Catalytic, atom-economical radical arylation of epoxides

Article abstract of DOI:10.1002/anie.201200431

A slow electron waltz: Catalysis in single-electron steps with radicals as intermediates is essential for an atom-economical arylation of epoxides with a proton-coupled electron transfer as a key step. Stabilization of the catalyst is essential for low catalyst loading, as demonstrated by reaction-progress kinetic analysis, cyclic voltammetry, and calculations. Copyright

Full text of DOI:10.1002/anie.201200431

Angewandte
Chemie
DOI: 10.1002/anie.201200431
Radical Chemistry
Catalytic, Atom-Economical Radical Arylation of Epoxides**
Andreas Gansꢀuer,* Maike Behlendorf, Daniel von Laufenberg, Andrꢁ Fleckhaus,
Christian Kube, Dhandapani V. Sadasivam, and Robert A. Flowers II*
The development of efficient catalytic reactions is one of the
central aspects of chemistry and arguably the most important
for the invention of novel sustainable processes.[1] Radical-
based transformations are among the most attractive methods
for use in catalytic cycles owing to the ease of radical
chloride has to be employed in stoichiometric amounts for
radical generation in rather acidic media (aqueous HCl).
Our catalytic cycle is shown in Scheme 1. It is initiated by
the single-electron oxidative addition of [Cp2TiCl] to the
substrate generating radical intermediate A. Addition of the
generation, high functional group tolerance,
[2]
À
and selectivity in C C bond formation.
Herein we present such a process, an
atom-economical
titanocene-catalyzed[3]
intramolecular arylation of epoxide-derived
radicals. Our approach exploits the innate
capability of the titanocene(III)/(IV) redox
couple to undergo reversible electron-trans-
fer reactions.[4] This allows the implementa-
tion of both oxidative additions and reduc-
tive eliminations in single-electron steps into
catalytic cycles. The key step of our method
is presumed to be a proton-coupled electron
transfer (PCET).[5] It constitutes the pivotal
single-electron reductive elimination, pro-
vides the driving force for efficient rear-
omatization of the radical s-complex, and
negates the need for sacrificial co-reductants
or oxidants necessary in radical-based chain
processes or catalytic reactions.[6] This issue
is critical in Minisci reactions,[7] radical
additions to electron deficient heteroarenes, Scheme 1. Proposed catalytic cycle. Cp=cyclopentadienyl, M=metal.
which often require stoichiometric amounts
of metal (Fe, Ag) salts and oxidants (H2O2 or
organic peroxides). More recently, significant progress
towards more sustainable radical arylation has been reported
by Heinrich et al.[8] In these reactions, aryl diazonium salts are
employed as radical precursors. Nevertheless, titanium tri-
radical to the pendant arene produces the pivotal radical s-
complex B in the radical translocation step. The single-
electron reductive elimination of [Cp2TiCl] can be accom-
plished by an electron transfer from the arene B to the
titanocene to form C. Subsequent proton transfer to the
titanocene(III)-bound alkoxy group yields product and cata-
lyst. As a consequence, the catalytic cycle is completely atom-
economical and does not require the use of stoichiometric
[*] Prof. Dr. A. Gansꢀuer, M. Behlendorf, D. von Laufenberg,
A. Fleckhaus, C. Kube
Kekulꢁ-Institut fꢂr Organische Chemie und Biochemie der
Universitꢀt Bonn
Gerhard Domagk Strasse 1, 53121 Bonn (Germany)
E-mail: andreas.gansaeuer@uni-bonn.de
À
amounts of an external acid for the protonation of a Ti O
bond, or a source (such as O2) for the oxidation of B to the
cationic s complex, and in principle requires only the amount
of a metal powder necessary for the initial reduction of the
precatalyst [Cp2TiCl2].[9]
Dr. D. V. Sadasivam, Prof. R. A. Flowers II
Department of Chemistry, Lehigh University
Bethlehem, PA 18015 (USA)
E-mail: rof2@lehigh.edu
Homepage: http://www.lehigh.edu/~rof2/
With 10 mol% 3, complete conversion of 1a to 2a was
realized in refluxing THF after 30 min and was isolated in
98% yield (Scheme 2). This result clearly demonstrates that
neither an external oxidant nor an acid are necessary for
turnover. Manganese is only required for the generation of
the active catalyst, as without [Cp2TiCl2] no reaction takes
place. However, catalyst loading is still rather high. To
overcome this limitation, the influence of the reaction
[**] We are grateful to support from the National Science Foundation of
the USA (CHE 1123815) and the Deutsche Forschungsgemein-
schaft (SFB 813 “Chemistry at Spin Centers”). We thank Todd A.
Maisano for initial contributions to rate studies.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201200431.
Angew. Chem. Int. Ed. 2012, 51, 4739 –4742
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4739

.
Angewandte
Communications
chloro-substituted complexes [(C5H4Cl)CpTiCl2] (5) and
[(C5H4Cl)2TiCl2] (6) turned out to be efficient catalysts,
resulting in complete conversion of 1c to 2c with 5 mol%
catalyst after 2 h and high yields of isolated 2c (92% and
93%). Compared to [Cp2TiCl], the oxidation potential of
[(C5H4Cl)CpTiCl] is 120 mV more positive and the oxidation
potential of [(C5H4Cl)2TiCl] is 290 mV more positive. Thus,
the relatively small change in the redox potential of 5 is
already sufficient for a significant improvement of the yield of
2c. It is clear, however, that for even more difficult substrates
the use of 6 will be highly attractive. Addition of 4 improves
the performance of the reaction further and reduction of
catalyst loading to 5 mol% is possible for both 5 and 6.
To explore the mechanism and role of catalyst, reaction-
progress kinetic analysis (RPKA) experiments were per-
formed.[10,11] Ideally, the concentration of the catalyst remains
constant over the course of the reaction. The deactivation of
the catalyst, if any, occurring during the course of the reaction
can be determined through “same excess” experiments.
Runs 1 and 2 are set up so that run 2 starts with concen-
trations at 50% of run 1. Lack of overlay shows deactivation
of catalyst under reaction conditions (Figure 1). Furthermore,
an induction period owing to the reduction of 3 by manganese
dust was observed. To activate the metal, 4 was added and the
induction period was shortened significantly (see Supporting
Scheme 2. Atom-economical aromatic substitution with 1a.
conditions and additives on the performance of the catalyst
was investigated, as summarized in Table 1.
Decreasing catalyst loading to 5 mol% 3 resulted in
incomplete conversion. Addition of 4 has a dramatic effect on
the reaction. The complete conversion of 1a to 2a can be
Table 1: Atom-economical catalytic radical aromatic substitution.[a]
Substr.
Mol%, cat.
Mol% 4
t
1/2
Yield [%]
1a
1a
1a
1a
1b
1c
1c
1c
1c
10, 3
5, 3
5, 3
–
–
10
5
5
–
–
20
20
30 min
30 min
30 min
2 h
0:100
25:75[b]
0:100[b]
0:100
98[b]
–
–
1, 3[c]
1, 3[c]
10, 3
5, 6
96[d]
91[d]
–
2 h
0:100
30 min
30 min
2 h
87:13[e]
40:60[e]
0:100
–
5, 5
5, 6
92[e]
93[e]
2 h
0:100
[a] Sub.=substrate, Coll·HCl=2,4,6-Me3Py·HCl (4). All concentrations
refer to 1. [b] 0.1m. [c] 10 mol% Mn. [d] 0.5m. [e] 0.03m.
achieved with as little as 1 mol% 3, 5 mol% of 4, and
10 mol% of Mn. The yield of isolated 2a was 96%. Thus, Mn-
reduced 3 in combination with 4 constitutes a highly efficient
catalytic system for the conversion of 1a to 2a. The same
results were obtained when excess manganese powder was
filtered off. The most practical way of carrying out the
reaction is by the addition of substrate, precatalyst, and
a slight excess manganese dust to THF and refluxing the
mixture until starting material is consumed. Neither MnCl2,
the protic acid 2,4,6-trimethylpyridinium hydrochloride
(Coll·HCl; 4), nor Cp2TiCl2 initiate the reaction on their
own. Therefore, a highly unlikely cationic epoxide opening
can be ruled out.
Epoxide 1b with two p-methyl substituents also consti-
tutes an excellent substrate. The yield of isolated 2b under
identical conditions as for 1a was 91%. Substrate 1c does not
constitute a suitable substrate for 3. Because the oxidative
addition to epoxides is usually unproblematic, we reasoned
that for 1c the radical transfer to convert B to C is critical. In
this case, titanocene catalysts that are better oxidants than
[Cp2TiCl2] should be superior catalysts. Such a redox tuning is
straightforward to achieve by introducing electron-withdraw-
ing substitutents to the cyclopentadienyl ligands. Indeed, the
Figure 1. Rate vs. concentration of epoxide for same-excess experi-
*
ments with and without Coll·HCl. a) Run 1 at 100%, no Coll·HCl ( ),
!
*
run 2 at 50%, no Coll·HCl ( ). b) Run 3 at 100% with Coll·HCl ( ),
~
and run 4 at 50% with Coll·HCl ( ).
4740 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4739 –4742

Angewandte
Chemie
Information). Moreover, same excess experiments containing
a twofold excess of Coll·HCl (based on [Cp2TiCl]) showed
that the acid also facilitated turnover. The same excess kinetic
runs 3 and 4 show overlay that is consistent with constant
concentration of catalyst.
more stable than the individual components after the zero-
point vibrational energy corrections (see the Supporting
Information). Taken together, CV and calculations suggest
that 4 stabilizes the catalyst through a unique hydrogen
bonding interaction.
The role of Coll·HCl for catalyst stability cannot be solely
tied to proton transfer as it is employed in substoichiometric
amounts. To further explore the role of Coll·HCl, cyclic
voltammetry (CV) was employed. The cyclic voltammogram
of Mn-reduced 3 exhibited peaks corresponding to the
[(Cp2TiCl)2]/[Cp2TiCl] couple and [Cp2Ti]+ that is formed in
a follow-up reaction during the sweep.[12] Upon addition of 4,
a new oxidation wave at À1.25 V appears that is due to
[Cp2TiCl2]À, a species that is well-known to be incapable of
opening epoxides.[12] Additionally, the generation of [Cp2Ti]+
is completely suppressed. Thus, the addition of 4 to Mn-
reduced 3 results in an equilibrium between [(Cp2TiCl)2],
[Cp2TiCl], and [Cp2TiCl2]À. Usually, addition of chloride salts
to solutions of Cp2TiCl results in the exclusive formation of
[Cp2TiCl2]À (Figure 2).
At first sight, it seems paradoxical that a reduction of the
concentration of the active catalyst results in more efficient
reactions. However, our data clearly show that the decom-
position of [(Cp2TiCl)2]/[Cp2TiCl] at elevated temperatures is
prevented by the accessibility of the [Cp2TiCl2]À resting state
of the catalyst.[17] As a consequence, the use of 4 as additive
provides an attractive opportunity for lowering catalyst
loading.
To further examine the scope of the reaction, examples of
the aromatic substitution catalyzed by 4 are summarized in
Table 2. Five- and six-membered, and even strained ring
systems (entries 2 and 6) can be readily prepared in high
yields and short reaction times. The control of diastereose-
lectivity (entry 7) was excellent. Moreover, a number of
functional groups can be incorporated. It is especially note-
worthy that with catalyst 6, even electron-withdrawing esters
To further examine the role of 4, DFT calculations were
carried out using Gausian03.[13] Geometry optimization of all
structures were performed employing a B3LYP[14,15] func-
tional and the def2-TZVP basis set.[16] Interestingly, the
calculations revealed formation of a complex between the
hydrochloride salt and the TiIII center that is 9.8 kcalmolÀ1
Table 2: Scope of the atom-economical catalytic aromatic substitution
by radicals with 5 mol% of catalyst 6 and 20 mol% 4.
Entry
Substrate
t [h]
Product [%]
1
2
2
2
3
4
5
1
1
1
6
7
2
2
Figure 2. a) Cyclic voltammograms showing the impact of 4 on tita-
nium-based intermediates present during redox process. Fc=ferro-
cene. b) Calculated gas-phase complex between 3 and 4 (Ti purple,
Cl green, N blue, C gray, H white).
Angew. Chem. Int. Ed. 2012, 51, 4739 –4742
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4741

.
Angewandte
Communications
Radicals in Organic Synthesis, Wiley-VCH, Weinheim, 2001;
c) D. P. Curran, N. A. Porter, B. Giese, Stereochemistry of
Radical Reactions, Wiley-VCH, Weinheim, 1996.
[3] a) A. Gansꢀuer, T. Lauterbach, S. Narayan, Angew. Chem. 2003,
115, 5714 – 5731; Angew. Chem. Int. Ed. 2003, 42, 5556 – 5573;
b) J. M. Cuerva, J. Justicia, J. L. Oller-Lꢁpez, J. E. Oltra, Top.
Curr. Chem. 2006, 264, 63 – 92; c) A. Gansꢀuer, C.-A. Fan, J.
Justicia, D. Worgull, F. Piestert, Top. Curr. Chem. 2007, 279, 25 –
52.
[4] a) A. Gansꢀuer, B. Rinker, M. Pierobon, S. Grimme, M.
Gerenkamp, C. Mꢂck-Lichtenfeld, Angew. Chem. 2003, 115,
3815 – 3818; Angew. Chem. Int. Ed. 2003, 42, 3687 – 3690;
b) B. M. Trost, H. C. Shen, J. P. Surivet, J. Am. Chem. Soc.
2004, 126, 12565 – 12579; c) A. Gansꢀuer, B. Rinker, N. Ndene-
Schiffer, M. Pierobon, S. Grimme, M. Gerenkamp, C. Mꢂck-
Lichtenfeld, Eur. J. Org. Chem. 2004, 2337 – 2351; d) D. Leca, L.
Fensterbank, E. Lacꢃte, M. Malacria, Angew. Chem. 2004, 116,
4316 – 4318; Angew. Chem. Int. Ed. 2004, 43, 4220 – 4222; e) A.
Gansꢀuer, A. Fleckhaus, M. Alejandre Lafont, A. Okkel, K.
Kotsis, A. Anoop, F. Neese, J. Am. Chem. Soc. 2009, 131, 16989 –
16999.
[5] a) M. H. V. Huynh, T. J. Meyer, Chem. Rev. 2007, 107, 5004 –
5064; b) J. J. Warren, T. A. Tronic, J. M. Mayer, Chem. Rev. 2010,
110, 6961 – 7001; c) J. M. Mayer, Acc. Chem. Res. 2011, 44, 36 –
46.
[6] A. L. J. Beckwith, V. W. Bowry, W. R. Bowman, E. Mann, J. Parr,
J. M. D. Storey, Angew. Chem. 2004, 116, 97 – 100; Angew. Chem.
Int. Ed. 2004, 43, 95 – 98.
[7] a) F. Minisci, A. Citterio, C. Giordano, Acc. Chem. Res. 1983, 16,
27 – 32; b) M. A. J. Duncton, Med. Chem. Commun. 2011, 2,
1135 – 1161.
[8] A. Wetzel, G. Pratsch, R. Kolb, M. R. Heinrich, Chem. Eur. J.
2010, 16, 2547 – 2556.
and nitriles that render the critical PCET more difficult are
tolerated.
The application of homolytic aromatic substitution is
experiencing a renaissance in synthetic organic chemistry
because this approach provides a useful and potentially much
milder alternative to classical electrophilic aromatic substitu-
tion. Recently, several groups have described the construction
of biaryl compounds with the assistance of organocatalysts.[18]
A recent essay presented evidence from literature precedent
showing that these reactions most likely proceed through
base-mediated homolytic aromatic substitution.[19] The dis-
tinguishing feature of base-mediated homolytic aromatic
substitution is that the reaction proceeds through a radical
chain mechanism. This inevitably necessitates the stoichio-
metric use of reagents for radical generation or the deproto-
nation of s complexes.
The unique characteristic of the process presented herein
is the catalytic role of titanium(III). Its regeneration by an
intramolecular electron transfer–proton transfer mechanism
negates the necessity to employ stoichiometric reagents other
than the substrate. Moreover, this allows a tailoring of the
reactivity of the catalyst by careful choice of the substitution
pattern of the cyclopentadienyl ligands.
In summary, we have described a conceptually novel
approach to conduct catalytic reactions by employing oxida-
tive additions and reductive eliminations in single electron
steps. The success of the radical aromatic substitution is
critically dependent on a mechanism based approach for
tuning the stability and electronic properties of the catalyst.
We are currently examining intermolecular reactions and
additions initiated through single electron reduction of other
functional groups to fully explore the breadth and scope of
the reaction. The results of these studies will be presented in
due course.
[9] P. Wipf, J. P. Maciejewski, Org. Lett. 2008, 10, 4383 – 4386.
[10] a) D. G. Blackmond, Angew. Chem. 2005, 117, 4374 – 4393;
Angew. Chem. Int. Ed. 2005, 44, 4302 – 4320; b) J. S. Mathew,
M. Klassmann, H. Iwarmura, F. Valera, A. Futran, E. A. C.
Emanuelsson, D. G. Blackmond, J. Org. Chem. 2006, 71, 4711 –
4722.
[11] J. J. Devery III, J. C. Conrad, D. W. MacMillan, R. A. FlowersII,
Angew. Chem. 2010, 122, 6242 – 6246; Angew. Chem. Int. Ed.
2010, 49, 6106 – 6110.
[12] R. J. Enemærke, J. Larsen, T. Skrydstrup, K. Daasbjerg, J. Am.
Chem. Soc. 2004, 126, 7853 – 7864.
[13] M. J. Frisch et al. Gaussian03, revisionE.01 Gaussian, Inc.,
Wallingford, CT, 2004.
[14] A. D. Becke, J. Chem. Phys. 1993, 98, 5648 – 5652.
[15] C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785 –
789.
Experimental Section
Epoxide 1a (956 mg, 4 mmol, 1 equiv), catalyst 3 (10 mg, 40 mmol,
1.00 mol%), Coll·HCl (31.9 mg, 0.202 mmol, 5.1 mol%), and man-
ganese (21.9 mg, 0.399 mmol, 0.1 equiv) were placed in an oven-dried
Schlenk flask. THF (5 mL) was added, and the mixture was refluxed
under an argon atmosphere for two hours. Chromatography (Alox;
eluent cyclohexane/ethyl acetate/triethylamine 80:20:0.01) gave
915.5 mg (96%) of 2a.
[16] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 –
3305.
[17] T. Chivers, E. D. Ibrahim, J. Organomet. Chem. 1974, 77, 241 –
246.
Received: January 17, 2012
Published online: March 27, 2012
[18] a) C.-L. Sun, H. Li, D.-G. Yu, M. Yu, X. Zhou, X.-Y. Lu, K.
Huang, S.-F. Zheng, B.-J. Li, Z.-J. Shi, Nat. Chem. 2010, 2, 1044 –
1049; b) E. Shirakawa, K. i. Itoh, T. Higashino, T. Hayashi, J.
Am. Chem. Soc. 2010, 132, 15537 – 15539; c) W. Liu, H. Cao, H.
Zhang, H. Zhang, K. H. Chung, C. He, H. Wang, F. Y. Kwong, A.
Lei, J. Am. Chem. Soc. 2010, 132, 16737 – 16740.
[19] A. Studer, D. P. Curran, Angew. Chem. 2011, 123, 5122 – 5127;
Angew. Chem. Int. Ed. 2011, 50, 5018 – 5022.
Keywords: catalysis · electron transfer · radicals ·
reaction mechanisms · titanium
.
[1] a) B. M. Trost, Science 1991, 254, 1471 – 1477; b) B. M. Trost,
Angew. Chem. 1995, 107, 285 – 307; Angew. Chem. Int. Ed. Engl.
1995, 34, 259 – 281.
[2] a) S. Z. Zard, Radical Reactions in Organic Synthesis, Oxford
University Press, Oxford, 2003; b) P. Renaud, M. P. Sibi,
4742 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4739 –4742