2036
H. Li, J. Balsells / Tetrahedron Letters 49 (2008) 2034–2037
Br
Br
Br
1) i-PrMgCl
2) DMF
1) n-Bu3MgLi
2) Boc2O
90%
Br
Br
Br
COOtBu
Scheme 3.
OHC
COOtBu
97%
is an excellent electrophile to trap triarylmagnesium ate
complexes affording the desired t-butyl esters with com-
plete selectivity and in very high yields. Contrary to the
report by Lee, our reaction conditions do not require
excess organometallics, sonication, or the use of Lewis
acids to achieve good yields.
Although chemists tend to rely mostly on methyl or
ethyl esters, the use of t-butyl esters may have advantages
in some cases. The possibility of t-butylester hydrolysis
under acidic conditions makes these compounds compli-
mentary to simpler esters such as methyl or ethyl. Within
the context of this chemistry, we have successfully used
the t-butyl moiety as a protecting group for the ester func-
tionality in subsequent Grignard chemistry. This is exem-
plified by the reaction of t-butyl-3,5-dibromobenzoate
obtained by our method with i-propylmagnesium chloride
followed by addition to dimethylformamide (Scheme 3).13
In summary, we have demonstrated that di-t-butyl
dicarbonate is a good electrophile for additions to triarene-
magnesium ate complexes. When coupled with the high
selectivity for single metal–halogen exchange, this method
is a practical solution to the limitation of transition
metal-catalyzed carbonylations of arenes bearing multiple
halogen substituents.
After some reaction optimization, we were able to per-
form the complete process in a one-pot operation under
non-cryogenic conditions. In most cases, the products
obtained did not require further purification and were
obtained as colorless oils or solids that could be used in sub-
sequent chemical steps. Lithium tri-n-butylmagnesium ate
complex is generated in situ at À10 °C in toluene by mixing
1/3 mol equiv of n-butyl magnesium chloride with
2/3 mol equiv of n-butyl lithium. The resulting organo-
metallic species has the capability to transfer all three butyl
groups readily, making the process very efficient. A slight
excess of organometallic species is typically used in our pro-
cedure to compensate for the use of reagent grade solvents
rather than anhydrous grade. Despite this excess reagent,
we have never observed byproducts arising from a double
metal–halogen exchange reaction. Subsequent quenching
of the triarylmagnesium ate complex with Boc anhydride
also takes place at À10 °C and is typically complete in
2–3 h. Both chemical transformations can be monitored
easily by HPLC. Once the reaction is judged complete by
HPLC (<1% bromobenzene remaining), the addition of a
10% (w/w) citric acid solution causes the addition inter-
mediate to collapse with evolution of CO2.
References and notes
1. See for example: (a) Skoda-Foldes, R.; Kollar, L. Curr. Org. Chem.
2002, 6, 1097–1119; (b) Beller, M.; Indolese, A. F. Chimia 2001, 55,
684–687.
2. Bratton, L. D.; Huh, H.; Bartsch, R. A. J. Heterocycl. Chem. 2000,
37, 815–819.
3. Amedio, J. C. J.; Lee, G. T.; Prasad, K.; Repic, O. Synth. Commun.
1995, 25, 2599–2612.
4. Larock, R. C. Comprehensive Organic Transformations: A Guide to
Functional Group Preparations, 2nd ed.; John Wiley & Sons: New
York, 1999.
5. Cai, C.; Rivera, N. R.; Balsells, J.; Sidler, R. R.; McWilliams, J. C.;
Shultz, C. S.; Sun, Y. Org. Lett. 2006, 8, 5161–5164.
6. Cassar, L.; Foa, M.; Gardano, A. J. Organomet. Chem. 1976, 121,
C55–C56.
7. Lee, A. S.-Y.; Wu, C.-C.; Lin, L.-S.; Hsu, H.-F. Synthesis 2004, 568–
572.
8. Iida, T.; Wada, T.; Tomimoto, K.; Mase, T. Tetrahedron Lett. 2001,
42, 4841–4844.
9. All compounds isolated gave consistent 1H NMR, 13C NMR and
GCMS or LCMS data.
We next examined the application of this protocol to
other di- and trihaloarenes and heteroarenes. The results
obtained are summarized in Table 1. The reaction per-
formed very well on di- and tribromobenzenes 1, 2 and 3.
The reaction of tribromobenzene 4 afforded t-butyl-2,5-
dibromobenzoate as the major product in 71% yield.11
Complete selectivity was observed on the reaction of 1,4-
dibromo-2-fluorobenzene 5 to afford t-butyl-4-bromo-2-
fluorobenzoate in 92% yield.
10. Typical reaction procedure; Synthesis of t-butyl-3,5-dibromobenzoate:
n-Butyl lithium (7.2 mL, 18.0 mmol, 2.5 M in hexanes) was added to
a 100 mL flask under nitrogen atmosphere containing 9 mL of
toluene at À10 °C. n-Butylmagnesuim chloride (4.5 mL, 9.0 mmol,
2 M in THF) was then added at such a rate to keep the temperature
<À5 °C. The resulting milky slurry was aged at À10 °C for 30 min.
To this slurry was added 1,3,5-tribromobenzene (6.67 g, 21.2 mmol)
dissolved in 20 mL of toluene. The rate of addition was such that
the temperature did not increase above À5 °C. After the addition
was complete, the mixture was kept at À10 °C until the metal–
halogen reaction was complete (30 min to 2 h depending on the
substrate). A solution of di-t-butyl dicarbonate (5.89 g, 27 mmol) in
7.5 mL of toluene was then charged such at a rate to keep the
temperature <À5 °C. After the addition was complete, the mixture
was kept at À10 °C until the aryl-Mg intermediate was completely
Substrates containing both chloro and bromo substitu-
ents (6–8) showed preference towards the bromide for
metal–halogen exchange, affording the corresponding benz-
oates in high yields. Mixed results were obtained with
dibromoheteroaromatic compounds. While furan 9, thio-
phene 10 and pyridine 11 behaved well in the reaction, pyr-
idines 12 and 13 decomposed completely upon the addition
of Boc anhydride resulting in complex mixtures of products.
This was a surprising outcome since both the substrates
behaved well when submitted to the same metal–halogen
exchange conditions followed by quenching with
dimethylformamide.8
Li, Hongmei
