Bromoform activation. TiCl4-Mg-promoted CHBr2 - and CBr3- transfer to a variety of aldehydes and ketones

Article abstract of DOI:10.1021/ol402861a

TiCl₄-Mg can mediate addition of CHBr₃ to a variety of aldehydes and ketones to form dibromomethyl carbinols and also be used to effect CBr₃ transfer to carbonyl groups to form tribromomethyl carbinols. The successful application of TiCl₄-Mg-promoted coupling of CHBr₃ with various carbonyl compounds, especially in the case of highly enolizable ketones such as 2-indanone and β-tetralone, highlights the extraordinary reactivity and selectivity and the weakly basic nature of this system.

Full text of DOI:10.1021/ol402861a

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Bromoform Activation. TiCl4À
À
À
Mg-Promoted CHBr2 and CBr3 Transfer
to a Variety of Aldehydes and Ketones
Tu-Hsin Yan,* Su-Haur Chang, Cheng-Ta Chang, Chia-Kuan Lin, and Chien-Yu Liu
Department of Chemistry, National Chung-Hsing University, Taichung 400, Taiwan,
Republic of China
thyan@mail.nchu.edu.tw
Received October 4, 2013
ABSTRACT
TiCl4ÀMg can mediate addition of CHBr3 to a variety of aldehydes and ketones to form dibromomethyl carbinols and also be used to effect CBr3
transfer to carbonyl groups to form tribromomethyl carbinols. The successful application of TiCl4ÀMg-promoted coupling of CHBr3 with various
carbonyl compounds, especially in the case of highly enolizable ketones such as 2-indanone and β-tetralone, highlights the extraordinary
reactivity and selectivity and the weakly basic nature of this system.
Polyhalomethyl carbinols are valuable building blocks
for further structural elaboration as well as interesting
substances in their own right. While the carbonyl addition
of polyhalomethyl anions generated in situ from a combi-
nation of polyhalomethane with strong bases such as
n-BuLi or lithium dicyclohexylamide at À78 °C and KOH
at 0 °C constitutes one of the fundamental approaches for
synthesis of polyhalomethyl carbinols,1 this process does
not generally extend to highly enolizable ketones and
aldehydes bearing R-protons, since complex mixtures re-
sult from extensive aldol side reactions.1b The other main
approaches to the synthesis of polyhalomethyl carbinols,
which avoid the use of strong base, involve an improve-
ment in polyhalomethyl anion generation, which is derived
from either the desilylation of polyhalomethylsilanes2
mediated by fluoride ion or the decarboxylation of sodium
trihaloacetate promoted by trihaloacetatic acid.3 However,
these two mild and efficient approaches to polyhalomethyl
carbinols do not generally extend to ketones. The significant
limitation on the general utility of polyhalocarbanion species
arises from their low chemoselectivity and high thermolability.
Compared with the polyhalomethyllithium and sodium
reagents, the higher thermostability and selectivity of poly-
halomethyltitanium compounds make their easy availability
by synthesis important.4 By far, either dihalomethyl-
or trihalomethyltitanium compounds have been gen-
erated by adding the titanium derivatives to the ex-
tremely thermolabile polyhalomethyllithium compounds
at very low temperatures (eÀ70 °C).4 Can the dibromo-
methyltitanium and tribromomethyltitanium reagents be
generated directly via oxidative addition of the CÀBr bond
to TiCl4ÀMg bimetallics? Earlier work in our laboratories
established the feasibility of CHnXmÀTiCl4ÀMg system as
highly nucleophilic methylene and halomethylene carbe-
noids to effect carbonyl-olefination.5 In searching for new
strategies based upon the concept of simple oxidative
(4) Kauffmann, T.; Fobker, R.; Wensing, M. Angew. Chem., Int. Ed.
Engl. 1988, 27, 943.
(5) (a) Yan, T.-H.; Tsai, C.-C.; Chien, C.-T.; Cho, C.-C.; Huang,
P.-C. Org. Lett. 2004, 6, 4961. (b) Yan, T.-H.; Chien, C.-T.; Tsai, C.-C.;
Lin, K.-W.; Wu, Y.-H. Org. Lett. 2004, 6, 4965. (c) Lin, K.-W.; Yan, S.;
Hsieh, I.-L.; Yan, T.-H. Org. Lett. 2006, 8, 2265. (d) Tsai, C.-C.; Hsieh,
I.-L.; Cheng, T.-T.; Tsai, P.-K.; Lin, K.-W.; Yan, T.-H. Org. Lett. 2006,
8, 2261. (e) Lin, K.-W.; Tsai, C.-H.; Hsieh, I.-L.; Yan, T.-H. Org. Lett.
2008, 10, 1927. (f) Bhorge, Y. R.; Chang, C.-T.; Chang, S.-H.; Yan,
T.-H. Eur, J. Org. Chem. 2012, 4805. (g) Bhorge, Y. R.; Chang, S.-H.;
Chang, C.-T.; Yan, T.-H. Tetrahedron 2012, 68, 4846.
(1) (a) Taguchi, H.; Yamamoto, H.; Nozaki, H. J. Am. Chem. Soc.
1974, 96, 3010. (b) Wyvratt, J. M.; ZHazen, G. G.; Weinstock, L. M.
J. Org. Chem. 1987, 52, 944.
(2) Fujita, M.; Hiyama, T. J. Am. Chem. Soc. 1985, 107, 4085.
(3) (a) Corey, E. J.; Link, J. O.; Shao, Y. Tetrahedron Lett. 1992, 33,
3435. (b) Corey, E. J.; Link, J. O. J. Am. Chem. Soc. 1992, 114, 1906.
r
10.1021/ol402861a
XXXX American Chemical Society

addition, we turned our attention to the concept of a
polyhalomethyltitanium synthesisbasedupon a titaniumÀ
magnesium bimetallic-mediated oxidative addition of the
halides. Effecting such elaborations by use of TiCl4ÀMg
may have the advantages of (1) extending the addition to
various carbonyl compounds such as aldehydes and en-
olizable ketones and (2) enhancing synthetic efficiency by
elaboration of bromoform to both dibromomethyl- and
tribromomethyltitanium reagents. Herein we record pro-
tocols whereby the TiCl4ÀMg-promoted direct coupling
of bromoform with carbonyl compounds can be directed
not only to dibromomethyl carbinols, which served as the
precursors of terminal alkynes6a and acetylenic alcohols6b,c
as well as important intermediates for ketone homologa-
tion and ring expansion,7 but also to tribromomethylcar-
binols, which are important building blocks for synthesis
of conjugated ketones8a and R- and β-bromoenol ethers.8b
Initial studies centered on the TiCl4ÀMg-promoted
CHBr2-transfer reactions. The direct coupling of2-methyl-
cyclohexanone 1a with CHBr3 promoted by TiCl4ÀMg
was chosen to test the feasibility of the process (Table 1).
When 1a (1.0 mmol) and CHBr3 (0.3 mL, 3.3 equiv) are
treated with TiCl4 (1 equiv) and magnesium powder
(5 equiv) in ClCH2CH2ClÀDME (1 mL) at 0À25 °C,
1-(dibromomethyl)-2-methylcyclohexanol 1b was indeed
produced but only in less than 30% conversions after 3 h
(entry 1). The desired dibromomethylcarbinol 1b was
found to be admixed with a trace of vinyl bromide 1d.
Increasing the amount of Mg dramatically improved the
dibromomethylcarbinol formation, with the yield varying
from 35% to 50% (entries 2 and 3). Most revealing was the
effect of the electron-pair-donor (EPD) additives on this
process. Replacing DME with THF proves most satisfac-
tory, giving an 85% yield of adduct 1b (entry 4). A NOE
difference experiment indicates a diequatorial orientation
of dibromomethyl and methyl substituents in 1b, substan-
tiating a highly stereocontrolled HCBr2 transfer. More
gratifyingly, the reaction directly scales up; thus, dibromo-
methylcarbinol 1b was obtained in 75% yield on a
10-mmol scale using 6 mmol of TiCl4 and 55 mmol of Mg.
It seemed reasonable to expect that the direct coupling of
CÀBr bond with a presumed TiCl4ÀMg complex accounts
for the generation of an active diibromomethyltitanium
species and the formation of 1-(dibromomethyl)-2-methyl-
cyclohexanol 1b. Notably, addition to a substituted cyclo-
hexanone 1a takes place with complete stereoselectivity.
Assuming that steric hindrance to attack is controlling, the
stereochemistry possessing the methyl group trans to the
CBr3 unit is assigned. Having established the feasibility of
Table 1. Reaction Conditions for the TiCl4ÀMg-Promoted
CHBr2 Transfer to Ketone 1a
entry
TiCl4/Mg (equiv)
EPD additive
yielda (%) 1b
1
2
3
4
5
1:5
DME
DME
DME
THF
THF
22b,c
35b,c
50b,c
87b,c
75b,c
1:8
1:10
1:10
6:55d
a Isolated yield. b Plus ∼50% recovered starting ketone. c Plus 5À7%
vinyl bromide 1d. d The reaction was performed on a 10 mmol scale.
the HCBr2-transfer reaction, its generality with respect to
the structure of the ketone was established. Reaction of
cyclopentanone and cycloheptanone with CHBr3 under
the standard conditions gave the desired adducts 2b (85%)
and 3b (82%), respectively (Table 2, entries 1 and 2).
Changing the ketone to acyclic ketones 4a and 5a led to
equally gratifying results (entries 3 and 4) with formation
of dibromomethyl alcohols 4b and 5b. The CHBr2 transfer
onto the unsaturated ketone 6a was equally effective by
replacing THF with DME (entry 5). The aromatic ketones
7a and 8a also gave satisfactory results with CHBr3ÀMgÀ
TiCl4ÀTHF (entries 6 and 7). On the other hand, using the
readily enolizable ketones9 also gave satisfactory results.
Thus, either 2-indanone 9a or β-tetralone 10a reacted
efficiently with CHBr3-derived dibromomethyltitanium
reagent to give the desired coupling products 9b (71%)
and 10b (62%), respectively (entries 8 and 9). The success-
ful application of TiCl4ÀMg-promoted CHBr2 transfer to
the readily enolizable 2-indanone and β-tetralone high-
lights the weakly basic nature of this system. More gratify-
ingly, increasing the degree of steric hindrance at the
carbonyl group does not impede coupling reaction. Thus,
reacting 2,6-dimethylcyclohexanone 11a and camphor-
derived ketoester 12a with CHBr3ÀMgÀTiCl4ÀTHF
produced the adducts 11b and 12b, respectively (entries
10 and 11). Notably, the ester group was completely un-
affected. Aldehydes, which are particularly prone to re-
duction to give alcohols, can also lead to equally gratifying
results by replacing THF with DME. Thus, in the reaction
ofaliphatic aldehydes(1 mmol) with CHBr3, use of 1 mmol
of TiCl4, 10 mmol of Mg, and 1 mL of DME also led to
smooth coupling, giving the desired adduct in 73À75%
yield (entries 12 and 13). The unsaturated aldehyde gave an
analogous result (entry 14). Exposing the aromatic alde-
hydes to the CHBr3ÀTiCl4ÀMgÀDME system also gave
(6) (a) Wang, Z.; Yin, J.; Campagna, S.; Pesti, J. A.; Fortunak, J. M.
J. Org. Chem. 1999, 64, 6918. (b) Smith, A. B., III; Adams, C. M.;
Kozmin, S. A.; Paone, D. V. J. Am. Chem. Soc. 2001, 123, 5925. (c)
Kowalski, C. J.; Haque, M. S.; Fields, K. W. J. Am. Chem. Soc. 1985,
107, 1429.
(7) Ward, H. D.; Teager, D. S.; Murray, R. K., Jr. J. Org. Chem.
1992, 57, 1926.
(8) (a) Falck, J. R.; He, A.; Reddy, L. M.; Kundu, A.; Barma, D. K.;
Bandtopadhyay, A.; Kamila, S.; Akella, R.; Bejot, R.; Mioskowski, C.
Org. Lett. 2006, 8, 4645. (b) Bejot, R.; Tisserand, S.; Reddy, L. M.;
Barma, D. K.; Baati, R.; Falck, J. R.; Mioskowski, C. Angew. Chem.,
Int. Ed. Engl. 2005, 44, 2008.
(9) (a) Johnson, C. R.; Tait, B. D. J. Org. Chem. 1987, 52, 281. (b)
Acetoacetic ester fails to react with CHBr3-derived dibromomethyltitanium
reagent to give the desired coupling product. (c) This TiCl4ÀMgÀCHBr3 is
also suitable for the CHBr2 transfer to 2-chlorocyclohexanone; see the
Supporting Information.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 2. CHBr3ÀTiCl4ÀMg-Mediated CHBr2-Transfer to
Various Ketone and Aldehyde Carbonyl Groups at 0À25 °C
Scheme 1. Selectivity in CHBr3ÀTiCl4ÀMg-Mediated
CHBr2-Transfer Reaction
Table 3. CHBr3ÀTiCl4ÀMgÀTHF-Mediated CBr3-Transfer to
Various Ketone and Aldehyde Carbonyl Groups at À30 °C
a Isolated yield. b Plus 5À7% bromomethylenation product. c Reaction
was performed by replacing THF with DME. c Assuming steric hindrance
to attack is controlling.
satisfactory results (entry 15). The chemoselectivity was
briefly explored with a series of aldehydes and ketones as
summarized in Scheme 1. As expected, exposure of a mix-
ture of aliphatic aldehyde and ketone to a 0.3 equiv of
TiCl4 and 3 equiv of Mg in ClCH2CH2Cl/DME at 0 °C
produced exclusively the aldehyde-derived adduct. The
aromatic aldehyde gavean analogous resultin the presence
of aliphatic ketone. Thus, using a 0.5:6 TiCl4/Mg ratio
a Isolated yield. b Reaction was performed on a 25 mmol scale.
c Reaction was performed by adding THF and DME.
Org. Lett., Vol. XX, No. XX, XXXX
C

proves most satisfactory, giving only a 70% yield of
benzaldehyde-derived dibromomethyl alcohol 16b. The
preference for coupling of aldehydes with CHBr3 pto-
moted by TiCl4ÀMg highlihgts the chemoselectivity of
the CHBr2-transfer process.
Scheme 2. TiCl4ÀMgÀBimetallic Complex-Promoded
CBr3-Transfer Reaction of CHBr3
Having established that CHBr3ÀMgÀTiCl4 showed
promise as a generally useful CHBr2-transfer reagent,
we envisioned the feasibility of the TiCl4ÀMg-promoted
CBr3 transfer reactions. Initial studies centered on a simple
2-methylcyclohexanone 1a. Gratifyingly, exposing 1a to
the CHBr3ÀMgÀTiCl4ÀTHF system at À30 °C led to
smooth CBr3 transfer to give the desired adduct (Table 3,
entry 1). Thus, using a 0.5:6 TiCl4/Mg ratio proves most
satisfactory, giving a 89% yield of the desired alcohol 1c in
addition to a trace (<5%) of dibromomethyl alcohol
adduct. Most delightfully, the reaction directly scales up;
thus, tribromomethylalcohol1cwas obtainedin83%yield
on a 25-mmol scale using a 5 equiv of TiCl4 and 60 equiv of
Mg. Extension of this TiCl4ÀMg-promoted CBr3 transfer
to other traps confirms their generality. Fortunately, the
unsaturated ketones also gave satisfactory results with
CHBr3ÀMgÀTiCl4ÀTHF system (entries 2À4). Not sur-
prisingly, in contrast to polyhalomethyllithium and mag-
nesium reagents, this CHBr3ÀMgÀTiCl4ÀTHF system
reacted efficiently even with the readily enolizable ketones.
Thus, reacting 2-indanone and β-tetralone with CHBr3
under the standard conditions produced the desired tri-
bromomethyl alcohol adducts 9c (87%) and 10c (85%),
respectively (entries 5 and 6). On the other hand, the
TiÀMgÀbimetallic complex promoted CBr3 transfer onto
the aldehydes was equally effective (entries 7À11). More
dramatically, this TiCl4ÀMg-promoted CBr3 transfer to
chloro ketone containing an active methine hydrogen and
chloro group was equally effective (entry 12).
complex on the relative amount of TiCl4 to Mg and THF.
Further mechanistic work is clearly required before any
definite conclusions can be reached.
The ability to elaborate CHBr3 into unusual dibromo-
methyl- and tribromomethyltitanium reagents via TiCl4ÀMg
bimetallic complex mediated transformation should prove
to be a particularly valuable strategy. The successful appli-
cation of TiCl4ÀMg-promoted coupling of CHBr3 with
various carbonyl compounds, especially in the case of
highly enolizable ketones such as 2-indanone and
β-tetralone, highlights the extraordinary reactivity and the
weakly basic nature of this system. On the other hand, this
CHBr3ÀMgÀTiCl4 seems highly selective and might be-
come practical CHBr2- and CBr3-transfer reagents applic-
able to large-scale synthesis. The efficiency of this chem-
istry suggested several intriguing directions which are
currently under active investigation.
Acknowledgment. We thank the National Science
Council of the Republic of China for generous support.
Supporting Information Available. Experimental pro-
cedures and spectra data for 1bÀ16b, 1c, 6cÀ10c, 13c, and
15cÀ19c. This material is available free of charge via the
Internet at http://pubs.acs.org.
The reaction is best envisioned as involving interception
of a presumed [Cl2TiBr(CHBr2)] complex by the CHBr3 to
generate an active [Cl2TiBr(CBr3)] complex and CH2Br2
(Scheme 2). Left unanswered by this proposal is the strong
dependence of the formation of polyhalomethyltitanium
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX