Post-Metalation Structural Insights into the Use of Alkali-Metal-Mediated Zincation for Directed ortho-Metalation of a Tertiary Aromatic Amide.

Communications
of alkyl benzoates, alkyl benzamides, and related azaaromatics.[2, 3] This important reagent is but one of a growing
number of alkali-metal zincate compounds that are attracting
much attention from synthetic chemists in respect of their
often special reactivities/selectivities,[4] and in certain cases,
catalytic action[5] towards organic substrates. Much of our
knowledge of these synthetic applications comes at the end of
the reaction sequence, when all active metal–organic materials have been exhausted,[6] and is usually derived from
isolated, metal-free products obtained by the in situ interception of metalated intermediates by electrophilic quenching. Scheme 1 illustrates this method for the sequential ortho-
Scheme 1. Synthesis of N,N-diisopropyl-2-iodobenzamide by a standard metalation/electrophilic-interception strategy.
ortho-Metalation (2)
DOI: 10.1002/anie.200503213
Post-Metalation Structural Insights into the Use of
Alkali-Metal-Mediated Zincation for Directed
ortho-Metalation of a Tertiary Aromatic Amide**
William Clegg, Sophie H. Dale, Ross W. Harrington,
Eva Hevia, Gordon W. Honeyman, and
Robert E. Mulvey*
Introduced in 1999 by Kondo et al.,[1] lithium di-tert-butyltetramethylpiperidinozincate (LiTMP-zincate) is a highly
chemoselective base for directed ortho-metalation (DoM)
[*] Dr. E. Hevia, Dr. G. W. Honeyman, Prof. R. E. Mulvey
WestCHEM
Department of Pure and Applied Chemistry
University of Strathclyde
Glasgow, G1 1XL (UK)
Fax: (+ 44) 141-552-0876
E-mail: [email protected]
Prof. W. Clegg, Dr. S. H. Dale, Dr. R. W. Harrington
School of Natural Sciences (Chemistry)
University of Newcastle
Newcastle upon Tyne, NE1 7RU (UK)
[**] This work was supported by the UK Engineering and Physical
Science Research Council through grant award no. GR/R81183/01
and by the EU through a Marie Curie Fellowship to E.H.
2374
metalation/iodination of N,N-diisopropylbenzamide by a
LiTMP–zincate/I2 combination,[1] a reaction sequence that is
relevant to the present study. In general, surprisingly little
well-defined information exists on the chemistry taking place
“at the coalface”, that is, about the natures of the starting
alkali-metal zincate reagents themselves and the metalated
intermediates they generate prior to the final, electrophilic
quenching, step. To fully understand this chemistry and
develop it rationally, such information is indispensable,
particularly in the case of LiTMP–zincate and related
ambident reagents which in theory could exhibit either alkyl
or amido (or both) reactivity towards organic substrates. In
the preceding Communication in this issue we discussed the
structures of the TMP–zincate reagent [(thf)Li(m-tmp)(mtBu)Zn(tBu)] and the related tertiary aromatic amide complex [{(iPr)2NC(Ph)(=O)}Li(m-tmp)(m-tBu)Zn(tBu)], thus
providing valuable structural information on zincate bases
prior to them taking part in metalation reactions.[7] Herein, in
reporting the first application of a sodium TMP-zincate
reagent in directed ortho-metalation, we shed light on the
selective ligand transfer effecting ortho-metalation of N,Ndiisopropylbenzamide, by successfully isolating and crystallographically characterizing the ortho-metalated “intermediate”. Furthermore, by similarly elucidating the ortho-metalated structure generated by the analogous lithium TMP–
zincate system (not the original Kondo reagent but a new
TMEDA-complexed variant), we have uncovered an intriguing “alkali-metal effect”, as it transpires that the sodium
zincate and lithium zincate reagents behave differently
towards the aromatic tertiary amide. We also discuss how
these mixed-metal ortho-metalated structures deviate in
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2374 –2377
Angewandte
Chemie
many key respects from the few recorded structures of related
ortho-lithiated aromatic amides.
Both alkali-metal zincate reagents, previously known
[(tmeda)Na(m-tmp)(m-tBu)Zn(tBu)] (1)[8] and the previously
unknown lithium congener [(tmeda)Li(m-tmp)(m-tBu)Zn(tBu)] (2) were generated in situ in hexane solution by
treating NaTMP and LiTMP, respectively, with equimolar
amounts of tBu2Zn then TMEDA. The formulation of 1 as
written was established earlier by X-ray crystallography,[8]
supporting KondoCs view that “precomplexation of MTMP
and tBu2Zn is essential for successful metalation”. Subsequent reaction of 1 with one molar equivalent of N,Ndiisopropylbenzamide yielded the colorless crystalline orthozincated product [(tmeda)Na(tmp){2-[1-C(O)N(iPr)2]C6H4}Zn(tBu)] (3; Scheme 2). The retention of TMP within this
Scheme 2. The distinct reaction pathways followed by sodium zincate
1 and lithium zincate 2 towards N,N-diisopropylbenzamide.
metalation product indicates that at least in part, and possibly
in whole, sodium zincate 1 functions as an alkyl base towards
the aromatic amide. This finding is consistent with the
reported metalating behavior of 1 towards benzene;[8] but
contrasts with the amido basicity exhibited by LiTMP–zincate
in its reaction with this amide[1] (Scheme 1) which, it should be
stressed, was carried out in neat THF solution. In the light of
these precedents, the outcome of the corresponding reaction
between the suspension of TMEDA-complexed lithium
zincate 2 and the amide is wholly surprising. Either
[(tmeda)Li(tmp){2-[1-C(O)N(iPr)2]C6H4}Zn(tBu)] (by analogy with 3) or [(tmeda)Li(tBu){2-[1-C(O)N(iPr)2]C6H4}Zn(tBu)] (by analogy with the LiTMP–zincate reaction) could
have been the product expected, but in practice neither is
found, and instead, unexpectedly, the bis(amide) [(tmeda)Li{2-[1-C(O)N(iPr)2]C6H4}2Zn(tBu)] (4) is produced. NMR
Angew. Chem. Int. Ed. 2006, 45, 2374 –2377
spectroscopic data suggest that production of 4 is accompanied by a significant quantity of residual 2, out of step with the
1:1 stoichiometry of reactants employed in the experiment. It
was then reasoned that 2 could be exhibiting dual alkyl and
amido basicity towards the aromatic amide, and thus that only
0.5 molar equivalents would be needed to convert all of the
amide into the ortho-metalated form, with the co-products
being tBuH and TMPH. This idea was duly confirmed on
repeating the reaction with a 1:2 stoichiometry of zincate 2:
aromatic amide as the yield of isolated, crystalline 4 obtained
increased from 38 % to 72 %. Furthermore, TMP(H) was
found to be the predominant component of the filtrate
following removal of 4. For stoichiometric efficiency 2 is
therefore the reagent solution of choice in this case, though it
is noted that four times as much TMEDA-free LiTMP-zincate
was employed in the aforementioned original reaction[1]
(Scheme 1), that is, a 2:1 as opposed to a 1:2 base:amide
stoichiometry. No explanation was given for using excess
base.
Retaining the “{(tmeda)Na(m-TMP)Zn(tBu)}” backbone
of the reactant zincate 1, the molecular structure of 3[9]
(Figure 1) is completed by the ortho-metalated benzamide
Figure 1. Molecular structure of 3 (hydrogen atoms and minor
disorder component are omitted for clarity). Selected interatomic
distances [A] and angles [8]: Zn1–C1 2.077(3), Zn1–C14 2.055(3), Zn1–
N2 2.018(2), Na1···C1 3.032(3), Na1···C7 3.102(3), Na1–O1 2.258(2),
Na1–N2 2.435(3), Na1–N3 2.576(3), Na1–N4 2.493(3), C7–O1
1.245(3); C1-Zn1-C14 113.66(12), C1-Zn1-N2 115.73(10), C14-Zn1-N2
130.14(11), O1-Na1-N2 112.94(9), O1-Na1-N3 90.41(9), O1-Na1-N4
111.44(9), N2-Na1-N3 127.85(9), N2-Na1-N4 128.71(10), N3-Na1-N4
75.32(10).
fragment which bridges asymmetrically through its oxygen
and ortho-carbon atoms to the sodium and zinc centers,
respectively. This results in a seven-membered, five-element
(NaNZnCCCO) ring. In contrast, having two ortho-metalated
benzamide bridges with the same chemoconnectivity (O to
alkali metal; ortho-C to Zn), 4 exhibits a larger 10-membered
[Li(OCCC)2Zn] ring within its molecular structure[10]
(Figure 2), which is completed by terminal TMEDA (N,N’attached) and tBu (tertiary C-attached) ligands on lithium
and zinc, respectively. A pseudo C2 axis runs through the
LiZnCC(H)3 plane. Further to providing the first insights into
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2375
Communications
Figure 2. Molecular structure of 4 (hydrogen atoms are omitted for
clarity). Selected bond lengths [A] and angles [8]: Zn1–C3 1.934(2),
Zn1–C16 1.971(2), Zn1–C27 2.255(3), Li1–O1 1.881(4), Li1–O2
1.891(4), Li1–N3 2.200(5), Li1–N4 2.185(5), O1–C1 1.280(3), O2–C14
1.250(3); C3-Zn1-C16 111.76(10), C3-Zn1-C27 122.61(11), C16-Zn1C27 125.59(10), O1-Li1-O2 112.0(2), O1-Li1-N3 114.58(18), O1-Li1-N4
115.7(2), O2-Li1-N3 117.1(2), O2-Li1-N4 116.83(19), N3-Li1-N4
76.61(15).
the products of tertiary amide-directed ortho metalation[11]
promoted by alkali-metal zincate reagents, the structures of 3
and 4 enable comparisons with the few known examples of
analogous ortho-lithiated (homometallic) structures. All
three such compounds reported, namely, N,N-diisopropyl-2lithiobenzamide·Et2O,[12]
N,N-diisopropyl-2-lithionaphthamide·THF
and
N,N-diisopropyl-2-ethyl-6-lithiobenzamide·THF[13] adopt essentially the same dimeric, trans 5.4.5
fused-ring, arrangement with a central (LiC)2 ring and
LiOCCC outer rings (5; Figure 3)—a motif markedly distinct
to that of either 3 or 4. The salient feature of 5 is that lithium
binds simultaneously to both the carbonyl oxygen atom (a
contact assumed to be important mechanistically in terms of
the “complex-induced proximity effect”)[7, 14] and the orthocarbon anion. This is a key distinction in comparison to the
situation in 3 and 4, where the alkali-metal binds to the
carbonyl oxygen atom but is distant from the ortho-carbon
anion. In 4, where a direct comparison is possible, the mean
Li O (carbonyl) length (1.886 K) is significantly shorter than
those in each structure of 5 (1.936(3), 1.974(4), and
1.972(5) K, respectively).[11, 12] Aided by zinc, lithium can
approach the carbonyl oxygen atom more closely than its
counterparts in 5 and at the same time zinc can engage in
shorter and stronger bonds with the ortho-carbon anion
(mean length in 4, 1.953 K; in 3, 2.077 K: the significant
lengthening of the latter bond could be due to the presence of
the bulky TMP co-ligand on this zinc atom), than the lithium
atoms can achieve in 5 (mean length over the three structures,
2.280 K). Thus the metalation process generating 3 and 4 is
best viewed as an “alkali-metal-mediated zincation”,[15] and
the synergic bimetallic component of the bonding within the
products appears to lead to an extra stabilization in comparison to the situation in the single-metal, ortho-lithiated
products. However, whether the synergic bonding would
have a similar beneficial effect on the stability of the
transition states and on the reaction rates of these orthometalation processes remains an open question. In the report
of
aforementioned
N,N-diisopropyl-2-lithiobenzamide·Et2O[11] it was noted that the dihedral angle of the amide
N C=O and aromatic Ph planes (578 in the parent amide)[16]
had to lessen to 46.88 to allow coordination of lithium to the
amide C=O group; interestingly the situation is reversed in 3
and 4 as this twist angle increases to 63.24(17)8 and 76.24(13)8,
respectively. These increases may reflect the fact that there is
no longer a requirement for the alkali metal to interact
simultaneously with the ortho-carbon anion, though caution is
required in attempting to rationalize differences in any bond
angle in isolation as the global stereochemistries of 3, 4, and 5,
are markedly different in each case. ortho-Lithiation reactions
utilizing sBuLi as the base are sensitive to the twist angle of
the amide, with coplanarity (twist angle, 08) leading to the
fastest rates.[16] Whether the same correlation would hold true
in the case of alkali-metal zincate bases that give rise to
metalated products with unique structures distinct from those
of ortho-lithiated products remains to be ascertained.
It is worth emphasizing that tBu2Zn, a feeble deprotonating agent on its own, is incapable of directly zincating tertiary
aromatic amides.[17] Hence, the description of the new
reactions outlined herein as “alkali-metal-mediated zincations” is both accurate and apt. These results also establish
that the identity of the alkali metal can have a profound effect
on the outcome of a zincate reaction, ruling out the possibility
that mechanistically the zincate reagents react as solventseparated ion pairs [M·(solvent)x]+[Zn(tmp)(tBu)2] , which
would be expected to yield equivalent [ZnR3] products.
Moreover, this study reinforces the added value of studying
the reactive metallo intermediates, because quenching the
reaction solutions of 3 or 4 with an electrophile such as I2
would give the same set of organic products and the
distinction between the lithium and sodium systems would
have remained invisible.
Experimental Section
Figure 3. Representative structure of the trans 5.4.5 fused-ring arrangement found in ortho-lithiated aromatic amides.
2376
www.angewandte.org
All reactions were carried out under a protective argon atmosphere.
3: A solution of tBu2Zn (0.358 g, 2 mmol) in hexane (10 mL) was
transferred by cannula to a suspension of NaTMP in hexane
[prepared in situ by reaction of BuNa (0.16 g, 2 mmol) with
TMP(H) (0.34 mL, 2 mmol)]. Subsequently a molar equivalent of
TMEDA (2 mmol, 0.30 mL) was added. The resulting suspension was
heated gently to form a yellow solution which was allowed to cool to
ambient temperature before the addition of a molar equivalent
(2 mmol, 0.41 g) of N,N-diisopropylbenzamide, forming a brightly
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2374 –2377
Angewandte
Chemie
colored yellow solution. The solution was transferred to the freezer to
aid crystallization. A crop (0.48 g, 40 %) of transparent crystals
formed in solution that were suitable for X-ray crystallographic
analysis. FT-IR (nujol): ñ = 1598 cm 1(nC=O). 1H NMR (400 MHz,
25 8C, C6D6): d = 7.98 (d, 1 H, Hmeta Ar’), 7.21 (t, 1 H, Hpara Ar’), 7.02 (t,
1 H, Hmeta* Ar’), 6.95 (d, 1 H, Hortho* Ar’), 4.17 (m, 1 H, CH, iPr, Ar’),
3.08 (m, 1 H, CH, iPr, Ar’), 1.98 (m, 6 H, broad, 2 H of g-TMP and 4 H
of b-TMP), 1.79 (s, 4 H, CH2 TMEDA), 1.71 (s, 12 H, CH3 TMEDA),
1.56 (s, 9 H, tBu), 1.54, 1.51, 1.30, 1.20 (s, 3H each, CH3-TMP), 1.48,
1.39, 0.91, 0.81 ppm (d, 3 H each, CH3, iPr, Ar’). 13C{1H} NMR
(100.63 MHz, 25 8C, [D6]benzene): d = 179.22 (C=O), 169.15 (Cortho),
148.06 (Cipso), 142.26 (Cmeta), 127.23 (Cmeta*), 124.04 (Cpara), 123.84
(Cortho*), 57.94 (CH2, TMEDA), 53.55, 53.23 (a-TMP), 52.00, 46.22
(CH, iPr), 46.32 (CH3, TMEDA), 41.18, 40.29 (b-TMP), 37.83, 36.16,
35.34, 34.87 (CH3-TMP), 36.64 (CH3, tBu), 22.89, 21.97 (CH3, iPr),
21.03 (g-TMP), 20.70 ppm (2 N CH3, iPr and C(CH3)3, tBu). Figure 4
shows the labeling scheme of benzamide (Ar’).
Figure 4. Atom-labeling scheme used in the NMR assignments of the
ortho-zincated benzamide structures.
4: A solution of tBu2Zn (0.358 g, 2 mmol) in hexane (10 mL) was
transferred by cannula to a solution of LiTMP in hexane [prepared
in situ by reaction of BuLi (1.25 mL of a 1.6 m solution in hexane,
2 mmol) with TMP(H) (0.34 mL, 2 mmol)]. The addition of TMEDA
(0.30 mL, 2 mmol) to the resulting pale yellow solution was accompanied by the precipitation of a white solid. N,N-diisopropylbenzamide (0.34 mL, 2 mmol) was then introduced. A bright yellow
solution is obtained which was stirred for 30 min at room temperature. Toluene (2 mL) was added to the solution that was slightly
cloudy, obtaining a transparent bright yellow solution that was placed
in the freezer at 20 8C. A crop of cubic colorless crystals was
obtained (0.51 g, 38 %); one of these crystals was used for an X-ray
crystallography experiment. FT-IR (nujol): ñ = 1603 cm 1(nC=O).
1
H NMR (400 MHz, 25 8C, C6D6): d = 8.26 (d, 2 H, Hmeta Ar’), 7.29
(t, 2 H, Hpara Ar’), 7.10 (t, 2 H, Hmeta* Ar’), 6.84 (d, 2 H, Hortho* Ar’), 4.23
(m, 2 H, CH, iPr, Ar’), 3.02 (m, 2 H, CH, iPr, Ar’), 1.89 (24 H, CH2 and
CH3 TMEDA), 1.80 (s, 9 H, tBu), 1.25, 1.15, 1.10, 0.67 ppm (d, 6 H
each, CH3, iPr, Ar’). 13C{1H} NMR (100.63 MHz, 25 8C, [D6]benzene):
d = 179.01 (C = O), 169.27 (Zn-Cortho), 146.78 (Cipso), 141.77 (Cmeta),
126.13 (Cmeta*), 123.52 (Cpara), 122.26 (Cortho*), 57.58 (CH2, TMEDA),
51.27 (CH, iPr), 46.43 (CH3, TMEDA), 45.82 (CH, iPr), 36.31 (CH3,
tBu), 22.87 (C(CH3)3, tBu), 21.32, 20.85, 20.83, 20.51 ppm (CH3, iPr).
7
Li NMR (155.50 MHz, 25 8C, C6D6, reference LiCl in D2O at
0.00 ppm): 0.12 ppm.
Received: September 9, 2005
Revised: November 8, 2005
Published online: March 2, 2006
.
Keywords: alkali metals · amides · directing groups · metalation ·
zinc
[1] Y. Kondo, M. Shilai, M. Uchiyama, T. Sakamoto, J. Am. Chem.
Soc. 1999, 121, 3539.
[2] T. Imahori, M. Uchiyama, T. Sakamoto, Y. Kondo, Chem.
Commun. 2001, 2450.
Angew. Chem. Int. Ed. 2006, 45, 2374 –2377
[3] M. Uchiyama, T. Miyoshi, Y. Kajihara, T. Sakamoto, Y. Otani, T.
Ohwada, Y. Kondo, J. Am. Chem. Soc. 2002, 124, 8514.
[4] W. TQckmantel, K. Oshima, H. Nozaki, Chem. Ber. 1986, 119,
1581; T. Harada, T. Katsuhira, A. Osada, K. Iwazaki, K.
Maejima, A. Oku, J. Am. Chem. Soc. 1996, 118, 11 377; M.
Uchiyama, M. Koike, M. Kameda, Y. Kondo, T. Sakamoto, J.
Am. Chem. Soc. 1996, 118, 8733; Y. Kondo, N. Takazawa, C.
Yamazaki, T. Sakamoto, J. Org. Chem. 1994, 59, 4717; M.
Uchiyama, S. Furumoto, M. Saito, Y. Kondo, T. Sakamoto, J.
Am. Chem. Soc. 1997, 119, 11 425; M. Uchiyama, M. Kameda, O.
Mishima, N. Yokoyama, M. Koike, Y. Kondo, T. Sakamoto, J.
Am. Chem. Soc. 1998, 120, 4934.
[5] F. F. Kneisel, M. Dochnahl, P. Knochel, Angew. Chem. 2004, 116,
1032; Angew. Chem. Int. Ed. 2004, 43, 1017; L.-Z. Gong, P.
Knochel, Synlett 2005, 267.
[6] For an NMR spectroscopic structural determination of organozincate reagents see: T. A. Mobley, S. Berger, Angew. Chem.
1999, 111, 3256; Angew. Chem. Int. Ed. 1999, 38, 3070.
[7] W. Clegg, S. H. Dale, E. Hevia, G. W. Honeyman, R. E. Mulvey,
Angew. Chem. 2006, 118, 2430–2434; Angew. Chem. Int. Ed.
2006, 45, 2371.
[8] P. C. Andrikopoulos, D. R. Armstrong, H. R. L. Barley, W.
Clegg, S. H. Dale, E. Hevia, G. W. Honeyman, A. R. Kennedy,
R. E. Mulvey, J. Am. Chem. Soc. 2005, 127, 6184.
[9] Crystal data for 3: C32H61N4NaOZn, Mr = 606.2, monoclinic,
space group P21/n, a = 14.946(3), b = 15.913(3), c = 15.158(3) K,
b = 90.18(3)8, V = 3605.1(12) K3, Z = 4, 1calcd = 1.117 g cm 3, m(MoKa) = 0.72 mm 1, T = 150 K, R = 0.053 (F2 > 2 s), Rw = 0.107
(for all F2 values) for 6942 data and 398 refined parameters; final
difference synthesis within 0.41 e K 3. The tBu group is
disordered over two orientations. H atoms were constrained
with a riding model.
[10] Crystal data for 4: C36H61LiN4O2Zn, Mr = 654.2, monoclinic,
space group P21/n, a = 14.192(4), b = 17.825(5), c = 15.641(5) K,
b = 90.041(7), V = 3957(2) K3, Z = 4, 1calcd = 1.098 g cm 3, m(MoKa) = 0.65 mm 1, T = 150 K, R = 0.034 (F2 > 2 s), Rw = 0.091
(for all F2 values) for 8018 data and 413 refined parameters; final
difference synthesis within 0.52 e K 3. CCDC 283196 and
CCDC 283197 contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
[11] For a minireview on how homo- and hetero-aggregation can
affect directed ortho-lithiation see: R. A. Gossage, J. T. B. H.
Jastrzebski, G. van Koten, Angew. Chem. 2005, 117, 1472;
Angew. Chem. Int. Ed. 2005, 44, 1448.
[12] J. Clayden, R. P. Davies, M. A. Hendy, R. Snaith, A. E. H.
Wheatley, Angew. Chem. 2001, 113, 1282; Angew. Chem. Int. Ed.
2001, 40, 1238.
[13] D. R. Armstrong, S. R. Boss, J. Clayden, R. Haigh, B. A.
Kirmani, D. J. Linton, P. Schooler, A. E. H. Wheatley, Angew.
Chem. 2004, 116, 2187; Angew. Chem. Int. Ed. 2004, 43, 2135.
[14] P. Beak, A. I. Meyers, Acc. Chem. Res. 1986, 19, 356; P. Beak,
J. E. Hunter, Y. M. Jun, J. Am. Chem. Soc. 1983, 105, 6350.
[15] For an example of alkali-metal-mediated magnesiation see: P. C.
Andrikopolous, D. R. Armstrong, D. V. Graham, E. Hevia,
A. R. Kennedy, R. E. Mulvey, C. T. OCHara, C. Talmard, Angew.
Chem. 2005, 117, 3525; Angew. Chem. Int. Ed. 2005, 44, 3459.
[16] P. Beak, S. T. Kerrick, D. J. Gallagher, J. Am. Chem. Soc. 1993,
115, 10 628.
[17] For the use of a mixed LiTMP–nBu2Zn system in the direct
zincation of ferrocene see: H. R. L. Barley, W. Clegg, S. H. Dale,
E. Hevia, G. W. Honeyman, A. R. Kennedy, R. E. Mulvey,
Angew. Chem. 2005, 117, 6172; Angew. Chem. Int. Ed. 2005, 44,
6018.
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2377