblock cooligomer.

The ¹³C-NMR spectrum (Fig. S40†) exhibits the absence of the signal assigned to the methine –CHI– at 18 ppm that underwent a low field shift to 20.55 ppm for the new central methylene group. In addition, the methylene group in –CF₂CH₂ shifted from 37.65 to 27.80 ppm due to the reduction of the iodine atom. The yield of this reaction was 77%. The positive mode of ASAP and MALDI-TOF-MS spectra (Fig. 6 and S41–42†) also confirmed the reduction of the material and showed the connectivity of oligo(HFPO) to oligo(PEG) evidenced by the presence of 166 and 44 m/z for HFPO and ethylene oxide repeat units, respectively.

Fig. 6 Positive ion mode MALDI-TOF-MS spectrum of oligo(HFPO)-CH₂CH₂CH₂-oligo(PEG) (using as a matrix DCTB and LiCl as the cationizing agent), 1807 m/z is x = 8 and y = 5. The insert expansion m/z between 1850 and 2100 displays 166 and 44 m/z repeat units for HFPO [CF(CF₃)CF₂O] and ethylene oxide (CH₂CH₂O), respectively.

2.5. Surface tension measurements (critical micellar concentration, CMC)

The oligo(HFPO)-b-oligo(PEG) diblock co-oligomer is as an attractive amphiphilic molecule, the surfactant behavior of which deserves to be studied. It is partially water-soluble up to ca. 0.3 g L⁻¹. Surface tension properties of this diblock co-oligomer were assessed by tensiometry and compared to the surface tension of commercially available ammonium perfluorooctanoate (APFO). Three surfactants were measured for comparison: APFO, oligo(HFPO)-CH₂CHICH₂-oligo(PEG) and oligo(HFPO)-b-oligo(PEG) diblock cooligomers. APFO, a commonly used surfactant, was regarded as a reference and was measured to have a surface tension of 37.7 mN m⁻¹ at the critical micellar concentration (CMC) of 3.77 g L⁻¹ (Fig. S43†). Oligo(HFPO)-CH₂CHICH₂-oligo(PEG) and oligo(HFPO)-b-oligo(PEG) diblocks were measured to have surface tensions of 38.5 mN m⁻¹ and 43.5 mN m⁻¹ at the cmc of ca. 0.13 g L⁻¹ and 0.04 g L⁻¹, respectively. The plot of their surfaces can be seen in Fig. 7. Interestingly, it can be noted that the surface tension properties of such a diblock co-oligomeric surfactant are much better than those of APFO in terms of the amount of material required to reach the critical micellar concentration. The oligo(HFPO)-b-oligo(PEG) diblock's cmc is ca. 94 times better than that of APFO and 3 times smaller than that of oligo(HFPO)-CH₂CHICH₂-oligo(PEG). In comparison, the cmcs of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS) are 4.45 and 15.70 g L⁻¹, respectively.⁶⁶ At room temperature, the oligo(HFPO)-b-oligo(PEG) diblock co-oligomer is a potential surfactant, competitor to APFO.

Fig. 7 Surface tension measurements of new surfactant (left) oligo(HFPO)-CH₂CHICH₂-oligo(PEG) [CMC = 0.13 g L⁻¹] (and right) oligo(HFPO)-b-oligo(PEG)[CMC = 0.04 g L⁻¹]. Reference: APFO [CMC = 3.77 g L⁻¹] (Fig. S42†).

3. Conclusion

An original diblock co-oligomer based on oligo(hexafluoropropylene oxide) and oligo(ethylene oxide) was successfully synthesized by the radical addition of oligo(HFPO)-CF(CF₃)CF₂I onto an allyl-O-PEG-OCH₃ derivative followed by the selective reduction of the iodine atom. Careful optimization of the experimental conditions such as the nature of the initiators (TBPPi, AIBN, or BPO), temperature, and reaction time was required to synthesize the desired diblock co-oligomer. BPO was shown to be the most efficient source of radicals with regard to both yield and reaction time. The best overall yield obtained from the oligo(HFPO)-CF(CF₃)CF₂I precursor was 71%. The oligo(HFPO)-b-oligo(PEG) diblock co-oligomer exhibits very interesting surfactant behavior. Indeed, the surface tension of water reached 32.5 mN m⁻¹ for a surfactant concentration of 0.11 g L⁻¹ and a low critical micellar concentration (0.04 g L⁻¹ at room temperature). This newly formed diblock molecule has demonstrated relevant properties compared to those of the commercially available ammonium perfluorooctanoate (APFO). Further studies dealing with the synthesis of homologue surfactants containing various oligo(HFPO) and PEG chain lengths delving into deeper surface activity studies, absorption kinetics, and static and dynamic interfacial properties, as well as their bioaccumulation, decomposition and micellar behavior in solution, are in progress.

4. Experimental details

4.1. Materials

All materials were used without further purification or drying. Krytox® primary iodide, 1-iodoperfluoropropyl-2-oligo(hexafluoropropylene oxide) (oligo(HFPO)-CF(CF₃)CF₂I) and Freon 113 were kindly offered by DuPont (Wilmington, USA) and iodoperfluorohexane (C₆F₁₃I, purity 95%) was generously supplied by Atofina (now Arkema, Pierre-Benite, France). Ammonium perfluoro-octanoate (Daikin Industries, Ltd), allyl alcohol and anhydrous copper(II) acetate (Fluka, 98%), tributyl stannane (Sigma Aldrich, 97%), sodium dithionite (Na₂S₂O₄, Fisher Chemical), chloroform (Sigma Aldrich, 98%), recrystallized azobisisobutyronitrile (AIBN) in methanol (Fluka Analytical), tert-butyl peroxypivalate (TBPPi, TRIGONOX 25-C75, 75% TBPPi diluted to 63% in isodecane, Akzo Nobel), di-benzoyl peroxide (BPO) (Sigma Aldrich), polyethylene glycol methyl ether (Fluka Analytical, 500 amw), and allyl bromide (Sigma Aldrich).

4.2. Analyses

4.2.1. Gas chromatography (GC)mass spectrometry (MS). The total product mixture was analyzed using a Shimadzu GC (GC-2010 Plus) and a quadripole MS (GCMS-QP2010 SE) equipped with a Zebron ZB-5 ms column, 20 m × 0.18 mm id, 0.18 μm 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 and then the heating rate was 25 °C min⁻¹ until reaching 250 °C and holding at 250 °C for 2 minutes. The total pressure was 108 kPa, the total flow 25.9 mL min⁻¹, 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.

4.2.2. Atmospheric pressure solids analysis probe (ASAP)-time-of-flight mass spectrometry (TOF-MS). Full MS analyses were performed on a SYNAPT G2 HDMS QTOF Mass Spectrometer fitted with an Atmospheric Solids Analysis Probe (Waters Corp., Manchester, UK). The samples were applied directly to the exterior of glass capillary that is attached to the ASAP probe. The sample in the gas phase was ionized by the proximity to a corona discharge needle. Ions were then passed from the atmospheric pressure region to the mass spectrometer. ASAP mass spectra were acquired in positive and negative ion modes over the m/z 50–4500 range for 3 min for MS spectra. A nitrogen gas flow of 500 L h⁻¹ was ramped from 50 °C to 650 °C at 200 °C min⁻¹ for thermal desorption. The corona discharge voltage was 4 μA and the sampling cone voltage was 40 V.

4.2.3. Matrix assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF-MS). Experiments were determined with a Bruker Ultraflex III with a positive ionization method for compounds containing a higher content of hydrocarbons and a negative ionization method for compounds higher in fluorocarbon content. For sample preparation, samples were 1/50 diluted in CH₂Cl₂. 1 mL of LiCl–MeOH (10 mg mL⁻¹) was deposited on the target first, and dried. Then, 0.5 mL of the sample solution was spotted on top of the LiCl layer and dried, and finally 0.5 mL of the trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB) matrix was applied on top and dried.

4.2.4. Nuclear magnetic resonance (NMR) spectroscopy. The structure of the products was determined by NMR spectroscopy at room temperature (25 °C). NMR spectra were recorded on Bruker AC-400 instruments using deuterated benzene or dimethylformamide capillary as internal references, respectively. The experimental conditions were accomplished using TopSpin 2.1 operating at 400.13 (¹H), 376.46 (¹⁹F), 100.62 (¹³C) MHz. Flip angle 90° for ¹H and ¹³C and 30° for ¹⁹F-NMR; acquisition time 3.96 s (¹H), 0.87 s (¹⁹F), 1.36 s (¹³C); pulse delay 2 (¹H), 4 s (¹⁹F and ¹³C); scans 128 (¹H), 16 (¹⁹F), 6144 (¹³C); and pulse width of 12.50 (¹H), 13.0 (¹⁹F), and 9.0 (¹³C) μs. The letters s, d, t, q, and sext stand for singlet, doublet, triplet, quartet, and sextet, respectively.

4.2.5. Surface tension for critical micelle concentration (CMC). The characterization of the surface tension was carried out on a Dataphysics DCAT tensiometer equipped with a DuNouy ring made of a platinum–iridium alloy. To assess the surface tension (SFT), the tensiometer first detected the surface of the test liquid (30 mL) by moving the sample vessel with the liquid until the balance detects a weight difference (since the probe is getting lighter when it dips into the liquid). The data are acquired as the position where the ring meets the surface. Then, the ring dips into the liquid to the defined position (immersion depth). The stage moved down to the stored position of the surface and waited until a constant SFT value was reached. The solution was allowed to equilibrate in the apparatus and then surface tension measurements were accomplished. The critical micelle concentration (CMC) was calculated as the intersection between the two straight lines emerging from high and low concentrations.

4.3. Syntheses

4.3.1 Synthesis of C₆F₁₃CH₂CHICH₂OH^9,47,48. Note: Oxygen was removed from the reactants (1-iodoperfluoroderivatives and allyl) by bubbling nitrogen through the systems for several minutes.

RXN #1: A round-bottomed flask (10 mL) was charged with C₆F₁₃I (0.5135 g, 1.15 mmol), allyl alcohol (0.1089 g, 1.87 mmol) and TBPPi (0.0264 g, 0.152 mmol), heated to 75 °C, for 2 h. After elimination of unreacted reactants by a rotary evaporator, the total product mixture was dissolved in CDCl₃ for NMR analysis. Yellow solid, yield by ¹⁹F-NMR = >99% yield, with >99% purity.

RXN #2: A round-bottomed flask (10 mL) was charged with C₆F₁₃I (0.501 g, 1.12 mmol), allyl alcohol (0.0667 g, 1.148 mmol), AIBN (0.0037 g, 0.0225 mmol), Na₂S₂O₄ (0.0503 g, 0.289 mmol), and H₂O (0.0663 mL), heated to 90 °C for 4 h. The product was dissolved in CDCl₃. Slightly yellow solid, yield by ¹⁹F-NMR = 92% yield, with 93% purity.

RXN #3: A round-bottomed flask (10 mL) fitted with a reflux condenser was charged with C₆F₁₃I (0.501 g, 1.12 mmol), allyl alcohol (0.0976 g, 1.68 mmol), BPO (0.2442 g, 1.008 mmol), copper(II) acetate (0.0102 g, 0.055 mmol), and glacial acetic acid (0.3 mL), heated to 90 °C for 8 h. The product was dissolved in CDCl₃. Yellow solid, yield by ¹⁹F-NMR = 0% yield.

4.3.1.1. Characterization. TBPPI: ¹H NMR (400 MHz, CDCl₃, 25 °C) (Fig. S3†): δ = 4.31 (quin, –CH₂CHICH₂OH, ³J_HH = 6.57 Hz, 1H), 3.80, 3.74 (–CH₂CHICH_aH_bOH, ²J_HaHb = 12.13 Hz, 1H), 3.78, 3.73(d, –CH₂CHICH_aH_bOH, ²J_HbHa = 12.13 Hz, 1H) 3.01 (m, –CF₂CH_aH_bCHI–, 1H), 2.65 (m, –CF₂CH_aH_bCHI–, 1H), 2.95(–CH₂OH, 1H); ¹⁹F NMR (376.41 MHz, CDCl₃, 25 °C) (Fig. S5†): δ = −81.06 (CF₃–, ³J_FF = 10.33, ⁴J_FF = 2.30 Hz, 3F), −126.36 (m, CF₃CF₂(CF₂)₄CH₂–, 2F), −123.77 (m, –CF₂(CF₂)₃CH₂–, 2F), −123.05 (m, –CF₂(CF₂)₂CH₂–, 2F), −121.97 (m, –CF₂CF₂CH₂–, 2F), −113.17, −114.20 (dm, –CF₂CH₂–, ²J_FF = 144.68 Hz, 2F); ¹³C NMR (101 MHz, CDCl₃, 25 °C) (Fig. S7†): δ = 118.49 (qt, CF₃CF₂– ¹J_CF = 288.34 Hz, ²J_CF = 33.66 Hz), 117.70 (tt, –CF₂CF₂CH₂–, ¹J_CF = 257.61 Hz, ²J_CF = 32.20 Hz), 110.87 (m, CF₃(CF₂)₄CF₂–, 4C), 67.78 (s, –CH₂OH, 1C), 37.18 (t, –CF₂CH₂CHI–, ²J_CF = 20.49 Hz, 1C), 20.4 (s, –CH₂CHICH₂OH, 1C).

4.3.2. Synthesis of allyl-polyethylene glycol [PEG]-methyl ether (CH₂=CHCH₂O[CH₂CH₂O]_9.5CH₃). The same equipment as above was used and dried and the reactions were carried out under a nitrogen atmosphere. To the three neck round-bottomed flask (100 mL) containing a magnetic stirrer, sodium hydride (80% by mass) in mineral oil (0.5125 g, 17.129 mmol) was weighed, dissolved in dry toluene (5 mL) and placed in an ice bath. A mixture composed of PEG₅₀₀ (5.004 g, 10.58 mmol) in 50 mL toluene (dried by distillation) was added drop-wise into the reaction medium under vigorous stirring and then stirred for 2 h at room temperature. Then, the total product mixture was cooled in an ice bath and allyl bromide (3.849 g, 31.81 mmol) was slowly added. The mixture was stirred at room temperature for 24 h. The resulting mixture was filtered over silica gel. The liquid part was dissolved in 100 mL water and extracted four times with 20 mL portions of ether. The product was dried under high vacuum (0.128 × 10⁻³ mbar). The product was a colourless liquid; the isolated yield was 3.5044 g, 65%.
4.3.2.1. Characterization. ¹H NMR (400 MHz, C₆D₆, 25 °C) (Fig. S1†) δ = 5.85 (ddt, CH_aH_b=CH_cCH₂–, ³J_HcHb(trans) = 17.34 Hz, ³J_HcHa(cis) = 10.36 Hz, ³J_HcH(CH2) = 5.81 Hz, 1H), 5.20 (ddt, CH_aH_b=CH_cCH₂–, ²J_HbHa = 1.77 Hz, ³J_HbHc(trans) = 17.18 Hz, ⁴J_HbH(CH2) = 1.77 Hz, 1H), 5.11 (dm, CH_aH_b=CH_cCH₂–, ³J_HaHc(cis) = 10.36 Hz, ²J_HaHb = 1.77 Hz, 1H), 3.96 (dm, CH_aH_b=CH_cCH₂O–, ³J_H(CH2)Hc = 5.56 Hz, 2H), 3.53 (t, –OCH₂CH₂OCH3, ³J_HH = 5.5 Hz, 2H), 3.55–3.61 (m, –CH₂O–, 19 × 2H), 3.31 (s, –OCH₃, ⁴J_HH = 0.72 Hz, 3H), 3.48 (t, –CH₂OCH₃, ³J_HH = 5.5 Hz, 2H); ¹³C NMR (101 MHz, C₆D₆, 25 °C) (Fig. S2†) δ = 135.97 (s, –CH=, 1C), 116.0 (s, =CH₂, 1C), 72.07 (s, 1C, CH₂–OCH₂CH=CH₂, 1C), 72.07 (s, 1C, CH₂–OCH₃, 1C), 70.74 (s, –CH₂–O, 19 × 1C), 58.65 (s, 1C, CH₃).

4.3.3. Synthesis of C₆F₁₃CH₂CHICH₂O(CH₂CH₂O)_9.5CH₃. RXN #4: C₆F₁₃I (0.160 g, 0.358 mmol), allyl-O-PEG-OCH₃ (0.2112 g, 0.417 mmol) and TBPPi (0.050 g, 0.1808 mmol) were heated to 75 °C in a round-bottomed flask. After 2 h, additional TBPPi was added to the reaction (0.01 g, 0.057 mmol) and the reaction was then stopped after 8 h. The product was dissolved in CDCl₃. Brown solid, yield by ¹⁹F-NMR = 79% yield, with 84% purity.

RXN #5: C₆F₁₃I (0.093 g, 0.208 mmol), allyl-O-PEG-OCH₃ (0.2113 g, 0.417 mmol), and AIBN (0.001 g, 0.0042 mmol), heated to 90 °C for 8 h. The product was dissolved in CDCl₃. Brown solid, yield by ¹⁹F-NMR = 37% yield, with 51% purity.

RXN #6: A round bottom flask (50 mL) equipped with a reflux condenser contained C₆F₁₃I (0.093 g, 0.2084 mmol), allyl-O-PEG-OCH₃ (0.2113 g, 0.417 mmol), benzoyl peroxide (0.101 g, 0.413 mmol), copper(II) acetate (0.0039 g, 0.0214 mmol), and glacial acetic acid (10 mL) was heated at 90 °C for 8 h. The product was dissolved in CDCl₃. Dark brown solid, yield by ¹⁹F-NMR = 0% yield.

4.3.3.1. Characterization. ¹H NMR (400 MHz, DMSO capillary, 25 °C) (Fig. S11†): δ = 4.35 (m, –CH₂CHICH₂OH, 1H), 3.80, 3.74 (–CH₂CHICH_aH_bO–, ²J_HaHb = 10.86 Hz, 1H), 3.65, 3.64 (d, –CH₂CHICH_aH_eOH, ²J_HbHa = 10.86 Hz, 1H), 3.16 (s, –OCH₃, ⁴J_HH = 0.72 Hz, 3H), 3.35 (t, –OCH₂CH₂OCH₃, ³J_HH = 5.3 Hz, 2H), 3.33 (t, –CH₂OCH₃, ³J_HH = 6.3 Hz, 2H), 3.5–3.4 (m, –CH₂O–, 19 × 2H), 3.14–3.03 (m, –CF₂CH_aH_bCHI–, 1H), 2.70–2.59 (m, –CF₂CH_aH_bCHI–, 1H); ¹⁹F NMR (376.41 MHz, DMSO capillary, 25 °C) (Fig. S12†): δ = −80 (t, CF₃–, ³J_FF = 9.4 Hz, 3F), −112.24 (d, –CF_aF_bCH₂–, ²J_FaFb = 273.06 Hz, 1F), −112.83 (d, –CF_aF_bCH₂–, ²J_FbFa = 273.06 Hz, 1F, 2F), −120.73 (m, –CF₂CF₂CH₂–, 2F), −121.78 (m, –CF₂(CF₂)₂CH₂–, 2F), −122.62 (m, CF₂(CF₂)₃CH₂–, 2F), −125.15 (s, CF₃CF₂(CF₂)₄CH₂–); impurity: −63.78 (m, CF₂I, 2F), −112.59 (m, –CF₂CF₂I), −120.08 (m, –(CF₂)₂I, 2F); ¹³C NMR (101 MHz, CDCl₃ capillary, 25 °C) (Fig. S13†): δ = −122 to −104 (m, C₆F₁₃, 6C), 76.05 (s, –CHICH₂O–, 1C), 71.85 (s, –CH₂OMe, 1C), 70.47 (s, –OCH₂–, 19 × 1C), 70.40 (s, –OCH₂CH₂OMe), 58.92 (s, –OCH₃, 1C), 14.39 (s, –CH₂CHICH₂O–, 1C), 37.34 (t, ²J_CF = 20.83 Hz, –CF₂CH₂CHI–, 1C).

4.3.4. Synthesis of F[CF(CF₃)CF₂O]_8.9CH₂CHICH₂OH.
4.3.4.1. General synthetic procedure. 1-Iodo-2 oligo(hexafluoropropylene oxide) perfluoropropane (oligoHFPO-I) (M = 1773 g mol⁻¹), allyl alcohol, and initiator were weighed in appropriate amounts. Oxygen was displaced by purging nitrogen into the reaction mixture for 10 minutes. All reactions were carried out under positive nitrogen pressure and heated using an oil bath. Initiators were added to the reaction mixtures in regular periods according to their half-lives and reaction temperatures. The progress of the reaction was monitored by GC/MS using the starting iodide as the reference. After reaction, all products were separated by an acetone extraction in the same reaction flask. The soluble side-products were first separated by decantation, followed by filtration through a Teflon® filter (2 mm porosity). The final product was dried at ca. 40 °C under vacuum (ca. 140 × 10⁻³ mbar) for 20 minutes.
4.3.4.2. Initiator (tert-butyl peroxypivalate, TBPPi). RXN #7: In a 10 mL round-bottomed flask, oligo(HFPO)-CF(CF₃)CF₂I (1.006 g, 0.564 mmol), allyl alcohol (0.0438 g, 0.754 mmol), and TBPPi (0.0020 g, 0.0113 mmol) were heated to 75 °C for 48 h. TBPPi was transferred (0.0020 g, 0.0115 mmol) every 4 h during the day for a total of six injections. Allyl alcohol (0.0438 g, 0.754 mmol) was also added 6 times. The reaction mixture was washed 3 times with 3 mL acetone and one time with 3 mL ethanol. Yellow liquid, yield by ¹⁹F-NMR = 88%, with 98% purity.

RXN #8: In a 10 mL round-bottomed flask, oligo(HFPO)-CF(CF₃)CF₂I (1.006 g, 0.564 mmol), allyl alcohol (0.0438 g, 0.754 mmol), and TBPPi (0.00197 g, 0.0113 mmol) were heated to 90 °C for 38 h. TBPPi was transferred (0.00263 g, 0.0151 mmol) every 4 h during the day for a total of 18 injections. Allyl alcohol (0.0503 g, 0.867 mmol) was also added 18 times. The reaction mixture was washed 3 times with 3 mL acetone and one time with 3 mL ethanol. Yellow liquid, yield by ¹⁹F-NMR = 80%, with 90.5% purity.

RXN #9: In a 25 mL round-bottomed flask, oligo(HFPO)-CF(CF₃)CF₂I (10.090 g, 5.694 mmol), allyl alcohol (0.438 g, 7.54 mmol), and TBPPi (0.0020 g, 0.0113 mmol) were heated to 75 °C for 32 h. TBPPi was transferred (0.0355 g, 0.204 mmol) every 4 h during the day for a total of 16 injections. Allyl alcohol (0.0820 g, 1.411 mmol) was also added 16 times. The reaction mixture was washed 3 times with 3 mL acetone and one time with 3 mL ethanol. Yield by ¹⁹F-NMR = 0%.

4.3.4.3. Initiator (azobisisobutyronitrile, AIBN). RXN #10: 1° oligo(HFPO)-CF(CF₃)CF₂I (2.034 g, 1.147 mmol), allyl alcohol (0.0883 g, 1.52 mmol), and AIBN (0.0037 g, 0.0226 mmol) were placed in a two necked pear-shaped flask (10 mL) and heated to 90 °C for 64 h. AIBN and allyl alcohol were both added 7 times at 4 h intervals. When the reaction was complete, extraction of by-products was carried out with acetone washings (5 × 25 mL). Yield by ¹⁹F-NMR = 86% yield, with 99% purity.

RXN #11: In a 10 mL round bottomed flask, oligo(HFPO)-CF(CF₃)CF₂I (1.004 g, 0.564 mmol), allyl alcohol (0.0437 g, 7.52 mmol), AIBN (20% in chloroform, AIBN (0.0019 g, 0.0116 mmol) and chloroform (0.0074 g, 0.0620 mmol) were heated to 90 °C for 72 h. AIBN in chloroform was added via a syringe 17 times every 4 h while allyl alcohol was added once after 32 h (0.044 g, 0.758 mmol). Extraction with acetone was achieved (5 times, 20 ml). Yield by ¹⁹F-NMR = 86% yield, with 97% purity.

RXN #12: In a 250 mL round bottomed flask, oligo(HFPO)-CF(CF₃)CF₂I (10.040 g, 5.6402 mmol), allyl alcohol (0.0437 g, 7.52 mmol), AIBN (20% in chloroform, AIBN (0.0185 g, 0.0116 mmol) and chloroform (0.074 g, 0.0620 mmol) were heated to 90 °C for 72 h. AIBN in chloroform was added via a syringe 35 times every 4 h while allyl alcohol was added one time after 32 h (0.044 g, 0.758 mmol). Extraction with acetone was achieved (5 times 20 ml). Yield by ¹⁹F-NMR = 89% yield, with 98% purity.

4.3.4.4. di-benzoyl peroxide (BPO). RXN #13: In a three-necked round-bottomed flask (50 mL) equipped with a reflux condenser, oligo(HFPO)-CF(CF₃)CF₂I (1.003 g, 0.564 mmol), allyl alcohol (0.065 g; 1.12 mmol), glacial acetic acid (solvent, 10 mL), copper(II) acetate (0.011 g, 0.0548 mmol), and BPO (0.273 g; 1.128 mmol) were heated to 90 °C for 8 h. The contents were cooled and the acetic acid was evaporated. A one-time extraction with water (10 mL) followed by a 4× extraction with ethanol (10 mL) achieved the final material. Yield by ¹⁹F-NMR = 0% yield.

RXN #14: In a three-necked round-bottomed flask (50 mL) equipped with a reflux condenser, oligo(HFPO)-CF(CF₃)CF₂I (1.003 g, 0.564 mmol), allyl alcohol (0.1310 g, 2.2559 mmol), glacial acetic acid (solvent, 10 mL), copper(II) acetate (0.005 g, 0.02738 mmol), and BPO (0.8767 g, 3.619 mmol) were heated to 90 °C for 8 h. The contents were cooled and the acetic acid was evaporated. A one-time extraction was accomplished with water (10 mL), followed by a 4× extraction with ethanol (10 mL). Yield by ¹⁹F-NMR = 0% yield.

4.3.4.5. Characterization. TBPPI: ¹H NMR (400 MHz, C₆D₆ capillary, 25 °C) (Fig. S17†): δ = 4.4 (s, –CH₂CHICH₂OH, 1H), 4.21 (s, –CH₂OH, 1H), 3.81 (–CH₂CHICH₂OH, 2H), 2.98 (m, –CF₂CH_aH_bCHI–, 1H), 2.72 (m, –CF₂CH_aH_bCHI–, 1H); (AIBN): ¹H NMR (400 MHz, C₆D₆ capillary, 25 °C) (Fig. S18†): δ = 4.31, 4.25 (s, –CH₂CHICH₂OH, 1H), 3.99 (s, –CH₂OH, 1H), 3.66 (–CH₂CHICH₂OH, 2H), 2.84 (m, –CF₂CH_aH_bCHI–, 1H), 2.54 (m, –CF₂CH_aH_bCHI–, 1H); ¹⁹F NMR (376.41 MHz, C₆D₆, 25 °C) (Fig. S20†): δ = −80 to −85 (m, CF(CF₃)CF₂O–), −80.15 (s, CF₃CF₂CF₂O–, 3F), −81.80 (s, CF₃CF₂CF₂O–, 2F), −110.92 (d, –CF(CF₃)CF_aF_bCH₂–, ²J_FF = 262.73 Hz), −112.54 (d, –CF(CF₃)CF_aF_bCH₂–, ²J_FF = 237.49 Hz), −113.99 (d, –CF(CF₃)CF_aF_bCH₂–, ²J_FF = 261.58 Hz), −129.80 (s, CF₃CF₂CF₂–, 2F), −146.80 (m, –CF(CF₃)CF₂–, 8.9 × 1F); ¹³C NMR (101 MHz, DMSO/C₆D₆ capillary, 25 °C) (Fig. S22†): δ = 118.0 (qd, ¹J_CF = 290.9, ²J_CF = 28.2 Hz, –OCF(CF₃)CF₂–), 117.6 (qt, ¹J_CF = 286.15 Hz, ²J_CF = 32.93 Hz, CF₃CF₂CF₂O–, 1C), 117.5 (qd, ¹J_CF = 286.15, ²J_CF = 34.40 Hz, CF₃CF₂CF₂O, 1C), 114.7 (td, ¹J_CF = 285.74, ²J_CF = 31.26 Hz, –OCF(CF₃)CF₂–, 8.9 × 1C), 105.2 (tsext, ¹J_CF = 267.03 Hz, ²J_CF = 36.68 Hz, CF₃CF₂CF₂O–, 1C), 101.8 (dsext, ¹J_CF = 270.7, ²J_CF = 36.7 Hz, –OCF(CF₃)CF₂–), 66.32 (s, –CH₂CHICH₂OH, 1C), 36.52 (m, –CF₂CH₂CHICH₂OH, 1C), 18.09, 17.90 (s, –CH₂CHICH₂OH, 1C).

4.3.5. Synthesis of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CH₂CH₂OH⁵⁷. To the three necked round-bottomed flask (250 mL), equipped with a dropping funnel and a reflux condenser, was added product from RXN #12, pHFPO-CH₂CHICH₂OH (10.012 g, 5.42 mmol), and AIBN (0.3166 g, 1.928 mmol) dissolved in trifluorotoluene (20 mL). The medium was saturated with nitrogen by a 10 minute bubbling. The reaction mixture was then heated at 80 °C. Tributyltin hydride (2.435 g, 8.12 mmol) was added drop-wise for 5 minutes. The reaction was monitored by GC/MS at 2 h and then at 5 h. Then, AIBN (0.037 g, 0.226 mmol) was added followed by additional tributyltin hydride (0.369 g, 1.265 mmol) again drop-wise over 5 minutes. The reaction mixture was stirred at 80 °C for 5 h and then cooled to room temperature. Then, trifluorotoluene was evaporated and the resulting product was extracted with acetone (10 mL) three times while the insoluble side-products were decanted. The solid side-products were removed by filtration and then all solvents were evaporated under high vacuum. Isolated yield = 5.4438 g, 58% yield.
4.3.5.1. Characterization⁵⁶. ¹H NMR (400 MHz, CDCl₃): δ = 3.8 (t, ³J_HH = 5.6 Hz, –CH₂CH₂OH), 2.3 (m, –CF₂CH₂CH₂–), 1.9 (m, –CH₂CH₂CH₂–); ¹³C NMR (101 MHz, CDCl₃ capillary, 25 °C) δ = 60.8 (s, –CH₂OH, 1C), 27.9 (t, ²J_CF = 24.3, –CF₂CH₂–, 1C), 23.5 (s, –CH₂CH₂CH₂–, 1C); the ¹⁹F-NMR is similar to that of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CHICH₂OH. ASAP-TOF-MS spectrum (positive and negative, Fig. S44–45†).

4.3.6. Synthesis of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CHICH₂O(CH₂CH₂O)_9.5CH₃.
4.3.6.1. General procedure. 1-Iodo-2-oligo(hexafluoropropylene oxide) perfluoropropane (oligo(HFPO)-CF(CF₃)CF₂I), polyethylene glycol allyl (allyl-O-PEG-OCH₃), and the initiator were weighed in appropriate amounts. Oxygen was displaced by purging nitrogen into the reaction mixture for 10 minutes. All reactions were carried out under positive nitrogen pressure and heated using an oil bath. Initiators were added to the reaction mixtures in regular periods according to their half-lives and temperatures. The progress of the reaction was monitored by GC/MS using the starting iodide as the reference. When the reaction was complete, the GC/MS showed no oligo(HFPO) iodide. After the reaction, all products were separated by acetone extraction in the same reaction flask. The soluble side-products were first separated by decantation, followed by filtration through a Teflon® filter (2 mm porosity). The final product was dried at ca. 40 °C under vacuum (ca. 140 × 10⁻³ mbar) for 20 minutes and resulted in a light yellow oil.
4.3.6.2. Initiator (tert-butyl peroxypivalate, TBPPi). RXN #15: In a 10 mL round-bottomed flask equipped with a reflux condenser, oligo(HFPO)-CF(CF₃)CF₂I (1.0 g, 0.564 mmol), allyl-O-PEG-OCH₃ (0.437 g, 0.8619 mmol), and TBPPi (0.312 g, 1.13 mmol) were heated to 75 °C. TBPPi (3.202 g, 0.23 mmol) was added every 4 h for a total of 19 additions. The product was washed 5× with ethanol (10 mL). A yellow liquid was obtained but the yield by ¹⁹F-NMR spectroscopy was 0%.

RXN #16: In a 3 neck 100 mL round-bottomed flask equipped with a reflux condenser, oligo(HFPO)-CF(CF₃)CF₂I (10.008 g, 5.64 mmol), allyl-O-PEG-OCH₃ (4.373 g, 8.61 mmol), and TBPPi (0.312 g, 1.13 mmol) were heated to 75 °C for 64 h. TBPPi (0.312 g, 1.13 mmol) was added every 4 h for a total of 15 additions. The product was washed 3 times with ethanol (20 mL). Light yellow liquid, yield by ¹⁹F-NMR = 0% yield.

4.3.6.3. Initiator (azobisisobutyronitrile, AIBN). RXN #17: In a 2 necked round-bottomed flask, oligo(HFPO)-CF(CF₃)CF₂I (1.005 g, 0.564 mmol), allyl-O-PEG-OCH₃ (1.3110 g, 2.5858 mmol), and AIBN (0.002 g, 0.012 mmol) were heated to 90 °C for 80 h. AIBN (0.020 g, 0.12 mmol) was added 20 times every 4 h. After reaction, extractions with acetone (5 × 10 mL) were performed. A light honey brown liquid was obtained. A yield of 48% and a purity of 75 % were calculated by ¹⁹F-NMR.

RXN #18: In a two-neck round-bottomed flask equipped with a Claisen condenser and a magnetic stir bar, oligo(HFPO)-CF(CF₃)CF₂I (10.012 g, 5.64 mmol), allyl-O-PEG-OCH₃ (3.8228 g, 7.54 mmol), AIBN (0.0185 g, 0.113 mmol) and chloroform (0.074 g, 6.197 mmol) were added. The reaction mixture was bubbled for ca. 10 minutes with nitrogen and then placed in an oil bath that was progressively heated up to 90 °C. Additional allyl-O-PEG-OCH₃ (3.8228 g, 7.54 mmol) was transferred into the reaction mixture after 40 h. AIBN in chloroform (0.0185 g, 0.113 mmol, 0.074 g, 0.618 mmol) was added 73 times in total with the additions occurring every 2 h. Because complete conversion was not detected with NMR more of allyl-O-PEG-OCH₃ was added (4.023 g, 7.93 mmol) and AIBN in chloroform (0.185 g, 1.13 mmol, 0.74 g, 6.197 mmol) was added 14 times in total occurring every 2 h. The product mixture was washed with 10 mL of water and then 3 times with 20 mL of acetone. The product was dried using a high-vacuum pump and filtered through a ring filter. A yellow-brown oil was obtained. The yield by ¹⁹F-NMR = 50%, with 76% purity.

4.3.6.4. Initiator (di-benzoyl peroxide (BPO)). RXN #19: In a two-neck 10 mL round-bottomed flask equipped with a reflux condenser, oligo(HFPO)-CF(CF₃)CF₂I (1.0007 g, 0.564 mmol), allyl-O-PEG-OCH3 (1.4003 g, 2.7613 mmol), copper(II)acetate (0.0052 g, 0.0274 mmol), glacial acetic acid (3 mL) and BPO (0.5675 g, 2.343 mmol) were added and heated to 90 °C. After the reaction, acetic acid was evaporated and washed 7 times with acetone (10 mL). Green oil, yield by ¹⁹F-NMR = 51% yield, with 88% purity.

RXN #20: In a three-neck 100 mL round-bottomed flask equipped with a reflux condenser, oligo(HFPO)-CF(CF₃)CF₂I (10.007 g, 5.64 mmol), allyl-O-PEG-OCH3 (4.291 g, 8.46 mmol), copper(II) acetate (0.052 g, 0.274 mmol), glacial acetic acid (3 mL), and BPO (1.641 g, 6.77 mmol) were added and then heated to 90 °C. After 8 h, additional allyl-O-PEG-OCH3 (1.431 g, 2.821 mmol) and BPO (1.192 g, 4.91 mmol) were added. The number of allyl-O-PEG-OCH₃ and BPO additions was 5 times every 8 h. After the reaction the acetic acid was evaporated, and washed 7 times with acetone (10 mL). Dark green oil, yield by ¹⁹F-NMR = 71% yield, with 94% purity.

4.3.6.5. Characterization. ¹H NMR (400 MHz, CDCl₃ capillary, 25 °C) (Fig. S28†): δ = 4.36 (b, –CHI–, 1H), 3.79 (b, –CHICH₂O, 2H), 3.59 (b, –CH₂O, 19 × 1H), 3.46 (b, –CH₂CH₂OCH₃, 4H), 3.28 (s, CH₃O–, 3H), 3.09 (vb, –CF₂CH_aH_bCHI–, 1H), 2.58 (vb, CF₂CH_aH_bCHI–, 1H); ¹⁹F NMR (376.41 MHz, C₆D₆, 25 °C) (Fig. S29†): δ = −80 to −84 (CF(CF₃)CF₂O–), −84.04 (CF₃CF₂CF₂O–, 3F), −82.37 (CF₃CF₂CF₂O–, 2F), −112 to –117.5 (b, –CF(CF₃)CF₂CH₂CHI–, 2F), −132.03 (s, CF₃CF₂CF₂–, 2F), −146.54 (m, –CF(CF₃)CF₂–, 8.9 × 1F); ¹³C NMR (101 MHz, C₆D₆, 25 °C) (Fig. S30†): δ = 118.0 (qd, ¹J_CF = 290.9, ²J_CF = 28.2 Hz –OCF(CF₃)CF₂–), 117.6 (qt, ¹J_CF = 286.15 Hz, ²J_CF = 32.93 Hz, CF₃CF₂CF₂O–, 1C), 117.5 (qd, ¹J_CF = 286.15, ²J_CF = 34.40 Hz, CF₃CF₂CF₂O, 1C), 114.7 (td, ¹J_CF = 285.74, ²J_CF = 31.26 Hz, –OCF(CF₃)CF₂–, 8.9 × 1C), 105.2 (tsext, ¹J_CF = 267.03 Hz, ²J_CF = 36.68 Hz, CF₃CF₂CF₂O–, 1C), 101.8 (dsext, ¹J_CF = 270.7, ²J_CF = 36.7 Hz, –OCF(CF₃)CF₂–), 76.27 (s, –CH₂CHICH₂O–, 1C), 72.17 (s, CH₂–CH₂–OMe, 2C), 70.79 (s, 19 × 1C, –CH₂–O), 70.54 (s, –CH₂CH₂OMe, 1C), 58.65 (s, 1C, CH₃), 37.65 (m, –CF₂CH₂CHICH₂O–, 1C), 14.30 (s, –CH₂CHICH₂O–, 1C).

4.3.7. Synthesis of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CH₂CH₂O(CH₂CH₂O)_9.5CH₃. To a 100 mL three neck round-bottomed flask equipped with a magnetic stirrer, a reflux condenser and a dropping funnel, the diblock iodo copolymer (RXN #20, 4.03 g, 1.6999 mmol), AIBN (0.1066 g, 0.649 mmol) and trifluorotoluene (6 mL) were added. Oxygen was displaced from the reaction mixture by bubbling nitrogen gas through the reaction mixture for several minutes. The reaction mixture was placed in an 80 °C hot oil bath. Then tributyltin hydride 97% solution (0.505 mL, 1.8699 mmol) was added drop by drop over 2 minutes. The colour of the reactant changed from green to orange and then finally black. The reaction was stopped after 5 h. The product with the black suspension was centrifuged for 45 min at 8000 rpm at room temperature. The product was then decanted and finally filtered. To the slightly yellow liquid was added KF (0.108 g, 1.8587 mmol), and intensively stirred over 24 h at room temperature (ca. 25 °C). Next, 2 mL of Freon 113 was added and the contents were centrifuged for 45 min at 8000 rpm. The liquid product was decanted and filtered through a ring filter (pore size 0.2 μm, diameter 13 mm). All solvents were removed under rotational evaporation (50 mbar, 60 °C). The product was then filtered a final time with a ring filter. The final product was dried under high vacuum (800 × 10⁻³ mbar). The product is a light lemon yellow oil. Isolated yield 3.786 g, 77% yield.
4.3.7.1. Characterization. 1H NMR (400 MHz, CDCl₃ capillary, 25 °C) (Fig. S35†): δ = 3.5 9 (b, –CH₂O, 23 × 1H), 3.40 (s, CH₃O–, 3H), 2.20 (b, –CF₂CH₂CH₂–, 1H), 1.89 (b, –CF₂CH₂CH₂–, 1H); ¹⁹F NMR (376.41 MHz, C₆D₆, 25 °C) (Fig. S36†): δ = −80 to −84 (CF(CF₃)CF₂O–), −84.04 (CF₃CF₂CF₂O–, 3F), −82.37 (CF₃CF₂CF₂O–, 2F), −112 to −118 (b, –CF(CF₃)CF₂CH₂CH₂–, 2F), −131.56 (s, CF₃CF₂CF₂–, 2F), −146.05 (m, –CF(CF₃)CF₂–, 9.9 × 1F); ¹³C NMR (101 MHz, C₆D₆, 25 °C) (Fig. S37†): δ = 118.0 (qd, ¹J_CF = 290.9, ²J_CF = 28.2 Hz, –OCF(CF₃)CF₂–), 117.6 (qt, ¹J_CF = 286.15 Hz, ²J_CF = 32.93 Hz, CF₃CF₂CF₂O–, 1C), 117.5 (qd, ¹J_CF = 286.15, ²J_CF = 34.40 Hz, CF₃CF₂CF₂O, 1C), 114.7 (td, ¹J_CF = 285.74, ²J_CF = 31.26 Hz, –OCF(CF₃)CF₂–, 8.9 × 1C), 105.2 (tsext, ¹J_CF= 267.03 Hz, ²J_CF = 36.68 Hz, CF₃CF₂CF₂O–, 1C), 101.8 (dsext, ¹J_CF = 270.7, ²J_CF = 36.7 Hz, –OCF(CF₃)CF₂–), 69.42 (bs, –CH₂O–, 21 × 1C), 57.59 (s, CH₃, 1C), 27.80 (m, –CF₂CH₂CH₂CH₂O–, 1C), 20.55 (s, –CH₂CH₂CH₂O–, 1C).

4.3.8 Initiator reactions.
4.3.8.1. Model reactions of 1-iodoperfluorohexane with radical initiators. General procedure: In a round-bottomed flask (10 mL), C₆F₁₃I (0.5002 g, 1.12 mmol) and the appropriate initiator (TBPPi, AIBN, or BPO) were added. Oxygen was removed by purging the mixture with nitrogen gas for at least 10 minutes. The reaction mixture was heated at the required temperature.

RXN #21: TBPPi (0.6702 g, 2.42 mmol) was added to C₆F₁₃I and heated at 75 °C for 3 h. ¹⁹F-NMR yield = 92% C₆F₁₃CH₂CI(CH₃)₂.

RXN #22: AIBN in chloroform (0.2002 g, 1.21 mmol; 1.002 g, 8.37 mmol) was added to C₆F₁₃I and heated to 90 °C for 3 h. No reaction.

RXN #23: AIBN in chloroform (0.3678 g, 2.24 mmol; 1.4712 g, 12.3 mmol), Na₂S₂O₄ (0.0436 g, 0.266 g), and H₂O (0.108) were added to C₆F₁₃I and heated to 90 °C for 3 h.

RXN #24: BPO (0.3315 g, 1.36 mmol), copper(II) acetate (0.0102 g, 0.056 mmol), and acetic acid (1 mL) were added to C₆F₁₃I heated to 90 °C for 3 h. ¹⁹F-NMR yield = >99% C₆F₁₃-Aromatic compounds (see Fig. S14†).

4.3.8.2. Model reactions of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂I with radical initiators. General procedure: In a round-bottomed flask (10 mL) equipped with a condenser, oligo(HFPO)-CF(CF₃)CF₂I (M = 1773 g mol⁻¹, 1.003 g, 0.564 mmol) and the appropriate initiator (TBPPi, AIBN, or BPO) were added. Oxygen was removed by purging the mixture with nitrogen gas for at least 5 minutes. The reaction mixture was heated at the required temperature.

RXN #25: TBPPi (0.329 g, 1.154 mmol) was added to oligo(HFPO)-CF(CF₃)CF₂I and heated to 75 °C for 4 h. ¹⁹F-NMR Yield = 95.5% of F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂CH₂CI(CH₃)₂ and 4.5% F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂H

RXN #26: AIBN (0.1895 g, 1.154 mmol) was added to oligo(HFPO)-CF(CF₃)CF₂I and heated to 90 °C for 4 h. No reaction.

RXN #27: AIBN in chloroform (0.1895 g, 1.154 mmol; 0.9475 g, 7.94 mmol) was added to oligo(HFPO)-CF(CF₃)CF₂I and heated to 90 °C for 4 h. No reaction.

RXN #28: AIBN in chloroform (0.1895 g, 1.128 mmol; 0.742 g, 6.22 mmol), Na₂S₂O₄ (0.0236 g, 0.133 g), H₂O (0.055 ml) was added to oligo(HFPO)-CF(CF₃)CF₂I and heated to 90 °C for 4 h. No reaction.

RXN #29: BPO (0.279 g, 1.154 mmol), copper(II) acetate (0.071 g, 0.389 mmol), and acetic acid (0.3 ml) were added to oligo(HFPO)-CF(CF₃)CF₂I and heated to 90 °C for 8 h. ¹⁹F-NMR yield = 7.5% F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂-Aromatics and 0.5% F[CF(CF₃)CF₂O]_8.9CF(CF₃)CF₂H.

Acknowledgements

The authors thank the E.I. du Pont de Nemours and Co., Inc. (especially Dr Jon Howell), the Daikin Industries, and the Elf Atochem for generous gifts of oligo(HFPO)-CF(CF₃)CF₂I, ammonium perfluorooctanoate, and C₆F₁₃I, respectively, the Fundation Balard-Chaire Total for supporting Professor Chadron Mark Friesen who spent one academic year at Ingénierie et Architectures Macromoléculaires, Institut Charles Gerhardt, and the European Commission for funding Jiří Lapčík with an Erasmus grant.

References

1. E. Kissa, Fluorinated Surfactants and Repellents, Marcel Dekker, Inc, New York, 2nd edn, 2001.
2. (a) C. Taylor, Annual Surfactants Review, 1999; (b) K. Skrabania, A. Laschewsky, H. von Berlepsch and C. Böttcher, Langmuir, 1999, 25, 7594–7606; (c) A. Laschewsky, Adv. Polym. Sci., 1995, 124, 1–39.
3. (a) M. P. Krafft and J. G. Riess, J. Polym. Sci., Part A: Polym. Chem., 2007, 45, 1185–1198; (b) M. P. Krafft and J. G. Riess, Chem. Rev., 2007, 109, 1714–1785; (c) J. G. Riess, Curr. Opin. Colloid Interface Sci., 2009, 14, 294–304.
4. (a) J. G. Riess, Chem. Rev., 2001, 101, 2797–2920; (b) J. G. Riess, Curr. Opin. Colloid Interface Sci., 2001, 14, 294–304.
5. G. Kostov, F. Boschet and B. Ameduri, J. Fluorine Chem., 2009, 130, 1192–1199.
6. (a) R. Renner, Environ. Sci. Technol., 2001, 35, 154A–160A; (b) H. J. Lehmler, Chemosphere, 2001, 58, 1471–1496.
7. E. I. du Pont de Nemours and Company . DuPont Global Strategy, Comprehensive Source Reduction, Presentation to EPA, January 31, 2005, U. S. EPA Administrative Record AR226-1914, 2005.
8. J. Kovarova and Z. Svobodova, Neuroendocrinol. Lett., 2008, 29, 599–608.
9. (a) R. Kaplánek, O. Paleta, I. Ferjentsiková and M. Kodíce, J. Fluorine Chem., 2009, 130, 308–316; (b) L. Caillier, E. Taffin de Givenchy, R. Levy, Y. Vandenberghe, S. Geribaldi and F. Guittard, J. Colloid Interface Sci., 2009, 332, 201–207.
10. L. Ahrens, J. Environ. Monit., 2011, 13, 20–31.
11. O. A. De Silva, C. Spencer, B. F. Scott, S. Backus and D. C. G. Muir, Environ. Sci. Technol., 2011, 45, 8060–8066.
12. A. A. Rand and S. A. Mabury, Environ. Sci. Technol., 2011, 45, 8053–8059.
13. (a) J. H. Simons, J. Electrochem. Soc., 1949, 95, 47–59; (b) J. Burdon, I. W. Parson and J. C. Tatlow, Tetrahedron, 1949, 28, 43–51; (c) F. G. Drakesmith and D. A. Hughes, J. Appl. Electrochem., 1949, 6, 23–38; (d) J. S. Clarke and A. T. Juhn, J. Electroanal. Chem., 1949, 85, 299–306.
14. B. Ameduri and B. Boutevin, Well-Architectured Fluoropolymers: Synthesis, Properties and Applications, Elsevier, Amsterdam, 2004.
15. A. B. Lindstrom, M. J. Strynar and E. L. Libelo, Environ. Sci. Technol., 2011, 45, 7954–7961.
16. J. C. D'Eon and S. A. Mabury, Environ. Sci. Technol., 2011, 45, 7974–7984.
17. United Nations Environment Programme Report of the Conference of the Parties of the Stockholm Convention on Persistent Organic Pollutants on the work of its fourth meeting. http://chm.pops.int/Portals/0/Repository/COP4/UNEP-POPS-COP. 4–38 (accessed November 10, 2002).
18. Y. Zushi, J. Hogarh and S. Masunaga, Clean Technol. Environ. Policy, 2011, 1–12.
19. O. Midasch, H. Drexler, N. Hart, M. W. Beckman and F. Angerer, J. Int. Arch Occup. Environ. Health, 2007, 80, 643–652.
20. E. Brede, M. Wilhelm, T. Göen, J. Müller, K. Rauchfuss and M. Kraft, et al. , Int. J. Hyg. Environ. Health, 2011, 213, 217–223.
21. K. Steenland, T. Fletcher and D. A. Savitz, Environ. Health Perspect., 2010, 118, 1100–1108.
22. J. R. Parsons, M. Saez, J. Dolfing and P. de Voogt, Reviews of Environmental Contamination Toxicology, Springer, New York, 2008, vol. 196, pp. 53–71.
23. G. W. Olsen, J. M. Burris, D. J. Ehresman, J. W. Froehlich, A. M. Seacat and J. L. Butenhoff, et al. , Environ. Health Perspect., 2007, 115, 1298–1305.
24. H. Hori, E. Hayakaiva, H. Einaga, S. Kutsuna, K. Koike and T. Ibusuki, et al. , Environ. Sci. Technol., 2004, 38, 6118–6124.
25. M. Houde, A. O. De Silva, D. C. G. Muir and R. J. Letcher, Environ. Sci. Technol., 2011, 45, 7962–7973.
26. G. Olsen, S. Chang, P. Noker, G. Gorman, D. Ehresman, P. Lieder and J. Butenhohh, Toxicology, 2009, 256, 65–74.
27. (a) H. Fromme, O. Midasch, D. Twardella, J. Angerer, S. Boehmer and B. Liebl, Int. Arch. Occup. Environ. Health, 2007, 80, 313–319; (b) T. Fromel and T. P. Knepper, Chemosphere, 2010, 80, 1387–1392.
28. H. Fromme, S. Tittlemier, W. Volkel, M. Wilhelm and D. Twardella, Int. J. Hyg. Environ. Health, 2009, 212, 239–270.
29. M. Lorber and P. P. Egeghy, Environ. Sci. Technol., 2011, 45, 8006–8014.
30. D. Trudel, L. Horowitz, M. Wormuth, M. Scheringer, I. T. Cousins and K. Hungerbuhler, Risk Anal., 2008, 28, 807–807.
31. L. S. Haug, C. Thomsen, A. L. Brantsaeter, H. E. Kvalem, M. Haugen and G. Becher, et al. , Environ. Int., 2010, 36, 772–778.
32. (a) U.S. Environmental Protection Agency. PFOA stewardship Program. http://www.epa.gov/oppt/pfoa/pubs/pfoarisk.htm (accessed April 5th, 2006); (b) J. M. Barroso, Off. J. Eur. Union Law, 2010, 223, 29–36.
33. V. Barry, A. Winquist and K. Steenland, Environ. Health Perspect., 2013, 121, 1313–1318.
34. C. Zhang, H. Yan, F. Li, X. Hu and Q. Zhou, J. Hazard. Mater., 2013, 260, 689–699.
35. R. C. Buck, J. Franklin, U. Berger, J. M. Conder, I. T. Cousins, P. de Voogt, A. A. Jensen, K. Kannan, S. A. Mabury and S. P. J. van Leeuwen, Integr. Environ. Assess. Manage., 2011, 7, 513–541.
36. T. Schuster, S. Schellenberger, R. Friedrich, M. Klapper and K. Müllen, J. Fluorine Chem., 2013, 154, 30–36.
37. R. Renner, Environ. Sci. Technol., 2006, 40, 12–13.
38. R. J. Dams, M. S. Terrazas, M. J. Sierakowski and G. G. I. Moore, US Patent US, 0148671 A1, 2006 (assigned to 3M).
39. S. K. Ritter, Chem. Eng. News, 2010, 88, 12–17.
40. A. Zaggia and B. Ameduri, Curr. Opin. Colloid Interface Surf., 2012, 17, 188–195.
41. M. Peschka, N. Fichtner, W. Hierse, P. Kirsch, E. Montenegro and M. Seidel, et al. , Chemosphere, 2008, 72, 1534–1540.
42. (a) M. Klapper, S. Nenov, R. Haschick and K. Müllen, Acc. Chem. Res., 2008, 41, 1190–1201; (b) F. E. Golling, T. Schuster, C. Geidel, L. Mammen, D. Vollmer, M. Klapper and K. Müllen, Advances in Fluorine containing polymers, Amer. Chem. Soc., Washington D.C., 2012; (c) T. Schuster, J. W. Krumpfer, S. Schellenberger, R. Friedrich, M. Klapper and K. Müllen, J. Colloid Interface Sci., 2014, 428, 276–285.
43. J. Eastoe, S. E. Rogers, L. J. Martin, A. Paul, F. Guittard, E. Guittard, R. K. Heenan and J. R. P. Webster, Langmuir, 2006, 22, 2034–2038.
44. A. Dramé, E. Taffin de Givenchy, S. Y. Dieng, S. Amigoni, M. Oumar, A. Diouf, T. Darmanin and F. Guittard, Langmuir, 2013, 29, 14815–14822.
45. P. Kappler and M. J. Lina, WO, 121060 A1, 2005 (assigned to Arkema).
46. G. Boutevin, D. Tiffes, C. Loubat, B. Boutevin and B. Ameduri, J. Fluorine Chem., 2012, 134, 77–84.
47. F. Boschet, G. K. Kostov, B. Boutevin, A. Jackson and B. Ameduri, Polym. Chem., 2012, 3, 217–223.
48. G. K. Kostov, F. Boschet, J. Buller, L. Badache, S. M. Brandstadter and B. Ameduri, Macromolecules, 2011, 44, 1841–1855.
49. (a) J. L. Howell and E. W. Pérez, US, 0073588 A1, 2003, assigned to Du Pont de Nemours; (b) L. Han, Y. Zhang, H. Li and L. Li, Colloids Surf., A, 2009, 34, 176–180; (c) S. V. Kostjuk, E. Ortega, F. Ganachaud, B. Ameduri and B. Boutevin, Macromolecules, 2009, 42, 612–619; (d) N. Durand, D. Mariot, B. Ameduri, B. Boutevin and F. Ganachaud, Langmuir, 2009, 27, 4057–4067.
50. United States Environmental Protection Agency, Computational Toxicology Research, http://actor.epa.gov/actor/GenericChemical?casrn=60164-51-4 (accessed June 20, 2014).
51. H. Li, H.-Q. Chen, S. Qing and Y.-M. Zhang, J. Polym. Res., 2011, 18, 645–650.
52. (a) C. M. Friesen, The Synthesis and Characterization of Lubricant Derivatives for New Industrial Applications and Improvements in Thermal Oxidative Environments, Ph.D. Dissertation, University of Alabama, Tuscaloosa, 2000; (b) G. A. Mountain, B. J. Jelier, C. Bagia, C. M. Friesen and J. M. Janjic, J. Fluorine Chem., 2014, 162, 38–44.
53. A. Bravo, H.-R. Bjørsvik, F. Fontana, L. Liguori, A. Mele and F. Minisci, J. Org. Chem., 1997, 62, 7128–7136.
54. M. P. Gelin and B. Ameduri, J. Polym. Sci., Part A: Polym. Chem., 2003, 41, 160–171.
55. (a) G. G. Furin, Russ. Chem. Rev., 2000, 69(6), 491–522; (b) P. M. Murphy, C. S. Baldwin and R. C. Buck, J. Fluorine Chem., 2012, 138, 3–23.
56. D. Lahiouhel, B. Ameduri and B. Boutevin, J. Fluorine Chem., 2001, 107(1), 81–88.
57. G. Gambaretto, L. Conte, G. Fornasieri, C. Zarantonello, D. Tonei, A. Sassi and R. Bertani, J. Fluorine Chem., 2003, 121(1), 57–63.
58. T. Nakamura, S. H. Thang, W. K. Busfield, S. Suyama, I. D. Jenkins and E. Rizzardo, J. Am. Chem. Soc., 1996, 118, 10824–10828.
59. C. M. Friesen, C. D. Montgomery and S. A. J. U. Temple, J. Fluorine Chem., 2012, 144, 24–32.
60. M. J. P. Smith, N. R. Cameron and J. A. Mosely, Analyst, 2012, 137(19), 4524–4453.
61. B. Ameduri and B. Boutevin, Macromolecules, 1990, 23, 2433–2439.
62. F. Ceretta, A. Zaggia, L. Conte and B. Ameduri, J. Fluorine Chem., 2012, 135, 220–224.
63. J. Guiot, B. Ameduri and B. Boutevin, Macromolecules, 2002, 35, 8694–8707.
64. G. Kostov, M. Holan, B. Ameduri and M. H. Hung, Macromolecules, 2012, 45, 7375–7387.
65. C. M. Friesen, K. N. Hay, J. L. Howell and D. A. Nyvall, US, 287559A1, 2006 (assigned to DuPont).
66. Q. Yu, R. Zhang, S. Deng, J. Huang and G. Yu, Water Res., 2009, 43, 1150–1158.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4py00965g

---