Nitrous oxide-dependent iron-catalyzed coupling reactions of grignard reagents

Article abstract of DOI:10.2533/chimia.2015.327

The formation of carbon-carbon bonds is one of the fundamental transformations in chemistry. In this regard the application of palladium-based catalysts has been extensively investigated during recent years, but nowadays research focuses on iron catalysis, due to sustainability, costs and toxicity issues; hence numerous examples for iron-catalyzed cross-coupling reactions have been established, based on the coupling of electrophiles (R1-X, X = halide) with nucleophiles (R2-MgX). Only a small number of protocols deals with the iron-catalyzed oxidative coupling of nucleophiles (R1-MgX + R2-MgX) with the aid of oxidants (1,2-dihaloethanes). However, some issues arise with these oxidants; hence more recently the potential of the industrial waste product nitrous oxide (N2O) was investigated, because the unproblematic side product N2 is formed. Based on that, we demonstrate the catalytic potential of easily accessible iron complexes in the oxidative coupling of Grignard reagents. Importantly, nitrous oxide was essential to obtain yields up to <99% at mild conditions (e.g. 1 atm, ambient temperature) and low catalyst loadings (0.1 mol%) Excellent catalyst performance is realized with turnover numbers of up to 1000 and turnover frequencies of up to 12000 h-1. Moreover, a good functional group tolerance is observed (e.g. amide, ester, nitrile, alkene, alkyne). Afterwards the reaction of different Grignard reagents revealed interesting results with respect to the selectivity of cross-coupling product formation.

Full text of DOI:10.2533/chimia.2015.327

CatalytiC aCtivation of Small moleCuleS
CHIMIA 2015, 69, No. 6 327
doi:10.2533/chimia.2015.327
Chimia 69 (2015) 327–330 © Schweizerische Chemische Gesellschaft
Nitrous Oxide-dependent Iron-catalyzed
Coupling Reactions of Grignard Reagents
Peter Döhlert, Maik Weidauer, and Stephan Enthaler*
Abstract: The formation of carbon–carbon bonds is one of the fundamental transformations in chemistry.
In this regard the application of palladium-based catalysts has been extensively investigated during recent
years, but nowadays research focuses on iron catalysis, due to sustainability, costs and toxicity issues; hence
numerous examples for iron-catalyzed cross-coupling reactions have been established, based on the coupling
of electrophiles (R1-X, X = halide) with nucleophiles (R2-MgX). Only a small number of protocols deals with the
iron-catalyzed oxidative coupling of nucleophiles (R1-MgX + R2-MgX) with the aid of oxidants (1,2-dihaloethanes).
However, some issues arise with these oxidants; hence more recently the potential of the industrial waste product
nitrous oxide (N2O) was investigated, because the unproblematic side product N2 is formed. Based on that, we
demonstrate the catalytic potential of easily accessible iron complexes in the oxidative coupling of Grignard
reagents. Importantly, nitrous oxide was essential to obtain yields up to >99% at mild conditions (e.g. 1 atm,
ambient temperature) and low catalyst loadings (0.1 mol%) Excellent catalyst performance is realized with
turnover numbers of up to 1000 and turnover frequencies of up to 12000 h–1. Moreover, a good functional group
tolerance is observed (e.g. amide, ester, nitrile, alkene, alkyne). Afterwards the reaction of different Grignard
reagents revealed interesting results with respect to the selectivity of cross-coupling product formation.
Keywords: Grignard reagents · Iron catalysts · Nitrous oxide
ppm) and therefore high costs (~1460 € per tion, and is a greenhouse gas with a warm-
Introduction
mole), and the high toxicity of this metal ing potential 300 times greater than CO2.[10]
creates problems with respect to pharma- The industrial emission is mainly reduced
The formation of carbon–carbon bonds
ceutical applications.[4,5] Based on these by the implementation of decomposition
negative issues, a current trend in catalysis protocols.[11] Hence the application of N O
is the development of more environmen- as oxidant can be an option to reduce t2he
tally benign and sustainable alternatives.[6] amounts of waste and add value to a green-
In particular replacement with cheaper er, sustainable and environmentally friend-
and low toxic iron catalysts is envisaged ly chemistry. More recently, we reported
(4th most abundant element of earth’s the catalytic application of a trispyrrole
crust, <0.01 €/mol). Indeed, numerous iron(ii) complex in coupling reactions us-
iron-catalyzed cross-coupling reactions ing N O as oxidant.[12] On the one hand ac-
have been reported so far, mainly based tivity2was observed in the homo-coupling
on the coupling of an electrophile (e.g. of sp2 hybridized Grignard reagents and
C(sp2, sp3 hybridization)-halogen) with a on the other hand cross-coupling of differ-
nucleophile (e.g. C(sp2, sp3 hybridization)- ent sp2 hybridized Grignard reagents were
Mg-halogen) (Fig. 1).[7] Moreover, a small studied. However, low selectivity for the
number of protocols described the iron- cross-coupling product was revealed and
catalyzed coupling of two nucleophiles significant amounts of the homo-coupling
(e.g. Grignard reagents), on the one hand products are formed (Fig. 1). Based on this
the oxidative homo-coupling and on the initial work we became interested in the
other hand the oxidative cross-coupling.[8] improvement of the catalyst activity and
Importantly, mainly 1,2-dihaloethanes the improvement of the selectivity towards
were applied as the oxidant to access the the cross- coupling products.
coupling products in reasonable yields.
is one of the fundamental transformations
in chemistry to produce pharmaceuticals,
bulk and fine chemicals, agrochemicals,
polymers etc.[1] In this regard, transition-
metal catalyzed coupling reactions offer
a versatile strategy and nowadays it is an
established methodology and success has
been proven in countless synthetic meth-
odologies.[2] However, due to an increasing
demand and modification of regulations
(e.g. sustainability and environmental is-
sues) novel protocols for C–C bond forma-
tions are still of interest and a challenging
task for industrial and academic research.[3]
For instance numerous impressive pro-
cesses have been reported for palladium
catalysis, but some issues arise from its
low abundance in the earth crust (~0.006
However, some sustainability issues arise
from the application of 1,2-dihaloethanes, Results and Discussion
e.g. origin from fossil resources, formation
of waste, reactivity of the side product.
The iron complexes 1a–1d and 2a,2b
Recently, the group of Severin reported an were synthesized in accordance to recently
alternative iron-catalyzed oxidative homo- reported procedures. In more detail, the re-
coupling approach by replacing 1,2-diha- action of iron(ii) chloride with the corre-
loethanes with the industrial waste prod- sponding substituted imidazole or pyridine
uct nitrous oxide (N2O), forming as side and subsequent addition of molecular oxy-
product unproblematic N .[9] Interestingly, gen allowed access to the complexes.[13–15]
anthropogenic N2O is pr2oduced as waste With this selection of precatalysts in
product, e.g. in industrial Nylon produc- hand we evaluated the potential in the
*Correspondence: Prof. Dr. S. Enthaler
Technische Universität Berlin
Institute of Chemistry: Cluster of Excellence
‘Unifying Concepts in Catalysis’
Straße des 17. Juni 115/C2
D-10623 Berlin, Germany
E-Mail: stephan.enthaler@tu-berlin.de

328 CHIMIA 2015, 69, No. 6
CatalytiC aCtivation of Small moleCuleS
to the Grignard reagent 3 and complex 1d
(0.01 mol%) under an atmosphere of N O.
For instance in the presence of differ2ent
organic acid derivatives (amide, ester, ni-
triles) complex 1d successfully performed
the homo-coupling of 3 with somewhat
reduced activity, e.g. for nitrile 7 only
64% yield of 4 were realized within 5 min
(Table 2, entry 3), while with the ester-con-
taining additive 6 full conversion with an
excellent selectivity was obtained (Table 2,
entry 2). Importantly, no conversion of the
functional group was monitored (Table 2,
entries 1–3). In contrast, the addition of an
additive containing a ketone or sulfoxide
function revealed the transformation of the
functional group (Table 2, entries 4 and 5).
In the case of addition of bromobenzene 10
thehomo-couplingoftheGrignardreaction
was exclusively performed by complex 1d,
while the cross-coupling product was not
observed (Table 2, entry 6). Interestingly,
this result can be attractive for controlling
the reaction outcome of reaction sequenc-
es, e.g. selective nucleophile–nucleophile
Fig. 1. Iron-catalyzed cross-coupling approaches.
Table 1. Nitrous oxide dependent iron-catalyzed oxidative homo-coupling of aryl Grignard reagent 3.
iron-catalyzed homo-coupling reaction of
4-methoxyphenylmagnesium bromide 3 to
produce 4,4'-dimethoxybiphenyl 4 (Table
1).[16] Initially, the reaction was performed
under a nitrogen atmosphere with cata-
lytic amounts (0.1 mol%) of complex 1d
at room temperature, but no product for-
mation was observed (Table 1, entry 1). In
contrast, changing the atmosphere to mo-
lecular oxygen revealed the formation of
64% coupling product after 5 min (Table 1,
entry 2), while with nitrous oxide the yield
of 4 was improved towards 95% (Table 1,
entry 3). Interestingly, a significant gas
evolution and heat production was noticed
in the presence of N O. A similar reactiv-
ity was observed in 2the presence of 0.05
mol% precatalyst 1d, which corresponds
to a turnover number of 950 and a turnover
Entrya Precatalyst Catalyst loading
Gas
Time [min] Yield [%]
frequency[17] of 11400 h–1 (Table 1, entry
[mol%]
6). In contrast to the work of Severin and
co-workers[9] with catalytic amounts of
metal salts (reaction time: 1 h) and our ear-
lier investigation with a trispyrrole iron(ii)
complex[12] (reaction time: 24 h) excellent
yields (>90%) were obtained within short
reaction times (5 min). Noteworthy, apply-
ing complexes 1a–1c and 2a as precata-
lysts similar yields (93–97%) were moni-
1
2
3
4
5
6
1d
1d
1d
1d
1d
1d
0.1
0.1
0.1
0.1
0.1
0.05
N2
5
5
<1
64
95
97
97
95
O2
N2O
N2O
N2O
N2O
5
10
30
5
7
1a
0.1
N O
5
93
tored for product 4 after 5 min (Table 1,
2
entries 7–10), while with complex 2b no
product formation was detected, due to low
8
9
1b
1c
0.1
0.1
N2O
N2O
5
5
97
95
solubility (Table 1, entry 11).
Afterwards the functional group toler-
ance of the iron-catalyzed oxidative homo-
10
11
2a
2b
0.1
0.1
N2O
N2O
5
5
96
<1b
coupling protocol was studied (Table 2).
Based on that, an intramolecular screen-
ing protocol was applied.[18] In more de-
tail, stoichiometric amounts of an additive
containing a functional group was added
aReaction conditions: 3 (1.0 mmol) in THF (0.5 M), subsequently the atmosphere was replaced
and a solution of 1 (0.001 mmol) in THF (0.1 mL) was added. The mixture was stirred for 5–30 min
at r.t. Yields were determined by GC-MS analysis with n-dodecane as internal standard. bPoor
solubility of 2b.

CatalytiC aCtivation of Small moleCuleS
CHIMIA 2015, 69, No. 6 329
Table 2. Nitrous oxide dependent iron-catalyzed oxidative homo-
coupling – intermolecular reaction evaluation.
product 18 (73%) as the major component.
Importantly, the amount of the cross-cou-
pling product 18 surpasses the statistic dis-
tribution, demonstrating the control of the
selectivity by the catalyst. A similar result
was obtained when replacing phenyl mag-
nesium bromide by phenyl magnesium
chloride (Table 3, entry 8). Moreover, an
increased yield (84% of the product mix-
Entrya
Additive
Yield [%] 4
Stability
of the Additive
ture) for the cross-coupling product was
noticed for the reaction of 3 and 13 (Table
1
87
Yes
3, entry 9). Next the reaction of sp2 hybrid-
ized Grignard reagents with sp3 hybrid-
ized Grignard reagents was studied. For
instance cyclohexyl magnesium bromide
21 was reacted with 3 in the presence of
5
6
2
3
>99
64
Yes
Yes
1d and N2O (Table 3, entry 10). As major
pathway the homo-coupling of 21 was ob-
served (61% of the product mixture), while
the cross-coupling product was observed
in 37%. In contrast to that, changing the sp3
hybridized carbon function from second-
ary to primary improved the yield of the
7
cross-coupling product (Table 3, entry 11).
Finally, the cross-coupling of sp3 hybrid-
4
<1
No
ized Grignard reagents was investigated
(Table 3, entry 12), revealing the cross-
coupling product as major component of
the product mixture.
8
9
5
6
76
83
No
Summary
Yes
In summary we have demonstrated
the catalytic potential of simple and eas-
ily accessible iron complexes in the oxi-
dative coupling of Grignard reagents.
Importantly, to force the reaction, addi-
tion of nitrous oxide was necessary to ob-
tain yields up to >99% at mild conditions
(e.g. 1 atm, ambient temperature) and low
catalyst loadings (0.1 mol%) within short
reaction times (5 min). Interestingly, ex-
cellent catalyst performance was realized
with turnover numbers of up to 1000 and
turnover frequencies of up to 12000 h–1.[17]
Moreover, an intramolecular screening
protocol with functional group containing
10
11
7
8
58
71
Yes
Yes
12
a Reaction conditions: 3 (1.0 mmol) in THF (0.5 M), additive (1.0 mmol),
subsequently the atmosphere was replaced and a solution of 1a (0.001
mmol) in THF (0.1 mL) was added. The mixture was stirred for 5 min
at r.t. Yields were determined by GC-MS analysis with n-dodecane as
internal standard.
coupling in the presence of an electrophile coupling of sp2-sp2 hybridized Grignard additives revealed in most cases excel-
under N O and nucleophile-electrophile reagents the coupling of sp3-sp3 hybridized lent selectivity for the oxidative homo-
Grignard reagents was investigated, result- coupling (e.g. amide, ester, nitrile, alkene,
Next the potential of complex 1d in ing in a somewhat lower yield of 56% for alkyne). Afterwards the reaction of differ-
the oxidative homo-coupling of different 17 (Table 3, entry 6).
ent Grignard reagents showed interesting
Grignard compounds was evaluated (Table After having established the oxidative results with respect to the selectivity of the
coupling2under N2.
3, entries 1–6). In more detail, an excellent homo-coupling of Grignard reagents with formation of the cross-coupling product.
performance (yield: >99%) was observed N O we focused on the transfer of the pro-
for the conversion of Grignard reagent 13 to2col towards the oxidative cross-coupling
Acknowledgements
Financial support from the Cluster of
containing a thioether function, which has of two different Grignard reagents (Table
Excellence “Unifying Concepts in Catalysis”
been reported as a useful leaving group 3, entries 7–12). Initially, in the presence
(funded by the Deutsche Forschungsgemein-
for C–C bond formations (Table 3, entry
of complex 1d (0.1 mol%) and N O equal
2).[19] Moreover, the sterically demanding amounts of 3 (0.5 mmol) and 214 (0.5
mesityl Grignard reagent 15 was applied as mmol) were reacted. Interestingly, after 5
schaft and administered by the Technische
Universität Berlin) is gratefully acknowledged.
substrate (Table 3, entry 4). Unfortunately,
minutes a mixture of three products was
Received: April 16, 2015
no coupling product was observed, prob- detected (Table 3, entry 7). On the one
ably due to steric reasons. A similar result hand the homo-coupling products of 3
was obtained for the radical clock substrate (5% of the product mixture) and 14 (22%)
16 (Table 3, entry 5). In addition to the and on the other hand the cross-coupling

330 CHIMIA 2015, 69, No. 6
CatalytiC aCtivation of Small moleCuleS
A. Mosier, C. Kroeze, C. Nevison, O. Oenema,
S. Seitzinger, O. van Cleemput, Nutr. Cycle.
Agroecosys. 1998, 52, 225.
Table 3. Iron-catalyzed oxidative homo-coupling and hetero-coupling of aryl Grignard
compounds.
[11] A. Shimizu, K. Tanaka,
M. Fujimori,
Entrya
Grignard
reagent A
Grignard Yield [%]
reagent B
Ratio of Productsc
Chemosphere - Global Change Science 2 2000,
425.
[12] M. Weidauer, E. Irran, S. Enthaler, Inorg. Chem.
A–A
A–B
B–B
1
–
–
–
–
95
>99
92
–
–
–
–
–
–
–
–
–
Commun. 2015, 54, 1.
[13] K. Schröder, S. Enthaler, B. Bitterlich, T.
Schulz, A. Spannenberg, M. K. Tse, K. Junge,
M. Beller, Chem.–Eur. J. 2009, 15, 5471.
[14] P. Döhlert, E. Irran, R. Kretschmer, S. Enthaler,
Inorg. Chem. Commun. 2015, 51, 4.
[15] M. Haberberger, K. Mayer, E. Irran, R.
Kretschmer, S. Enthaler, submitted.
3
13
14
2
3
4
–
–
–
[16] General procedure for iron-catalyzed coupling
of aryl Grignard reagents: A Schlenk tube
was charged with the corresponding Grignard
reagent (1.0 mmol, 0.5 M in THF, 2.0 mL), and
a stirring bar, subsequently the N2-atmosphere
was replaced by N2O and the mixture was
<1
stirred for 5 minutes. From a stock solution
(0.01 M) of precatalyst 1d (0.1 mol%) 0.1 mL
were transferred via a syringe into the tube
containing the Grignard reagent. The tube was
sealed and the mixture was stirred for 5 minutes
at room temperature. After the reaction was
completed an aliquot was taken and diluted in
dichloromethane, the yields were determined
by GC-MS analysis applying n-dodecane as
internal standard. 4: MS (EI) m/z = 214 (100,
M+), 200 (12), 199 (84), 171 (29), 156 (9), 128
(18). 13a: MS (EI) m/z = 246 (100, M+), 231
(82), 216 (15), 184 (24), 171 (11), 152 (24), 139
(19), 123 (17). 14a: MS (EI) m/z = 154 (100,
M+), 153 (38), 152 (27), 77 (10), 76 (21). 17a:
MS (EI) m/z = 198 (1, M+), 85 (36), 71 (61), 57
(100). 18: MS (EI) m/z = 184 (100, M+), 169
(46), 141 (38), 115 (31). 20: MS (EI) m/z = 230
15
16
17
5
6
–
–
<1
56
–
–
–
–
–
–
7
8
3
3
14
3
42b
34b
4: 5
18: 73 14a: 22
4: <1
14a:29 18: 70
(100, M+), 215 (87), 187 (23), 172 (23), 139
(27), 128 (11), 115 (23), 63 (16). 21a: MS (EI)
m/z = 166 (1, M+), 82 (100), 67 (61), 55 (72).
22: MS (EI) m/z = 190 (27, M+), 147 (69), 134
(29), 121 (69), 103 (15), 91 (100), 89 (12), 77
(46), 65 (40). 23: MS (EI) m/z = 206 (1, M+),
121 (100), 91 (15), 77 (15). 24: MS (EI) m/z =
182 (1, M+), 83 (100), 67 (30), 55 (95).
19
21
9
13
3
80b
46b
4: 5
20: 84 13a: 11
4: 2
10
21a:61 22: 37
[17] For calculation of the turnover frequency the
yield after 5 min was taken into account.
[18] K. D. Collins, F. Glorius, Acc. Chem. Res. 2015,
11
12
3
17
21
39b
74d
4: 7
23: 53 17a: 40
17
17a: 5 24: 53 21a: 35
DOI: 10.1021/ar500434f.
aReaction conditions: Grignard reagent (1.0 mmol for homo coupling) or Grignard reagent A (0.5
mmol) and Grignard reagent B (0.5 mmol) (for hetero coupling) in THF (0.5 M), subsequently the
atmosphere was replaced and a solution of 1a (0.001 mmol) in THF (0.1 mL) was added. The
mixture was stirred for 5 min at r.t. Yields/conversions were determined by GC-MS analysis
applying n-dodecane as internal standard. bValue corresponds to the conversion of compound 3.
cRatio of the homo-coupling products A-A, B-B and the hetero-coupling product A-B determined
by GC-MS analysis. dValue corresponds to the conversion of compound 17.
[19] C. I. Someya, M. Weidauer, S. Enthaler, Catal.
Lett. 2013, 143, 424.
[7] See for instance: I. Bauer, H.-J. Knölker, Chem.
[1] See for instance: ‘Metal-Catalyzed Cross-
Coupling Reactions and More’, Eds. A. de
Rev. 2015, dx.doi.org/10.1021/cr500425u and
references therein.
Meijere, S. Bräse, M. Oestreich, 2014, Wiley- [8] See for instance: a) G. Cahiez, C. Chaboche, F.
VCH Weinheim.
Mahuteau-Betzer, M. Ahr, Org. Lett. 2005, 7,
1943; b) T. Nagano, T. Hayashi, Org. Lett. 2005,
7, 491; c) G. Cahiez, A. Moyeux, J. Buendia, C.
[4] ‘List of Periodic Table Elements Sorted by Duplais, J. Am. Chem. Soc. 2007, 129, 13788;
Abundance in Earth’s crust’, Israel Science and d) W. Liu, A. Lei, Tetrahedron Lett. 2008, 49,
Technology Homepage. Retrieved 2007-04-15. 610; e) T. Itoh, K. Kude, A. Ishioka, S. Hayase,
[5] ‘Platinum Today’, www.platinum.matthey.com, M. Kawatsura, ECS Trans. 2008, 13, 47; f) G.
accessed 15 September 2014. Cahiez, L. Foulgoc, A. Moyeux, Angew. Chem.,
[6] See for instance: a) P. T. Anastas, M. M. Int. Ed. 2009, 48, 2969; g) W. Yang, J. Zhou, B.
Kirchhoff, Acc. Chem. Res. 2002, 35, 686; b) Wang, H. Ren, Chem.–Eur. J. 2011, 17, 13665.
J. L. Tucker, Org. Process Res. Dev. 2010, 14, [9] G. Kiefer, L. Jeanbourquin, K. Severin, Angew.
328; c) R. Dach, J. J. Song, F. Roschangar, W. Chem. Int. Ed. 2013, 52, 6302.
Samstag, C. H. Senanayake, Org. Process Res. [10] See for instance: a) J. C. Kramlich, W. P. Linak,
Dev. 2012, 16, 1697. Prog. Energ. Combust. Sci. 1994, 20, 149; b)
[2] M. Beller, Chem. Soc. Rev. 2011, 40, 4891.
[3] P. T. Anastas, ChemSusChem 2009, 2, 391.