Powerful fluoroalkoxy molybdenum(V) reagent for selective oxidative arene coupling reaction

Article abstract of DOI:10.1002/anie.201309287

We introduce the novel fluoroalkoxy molybdenum(V) reagent 1 which has superior reactivity and selectivity in comparison to MoCl₅ or the MoCl₅/TiCl₄ reagent mixture in the oxidative coupling reactions of aryls. Common side reactions, such as chlorination and/or oligomer formation, are drastically diminished creating a powerful and useful reagent for oxidative coupling. Theoretical treatment of the reagent interaction with 1,2-dimethoxybenzene-type substrates indicates an inner-sphere electron transfer followed by a radical cationic reaction pathway for the oxidative-coupling process. EPR spectroscopic and electrochemical studies, X-ray analyses, computational investigations, and the experimental scope provide a highly consistent picture. The substitution of chlorido ligands by hexafluoroisopropoxido moieties seems to boost both the reactivity and selectivity of the metal center which might be applied to other reagents as well. Wait a Mo: A novel molybdenum(V) reagent (see scheme) is superior to the performance of MoCl₅ in the oxidative coupling reaction of aryls leading to less by-products and significantly increased yields. Copyright

Full text of DOI:10.1002/anie.201309287

.
Angewandte
Communications
DOI: 10.1002/anie.201309287
Oxidative Coupling
Powerful Fluoroalkoxy Molybdenum(V) Reagent for Selective
Oxidative Arene Coupling Reaction
Moritz Schubert, Jana Leppin, Kathrin Wehming, Dieter Schollmeyer, Katja Heinze,* and
Siegfried R. Waldvogel*
Dedicated to Professor Rudolf Zentel on the occasion of his 60th birthday
Abstract: We introduce the novel fluoroalkoxy molybde-
num(V) reagent 1 which has superior reactivity and selectivity
in comparison to MoCl5 or the MoCl5/TiCl4 reagent mixture in
the oxidative coupling reactions of aryls. Common side
reactions, such as chlorination and/or oligomer formation,
Scheme 1. Conversion of the organic substrate into the desired cou-
pling product and the undesired chlorinated by-product.
are drastically diminished creating a powerful and useful
reagent for oxidative coupling. Theoretical treatment of the
reagent interaction with 1,2-dimethoxybenzene-type substrates
indicates an inner-sphere electron transfer followed by a radical
cationic reaction pathway for the oxidative-coupling process.
EPR spectroscopic and electrochemical studies, X-ray analy-
ses, computational investigations, and the experimental scope
provide a highly consistent picture. The substitution of chlorido
ligands by hexafluoroisopropoxido moieties seems to boost
both the reactivity and selectivity of the metal center which
might be applied to other reagents as well.
ation reactions (Scheme 1).[6] The unique feature of this
reagent is that the coupling process is fast and clearly
completed before typical side reactions, that is, protodeioda-
tion reactions in iodoaryl substrates or cleavage of ketals can
occur.[7] MoCl5 was successfully applied to form five-,[8] six-
,
[4,9] seven-,[5,10] and eight-membered rings.[11] Yet, chlorination
is a typical undesired side reaction (Scheme 1).[1,12] It can even
become the dominate pathway when very electron-rich
À
O
xidative coupling reactions of aryls are twofold CH
substrates are employed or the C C coupling is rather
activations. Although known for a long time, this trans-
formation is experiencing a renaissance in synthetic organic
chemistry, since no leaving functionalities are required.[1]
Many metal salts in high oxidation states or strong electron
acceptors are capable to efficiently induce this conversion.[2]
Modern applications focus on oxidants based either on salts of
biocompatible metals, that is, iron or molybdenum, or metal-
free systems, such as hypervalent iodine reagents.[3] Currently,
the bis(trifluoroacetoxy)iodobenzene-based compounds
seem to be the most versatile reagents. However, in several
comparative studies, the oxidative power of this iodine(III)
reagent was outperformed by MoCl5.[4] Moreover, domino
oxidations exclusively induced by MoCl5 underline the unique
reactivity.[5] MoCl5 was employed in several oxidative aryl-
slow.[13] Addition of Lewis acids (usually TiCl4) which are
assumed to bind the co-formed chloride ions has proven to be
helpful in some of these cases, but the beneficial effect is not
general.[14] Therefore, a novel molybdenum reagent with less
chlorinating potential is highly desirable.
We report the synthesis and properties of a next-gener-
ation molybdenum(V) reagent 1 (Figure 1, Scheme 2) and
elucidate mechanistic details. The superior reactivity of 1 in
[*] M. Schubert, Dr. K. Wehming, Dr. D. Schollmeyer,
Prof. Dr. S. R. Waldvogel
Figure 1. Molecular structures of 1 and 2 determined by single-crystal
X-ray diffraction.
Institut fꢀr Organische Chemie
Johannes Gutenberg-Universitꢁt Mainz
Duesbergweg 10–14, 55128 Mainz (Germany)
E-mail: waldvogel@uni-mainz.de
Homepage: http://www.chemie.uni-mainz.de/OC/AK-Waldvogel/
comparison to MoCl5 in various challenging substrate con-
versions (Table 1) is demonstrated.
J. Leppin, Prof. Dr. K. Heinze
Institut fꢀr Anorganische Chemie und Analytische Chemie
Johannes Gutenberg-Universitꢁt Mainz
Duesbergweg 10–14, 55128 Mainz (Germany)
E-mail: katja.heinze@uni-mainz.de
Homepage: http://www.ak-heinze.chemie.uni-mainz.de/
Clean ligand exchange at MoCl5 is only successful by using
a highly oxidation stable anion which has electronic proper-
ties comparable to the chlorido ligand. This is vital to
conserve the unique reactivity of the MoV reagent.
1,1,1,3,3,3-hexafluoroisopropanol (HFIP) is a very stable
solvent and has beneficial effects on the oxidative coupling
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201309287.
2494
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 2494 –2497

Angewandte
Chemie
compared to HFIP led to a further selective substitution of all
but two chlorido ligands.
Both, 1 and 2, are dinuclear complexes in the solid state
(Figure 1). Similar to Mo2Cl10, the MoV centers are bridged by
chlorido ligands in 1 while alkoxido bridges are found in 2.
The smaller alkoxido bridge necessarily leads to a much
smaller Mo–Mo distance in
2 (2.832 ꢀ) compared to
1 (3.805 ꢀ) or Mo2Cl10 (3.825 ꢀ–3.886 ꢀ).[16]
Dissolution of the diamagnetic dinuclear complexes
Mo2Cl10, 1, or 2 in dichloromethane leads to formation of
EPR-active species in all cases. The X-band EPR spectra
show characteristic resonances at g values below 2.0 with
95/97Mo satellites (A = 46–51 G) clearly supporting the pres-
ence of magnetically uncoupled MoV (Figure 2) (95/97Mo: I =
5/2 with 25% natural abundance; all other isotopes (75%
Scheme 2. Synthesis of 1 and 2 from MoCl5.
À
Table 1: Conversions with MoCl5, MoCl5/TiCl4, and 1 (formed C C bonds
are indicated in bold).
Product
Yield [%][a]
MoCl5
26
MoCl5/TiCl4
1
58
41
44
71
47[7a]
60
82
Figure 2. X-band EPR spectra (in black) and simulations (in gray) of
a) Mo2Cl10, b) 1, and c) 2 in dichloromethane at 243 K (Mo2Cl10
g1 =1.9532, A1 =50.5 G, 62%; g2 =1.9472, A2 =50.0 G, 38%; 1:
g1 =1.9466, A1 =50.0 G, 75%; g2 =1.9407, A2 =46.0 G, 25%; 2:
g1 =1.9505, A1 =51.0 G, 25%; g2 =1.9473, A2 =50.5 G, 75%).
:
67[20]
53[4a]
40
89
78
natural abundance): I = 0). Hence, the two bridges (chlorido
in Mo2Cl10 and 1; alkoxido in 2) are at least partially disrupted
and free coordination sites for substrates should be available
in all cases (see the Supporting Information for possible MoV
species and EPR spectra at 298 K including simulations).[17]
Dissolved in the respective alcohols (HFIP or TFE), com-
plexes 1 and 2 fully dissociate into mononuclear, six-
coordinate species by coordination of the alcohol (1: g =
1.9501; 2: g1 = 1.9480 (80%), g2 = 1.9403 (20%); 1,2:
A(95/97Mo) = 51.0 G). These values are similar to those
reported for MoCl5 dissolved in ethanol (see the Supporting
Information for EPR spectra and simulations) although this is
believed to be a much more complicated reaction.[17b,18]
To evaluate the oxidative power of all three compounds,
their peak potentials versus ferrocene were determined in
dichloromethane/[nBu4N][B(C6F5)4]. While Mo2Cl10 and
1 show the expected high potentials (Ep = 1.16 V, 1.22 V), 2
has a much lower value of Ep = 0.31 V. Hence, 2 should be
thermodynamically less competent to oxidize aromatic sub-
strates. This view is clearly supported by DFT calculations
(B3LYP, LANL2DZ, PCM CH2Cl2) of MoCl5, MoCl3(OCH-
(CF3)2)2, and MoCl(OCH2CF3)4, in the presence of 3,4-
dimethoxytoluene (Figure 3).
34[13]
<5[b]
[a] Yields are of isolated products. [b] Yield determined by GC (see the
Supporting Information).
of aryls.[15] The hexafluoroisopropoxy as well as the trifluoro-
ethoxy ligands were chosen because of their good availability.
Therefore, ligand substitution of chloride by 1,1,1,3,3,3-
hexafluoroisopropanolate seemed to be useful and was
easily achieved by heating both reaction partners (Sche-
me 2a). Upon evolution of hydrogen chloride a reddish-
brown crystal slurry was obtained providing 1 selectively in
95% yield.
Under the same conditions the common less-fluorinated
alcohol 2,2,2-trifluoroethanol (TFE) delivered 2 selectively in
an isolated yield of 97% (Scheme 2b). The more electron-
rich nature and smaller size of 2,2,2-trifluoroethanolate as
Angew. Chem. Int. Ed. 2014, 53, 2494 –2497
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2495

.
Angewandte
Communications
kinetic barrier to the already unfavorable electron transfer to
2.
The abovementioned undesired chlorination reactions
during the oxidative coupling can be significantly reduced by
the use of 1. The major reasons are the lower chlorine content
of the novel reagent and the lower nucleophilicity of the HFIP
ligands compared to chloride.[15a,19] In addition, the use of
1 decreases the amount of undesired oligomeric materials.
This observation might be explained by the higher oxidation
potential of 1 in comparison to molybdenum pentachloride
which possibly leads to a higher reaction rate of the substrate
and subsequent conversions resulting in smaller amounts of
by-products. Table 1 depicts the selection of substrates which
were subjected to the oxidative coupling reaction and which
had been problematic in terms of yield and selectivity using
MoCl5.
Reagent 1 considerably increases the yield of the desired
product in all cases in our study (Table 1) whereas reagent 2
was unsuitable. The oxidative transformation of the extra-
ordinarily sensitive substrate a-aryl cinnamate using 1 deliv-
ered phenanthrene derivative 3 isolated in 71% yield and
only 18% of an undesired chlorinated by-product (Table 1,
see the Supporting Information).[21] Comparative experi-
ments with MoCl5 gave 3 in a yield of only 26% but the
chlorinated product was isolated in 63% yield. The use of
MoCl5/TiCl4, as a reagent mixture can reduce the chlorination
reaction to some extent but the total yield for phenanthrene
products is significantly lower. The oxidative coupling of 1,5-
diiodo-2,3-dimethoxybenzene by 1 allowed 4 to be isolated in
60% yield. Previous investigations with MoCl5 of this
particular conversion allowed isolation in only 47% yield. It
is noteworthy, that MoCl5 had to be added in small portions to
reduce the yield of proto-deiodinated by-products.[7] The use
of the MoCl5/TiCl4 mixture strongly promoted the proto-
deiodination process. The reaction of 3,4-dimethoxytoluene
with 1 to the corresponding biaryl yielded 82% of 5 (MoCl5:
67%; Table 1).[16] Using 1 the formation of chlorinated by-
products can even be completely suppressed in this case. The
oxidation of 3,3’’-dimethoxy-o-terphenyl with 1 gave 2,7-
dimethoxytriphenylene (6), isolated in 89% yield. King and
co-workers report an isolation in 53% yield using a similar
method with MoCl5.[4a] The MoCl5/TiCl4 mixture provided no
amelioration. The yield is inferior (40%) and oligomeric
materials seem to be formed. The oxidative coupling of 1,3,5-
trimethylbenzene using 1 gave 7, isolated in 78% yield.
Kovacic et al. reported isolation in 34% yield with MoCl5.
Furthermore, they report several chlorinated and oligomeric
by-products that are made responsible for the low yield.[16]
The use of TiCl4 did not improve the situation and the desired
compound is only traced by GC methods, hence HFIP ligand
is clearly preferred.
Figure 3. DFT-optimized geometries and spin densities of a) MoCl5,
b) MoCl3(OCH(CF3)2)2 with equatorial alkoxido ligands, and c) MoCl-
(OCH2CF3)4 with equatorial alkoxido ligands in the presence of 3,4-
dimethoxytoluene (spin multiplicity=2; yellow/blue isosurface value
0.02 a.u.; Mulliken spin densities at Mo given in red; Mo–O distan-
ces [ꢂ]).
For MoCl5 and 1, electron transfer from 3,4-dimeth-
oxytoluene to the Mo center has clearly occurred with
Mulliken spin densities at Mo of 1.89 (MoCl5 + 3,4-
dimethoxytoluene) and 1.84 (MoCl3(OCH(CF3)2)2 + 3,4-
dimethoxytoluene) confirming the presence of d2 MoIV and
negative spin density at the 3,4-dimethoxytoluene substrate
(Figure 3a and b). This result clearly indicates the formation
of an organic radical cation which enters the oxidative
pathway. For the mononuclear species MoCl(OCH2CF3)4
derived from 2 no such electron transfer from 3,4-dimethox-
ytoluene can be discerned (Mulliken spin density at Mo: 0.91)
confirming the presence of d1 MoV (Figure 3c).
In addition to the thermodynamic arguments presented,
we were able to calculate stationary structures with 3,4-
dimethoxytoluene kO- and kOkO’-coordinated to MoCl5 and
MoCl3(OCH(CF3)2)2 leading to six- and seven-coordinate
complexes, respectively (see the Supporting Information).
Coordination of the substrate to Mo proved to be impossible
for MoCl(OCH2CF3)4 according to DFT calculations. Hence,
the steric protection by four alkoxido ligands per Mo center in
2 hinders substrate coordination and poses an additional
Received: October 24, 2013
Revised: December 9, 2013
Published online: January 29, 2014
À
Keywords: arene coupling · C C coupling · cyclization ·
molybdenum · oxidative coupling
.
2496 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 2494 –2497

Angewandte
Chemie
O. Kataeva, S. R. Waldvogel, Eur. J. Org. Chem. 2005, 2987 –
2999.
[8] S. Trosien, D. Schollmeyer, S. R. Waldvogel, Synthesis 2013,
1160 – 1164.
[1] M. Grzybowski, K. Skonieczny, H. Butenschçn, D. T. Gryko,
Angew. Chem. 2013, 125, 10084 – 10115; Angew. Chem. Int. Ed.
2013, 52, 9900 – 9930.
[2] G. Lessene, K. S. Feldman in Modern Arene Chemistry (Ed.: D.
Astruc), Wiley-VCH, Weinheim, 2002, pp. 479 – 538.
[9] S. Trosien, S. R. Waldvogel, Org. Lett. 2012, 14, 2976 – 2979.
[10] B. Kramer, S. R. Waldvogel, Angew. Chem. 2004, 116, 2501 –
2503; Angew. Chem. Int. Ed. 2004, 43, 2446 – 2449.
[11] B. Kramer, A. Averhoff, S. R. Waldvogel, Angew. Chem. 2002,
114, 3103 – 3104; Angew. Chem. Int. Ed. 2002, 41, 2981 – 2982.
[12] a) J. San Filippo, Jr., A. F. Sowinski, L. J. Romano, J. Org. Chem.
1975, 40, 3295 – 3296; b) J. San Filippo, Jr., A. F. Sowinski, L. J.
Romano, J. Org. Chem. 1975, 40, 3463 – 3464; c) E. L. McCan-
n III, T. M. Brown, Inorg. Synth. 1970, 12, 181 – 186; d) J.
San Filippo, Jr., A. F. Sowinski, L. J. Romano, J. Am. Chem.
Soc. 1975, 97, 1599 – 1600.
[3] a) H. Tohma, H. Morioka, S. Takizawa, M. Arisawa, Y. Kita,
Tetrahedron 2001, 57, 345 – 352; b) T. Dohi, A. Maruyama, M.
Yoshimura, K. Morimoto, H. Tohma, Y. Kita, Angew. Chem.
2005, 117, 6349 – 6352; Angew. Chem. Int. Ed. 2005, 44, 6193 –
6196; c) T. Dohi, K. Morimoto, Y. Kiyono, A. Maruyama, H.
Tohma, Y. Kita, Chem. Commun. 2005, 2930 – 2932; d) K.
Morimoto, N. Yamaoka, C. Ogawa, T. Nakae, H. Fujioka, T.
Dohi, Y. Kita, Org. Lett. 2010, 12, 3804 – 3807; e) K. Morimoto,
T. Nakae, N. Yamaoka, T. Dohi, Y. Kita, Eur. J. Org. Chem. 2011,
6326 – 6334; f) T. Dohi, M. Ito, K. Morimoto, M. Iwata, Y. Kita,
Angew. Chem. 2008, 120, 1321 – 1324; Angew. Chem. Int. Ed.
2008, 47, 1301 – 1304; g) E. Faggi, R. M. Sebasyiꢁn, R. Pleixats,
A. Vallribera, A. Shafir, A. Rodrꢂguez-Gimenco, C. R. de Ar-
ellano, J. Am. Chem. Soc. 2010, 132, 17980 – 17982; h) T. Dohi,
M. Ito, I. Itani, N. Yamaoka, K. Morimoto, H. Fujioka, Y. Kita,
Org. Lett. 2011, 13, 6208 – 6211; i) T. Dohi, Y. Minamitsuji, A.
Maruyama, S. Hirose, Y. Kita, Org. Lett. 2008, 10, 3559 – 3562,
and references therein.
[13] P. Kovacic, R. M. Lange, J. Org. Chem. 1965, 30, 4251 – 4254.
[14] B. Kramer, R. Frçhlich, S. R. Waldvogel, Eur. J. Org. Chem.
2003, 3549 – 3554.
[15] a) T. Dohi, N. Yamaoka, Y. Kita, Tetrahedron 2010, 66, 5775 –
5785; b) A. Kirste, M. Nieger, I. M. Malkowsky, F. Stecker, A.
Fischer, S. R. Waldvogel, Chem. Eur. J. 2009, 15, 2273 – 2277;
c) A. Kirste, G. Schnakenburg, F. Stecker, A. Fischer, S. R.
Waldvogel, Angew. Chem. 2010, 122, 983 – 987; Angew. Chem.
Int. Ed. 2010, 49, 971 – 975; d) A. Kirste, G. Schnakenburg, S. R.
Waldvogel, Org. Lett. 2011, 13, 3126 – 3129; e) A. Kirste, B.
Elsler, G. Schnakenburg, S. R. Waldvogel, J. Am. Chem. Soc.
2012, 134, 3571 – 3576.
[16] J. Beck, F. Wolf, Acta Crystallogr. Sect. B 1997, 53, 895 – 903.
[17] a) K. Dyrek, K. Kruczala, Z. Sojka, S. Schlick, J. Phys. Chem.
1993, 97, 9196 – 9200; b) Z. Gao, S. Schlick, J. Chem. Soc.
Faraday Trans. 1996, 92, 4239 – 4245.
[18] C. Limberg, R. Boese, B. Schiemenz, J. Chem. Soc. Dalton Trans.
1997, 1633 – 1637.
[19] R. Francke, D. Cericola, R. Kçtz, S. R. Waldvogel, Electrochim.
Acta 2012, 59-86, 372 – 380.
[20] S. R. Waldvogel, Synlett 2002, 622 – 624.
[21] According to: A. Ianni, S. R. Waldvogel, Synthesis 2006, 2103 –
2112.
[4] a) B. T. King, J. Kroulik, C. R. Robertson, P. Rempala, C. L.
Hilton, J. D. Korinek, L. M. Gortari, J. Org. Chem. 2007, 72,
2279 – 2288; b) K. Hackelçer, G. Schnakenburg, S. R. Waldvo-
gel, Eur. J. Org. Chem. 2011, 6314 – 6319.
[5] K. Hackeloer, G. Schnakenburg, S. R. Waldvogel, Org. Lett.
2011, 13, 916 – 919.
[6] S. R. Waldvogel, S. Trosien, Chem. Commun. 2012, 48, 9109 –
9119.
[7] a) S. R. Waldvogel, E. Aits, C. Holst, R. Frçhlich, Chem.
Commun. 2002, 1278 – 1279; b) S. R. Waldvogel, A. R. Wartini,
P. H. Rasmussen, J. Rebek, Tetrahedron Lett. 1999, 40, 3515 –
3518; c) S. R. Waldvogel, R. Frçhlich, C. A. Schalley, Angew.
Chem. 2000, 112, 2580 – 2583; Angew. Chem. Int. Ed. 2000, 39,
2472 – 2475; d) M. C. Schopohl, A. Faust, D. Mirk, R. Frçhlich,
Angew. Chem. Int. Ed. 2014, 53, 2494 –2497
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2497