Activation of weak nucleophiles: Polyfluorocarbamates from polyfluoroalcohols via a fast radical reaction

Article abstract of DOI:10.1016/j.tetlet.2013.09.037

A new fast radical mechanism has been observed for the reaction of polyfluorinated alcohols and phenylisocyanate, very sensitive to the change of solvents and the concentration of reactants. The acidity of polyfluoroalcohols seems to be responsible for the observed new reactivity and evidences from kinetic studies, electron paramagnetic resonance, cyclic voltammetry, and photostimulation suggest that polyfluoroalkoxy radical is the key intermediate in the chain. To the best of our knowledge, it is the first time that a radical mechanism is described for the preparation of carbamates.

Full text of DOI:10.1016/j.tetlet.2013.09.037

Tetrahedron Letters 54 (2013) 6310–6313
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Activation of weak nucleophiles: polyfluorocarbamates
from polyfluoroalcohols via a fast radical reaction
Marc Soto a, Helena Comalrena a, Ursula Balduzzi a, Gonzalo Guirado a, Vega Lloveras b,c
,
José Vidal-Gancedo b,c, Rosa María Sebastián a, , Jordi Marquet a,
⇑
⇑
a Departament de Química, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193 Barcelona, Spain
b Institut de Ciència de Materials de Barcelona, Consejo Superior de Investigaciones Científicas, Cerdanyola del Vallès, 08193 Barcelona, Spain
c CIBER-BBN, Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 June 2013
Revised 9 September 2013
Accepted 11 September 2013
Available online 18 September 2013
A new fast radical mechanism has been observed for the reaction of polyfluorinated alcohols and phen-
ylisocyanate, very sensitive to the change of solvents and the concentration of reactants. The acidity of
polyfluoroalcohols seems to be responsible for the observed new reactivity and evidences from kinetic
studies, electron paramagnetic resonance, cyclic voltammetry, and photostimulation suggest that poly-
fluoroalkoxy radical is the key intermediate in the chain. To the best of our knowledge, it is the first time
that a radical mechanism is described for the preparation of carbamates.
Keywords:
Ó 2013 Elsevier Ltd. All rights reserved.
Polyfluorocarbamates
Single electron transfer
Radical chain reaction
Electron paramagnetic resonance
Cyclic voltammetry
A lot of attention has been given to the synthesis of fluorinated
organic materials.1 Polyfluorinated organic compounds have very
interesting properties such as high thermal and chemical stability,
hydrophobicity and oleophobicity, and low surface energy,2 and
their applications range from anesthetics to oxygen transport
agents among many others, especially in the fields of pharmaceutics
and materials science.3 In addition, the demonstrated usefulness of
biphasic fluorinated chemistry has increased the interest in their
synthesis and the study of their properties.4 On the other hand, car-
bamate fragments enter into the composition of numerous drugs
and pesticides and polyfluoroalkyl carbamates have been synthe-
sized and shown to have potential applications in these fields.5 In
the area of materials, highly fluorinated polymers have found appli-
cations as thermoplastics, elastomers, coatings, membranes, etc.
Among those highly fluorinated polymers, fluorinated polyure-
thanes represent a promising class of compounds.6 Therefore, the
study of fluorinated polyurethanes has attracted considerable
interest in recent years.7 On the whole, however, and with a few
exceptions, these compounds have not yet developed all their
potential, among other reasons, due to a shortage of basic knowl-
edge on the fundamental chemistry involved in their synthesis.
The standard approach to produce carbamates consists in the
reaction between an isocyanate and an alcohol.8 However,
polyfluorinated alcohols are significantly more acidic than normal
alcohols (ethanol pKa = 29.8; 2,2,2-trifluoroethanol pKa = 23.5; iso-
propanol pKa = 30.3; hexafluoro-iso-propanol pKa = 17.9 in DMSO)9
and show a very low nucleophilicity in polar reactions. Thus, poly-
fluorinated alcohols do not react with alkyl isocyanates and react
slowly with aromatic isocyanates at room temperature, catalysis
(amines5b,12c or dibutyltin derivatives10), or highly electrophilic
carbamates (aroxysulfonylcarbamates),11 being necessary for the
reactions to proceed. Alternative approaches include the catalyzed
reaction of fluorinated carbonates with amines12 or the electro-
chemically induced Hofmann rearrangement.13
In 2005, some of us reported a fast radical chain mechanism in
the polyfluoroalkoxylation of aromatics through NO2 group dis-
placement.14 Based mainly on theoretical studies we proposed a
mechanism that included the intermediacy of the polyfluoroalkoxy
radical. In the present Letter, we describe that in certain conditions,
polyfluorinated alcohols react with phenylisocyanate in a very fast
radical reaction that includes the polyfluoroalkoxy radical as a key
intermediate. Full evidences come from cyclic voltammetry (CV)
and Electron Paramagnetic Resonance (EPR) experiments. This
finding opens a new avenue for the reactivity of very non nucleo-
philic polyfluorinated alcohols.
Trifluoroethanol (TFE), 1H,1H-perfluoro-1-octanol, and hexa-
fluoro-iso-propanol (HFIP) react very fast with phenylisocyanate
(PhNCO) at room temperature in DMF (Scheme 1). In Table 1,
yields and reaction conditions are described.
⇑
Corresponding authors. Tel.: +34 935811029; fax: +34 935812477 (J.M.).
E-mail address: jordi.marquet@uab.cat (J. Marquet).
0040-4039/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2013.09.037

M. Soto et al. / Tetrahedron Letters 54 (2013) 6310–6313
6311
H
N
completely stopped (Table 1, entry 2), and no conversion at all was
observed even after 45 min. On the contrary, the reference reaction
of ethanol with phenylisocyanate in the same conditions showed
no effect upon the presence of the radical scavenger (Table 1, entry
4). It reached a 50% conversion after 45 min, thus confirming the
operation of a polar mechanism in this case, and suggesting the
existence of a radical chain mechanism in the reaction of TFE with
phenylisocyanate. This proposal of a fast radical reaction was sup-
ported by the fact that the reaction of TFE with phenylisocyanate
that failed completely in ACN, could be photostimulated (30% con-
version after 10 min under UV irradiation, medium vapor Hg lamp,
Pyrex filter).
N
O
C
C
O
RF
+
HO RF
O
Scheme 1. Synthesis of fluorinated phenylcarbamates.
Table 1
Conversion percentagea at 10 min for the reaction of phenylisocyanate with alcohols
in DMF
#
Alcoholb
Galvinoxylc (%) Conversion of PhNCOa
100d (mM) 10d (mM) 1d (mM)
1
2
3
4
5
6
7
8
9
CF3CH2OH
0
15
0
15
0
15
0
15
0
30
26
29
—
12
15
55
50
—
55
ꢀ0
15
17
80
10
63
19
19
85
1,7
—
This behavior was general for other polyfluorinated alcohols as
it is described in Table 1 (entries 5–8).
CH3CH2OH
—
35
7
47
2
—
The hypothesis of a radical chain mechanism in the reaction of
polyfluorinated alcohols with phenylisocyanate was deeply inves-
tigated for the case of TFE. EPR experiments supported the hypoth-
esis since radical intermediates were detected by means of a spin
trap. We used two different spin traps. 2-methyl-2-nitrosopropane
(MNP) is a general spin trap and is particularly suitable for the
detection of carbon-centered radicals while 3,3,5,5-tetramethyl-
pyrroline-N-oxide (TMPO) works very well with oxygen-centered
radicals.15 Using TMPO as a spin trap, a double triplet absorption
could be observed (a(1 N) = 12.70 G; a(1H) = 6.91 G; g = 2.0062)
that once properly simulated was attributed to the radical adduct
of TMPO with the trifluoroethoxy radical (CF3CH2OÅ), (Fig. 1a).16
The behavior, stability, and electrochemical properties of the
trifluoroethoxy radical were studied using CV (DMF + 0.1 M n-Bu4-
NBF4, see Supplementary data). A one electron oxidation wave at
1.05 V versus SCE appears in the forward scan (from 0.0 to 1.6 V),
whereas an irreversible reduction wave at À0.93 V versus SCE is
observed in the cathodic back scan (from 1.6 to À1.6 V). It is impor-
tant to remark that when the CV is scanned first in cathodic direc-
tion, the irreversible reduction wave at À0.93 V does not exist
while the oxidation wave at 1.05 V appears uncharged on the re-
verse scan. The voltammogram of trifluoroethoxy anion at higher
scan rates (5.0 V sÀ1, see Supplementary data), presents a revers-
ible one-electron oxidation wave with E° = 0.95 V versus SCE. That
is, if there are no chemical reactions linked to electron transfer,
one-electron wave is observed which corresponds to the formation
of the specie A (t1/2 = 1 ms appears).17 The specie A evolves to a
second one, more stable (B), at longer times (t1/2 > 1 ms). This sec-
ond specie, B, undergoes one electron reduction giving back the tri-
fluoroethoxide anion (being also TFE detected). Hence, at this point
it is fairly to think that specie A is the trifluoroethoxy radical
(where the radical is centered on oxygen), whereas specie B should
be the most stable isomeric radical specie where the radical is cen-
tered on the carbon (Scheme 2).18
CF3(CF2)6CH2OH
(CF3)2CHOH
(CH3)2CHOH
a
b
c
Conversion determined by GC using hexadecane as internal standard.
Initial ratio alcohol vs. phenyl isocyanate (2:1).
Molar percentage of the radical scavenger galvinoxyl.
Initial concentration of PhNCO, [PhNCO]0.
d
The efficiency shown by these reactions constituted a surprise
since polyfluorinated alcohols are well known poor nucleophiles.
Therefore, a preliminary mechanistic study was undertaken on
the reaction of TFE with phenylisocyanate, using as a reference
the corresponding reaction of ethanol (Table 1, entries 1 and 3).
Thus, no reaction was observed between ethanol and phenylisocy-
anate (1 mM alcohol, entry 3), whereas a 85% conversion of phen-
ylisocyanate into the corresponding phenylcarbamate was
obtained when using TFE in the same conditions and reaction time
(Table 1, entry 1).
The reaction of TFE with phenylisocyanate proceeds at high
rates in some polar aprotic solvents as DMF, DMSO, and HMPA,
but fails completely in acetonitrile, propylenecarbonate, and ace-
tone. On the contrary, the reference reaction of ethanol was much
less sensitive to the solvent. In addition, on increasing the reactants
concentration, from 10 to 100 mM (phenyl isocyanate concentra-
tion, keeping the relationship isocyanate/alcohol 1:2), the observed
effect in the reaction conversion is opposite for TFE and ethanol
(Table 1, entries 1 and 3).
The reaction of TFE with phenylisocyanate was carried out in
the standard conditions (10 mM of phenylisocyanate, and DMF at
room temperature) in the presence of a substoichiometric amount
(15% molar) of galvinoxyl as a radical scavenger. The reaction was
Figure 1. Experimental and simulated EPR spectra of (a) the TMPO adduct resulting from the reaction of TFE with PhNCO in the presence of TMPO, and (b) MNP and B
(Scheme 2) adduct.

6312
M. Soto et al. / Tetrahedron Letters 54 (2013) 6310–6313
- 1 e-
process (step 2 in the Scheme 3). Our results (effect of the solvent)
•
O
CF3
(A)
O
CF3
CF3
indicate that the solvent can help the proton transfer step through
its ‘donor properties’ (Lewis basicity). Thus, the influence of the
solvent in the success of the reaction does not correlate with any
of the typical solvent polarity parameters.20 Instead the results
correlate with the ‘Donor Number (DN)’ scale,21 and with the
‘hydrogen bonding index’ as defined in table 1 of Ref. 8. Following
previously proposed schemes for urethane formation reactions,8,22
the aprotic solvent is considered to solvate the complex of phenyl-
isocyanate/alcohol at the active hydrogen to form an ion pair
which in our case would undergo the electron transfer process
more easily.
The radical mechanism proposed in Scheme 3 can also justify
the peculiar concentration effects observed in our reactions
(Table 1). They are probably related with the termination steps
of the chain mechanism. Thus, the major termination steps in
our mechanism will be the isomerization process from oxygen
radical to carbon radical, or the dismutation reactions. All termina-
tion processes are bimolecular reactions and therefore their rates
are very sensitive to the concentration of the reaction mixture,
the chain being sustainable only in a narrow range of concentra-
tions. Our results (Table 1) indicate that a balance exists between
‘initiation-propagation’ and ‘termination’ steps. These groups of
steps must have an opposite effect on the variation of the global
rate with the concentration and on the sustainability of the radical
chain.
We have demonstrated that the reactions of polyfluoroalcohols
with phenylisocyanate can be carried out in very mild specific
conditions that allow a new fast radical chain mechanism to
become apparent. The acidity of polyfluoroalcohols seems to be
responsible for the observed new reactivity and preliminary mech-
anistic studies indicate that the polyfluoroalkoxy radical is the key
intermediate in the chain. Therefore, it seems we are in the pres-
ence of an example of a reaction type that is activated by Single
Electron Transfer and that by using the appropriate conditions
could be rather general for acidic poor nucleophiles. To the best
of our knowledge, it is the first time that a radical mechanism
has been proposed for the preparation of carbamates.
t1/2 < 1 ms
+ 1 e-
HO
HO
CF3
(B)
Scheme 2. Proposed voltammetric cycle for CF3CH2ONa.
The nature of species A was confirmed by EPR. A controlled
potential electrolysis of a 25 mM sodium trifluoroethoxide:MNP
spin trap (1:5) solution in DMF + 0.1 M n-Bu4NBF4 was electrolyzed
at 1.3 V for 10 min at 13 °C. Immediately after, an EPR spectrum of
the electrolyzed solution was recorded. The spectrum showed the
presence of a radical, that after simulation was attributed to the
adduct of species B (radical centered on the
a-carbon) and MNP,
(a(1N) = 11.21 G; a(1H) = 2.33 G; a(3F) = 2.39 G; g = 2.0062,
Fig. 1b), thus strongly supporting the voltammetric cycle proposed
in Scheme 2.
The radical scavenger experiments in the presence of galvinoxyl
were also carried out in the reactions of phenylisocyanate with
1H,1H-perfluorooctanol and with hexafluoro-iso-propanol (HFIP)
with the same result obtained in the reaction of PhNCO with TFE
and previously described (Table 1, entries 6 and 8 to compare with
entries 5 and 7, respectively).
In Scheme 3 a mechanistic proposal that agrees with the re-
ported facts is advanced. The propagation part of the mechanistic
scheme has a key intermediate, the alkoxy radical. The presence
of this radical in the reaction mixture has been shown by EPR
experiments and its properties established by CV studies. Thus,
our results suggest that this radical reacts with phenylisocyanate
or, in the absence of phenylisocyanate, quickly evolves to the
isomer with the radical centered on the
a-carbon. This a-carbon
centered radical seems to be more stable than the oxygen centered
isomer in agreement with what has been reported in the
literature.19
Theoretical calculations reported by some of us,14 indicate that
the trifluoroethoxy radical has nucleophilic properties and can
compete with the corresponding anion in the elementary step of
the propagation cycle. Therefore, the chain mechanism can have
a kinetic advantage over the polar mechanism once the proper
initiation conditions are reached.
Currently we are working in establishing the scope, significance,
and limitations of the reaction.
Acknowledgments
One of the main differences between ethanol and TFE relies in
their acidity, polyfluorinated alcohols being more acidic by more
than five orders of magnitude.9 Protonation of the phenylisocya-
nate by TFE (step 1 in the Scheme 3), even though it is not a
thermodynamic favorable process, can trigger the radical mecha-
nism since it would be coupled to an exothermic electron transfer
We acknowledge the financial support from the Ministerio de
Ciencia e Innovación and Ministerio de Ciencia y Tecnologia of
Spain (Projects CTQ2009-07469, CTQ2011-22649, Consolider
INGENIO 2010: CSD2007-00006) and Generalitat de Catalunya
(Projects 2009 SGR 1441 and 2009 SGR 1084). We also acknowl-
edge the support from the Universitat Autònoma de Barcelona
for the PIF scholarship to M.S.
Initiation
H
N
N
Supplementary data
(1)
HO
O
CF3
CF3
Ph
Ph
+
+
+
+
C
C
O
CF3
CF3
Ph
Ph
C
C
O
O
O
O
Supplementary data (experimental procedures, spectroscopic
data of products, EPR spectra, data table of kinetic studies and
conditions for electrochemical measurements) associated with this
article can be found, in the online version, at http://dx.doi.org/
10.1016/j.tetlet.2013.09.037. These data include MOL files and InChi-
Keys of the most important compounds described in this article.
H
N
H
N
(2)
O
Propagation
N
(3)
N
O
CF3
+
+
Ph
C
O
CF3
CF3
Ph
Ph
C
O
O
References and notes
H
N
N
O
CF3
O
CF3
(4)
HO
Ph
C
O
C
O
+
O
CF3
1. (a)Organofluorine Chemistry, Principles and Commercial Applications; Banks, R. E.,
Smart, B. E., Tatlow, J. C., Eds.; Plenum: New York, 1994; (b)Organofluorine
Chemistry; Uneyama, K., Ed.; Wiley-Blackwell, 2006; (c)Organofluorine
Chemistry: Fluorinated Alkenes and Reactive Intermediates; Chambers, R. D.,
Ed.; Springer, 2010.
Scheme 3. Proposed radical chain mechanism for the reaction of TFE with
phenylisocyanate.

M. Soto et al. / Tetrahedron Letters 54 (2013) 6310–6313
6313
2. (a)Handbook of Fluorous Chemistry; Gladysz, J. A., Curran, D. P., Horváth, I. T.,
Eds.; Wiley-VCH: Weinheim, 2004.
3. Hiyama, W. Organofluorine Compounds: Chemistry and Applications; Springer,
2000.
4. (a) Horwáth, I. T. Acc. Chem. Res. 1998, 31, 641–650. and references therein; (b)
Moreno-Mañas, M.; Pleixats, R. Acc. Chem. Res. 2003, 36, 638–643. and
references therein.
12. (a) Padilla, A.; Corma, A.; García, H.; Juarez, R. PCT Int. Appl., WO2009098327.;
(b) Padilla, A.; Ruiz, V.; Corma, A.; García, H.; Juarez, R. Eur. Pat. Appl.,
EP2036884, 2009.; (c) Bogolubsky, A. V.; Ryabukhin, S. V.; Pipko, S. E.; Lukin,
O.; Shivanyuk, A.; Mykytenko, D.; Tolmachev, A. Tetrahedron 2011, 67, 3619–
3623.
13. Matsumura, Y.; Satoh, Y.; Maki, T.; Onomura, O. Electrochim. Acta 2000, 45,
3011–3020.
5. (a) Authors not available, Abbott Laboratories, GB Patent No. 891567, 1962;
Chem. Abstr. 1962, 57, 5808.; (b) Studnev, Yu. N.; Frolovskii, V. A.; Krasnov, A.
P.; Rozantsev, G. G.; Stolyarov, V. P.; Kinash, O. F.; Fokin, A. V. Pharm. Chem. J.
2002, 36, 654–657.
14. Tejero, I.; Huertas, I.; González-Lafont, A.; Lluch, J. M.; Marquet, J. J. Org. Chem.
2005, 70, 1718–1727.
15. Alberti, A.; Macciantelli, D. In Electron Paramagnetic Resonance: A Practitioner’s
Toolkit; Brusolon, M., Giamello, E., Eds.; John Wiley & Sons, Inc.: Hoboken, New
Jersey, 2009; pp 287–323; (b) Castellanos, M. M.; Reyman, D.; Sieiro, C.; Calle,
P. Ultrason. Sonochem. 2001, 8, 17–22.
16. After 5 min from the preparation of a solution of PhNCO (20 mM) and TFE
(40 mM) in DMF, we added the spin trap TMPO in excess (>40 mM). The
resulting solution was carefully degassed with pure Argon and the EPR spectra
were recorded in a flat cell.
17. Andrieux, C. P.; Saveant, J. M. Techniques of Chemistry. In Investigation of Rates
and Mechanism of Reactions; Bernasconi, C. F., Ed.; Wiley: New York, 1986. Vol.
6, Chapter 2.1.
18. The reduction potential value is similar to previously described radical species
where the radical is centered on the carbon atom: Andrieux, C. P.; Gelis, L.;
Medebielle, M.; Pinson, J.; Saveant, J. M. J. Am. Chem. Soc. 1990, 112, 3509–3520.
19. (a) Papadimitriou, V. C.; Prosmitis, A. V.; Lazarou, Y. G.; Papagiannakopoulos, P.
J. Phys. Chem. A 2003, 107, 3733–3740; (b) Wang, Y.; Liu, J. Y.; Li, Z. S.; Wang, L.;
Sun, C. C. J. Comput. Chem. 2007, 28, 802–810.
6. (a) Yu, X. H.; Okkema, A. Z.; Cooper, S. L. J. Appl. Polym. Sci. 1990, 41, 1777–
1795; (b) Closkey, C. B.; Yip, C. M.; Santerre, J. P. Macromolecules 2002, 35, 924–
933; (c) Kim, Y. S.; Lee, J. S.; Ji, Q. Polymer 2002, 43, 7161–7170; (d) Tan, H.; Li,
J.; Guo, M.; Du, R.; Xie, X. Polymer 2005, 46, 7230–7239; Massa, T. M.; Yang, M.
L.; Ho, J. C.; Brash, J. L.; Santerre, J. P. Biomaterials 2005, 26, 7367–7376; (f)
Krafft, M. P.; Riess, J. G. J. Polym. Sci. Part A: Polym. Chem. 2007, 45, 1185–1198.
7. (a) Chen, K. Y.; Kuo, J. F. Macromol. Chem. Phys. 2000, 201, 2676–2686; (b)
Rolland, J. P.; Van Dam, R. M.; Schorzman, D. A.; Quake, S. R.; DeSimone, J. M. J.
Am. Chem. Soc. 2004, 126, 2322–2332; (c) Wang, L. F. Polymer 2007, 48, 7414–
7418; (d) Ge, Z.; Zhang, X.; Dai, J.; Li, W.; Luo, Y. J. Macromol. Sci. Part A: Pure and
Appl. Chem. 2009, 46, 215–221; (e) Jiang, G.; Tuo, X.; Wang, D.; Li, Q. J. Polym.
Sci.: Part A Polym. Chem. 2009, 47, 3248–3256; Turri, S.; Levi, M.; Trombetta, T.
US Patent: US7323435B1, 2008.; Hu, Z.; Finlay, J. A.; Chen, L.; Betts, D. E.;
Hillmyer, M. A.; Callow, M. E.; Callow, J. A.; DeSimone, J. M. Macromolecules
2009, 42, 6999–7007.
8. Delebecq, E.; Pascault, J. P.; Boutevin, B.; Ganachaud, F. Chem. Rev. 2013, 113,
80–118.
20. Reichardt, C. Solvent and Solvent Effects in Organic Chemistry, 2nd. Ed;
Weinheim: Wiley-VCH, 1988. pp 408–410.
9. Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456–463.
21. Gutmann, V. Coord. Chem. Rev. 1976, 18, 225–255.
10. Mashlyakovskiy, L.; Khomko, E.; Volynkina, N.; Tonelli, C. J. Polym. Sci.: Part A:
Polym. Chem. 2002, 40, 3771–3795.
22. Chang, M. C.; Chen, S. A. J. Polym. Sci. Part A: Polymer Chem. 1987, 25, 2543–
2559.
11. (a) El Kateb, M.; Beji, M.; Baklouti, J. J. Fluorine Chem. 1997, 81, 139–141; (b)
Beji, M.; Sbihi, H.; Baklouti, A.; Cambon, A. J. Fluorine Chem. 1999, 99, 17–24; (c)
Sbihi, H.; Beji, M.; Baklouti, A. J. Fluorine Chem. 2002, 115, 161–164.