Bismuth Compounds in Radical Catalysis: Transition Metal Bismuthanes Facilitate Thermally Induced Cycloisomerizations

Article abstract of DOI:10.1002/anie.201904365

The controlled radical chemistry of bismuth compounds is still in its infancy. Further developments are fueled by the properties of these complexes (e.g., low toxicity, high functional group tolerance, low homolytic bond dissociation energies, and reversible homolytic bond dissociations), which are highly attractive for applications in synthetic chemistry. Here we report the first catalytic application of transition metal bismuthanes (i.e. compounds with a Bi–TM bond; TM=transition metal). Using the catalyzed radical cyclo-isomerization of δ-iodo-olefins as a model reaction, characteristics complementary or superior to known B, Mn, Cu, Zn, Sn, and alkali metal reagents are demonstrated (including a different crucial intermediate), establishing transition metal bismuthanes as a new class of (pre-)catalysts for controlled radical reactions.

Full text of DOI:10.1002/anie.201904365

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Bismuth Compounds in Radical Catalysis: Transition Metal
Bismuthanes Facilitate Thermally-Induced Cyclo-Isomerizations
Authors: Jacqueline Ramler, Ivo Krummenacher, and Crispin
Lichtenberg
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201904365
Angew. Chem. 10.1002/ange.201904365
Link to VoR: http://dx.doi.org/10.1002/anie.201904365
http://dx.doi.org/10.1002/ange.201904365

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
Bismuth Compounds in Radical Catalysis: Transition Metal
Bismuthanes Facilitate Thermally-Induced Cyclo-Isomerizations
Jacqueline Ramler,[a] Ivo Krummenacher, and Crispin Lichtenberg*[a]
Abstract: The controlled radical chemistry of bismuth compounds is
still in its infancy. Further developments are fueled by the properties
of these complexes (e.g.: low toxicity, high functional group
tolerance, low homolytic bond dissociation energies, and reversible
homolytic bond dissociations), which are highly attractive for
applications in synthetic chemistry. Here we report the first catalytic
application of transition metal bismuthanes (i.e. compounds with a
Bi–TM bond; TM = transition metal). Using the catalyzed radical
hexene as a model reaction, we demonstrate here the first
catalytic version of this reaction under thermal conditions,
revealing the great potential of transition metal bismuthanes to
be exploited as a new class of readily tunable catalysts in
controlled radical reactions.
Starting from the diaryl bismuth chloride [Bi{(C6H4CH2)2S}Cl]
(1)[19] a series of closely related transition metal bismuthanes 2-4
was synthesized and isolated in 73-80% yield (Scheme 1).
According to single-crystal X-ray diffraction analyses,[47] 2-4
possess typical molecular structures in the solid state without
directed intermolecular interactions (4 is exemplarily shown in
Scheme 2; for details see Supp. Inf.). The transition metal atoms
are found in distorted tetrahedral (3), trigonal bipyramidal (2; Bi1
and P1 in the axial positions; τ5 = 0.91), and octahedral (4)
coordination geometries, respectively. The Bi–C (2.27-2.29 Å)
and Bi–TM bond lengths (TM = Co, Fe, Mn: 2.70, 2.69, 2.86 Å)
are in the expected ranges.[15-18] It is important to note that in all
cases, the thioether functionality does not interact with the
bismuth center. This type of interaction is observed when the
bismuth atom in such complexes is part of a Biδ+–Xδ– bond which
is polarized towards X (e.g. X = Cl, Br, I, NO3, O3SCF3).[19] Thus,
the absence of Lewis acid/base interactions between the
bismuth atom and the thioether group is a first hint towards non-
or only weakly polarized Bi–TM bonds in compounds 2-4.
Compounds 2-4 were analyzed by solution IR spectroscopy.
In particular, the CO stretching frequencies were used as a tool
for the experimental approximation of the partial charge at the
transition metal centers. Thus, the CO stretching frequencies of
2-4 were compared to those of [TM]δ+–Cl/Br, [TM]2, and [Cat]–
[TM]δ– ([TM] = Co(CO)3(PPh3), Fe(C5H5)(CO)2, Mn(CO)5; Cat =
countercation). This comparison is visualized for the relevant
iron compounds in Figure 1.[20,21] Accordingly, the Bi–Fe bond in
3 and the Bi–Mn bond in 4 are essentially non-polar or very
weakly polarized, whereas for the Bi–Co bond in 2, the data
analysis only allowed to rule out a polarization towards Co (for
details see Supp. Inf.).
cyclo-isomerization of δ-iodo-olefins as
a
model reaction,
characteristics complementary or superior to known B, Mn, Cu, Zn,
Sn, and alkali metal reagents are demonstrated (including a different
crucial intermediate), establishing transition metal bismuthanes as a
new class of (pre-)catalysts for controlled radical reactions.
Complexes based on the heavy main group element bismuth
have been associated with low homolytic bond dissociation
energies of Bi–X bonds (X = H, alkyl, allyl, aryloxide, amide).[1-5]
This is due to (i) an inefficient orbital overlap of the substituents
with large and diffuse bismuth-centered orbitals, (ii) a relatively
low polarity of the Bi–X bond, and (iii) a stabilization of bismuth-
centered radical species by relativistic effects.[6] Such homolytic
bond dissociations have been exploited in synthetic chemistry,
polymerization catalysis, and materials science. Examples
include the synthesis of dibismuthanes [Bi(aryl)2]2 in
dehydrogenative coupling reactions from in situ generated
HBi(aryl)2,[1a,b] the polymerization of activated olefins,[2] and the
deposition of thin films of BiOx, SrxTayOz, and strontium bismuth
tantalate in metal-organic chemical vapor deposition.[1c,d]
Reactive radical species that are generated in the course of
such reactions can be extremely difficult or even impossible to
detect experimentally. Hence, their existence was often
suggested merely based on the analysis of stable diamagnetic
products.[2,4,7] Only recently, the first examples of well-defined
persistent[8] or even isolable bismuth radical compounds have
been reported.[9-11] To date, the only catalytic application of an
isolable bismuth radical is the dehydrogenative coupling of
SiPhH3 with TEMPO to give SiPhH2(TEMPO) (TEMPO =
tetramethylpiperidinyl-N-oxide).[12,13] While of great fundamental
significance, the bismuth radical showed a low catalytic activity
(75% conversion after 15 d)[14] and the use of the isolable radical
TEMPO as a substrate was necessary to facilitate the reaction.
In the context of controlled radical reactions, we became
interested in transition metal bismuthanes, i. e. molecules
containing the structural motif R2Bi–[TM] ([TM] = transition metal
complex fragment). This large class of compounds shows a
broadly tunable set of properties – an ideal setting for catalytic
applications.[15-17] However, the stoichiometric or even catalytic
reactivity of Bi–[TM] bonds towards organic substrates is largely
unexplored. Using the radical cyclo-isomerization of 6-iodo-1-
The reactivity of compound 4 towards the radical scavengers
diphenyldisulfide and CCl4 was investigated (Scheme 2 and
Supp. Inf.). Reactions of 4 with 1 equiv. of (PhS)2 resulted in
formation of [Bi{(C6H4CH2)2S}(SPh)] (6) in 94% spectroscopic
yield, which was isolated and fully characterized. Single-crystal
X-ray analysis confirmed an S1→Bi1 donor/acceptor interaction
as a result of the polarized Bi1δ+−S2δ− bond (vide supra; for
details see Supp. Inf.).[47] Reactions of 4 with CCl4 gave
[a]
J. Ramler, Dr. I. Krummenacher, Dr. C. Lichtenberg
Institute of Inorganic Chemistry,
Julius-Maximilians-University Würzburg
Am Hubland, 97074 Würzburg, Germany
E-mail: crispin.lichtenberg@uni-wuerzburg.de
Scheme 1. Synthesis of transition metal bismuthanes 2-4. Compound 2 was
synthesized from 1 and Na[Co(CO)4] in the presence of PPh3. L = thf.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
sunlamp) are necessary to facilitate these reactions. In contrast,
there are no reports of the thermally-induced catalyzed radical
cyclo-isomerization of 6-iodo-1-hexene. This would allow the use
of substrates with functional groups that do not tolerate strongly
basic/nucleophilic conditions and facilitate this reaction under
simplified conditions (i.e. without the need for a photochemical
apparatus). We set out to validate the catalytic competence of
transition metal bismuthanes in the cyclo-isomerization of 6-
iodo-1-hexene (S1) under thermal conditions. In order to
contribute towards a comprehensive understanding of this
reaction, we reproduced the results that were previously
reported with 10 mol% [Sn(nBu)3]2 as a catalyst (photochemical
conditions with λmax < 400 nm),[24] which resulted in formation of
the cyclo-isomerized product P1 in 90% yield (Table 1, entry
Figure 1. CO stretching frequencies of compounds Fp–E (Fp
Fe(C5H5)(CO)2), where E is an electropositive or electronegative binding
partner, another Fp moiety,[21] or a bismuthyl group.
=
1).[28] When the wavelength of the light source was set to λmax
=
400 nm (light emitting diode, P = 3 W), no significant conversion
of S1 was observed (entry 2). Importantly, [Sn(nBu)3]2 also failed
to catalyze the formation of P1 under thermal conditions (60 °C,
20 h; entry 3).[29] Common radical sources such as AIBN were
also tested as catalysts under thermal conditions, but gave only
low conversions (entry 4; AIBN = azo-bis-isobutyronitrile).[30]
Simple Cu(I) salts in combination with suitable ligand platforms
have recently been shown to effectively catalyze the cyclo-
isomerization of activated halo-olefins,[22g] but mixtures of CuCl
or CuI with 2,2’-bipyridine did not catalyze the cyclo-
isomerization of S1 (entries 5,6). Transition metal carbonyl
species such as [Mn(CO)5]2 have been reported as
stoichiometric reagents in a hydro-dehalogenative cyclization of
S1.[31] [Mn(CO)5]2 also catalyses related carboacylations of
unactivated iodo-olefins, when pressurized with 10 atm of CO.[32]
However, both types of transformations require irradiation of the
sample with a sunlamp or ambient light. When 10 mol% of
[Mn(CO)5]2 were reacted with S1 at 60 °C, only a low yield (17%)
of the cyclo-isomerization product P1 was detected in the
presence of ambient light, and no conversion was observed
when the reaction was performed in the dark (entries 7,8).
Triethylborane in combination with traces of O2 has been
introduced as a reagent for the cyclo-isomerization of iodo-
olefins in water or water/MeOH as the most suitable
solvents.[33a,b] So far, activated and secondary iodides have been
reported as substrates.[33] Using these standard protocols, BEt3
did not effectively catalyze the cyclo-isomerization of S1 (entries
9,10). In contrast to previous reports on other substrates, a
higher conversion was obtained with benzene as the solvent. In
our hands, moderate yields of up to 55% P1 were obtained,
when S1 was treated with 10 mol% BEt3 in benzene for 20 h at
23 °C in the presence of traces of O2 (entry 11). Finally, the
transition metal bismuthanes 2-4 were tested as catalysts in the
cyclo-isomerization of S1. Whereas the cobalt species 2 showed
low activity, the iron compound 3 and the manganese compound
4 led to yields of 24% and 73%, respectively (entries 12-14).
Focusing on manganese bismuthanes as the most active
species, the synthesis of literature-known[15a] [BiPh2(Mn(CO)5)]
(7) was improved (Supp. Inf.). In the presence of 7 (10 mol%),
S1 smoothly cyclo-isomerized at 60 °C to give P1 in 96% yield
after 20 h (entry 15; for time-conversion-plot see Supp. Inf.). The
differences in the catalytic activity of 2, 3, 4, and 7 were ascribed
to the presence/absence of the thioether bridge in the ligand
backbone and to the variation of the transition metal complex
fragments, each of which may enter different reaction pathways
under catalytic conditions (for details see Supp. Inf.). The high
catalytic competence of 7 in thermally-induced reactions is in
Scheme 2. Reactions of [Bi{(C6H4CH2)2S}(Mn(CO)5)] (4) with (PhS)2 and CCl4
to give [Bi{(C6H4CH2)2S}(SPh)] (5) and [Bi{(C6H4CH2)2S}(CCl3)] (6),
respectively. Molecular structures of 4 and 5 in the solid state: selected bonds
lengths (Å): Compound 4: Bi1–Mn1, 2.8576(8); Bi1–C, 2.275(5)-2.277(5);
Mn1–CO(cis), 1.844(6)-1.871(6); Mn1–CO(trans), 1.818(6); C–Bi1–C,
105.56(18);
Mn1–Bi1–C,
96.73(12)-101.12(12);
Bi1–Mn1–CO(trans),
176.77(17); Bi1–Mn1–CO(cis), 82.19(16)-88.37(15). Compound 5: Bi1−C,
2.254(5)-2.287(5); Bi1−S1, 2.9401(15); Bi1−S2, 2.6149(16).
Mn(CO)5Cl and [Mn(CO)4Cl]2 (identified by comparison of IR
spectra and single crystal XRD data with the literature) along
with [Bi(C6H4CH2S)2(CCl3)] (6) (identified by NMR spectroscopy
and HRMS analysis). These results suggest that 4 undergoes
radical reactions with suitable substrates.
The cyclo-isomerization of 6-iodo-1-hexene and related
compounds is a powerful tool for the synthesis of cyclo-pentane
structural motifs while preserving the synthetically valuable iodo
functionality (see reaction in Table 1).[22,23] From a mechanistic
point of view, radical and polar reaction pathways have been
suggested, depending on the reagents used (vide infra). It is
important to note that the substitution pattern at the substrate
also crucially influences the feasibility and mechanistic pathways
of this reaction. For the parent substrate, 6-iodo-1-hexene (S1),
a limited number of protocols is available for the cyclo-
isomerization to give iodo-methyl-cyclopentane (P1). One
stoichiometric reaction relies on Zn and I2 (along with small
amounts of O2) and proceeds via a radical pathway.[24]
A
catalytic version was realized with strongly basic/nucleophilic
organolithium compounds and most likely proceeds via a polar
reaction mechanism.[25a] A catalyzed radical reaction can be
considered a standard protocol and is based on toxic organotin
species [SnR3]2.[5,24a,26,27] Photochemical conditions (e.g.: 275 W
This article is protected by copyright. All rights reserved.

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
Table 1. Potential catalysts for the radical cyclo-isomerization of S1 to give P1.
(alkyl)(diphenyl)bismuth compound [BiPh2((CH2)4CH=CH2)] (9)
(Table 2).[34] Addition of 9 equiv. S1 followed by heating to 60 °C
for 20 h led to formation of P1 in near-quantitative yield,
demonstrating the catalytic competence of the in situ generated
species (Table 2). Compound 9 was also detected as an
intermediate by NMR spectroscopy under catalytic conditions
(Supp. Inf.), demonstrating fundamental differences to related
reactions facilitated by tin species under photochemical
conditions. The manganese compound 8 was also synthesized
via an independent route[35] and tested for its catalytic
competence in the cyclo-isomerization reaction, where it led to
low conversions of only 7% (entry 4). Attempts to synthesize and
isolate 9 via Mn-free routes were unsuccessful so far; however,
the structurally related compound [Bi((CH2)4CH=CH2)3] could be
obtained in 85% purity by reaction of BiCl3 with in situ generated
[Li(CH2)4CH=CH2] at –78 °C.[36] Indeed, [Bi((CH2)4CH=CH2)3]
showed a significant catalytic activity in the cyclo-isomerization
of S1, albeit extended reaction times of 66 h were necessary to
reach high yields of P1 (86%; entry 5; cf. entry 3).[37] Thus, the
transition metal bismuthane 7 represents a new type of readily
available, storable pre-catalyst for the operationally simple
generation of bismuth species that are active in the catalyzed
cyclo-isomerization of S1. The reaction can also be performed
on the 1 mmol scale (Supp. Inf.).
Catalyzed 5-exo-trig-cyclizations (such as S1 → P1) may
proceed via polar or radical type pathways, depending on the
nature of the substrate, the choice of catalyst, and the reaction
conditions. This can have a dramatic influence on the substrate
scope and the stereochemical outcome of these reactions. Thus,
the cyclo-isomerization of S1 catalyzed by 7 was carried out in
the presence of CCl4 as a radical scavenger. This led to
significantly lower yields of P1 (50%), suggesting that radical
pathways are operative (entry 6 vs. entry 1). To gain direct
evidence for radical participation, the cyclo-isomerization of S1
catalyzed by 7 was performed at 60 °C in the presence of the
nitrone 10 as a radical trap and analyzed by EPR spectroscopy
at 23 °C. Indeed, an isotropic resonance with g = 2.006 and
#
1
Catalyst
[Sn(nBu)3]2
Conditions
20 °C, 1.25 h, h·ν[b]
20 °C, 1.0 h, h·ν[c]
60 °C, 20 h[d]
60 °C, 20 h[d]
60 °C, 20 h[d]
60 °C, 20 h[d]
60 °C, 20 h[d]
60 °C, 20 h[f]
Yield[a]
90
2
2
[Sn(nBu)3]2
3
[Sn(nBu)3]2
<1
10
<1
<1
17
<1
5
4
AIBN
5[e]
6[e]
7
CuCl, 2,2‘-bipy
CuI, 2,2‘-bipy
[Mn(CO)5]2
8
[Mn(CO)5]2
9[g]
10[i]
11
12[k]
13
14
15
BEt3 (solvent: H2O)
BEt3 (solvent: H2O/MeOH (97:3))
BEt3
23 °C, 20 h[d,h]
23 °C, 20 h[d,h]
23 °C, 20 h[d,h,j]
80 °C, 20 h[d]
60 °C, 20 h[d]
60 °C, 20 h[d]
60 °C, 20 h[d]
4
55
4
[Bi(C6H4CH2)2S(Co(CO)3PPh3)] (2)
[Bi(C6H4CH2)2S(Fe(CO)2(C5H5))] (3)
[Bi(C6H4CH2)2S(Mn(CO)5] (4)
[BiPh2(Mn(CO)5)] (7)
24
73
96
[a] given in %, determined by 1H NMR spectroscopy and/or GC/MS. [b] Hg
lamp. [c] 400 nm (LED). [d] ambient light. [e] The same yield of P1 was
obtained with MeCN as the solvent. [f] in the dark. [g] Water was used as the
solvent. [h] traces of O2. [i] Water/MeOH (97:3) was used as the solvent. [j]
50% yield at 60°C (Supp. Inf.). [k] a yield of <1% was detected, when the
reaction was run for 20 h at 60 °C.
Table 2. Cyclo-isomerization of S1 catalyzed by [BiPh2(Mn(CO)5)] (7) and
related compounds
stark contrast to the performance of established reagents (Table
1), offering potential to be exploited in orthogonal synthetic
strategies and multi-catalysis. These results provide the first
examples for the use of well-defined, molecular transition metal
bismuthanes, R2Bi-[TM], as readily accessible and tunable (pre-
)catalysts in an organic transformation ([TM] = transition metal
complex fragment).
The remarkable activity of 7 as a (pre-)catalyst in the cyclo-
isomerization of S1 under ambient light was maintained, when
the reaction was carried out in the dark (Table 2, entries 1,2).
This is in contrast to the related [Mn(CO)5]2-catalyzed
carboacylations of non-activated iodo-olefins, which require at
least ambient light in order reach significant conversion.[32] The
catalyst loading could be lowered to 1 mol%, albeit longer
reaction times of 66 h were necessary to reach high yields of
86% (entry 3). In order to gain further insights into the cyclo-
isomerization of S1 catalyzed by compound 7, the reaction was
carried out in a 1:1 stoichiometry at 60 °C. IR spectroscopic
analysis of the reaction mixture indicated the presence of the by-
product [MnI(CO)5] (8), which could also be isolated and
unequivocally identified by single-crystal X-ray analysis and IR
spectroscopy of the isolated sample (Table 2). In agreement
with these observations, NMR spectroscopic analyses of the
Catalyst
Conditions
Yield[a] [%]
#
[BiPh2(Mn(CO)5)] (7)
[BiPh2(Mn(CO)5)] (7)
[BiPh2(Mn(CO)5)] (7)[b]
[MnI(CO)5)] (8)
20 h, ambient light
20 h, in the dark
56 h, ambient light
20 h, ambient light
66 h, ambient light
20 h, CCl4 added
96
95
86
7
1
2
3
4
5
6
[Bi((CH2)4CH=CH2)3][c]
[BiPh2(Mn(CO)5)] (7)
86
50
[a] yield of P1 determined by 1H NMR spectroscopy and/or GC/MS analysis.
[b] 1 mol% catalyst was used. [c] [Bi((CH2)4CH=CH2)3] was synthesized via a
Mn-free route in 85% purity according to NMR spectroscopic analyses.
reaction
mixture
revealed
the
formation
of
the
This article is protected by copyright. All rights reserved.

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
operating via radical pathways in this reaction. Polar and
strongly basic organometallic compounds such as MeLi or PhLi
have been reported to catalyze the cyclo-isomerization of S1
and S2.[25] However, when the allyl ethers S3 and S4 were used
as substrates, no cyclo-isomerization was observed and
compounds LiO(C3H5), CH2=CHR (R = H, Me), and MeI/PhI
were detected instead (Table 3, #2,3).[25] These product
distributions clearly demonstrate that polar degradation
pathways are energetically accessible for S3 and S4.
Consequently, these allyl ethers were also tested as substrates
towards 7. Indeed, S3 and S4 underwent smooth cyclo-
isomerization reactions to give P3 and P4 in the presence of 10-
20 mol% of compound 7 (Table 3, #4,5). The slight preference
for formation of cis-P4 over trans-P4 (2.7:1.0) is in agreement
with a radical intermediate.[44]
Figure 2. Left: Use of nitrone 10 as a radical trap in the cyclo-isomerization of
S1 catalyzed by 7. Right: X-band EPR spectrum of 11 in benzene (black =
experimental, red = simulated).
Amide functional groups are also tolerated by the transition
metal bismuthane (pre-)catalyst, as demonstrated by the smooth
quantitative cyclization of S5 to give pyrrolidinone P5 within 4.5
h at 60°C in the presence of 10 mol% 7 (Table 3, #6).[45] The
more reactive tertiary iodide S6 cyclo-isomerized to give P6 in
the presence of 10 mol% 7 at ambient temperature in 5 h with a
yield of 90% (Table 3, #7).[46] Overall, these results further
corroborate the applicability of transition metal bismuthane 7 as
a (pre-)catalyst in cyclo-isomerization reactions of iodo-olefins,
operating via radical pathways.
Table 3. Radical cyclo-isomerization catalyzed by 7 (this work) as compared to
polar degradation (Bailey et al.).
In conclusion, we have demonstrated that transition metal
bismuthanes R2Bi−[TM] (R
= aryl, [TM] = Fe(C5H5)CO2,
Mn(CO)5) undergo highly selective radical reactions. Proof of
principle was delivered using the catalyzed cyclo-isomerization
of δ-iodo-olefins as a model reaction. For the first time, this
catalyzed radical reaction of the parent substrate (6-iodo-1-
hexene) was achieved under thermal conditions. This is the first
example of transition metal bismuthanes to act as highly
selective radical (pre-)catalysts in organic reactions. Thereby,
this class of compounds is established as a new type of (pre-
)catalyst for radical transformations, with properties comple-
mentary or superior to Cu, Zn, Sn, and alkali metal reagents.
#
Substr.
R
R’
Z
E
Li-R’’/Cat., Pathway
7, rad. cycl.
Yield[a]
1[b]
2[c]
3[d]
4[e]
5[f]
6
S2
S3
S4
S3
S4
S5
S6
Me
H
H
CH2
CH2
CH2
CH2
CH2
CO
CH2
O
>99
≈100
≈100
83
H
Li-Me, polar degr.
Li-Ph, polar degr.
Me
H
H
O
H
O
7,
7,
7,
7,
rad. cycl.
rad. cycl.
rad. cycl.
rad. cycl.
Me
H
H
O
>99
>95
90
H
NC3H5
CH2
7[g]
Me
Me
CH2
Acknowledgements
[a] yield of P2-P6 determined by 1H NMR spectroscopy and/or GC/MS
analysis. [b] P2: cis/trans = 3.4:1.0. [c] R’’ = Me, ref. [25]. [d] R’’ = Ph, ref. [25].
[e] 20 mol% catalyst. [f] P4: cis/trans = 2.7:1.0. [g] T = 23 °C.
The authors thank Dr. R. D. Dewhurst for helpful discussions,
Prof. H. Braunschweig for continuous support, as well as the FCI
(Liebig fellowship) and the DFG for generous financial support.
hyperfine couplings of a(14N) = 41 MHz, a(1H) = 8.0 MHz, and
a(13C) = 15.1 MHz was detected (Figure 2).[38] These parameters
are in good agreement with those expected for radical species
11,[39,40] indicating the presence of the cyclopentylmethyl radical
during cyclo-isomerization of S1 catalyzed by 7.
The cyclo-isomerization of 6-iodo-1-heptene (S2) gives two
diastereomers of P2, the cis- and trans- isomer (Table 3). It has
been reported that the diastereoselectivity of this and related
cyclizations is significantly affected by the mechanistic pathway
that dominates the reaction: whereas polar reactions favor
formation of the trans-isomer (cis/trans < 1.0 : 40),[25b,41-43] the
observation of cis/trans ratios > 2.5 : 1.0 have been attributed to
radical pathways.[24a,44] 10 mol% of 7 smoothly catalyzed the
transformation of S2 into P2 with >99% conversion after 3.5 h at
60 °C (Table 3, #1). P2 was formed with a cis/trans ratio of
3.4 : 1.0. This is in line with results reported for the same
Keywords: bismuth • transition metal bismuthanes • radical
reactions • catalysis • cyclo-isomerizations
[1]
[2]
a) L. Balázs, H. J. Breunig, E. Lork, Z. Naturforsch. 2005, 60b, 180-182;
b) G. Balázs, L. Balázs, H. J. Breunig, E. Lork, Organometallics 2003,
22, 2919-2914; c) J.-Y. Heyeon, M. Lisker, M. Silinskas, E. Burte, F. T.
Edelmann, Chem. Vap. Deposition 2005, 11, 213-218; d) M.
Vehkamäki, T. Hatanpää, M. Titala, M. Leskelä, J. Mater. Chem. 2004,
14, 3191-3197.
a) S. Yamago, E. Kayahara, M. Kotani, B. Ray, Y. Kwak, A. Goto, T.
Fukuda, Angew. Chem. Int. Ed. 2007, 46, 1304-1306; b) C. Lichtenberg,
F. Pan, T. P. Spaniol, U. Englert, J. Okuda, Angew. Chem. Int. Ed.
2012, 51, 13011-13015.
[3]
[4]
T. A. Hanna, A. L. Rieger, P. H. Rieger, X. Wang, Inorg. Chem. 2002,
41, 3590-3592.
a) I. J. Casely, J. W. Ziller, M. Fang, F. Furche, W. J. Evans, J. Am.
Chem. Soc. 2011, 133, 5244-5247; b) C. Hering-Junghans, A. Schulz,
M. Thomas, A. Villinger, Dalton Trans. 2016, 45, 6053-6059; c) C.
reaction using [Sn(nBu)3]2 as
a
radical catalyst under
photochemical conditions,[24a] and thus also points towards 7
This article is protected by copyright. All rights reserved.

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
Knispel, C. Limberg, C. Tschersich, Chem. Commun. 2011, 47, 10794-
[22] a) D. P. Curran, Synthesis 1988, 417-439; b) P. A. Baguley, J. C.
Walton, Angew. Chem. Int. Ed. 1998, 37, 3072-3082; c) K. C.
Majumdar, G. V. Karunakar, B. Sinha, Synthesis 2012, 44, 2475-2505;
d) K. C. Majumdar, P. K. Basu, P. P. Mukhopadhyay, Tetrahedron 2005,
61, 10603-10642; e) B. Giese, B. Kopping, T. Göbel, J. Dickhaut, G.
Thoma, K. J. Kulicke, F. Trach, Org. React. 1996, 48, 301-856; f) G. J.
Rowlands, Tetrahedron 2010, 66, 1593-1636; g) A. J. Clark, Chem. Soc.
Rev. 2002, 31, 1-11.
10796.
[5]
The toxicity of bismuth compounds has been reported to be typically
low, especially when compared to related Sn, Pb, Sb, or Te species.
E.g.: J. M. Bothwell, S. W. Krabbe, R. S. Mohan, Chem. Soc. Rev.
2011, 40, 4649-4707 and references therein.
[6]
[7]
P. Pyykkö, Chem. Rev. 1988, 88, 563-594.
A low-valent bismuth compound has been generated in situ, suggested
to have biradical character based on CASSCF calculations, and
trapped by reaction with an alkyne: J. Bresien, A. Hinz, A. Schulz and A.
Villinger, Dalton Trans. 2018, 47, 4433-4436.
[23] For the related, stoichiometric radical-cyclo-isomerization of δ-
alkoxamino-olefins see: K. Molawi, T. Schulte, K. O. Sigenthaler, C.
Wetter, A. Studer, Chem. Eur. J. 2005, 11, 2335-2350.
[8]
[9]
S. Ishida, F. Hirakawa, K. Furukawa, K. Yoza, T. Iwamoto, Angew.
Chem. Int. Ed. 2014, 53, 11172-11176.
[24] a) T. Cohen, H. Gibney, R. Ivanov, E. A.-H. Yeh, I. Marek, D. P. Curran,
J. Am. Chem. Soc. 2007, 129, 15405-15409; b) C. Meyer, I. Marek, G.
Courtemanche, J.-F. Normant, Synlett 1993, 266-268; c) C. Meyer, I.
Marek, G. Courtemanche, J.-F. Normant, Tetrahedron Lett. 1993, 34,
6053-6056; d) C. Meyer, I. Marek, G. Courtemanche, J.-F. Normant,
Tetrahedron 1994, 50, 11665-11692.
a) R. J. Schwamm, J. R. Harmer, M. Lein, C. M. Fitchett, S. Gramville,
M. P. Coles, Angew. Chem. Int. Ed. 2015, 54, 10630-10633; b) C.
Ganesamoorthy, C. Helling, C. Wölper, W. Frank, E. Bill, G. E. Cutsail
III, S. Schulz, Nat. Commun. 2018, 9, 87.
[10] C. Lichtenberg, Angew. Chem. Int. Ed. 2016, 55, 484-486.
[11] For reversible electron transfer at Bi see: J. Ramler, J. Poater, F. Hirsch,
B. Ritschel, I. Fischer, F. M. Bickelhaupt, C. Lichtenberg, Chem. Sci.
2019, 10, 4169-4176.
[25] a) W. F. Bailey, M. W. Carson, J. Org. Chem. 1998, 63, 361-365; b) W.
F. Bailey, M. W. Carson, J. Org. Chem. 1998, 63, 9960-9967.
[26] D. P. Curran, D. Kim, Tetrahedron Lett. 1986, 27, 5821-5824.
[27] Distannanes in radical cyclo-isomerizations of iodo-olefins and iodo-
alkynes have been suggested to act as initiators and/or trapping agents
for I and I2 (depending on substrate and conditions): a) D. P. Curran,
M.-H. Chen, D. Kim, J. Am. Chem. Soc. 1989, 111, 6265-6276; b) M.
Newcomb, D. P. Curran, Acc. Chem. Res. 1988, 21, 206-214; c) A.
Studer, D. P. Curran, Angew. Chem. Int. Ed. 2016, 55, 58-102.
[28] It has been shown that the radical cyclization step of S1 proceeds with
an exo/endo selectivity of 50:1 (ref. [22]); for simplicity, only the major
product (P1) is shown and discussed throughout this work.
[12] R. J. Schwamm, M. Lein, M. P. Coles, C. M. Fitchett, Chem. Commun.
2018, 54, 916-919.
[13] For stoichiometric reactivity studies with bismuth radicals see: a) R. J.
Schwamm, M. Lein, M. P. Coles, C. M. Fitchett, J. Am. Chem. Soc.
2017, 139, 16490-16493; b) R. J. Schwamm, M. Lein, M. P. Coles, C.
M. Fitchett, Angew. Chem. Int. Ed. 2016, 55, 14798-14801.
[14] D. J. Liptrot, M. S. Hill, M. F. Mahon, Angew. Chem. Int. Ed. 2014, 53,
6224-6227.
[15] Selected examples of compounds with BiIII–Mn bonds: a) J. M. Cassidy,
K. H. Whitmire, Inorg. Chem. 1991, 30, 2788-2795; b) A. J. Ashe III, J.
W. Kampf, D. B. Puranik, J. Organomet. Chem. 1993, 447, 197-201; c)
N. A. Compton, R. J. Errington, G. A. Fisher, N. C. Norman, P. M.
Webster, P. S. Jarrett, S. J. Nicholls, A. G. Orpen, S. E. Stratford, N. A.
L. Williams, J. Chem. Soc. Dalton Trans. 1991, 669-676; d) J. M. Wallis,
G. Müller, H. Schmidbaur, Inorg. Chem. 1987, 26, 458-459.
[29] [SnMe3]2 has been reported to act as a catalyst in the cyclo-
isomerization of an activated δ-iodo-olefin under thermal conditions: D.
P. Curran, C.-T. Chang, J. Org. Chem. 1989, 54, 3140-3157.
[30] Also see closely related reactions in ref. [26].
[31] B. Giese, G. Thoma, Helv. Chim. Acta 1991, 74, 1135-1142.
[32] C. M. McMahon, M. S. Renn, E. J. Alexanian, Org. Lett. 2016, 18,
4148-4150.
[16] Selected examples of compounds with BiIII–Fe bonds: a) R. E.
Bachmann, K. H. Whitmire, Inorg. Chem. 1995, 34, 1542-1551; b) J. M.
Wallis, G. Müller, H. Schmidbaur, J. Organomet. Chem. 1987, 325,
159-168; c) M. Shieh, Y. Liou, M.-H. Hsu, R.-T. Chen, S.-J. Yeh, S.-M.
Peng, G.-H. Lee, Angew. Chem. Int. Ed. 2002, 41, 2384-2386; d) R.
Schiwon, R. Metzinger, C. Pfirrmann, C. Limberg, Z. Naturforsch. 2013,
68b, 561-568; e) T. Gröer, M. Scheer, J. Chem. Soc., Dalton Trans.
2000, 647-653; f) T. Gröer, M. Scheer, Organometallics 2000, 19,
3683-3691; g) R. Schiwon, F. Schax, B. Braun, C. Limberg, J.
Organomet. Chem. 2016, 821, 71-77; h) R. Schiwon, B. Braun, R.
Methinger, C. Limberg, Z. Allg. Anorg. Chem. 2016, 642, 1198-1206; i)
G. Bendt, R. Schiwon, S. Salamon, J. Landers, U. Hagemann, C.
Limberg, H. Wende, S. Schulz, Inorg. Chem. 2016, 55, 7542-7549.
[17] Selected examples of compounds with BiIII–Co bonds: a) F. Calderazzo,
R. Poli, G. Pelizzi, J. Chem. Soc. Dalton Trans. 1984, 2535-2540; b) S.
Martinengo, G. Ciani, J. Chem. Soc. Chem. Commun. 1987, 1589-
1591; c) G. Ciani, M. Moret, A. Fumagalli, S. Martinengo, J. Organomet.
Chem. 1989, 362, 291-296; d) also see ref. [16f].
[33] a) H. Yorimitsu, T. Nakamura, H. Shinokubo, K. Oshima, J. Org. Chem.
1998, 63, 8604-8605; b) H. Yorimitsu, T. Nakamura, H. Shinokubo, K.
Oshima, K. Omoto, H. Fujimoto, J. Am. Chem. Soc. 2000, 122, 11041-
11047; c) A. Fusano, S. Sumino, S. Nishitani, T. Inouye, K. Morimoto, T.
Fukuyama, I. Ryu, Chem. Eur. J. 2012, 18, 9415-9422.
[34] With extended reaction times of 20 h, reaction of S1 with one
equivalent of 7 also gives small amounts of the cyclo-isomerization
product P1. Larger quantities of P1 (65%) were detected, when 2 equiv.
MeI were added to the in situ generated mixture of 8 and 9.
[35] K. Oshima, Science of Synthesis 2003, 2, 13-89.
[36] The impurities are compounds, in which olefin functionalities have been
carbometalated by Bi((CH2)4CH=CH2) groups according to NMR
spectroscopy and GC/MS analysis after aqueous work-up.
[37] A slow cyclo-isomerization of [In((CH2)4CH=CH2)3] has been reported: T.
W. Dolzine, J. P. Oliver, J. Organomet. Chem. 1974, 78, 165-176.
[38] A bismuth radical was not directly detected in reactions of S1 catalyzed
by 7, which was ascribed to the low concentration and short life-time of
such species; fast relaxation times may generally prohibit their
detection, even at low temperature (c.f. ref. [8] and Supp. Inf.). No
resonance could be detected by EPR spectroscopy, when S1 and 10
were heated to 60 °C in benzene in the absence of 7.
[18] Compounds of the general formula [R2BiFe(C5H5)(CO)2] (R = aryl, alkyl)
have not been structurally characterized. The Bi–Fe bond length in 3
(2.69 Å) is similar to that in [Bi(Fe(C5H5)(CO)2)3] (2.72 Å); see ref. [16e].
[19] a) R. Qiu, S. Yin, X. Zhang, J. Xia, X. Xu, S. Luo, Chem. Commun.
2009, 31, 4759-4761; b) X.-W. Zhang, J. Xia, H.-W. Yan, S.-L. Luo, S.-
F. Yin, C.-T. Au, W.-Y. Wong, J. Organomet. Chem. 2009, 694, 3019-
3026; c) A. Toma, C. I. Raţ, A. Silvestru, T. Rüffer, H. Lang, M. Mehring,
J. Organomet. Chem. 2016, 806, 5-11.
[39] D. Rehorek, Z. Chem. 1980, 20, 325-332.
[40] Formation of the corresponding nitroxyl radical with a 1-hexene-6-yl
substituent rather than 11 cannot be formally ruled out.
[41] a) W. F. Bailey, T. T. Nurmi, J. J. Patricia, W. Wang, J. Am. Chem. Soc.
1987, 109, 2442-2448; b) E. C. Ashby, T. N. Pham, J. Org. Chem. 1987,
52, 1291-1300; c) K. Deng, A. Bensari-Bouguerra, J. Whetstone, T.
Cohen, J. Org. Chem. 2006, 71, 2360-2372; d) H. Liu, K. Deng, T.
Cohen, K. D. Jordan, Org. Lett. 2007, 9, 1911-1914.
[20] a) A. R. Manning, J. Chem. Soc. A 1968, 1319-1324; b) X. Jiang, L.
Chen, X. Wang, L. Long, Z. Xiao, X. Liu, Chem. Eur. J. 2015, 21,
13065-13072; c) K. Inkrott, R. Goetze, S. G. Shore, J. Organomet.
Chem. 1978, 154, 337-342; d) K. H. Pannell, D. Jackson, J. Am. Chem.
Soc. 1976, 98, 4443-4446.
[42] Adding 6-iodo-1-hexene (S2) to a solution of tBuLi in pentane/Et2O at –
78 °C (rather than the reverse order of addition) gave a cis/trans
product ratio of ca. 1.5-1.9 : 1.0 with low yields of 11-13% (ref. [41a,b]).
[43] Grignard reagents generated from 6-chloro-1-hexene also cyclize with
a preference for the trans-product: a) H. G. Richey Jr., T. C. Rees,
[21] In the discussion of [Fe(C5H5)(CO)2]2, IR spectroscopic data ascribed to
the isomer without bridging CO ligands has been used (ref. [20]).
This article is protected by copyright. All rights reserved.

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
Tetrahedron Lett. 1966, 7, 4297-4301; b) W. C. Kossa Jr., T. C. Rees,
H. G. Richey Jr., Tetrahedron Lett. 1971, 12, 3455-3458.
[44] a) A. L. J. Beckwith, I. Blair, G. Phillipou, J. Am. Chem. Soc. 1974, 96,
1613-1614; b) D. C. Spellmeyer, K. N. Houk, J. Org. Chem. 1987, 52,
959-974.
[45] For cyclization of S5 with initiators such as BEt3, TiCl3, [Sn(nBu)3]2,
organic radical precursors, and related species see: a) M. Ikeda, H.
Teranishi, K. Nozaki, H. Ishibashi, J. Chem. Soc., Perkin Trans. 1 1998,
1691-1697; b) K. Wakabayashi, H. Yormitsu, H. Shinokubo, K. Oshima,
Bull. Chem. Soc. Jpn. 2000, 73, 2377-2378; c) H. Yorimitsu, K. Oshima,
Bull. Chem. Soc. Jpn. 2002, 75, 853-854; d) L. Cao and C. Li,
Tetrahedron Lett. 2008, 49, 7380-7382; e) D. P. Curran, J. Tamine, J.
Org. Chem. 1991, 56, 2746-2750; f) A. G. Hernán, J. Kilburn,
Tetrahedron Lett. 2004, 45, 831-834; g) H. Yorimitsu, K. Wakabayashi,
H. Shinokubo, K. Oshima, Bull. Chem. Soc. Jpn. 2001, 74, 1963-1970.
[46] Here, 7 is an alternative to initiators like polar and reducing LiAlH4 or
toxic [Sn(nBu)3]2, which requires irradiation conditions: a) E. C. Ashby
and C. O. Welder, J. Org. Chem. 1998, 63, 7707-7714; b) see ref. [26].
[47] CCDC 1884682-1884685 contain the crystallographic information for
this work.
This article is protected by copyright. All rights reserved.

10.1002/anie.201904365
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
COMMUNICATION
Bismuth compounds featuring a
bismuth–transition metal bond
catalyze selective radical cyclo-
isomerization reactions of δ-iodo-
olefins. These complexes show
properties complementary or superior
to those of more traditional B, Sn, Cu,
Mn, or alkali metal reagents, which
establishes them as a new type of
(pre-)catalysts for radical
Jacqueline. Ramler, Ivo Krummenacher,
Crispin Lichtenberg*
Page No. – Page No.
Bismuth Compounds in Radical
Catalysis: Transition Metal
Bismuthanes Facilitate Thermally-
Induced Cyclo-Isomerizations
transformations.
This article is protected by copyright. All rights reserved.