Aromatic substitution with hypercloso C(BCH3)11: A new mechanism

Article abstract of DOI:10.1021/ja066520x

It is proposed that the solvolytic intermediate C(BMe)₁₁ performs aromatic substitution by a novel mechanism. Copyright

Full text of DOI:10.1021/ja066520x

Published on Web 03/15/2007
Aromatic Substitution with Hypercloso C(BCH3)11: A New Mechanism
Kamesh Vyakaranam,† Zdenek Havlas,‡ and Josef Michl*,†,‡
Department of Chemistry and Biochemistry, UniVersity of Colorado, Boulder, Colorado 80309-0215, and
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, FlemingoVo na´m. 2,
166 10 Prague, Czech Republic
Received October 24, 2006; E-mail: michl@eefus.colorado.edu
Scheme 1. Aromatic Substitution with 1
We report aromatic substitution with the reactive intermediate
+
C(BCH3)11 (1, Chart 1),1 formally a hypercloso cation HC(BCH3)11
with a deprotonated carbon vertex but, according to calculations,2
better viewed in the ground state as a closo cluster with a 5-fold
symmetric carbenoid/carbonium ylide structure C(BMe)11
T
+C-(BMe)11 (1a) carrying a vacant exocyclic orbital on carbon.3
The reaction yields highly hindered C-arylated and B-permethylated
carborate anions (Me ) CH3):
ArH + 1 f Ar-C(BMe)11- + H+
(1)
In the process, 1 exhibits a novel selectivity, unrelated to its
known1 electrophilic affinity for lone pairs. To account for this
result, we propose that this aromatic substitution proceeds by a new
mechanism with a norcaradiene-like transition state. The properties
of 1 may be relevant for the reactions and structures of the isomeric
“boronium ylides”2 with a naked boron vertex, which occur as
intermediates in electrophile-induced nucleophilic substitution
(EINS) on polyhedral boranes.4-6
the same products, now with C2H4 and HBr as byproducts. The
latter was quenched with 2,6-di-tert-butylpyridine (4), added either
before or after the reaction, with identical results. The trapping
was nearly quantitative, and as expected, the product was mostly
2-.
Lone-pair nucleophiles gave the expected1 products. Reactions
-
of 3- with pyridine and NaN3 yielded C5H5N+-C(BMe)11 and
1-N3-C(BMe)11-, respectively; in pure (CF3)2CHOH, 3- produced
-
2-, and in CF3CH2OH, it afforded CF3CH2O-C(BMe)11
.
Chart 1. Structure of 1: Left, Carbenoid/Carbonium Ylide. Right,
All monosubstituted benzenes, o-xylene (5), naphthalene (6), and
biphenylene (7) gave substitution products according to eq 1.
However, 4, m-xylene (8), p-xylene (9), acenaphthylene (10),
pleiadiene (11),7 and pyrene (12) did not react (Chart 2).
A Possible Biradicaloid Form
Chart 2. Aromatic Substitution with 1: Reactive Sites in
Hydrocarbons
Two sources of 1 were used (Scheme 1) and yielded indistin-
guishable results, confirming the intermediacy of free 1, deduced
previously from the observed solvolytic zero kinetic order in the
nucleophile:1 (i) Solvolysis of 1-(CF3)2CHO-C(BMe)11- (2-) with
excess trapping agent yielded substitution products in high
isolated yield. Addition of hydroquinone in one run had no
effect. (ii) Solvolytic Grob fragmentation of ∼3-8-fold excess
The absence of detectable ortho substitution with 1 is attributed
to the steric demands of the methyl groups adjacent to its carbon
vertex. This also explains the lack of substitution in 9 and in
the peri positions of 6, 7, and 10-12, in which a second ring
plays the role of an ortho substituent. Remarkably, starting with
-
of 1-Br(CH2)2-C(BMe)11 (3-) over the trapping partner yielded
† University of Colorado.
‡ Academy of Sciences of the Czech Republic.
9
4172
J. AM. CHEM. SOC. 2007, 129, 4172-4174
10.1021/ja066520x CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Trapping of 1 with Substituted Benzenes (PhX) and with
given the reactivity of 5 and toluene, and the inertness of 10-12,
given the reactivity of 6.
Aromatic Hydrocarbonsa
b
m/p
rel
m/p(Ph
c,j
‚
)
rel(Ph
h,j
‚
)
We propose that 1 cycloadds in a rate-determining step to two
adjacent vertices of the substrate, neither of which is allowed to
have an ortho substituent. Removal of a proton from one of the
attacked vertices and a loss of one of the two partial bonds between
the reagent and the substrate then convert the cyclic intermediate
into the product. The choice of the aromatic vertex to be
deprotonated determines which of the two new bonds is kept and,
hence, the regioselectivity. The need for both ortho positions to be
free is rationalized by the steric demands of the methyl substituents.
Inspection of computer models also makes it clear that ipso
substitution of the trimethylsilyl group is not hindered by its
bulkiness.
If the reaction proceeds by an attack on a bond, it should be
favored by high π bond order at the center of the reacting
HCHCCHCH unit. This indeed seems to be so: (i) one of the larger
rate effects found in Table 1 is the reactivity ratio of 7 over 6 (the
order of the reacting 2-3 bond is higher in 7 because resonance
structures with a cyclobutadiene central ring have smaller weight);
and (ii) the reaction fails with 11, whose 2-3 bond is sterically
qualified but is nearly single.
Both new partial bonds formed in the rate-determining step could
originate in the naked carbon of 1 (Scheme 2A), in which case the
intermediate would have a partially formed three-membered ring.
Then, in symmetric substrates such as benzene, both aromatic CH
vertices under attack would remain equivalent, and either could be
deprotonated in the next step with the same likelihood. Since the
isotopic effect in the C6H6/C6D6 competition is only ∼1.15, an even
smaller effect would be expected on the rate of the reaction of
benzene-1,3,5-d3. However, one would expect a significant isotope
effect on product ratio, as the removal of a proton from the attacked
DCCH bond should be faster than the removal of a deuteron. For
instance, the rate-determining hydrogen removal step in the reaction
of singlet oxygen or triazolinedione with alkenes exhibits an isotopic
effect of 1.4.11
Alternatively (Scheme 2B), 1 could participate through one of
its five edges or trigonal faces adjacent to the naked carbon vertex.
One of the new partial bonds would be made by the vertex carbon,
and one would be made by one of its five boron neighbors, or two
adjacent neighbors. In the cycloadduct, the two aromatic CH vertices
would be nonequivalent and it would be the one attached to the
carbon of 1 that would have to lose a proton or a deuteron to form
the product. The isotope effect would be determined already in the
first reaction step and should be less than 1.15.
X
2c,d,e
3c,f,g
2h,i
3h
c,k
h,k
NO2
CF3
CN
CH3
Cl
Br
I
CH3O
(CH3)2N
1.4
1.3
3.4
1.6
1.3
1.4
1.3
3.6
1.5
1.4
0.9
0.8l
0.4 0.3 2.0 3.0
1.0 1.4
0.5 0.3 2.9 2.6
1.4 1.2
1.7 1.7 1.0 0.9
1.7 1.9 0.8 1.6
0.9
1
1
1
1.2m 1.0
1.7n
1.1
0.6
0.5
0.4
o
0.7
0.4
0.3
o
1.1l
1.2 1.2 0.9 0.7
5
6
p
p
(1q)
7
p
p
(2.1q)
1.2r
C6D6
1.1r
1,3,5-C6H3D3 1.3s
1.4s
a The products were separated by reverse phase HPLC. Structures were
assigned by 1H and 13C NMR. Product ratios were determined by NMR of
reaction mixtures. Fluorobenzene produced a complex mixture of cage
fluorinated products that we were unable to separate by HPLC and was
excluded from the study. b Only the meta and para isomers are counted as
products in total yield evaluation. c The ratio of meta over para isomer.
Not corrected for the 2:1 statistical advantage favoring the meta position.
d In (CF3)2CHOH/1% triflic acid at 40 °C, [2] ) 17 mM, [ArH] ) 0.5 M,
in competition experiments [ArH] ) [toluene] ) 0.25 M. e Total isolated
yield of both isomers, based on 2, 80-90%. f In (CF3)2CHOH at 60 °C,
[3] ) 8-25 mM, [ArH] ) 3-11 mM, in competition experiments
[ArNO2] ) [ArOMe] and [6] ) 1.5[7]. g Total reactivity (both products)
relative to toluene. h Total isolated yield based on ArH, 60-90%. The
remainder of 3 was isolated as 2. i Relative reactivities from competi-
tion experiments. j From ref 9. k From ref 8. l From a competition between
X ) NO2 and X ) CH3O. m In the presence of hydroquinone, 1.25. n In
the presence of hydroquinone, 1.6. o Only the 4-substituted isomer is
formed. p Only the 2-substituted isomer is formed. q From a competition
between 7 and 6, corrected for the 1.5 to 1 ratio of trap concentrations.
r Reactivity ratio of C6H6 over C6D6. s Product ratio of Ar ) C6H2D3 over
Ar ) C6H3D2.
3- and 4-bromotrimethylsilylbenzene, ipso substitution of the
trimethylsilyl group took place in a ∼30% yield, and 60% of 2-
was also isolated.
The meta to para substitution ratios in monosubstituted benzenes
suggest that in intramolecular competition 1 is slightly electrophilic,
but this is not reflected in the relative substrate reactivity. Together,
the data in Table 1 exclude polar mechanisms, especially the
anticipated electrophilic aromatic substitution. A radical mechanism
can be envisaged: (i) the ground state of a Wheland complex might
•
be a biradical, cyclohexadienyl attached to -C(BMe)11 , instead
of an ylide, benzenonium attached to -C(BMe)11-; (ii) 1 might
have a triplet ground state (instead of the calculated2 singlet).
However, an ordinary radical aromatic substitution mechanism does
not account for the failure of 4, 8, and 10-12 to react, nor is it
particularly favored by a comparison with the selectivity in aromatic
substitution by the phenyl radical8,9 (Table 1) and by the finding
that addition of a radical scavenger had no effect.
The other known nonpolar mechanism is insertion into CH
bonds.10 The nearly identical reactivity of C6H6 and C6D6 and the
small isotope effect in benzene-1,3,5-d3 show that such insertion
is not involved in the rate-determining step. Rather, the rate-
determining step of the substitution appears to generate an
intermediate that is subsequently deprotonated.
The isotopic effect actually observed in the trapping with
benzene-1,3,5-d3 is ∼1.35 and argues in favor of Scheme 2A.
Initial computational results are also more compatible with
Scheme 2A. A diligent search up to the CCSD(T)/cc-pVDZ//
B3LYP/6-31G* level of optimization of a simplified model,
CB11H11, produced structures with a stretched C-B edge resembling
1b, but only at energies at least 45 kcal/mol above that of the C5V
symmetry structure. B3LYP/631G* optimizations of the geometry
of CB11H11 at 50 different starting points generated by random
displacements of atoms from the C5V symmetric structure identified
11 low-symmetry minima in the potential energy surface, but all
were at least 27 kcal/mol higher in energy. Results of a few
calculations on the CB11Me11 analogues gave no reason to expect
that this number will be reduced significantly by the presence of
the methyl groups. The results make it very unlikely that a structure
Most important, all mechanisms invoking an attack by 1 on one
of the aromatic CH vertices fail to account for the inertness of 8,
9
J. AM. CHEM. SOC. VOL. 129, NO. 14, 2007 4173

C O M M U N I C A T I O N S
Scheme 2. Aromatic Substitution with 1: Cluster Participation
through a Vertex (A) and through an Edge (B)
resemble the addition of singlet carbene to an aromatic ring. The
separation of charges in the first intermediate in Scheme 2A is
undoubtedly also only formal, making it compatible with the near
absence of polar effects on the reaction rate. Attempts to optimize
the geometry of the proposed reaction intermediate and to examine
its electronic structure have so far been fruitless and may require
the inclusion of solvent.
We conclude that 1 performs aromatic substitution to yield
sterically crowded products not easily accessible otherwise and does
so by an unprecedented mechanism, compatible with its previously
proposed unusual electronic structure.
Acknowledgment. This work was supported by the NSF (CHE-
0446688 and OISE-0532040) and the Czech Ministry of Education
(KONTAKT No. ME 857). We are grateful to anonymous review-
ers, especially for drawing our attention to ref 11.
Supporting Information Available: Experimental details and
results of ab initio computations. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Vyakaranam, K.; Ko¨rbe, S.; Divisˇova´, H.; Michl, J. J. Am. Chem. Soc.
2004, 126, 15795.
(2) Zharov, I.; Havlas, Z.; Orendt, A. M.; Barich, D. H.; Grant, D. M.; Fete,
M. G.; Michl, J. J. Am. Chem. Soc. 2006, 128, 6089.
(3) A somewhat analogous formal doubly deprotonated hypercloso dication
2+
(HC)2(BH)10 has been also identified as a reaction intermediate: (a)
Ghosh, T.; Gingrich, H. L.; Kam, C. K.; Mobraaten, E. C.; Jones, M., Jr.
J. Am. Chem. Soc. 1991, 113, 1313. (b) Kiran, B.; Anoop, A.; Jemmis, E.
D. J. Am. Chem. Soc. 2002, 124, 4402.
(4) Plesˇek, J.; Plza´k, Z.; Stuchl´ık, J.; Herˇma´nek, S. Collect. Czech. Chem.
Commun. 1981, 46, 1748.
(5) Jel´ınek, T.; Plesˇek, J.; Maresˇ, F.; Heˇrma´nek, S.; Sˇt´ıbr, B. Polyhedron 1987,
6, 1981.
(6) Ko¨rbe, S.; Schreiber, P. J.; Michl, J. Chem. ReV. 2006, 106, 5208.
(7) Meinwald, J.; Samuelson, G. E.; Ikeda, M. J. Am. Chem. Soc. 1970, 92,
7604.
(8) Rondestvedt, C. S., Jr.; Blanchard, H. S. J. Org. Chem. 1956, 21,
229.
such as 1b represents the ground state of 1 or even that it intervenes
along the substitution reaction path.
The structure 1a fits the observed reactions with lone-pair
carrying electrophiles, but at first sight suggests that aromatic
substitution with 1 will proceed along the standard electrophilic
substitution path. However, density functional theory (DFT)
calculations2 suggested that the formal positive and negative charges
actually mostly reside in different orbitals of the same carbon atom,
such that the electronic structure is reminiscent of an S2 state of a
carbene more than the charge-separated state of an ylide. It is then
reasonable that the first step of the aromatic substitution should
(9) Vernin, G.; Jauffred, R.; Ricard, C.; Dou, H. J. M.; Metzger, J. J. Chem.
Soc., Perkin Trans. 2 1972, 1145.
(10) (a) Waltz, K. M.; Muhoro, C. N.; Hartwig, J. F. Organometallics 1999,
18, 3383. (b) Watson, P. L. J. Am. Chem. Soc. 1983, 105, 6491. (c)
Labinger, J. A.; Bercaw, J. E. Nature 2002, 417, 507. (d) Thompson, M.
E.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2000,
122, 10846.
(11) Elemest, Y.; Foote, C. S. J. Am. Chem. Soc. 1992, 114, 6044.
JA066520X
9
4174 J. AM. CHEM. SOC. VOL. 129, NO. 14, 2007