Fluorocyclisation via I(I)/I(III) catalysis: A concise route to fluorinated oxazolines

Article abstract of DOI:10.3762/BJOC.14.88

Herein, we describe a catalytic fluorooxygenation of readily accessible N-allylcarboxamides via an I(I)/I(III) manifold to generate 2-oxazolines containing a fluoromethyl group. Catalysis is conditional on the oxidation competence of Selectfluor, whilst HF serves as both a fluoride source and Br?nsted acid activator. The C(sp³)–F bond of the mono-fluoromethyl unit and the C(sp³)–O bond of the ring are aligned in a synclinal relationship thereby engaging in stabilising hyperconjugative interactions with vicinal, electron-rich σ-bonds (σ_C–C→σ*_C–F and σ_C–H→σ*_C–O). This manifestation of the stereoelectronic gauche effect was established by X-ray crystallographic analysis of a representative example. Given the importance of fluorine in drug discovery, its ability to modulate conformation, and the prevalence of the 2-oxazoline scaffold in Nature, this strategy provides a rapid entry into an important bioisostere class.

Full text of DOI:10.3762/BJOC.14.88

Fluorocyclisation via I(I)/I(III) catalysis: a concise route to
fluorinated oxazolines
Felix Scheidt‡, Christian Thiehoff‡, Gülay Yilmaz, Stephanie Meyer,
Constantin G. Daniliuc§, Gerald Kehr and Ryan Gilmour*¶
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2018, 14, 1021–1027.
Organisch-Chemisches Institut, Westfälische Wilhelms-Universität
Münster, Corrensstraße 40, 48149 Münster, Germany
doi:10.3762/bjoc.14.88
Received: 02 March 2018
Accepted: 27 April 2018
Published: 09 May 2018
Email:
Ryan Gilmour* - ryan.gilmour@uni-muenster.de
* Corresponding author ‡ Equal contributors
§ X-ray crystallographer
This article is part of the Thematic Series "Hypervalent iodine chemistry in
organic synthesis".
¶ Homepage: http://www.uni-muenster.de/Chemie.oc/gilmour
Guest Editor: T. Wirth
Keywords:
catalysis; cyclisation; fluorination; gauche effect; hypervalent iodine;
© 2018 Scheidt et al.; licensee Beilstein-Institut.
License and terms: see end of document.
oxazolines
Abstract
Herein, we describe a catalytic fluorooxygenation of readily accessible N-allylcarboxamides via an I(I)/I(III) manifold to generate
2-oxazolines containing a fluoromethyl group. Catalysis is conditional on the oxidation competence of Selectfluor®, whilst HF
serves as both a fluoride source and Brønsted acid activator. The C(sp3)–F bond of the mono-fluoromethyl unit and the C(sp3)–O
bond of the ring are aligned in a synclinal relationship thereby engaging in stabilising hyperconjugative interactions with vicinal,
electron-rich σ-bonds (σC–C→σ*C–F and σC–H→σ*C–O). This manifestation of the stereoelectronic gauche effect was established
by X-ray crystallographic analysis of a representative example. Given the importance of fluorine in drug discovery, its ability to
modulate conformation, and the prevalence of the 2-oxazoline scaffold in Nature, this strategy provides a rapid entry into an impor-
tant bioisostere class.
Introduction
Marine and terrestrial natural product bioprospecting has estab- based on the 2-oxazoline building block constitute versatile
lished a broad spectrum of structurally complex, bioactive platforms for a range of biomedical applications ranging from
metabolites containing the venerable 2-oxazoline unit [1,2]. drug delivery through to tissue engineering [3,4]. Collectively,
This diversity is exemplified by the siderophore antibiotic the importance of the 2-oxazoline scaffold for translational
D-fluviabactin, the cytotoxic agent westiellamide, the anti- research, together with its strategic value in the design of chiral
fungal macrodiolide leupyrrin A1 and the antitumour com- ligands and auxiliaries [5-7], has culminated in a rich and inno-
pound BE-70016 (Figure 1). In addition, synthetic polymers vative arsenal of synthetic methods.
1021

Beilstein J. Org. Chem. 2018, 14, 1021–1027.
Figure 1: Selected examples of bioactive compounds containing the 2-oxazoline motif.
To contribute to the current catalysis ordnance for the prepara-
tion of 2-oxazolines, and provide a direct route to 5-fluoro-
methyl derivatives from simple unactivated alkenes, the fluoro-
cyclisation of N-allylcarboxamides facilitated by the in situ gen-
eration of p-TolIF2 was envisaged [8-11]. Since hydrogen and
hydroxy groups are often substituted by fluorine in molecular
editing processes [12], this transformation would provide facile
access to a bioisostere of the parent scaffold (Figure 1, right).
In recent years, I(I)/I(III) catalysis has emerged as a powerful
and expansive platform for the generation of structural com-
plexity [13-24]. Motivated by the noticeable absence of mild,
catalysis-based strategies to generate the vicinal difluoro-
ethylene motif directly from simple alkenes [25-27], we
recently exploited I(I)/I(III) catalysis to enable this transformat-
ion [28,29]. Employing Selectfluor® as the terminal oxidant, it
was possible to generate p-TolIF2 in situ from p-iodotoluene
Figure 2: The catalytic difluorination of alkenes (top) and the pro-
posed fluorocyclisation via the same I(I)/I(III) manifold (bottom).
and an inexpensive HF source [30-35]. This strategy proved to
be mild and general, smoothly converting terminal olefins to the
corresponding 1,2-difluoroethylene unit; a substructure that
may be considered a chiral, hybrid bioisostere of the Et and CF3 would allow the original reaction path to be intercepted to
groups (Figure 2, top) [36]. generate a 2-oxazoline with an exocyclic fluoromethyl unit.
Since the success of this process is contingent on the efficient Results and Discussion
generation of p-TolIF2 in situ, the platform lends itself to Optimisation: As a starting point, the conversion of N-allyl-
related oxidative transformations. To that end, it was envisaged benzamide (1) to the corresponding 2-phenyloxazoline 2 was
that the protocol could be effectively translated to the fluoro- investigated (Table 1). Reactions were performed in DCE
cyclisation of readily accessible N-allylcarboxamides (Figure 2, (0.2 mol·L−1) with 20 mol % catalyst loading, and using Select-
bottom). Whilst the initial phase of catalysis would resemble fluor® as the oxidant. An initial reaction screen, based on the
that of the catalytic difluorination, the presence of the amide conditions reported for our vicinal difluorination study [9],
1022

Beilstein J. Org. Chem. 2018, 14, 1021–1027.
Table 1: Optimisation of reaction conditions for a benchmark transformation (1→2).a
Entry
Concentration
[mol·L−1]
Solvent
Catalyst loading
[mol %]
Amine/HF ratio
Conversionb
[%]
Yield
[%]c
1
2
3
4
0.2
0.2
0.2
0.2
DCE
DCE
DCE
DCE
20
20
20
20
1:4.5
1:3
>95
50
46
<5
44
46
1:7.5
1:9.23
>95
>95
Pyr·HF
(6 equiv)d
5
0.2
DCE
20
>95
27
6
7
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
DCE
toluene
MeCN
THF
20
20
20
20
20
10
2.5
0
1:4.5
1:4.5
1:4.5
1:4.5
1:4.5
1:4.5
1:4.5
1:4.5
>95
>95
78
81
72
8
47
9
34
<5
10
11
12
13
DCM
DCM
DCM
DCM
>95
>95
40
>95
>95 (67)e
30
<5
<5
aStandard reaction conditions: N-allylbenzamide (200 µmol), catalyst p-iodotoluene, solvent, amine/HF source 1:1 (v/v), Selectfluor® (1.5 equiv),
ambient temperature, 24 h; bDetermined from the 1H NMR spectrum using ethyl fluoroacetate (1.0 equiv) as internal standard; cDetermined from the
19F NMR spectrum using ethyl fluoroacetate (1.0 equiv) as an internal standard; dReaction conducted with 1 mL of solvent; eYield after column chro-
matography on silica gel. Reduction in yield is due to hydrolysis. DCE: 1,2-dichloroethane.
began with an exploration of the effect of amine/HF ratio. This er, further decreasing the loading to 2.5 mol % demonstrates the
was deemed prudent due to the perceived likelihood that HF limits of the system (30% yield, Table 1, entry 12). Finally, for
also functions as a Brønsted acid activator in catalysis. Employ- completeness, the control experiment in the absence of p-TolI
ing an amine/HF ratio of 1:4.5, product formation was ob- was performed and confirms the role of this species in catalysis.
served (Table 1, entry 1, 46%). Reducing this ratio to 1:3 had a
detrimental effect on catalysis efficiency, generating the prod- Establishing scope: With a general procedure having been de-
uct in <5% yield (Table 1, entry 2). Increasing the ratio to 1:7.5 veloped, attention was then focused on establishing the scope of
and 1:9.23 (Olah’s reagent) restored catalysis efficiency but did the transformation (Figure 3). To explore the effect of changes
not surpass previous observations (44 and 46% yields, Table 1, to the aryl ring, compared to the parent scaffold 2a, representa-
entries 3 and 4, respectively). For comparison, the reaction was tive N-allylcarboxamides containing the p-OCH3, p-NO2 and
attempted using Pyr·HF (6 equiv) but this alteration had an p-CF3 substituents were exposed to the general conditions (to
adverse effect on yield (27%, Table 1, entry 5). Based on these generate 2b, 2c and 2d, respectively). These transformations
findings, the remainder of the study was performed with an proceeded smoothly to deliver the target 2-oxazolines in good
amine/HF ratio of 1:4.5. Reducing the concentration from yields (up to 69%) and in the case of compound 2c, the fluoro-
0.2 mol·L−1 to 0.1 mol·L−1 led to a large increase in yield (81%, cyclisation was performed on a 1 mmol scale with no impact on
Table 1, entry 6).
the yield. However, the aldehyde derivative 2e proved to be
more challenging and was isolated in a modest 31% yield.
Whilst solvents such as toluene, acetonitrile and THF were less Systems containing ortho-substituents (2f and 2g) were also
effective than DCE (Table 1, entries 7–9), switching to DCM well tolerated but in the case of 2f it was necessary to extend
led to full consumption of the starting material and a quantita- the reaction time to 40 h. Disubstitution patterns such as in 2h
tive NMR yield (Table 1, entry 10). In a final optimisation and 2i, the latter of which contains a free phenol moiety, were
round, the catalyst loading was reduced to 10 mol % with no also tolerated (69% and 65% yield, respectively), as was the
discernable effect on performance (Table 1, entry 11). Howev- highly deactivated pentafluorophenyl analogue 2j (48%). To
1023

Beilstein J. Org. Chem. 2018, 14, 1021–1027.
Figure 3: Substrate scope. aReaction conducted on 1 mmol scale. bReaction time increased to 40 hours. cReaction time increased to 32 hours.
Yields refer to isolated values whilst NMR yields are given in parentheses (19F NMR using ethyl fluoroacetate as an internal standard).
briefly explore the effect of chain length on efficiency, the Finally, to explore possible diastereocontrol in the cyclisation
cyclisation of N-(but-3-en-1-yl)benzamide and N-(pent-4-en-1- event, oxazoline 2r was generated from the corresponding
yl)benzamide was explored (to generate 2k and 2l, respec- α-chiral amide under standard conditions. Analysis of the crude
tively). Unsurprisingly, whilst the 6-membered ring formed in reaction mixture by 19F NMR allowed a yield of >95% to be
42% yield, cyclisation to form the analogous 7-membered ring determined and a 1:1 dr. This is to be expected given the remote
failed. It was, however, possible to generate heterocyclic nature of the stereocentre. It is important to note that attempts to
species such as the 9-fluorenonyl-substituted oxazoline 2m separate this compound by column chromatography resulted in
(48%) and the furan 2n (59%). Whilst it was not possible to significant hydrolysis. Consequently, the oxazoline was
generate the bisoxazoline 2o (X = N), the analogous carbogenic exposed to acidic media and quantitatively hydrolysed to the
scaffold 2p (X = CH) formed in 46% yield. Finally, although fluorohydrin 3 in 61% yield (Scheme 1). In contrast, the cycli-
more challenging, it was also possible to generate an aliphatic sation to form 2s was highly diastereoselective on account of
2-oxazoline (2q) in a modest 44% yield.
the proximal nature of the stereocentre (65%, dr >95:5).
1024

Beilstein J. Org. Chem. 2018, 14, 1021–1027.
Scheme 1: Exploring diastereocontrol and the synthesis of the fluorohydrin 3. Yields in parentheses were determined by 19F NMR using ethyl fluo-
roacetate as an internal standard. Unless otherwise stated, yields refer to isolated values.
Compound 3 is noteworthy on account of the β-amino alcohol
Table 2: Crystallographic data for compound 2c.
and β-fluoro alcohol motifs that collectively preorganised the
Entry
Data
propyl chain. Stabilising hyperconjugative interactions mani-
fest themselves in the characteristic gauche conformations
around the two respective torsion angles [37,38]. In this case, it
is also highly probable that hydrogen bonding will reinforce
these conformational preferences. Whilst it was not possible to
isolate crystals of 3 that were suitable for X-ray analysis, it was
possible to unambiguously establish the structure of oxazoline
2c bearing a p-NO2 group (Figure 4 and Table 2) [39]. The mo-
lecular structure reveals the expected gauche arrangement with
a dihedral angle FCCO ≈ −73.4° due to σC–C→σC–F* and
σC–H→σC–O* interactions. This observation is in agreement
with the fluorine gauche effect.
formula
C10H9FN2O3
224.19
Mr
crystal size, mm3
crystal system
space group
cell constants
a, Å
0.032 × 0.162 × 0.247
orthorhombic
Pna21
10.0315(3)
15.4164(5)
6.5161(2)
1007.71(5)
4
b, Å
c, Å
V, Å3
Z
Dx, Mg m−3
μ, mm−1
1.48
1.06
F(000), e
T, K
464
100(2)
λ, Å
1.54178
137
2θmax, deg
transmissions
refl. meas./indep./Rint
ref. parameters
restraints
R [F ≥ 4 σ(F)]
wR (F2, all refl.)
S
0.78–0.97
10003/1813/0.034
182
118
0.032
0.086
1.05
Δρmax, e Å−3
0.145/−0.194
Figure 4: X-ray molecular structure of compound 2c. Thermal ellip-
soids shown at the 50% propability level. Torsion angle
Conclusion
(
F1–C10–C9–O1 −73.4°) consistent with the fluorine gauche effect.
An operationally simple route to 5-fluoromethyl-2-oxazolines
CCDC number 1815371.
from readily accessible N-allylcarboxamides is disclosed based
1025

Beilstein J. Org. Chem. 2018, 14, 1021–1027.
12.Meanwell, N. A. J. Med. Chem. 2018.
on an I(I)/I(III) catalysis manifold. This metal-free fluoro-
cyclisation employs p-iodotoluene (10 mol %) as an inexpen-
sive organocatalyst and Selectfluor® as oxidant. The optimal
amine/HF ratio (1:4.5) is easily obtained by combining com-
mercially available triethylamine tris(hydrogenfluoride)
(Et3N·3HF) and Olah’s reagent (Pyr·HF). Broad functional
group tolerance is observed in the products, the structures of
which display the stereoelectronic fluorine gauche effect.
doi:10.1021/acs.jmedchem.7b01788
13.Singh, F. V.; Wirth, T. Catalytic Oxidations with Hypervalent Iodine. In
Catalytic Oxidation in Organic Synthesis; Muñiz, K., Ed.; Science of
Synthesis; 2018; pp 29–62. doi:10.1055/sos-SD-225-00023
14.Hypervalent Iodine Chemistry; Wirth, T., Ed.; Topics in Current
Chemistry, Vol. 373; 2016.
15.Martín Romero, R.; Wöste, T. H.; Muñiz, K. Chem. – Asian J. 2014, 9,
972–983. doi:10.1002/asia.201301637
16.Singh, F. V.; Wirth, T. Oxidative Functionalization with Hypervalent
Halides. In Reference Module in Chemistry, Molecular Sciences and
Chemical Engineering, 2nd ed.; Molander, G. A.; Knochel, P., Eds.;
Comprehensive Organic Synthesis II, Vol. 7; Elsevier: Oxford, 2014;
pp 880–933. doi:10.1016/B978-0-08-097742-3.00735-7
17.Singh, F. V.; Wirth, T. Chem. – Asian J. 2014, 9, 950–971.
doi:10.1002/asia.201301582
Supporting Information
Supporting Information File 1
Experimental part.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-14-88-S1.pdf]
18.Brown, M.; Farid, U.; Wirth, T. Synlett 2013, 24, 424–431.
doi:10.1055/s-0032-1318103
19.Quideaux, S.; Wirth, T. Tetrahedron 2010, 66, 5737–5738.
doi:10.1016/j.tet.2010.06.026
20.Parra, A.; Reboredo, S. Chem. – Eur. J. 2013, 19, 17224–17260.
doi:10.1002/chem.201302220
Acknowledgements
We acknowledge generous financial support from the WWU
Münster and the Deutsche Forschungsgemeinschaft (SFB 858
and Excellence Cluster EXC 1003 “Cells in Motion”). We
thank Dr. Nicola Lucchetti for helpful discussions.
21.Richardson, R. D.; Wirth, T. Angew. Chem., Int. Ed. 2006, 45,
4402–4404. doi:10.1002/anie.200601817
22.Ochiai, M.; Miyamoto, K. Eur. J. Org. Chem. 2008, 4229–4239.
doi:10.1002/ejoc.200800416
23.Uyanik, M.; Ishihara, K. Chem. Commun. 2009, 2086–2099.
doi:10.1039/b823399c
ORCID® iDs
Constantin G. Daniliuc - https://orcid.org/0000-0002-6709-3673
Gerald Kehr - https://orcid.org/0000-0002-5196-2491
Ryan Gilmour - https://orcid.org/0000-0002-3153-6065
24.Dohi, T.; Kita, Y. Chem. Commun. 2009, 2073–2085.
doi:10.1039/b821747e
25.Hara, S.; Nakahigashi, J.; Ishi-i, K.; Sawaguchi, M.; Sakai, H.;
Fukuhara, T.; Yoneda, N. Synlett 1998, 495–496.
doi:10.1055/s-1998-1714
26.Sawaguchi, M.; Hara, S.; Fukuhara, T.; Yoneda, N. J. Fluorine Chem.
2000, 104, 277–280. doi:10.1016/S0022-1139(00)00241-4
27.Cresswell, A. J.; Eey, S. T.-C.; Denmark, S. E. Angew. Chem., Int. Ed.
2015, 54, 15642–15682. doi:10.1002/anie.201507152
See for a discussion of this transformation.
References
1. Davidson, B. S. Chem. Rev. 1993, 93, 1771–1791.
doi:10.1021/cr00021a006
2. Palanisamy, S. K.; Rajendran, N. M.; Marino, A.
Nat. Prod. Bioprospect. 2017, 7, 1–111.
28.Molnár, I. G.; Gilmour, R. J. Am. Chem. Soc. 2016, 138, 5004–5007.
doi:10.1021/jacs.6b01183
doi:10.1007/s13659-016-0115-5
3. Hoogenboom, R. Angew. Chem., Int. Ed. 2009, 48, 7978–7994.
doi:10.1002/anie.200901607
29.Banik, S. M.; Medley, J. W.; Jacobsen, E. N. J. Am. Chem. Soc. 2016,
138, 5000–5003. doi:10.1021/jacs.6b02391
4. de la Rosa, V. R. J. Mater. Sci.: Mater. Med. 2014, 25, 1211–1225.
doi:10.1007/s10856-013-5034-y
30.Weinland, R. F.; Stille, W. Ber. Dtsch. Chem. Ges. 1901, 34,
2631–2633. doi:10.1002/cber.190103402213
5. Helmchen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336–345.
doi:10.1021/ar9900865
31.Edmunds, J. J.; Motherwell, W. B. J. Chem. Soc., Chem. Commun.
1989, 881–883. doi:10.1039/C39890000881
6. Desimoni, G.; Faita, G.; Jørgensen, K. A. Chem. Rev. 2006, 106,
3561–3651. doi:10.1021/cr0505324
32.Frohn, H.-J.; Bailly, F.; Welting, D.; Bardin, V. V. J. Fluorine Chem.
2009, 130, 301–307. doi:10.1016/j.jfluchem.2008.11.004
33.Wilkinson, J. A. Chem. Rev. 1992, 92, 505–519.
doi:10.1021/cr00012a002
7. Hargaden, G. C.; Guiry, P. J. Chem. Rev. 2009, 109, 2505–2550.
doi:10.1021/cr800400z
8. Rauniyar, V.; Lackner, A. D.; Hamilton, G. L.; Toste, F. D. Science
2011, 334, 1681–1684. doi:10.1126/science.1213918
9. Han, Y.-C.; Zhang, Y.-D.; Jia, Q.; Cui, J.; Zhang, C. Org. Lett. 2017, 19,
5300–5303. doi:10.1021/acs.orglett.7b02479
See for the first discussion of the X-ray structure of p-TolIF2 and
reference [134] therein.
34.Ye, C.; Twamley, B.; Shreeve, J. Org. Lett. 2005, 7, 3961–3964.
doi:10.1021/ol051446t
10.Galeazzi, R.; Martelli, G.; Mobbili, G.; Orena, M.; Rinaldi, S. Org. Lett.
2004, 6, 2571–2574. doi:10.1021/ol049146j
35.Sarie, J. C.; Thiehoff, C.; Mudd, R. J.; Daniliuc, C. G.; Kehr, G.;
Gilmour, R. J. Org. Chem. 2017, 82, 11792–11798.
doi:10.1021/acs.joc.7b01671
See for a synthesis route to 5-fluoromethyl-2-oxazoline.
11.Kohlhepp, S. V.; Gulder, T. Chem. Soc. Rev. 2016, 45, 6270–6288.
doi:10.1039/C6CS00361C
36.Molnár, I. G.; Thiehoff, C.; Holland, M. C.; Gilmour, R. ACS Catal.
2016, 6, 7167–7173. doi:10.1021/acscatal.6b02155
See for an excellent review on hypervalent iodine fluorocyclisations.
1026

Beilstein J. Org. Chem. 2018, 14, 1021–1027.
37.Zimmer, L. E.; Sparr, C.; Gilmour, R. Angew. Chem., Int. Ed. 2011, 50,
11860–11871. doi:10.1002/anie.201102027
38.Thiehoff, C.; Rey, Y. P.; Gilmour, R. Isr. J. Chem. 2017, 57, 92–100.
doi:10.1002/ijch.201600038
39.Crystallographic data: CCDC 1815371 contains the supplementary
crystallographic data. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via
http://www.ccdc.cam.ac.uk/data_request/cif.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.14.88
1027