Radical aromatic substitution with benzene chromiumtricarbonyl

Article abstract of DOI:10.1016/j.tetlet.2007.08.084

Radical aromatic substitution of tert-butyl groups for hydrogen on simple arene chromiumtricarbonyl complexes was accomplished. Preexisting tert-butyl groups and methoxy groups directed incoming radicals primarily to the meta-position, and methoxy groups diminished the rate of substitution by roughly tenfold.

Full text of DOI:10.1016/j.tetlet.2007.08.084

Tetrahedron Letters 48 (2007) 7903–7905
Radical aromatic substitution with benzene chromiumtricarbonyl
*
Jeffrey H. Byers, Nathan R. Neale, J. Bradford Alexander and Stephen P. Gangemi
Department of Chemistry and Biochemistry, Middlebury College, Middlebury, VT 05753, United States
Received 24 July 2007; accepted 23 August 2007
Available online 29 August 2007
Abstract—Radical aromatic substitution of tert-butyl groups for hydrogen on simple arene chromiumtricarbonyl complexes was
accomplished. Preexisting tert-butyl groups and methoxy groups directed incoming radicals primarily to the meta-position, and
methoxy groups diminished the rate of substitution by roughly tenfold.
Ó 2007 Elsevier Ltd. All rights reserved.
Arene chromiumtricarbonyl complexes have been used
extensively as electrophiles in organic syntheses for
many years. More recently, several groups have demon-
esting reagent for a study of this type. With this goal
in mind, we considered the addition of nucleophilic alkyl
radicals to benzene chromiumtricarbonyl, following our
previously established photolytic protocol for radical
aromatic substitution. Atom transfer reactions of simple
alkyl halides are not particularly common. Successful
photolytic atom-transfer additions of tert-butyl iodide
and iso-propyl iodide to electron-deficient olefins have
1
strated that arene chromiumtricarbonyl complexes are
also susceptible to attack by nucleophilic radicals. Sch-
2
3
malz and Li have demonstrated SmI -promoted ketyl
2
addition to arene chromiumtricarbonyl complexes,
which only rearomatized with the loss of previously
incorporated leaving groups, such as methoxy, fluoride,
8
been reported. Atom transfer addition of 1°, 2°, and
4
or chloride. Merlic and Houk demonstrated inter-
3° alkyl iodides to CO have been demonstrated, leading
molecular addition of the acetone–SmI ketyl, forming
to acyl iodides, which in turn were intercepted with alco-
2
9
mixtures of isomeric cyclohexadienes. They found a
hol solvents, leading to generation of esters.
1
00,000:1 selectivity for radical attack at benzene chro-
miumtricarbonyl over uncomplexed benzene, indicating
significant rate acceleration resulting from metal com-
plexation of the arene. Our group, and others, have
demonstrated the generation and trapping of radicals
alpha to arene chromiumtricarbonyl complexes.
The specific conditions utilized in this study, the details
of which have been described in greater detail in previ-
ous work from our laboratory, will be briefly summa-
5
6
7
rized herein: suitable alkyl halides and olefins (in this
case, arene–chromium complexes) were photolyzed in
a CH Cl solution to which propylene oxide was added
2
2
8
In a series of previous publications we described novel
synthetic methodology whereby radical aromatic sub-
stitution could be accomplished by a process involving
iodine and bromine-transfer radical addition of electro-
philic radical precursors to electron-rich heteroaromat-
ics, accompanied by spontaneous rearomatization
in excess to serve as an HI trap. Curran has shown that
a substoichiometric quantity of Bu SnSnBu is required
3
3
in I-transfer radical addition reactions in order to
consume I , a radical chain suppressant which is gener-
2
ated as a byproduct of these reactions. In the course of
7
our previous work, we found that the addition of
7
through loss of HI. We were curious to see whether this
Na S O as an I2 reductant in the presence of the
2
2
3
+
À
atom-transfer substitution process would prove useful in
a situation where the polarity of the radicals and olefins
was reversed, and the aforementioned precedent for
nucleophilic radical attack on arene–chromium com-
plexes made them an appropriate and potentially inter-
phase-transfer catalyst Bu N I to aid in thiosulfate
4
solubility, provided an effective alternative to the use
of distannanes.
The proposed atom-transfer radical addition and associ-
ated non-radical rearomatization step are shown in
Scheme 1. tert-Butyl iodide was chosen as a radical pre-
cursor for addition reactions with benzene chromiumtri-
carbonyl (1), as it was a source of highly nucleophilic
*
Corresponding author. Tel.: +1 802 443 5207; fax: +1 802 443
072; e-mail: byers@middlebury.edu
2
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.08.084

7904
J. H. Byers et al. / Tetrahedron Letters 48 (2007) 7903–7905
In a critical control experiment, when benzene lacking
chromium complexation was subjected to identical reac-
tion conditions, none of the alkylation products listed in
Table 1 were obtained. Benzene is known to be virtually
t
-Bu
+
t
-Bu
(
CO) Cr
(CO) Cr
12
3
3
1
unreactive with alkyl radicals, and is frequently used
as a solvent for reactions involving the generation and
propagation of alkyl radicals. Since there are no known
examples of arene chromiumtricarbonyl complexes
I
+
t-BuI
+
t-Bu
t-Bu
t-Bu
4
(
CO) Cr
(CO)
3
Cr
undergoing Friedel–Crafts alkylation, and benzene
3
was unreactive under these reaction conditions, we can
conclude that the observed products were indeed the
result of radical addition to the chromium complex.
Since uncomplexed benzene is unreactive to these condi-
tions, the varying degrees of multiple alkylation implied
that some of the reactive arene chromiumtricarbonyl
intermediates remained complexed to the metal until
the highest possible degree of substitution was complete.
In the proposed mechanism (Scheme 1), coordinatively
unsaturated 16e and 17e intermediates are generated,
and it is likely that decomplexation occurred most read-
ily at these intermediates, allowing for mono- and di-
substituted products. Eventually, the alkyl-substituted
complexes formed proved too labile under the reaction
conditions, with the end result that no complexed arenes
were detected at the conclusion of the reaction. Finally,
the fact that aromatic substitution products, rather than
addition products, were identified provided indirect
evidence that the key atom-transfer step of atomic
iodine to the Cr-bound pentadienyl radical, which is
necessary for both rearomatization as well as radical
chain propagation, was occurring as proposed.
I
-
HI
t-Bu
t-Bu
(
CO) Cr
(CO) Cr
3
3
2
Scheme 1. Proposed mechanism for atom-transfer radical aromatic
substitution with benzene chromiumtricarbonyl.
tertiary radicals, leading to enhanced addition rates to
arene chromiumtricarbonyl complexes.
À
À
When these reaction conditions10 were employed with a
.5:1 ratio of 1:t-BuI (Table 1, entry 1) and photolysis
1
for 12 h, a mixture of mono-, di-, and tri-substituted
products was obtained, with tert-butyl benzene as the
major product. Yields were determined by GC, with
product retention times determined by authentic, com-
mercially available samples, with dodecane as an inter-
nal standard. When the experiment was repeated with
an excess of t-BuI (1:10 1:t-BuI), the same products were
observed, but with 1,3,5-tri-tert-tertbutylbenzene as the
major product (Table 1, entry 2). No arene–chromium
complexes could be observed in or isolated from the
crude reaction mixture. When shorter reaction times
were employed, mixtures of complexed and uncom-
plexed arenes were obtained, complicating the analysis.
When isopropyl iodide was used as radical precursor
under similar reaction conditions no substitution prod-
ucts were observed.
The regiochemistry of nucleophilic additions to substi-
tuted arene–chromium complexes have been extensively
studied. Prediction of regiochemistry proved quite com-
plex, depending on sterics, LUMO coefficients, charge,
nucleophile strength, and rotamer conformations of
13
the arene–Cr(CO) bond. In our examples, the absence
3
of any ortho-disubstitution is not surprising due to steric
effects. meta-Substitution accounted for most of the
polyalkylated products observed, with meta:para ratios
It seemed probable that the polysubstituted products
formed from reaction of excess tert-butyl radical with
tert-butylbenzene chromiumtricarbonyl (2), generated
in situ. To test this premise, we subjected independently
(
4 + 6:5) of 6.5:1 (entries 1 and 3) and 7.8:1 (entry 2)
being observed. The LUMO coefficients of alkylbenz-
1
1
synthesized tert-butylbenzene chromiumtricarbonyl to
the reaction conditions (Table 1, entry 3), and found the
ratios of 1,3-, 1,4-, disubstituted and 1,3,5-trisubstituted
benzenes formed were comparable to those observed in
entry 2, adding support to this presumption. We were
also able to quantify decomplexed tert-butylbenzene,
allowing us to account for most (89%) of the arene-
derived products in the reaction.
OCH
3
t
-Bu
t
-Bu
10 eq
t-BuI
propylene oxide
+
-
H CO
3
3
Bu N I /Na S O
H
3
CO
t-Bu
(CO)
Cr
7 (1 eq)
4
2
2
3
CH Cl /hv
2
2
5
%
11%
Scheme 2. Addition to anisole chromiumtricarbonyl.
Table 1. Products and yields
Entry
Precursor (equiv)
t-Bul (equiv)
Yields (%)
t-Bu benzene
3)
1,3-Di-t-Bu
benzene (4)
1,4-Di-t-Bu
benzene (5)
1,3,5-Tri-t-Bu
benzene (6)
(
1
2
3
1 (1.5)
1 (1)
2 (1)
1
10
10
t-BuI
23
12
29
7
5
9
2
5
8
6
34
43
propylene oxide
+ -
4 2 2 3
Bu N I /Na S O
CH Cl /hv
2
2 h
2
1

J. H. Byers et al. / Tetrahedron Letters 48 (2007) 7903–7905
7905
OCH
3
t
-Bu
t
-Bu
1
eq
t-BuI
+
+
propylene oxide
Bu N I /Na S O
CH Cl /hv
+
-
OCH3
4
2
2
3
(
CO) Cr
(CO) Cr
3
3
1.3%
13%
2
2
7
(5 eq)
1 (5 eq)
Scheme 3. Competitive radical additions.
ene–Cr(CO) complexes favored ortho and meta addi-
S.; Schwarz, A. Tetrahedron Lett. 1996, 37, 2947; (c)
Schmalz, H.-G.; Kiehl, O.; Gotov, B. Synlett 2002, 1253;
(d) Hoffmann, O.; Schmalz, H.-G. Synlett 1998, 1426.
. (a) Lin, H.; Yang, L.; Li, C. Organometallics 2002, 21,
3
tion of nucleophiles, and polarization effects favored
1
3
para attack. The predominance of meta-substitution
seen in our work with tert-butyl radicals can be under-
stood in light of this model, by recognizing the neutral
nature of the radical, greatly diminishing the influence
of polarization effects. Predominate meta-direction by
3
3
2
848; (b) Lin, H.-C.; Yang, L.; Li, C.-Z. Chin. J. Chem.
003, 21, 797.
4
5
. Merlic, C. A.; Miller, M. M.; Hietbrink, B. N.; Houk, K.
N. J. Am. Chem. Soc. 2001, 123, 4904.
. Byers, J. H.; Janson, N. J. Org. Lett. 2006, 8, 3453.
methyl groups has also been observed in the radical sub-
3
stitution of SmI ketyls for halides.
6. (a) Merlic, C. A.; Walsh, J. C. Tetrahedron Lett. 1998, 39,
083; (b) Taniguchi, N.; Uemura, M. Tetrahedron Lett.
2
2
When anisole chromiumtricarbonyl (7) was submitted to
these reaction conditions, 3-tert-butyl and 3,5-di-tert-
butyl anisole were obtained (Scheme 2), accompanied
by significant quantities of unreacted anisole chromium-
tricarbonyl. Arene complexes bearing electron donors
are apparently more robust than those lacking such
functionality, allowing for recovery of some unreacted
starting material 7. In order to assess the electronic
influence of a methoxy substituent on the rate of radical
attack, we subjected an equimolar mixture of 7 and 1 to
limiting tert-butyl iodide, leading to the formation of a
1997, 38, 7199; (c) Creary, X.; Mehrsheikh-Mohammadi,
M. E.; McDonald, S. J. Org. Chem. 1989, 54, 2904.
. (a) Byers, J. H.; Campbell, J. E.; Knapp, F. H.; Thissell, J.
G. Tetrahedron Lett. 1999, 40, 2677–2680; (b) Byers, J. H.;
Duff, M. P.; Woo, G. W. Tetrahedron Lett. 2003, 44,
7
6
853–6855; (c) Byers, J. H.; DeWitt, A.; Nasveschuk, C.
G.; Swigor, J. E. Tetrahedron Lett. 2004, 45, 6587–
590.
6
8
. Curran, D. P.; Dooseop, K. Tetrahedron 1991, 47, 6171–
6188.
9. (a) Nagahara, K.; Ryu, I.; Komatsu, M.; Sonoda, N. J.
Am. Chem. Soc. 1997, 119, 5465; (b) Kreimerman, S.; Ryu,
I.; Minakate, S.; Komatsu, M. Org. Lett. 2000, 2, 389; (c)
Ryu, I.; Nasgahara, K.; Kambe, N.; Sonoda, N.; Kreim-
erman, S.; Komatsu, M. Chem. Commun. 1998, 1953.
0. Typical reaction conditions: In a screw-top Pyrex test
tube, the chromium complex 1 (0.270 g, 1 mmol), tert-
1
0:1 ratio of tert-butylbenzene to 3-tert-butylanisole
(
Scheme 3). Thus, we concluded that the rate of radical
attack on 1 occurs at 10 times the rate as that observed
with 7. This makes qualitative sense in light of the fact
that nucleophilic radicals are known to react more
rapidly with more electron-deficient alkenes, making
the deactivating effect of a methoxy substituent
reasonable.
1
butyl iodide (1.839 g, 10 mmol), propylene oxide (0.515 g,
+ À
1
0 mmol), Na S O (0.158 g, 1 mmol), and Bu N I
2 2 3 4
(
0.0369 g, 0.1 mmol) were combined with 3 mL of dichlo-
romethane. The reaction mixture was deoxygenated at
°C for 10 min and photolyzed with a 450-W medium
0
In summary, we have found that our methodology for
radical aromatic substitution can be used effectively with
benzene chromiumtricarbonyl and highly nucleophilic
tert-butyl radicals. The observed regioselectivity is rea-
sonable in light of the established literature dealing with
nucleophilic attack on arene chromiumtricarbonyl data.
pressure Hanovia lamp for 12 h. Analysis was performed
by GC/FID on an HP-5 capillary column. The identities
of all products were determined by comparison of
retention times with authentic samples. Response factors
and yields were determined relative to dodecane
(ꢀ100 mg) added as an internal standard prior to GC/
FID analysis (Hewlett-Packard 5890) gas chromatograph
for yield determination.
1
1. The procedure used to synthesize 2: Toma, S.; Hudecek,
M. J. Organomet. Chem. 1991, 406, 147; The sample of 2
provided spectra identical to those obtained for 2 previ-
ously generated through other methods: Djukic, J.-P.;
Rose-Munch, F.; Rose, E.; Simon, F.; Dromzee, Y.
Organometallics 1995, 14, 2027.
Acknowledgments
Financial support is acknowledged from the NSF
(
CHE-0315745). A summer research fellowship for
J.B.A. was provided by the Vermont Genetics Network.
1
2. Citterio, A.; Minisci, F.; Porta, O.; Sesana, G. J. Am.
Chem. Soc. 1977, 99, 7960.
1
3. (a) Solladie-Cavallo; Wipf, G. Tetrahedron Lett. 1980, 21,
3047; (b) Kundig, E. P.; Desobry, V.; Simmons, D. P.;
Wenger, E. J. Am. Chem. Soc. 1989, 111, 1804; (c)
Semmelhack, M. F.; Clark, G. R.; Farina, R.; Saeman, M.
J. Am. Chem. Soc. 1979, 101, 217; (d) Semmelhack, M. F.;
Garcia, J. L.; Cortes, D.; Farina, R.; Hong, R.; Carpenter,
B. K. Organometallics 1983, 2, 467.
References and notes
1
. Pape, A. R.; Kaliappan, K. P.; Kundig, E. P. Chem. Rev.
000, 100, 2917.
2
2
. (a) Schmalz, H.-G.; Siegel, S.; Bats, J. W. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2383; (b) Schmalz, H.-G.; Siegel,