Iron-mediated and -catalyzed metalative cyclization of electron- withdrawing-group-substituted alkynes and alkenes with grignard reagents

Article abstract of DOI:10.1002/chem.201101273

Treatment of ethyl (E)-5,5-bis[(benzyloxy)methyl]-8-(N,N-diethylcarbamoyl)- 2-octen-7-ynoate with an iron reagent generated from FeCl₂ and tBuMgCl in a ratio of 1:4 (abbreviated as FeCl₂/4 tBuMgCl) afforded ethyl [4,4-bis[(benzyloxy)methyl]-2-[(E)-(N,N-diethylcarbamoyl)methylene] cyclopent-1-yl]acetate in good yield. Deuteriolysis of an identical reaction mixture afforded the bis-deuterated product ethyl [4,4-bis[(benzyloxy)methyl]-2- [(E)-(N,N-diethylcarbamoyl)deuteriomethylene]cyclopent-1-yl]deuterioacetate, thus confirming the existence of the corresponding dimetalated intermediate. The latter intermediate can react with halogens or aldehydes to facilitate further synthetic transformations. The amount of FeCl₂ was reduced to catalytic levels (10 mol% relative to enyne), and catalytic cyclizations of this sort proceeded with yields comparable to those of the aforementioned stoichiometric reactions. The cyclization of diethyl (E,E)-2,7-nonadienedioate with a stoichiometric amount of FeCl₂/4 tBuMgCl, followed by the addition of sBuOH as a proton source, afforded a mixture of 2-(ethoxycarbonyl)- 3-bicyclo[3.3.0]octanone and its enol form in good yield. The use of aldehyde or ketone in place of sBuOH afforded 2-(ethoxycarbonyl)-3-bicyclo[3.3.0]octanone, which has an additional hydroxyalkyl side chain. Additionally, the metalation of a carbon-carbon unsaturated bond in N,N-diethyl-5,5-bis[(benzyloxy)methyl]-7,8- epoxy-2-octynamide or (E)-3,3-dimethyl-6-(N,N-diethylcarbamoyl)-5-hexenyl p-toluenesulfonate with FeCl₂/4 tBuMgCl or FeCl₂/4 PhMgBr was followed by an intramolecular alkylation with an epoxide or alkyl p-toluenesulfonate to afford 5,5-bis[(benzyloxy)methyl]-3-[(E)-(N,N- diethylcarbamoyl)methylene]-1-cyclohexanol or N,N-diethyl(3,3- dimethylcyclopentyl)acetamide after hydrolysis. In both cases, the remaining metalated portion α to the amide group was confirmed by deuteriolysis and could be utilized for an alkylation with methyl iodide.

Full text of DOI:10.1002/chem.201101273

FULL PAPER
DOI: 10.1002/chem.201101273
Iron-Mediated and -Catalyzed Metalative Cyclization of Electron-
Withdrawing-Group-Substituted Alkynes and Alkenes with Grignard
Reagents
Takeshi Hata, Shiro Sujaku, Naoki Hirone, Kirihiro Nakano, Junsuke Imoto,
Haduki Imade, and Hirokazu Urabe*[a]
Abstract: Treatment of ethyl (E)-5,5-
bis[(benzyloxy)methyl]-8-(N,N-diethyl-
carbamoyl)-2-octen-7-ynoate with an
iron reagent generated from FeCl2 and
tBuMgCl in a ratio of 1:4 (abbreviated
as FeCl2/4 tBuMgCl) afforded ethyl
[4,4-bis[(benzyloxy)methyl]-2-[(E)-
duced to catalytic levels (10 mol% rel-
ative to enyne), and catalytic cycliza-
tions of this sort proceeded with yields
comparable to those of the aforemen-
tioned stoichiometric reactions. The
cyclization of diethyl (E,E)-2,7-nona-
additional hydroxyalkyl side chain. Ad-
ditionally, the metalation of a carbon–
carbon unsaturated bond in N,N-dieth-
yl-5,5-bis[(benzyloxy)methyl]-7,8-
epoxy-2-octynamide or (E)-3,3-dimeth-
yl-6-(N,N-diethylcarbamoyl)-5-hexenyl
dienedioate with
a
stoichiometric
p-toluenesulfonate
with
FeCl2/4
(N,N-diethylcarbamoyl)methylene]cy-
clopent-1-yl]acetate in good yield.
Deuteriolysis of an identical reaction
mixture afforded the bis-deuterated
product ethyl [4,4-bis[(benzyloxy)meth-
yl]-2-[(E)-(N,N-diethylcarbamoyl)deu-
teriomethylene]cyclopent-1-yl]deuter-
ioacetate, thus confirming the existence
of the corresponding dimetalated inter-
mediate. The latter intermediate can
react with halogens or aldehydes to fa-
cilitate further synthetic transforma-
tions. The amount of FeCl2 was re-
amount of FeCl2/4 tBuMgCl, followed
by the addition of sBuOH as a proton
source, afforded a mixture of 2-(ethox-
ycarbonyl)-3-bicycloACTHNUGTRNEUNG[3.3.0]octanone
and its enol form in good yield. The
use of aldehyde or ketone in place of
tBuMgCl or FeCl2/4 PhMgBr was fol-
lowed by an intramolecular alkylation
with an epoxide or alkyl p-toluenesul-
fonate to afford 5,5-bis[(benzyloxy)-
methyl]-3-[(E)-(N,N-diethylcarba-
moyl)methylene]-1-cyclohexanol
N,N-diethyl(3,3-dimethylcyclopentyl)-
acetamide after hydrolysis. In both
or
sBuOH afforded 2-(eth
ACHUTGTNRENoNUG xyCAHUTNGTRENcNNGU arbonyl)-3-
bicyclo[3.3.0]octanone, which has an
T
A
ACHTUNGTRENNUNG
cases, the remaining metalated portion
a to the amide group was confirmed by
deuteriolysis and could be utilized for
an alkylation with methyl iodide.
Keywords: alkenes
·
alkynes
·
Grignard reaction · iron · metallacy-
cles
Introduction
have attracted much attention as safe, abundant, and eco-
nomical reagents for a wide range of organic reactions, in-
cluding coupling reactions, cyclizations, isomerizations, met-
alations, oxidations, reductions, and other types of reac-
tions.[3] Considering the current rapid growth in the number
of iron-mediated reactions, we were also prompted to ex-
plore a new utility of iron, particularly in the formation of
metallacycles from unsaturated compounds. Herein, we
report a synthetic methodology, according to which, an iron
salt is utilized for the metalation of functionalized carbon–
carbon multiple bonds to enable the aforementioned cycliza-
tion and even alkylation with an electrophile. (Scheme 1).
Another advantage is that iron-mediated reactions consti-
tute synthetic methods that are economical and sustaina-
ble.[3]
When the intramolecular cyclization of acetylenes and ole-
fins is performed with a stoichiometric amount of metal spe-
cies, the metal portion may be incorporated into the cyclized
products, producing new and versatile organometallic com-
pounds.[1] However, the applicability of this process to func-
tionalized olefins and acetylenes is limited, probably be-
cause of the poor compatibility of the reactive metallic por-
tion and the functional group.[2] Recently, iron compounds
[a] Dr. T. Hata, S. Sujaku, N. Hirone, K. Nakano, J. Imoto, H. Imade,
Prof. Dr. H. Urabe
Department of Biomolecular Engineering,
Graduate School of Bioscience and Biotechnology
Tokyo Institute of Technology, 4259-B-59 Nagatsuta-cho
Midori-ku, Yokohama, Kanagawa 226-8501 (Japan)
Fax : (+81)045-924-5849
Results and Discussion
E-mail: hurabe@bio.titech.ac.jp
Homepage: http://www.urabe-lab.bio.titech.ac.jp
Stoichiometric enyne and diene cyclizations:[4] We first in-
vestigated the cyclization of doubly functionalized enyne 1
with iron reagents generated by the reduction of FeCl2 with
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101273.
Chem. Eur. J. 2011, 17, 14593 – 14602
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14593

Scheme 4.
Scheme 1. Iron-mediated formation of metallacycles or metalated species
from functionalized unsaturated compounds. FG=functional group,
[Fe]=iron reagent or iron moiety, El+ =electrophile.
rich, low-valent iron reagent probably interacts more strong-
ly with electron-deficient unsaturated bonds in 1 during the
cyclization process.
Overall results of the enyne cyclization are summarized in
Table 1. Functionalized enynes with tert-butyl ester 9, nitrile
10, amide 11, or sulfone 12 moieties reacted equally well to
afford the corresponding cyclopentanes 19–22 after aqueous
workup (Table 1, entries 2–5). Substitution to the tether por-
tion did not impede cyclization, and thus compounds 13 and
14 afforded 23 and 24, respectively (Table 1, entries 6 and
7), with bicyclic heterocycle 24 notably being produced as a
single stereoisomer. Enyne 15, which incorporates two ethyl
ester appendages, generated diester 25 (Table 1, entry 8)
through a cyclization process. The cyclohexane framework
26 was formed in good yield, regardless of the cis- or trans-
olefinic stereochemistry of the starting materials 16 and 17
(Table 1, entries 9 and 10). Finally, the cyclization of func-
tionalized dienyne 18 produced vinyl-substituted cyclopen-
tane 27 in good yield (Table 1, entry 11).
various Grignard reagents.[5] Scheme 2 summarizes its out-
come. Among the Grignard reagents tested, the combina-
tion of FeCl2 and tBuMgCl in a molar ratio of 1:4[6] was opti-
mal, giving the desired cyclization product 2 in 83% yield
with the functional groups intact.[7–9]
The intermediate metallacycle 4, which yielded the cyclic
products shown in Table 1 below through the hydrolysis
route, was additionally able to react with external electro-
philes (such as halogens, aldehydes, or ketones) selectively
at the position shown in Scheme 5 to give the coupling prod-
Scheme 2. Bn=benzyl.
Scheme 3 shows detailed data on the highest-yielding re-
action depicted in Scheme 2. Upon reaction with FeCl2/
4 tBuMgCl (3), enyne 1 afforded 2 as a single olefinic
isomer as depicted in Scheme 3. Deuteriolysis after the reac-
tion gave the deuterated product [D2]2, confirming the pres-
ence of a bismetalated intermediate 4.[10] Interestingly, less
extensively functionalized enynes
5 and 6, shown in
Scheme 4, underwent cyclization to give 7 and 8 in lower
yields, suggesting that the iron reagent 3 prefers functional-
ized to unfunctionalized enynes. Therefore, the electron-
Scheme 5.
ucts 28–32.[11] Thus, further synthetic transformations, such
as functionalization and carbon–carbon bond formation, are
possible from the intermediate metallacycles, demonstrating
their utility as versatile organometallic compounds.
Cyclization of dienedioate 33 with the same iron reagent
3 took a different reaction course, through which the expect-
ed cyclopentane 34 was not isolated after aqueous workup
(Scheme 6). Instead, bicyclic ketoester 35 was formed as a
1:1 mixture of tautomeric keto and enol forms (35 and
Scheme 3.
14594
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14593 – 14602

Iron-Mediated and -Catalyzed Metalative Cyclization
FULL PAPER
36)[12] in unstable yields up to
32%. The ketoester was pro-
duced as a single stereoisomer
Table 1. Cyclization of functionalized enynes according to the reaction shown in Scheme 3.
Entry
Substrate
Period
[h]
Product[a]
Isolated
yield [%]
with
a
thermodynamically
stable
exo-alkoxycarbonyl
group. The cis junction of bicy-
clic compound 35 was spectro-
scopically confirmed in compar-
ison to data of an authentic
sample. A rationalization for
the formation of 35 is also
shown in Scheme 6. The iron-
mediated cyclization of 33 gen-
1
2
3
4
5
R=CO2Et
CO2tBu
CN
(1)
(9)
(10)
(11)
(12)
3.5
4
2
4.5
3
R=CO2Et
CO2tBu
CN
(2)
(19)
(20)
(21)
(22)
83
85
49
92
76
CONEt2
SO2Ph
CONEt2
SO2Ph
6
(13)
4
(23)
66 (d.r.=2:1)[b]
erated meACTHNUGTRENNUGtallaACHTUNGTRENNcUGN ycle 37, one
carbon–metal bond of which
was protonated during aqueous
workup to give 38. The mono-
enolate of diester 38 rapidly un-
derwent Dieckmann condensa-
tion to give the observed prod-
uct 35. Because the usual
workup is not an appropriate
way to perform controlled pro-
tonation of 37, the proper
choice of the proton source, re-
garding its reactivity and stoi-
chiometry, appeared critical for
the efficient conversion of 37 to
38. After investigations along
this line, we found that the use
of sBuOH under the conditions
shown in Scheme 6 led to a
stable and improved yield of 35
(50%).[13–15]
7
(14)
0.5
(24)
64[c]
8
9
(15)
(16)
3
(25)
(26)
61
65
4.5
10
11
(17)
(18)
2
3
(26)
(27)
71
73 (E/Z=1:1)[d]
Table 2 shows various ke-
toesters prepared from diene-
dioates according to Scheme 6.
The yields of 45b from the cor-
[a] Products were isolated after workup with aqueous HCl. [b] d.r.=diastereomeric ratio. [c] Reagent 3
(1.57 equiv) was used. [d] For the disubstituted double bond.
responding E,E- and Z,E-dienes 39b and 40 (61 and 35%,
respectively) show that the former is a more preferable
starting material than the latter (Table 2, entries 2 and 4).
This was again seen in the cyclization of E,E- and Z,Z-dien-
edioates 42 and 43, for which the latter no longer produced
the desired compound 47 in an acceptable yield (Table 2, en-
tries 7 and 8). A substituted ketoester 46 and those fused to
oxa- or aza-heterocycles 47 and 48 were produced from 41,
42, and 44, respectively, in satisfactory yields (Table 2, en-
tries 6, 7, and 9).[16]
While the transformation shown in Scheme 6 is triggered
by protonation of metallacycle 37, this proton could be re-
placed by other electrophiles. In fact, when a ketone or al-
dehyde was added to metallacycle 49 (Scheme 7), an aldol-
type reaction took place instead of a protonation, which was
followed by Dieckmann condensation to give bicyclic com-
pounds 50–53. Ketoester 50 was generated as a single ste-
reoisomer with the hydroxyisopropyl group at the exo posi-
tion, and the stereochemistry assigned to products 51–53
Scheme 6. Tandem cyclization of dienedioate.
Chem. Eur. J. 2011, 17, 14593 – 14602
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14595

H. Urabe et al.
Table 2. Bicyclic ketoesters prepared according to the reaction shown in Scheme 6.
low cost and abundance, as described in the Intro-
duction. Nonetheless, its catalytic use is still attrac-
tive from the following points of view: 1) to mini-
mize transition-metal waste and 2) to modify the re-
activity of the resultant organometallic intermedi-
ates. At the outset, we theorized that the reduction
of the amount of iron would facilitate the 1,2- and
1,4-addition of the highly nucleophilic Grignard re-
agents and, as a result, inhibit cyclization.[18] In fact,
when we attempted the enyne cyclization of 1 with
1 mol% of FeCl2 (and 4 equiv of tBuMgCl relative
to 1), only a small amount of the desired compound
2 was observed (Scheme 8). However, to our de-
light, when the quantity of iron was increased to
5 mol%, the yield of 2 was significantly improved.
Eventually, with 10 mol% of FeCl2, the yield was
further amplified to reach a level comparable to
that of the stoichiometric reaction (cf. 125 mol% of
FeCl2).
Under the optimum conditions given in
Scheme 8, the cyclization generally took place in
good yields, as seen from the results in Table 3. A
variety of functional groups survived these catalytic
conditions to give desired products 2 and 19–22
(Table 3, entries 1–5). Diester 15 cyclized to 25 in a
somewhat better yield than the corresponding stoi-
chiometric reaction (Table 3, entry 6 vs. Table 1,
entry 8). The cyclization of 16 to 26 (containing a
six-membered ring) and the dienyne cyclization of
18 to 27 were as efficient as the stoichiometric reac-
tions (Table 3, entries 7 and 8 vs. Table 1, entries 9
and 11).
Entry
Dienedioate
Product[a]
Isolated
yield [%][b]
1
2
3
R=Me (39a)
Et (39b)
R=Me (45a)
Et (45b)
53
61
iPr (39c)
iPr (45c)
47[c]
4
36[d]
(40)
(33)
ACHTUNGTNER(NUNG 45b)
5
50
(35)
(46)
6
7
57[c]
(41)
(42)
50
(47)
(47)
8
9
(43)
(44)
Because the content of iron is low in the catalytic
reactions, the metalated intermediates mostly con-
sist of magnesium derivatives such as 54
(Scheme 9). Whereas the aforementioned stoichio-
metric iron-based metallacycle 4 in Scheme 5 react-
ed once with halogens, aldehydes, or ketones to
9[e]
62
(48)
[a] Only the keto form is shown. [b] Yield of a mixture of keto and enol forms. [c] Re-
agent 3 (1.5 equiv) was used. [d] Reagent 3 (1.42 equiv) was used. [e] Ts=tosyl.
Scheme 7.
was based on that of 50. This transformation could be con-
sidered as analogous to the generation and reaction of the
dianion of ketoesters 45a–c,[17] which gives derivatives that
have a side chain.
Catalytic enyne cyclization: The iron reagent used in this
study is suitable for stoichiometric reactions, because of its
Scheme 8.
14596
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14593 – 14602

Iron-Mediated and -Catalyzed Metalative Cyclization
FULL PAPER
Table 3. Iron-catalyzed cyclization of functionalized enynes according to the reaction shown in Scheme 8.
lents of the Grignard reagent.
However, higher amounts of
Grignard reagent (up to
4 equiv) are necessary in prac-
tice to break metallacycle 58,
which is stabilized by iron-cen-
tered chelation, into its constit-
uents parts 54 and 56.[19]
Entry
Substrate
Period
[h]
Product
Isolated
yield [%]
1
2
3
4
5
R=CO2Et
CO2tBu
CN
(1)
(9)
(10)
(11)
(12)
4.5
4.5
3.5
4.5
3.5
R=CO2Et
CO2tBu
CN
(2)
(19)
(20)
(21)
(22)
82
80
54
70
70
Stoichiometric cyclization by al-
kylation: Besides the cycliza-
tion of bis-unsaturated com-
pounds described above, the al-
kylation of metalated carbon–
carbon multiple bonds in an in-
tramolecular manner consti-
tutes an alternative method to
CONEt2
SO2Ph
CONEt2
SO2Ph
6
7
8
(15)
(16)
(18)
3.5
3.5
3.5
(25)
(26)
(27)
80
71
prepare
cyclic
compounds.
Whereas intramolecular conju-
gate addition of a nucleophile
to electron-deficient unsaturat-
ed bonds, followed by addition
of an appropriate electrophilic
80 (E/Z=1:1)[a]
[a] For the disubstituted double bond.
Scheme 9. HMPA=hexamethylphosphoric triamide.
give mono-adducts 28–32, the magnesium derivative 54 un-
derwent alkylation twice with allyl bromide to give triene 55
as a 1:1 mixture of two isomers arising from the olefin ge-
ometry or diastereomeric sp3-carbon centers. Alkylations
are generally less efficient than halogenations or aldehyde
additions, and thus, the double alkylation of 54 serves to
prove that 54 is more reactive than the iron-based metalla-
cycle 4 generated by stoichiometric reaction.
Scheme 10. Proposed mechanism of iron-mediated and -catalyzed metala-
tive cyclization.
The proposed catalytic cycle accompanied by the stoichio-
metric reaction pathway is shown in Scheme 10. In the stoi-
chiometric reaction, for which four equivalents of tBuMgCl
are essential for good yields of products, the active iron spe-
cies is probably 57, which is generated from FeCl2 via 56.[6]
The low-valent iron compound 57 reductively combines two
electron-deficient unsaturated bonds of enyne 1 to give the
metallacycle 58, which is analogous to 4 (Ln =(MgCl)2) in
Scheme 3 or 37 (Ln =(MgCl)2) in Scheme 6. Subsequently,
the excess tBuMgCl expels the iron portion from 58 to gen-
erate the magnesium compound 54 and the iron species 56,
which enters a new catalytic cycle to produce 54. As seen in
Scheme 10, the catalytic reaction requires only two equiva-
trap, is a well-established cyclization method (Scheme 11),[20]
the reverse version, which would involve the addition of an
electrophile to 59 (Scheme 12 and thus, Scheme 1b), has not
Scheme 11. Cyclization by addition of nucleophile to electron-deficient
unsaturated carbonyl compounds.
Chem. Eur. J. 2011, 17, 14593 – 14602
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14597

H. Urabe et al.
D2O afforded the deuterated product [D]66, confirming the
presence of a metalated intermediate such as 65. Moreover,
when the reaction was terminated at a low temperature, cis-
olefinic amide 67, which maintained an epoxide moiety, was
recovered as an isolable product (Scheme 15). These obser-
vations are consistent with the reaction course shown in
Scheme 14. It is worth noting that this proposed pathway, in-
volving a substitution of the epoxide group in 63 with the in-
termediate acetylene–iron complex 64, is mechanistically
different from that of a similar nickel-catalyzed reaction of
unfunctionalized acetylene 68 (Scheme 16). In this case, to
account for the formation of cyclohexanol 70,[25] it was pre-
sumed that the nickel complex was inserted to the epoxide
Scheme 12. Cyclization by addition of electrophile to electron-deficient
unsaturated carbonyl compounds.
been documented,[21] even though both of these transforma-
tions are conceptually and practically of equal importance.
To assess the feasibility of the reaction given in
Scheme 12, we first investigated whether the iron reagent
utilized in the preceding sections could actually metalate a
functionalized acetylenic bond (Scheme 13). Among several
ꢀ
C O bond to form 69.
Scheme 15.
Scheme 13.
Scheme 16. COD=1,5-cyclooctadiene.
Grignard reagents tested, the combination of FeCl2 and
tBuMgCl in a molar ratio of 1:4 again proved to be the best.
Thus, treatment of acetylenic amide 60 with 3 (1.25 equiv)
produced cis-olefinic amide 62 in 57% yield after aqueous
workup. Deuteriolysis of the same reaction mixture afforded
the deuterated product [D2]62, confirming the presence of a
dimetallic species 61, which could be formally drawn as an
iron-containing metallacyclopropene.[22]
Encouraged by the above result, we returned to the reac-
tion shown in Scheme 12 and performed the cyclization of
acetylenic amide 63, which contains an epoxide moiety, with
3 (Scheme 14).[23,24] This reaction afforded cyclohexanol 66
with an E/Z ratio of 83:17 in good yield after aqueous
workup. Quenching of the reaction mixture with DCl in
Table 4 shows overall results of the intramolecular substi-
tution of epoxide with functionalized acetylenes. Substrate
72, which had a benzo-fused tether, afforded the bicyclic
product 77 (Table 4, entry 3), and those substrates with an
oxa or aza tether (73 and 74, respectively) afforded the cor-
responding heterocyclic compounds 78 and 79 (Table 4, en-
tries 4 and 5). The elimination of the heteroatom group b to
the carbon–iron bond in the intermediates 73 and 74 did not
appear to be a serious side reaction, as judged by the prod-
uct yields of 78 and 79. The ring closure always occurs in 6-
endo-Tet (or Trig) fashion, and this endo selectivity was also
observed for the cyclopentane annulation of 75 to 80
(Table 4, entry 6).[16]
Alkyl tosylates 81 and 82[23] behaved similarly to give cy-
clopentanes 83 and 84 (Scheme 17). Upon deuteriolysis of
the reaction mixture of 81, product [D]83 was formed with a
high deuterium content, which showed that this reaction
also followed the route illustrated in Scheme 12.[16]
Extension of the above tosylate cyclization to an olefinic
substrate 85 was not so straightforward and we could not
obtain the desired product 86 with 3 (Scheme 18). We subse-
quently employed an array of Grignard reagents
(Scheme 18) and found that FeCl2/4 PhMgBr (3’), which did
not promote the cyclization of acetylenic substrates
Scheme 14.
14598
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14593 – 14602

Iron-Mediated and -Catalyzed Metalative Cyclization
FULL PAPER
Table 4. Cyclization of functionalized acetylene and epoxide according to the reaction shown in Scheme 14.
(Scheme 17) or the enyne cycli-
zation shown in Scheme 2, re-
acted smoothly with olefins to
Entry
Substrate
Period [h]
Product[a,b]
Isolated
E/Z[d]
yield [%][c]
give the corresponding cy
zation products in excellent
yields.
ACHTUNGTRENNUNGcli-
AHCTUNGTRENNUNG
Scheme 19 shows more de-
tails on the cyclization of 85.
The olefin–iron complex 87
formed from 85 and 3’ under-
went cyclization to generate 89,
the hydrolysis (or deuteriolysis)
of which, gave [D]86. Intercep-
tion of 87 at a low temperature
with DCl in D2O afforded the
saturated amide 88, which bore
two deuterium atoms, and thus,
verified the role of intermediate
dimetalated species 87. Interest-
ingly, the metalation of the ole-
finic bond of 85 with 3’ and the
subsequent intramolecular alky-
lation were faster than the iron-
catalyzed phenylation of the to-
sylate moiety,[3,27] because no
phenylated product was isolated
in this transformation. Other
examples of this cyclization are
listed in Schemes 20 and 21.
1
2
NR2 =NEt2
(63)
(71)
3
5
(66)
(76)
64
46
83:17
75:25
3
(72)
2
(77)
47
14:86
[e]
4
5
6
(73)
(74)
(75)
4
4
3
(78)
(79)
(80)
46
47
37
–
17:83
21:79
[a] Products were isolated after workup with aqueous HCl. [b] Structure of the major isomer is shown.
[c] Combined yields of E and Z isomers. [d] Olefinic stereochemistry was assigned by 1H NMR spectroscopic
analysis in all cases except for entry 2. [e] An exact E/Z ratio could not be determined because of the presence
of diastereoisomers.
Scheme 17.
Scheme 19. Alkylation of functionalized olefin with iron reagent 3’.
Scheme 20 exemplifies that the intermediate metalated spe-
cies generated from 90 (similar to 89 in Scheme 19) was suc-
cessfully alkylated to give the carbon-elongated product 93
in good yield.
Scheme 18.
Chem. Eur. J. 2011, 17, 14593 – 14602
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14599

H. Urabe et al.
Typical procedure for compounds included in Table 3: Compound 2:
tBuMgCl (0.392 mL, 1.02m solution in THF, 0.400 mmol) was added to a
solution of 1 (50.6 mg, 0.100 mmol) and FeCl2 (1.3 mg, 0.010 mmol) in
THF (1.0 mL) at ꢀ788C under argon. The solution was slowly warmed to
08C over 4.5 h and the reaction was terminated by addition of aqueous
HCl (1.0 mL, 1.00n). The organic layer was separated and the aqueous
layer was extracted with AcOEt. The combined organic layers were
washed with a saturated aqueous solution of NaHCO3 (2.0 mL), dried
1
over Na2SO4, and concentrated in vacuo to give a crude oil, the H NMR
spectroscopic analysis of which did not show the signals of another ole-
finic stereoisomer. The crude product was purified by chromatography
on silica gel (hexane/ethyl acetate as the eluent) to afford 2 as an oil
(41.7 mg, 82%).
Scheme 20.
Typical procedure for compounds in Table 4: An 83:17 E/Z mixture of
66: tBuMgCl (0.500 mL, 1.00m solution in THF, 0.500 mmol) was added
to
a solution of 63 (45.0 mg, 0.100 mmol) and FeCl2 (15.8 mg,
0.125 mmol) in THF (1.0 mL) at ꢀ788C under argon. The solution was
gradually warmed to ꢀ208C over 3 h, followed by termination of the re-
action by addition of aqueous HCl (2.0 mL, 1.00n). The organic layer
was separated and the aqueous layer was extracted with AcOEt. The
combined organic layers were washed with a saturated aqueous solution
of NaHCO3 (2.0 mL), dried over Na2SO4, and concentrated in vacuo to
give a crude oil, the 1H NMR spectroscopic analysis of which revealed
that the olefinic stereoselectivity was 83:17. The crude product was puri-
fied by chromatography on silica gel (hexane/ethyl acetate as the eluent)
to afford 66 as an oil of the same isomeric composition (28.9 mg, 64%).
Scheme 21.
Conclusion
We reported a series of effective metalative cyclizations of
functionalized enynes and dienes with a simple iron reagent,
3. The enyne cyclization could be made catalytic with re-
spect to iron. In addition, intramolecular alkylation of func-
tionalized acetylenes and olefins with an epoxide or tosylate
was achieved through metalation with analogous iron re-
agents. Further investigations into new synthetic applica-
tions of iron reagents are now in progress.
Typical synthetic procedure for compounds in Scheme 17: An 84:16 Z/E
mixture of 83: tBuMgCl (0.784 mL, 1.02m solution in THF, 0.800 mmol)
was added to a solution of 81 (59.2 mg, 0.100 mmol) and FeCl2 (25.4 mg,
0.200 mmol) in THF (1.0 mL) at ꢀ788C under argon. The solution was
slowly warmed to ꢀ208C over 5 h, followed by termination of the reac-
tion by addition of aqueous HCl (1 mL, 1.00n). The organic layer was
separated and the aqueous layer was extracted with AcOEt. The com-
bined organic layers were neutralized with a saturated aqueous solution
of NaHCO3 (2 mL), dried over Na2SO4, and concentrated in vacuo to
give a crude oil, the 1H NMR spectroscopic analysis of which revealed
that the Z/E ratio of the product was 60:40. The crude oil was purified
by chromatography on silica gel (hexane/ethyl acetate as the eluent) to
afford 83 as an oil (38.9 mg, 60%; 84:16 mixture of Z and E isomers).
Typical synthetic procedure for compounds in Scheme 19: Compound 86:
PhMgBr (0.730 mL, 1.09m solution in THF, 0.800 mmol) was added to a
solution of 85 (38.2 mg, 0.100 mmol) and FeCl2 (25.4 mg, 0.200 mmol) in
THF (1.0 mL) at ꢀ788C under argon. The solution was slowly warmed to
08C over 4 h, followed by termination of the reaction by addition of
aqueous HCl (1 mL, 1.00n). The organic layer was separated and the
aqueous layer was extracted with AcOEt. The combined organic layers
were neutralized with a saturated aqueous solution of NaHCO3 (2 mL),
dried over Na2SO4, and concentrated in vacuo to give a crude oil, which
was purified by chromatography on silica gel (hexane/ethyl acetate as the
eluent) to afford 86 as an oil (20.2 mg, 95%).
Experimental Section
Typical procedure for compounds included in Table 1: Compound 2:
tBuMgCl (0.500 mL, 1.00m solution in THF, 0.500 mmol) was added to a
solution of 1 (50.6 mg, 0.100 mmol) and FeCl2 (15.8 mg, 0.125 mmol) in
THF (1.0 mL) at ꢀ788C under argon. The resulting solution was slowly
warmed to ꢀ208C over 3.5 h, followed by termination of the reaction by
the addition of aqueous HCl (1.0 mL, 1.00n). The organic layer was sepa-
rated and the aqueous layer was extracted with AcOEt. The combined
organic layers were washed with
a saturated aqueous solution of
NaHCO3 (2.0 mL), dried over Na2SO4, and concentrated in vacuo to give
a crude oil, the 1H NMR spectroscopic analysis of which did not show
signals of the olefinic stereoisomer. The crude product was purified by
chromatography on silica gel (hexane/ethyl acetate as the eluent) to
afford 2 as an oil (42.3 mg, 83%).
Products 2, 45b, 66, 83, and 86 were fully characterized by 1H and
13C NMR spectroscopy, IR spectroscopy, and elemental analysis.
Typical procedure for compounds included in Table 2: A 1:1 mixture of
keto and enol forms of 45b from 39b: tBuMgCl (0.495 mL, 1.01m solu-
tion in THF, 0.500 mmol) was added to a solution of 39b (24.0 mg,
0.100 mmol) and FeCl2 (15.8 mg, 0.125 mmol) in THF (1.0 mL) at ꢀ788C
under argon. The solution was warmed to ꢀ208C over 1 h, 2-butanol
(0.012 mL, 0.130 mmol) was added, and the reaction mixture was
warmed to room temperature over 1 h. The reaction was terminated by
addition of aqueous HCl (3 mL, 1.00n). The organic layer was separated
and the aqueous layer was extracted with AcOEt. The combined organic
layers were washed successively with a saturated aqueous solution of
NaHCO3 and brine, dried over Na2SO4, and concentrated in vacuo to
give a crude oil, which was purified by chromatography on silica gel
(hexane/ethyl acetate as the eluent) to afford 45b as an oil (12.0 mg,
61%).
Acknowledgements
This work was supported by a Grant-in-Aid for Scientific Research (B)
(21350027) from JSPS,
a Grant-in-Aid for Young Scientists (B)
(20750071 to T.H.) from MEXT, and a Grant-in-Aid for Young Scientists
(B) (23750102 to T.H.) from JSPS.
[1] Group 4 metal (Ti, Zr)-mediated reactions are representative of
these stoichiometric transformations among many metal-promoted
cyclizations; a) Titanium and Zirconium in Organic Synthesis (Ed.:
I. Marek), Wiley-VCH, Weinheim, 2002; b) Symposia-in-Print (Eds.:
K. Suzuki, P. Wipf): Tetrahedron 2004, 60, 1269–1424; c) Symposia-
14600
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14593 – 14602

Iron-Mediated and -Catalyzed Metalative Cyclization
FULL PAPER
in-Print (Ed.: E. Negishi): Tetrahedron 1995, 51, 4255–4570; for the
latest reviews on other metal-catalyzed cyclizations, see: d) Hand-
book of Cyclization Reactions, Vols. 1 and 2 (Ed.: S. Ma), Wiley-
VCH, Weinheim, 2010; e) V. Michelet, P. Y. Toullec, J.-P. GenÞt,
Angew. Chem. 2008, 120, 4338–4386; Angew. Chem. Int. Ed. 2008,
47, 4268–4315; f) C. Aubert, O. Buisine, M. Malacria, Chem. Rev.
2002, 102, 813–834.
Kauffmann, Angew. Chem. 1996, 108, 401–418; Angew. Chem. Int.
Ed. Engl. 1996, 35, 386–403.
[7] Even with the metal species listed in Ref. [2], the cyclization of bis-
functionalized enynes, such as those in Tables 1 and 3, has not been
reported.
[8] Other recently reported iron-mediated cyclizations of (unfunctiona-
lized)dienes, enynes, and diynes are as follows: for [2+2] cyclization
and polymerization of dienes, see: a) M. W. Bouwkamp, A. C.
Bowman, E. Lobkovsky, P. J. Chirik, J. Am. Chem. Soc. 2006, 128,
13340–13341; b) D. Takeuchi, R. Matsuura, S. Park, K. Osakada, J.
Am. Chem. Soc. 2007, 129, 7002–7003; for a few types of enyne cy-
[2] Hereafter, the term “functionalized” refers to “electron-withdraw-
ing-group-substituted” and particularly to “ester- and amide-substi-
tuted”. Synthetically useful functionalized metallacycles are known;
for titanacycles, see: a) A. Wolan, Y. Six, Tetrahedron 2010, 66,
3097–3133; b) F. Sato, H. Urabe in Titanium and Zirconium in Or-
ganic Synthesis (Ed.: I. Marek), Wiley-VCH, Weinheim, 2002;
pp. 319–354; c) F. Sato, S. Okamoto, Adv. Synth. Catal. 2001, 343,
759–784; d) F. Sato, H. Urabe, S. Okamoto, Chem. Rev. 2000, 100,
2835–2886; e) H. Urabe, K. Suzuki, F. Sato, J. Am. Chem. Soc.
1997, 119, 10014–10027; f) R. Tanaka, A. Yuza, Y. Watai, D. Suzuki,
Y. Takayama, F. Sato, H. Urabe, J. Am. Chem. Soc. 2005, 127, 7774–
7780; for nickelacycles, see: g) J. Montgomery, Acc. Chem. Res.
2000, 33, 467–473; h) J. Montgomery, Angew. Chem. 2004, 116,
3980–3998; Angew. Chem. Int. Ed. 2004, 43, 3890–3908; for tantala-
cycles, see: i) Y. Kataoka, J. Miyai, M. Tezuka, K. Takai, K. Utimo-
to, J. Org. Chem. 1992, 57, 6796–6802.
[3] For reviews on iron-mediated organic reactions, see: a) Iron Cataly-
sis in Organic Chemistry: Reactions and Applications (Ed.: B. Plietk-
er), Wiley-VCH, Weinheim, 2008; b) A. Correa, O. G. MancheÇo, C.
Bolm, Chem. Soc. Rev. 2008, 37, 1108–1117; c) E. B. Bauer, Curr.
Org. Chem. 2008, 12, 1341–1369; d) S. Enthaler, K. Junge, M.
Beller, Angew. Chem. 2008, 120, 3363–3367; Angew. Chem. Int. Ed.
2008, 47, 3317–3321; e) C. Bolm, J. Legros, J. Le Paih, L. Zani,
Chem. Rev. 2004, 104, 6217–6254; for enyne cyclization, see:
f) Ref. [1e]; for coupling reactions, see: g) B. Plietker, Synlett 2010,
2049–2058; h) B. Plietker, A. Dieskau, Eur. J. Org. Chem. 2009,
775–787; i) W. M. Czaplik, M. Mayer, J. Cvengros, A. J. von Wange-
lin, ChemSusChem 2009, 2, 396–417; j) B. D. Sherry, A. Fꢁrstner,
Acc. Chem. Res. 2008, 41, 1500–1511; k) A. Fꢁrstner, M. Rubꢂn,
Chem. Lett. 2005, 34, 624–629; for iron carbonyl complexes, see:
l) M. F. Semmelhack in Organometallics in Organic Synthesis. A
Manual, 2nd ed. (Ed.: M. Schlosser), Wiley, New York, 2002,
pp. 1006–1121; m) Comprehensive Organometallic Chemistry, Vol. 4
(Eds.: G. Wilkinson, F. G. A. Stone, E. W. Abel), Pergamon, Oxford,
1982, pp. 243–649.
[4] Portions of this section have been preliminarily communicated, see:
T. Hata, N. Hirone, S. Sujaku, K. Nakano, H. Urabe, Org. Lett.
2008, 10, 5031–5033.
[5] For our relevant work on iron-catalyzed reactions of functionalized
unsaturated compounds and Grignard reagents, see: a) K. Fukuhara,
H. Urabe, Tetrahedron Lett. 2005, 46, 603–606; b) S. Okada, K. Ara-
yama, R. Murayama, T. Ishizuka, K. Hara, N. Hirone, T. Hata, H.
Urabe, Angew. Chem. 2008, 120, 6966–6970; Angew. Chem. Int. Ed.
2008, 47, 6860–6864; c) T. Hata, R. Bannai, M. Otsuki, H. Urabe,
Org. Lett. 2010, 12, 1012–1014.
[6] For some reactions using an iron catalyst and tert-butyl Grignard re-
agents, see: a) R. S. Smith, J. K. Kochi, J. Org. Chem. 1976, 41, 502–
509; b) E. C. Ashby, T. L. Wiesemann, J. Am. Chem. Soc. 1978, 100,
189–199; c) G. Cahiez, H. Avedissian, Synthesis 1998, 1199–1205;
d) H. Guo, K. Kanno, T. Takahashi, Chem. Lett. 2004, 33, 1356–
1357; e) L. K. Ottesen, F. Ek, R. Olsson, Org. Lett. 2006, 8, 1771–
1773; f) B. D. Sherry, A. Fꢁrstner, Chem. Commun. 2009, 7116–
7118; it has been reported that FeCl2 and 4 equiv of RCH2CH2MgX
AHCTUNGTREUNGcNN liACHTNUTRGEGzNNUN aACTHNUGTRENtNUNG ions, see: c) A. J. Pearson, R. A. Dubbert, Organometallics
1994, 13, 1656–1661; d) C. Nieto-Oberhuber, M. P. MuÇoz, S.
Lꢃpez, E. Jimꢂnez-NfflÇez, C. Nevado, E. Herrero-Gꢃmez, M. Radu-
can, A. M. Echavarren, Chem. Eur. J. 2006, 12, 1677–1693; for
tandem carbometalation of diynes, see: e) D. Zhang, J. M. Ready, J.
Am. Chem. Soc. 2006, 128, 15050–15051; for the Reppe-type alkyne
trimerization, see: f) C. Breschi, L. Piparo, P. Pertici, A. M. Caporus-
so, G. Vitulli, J. Organomet. Chem. 2000, 607, 57–63.
[9] Iron-catalyzed Diels–Alder and Alder–ene reactions are also
known, see: a) J. P. Genet, J. Ficini, Tetrahedron Lett. 1979, 20,
1499–1502; b) H. tom Dieck, R. Diercks, Angew. Chem. 1983, 95,
801–802; Angew. Chem. Int. Ed. Engl. 1983, 22, 778–779; c) K.-U.
Baldenius, H. tom Dieck, W. A. Kçnig, D. Icheln, T. Runge, Angew.
Chem. 1992, 104, 338–340; Angew. Chem. Int. Ed. Engl. 1992, 31,
305–307; d) J. M. Takacs, L. G. Anderson, G. V. B. Madhavan, F. L.
Seely, Angew. Chem. 1987, 99, 1073–1075; Angew. Chem. Int. Ed.
Engl. 1987, 26, 1013–1015; e) J. M. Takacs, L. G. Anderson, P. W.
Newsome, J. Am. Chem. Soc. 1987, 109, 2542–2544; f) A. Fꢁrstner,
K. Majima, R. Martin, H. Krause, E. Kattnig, R. Goddard, C. W.
Lehmann, J. Am. Chem. Soc. 2008, 130, 1992–2004; g) B. Moreau,
J. Y. Wu, T. Ritter, Org. Lett. 2009, 11, 337–339. See also: h) J. R.
Cabrero-Antonino, A. Leyva-Pꢂrez, A. Corma, Adv. Synth. Catal.
2010, 352, 1571–1576.
[10] A metalated carbon a to a carbonyl group is in equilibrium with its
enolate form, but the latter is not drawn for simplicity throughout
this article.
[11] Two diastereoisomers out of four were exclusively formed for 30
and 31. These diastereoisomers should be attributable to the stereo-
chemistry between the cyclopentane carbon and the carbon a to the
ester group, as also observed for products 28, 29, and 32. Assign-
ment of stereochemistry to major and minor isomers has not been
carried out.
[12] Ketoesters in Scheme 6, Table 2, and Scheme 7 exist as approximate-
ly a 1:1 mixture of keto and enol forms, of which only the former is
depicted for simplicity.
[13] Bicyclic ketoester 35 is a known key compound to pentalenolac-
tone F, see: D. E. Cane, P. J. Thomas, J. Am. Chem. Soc. 1984, 106,
5295–5303.
[14] For reviews on the preparation and synthetic utility of five-mem-
bered bicyclic compounds, see: a) G. Mehta, A. Srikrishna, Chem.
Rev. 1997, 97, 671–719; b) V. Singh, B. Thomas, Tetrahedron 1998,
54, 3647–3692.
[15] The dimerization of cinnamate esters to ketoesters has been report-
ed, see: a) N. Kise, S. Iitaka, K. Iwasaki, N. Ueda, J. Org. Chem.
2002, 67, 8305–8315 and references cited therein; b) K. Takaki, F.
Beppu, S. Tanaka, Y. Tsubaki, T. Jintoku, Y. Fujiwara, J. Chem. Soc.
Chem. Commun. 1990, 516–517; c) S. R. Jensen, A.-M. Kristiansen,
J. Munch-Petersen, Acta Chem. Scand. 1970, 24, 2641–2647; for in-
tramolecular cyclizations of a,b-olefinic esters, see: d) I. Shinohara,
M. Okue, Y. Yamada, H. Nagaoka, Tetrahedron Lett. 2003, 44,
4649–4652; for a related coupling of chalcones with titanocene di-
carbonyl, see: e) R. Schobert, F. Maaref, S. Dꢁrr, Synlett 1995, 83–
84; however, these methods did not result in the concomitant incor-
poration of a second constituent, such as a carbonyl compound, to
give products 50–53 (Scheme 7).
form a species of the formal composition of [FeACTHNUGRTNEUNG(MgX)2]n; g) B. Bog-
´
danovic, M. Schwickardi, Angew. Chem. 2000, 112, 4788–4790;
Angew. Chem. Int. Ed. 2000, 39, 4610–4612; h) A. Fꢁrstner, A. Leit-
ner, M. Mendez, H. Krause, J. Am. Chem. Soc. 2002, 124, 13856–
13863; i) A. Fꢁrstner, H. Krause, C. W. Lehmann, Angew. Chem.
2006, 118, 454–458; Angew. Chem. Int. Ed. 2006, 45, 440–444 and
references cited therein; some ate complexes of iron with tBuMgCl
are also possible for the iron intermediates in Scheme 10, see: j) T.
[16] Our attempts to execute the diene cyclization (Scheme 6), the cycli-
zations with epoxides (Scheme 14), and tosylates (Scheme 17) with
Chem. Eur. J. 2011, 17, 14593 – 14602
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14601

H. Urabe et al.
catalytic amounts of the iron reagent have so far proven unsuccess-
ful.
converted to carbanions by halogen/metal exchange and can be uti-
lized along the traditional reactions given in Scheme 11 rather than
that in Scheme 12.
[17] For the generation and reactions of dianions of ketoesters, see:
C. M. Thompson, D. L. C. Green, Tetrahedron 1991, 47, 4223–4285.
[18] This was what we actually experienced in the titanium alkoxide-
based metalative cyclizations of enynes, where attempted catalytic
reactions with respect to titanium were always unsuccessful
(Ref. [2a–d]).
[19] When the catalytic cyclization of 1 (Table 3, entry 1) was attempted
with 2 equiv of tBuMgCl, the desired product 2 was not obtained.
The similar tendency is also seen in Scheme 2 and Scheme 13.
[20] For reviews on three-component coupling reactions based on conju-
gate addition of nucleophiles, see: a) L. F. Tietze, G. Brasche, K. M.
Gericke, Domino Reactions in Organic Synthesis, Wiley-VCH, Wein-
heim, 2006, pp. 48–218; b) H.-C. Guo, J.-A. Ma, Angew. Chem.
2006, 118, 362–375; Angew. Chem. Int. Ed. 2006, 45, 354–366;
c) M. J. Chapdelaine, M. Hulce in Organic Reactions, Vol. 38 (Ed.:
L. A. Paquette), Wiley, New York, 1990, pp. 225–653.
[21] This transformation could be executed by the metalation of unsatu-
rated bond and a suitable electrophilic trap. Despite their limited
number, such examples can be found for unfunctionalized olefins
and acetylenes (see the Ref. below), but the cyclization of function-
alized acetylenes or olefins shown in Scheme 12 has no precedent.
For carbonyl compounds, see: a) Ref. [2a–d]; for halides, see: b) N.
Suzuki, D. Y. Kondakov, M. Kageyama, M. Kotora, R. Hara, T. Ta-
kahashi, Tetrahedron 1995, 51, 4519–4540; for tosylates, see:
c) R. R. Cesati, III, J. de Armas, A. H. Hoveyda, Org. Lett. 2002, 4,
395–398.
[24] For ring-opening reactions of epoxides, see: Aziridines and Epoxides
in Organic Synthesis (Ed.: A. K. Yudin), Wiley-VCH, Weinheim,
2006.
[25] a) C. Molinaro, T. F. Jamison, J. Am. Chem. Soc. 2003, 125, 8076–
8077; for its synthetic application, see: b) J. D. Trenkle, T. F. Jamison,
Angew. Chem. 2009, 121, 5470–5472; Angew. Chem. Int. Ed. 2009,
48, 5366–5368. In this nickel-catalyzed reductive coupling of acety-
lene and epoxide, functionalized acetylenes were not reported as the
substrate; and the reaction shown in Scheme 14 did not proceed
with a low-valent nickel catalyst in our laboratory. Thus, the success-
ful ring-opening reaction of the epoxide with a functionalized acety-
lene in this study may stem from the mechanistic differences be-
tween the Ni- and Fe-mediated reactions.
[26] For Baldwinꢄs rule, see: a) J. E. Baldwin, J. Chem. Soc. Chem.
Commun. 1976, 734–736. The epoxide moiety is claimed to behave
in an intermediate sense between the Tet and Trig categories. As 6-
endo-Tet is disfavored and other possible 6-endo-Trig, 5-exo-Tet, and
5-exo-Trig are favored, the alternative exo opening of the epoxide
moiety, which was not seen in the reaction shown in Scheme 14,
seems more common; b) R. J. Madhushaw, C.-L. Li, K.-H. Shen, C.-
C. Hu, R.-S. Liu, J. Am. Chem. Soc. 2001, 123, 7427–7428. Analo-
gously, the reaction of 75!80 (Table 4, entry 6) is anomalous be-
cause both 5-endo-Tet or -Trig modes are disfavored, whereas anoth-
er set of 4-exo-Tet and -Trig is favored by this rule.
[27] H. Urabe, F. Sato in Handbook of Grignard Reactions (Eds.: G. S.
Silverman, P. E. Rakita), Marcel Dekker, New York, 1996, pp. 577–
632.
[22] Acetylene–iron carbonyl complexes are known, but they react with
nucleophiles rather than electrophiles: D. L. Reger, E. Mintz, L.
Lebioda, J. Am. Chem. Soc. 1986, 108, 1940–1949; see also
Ref. [3m].
[23] We chose epoxides and tosylates as the electrophiles for this study
because other typical electrophiles, such as alkyl halides, are readily
Received: April 27, 2011
Revised: August 10, 2011
Published online: November 23, 2011
14602
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14593 – 14602