Synthesis and reactivity of (C6F5)3B-N-heterocycle complexes. 1. Generation of highly acidic sp3 carbons in pyrroles and indoles

Syn th esis a n d Rea ctivity of (C₆F ₅)₃B-N-Heter ocycle Com p lexes. 1.

Gen er a tion of High ly Acid ic sp ³Ca r bon s in P yr r oles a n d In d oles

Simona Guidotti,^†Isabella Camurati,^†Francesca Focante,^‡Luca Angellini,^‡

Gilberto Moscardi,^†Luigi Resconi,*^,†Rino Leardini,^‡Daniele Nanni,^‡Pierluigi Mercandelli,^§

Angelo Sironi,^§Tiziana Beringhelli,^#and Daniela Maggioni^#

Basell Polyolefins, Centro Ricerche G. Natta, I-44100 Ferrara, Italy; Dipartimento di Chimica Organica

“A. Mangini”, Universita` di Bologna, Viale Risorgimento 4, I-40136 Bologna, Italy; and Dipartimento

di Chimica Strutturale e Stereochimica Inorganica and Dipartimento di Chimica Inorganica,

Metallorganica e Analitica, Universita` di Milano, Via Venezian 21, I-20133 Milano, Italy

luigi.resconi@basell.com

Received October 14, 2002

The reaction of pyrroles and indoles with B(C₆F₅)₃and BCl₃produces 1:1 B-N complexes containing

highly acidic sp³carbons, for example, N-[tris(pentafluorophenyl)borane]-5H-pyrrole (1) and N-[tris-

(pentafluorophenyl)borane]-3H-indole (2), that are formed by a new formal N-to-C hydrogen shift,

the mechanism of which is discussed. With some derivatives, restricted rotation around the B-N

bond and/or the B-C bonds was observed by NMR techniques, and some rotational barriers were

calculated from experimental data. The acidity of the sp³carbons in these complexes is shown by

their ability to protonate NEt₃, with formation of pyrrolyl- and indolyl-borate ammonium salts.

The driving force for this reaction is given by the restoration of the aromaticity of the heterocycle.

SCHEME 1

In tr od u ction

There has recently been a growing interest in the

chemistry of perfluoroarylboranes, due to both their high

Lewis acidity and chemical stability.^1-4

The Lewis acidity of B(C₆F₅)₃can be converted into

Brønsted acidity by complexation with water or alcohols

(Scheme 1). The complexity of the water-B(C₆F₅)₃system

has been recently investigated in detail.⁵The acidity of

boron-coordinated XOH has been demonstrated and

applied to catalytic reactions.^6,7

SCHEME 2

On the other hand, there is little known about aryl

borane/secondary amine complexes and of how the reac-

tivity of the N-bound proton would be affected by

coordination of the same nitrogen to a highly Lewis acidic

borane such as B(C₆F₅)₃. Recently, Klosin and co-workers

at Dow have reported the coordination of 2 equiv of

B(C₆F₅)₃to imidazole and the formation of delocalized,

stable bis[B(C₆F₅)₃]-imidazolate complexes, by reaction

with amines.⁸They also have shown that the pyrrole-

B(C₆F₅)₃complex, although neither isolated nor charac-

terized, is able to protonate triethylamine, with formation

of the [(pyrrolyl)B(C₆F₅)₃]^-[HNEt₃]⁺salt.

Erker has synthesized the 5H-pyrrole-B(C₆F₅)₃com-

plex, 1, by protonation of the Li salt of the [B(1-pyrrolyl)-

(C₆F₅)₃] borate (Scheme 2). The C5 methylene group of 1

is sufficiently acidic to protonate a CH₃of the metallocene

dimethyl catalyst with formation of methane, the [B(1-

pyrrolyl)(C₆F₅)₃] borate anion, and the metallocenium

methyl cation.⁹

^†Basell Polyolefins.

^‡Universita` di Bologna.

^§Dipartimento di Chimica Strutturale e Stereochimica Inorganica,

Universita` di Milano.

#

Dipartimento di Chimica Inorganica, Metallorganica e Analitica,

Universita` di Milano.

(1) (a) Massey, A. G.; Park, A. J . J . Organomet. Chem. 1964, 2, 245.

(b) Massey, A. G.; Park, A. J . J . Organomet. Chem. 1966, 5, 218.

(2) (a) J ordan, R. F. J . Chem. Educ. 1988, 65, 285. (b) J ordan, R. F.

Adv. Organomet. Chem. 1991, 32, 325. (c) Piers, W. E.; Chivers, T.

Chem. Soc. Rev. 1997, 26, 345. (d) Chen, E.; Marks, T. J . Chem. Rev.

2000, 100, 1391. (e) Bochmann, M. J . Chem. Soc., Dalton Trans. 1996,

255.

(3) J acobsen, H.; Berke, H.; Do¨ring, S.; Kehr, G.; Erker, G.; Fro¨hlich,

R.; Meyer, O. Organometallics 1999, 18, 1724.

(4) Chase, P. A.; Piers, W. E.; Patrick, B. O. J . Am. Chem. Soc. 2000,

122, 12911 and ref 13 therein.

(5) (a) Bergquist, C.; Bridgewater, B. M.; Harlan, C. J .; Norton, J .

R.; Friesner, R. A.; Parkin, G. J . Am. Chem. Soc. 2000, 122, 10581. (b)

Beringhelli, T.; Maggioni, D.; D’Alfonso, G. Organometallics 2001, 20,

4927.

(6) Ishihara, K.; Hanaki, N.; Funahashi, M.; Miyata, M.; Yamamoto,

H. Bull. Chem. Soc. J pn. 1995, 68, 1721.

(7) (a) Siedle, A. R.; Newmark, R. A.; Lamanna, W. M.; Huffman,

J . C. Organometallics 1993, 12, 1491. (b) Siedle, A. R.; Miller, J . A.;

Lamanna, W. M. WO 96/26967 to 3M, 1996.

(8) LaPointe, R. E.; Roof, G. R.; Abboud, K. A.; Klosin, J . J . Am.

Chem. Soc. 2000, 122, 9560; Chem. Eng. News 2000, Oct 9, 47.

(9) Kehr, G.; Roesmann, R.; Froehlich, R.; Holst, C.; Erker, G. Eur.

J . Inorg. Chem. 2001, 535.

Published on Web 06/18/2003

J . Org. Chem. 2003, 68, 5445-5465

5445

Guidotti et al.

SCHEME 3

CHART 1

SCHEME 4

CHART 2

Piers and co-workers recently reported the synthesis

and structural features of imine-B(C₆F₅)₃complexes,

notably the restricted rotation around the N-B bond.¹⁰

In an independent investigation, we have found that

1 and related B-N complexes can be obtained directly

by complexation between B(C₆F₅)₃and aromatic, five-

membered NH-heterocycles, such as indole and pyrrole,

by a new reaction that produces B-N complexes with

highly acidic sp³CH.¹¹In this paper we describe the

synthesis, characterization, and origin of the high acidity

of such complexes, which is unparalleled in known

borane-imine complexes.

from 360 to 271 K. At the lowest temperature, the C3

methylene gives rise to a clean AB-type spectrum that

clearly indicates restricted rotation around the B-N bond

and/or the B-C bonds. This behavior will be discussed

further on. The structure of 2 was confirmed by a single-

crystal X-ray analysis (see Solid State Structure section).

Both 1 and 2 are formed quantitatively and instanta-

neously, irrespective of the solvent or the order of

addition (see Experimental Section for details). It is worth

noting that 2 is stable in air for several days.

Resu lts a n d Discu ssion

Syn th esis a n d Ch a r a cter iza tion of Bor a n e-P yr -

r ole a n d Bor a n e-In d ole Com p lexes. Reaction of

pyrrole with 1 equiv of B(C₆F₅)₃in pentane, toluene, Et₂O,

or CH₂Cl₂gives quantitative and instantaneous conver-

sion to the 5H-pyrrole-B(C₆F₅)₃complex, 1 (Scheme 3).

The fact that complexation of pyrrole with B(C₆F₅)₃

does not give the tetrahedral B-NHC₄H₄complex 1′ is

Several other pyrrole- (Chart 1) and indole-borane

(Chart 2) complexes were synthesized in nearly quantita-

tive yields. 2,5-Dimethylpyrrole-B(C₆F₅)₃(3) also con-

1

demonstrated by its H NMR spectrum, which shows (in

1

tained ∼10% of the 3H-isomer 3′. The H NMR spectra

CD₂Cl₂) a diagnostic, broad singlet (2H) at 4.7 ppm for

the C5 methylene and three peaks for the three vinylic

protons at 8.6 (H2, m), 7.9 (H4, dq), and 6.9 (H3, dq) ppm.

NOESY experiments and single-crystal X-ray diffraction

confirmed the structure of complex 1 as being identical

to that reported by Erker and ruled out the formation of

the 3H-pyrrole-B(C₆F₅)₃complex. Therefore 1 can be

conveniently obtained by one-step direct complexation of

pyrrole and B(C₆F₅)₃, thus avoiding both previous depro-

tonation of the heterocycle and final protonation of

Erker’s [(1-pyrrolyl)B(C₆F₅)₃]Li reagent.⁹

of the pyrrole-B(C₆F₅)₃complexes 4-6, contrary to the

¹H NMR spectrum of 1, show a broad AB-system for the

methylene group in the R-position to the nitrogen atom,

indicating the presence of stereolabile conformational

enantiomers.

The reaction between B(C₆F₅)₃and 2-methyl- or 3-

methylindole cleanly gave the related 2-methyl- (7) and

3-methyl-3H-indole-B(C₆F₅)₃(8) complexes. Complex 7

is already “frozen” at room temperature, showing at 297

K a clear AB-type spectrum for the C3 methylene. At

about the same temperature (291 K), complex 8 clearly

shows two diastereoisomeric species. Other indole deriva-

tives (9-12) behave analogously.

BCl₃, which has Lewis acidity comparable to that of

B(C₆F₅)₃,^3,5,12reacts with indole to give 3H-indole-BCl₃

(13), the structure of which was assigned by analogy with

that of 2, due to the presence of the diagnostic meth-

ylene signal, in this case a singlet at 4.3 ppm (CD₂Cl₂).

Analogously, indole reacts with B(C₆F₅)₃in toluene,

Et₂O, or CH₂Cl₂to give quantitative and instan-

taneous conversion to the 3H-indole-B(C₆F₅)₃complex

2 (Scheme 4).

Unlike 1, the spectrum of which does not change even

1

when the temperature is lowered to 173 K, the H NMR

spectrum of 2 shows significant changes in the range

(10) Blackwell, J . M.; Piers, W. E.; Parvez, M.; McDonald, R.

Organometallics 2002, 21, 1400.

(11) Resconi, L.; Guidotti, S. Int. Pat. Appl. WO 01/62764 to Basell,

2001.

(12) (a) Beckett M. A.; Brassington, D. S.; Coles, S. J .; Hursthouse,

M. B. Inorg. Chem. Commun. 2000, 3, 530. (b) Do¨ring, S.; Erker, G.;

Fro¨hlich, R.; Meyer, O.; Bergander K. Organometallics 1998, 17, 2187.

5446 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

CHART 3 SCHEME 5

CHART 4

CHART 5

The reaction between B(C₆H₅)₃and indole did not

afford any adduct, due to the weak Lewis acidity of

triphenylborane; on the contrary, BF₃reacted with indole,

but no characterizable products were obtained.

3,3′-Dimethyl-2,2′-diindolylphenylmethane was syn-

thesized from 3-methylindole and benzaldehyde in nearly

quantitative yield by acid catalysis.¹³Reaction of BCl₃

(2 equiv) with the bisindolylmethane gave in ∼87% yield

compound 14: its ¹H NMR spectrum (CDCl₃) shows a

quartet at 4.2 ppm for the 3- and 3′-H protons and a

doublet at 1.0 ppm for the methyl groups. Its ¹³C NMR

(CDCl₃) confirmed the occurrence of the N-to-C hydrogen

shift showing a CH₃signal at 16.0 ppm and a CH signal

at 43.4 ppm for the C3 and C3′. On the contrary, the

reaction between B(C₆F₅)₃(2 equiv) and 3,3′-dimethyl-

2,2′-diindolylphenylmethane did not afford any adduct,

probably due to the steric hindrance of the phenyl group,

which blocks the borane approach.

Attempts were also carried out to synthesize the

7-methylindole and the 2-phenylindole adducts

(Chart 3). The reaction of the former with B(C₆F₅)₃in

dichloromethane is slow. After 5 min at room tempera-

ture, a 1.4:1 mixture of the expected product 15 and

starting material was formed together with lower amounts

of byproducts. The lower conversion rate observed is

probably due to the steric hindrance of the methyl group

on C7, which hinders B(C₆F₅)₃coordination to nitrogen.

Consumption of 7-methylindole proceeds together with

the decomposition of the expected adduct to give many

unidentified byproducts. Similarly, 2-phenylindole reacts

with B(C₆F₅)₃, yielding a ca. 1:1 mixture of the 2-phe-

nylindole-B(C₆F₅)₃complex 16 and starting material.

This product/reagent ratio does not change for 24 h, and

attempts to shift the equilibrium to the product by adding

excess borane failed.

generates a quite strong acidity of the proton(s) on the

sp³C5 or C3 carbon: for example, 1 and 2 react with

NEt₃to give quantitatively the known [B(1-pyrrolyl)-

(C₆F₅)₃]^-[HNEt₃]⁺salt 1a ⁸and the [B(1-indolyl)(C₆F₅)₃]^--

[HNEt₃]⁺salt 2a , respectively (Scheme 5). This reaction

can be used to probe the acidity of the borane-imine

complexes.

In addition to 1a and 2a , we obtained the salts [B(2,5-

dimethyl-1-pyrrolyl)(C₆F₅)₃]^-[HNEt₃]⁺(3a ), [B(2-ethyl-1-

pyrrolyl)(C₆F₅)₃]^-[HNEt₃]⁺(5a ), [B(2-methyl-1-indolyl)-

(C₆F₅)₃]^-[HNEt₃]⁺(7a ), [B(3-methyl-1-indolyl)(C₆F₅)₃]^--

[HNEt₃]⁺(8a ), [B(5-methoxy-1-indolyl)(C₆F₅)₃]^-[HNEt₃]⁺

(9a), [B(2-methylindeno[2,1-b]indol-5-yl)(C₆F₅)₃]^-[HNEt₃]⁺

(12a ), [B(1-indolyl)Cl₃]^-[HNEt₃]⁺(13a ), [B(1-imidazolyl)-

(C₆F₅)₃]^-[HNEt₃]⁺(17a ), and [B(1-pyrazolyl)(C₆F₅)₃]^--

[HNEt₃]⁺(18a ).

The pyrrolidine- (19) and indoline-B(C₆F₅)₃(20)

complexes were originally prepared to check whether the

H-shift is necessary to generate proton acidity in the

aromatic N-heterocycle-B(C₆F₅)₃complexes. As expected,

neither 19 nor 20 is able to protonate Et₃N, indicating

that, indeed, simple B-N coordination is not sufficient

to impart acidity to the N-H bond. On the other hand,

steric shielding by the borane cannot justify the lower

acidity, because it is not likely to be sufficient to prevent

deprotonation of the adduct by the small amine. In

addition, also 2-methyl-1-pyrroline-B(C₆F₅)₃21 does not

react with NEt₃: this shows that one double bond is not

sufficient for inducing proton acidity (Chart 5).

The strong acidity of the sp³CH group in 1- and 2-like

adducts is therefore very likely induced by the re-

aromatization of the heterocyclic ring and localization of

the negative charge on the boron atom, upon formation

of the anion. As a consequence, the resulting borate

anions are quite stable and are not expected to interact

with a strong Lewis acid, such as a metallocene alkyl

cation (other than, possibly, by a labile metal-F dipolar

interaction). This is also confirmed by a previous inves-

tigation by Erker, who has shown that Cp₂ZrMe(NC₄H₄)

transfers the pyrrolyl anion to B(C₆F₅)₃, rather than

the CH₃group, and that the resulting methylzirconoce-

nium cation is an active ethylene polymerization cata-

lyst.¹⁵

A few experiments were carried out with heterocycles

containing two nitrogen atoms (Chart 4). The reaction

of B(C₆F₅)₃with imidazole and pyrazole gave the known

17¹⁴and the new 18, respectively, as expected on the

basis of the higher basicity of the iminic nitrogen with

respect to the NH moiety.

In the pyrrole (1-like) and indole (2-like) adducts, the

presence of the NdC double bond combined with that of

the strong electron-withdrawing B(C₆F₅)₃Lewis acid

(13) (a) Pindur, U. Arch. Pharm. (Weinheim) 1980, 313, 790. (b)

Pindur, U.; Dittmann, K. Arch. Pharm. (Weinheim) 1985, 318, 340.

(c) Pindur, U.; Dittmann, K. Heterocycles 1986, 24, 1079.

(14) Ro¨ttger, D.; Erker, G.; Fro¨hlich, R.; Kotila, S. J . Organomet.

Chem. 1996, 518, 17.

J . Org. Chem, Vol. 68, No. 14, 2003 5447

Guidotti et al.

fluorine atoms of the borane moiety. Analogous features

1

can be observed in the H NMR spectra of adduct 7, the

methylene protons of which, even at room temperature,

give rise to an unambiguous AB spectrum (2.59 ppm),

which coalesces only above 381 K (see Supporting Infor-

1

mation). In addition, at 239 K, the H NMR spectrum of

adduct 8 clearly shows the presence of two diastereomeric

species in a 55:45 ratio, which are evidenced by the two

overlapped quartets arising from the H3 proton (4.23 and

4.36 ppm) as well as the two doublets originated by the

methyl group (1.53 and 1.65 ppm); these signals begin

to coalesce above room temperature (see Supporting

Information). As for compound 2, a proton-fluorine

through-space coupling gives rise at 297 K, for the H2

proton, to a doublet (8.68 ppm) that transforms into a

triplet on cooling to 239 K.

As far as the pyrrole adducts are concerned, compounds

4 and 5 (see Supporting Information) behave like the

indole analogues. Indeed, once again, anisochronous

methylene signals can be observed on cooling, and

decoalescence of the methylene singlet into an AB-type

spectrum clearly occurs at 239 K for both adducts.

Compound 5 additionally shows decoalescence of the

ethyl signals into an ABX₃spectrum (0.70-2.87 ppm),

due to diastereotopicity of the methylenic protons of the

1

ethyl group. It is worth noting that the H NMR spectra

of the parent compound 1 did not change at all when the

temperature was varied in the range of 173-353 K.

The spectroscopic behavior of adducts 2, 4, 5, 7, and 8

can be explained by assuming that these compounds are

chiral molecules: this hypothesis was confirmed by the

X-ray structure of 2 (see Solid State Structures section),

which clearly shows the lack of a plane of symmetry

because of the spatial arrangement of the fluorinated

rings of the borane moiety. Compound 2 and its ana-

logues therefore exist as pairs of stereolabile conforma-

tional enantiomers (atropisomers) that can interconvert

through rotation about the B-N and/or the B-C bonds.

When the interconversion barriers are sufficiently large,

that is, the bond rotations are slow enough, the two

enantiomeric species can be detected by NMR techniques

through the diastereotopicity of the methylene protons

(2, 4, 5, and 7). If the molecule contains a further

asymmetry element, two stereolabile diastereoisomers

are produced. In principle, they could form in nonequal

amounts and should be spectroscopically detected as

distinct species: indeed, due to its asymmetric C3 atom,

this is exactly what was observed for adduct 8.

F IGURE 1. ¹H NMR spectra of compound 2 at 360, 319, 297,

and 271 K in C₂D₂Cl₄.

Con for m a tion a l Stu d ies of In d ole-B(C₆F ₅)₃a n d

P yr r ole-B(C₆F ₅)₃Ad d u ct s b y Dyn a m ic NMR in

1

C₂D₂Cl₄. Figure 1 shows the H NMR spectra of adduct

The dynamic behavior of congested N-bound imine-

B(C₆F₅)₃adducts due to restricted rotation about the

B-N and B-C bonds has been recently evidenced by

Piers and cowokers using variable-temperature ¹⁹F and

¹H NMR spectroscopy.¹⁰

2 recorded at various temperatures. For this compound,

anisochronous methylene signals can be observed on

cooling, so that the singlet recorded for the methylene

protons at 319 K (4.19 ppm) decoalesces below 271 K into

an AB-type spectrum. At the same time, the singlet due

to H2 (8.73 ppm) changes first into a doublet and finally

into a triplet when the temperature is lowered from 319

to 297 and 271 K, respectively. A COSY experiment

carried out on compound 2 showed that there is no

coupling between H2 and the two geminal H3 protons;

therefore, the AB spectrum must be ascribed to the

diastereotopicity of the methylene protons, whereas the

multiplicity of H2 can be explained by spin-spin through-

space couplings of H2 with one (297 K) or two (271 K)

The interconversion of the two enantiomers was quan-

titatively studied for adduct 2. Simulation (Figure 2,

1

right) of the H NMR spectra in C₂D₂Cl₄(Figure 2, left)

of 2 at various temperatures allowed determination of

the kinetic constant of bond rotation, and thence of the

enantiomerization process, at each temperature. As one

can see, at 360 K the rotation is practically free (k > 10⁵

s^-1), whereas at 271 K the two enantiomeric conformers

do not interconvert in the NMR time scale. In the cases

of the two intermediate temperatures (319 and 297 K)

values of 700 and 41 s^-1were found, respectively, from

(15) Temme, B.; Erker, G. J . Organomet. Chem. 1995, 488, 177.

5448 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

F IGURE 2. (Left) Experimental ¹H NMR signal of the

methylene hydrogens of 2 as a function of temperature. (Right)

Computer simulations obtained with the rate constants indi-

cated.

which we calculated a value of 14.9 ( 0.2 kcal mol^-1for

the activation barrier (∆G^q) of the enantiomerization.

The barrier is probably originated by the steric hin-

drance experienced by the 2- and, particularly, 7-hydro-

gen during transit of the fluorinated rings. In the case

of 7, the methyl group in the 2-position is an additional

obstacle for rotation of the borane moiety and the

enantiomerization barrier rises to a significant extent

(∆G^q) 18.4 ( 0.2 kcal mol^-1, calculated by line-shape

simulation as above). The barrier was also likewise

determined for the pyrrole adduct 4. The calculated value

(∆G^q) 14.5 ( 0.2 kcal mol^-1) is more comparable to that

of compound 2: this let us conclude that the very high

barrier observed for adduct 7 is the result of the overall

steric hindrance due to the simultaneous presence of

(C2)Me and H7.

F IGURE 3. ¹H NMR spectra at variable temperature of 2

(C₇D₈, 7.1 T, / ) solvent signals, # ) impurity).

Str u ctu r e a n d Dyn a m ics in Tolu en e Solu tion of

1

Com p ou n d 2. In this solvent the H spectrum is similar

to the one in C₂D₂Cl₄, but a substantial low-field shift of

hydrogen H7 of the six-membered ring has been observed

(see Experimental Section and Supporting Information,

Figures S22 and S23, for the assignments) (Figure 3).

At further variance with C₂D₂Cl₄, a significant tem-

perature dependence of the chemical shifts was observed

for H7, H2, and the two diastereotopic methylenic (H3/

H3′) resonances. These latter signals, as in C₂D₂Cl₄, with

rising temperature broaden and coalesce above 313 K

(Figure 3). The proton NMR alone cannot discriminate

among the many dynamic processes that could give rise

to this behavior. A ¹⁹F NMR study was therefore per-

formed to obtain more detailed information.

F IGURE 4. ¹⁹F NMR spectra of 2 at variable temperature

(C₇D₈, 7.1 T, # ) impurity). The signals are labeled according

to Figure 5.

At 300 K, the 1D ¹⁹F NMR spectrum showed six

resonances in the ortho region (some accidental overlaps

occur in the para and meta regions), indicating a C₁

symmetry for compound 2. The variable-temperature

spectra (Figure 4) confirmed the occurrence of dynamic

processes. However, the resonances were not homoge-

neously broadened (e.g., at 332 K in the ortho region two

signals were still sharper than the others). This indicates

that rotation around the B-N bond is not the dominant

J . Org. Chem, Vol. 68, No. 14, 2003 5449

Guidotti et al.

F IGURE 6. Selected regions of 2D ¹⁹F-¹H correlated experi-

ments on compound 2 (C₇D₈, 7.1 T, 265 K, / ) impurity): (a)

scalar correlation experiment optimized for J _HF) 10 Hz (64

fid, SW 16560 Hz, TD 4K, D1 ) 1.5 s, 128 experiments, zero-

filled once in F1, weighting functions: squared shifted sine

bell in F2 and sine in F1); (b) dipolar correlation experiment

(τ_m) 0.55 s, 128 fid, SW 16560 Hz, TD 4K, D1 ) 1.5 s, 128

experiments, zero-filled once in F1, weighting functions:

squared shifted sine bell in F2 and sine in F1).

respect to the three phenylic rings came from the

observed scalar and dipolar coupling between some of the

ortho fluorines and the H2 or H7 hydrogen atoms (Figure

6), whereas the occurrence of through-space couplings¹⁶

and dipolar correlations between ortho fluorines belong-

ing to different rings probed the relative conformation

of the aryl rings (Figure 7). All of the spatial relationships

found from these experiments are indicated in Figure 5.

To take into account all of the experimental scalar and

dipolar interactions, the three phenyl rings should have

different tilts with respect to the heterocyclic moiety, as

sketched in Figure 5. Such a conformation is similar to

the one established in the solid state (see below) for

compound 2 and related derivatives.

Two-dimensional ¹⁹F-¹⁹F EXSY experiments showed

that the most relevant enantiomerization pathway of

compound 2, occurring already at 275 K, is the exchange

of rings B and C, whereas ring A maintains its identity

(Figure 8a). It is worth noting that in this process ring A

does not flip: in fact, the 2D maps do not show direct

exchange between the ortho fluorines atoms of A, and in

the ¹H variable-temperature spectra H2 (that is, a triplet

at low temperature for its interaction with A2 and B6

fluorine atoms) becomes a doublet, albeit broadened, up

to 313 K, indicating the persistence of the scalar H-F

correlation with A2. Likely, the B/C interconversion is

accomplished through the libration of the heterocycle and

of ring A with respect to the B-N and B-C bonds,

respectively, whereas rings B and C rotate as and

opposite to ring A, respectively.

F IGURE 5. Solution structure of compound 2 as derived from

2D ¹⁹F-¹⁹F and ¹⁹F-¹H correlation experiments. The experi-

mental dipolar and scalar interactions are indicated with

dotted lines.

TABLE 1. Assign m en ts of th e ¹⁹F Reson a n ces of

Com p ou n d 2 a t 233 K (δ)

ring

A

ortho

para

meta

-129.76(A2)

-131.65(A6)

-133.10(B2)

-135.25(B6)

-129.51(C2)

-134.46(C6)

-154.84(A4)

-161.35(A5)

-162.59(A3)

-162.34(B3)

-163.95(B5)

-163.54(C3)

-163.46(C5)

B

C

-154.63(B4)

-156.10(C4)

At 278 K 2D ¹⁹F-¹⁹F EXSY experiments showed

unambiguous much weaker exchange cross-peaks among

the ortho signals of rings A and B (but no cross-peaks

among ortho signals of A and C and between those of A)

dynamic process, because it would interconvert all three

rings and therefore affect all of the resonances to the

same extent.

The solution structure of compound 2, as established

through low-temperature 2D ¹⁹F-¹⁹F and ¹⁹F-¹H scalar

and dipolar correlation experiments (see Supporting

Information Figures S24, S25, and S26), is shown in

Figure 5, and the assignments of the fluorine resonances

of the three rings are reported in Table 1. Clues to

establish the orientation of the heterocyclic molecule with

(16) For other experimental examples of through-space ¹⁹F-¹⁹F

couplings see, for instance: (a) Mallory, F. B.; Mallory, C. W.; Butler,

K. E.; Lewis, M. B.; Xia, A. Q.; Luzik, E. D., J r.; Fredenburg, L. E.;

Ramanjulu, M. M.; Van, Q. N.; Francl, M. M.; Freed, D. A.; Wray, C.

C.; Hann, C.; Nerz-Stormes, M.; Carroll, P. J .; Chirlian, L. E. J . Am.

Chem. Soc. 2000, 122, 4108. (b) Ernst, L.; Ibrom, K.; Marat, K.;

Mitchell, R. H.; Bodwell, G. J .; Bushnell, G. W. Chem. Ber. 1994, 127,

1119.

5450 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

F IGURE 7. (a) Ortho region of a 2D ¹⁹F COSY45 experiment

(C₇D₈, 7.1 T, 233 K, 8 fid, SW 14120 Hz, TD 2K, D1 ) 0.5 s,

512 experiments, zero-filled to 1 K in F1, weighting func-

tions: squared shifted sine bell in F1 and sine in F2) showing

“through-space” correlations between fluorines belonging to

different rings (namely, C2 with A6 and B2; A2 with C6), and

one “through-bond” cross-peak (B2 T B6). (b) Ortho region of

a 2D ¹⁹F NOESY experiment (C₇D₈, 7.1 T, 233 K, τ_m) 300

ms, 16 fid, SW 14120 Hz, TD 2K, D1 ) 0.5 s, 512 experiments,

zero-filled to 1 K in F1, weighting functions: shifted sine bell

in both dimensions) showing dipolar correlations between

different rings (cross-peaks of opposite sign with respect to

diagonal peaks).¹⁷

F IGURE 8. Ortho region of a 2D EXSY experiment performed

on compound 2 (C₇D₈, 7.1 T, 16 fid, SW 12820 Hz, TD 2K, D1

) 1 s, 512 experiments, zero-filled to 2K in F1, weighting

functions: shifted sine bell in both dimensions): (a) 275 K and

τ_m) 100 ms, showing the exchange of ring C and B (C2/B2,

C2/B6, B2/C6, C6/B6, and their symmetry-related cross-peaks

are of the same sign as the diagonal peaks); (b) 278 K and τ_m

) 100 ms, showing the exchange cross-peaks between rings A

and B.

of various 2D EXSY experiments in the range of 275-

300 K gave the rate constants for the low-energy and the

high-energy processes (Table S3; Figure S28). The activa-

tion parameters calculated accordingly are E_a) 16.7(1)

kcal mol^-1, ∆H^#) 16.0(1) kcal mol^-1, and ∆S^#) 2.2(1)

cal mol^-1K^-1for B/C exchange and E_a) 18.0(2), ∆H^#)

17.4(2) kcal mol^-1, and ∆S^#) 1.7(1) cal mol^-1K^-1for A/B

exchange. The rate constants obtained from the simula-

tion of ¹H spectra account for both of the processes;

(Figure 8b). The occurrence of this second slower process

involving ring A leads eventually to the overall exchange

of all the ortho resonances as observed in a 2D experi-

ment at 300 K and τ_m) 100 ms (Figure S27). At 300 K,

the kinetic constants of these processes are 39 and 3 s^-1

respectively.

,

Simulation of the para region of the 1D ¹⁹F spectra in

the range from 283 to 350 K together with the analysis

J . Org. Chem, Vol. 68, No. 14, 2003 5451

Guidotti et al.

F IGURE 10. ORTEP view of the structure of the (1H-indol-

1-yl)tris(pentafluorophenyl)borate anion of 2a in the crystal

of its triethylammonium salt. Thermal ellipsoids are drawn

at the 30% probability level; hydrogen atoms are given

arbitrary radii.

F IGURE 9. ORTEP view of the structure of (3H-indole)tris-

(pentafluorophenyl)boron (2) in the crystal. Thermal ellipsoids

are drawn at the 30% probability level; hydrogen atoms are

given arbitrary radii.

however, the rate of the second process being 1 order of

magnitude smaller, their values and the activation

parameters are the same as those obtained from ¹⁹F

(Figure S28). The dynamic behavior of congested N-bound

imine-B(C₆F₅)₃adducts due to restricted rotation about

the B-N and the B-C bonds has been recently evidenced

by Piers and co-wokers using variable-temperature ¹⁹F

1

and H NMR spectroscopy.¹⁰These imine adducts exhibit

(at least in the solid state) conformations of the B-bound

phenyl rings very similar to that observed in the present

case (both in solution and in the solid state, see below).

However, the enantiomerization activation parameters

found for 2 are slightly larger than those reported by

Piers et al. (∆G^#∼ 66 vs 60 kJ mol^-1) despite the π

stacking interactions between fluorinated and hydroge-

nated phenyl rings observed in their compounds. Likely

in compound 2 the overall H-F interactions help in

maintaining the preferred conformation of the perfluori-

nated rings and, indeed, the low-energy enantiomeriza-

tion process in 2 implies the cleavage of most of these

H‚‚‚F interactions (all of those involving H7 and one of

the two involving H2).

Solid Sta te Str u ctu r es of Com p lexes betw een

In d oles a n d B(C₆F ₅)₃. The molecular structures of 2,

2a , and 20 have been determined in the solid state by

X-ray diffraction analysis. Figures 9, 10, and 11 show

ORTEP views of 2, the indolate anion of 2a , and 20,

respectively. Tables 2 (distances) and 3 (angles) contain

the most relevant bonding parameters for the three

compounds.

F IGURE 11. ORTEP view of the structure of (2,3-dihydro-

1H-indole)tris(pentafluorophenyl)boron (20) in the crystal.

Thermal ellipsoids are drawn at the 30% probability level;

hydrogen atoms are given arbitrary radii.

tetrahedral coordination geometry, in which four bond

angles are larger and two (opposite) are smaller than the

idealized value of 109.5°. The nitrogen atoms in both 2

and 2a have a trigonal planar coordination geometry, as

indicated by the sum of bond angles at N(1) (359.6 and

358.7° for 2 and 2a , respectively), whereas in 20 the

nitrogen atom is tetrahedrally coordinated.

Compound 2 is a rare example of the 3H-indole

tautomeric form.¹⁸As expected, the observed bond dis-

tances are in accordance with a localization of the double

bond between N(1) and C(2) [1.284(3) Å] in the five-

membered ring.¹⁹

The formulation of 2, 2a , and 20 as equimolecular

adducts between the indoles and tris(pentafluorophenyl)-

borane is confirmed. The borane and the heterocyclic

moieties are connected by a strong B-N bond. In all three

compounds the boron atom shows a distorted (pseudo D_2d)

In complex 2a the bond distances within the hetero-

cyclic moiety are very similar to that observed in the free

indolate anion²⁰and are consistent with an extensive

delocalization of the π electrons in the rings.

5452 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

TABLE 2. Selected Bon d Dista n ces (Å) for 2, 2a , a n d 20

TABLE 4. Selected F ‚‚‚F a n d F ‚‚‚H Dista n ces (Å) for 2^a

F(22) F(26) F(32) F(36) H(2) H(7)

2

2a

20

N(1)-C(2)

N(1)-C(7a)

N(1)-B(1)

C(2)-C(3)

C(3)-C(3a)

C(3a)-C(4)

C(3a)-C(7a)

C(4)-C(5)

C(5)-C(6)

C(6)-C(7)

C(7)-C(7a)

B(1)-C(11)

B(1)-C(21)

B(1)-C(31)

1.284(3)

1.447(3)

1.613(3)

1.471(3)

1.489(3)

1.376(3)

1.386(3)

1.372(4)

1.373(4)

1.390(3)

1.374(3)

1.641(3)

1.647(3)

1.641(3)

1.374(4)

1.390(4)

1.565(4)

1.355(4)

1.413(5)

1.407(5)

1.421(4)

1.374(6)

1.403(6)

1.374(5)

1.398(4)

1.663(4)

1.661(4)

1.662(4)

1.530(2)

1.478(2)

1.650(2)

1.525(3)

1.495(3)

1.387(3)

1.379(3)

1.378(4)

1.381(4)

1.386(3)

1.376(3)

1.649(3)

1.642(3)

1.651(2)

F(12) 4.650(3) 5.763(4) 4.719(3) 2.796(2) 2.61 4.98

F(16) 2.850(2) 4.429(3) 2.940(2) 5.947(4) 4.45 5.62

F(22)

F(26)

F(32)

F(36)

4.748(3) 5.599(4) 2.84 4.85

2.784(3) 4.629(3) 5.70 2.48

5.48 4.38

3.97 3.10

a

Close contacts are indicated in bold (the sums of the van der

Waals radii for the F‚‚‚F and F‚‚‚H contacts are 3.0 and 2.7 Å,

respectively).

Boron-nitrogen bond distances show a sensible de-

pendence on the electronic environment: the longer

[1.650(2) Å] is, as expected, the one involving the sp³

hybridized nitrogen atom in 20. In complexes 2 and 2a ,

in which the nitrogen atoms are sp²hybridized, the B-N

distances are 1.613(3) and 1.565(4) Å, the shorter one

involving the negatively charged indolate moiety (which

is a better Lewis base than the neutral 3H-indole

tautomer). The same trend has been reported for the

pyrrole derivatives 2H-pyrrole-B(C₆F₅)₃(1) and [1H-

pyrrol-1-yl-B(C₆F₅)₃]^-(1a ), in which the B-N distances

are 1.608(2) and 1.576(4) Å, respectively.⁹

As can be seen from Figures 9-11 and from the

torsional angles reported in Table 3, in the three com-

plexes the B(C₆F₅)₃moiety adopts a very similar confor-

mation. In particular, one of the three phenyl rings [the

one labeled C(31)-C(36)] eclipses the B-N bond, whereas

the other two exhibit a chiral two-bladed propeller-like

conformation. This very same conformation can be found

in most of the crystal structures of tris(pentafluorphenyl)-

borane derivatives collected in the Cambridge Structural

Database:²¹from this evidence it can be inferred that this

molecular fragment is a very rigid one, so that the

presence of a somehow high-energy barrier to the enan-

tiomerization process of the chiral propeller can be

expected.

This conformation results in some short F‚‚‚F non-

bonded contacts between the ortho-fluoro substituents of

the three phenyl rings and in some F‚‚‚H contacts

involving the hydrogen atoms bound to C(2) and C(7). A

list of these interactions for compound 2 (in which they

can also be evidenced in solution, at low temperature,

by NMR spectroscopyssee previous section) is given in

Table 4. Moreover, in compound 20 the hydrogen atom

bound to nitrogen is involved in a bifurcated intramo-

lecular hydrogen bond with the fluorine atoms F(12) and

F(36) (the two NH‚‚‚F distances are 2.15 Å).

TABLE 3. Selected Bon d a n d Tor sion a l An gles

(Degr ees) for 2, 2a , a n d 20

2

2a

20

C(2)-N(1)-C(7a)

C(2)-N(1)-B(1)

C(7a)-N(1)-B(1)

N(1)-C(2)-C(3)

108.28(18)

126.71(18)

124.61(16)

113.4(2)

101.82(18)

132.9(2)

107.71(19)

119.4(2)

118.5(2)

121.5(2)

121.1(2)

116.4(2)

108.77(18)

128.24(19)

123.0(2)

109.75(16)

101.32(16)

111.53(17)

114.27(17)

104.38(17)

115.71(17)

120.99(18)

125.69(18)

113.01(18)

119.31(18)

127.84(19)

112.8(2)

118.29(18)

127.39(18)

113.9(2)

53.3(3)

67.3(2)

106.3(2)

128.0(2)

124.4(2)

111.6(3)

107.3(3)

135.2(3)

106.2(3)

118.5(3)

119.6(4)

120.7(3)

121.7(4)

117.9(3)

108.5(3)

129.9(3)

121.6(3)

112.0(2)

103.1(2)

112.9(2)

111.5(2)

103.0(2)

114.5(2)

120.2(3)

126.1(3)

113.3(3)

118.4(3)

128.8(3)

112.6(3)

119.0(3)

127.5(3)

113.2(3)

55.9(4)

103.32(13)

113.17(13)

117.72(13)

106.28(15)

102.47(16)

129.4(2)

111.30(16)

119.2(2)

118.6(2)

C(2)-C(3)-C(3a)

C(3)-C(3a)-C(4)

C(3)-C(3a)-C(7a)

C(4)-C(3a)-C(7a)

C(3a)-C(4)-C(5)

C(4)-C(5)-C(6)

121.3(2)

120.9(2)

116.9(2)

C(5)-C(6)-C(7)

C(6)-C(7)-C(7a)

N(1)-C(7a)-C(3a)

N(1)-C(7a)-C(7)

C(3a)-C(7a)-C(7)

N(1)-B(1)-C(11)

N(1)-B(1)-C(21)

N(1)-B(1)-C(31)

C(11)-B(1)-C(21)

C(11)-B(1)-C(31)

C(21)-B(1)-C(31)

B(1)-C(11)-C(12)

B(1)-C(11)-C(16)

C(12)-C(11)-C(16)

B(1)-C(21)-C(22)

B(1)-C(21)-C(26)

C(22)-C(21)-C(26)

B(1)-C(31)-C(32)

B(1)-C(31)-C(36)

C(32)-C(31)-C(36)

N(1)-B(1)-C(11)-C(12)

N(1)-B(1)-C(21)-C(22)

N(1)-B(1)-C(31)-C(32)

110.67(16)

126.20(18)

123.12(18)

110.85(13)

104.01(13)

108.39(13)

116.66(14)

102.39(13)

114.49(14)

121.08(15)

125.00(15)

112.83(16)

122.25(16)

124.12(16)

113.45(17)

118.95(14)

127.69(15)

113.10(15)

47.2(2)

60.9(3)

176.0(2)

77.00(19)

170.43(14)

169.98(18)

The heterocyclic moiety in the 2,3-dihydro-1H-indole

complex 20 does not present any unusual feature. The

pentagonal ring assumes an envelope conformation on

C(2).

P ossible Mech a n ism of F or m a tion of Bor a n e-

P yr r ole a n d Bor a n e-In d ole Com p lexes. The forma-

tion of the B-N adducts can be considered as the reaction

(17) In the ortho region, NOE cross-peaks among resonances of

different rings supported the occurrence of the through-space couplings

observed in the COSY spectrum. The correlation between A6 and B6,

observable only in the NOESY spectrum, indicates that likely the

closeness in the space of these two fluorine atoms does not give rise to

an effective overlap of fluorine p orbitals, as required for through-space

coupling.¹⁶On the other hand, the strongly scalar coupled C2/B2

signals do not show NOE cross-peaks, likely for the occurrence of

double and zero quantum J modulation; see: Ernst, R. R.; Boden-

hausen, G.; Wokaun, A. In Principles of Nuclear Magnetic Resonance

in One and Two Dimensions; Oxford Science Publications: Oxford,

U.K., 1987; Chapter 9.

(19) The observed bond distances fully agree with the ones predicted

for the 3H-indole tautomer by density functional computation (B3LYP/

cc-pVDZ): Dubnikova, F.; Lifshitz, A. J . Phys. Chem. A 2001, 105,

3605.

(20) Bond distances (Å) for the 7-methylindolate anion are as

follows: N(1)-C(2), 1.383(4); N(1)-C(7a), 1.392(4); C(2)-C(3), 1.390-

(5); C(3)-C(3a), 1.420(5); C(3a)-C(4), 1.409(5); C(3a)-C(7a), 1.435-

(4); C(4)-C(5), 1.370(6); C(5)-C(6), 1.402(6); C(6)-C(7), 1.385(5); C(7)-

C(7a), 1.406(5). Karl, M.; Harms, K.; Seybert, G.; Massa, W.; Fau, S.;

Frenking, G.; Dehnicke; K. Z. Anorg. Allg. Chem. 1999, 625, 2055.

(21) CSD version 5.22 (Oct 2001). Allen, F. H.; Davies, J . E.; Galloy,

J . J .; J ohnson, O.; Kennard, O.; Macrae, C. F.; Mitchell, C. F.; Mitchell,

G. F.; Smith, J . M.; Watson, D. G. J . Chem. Inf. Comput. Sci. 1991,

31, 187.

(18) To the best of our knowledge, the crystal structure of only

another derivative of this tautomer has been reported, namely, the

complex [PdCl₂(2-methyl-3H-indole)₂]. Yamauchi, O.; Takani, M.;

Toyada, K.; Masuda, H. Inorg. Chem. 1990, 29, 1856.

J . Org. Chem, Vol. 68, No. 14, 2003 5453

Guidotti et al.

SCHEME 6

SCHEME 7

between a nucleophile (the heterocycle) and an electro-

phile (the borane). The attack of electrophiles on the 2-

or 3-position of pyrrole and indole, respectively, is a well-

known process.²²Nevertheless, formation of our pyrrole

and indole adducts cannot be accounted for simply by the

intervention of the borane as an electrophile, because it

would not be easy to explain the subsequent, concomitant

migrations of the N-H bond and the borane moiety

between nitrogen and carbon.

In the case of pyrrole, a simple explanation could be

coordination of the borane to nitrogen followed by 1,2

intra-(or inter-)molecular translocation of the N-H hy-

drogen in the adduct 1′ (Scheme 6). The latter step can

be promoted by a polarization of the N-H bond induced

by the solvent or a fluorine atom. Supporting this

hypothesis, the X-ray structure of the indoline-B(C₆F₅)₃

complex 20 shows two strong intramolecular NH-F_ortho

interactions (d_H-F) 2.15 Å) (see Solid State Structures

section). However, this mechanism cannot effectively

justify the formal 1,3-H shift observed with indole

derivatives. Furthermore, direct coordination of B(C₆F₅)₃

to the nitrogen atom of pyrrole or indole is unlikely, given

the aromaticity of the heterocyclic ring. Trying to answer

these questions, we have carried out DFT calculations

in the gas phase. They indicate that, taking the energy

of the reactants at infinity as a reference, adduct 1′ is

stabilized only by 0.9 kcal mol^-1, despite a rather low-

energy barrier of 3.4 kcal mol^-1needed for its formation.

Taking into account possible relatively large entropic

effects, the formation of adduct 1′ can be reasonably

thought to be unlikely.

k₂₃. k₂₁), by assuming the steady state approximation

on 3H-indole, we can estimate that d[2]/dt ) k₁₂[indole],

with k₁₂≈ 10^-24s^-1at 25 °C. Although the inclusion of

solvent effects may reduce the calculated barrier for

indole tautomerization, we believe that an alternative

mechanism may be involved. In particular, because the

tautomerization reaction of both pyrrole and indole is

catalyzed by acids,²⁵we make the hypothesis that B-N

adducts can be more conveniently accounted for through

nucleophilic attack of boron-coordinated water to a pyr-

role (or indole) double bond, followed by reorganization.

The coordination of B(C₆F₅)₃with water is very fast. Any

experiment aimed at having water-free borane showed

the presence of residual water in amounts depending on

the purification process used (see Experimental Section

for details). Because it is practically impossible to have

fully water-free B(C₆F₅)₃, the water-catalyzed reaction

would seem to be a likely possibility and is in accordance

with the different structures of pyrrole (electrophilic

attack on C_R) and indole (electrophilic attack on C_â)

derivatives. After protonation of the heterocycle, the

resulting species can evolve to the final adduct by

intramolecular elimination of water (path a of Scheme

8, shown for indole) or by attack of a non-coordinated

borane molecule and concomitant elimination of H₂O-

B(C₆F₅)₃(path b of Scheme 8), which hence acts as a

catalyst.

Moreover, calculations aimed at evaluating the possible

assistance of the ortho-F to the 1,2-H shift (step 1′-1 in

Scheme 6) indicated that the fluorine atom should not

play a role, at least in the early steps of this process.

Indeed, only after the system has overcome an energy

barrier of nearly 35 kcal mol^-1is a favorable H-F

interaction observed. Finally, test calculations on the step

1′-1 performed on the simplified system BF₃-pyrrole

and C₆F₆, aiming at evaluating the assistance of a F atom

by a further incoming B(C₆F₅)₃molecule, gave no results.

A simple explanation for the formation of 1 would be

the presence of minor amounts of the 3H-indole and 5H-

pyrrole tautomers at equilibrium in solution, which would

be immediately trapped by the borane, thus driving the

equilibrium to the right (Scheme 7). However, the

computed intramolecular isomerization barrier in the gas

phase from 1H-indole to 3H-indole is very high, on the

DFT calculations showed that formation of species 22

could be a viable process. Indeed, both indole and pyrrole

form a preliminary adduct with H₂O-B(C₆F₅)₃, stabilized

by 2.5 and 6.4 kcal mol^-1, respectively, with respect to

the reactants. The proton-transfer process has been

investigated for pyrrole only. The transition state to

species 22_{p yr r}from the initial adduct is at 6.2 kcal mol^-1

;

22_{p yr r}and 22_{in d}are at +0.14 and -11.4 kcal mol^-1with

respect to the adducts, respectively. The optimized

structures (Figure 12) display a bridging hydrogen atom

between nitrogen and oxygen.

Both species 22_{p yr r}and 22_{in d}are nearly 23 kcal mol^-1

more stable than the isolated H₂O-B(C₆F₅)₃adduct and

5H-pyrrole and 3H-indole, respectively. Moreover, 22_{p yr r}

is 15.1 and 8 kcal mol^-1more stable than the two limit

species in which the bridging H atom is at bond distance

only with N or O, respectively. This may be related to a

high activation energy, required for the completion of

proton transfer to oxygen, if no other molecule takes part

in the reaction. This step was actually found to be effic-

iently assisted by a free B(C₆F₅)₃[or H₂O-B(C₆F₅)₃],

which coordinates the nitrogen atom while simulta-

neously displacing H₂O-B(C₆F₅)₃. We simulated this pro-

cess for 22_{p yr r}by choosing a BF₃molecule (for computa-

order of 50 kcal mol^{-1 23,24}

Because the complexation of

.

B(C₆F₅)₃to the 3H-indole can be assumed to be much

faster than the 3H-indole f 1H-indole reaction (that is,

(22) (a) J oule, J . A.; Mills, K. Heterocyclic Chemistry, 4th ed.;

Blackwell: London, U.K., 2000; Chapters 13 and 17. (b) Catala´n, J .;

Ya´n˜ez, M. J . Am. Chem. Soc. 1984, 106, 421. (c) Hinman, R. L.;

Whipple, E. B. J . Am. Chem. Soc. 1962, 84, 2534.

(23) Dubnikova, F.; Lifshitz, A. J . Phys. Chem. A 2001, 105, 3605.

(24) Smith, B. J .; Liu, R. J . Mol. Struct. (THEOCHEM) 1999, 491,

211.

(25) Gut, I. G.; Wirz, J . Angew. Chem., Int. Ed. Engl. 1994, 33, 1153.

5454 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

F IGURE 12. Optimized structures of species 22_{in d}and 22_{p yr r}

.

SCHEME 8

Con clu sion s

The reaction of pyrroles and indoles with tris(pen-

tafluoroaryl)borane produces 1:1 B-N complexes that

contain highly acidic sp³carbons, generated by a formal,

nitrogen to carbon 1,3-H shift. A seemingly related

process has been observed by Erker in the formation of

B(C₆F₅)₃keto-O-adducts of naphthols.²⁶The (C₆F₅)₃B-

N-heterocycle complexes are able to protonate Lewis

bases such as trialkylamines. The mechanism of forma-

tion of these adducts probably entails protonation of the

heterocycle by catalytic amounts of H₂O-B(C₆F₅)₃, fol-

lowed by displacement of H₂O-B(C₆F₅)₃from the result-

ing complex by a free borane molecule. Restricted rotation

about the B-N and B-C bonds was observed in some

derivatives, and the corresponding barriers were calcu-

lated by NMR line-shape simulation. The activating

ability of these complexes in olefin polymerization^9,11will

be the subject of a future publication.

Exp er im en ta l Section

Gen er a l P r oced u r es. All operations were performed under

nitrogen by using conventional Schlenk-line techniques. Sol-

vents were purified by degassing with N₂and passing over

activated (8 h, N₂purge, 300 °C) Al₂O₃and stored under

nitrogen. Indole, 2-methylindole, 3-methylindole, 5-methoxy-

indole, 5-benzyloxyindole, 5-chloroindole, pyrrole, 2,5-dimeth-

ylpyrrole, 2,4-dimethylpyrrole, 2-ethylpyrrole, 4,5,6,7-tetrahy-

droindole, pyrrolidine, indoline, 2-methyl-1-pyrroline, imidazole,

pyrazole, BPh₃, BCl₃(1.0 M solution in heptane), BF₃‚Et₂O,

benzaldehyde, and B(C₆F₅)₃(Boulder Scientific Co.) were used

as received. 2-Methyl-5,6-dihydroindeno[2,1-b]indole²⁷was

synthesized according to the literature. Triethylamine was

dried over KOH pellets and stored under nitrogen over

activated 4 Å molecular sieves before use.

The ¹H and ¹³C NMR spectra of the compounds were

obtained using a Bruker DPX 200 spectrometer operating in

the Fourier transform mode at room temperature at 200.13

and 50.33 MHz, respectively. The samples were dissolved in

CDCl₃, CD₂Cl₂, C₂D₂Cl₄, C₆D₆, or C₆D₅CD₃. As reference, the

residual peak of CHCl₃, CHDCl₂, C₂HDCl₄, C₆HD₅, or C₆D₅-

CHD₂in the ¹H spectra (7.25, 5.35, 5.95, 7.15, or 2.10 ppm,

respectively) and the peak of the solvent in the ¹³C spectra

(53.80 ppm for CD₂Cl₂and 128.00 ppm for C₆D₆) were used.

tional reasons): the energy barrier is only 0.5 kcal mol^-1

with respect to the isolated 22_{p yr r}and BF₃. Finally, 1 and

2 are 11.9 and 14.5 kcal mol^-1more stable than the

starting B(C₆F₅)₃and indole or pyrrole, respectively.

The occurrence of path b (Scheme 8) as the main

product-forming mechanism was also suggested by ex-

periments carried out on indole with variable amounts

of preformed H₂O-B(C₆F₅)₃1:1 complex. In the presence

of 1 equiv of H₂O-B(C₆F₅)₃, formation of adduct 2 was a

very slow process that took 3 days to go to completion.

In addition, formation of 2 was accompanied by an

1

intermediate species (revealed by H NMR) that gradu-

ally disappeared during the course of the reaction.

Formation of 2 was instead almost as fast as usual by

using 0.7 equiv of H₂O-B(C₆F₅)₃together with 0.3 equiv

of “free” borane.

These data are clearly consistent with a mechanism

that entails preliminary formation of adduct 22_{in d}fol-

lowed by bimolecular elimination of H₂O-B(C₆F₅)₃by a

borane molecule (Scheme 8, path b). The unimolecular

alternative (Scheme 8, path a) could operate to some

extent but is probably a very minor process in the

presence of free borane.

(26) Vagedes, D.; Fro¨hlich, R.; Erker, G. Angew. Chem., Int. Ed.

Engl. 1999, 38, 3362.

(27) Nifant’ev, I. E.; Bagrov, V. V. Int. Pat. Appl. WO 99/24446 to

Montell, 1999.

(28) Sheldrick, G. M. SADABS; Universita¨t Go¨ttingen: Go¨ttingen,

Germany, 1996.

J . Org. Chem, Vol. 68, No. 14, 2003 5455

Guidotti et al.

Proton spectra were acquired with a 15° pulse and 2 s of delay

between pulses; 32 transients were stored for each spectrum.

The carbon spectra were acquired with a 45° pulse and 6 s of

delay between pulses; ∼512 transients were stored for each

spectrum. CD₂Cl₂and C₂D₂Cl₄were used as received, whereas

CDCl₃, C₆D₆, and C₆D₅CD₃were dried over activated 4 Å

molecular sieves before use. Preparation of the samples was

carried out under nitrogen by using standard inert atmosphere

techniques.

The NMR spectra of the dynamics study in toluene solution

of compound 2 were recorded on Bruker DRX 300 and AMX

500. Most of the ¹⁹F spectra were acquired on the DRX300

spectrometer without proton decoupling, because the detuned

proton channel was used for pulsing (PW_90°) 26 µs). The

spectra were referenced to coaxial tube CFCl₃(δ ) 0.0). The

temperature was calibrated with standard CH₃OH/CD₃OD and

ethylene glycol/DMSO solutions. All of the fluorine signals are

complex multiplets due to intra-ring ¹⁹F-¹⁹F couplings and the

occurrence also of through-space inter-ring ¹⁹F-¹⁹F coupling

and of ¹H-¹⁹F couplings, for some of the ortho resonances. Only

in the para region do the signals appear as deceptively simple

triplets of triplets.

GC-MS analyses were carried out on an HP 5890 series 2

gas cromatograph and an HP 5989B quadrupole mass spec-

trometer. MS analysis of N-[tris(pentafluorophenyl)borane]-

3H-indole (2) was carried out on a WATERS micromass ZQ

4000, by using the electrospray ionization technique (ESI).

The melting points of the compounds were obtained by using

a capillary electrothermal instrument.

For calculation of the barriers of the B-N bond torsion, total

line-shape simulations were performed by means of a PC

version of the DNMR-6 program (QCPE program 633, Indiana

University, Bloomington, IN).

removed in vacuo to give a whitish powder as product (yield

100%). Traces of the N-[tris(pentafluorophenyl)borane]-3H-

pyrrole isomer were noticed.

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](1H-pyr -

r ol-1-yl)bor a te (1a ). A solution of triethylamine (0.1657 g,

1.63 mmol) in 5 mL of dichloromethane was added dropwise

at room temperature under nitrogen atmosphere to a solution

of 1 (0.9780 g, 1.69 mmol) in 12 mL of dichloromethane in a

25 mL Schlenk flask. Exothermicity was not observed. The

light yellow solution was stirred for 1 h at room temperature,

and then the solvent was removed in vacuo to give a whitish

solid as product (1.14 g, 100%): mp 151.1-152.9 °C.

An aliquot of the product was dissolved into a few milliliters

of CH₂Cl₂, and the resulting solution was stored at 5 °C for 20

1

h: crystals were isolated and analyzed by H NMR: ¹H NMR

(CD₂Cl₂) δ 1.27 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 3.03 [q, 6H,

J ) 7.24 Hz, N(CH₂CH₃)₃], 4.85 (bs, 1H, NH), 5.95 (t, 2H, J )

2.15 Hz, H4, H3), 6.80 (m, 2H, H5, H2); ¹³C NMR (CDCl₃) δ

8.46 [N(CH₂CH₃)₃], 46.79 [N(CH₂CH₃)₃], 104.13 (C4, C3),

127.03 (C5, C2).

N-[Tr is(p en ta flu or op h en yl)bor a n e]-3H-In d ole (2). (a )

Exp er im en t a t Room Tem p er a tu r e in CH₂Cl₂by Ad d in g

B(C₆F ₅)₃to In d ole. Indole (1.07 g, 9.0 mmol) was dissolved

in 10 mL of CH₂Cl₂and charged into a 50 mL Schlenk under

nitrogen atmosphere. A solution of B(C₆F₅)₃(4.61 g, 9.0 mmol)

in 25 mL of CH₂Cl₂was added at room temperature under

stirring. During the addition, the color of the solution turned

immediately from yellowish to amber yellow; exothermicity

was not observed. The reaction mixture was stirred at room

temperature for 1 h, and then the solvent was removed in

vacuo to give a whitish solid as product (5.32 g, 94.4%): mp

1

203.9-206.7 °C; H NMR (CD₂Cl₂) δ 4.30 (broad AB system,

2H, H3,H3′), 7.39-7.72 (m, 3H, Ar), 7.72 (d, 1H, J ) 7.6 Hz,

H4), 8.83 (d, 1H, J _HF) 5.0 Hz, H2); ¹³C NMR (CD₂Cl₂) δ 42.18

(C3), 118.26 (CH), 125.35 (CH), 129.16 (CH), 129.20 (CH),

133.07 (C), 147.97 (C), 175.43 (C2) (peak assigned by a DEPT

experiment); NOESY (CD₂Cl₂) δ¹H/δ¹H 4.30/7.72 (H3/H4), 8.83/

4.30 (H2/H3).

N-[Tr is(p en t a flu or op h en yl)b or a n e]-5H -P yr r ole (1).

(a ) Exp er im en t in CH₂Cl₂. A yellow-orange solution of

pyrrole (0.35 g, 5.11 mmol) in 10 mL of dichloromethane was

added at room temperature under nitrogen atmosphere to a

light yellow solution of tris(pentafluorophenyl)borane (2.64 g,

5.12 mmol) in 40 mL of dichloromethane in a 100 mL Schlenk

flask. Exothermicity was not observed. The obtained yellow

reaction mixture was stirred for 2 h at room temperature, and

then the solvent was removed in vacuo to give a whitish

powder as product (yield 100%): mp 181.1-183.6 °C; ¹H NMR

(CD₂Cl₂) δ 4.71 (bs, 2H, H5,H5′), 6.94 (dq, 1H, J ) 5.48 Hz, J

) 1.08 Hz, H3), 7.90 (dq, 1H, J ) 5.48 Hz, J ) 1.08 Hz, H4),

8.58 (m, 1H, J ) 1.08 Hz, H2); ¹³C NMR (CD₂Cl₂) δ 66.72 (m,

C5), 128.61 (C3), 156.98 (C4), 172.04 (C2); NOESY (CD₂Cl₂)

δ¹H/δ¹H 4.71/7.90 (H5/H4), 7.90/6.94 (H4/H3), 6.94/8.58 (H3/

1

¹H NMR analysis at variable temperatures: T ) 271 K, H

NMR (C₂D₂Cl₄) δ 4.19 (AB system, 2H, J ) 26.2 Hz, H3,H3′),

7.31-7.62 (m, 4H, Ar), 8.72 (t, 1H, J _HF) 4.3 Hz, H2); T ) 297

1

K, H NMR (C₂D₂Cl₄) δ 4.19 (broad AB system, 2H, H3,H3′),

7.32-7.62 (m, 4H, Ar), 8.72 (t, 1H, J _HF) 4.9 Hz, H2); T ) 319

1

K, H NMR (C₂D₂Cl₄) δ 4.19 (bs, 2H, H3,H3′), 7.34-7.63 (m,

4H, Ar), 8.73 (s, 1H, H2); T ) 360 K, ¹H NMR (C₂D₂Cl₄) δ

4.18 (s, 2H, H3,H3′), 7.36-7.63 (m, 4H, Ar), 8.73 (s, 1H, H2).

ESI, m/z 628 [M⁺- 1].

(b) Exp er im en t a t 0 °C in CH₂Cl₂by Ad d in g In d ole to

B(C₆F ₅)₃. A solution of indole (0.18 g, 1.56 mmol) in 2 mL of

dichloromethane was added at 0 °C under nitrogen atmosphere

to a solution of tris(pentafluorophenyl)borane (0.81 g, 1.59

mmol) in 6 mL of dichloromethane in a 25 mL Schlenk flask.

Exothermicity was not observed. The reaction mixture was

allowed to warm to room temperature and stirred for 20 h. A

¹H NMR analysis showed that the reaction was already

complete after 1 h of stirring at room temperature. The solvent

was evaporated in vacuo to give a whitish solid as product

(yield ) 95.0%).

1

H2); H NMR (C₆D₆) δ 3.70 (bs, 2H, H5,H5′), 5.62 (dq, 1H, J

) 6.16 Hz, J ) 1.08 Hz, H3), 6.51 (dq, 1H, J ) 6.16 Hz, J )

1.08 Hz, H4), 7.51 (m, 1H, J ) 1.08 Hz, H2); ¹³C NMR (C₆D₆)

δ 65.76 (m, C5), 127.38 (C3), 155.67 (C4), 171.38 (C2); NOESY

(C₆D₆) δ¹H/δ¹H 3.70/6.51 (H5/H4), 6.51/5.62 (H4/H3), 5.62/7.51

(H3/H2).

The ¹H NMR spectrum in C₆D₆of complex 1 does not change

when the temperature is varied from 173 to 353 K.

(b) Exp er im en t in Tolu en e. A light yellow solution of tris-

(pentafluorophenyl)borane (1.182 g, 2.31 mmol) in 8 mL of

toluene was added at room temperature to a yellow solution

of pyrrole (0.158 g, 2.30 mmol) in 2 mL of toluene under

nitrogen atmosphere in a 25 mL Schlenk flask. Exothermicity

was not observed. The obtained yellow reaction mixture was

stirred for 2 h at room temperature, and then the solvent was

removed in vacuo to give a yellow powder as product (1.255 g,

purity ) 99.5%, yield ) 93.8%).

(c) Exp er im en t in P en ta n e. Pyrrole (0.270 g, 3.94 mmol)

was added at room temperature to a light yellow suspension

of tris(pentafluorophenyl)borane (2.03 g, 3.94 mmol) in 45 mL

of pentane in a 50 mL Schlenk flask. Exothermicity was not

observed. After 30 min of stirring at room temperature, a milky

suspension was formed and analyzed by ¹H NMR. The pyrrole

conversion was found to be quantitative. The solvent was then

(c) Exp er im en t a t Room Tem p er a tu r e in CH ₂Cl₂by

Ad d in g In d ole to B(C₆F ₅)₃. A solution of indole (0.41 g, 3.42

mmol) in 6 mL of dichloromethane was added at room

temperature under nitrogen atmosphere to a solution of tris-

(pentafluorophenyl)borane (1.76 g, 3.43 mmol) in 9 mL of

dichloromethane in a 25 mL Schlenk flask. Exothermicity was

not observed. During the addition the color of the solution

turned from light yellow to yellow. After 6 h of stirring at room

temperature, the solvent was evaporated in vacuo to give a

whitish solid as product (2.18 g, purity ) 99%, yield ) 100%).

The ¹H NMR analysis showed the presence of 1 wt % of

1

unreacted indole. H NMR (CD₂Cl₂) δ 4.28 (broad AB system,

2H, H3,H3′), 7.39-7.72 (m, 4H, Ar), 8.81 (d, 1H, J _HF) 4.9

Hz, H2).

5456 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

(d ) Exp er im en t a t -20 °C in Dieth yl Eth er by Ad d in g

In d ole to B(C₆F ₅)₃. A solution of indole (0.72 g, 6.05 mmol)

in 5 mL of diethyl ether was added at -20 °C under nitrogen

atmosphere to a suspension of tris(pentafluorophenyl)borane

(3.13 g, 6.07 mmol) in 20 mL of diethyl ether in a 50 mL

Schlenk flask. During the addition the color of the suspension

turned from whitish to yellow. The reaction mixture was then

allowed to warm to room temperature and stirred for 2 h with

final formation of a yellow solution. A ¹H NMR analysis

showed that the reaction was already complete after 1 h of

stirring at room temperature. The solvent was evaporated in

The ¹⁹F spectrum at 233 K was simulated with the program

WINDAISY to obtain the coupling constants reported in Tables

S1 and S2 (Supporting Information). To the best of our

knowledge, there are no programs that can handle the whole

size of the exchange problem (exchange between three sites

with five nonequivalent coupled spins each, or even more if

also heteronuclear coupling is included), so the analysis was

simplified to the exchange among three ABX sites (i.e., only

the para and meta resonances for each ring were considered).

In this assumption, computer simulations of the para region

of the 1D ¹⁹F variable temperature spectra were performed

with WINDYNAMICS. The rate constants obtained accord-

ingly and the Arrhenius plots are available in the Supporting

Information (Table S3 and Figure S28).

1

vacuo to give a light yellow solid as product (yield 100%). H

NMR (CDCl₃) δ 4.22 (broad AB system, 2H, H3,H3′), 7.34-

7.66 (m, 4H, Ar), 8.77 (d, 1H, J _HF) 5.0 Hz, H2).

(e) Exp er im en t a t Room Tem p er a tu r e in Tolu en e by

Ad d in g In d ole to B(C₆F ₅)₃. A solution of indole (0.65 g, 5.55

mmol) in 5 mL of toluene was added at room temperature

under nitrogen atmosphere to a solution of tris(pentafluo-

rophenyl)borane (2.82 g, 5.51 mmol) in 15 mL of toluene in a

50 mL Schlenk flask. Exothermicity was not observed. After

1 h of stirring at room temperature, a ¹H NMR analysis

showed complete conversion of starting indole to the complex

2. The solvent was evaporated in vacuo to give a whitish solid

as product (yield ) 100%): ¹H NMR (CD₂Cl₂) δ 4.29 (broad

AB system, 2H, H3,H3′), 7.40-7.72 (m, 4H, Ar), 8.82 (d, 1H,

J _HF) 5.1 Hz, H2).

(f) Exp er im en t a t 0 °C in Tolu en e by Ad d in g B(C₆F ₅)₃

to In d ole. A solution of tris(pentafluorophenyl)borane (1.31

g, 2.56 mmol) in 8 mL of toluene was added at 0 °C under

nitrogen atmosphere to a solution of indole (0.30 g, 2.54 mmol)

in 2 mL of toluene in a 25 mL Schlenk flask. The reaction

mixture was then allowed to warm to room temperature and

stirred for 4 h with final formation of a yellow solution. The

solvent was evaporated in vacuo to give a whitish solid as

product (1.49 g, purity ) 97%, yield ) 90.4%). The ¹H NMR

analysis showed the presence of 3 wt % of unreacted indole.

¹H NMR (CD₂Cl₂) δ 4.28 (broad AB system, 2H, H3,H3′), 7.38-

7.72 (m, 4H, Ar), 8.81 (d, 1H, J _HF) 4.7 Hz, H2).

Tr ieth yla m m on iu m [Tr is(p en ta flu or op h en yl)](1H-in -

d ol-1-yl)bor a te (2a ). Triethylamine (0.22 mL, 1.60 mmol) was

added dropwise at room temperature under nitrogen atmo-

sphere to a solution of 2 (1.01 g, 1.60 mmol) in 5 mL of

dichloromethane in a 25 mL Schlenk flask. During the addition

the solution turned from yellow to pink. Exothermicity was

not observed. After 1 h of stirring at room temperature, the

solvent was removed in vacuo to give a pinkish solid, which

was determined to be the desired product by ¹H NMR analysis

in CD₂Cl₂(1.16 g, 99.3%): mp 178.6-181.2 °C; ¹H NMR (CD₂-

Cl₂) δ 1.03 [t, 9H, J ) 7.3 Hz, N(CH₂CH₃)₃], 2.73 [q, 6H, J )

7.3 Hz, N(CH₂CH₃)₃], 4.70 (bs, 1H, NH), 6.33 (m, 1H, H3),

6.89-6.98 (m, 2H, Ar), 7.27-7.33 (m, 2H, H2 and Ar), 7.45-

7.54 (m, 1H, Ar); ¹³C NMR (CD₂Cl₂) δ 8.77 [N(CH₂CH₃)₃], 47.71

[N(CH₂CH₃)₃], 98.01 (C3), 114.38 (C2), 118.13, 119.20, 120.03

(C4, C5, C6), 130.48 (C3a), 135.26 (C7), 141.17 (C7a).

(a ) Exp er im en t in On e Step . Indole (0.292 g, 2.47 mmol)

and triethylamine (0.252 g, 2.48 mmol) were dissolved under

nitrogen atmosphere into 2 mL of dichloromethane in a 25 mL

Schlenk flask. A solution of tris(pentafluorophenyl)borane

(1.26 g, 2.46 mmol) in 6 mL of dichloromethane was added at

0 °C under stirring. At the end of the addition the yellow

solution was allowed to warm to room temperature and stirred

for 17 h. A ¹H NMR analysis showed that the reaction was

already complete after 1 h of stirring at room temperature.

The solvent was then removed in vacuo to give a pink solid as

product (1.54 g, 85.7%). The product was dissolved into 15 mL

of CH₂Cl₂, and the resulting solution was stored at 5 °C for 4

h and then at -20 °C for 12 days. Crystals were isolated and

The stability in air of complex 2 was checked both in solution

and in the solid state. The complex was found to be quite stable

in air, more in the solid state than in solution. Indeed, after 9

days, a CD₂Cl₂solution of a sample showed little decomposition

to a 5 wt % of indole and to a few weight percent of a second

compound not identified, whereas after the same time an

analogous powdery sample showed <2 wt % of indole and

traces of this unknown compound. After 15 days, the powdery

1

analyzed by H NMR in CD₂Cl₂at variable temperatures: T

) 298 K, ¹H NMR (CD₂Cl₂) δ 1.03 [t, 9H, J ) 7.3 Hz,

N(CH₂CH₃)₃], 2.73 [q, 6H, J ) 7.3 Hz, N(CH₂CH₃)₃], 4.70 (bs,

1H, NH), 6.33 (m,1H, H3), 6.89-6.98 (m, 2H, Ar), 7.27-7.33

(m, 2H, H2 and Ar), 7.45-7.54 (m, 1H, Ar); T ) 263 K, ¹H

NMR (CD₂Cl₂) δ 0.99 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 2.75

[q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃], 4.84 (bs, 1H, NH), 6.33 (d,

1H, J ) 2.84 Hz, H3), 6.89-6.98 (m, 2H, Ar), 7.24-7.34 (m,

1

sample became pink from white, but its H NMR spectrum in

CD₂Cl₂did not change.

N-[Tr is(p en ta flu or op h en yl)bor a n e]-3H-In d ole (2) in

Tolu en e-d ₈. N-[Tris(pentafluorophenyl)borane]-3H-indole (2)

(30.9 mg) was directly dissolved in the NMR tube under

nitrogen. The assigments of the resonances were performed

through appropriate standard 2D experiments (Supporting

Information): ¹H-¹H NOESY and COSYGS for ¹H; ¹H-¹³C

HMQC for ¹³C; ¹⁹F-¹⁹F COSY45 and NOESY. ¹H NMR (C₇D₈,

1

2H, H2 and Ar), 7.45-7.54 (m, 1H, Ar); T ) 248 K, H NMR

(CD₂Cl₂) δ 0.97 [t, 9H, J ) 7.34 Hz, N(CH₂CH₃)₃], 2.74 [bq,

6H, J ) 7.34 Hz, N(CH₂CH₃)₃], 4.88 (bs, 1H, NH), 6.33 (d, 1H,

J ) 2.84 Hz, H3), 6.89-6.97 (m, 2H, Ar), 7.24-7.36 (m, 2H,

H2 and Ar), 7.45-7.53 (m, 1H, Ar); T ) 238 K, ¹H NMR (CD₂-

Cl₂) δ 0.96 [t, 9H, J ) 7.34 Hz, N(CH₂CH₃)₃], 2.73 [m, 6H, J )

7.34 Hz, N(CH₂CH₃)₃], 4.82 (bs, 1H, NH), 6.33 (d, 1H, J ) 2.84

Hz, H3), 6.88-6.97 (m, 2H, Ar), 7.24-7.35 (m, 2H, H2 and

Ar), 7.44-7.53 (m, 1H, Ar); T ) 203 K, ¹H NMR (CD₂Cl₂) δ

0.90 [t, 9H, J ) 7.34 Hz, N(CH₂CH₃)₃], 2.72 [m, 6H, J ) 7.34

Hz, N(CH₂CH₃)₃], 4.86 (bs, 1H, NH), 6.32 (d, 1H, J ) 2.93 Hz,

H3), 6.87-6.96 (m, 2H, Ar), 7.23-7.34 (m, 2H, H2 and Ar),

7.43-7.52 (m, 1H, Ar). The CD₂Cl₂solution of the crystals was

stable for 1 day under nitrogen atmosphere.

2

290 K, 11 T) δ 2.60 (AB system, J ) 26.42 Hz, 1H, H3), 2.68

(AB system, 1H, H3′), 6.75 (m, 1H, H4), 6.79 (m, 1H, H6), 6.81

(m, 1H, H5), 7.58 (pseudo-d, 1H, J ) 7.76 Hz, H7), 7.64

(pseudo-d, br, 1H, H2); ¹³C NMR (C₇D₈, 300 K, 7.1 T) δ 40.3

(1CH₂, C3), 117.96 (1CH, C7), 124.6 (1CH, C4), 128.66, 128.59

(2CH, C5/C6), 132.5, 147.6 (2Cq, C3a/C7a), 174.8 (1CH, C2);

¹¹B NMR (C₇D₈, 300 K, 7.1 T) δ -3.47 (s, br); ¹⁹F NMR (C₇D₈,

300 K, 7.1 T) ring A, δ -129.90 (m, ortho), -131.31 (m, ortho),

-155.32 (ps-t, para), -161.55 (m, meta), -162.94 (m, meta);

ring B, δ -133.04 (m, ortho), -135.23 (m, ortho), -155.44 (ps-

t, para), -163.49 (m, meta), -164.32 (m, meta); ring C, δ

-129.37 (m, ortho), -134.52 (m, ortho), -156.76 (ps-t, para),

-163.81 (m, meta), -164.12 (m, meta).

N-[Tr is(p en ta flu or op h en yl)bor a n e]-2,5-Dim eth yl-5H-

p yr r ole (3). A pink solution of 2,5-dimethylpyrrole (0.313 g,

3.22 mmol) in 8 mL of dichloromethane was added at room

temperature under nitrogen atmosphere to a light yellow

solution of tris(pentafluorophenyl)borane (1.659 g, 3.22 mmol)

in 15 mL of dichloromethane in a 25 mL Schlenk flask.

The heteronuclear ¹⁹F-¹H HETCOR and HOESY experi-

ments were performed at 263 K on a DRX300 spectrometer

equipped with a QNP probe (PW_90°) 9.5 µs).

J . Org. Chem, Vol. 68, No. 14, 2003 5457

Guidotti et al.

Exothermicity was not observed. The reaction mixture was

stirred for 5 h at room temperature and analyzed by H NMR

at different times. The final light-orange solution was dried

in vacuo, giving a yellow powder as product (1.878 g, 96.1%):

mp 145.8-146.9 °C.

Hz, H5), 4.90 (AB system, 1H, J ) 24.26 Hz, H5′), 6.39 (bq,

1H, J ) 1.22 Hz, H3). ¹³C NMR (CD₂Cl₂) δ 14.52 (CH₃in 4),

18.14 (d, J _CF) 3.12 Hz, CH₃in 2), 69.83 (t, J _CF) 9.86 Hz,

C5), 128.11 (C3), 169.16 (C4), 184.94 (d, J _CF) 3.12 Hz, C2). T

1

) 313 K, H NMR (C₂D₂Cl₄) δ 2.14 (m, 3H, CH₃in 2), 2.23 (d,

3H, J ) 1.33 Hz, CH₃in 4), 4.72 (bs, 2H, H5, H5′), 6.32 (m,

1H, H3).

The product was found to be, by NMR analysis, a mixture

of N-[tris(pentafluorophenyl)borane]-2,5-dimethyl-5H-pyrrole

(3, 90%) and N-[tris(pentafluorophenyl)borane]-2,5-dimethyl-

3H-pyrrole (3′, 10%). N-[Tris(pentafluorophenyl)borane]-2,5-

dimethyl-5H-pyrrole (3): ¹H NMR (CD₂Cl₂) δ 1.23 (bt, 3H, J

) 7.14 Hz, CH₃in 5), 2.20 (d, 3H, J ) 2.84 Hz, CH₃in 2), 5.41

(bs, 1H, H5), 6.62 (dd, 1H, J ) 5.48 Hz, J ) 1.17 Hz, H3), 7.67

(m, 1H, J ) 5.48 Hz, H4); ¹H NMR (C₆D₆) δ 0.50 (m, 3H, CH₃

in 5), 1.29 (d, 3H, J ) 2.74 Hz, CH₃in 2), 4.70 (bs, 1H, H5),

5.27 (dd, 1H, J ) 5.38 Hz, J ) 1.17 Hz, H3), 6.21 (dm, 1H, J

) 5.38 Hz, H4); ¹³C NMR (CD₂Cl₂) δ 15.94 (d, J _CF) 15.3 Hz,

CH₃in 5), 19.36 (bs, CH₃in 2), 77.02 (d, J _CF) 15.3 Hz, CH5),

130.31 (C3), 161.43 (C4), 185.86 (d, J _CF) 3.70 Hz, C2); NOESY

(CD₂Cl₂) δ¹H/δ¹H 5.41/1.23 (H5/CH₃in 5), 2.20/6.62 (CH₃in

2/H3), 6.62/7.67 (H3/H4), 7.67/5.41 (H4/H5). N-[Tris(pentafluo-

rophenyl)borane]-2,5-dimethyl-3H-pyrrole (3′): ¹H NMR (CD₂-

Cl₂) δ 2.03 (bs, 3H, CH₃), 2.44 (m, 3H, J ) 2.05 Hz, CH₃), 3.71

(broad AB system, 2H, J ) 26.8 Hz, H3,H3′), 6.10 (bs, 1H,

N-[Tr is(p en t a flu or op h en yl)b or a n e]-5-et h yl-5H -p yr -

r ole (5). An orange solution of 2-ethylpyrrole (0.367 g, 3.47

mmol) in 5 mL of dichloromethane was added at room

temperature under nitrogen atmosphere to a light yellow

solution of tris(pentafluorophenyl)borane (1.800 g, 3.49 mmol)

in 15 mL of dichloromethane in a 25 mL Schlenk flask. During

the addition the color of the solution turned immediately from

orange to dark orange; exothermicity was not observed. The

reaction mixture was stirred overnight at room temperature:

a

¹H NMR analysis showed the presence of ∼11 mol % of

unreacted 2-ethylpyrrole. Then 0.21 g (0.41 mmol) of tris-

(pentafluorophenyl)borane was added to complete the reaction.

After a few minutes of stirring, the solvent was removed in

vacuo to give a dark yellow powder as product (yield ) 100%):

mp 185.4-186.8 °C; ¹H NMR (CD₂Cl₂) δ 0.88 (t, 3H, J ) 7.43

Hz, CH₃), 2.67 (bm, 2H, CH₂), 4.99 (broad AB system, J )

25.24 Hz, 2H, H5,H5′), 6.88 (dt, 1H, J ) 5.58 Hz, J ) 1.27 Hz,

H3), 7.77 (d, 1H, J ) 5.58 Hz, H4); ¹H NMR (C₆D₆) δ 0.075 (t,

3H, J ) 7.43 Hz, CH₃), 2.00 (m, 2H, J ) 7.43 Hz, CH₂), 4.14

(broad AB system, J ) 25.14 Hz, 2H, H5,H5′), 5.54 (dt, 1H, J

) 5.48 Hz, J ) 1.27 Hz, H3), 6.31 (d, 1H, J ) 5.48 Hz, H4);

NOESY (CD₂Cl₂) δ¹H/δ¹H 2.67/6.88 (CH₂/H3), 6.88/7.77 (H3/

H4), 7.77/4.99 (H4/H5); ¹³C NMR (CD₂Cl₂) δ 9.80 (CH₃), 25.48

(CH₂), 68.36 (m, C5), 130.30 (C3), 154.37 (C4), 189.38 (C2).

1

H4); H NMR (C₆D₆) δ 1.53 (m, 3H, CH₃), 1.61 (bs, 3H, CH₃),

2.09 (broad AB system, 2H, J ) 27.1 Hz, H3,H3′), 4.98 (bs,

1H, H4).

The ¹H NMR analysis of 3 at variable temperatures in CD₂-

Cl₂showed a significant change only in the signal of the methyl

group in the 2-position, which resulted in a clear triplet at

238 K, with chemical shift of 1.18 ppm and J ) 7.24 Hz.

Tr ieth yla m m on iu m [Tr is(p en ta flu or op h en yl)](2,5-d i-

m eth yl-1H-p yr r ol-1-yl)bor a te (3a ). A solution of triethyl-

amine (0.2209 g, 2.17 mmol) in 5 mL of dichloromethane was

added dropwise at room temperature under nitrogen atmo-

sphere to a solution of 3 (1.3080 g, 2.15 mmol) in 12 mL of

dichloromethane in a 25 mL Schlenk flask. Exothermicity was

not observed. The yellow solution was stirred for 1 h at room

temperature, and then the solvent was removed in vacuo to

give a yellow-light orange solid as product (yield 100%): mp

119.8-121.5 °C.

1

¹H NMR analysis at variable temperatures: T ) 239 K, H

NMR (CD₂Cl₂) δ 0.79 (t, 3H, J ) 7.34 Hz, CH₃), 2.34-2.87 (m,

2H, CH₂), 4.94 (AB system, 2H, J ) 26.21 Hz, H5,H5′), 6.84

(d, 1H, J ) 5.48 Hz, H3), 7.76 (d, 1H, J ) 5.48 Hz, H4); T )

313 K, ¹H NMR (C₂D₂Cl₄) δ 0.80 (t, 3H, J ) 7.52 Hz, CH₃),

2.42-2.90 (broad AB system, 2H, CH₂), 4.91 (broad AB system,

2H, H5,H5′), 6.78 (d, 1H, J ) 5.55 Hz, H3), 7.69 (d, 1H, J )

5.55 Hz, H4); T ) 320 K, ¹H NMR (C₂D₂Cl₄) δ 0.81 (t, 3H, J )

7.52 Hz, CH₃), 2.56 (bs, 2H, CH₂), 4.91 (bs, 2H, H5,H5′), 6.78

(d, 1H, J ) 5.53 Hz, H3), 7.69 (d, 1H, J ) 5.53 Hz, H4); T )

332 K, ¹H NMR (C₂D₂Cl₄) δ 0.83 (t, 3H, J ) 7.51 Hz, CH₃),

2.58 (bs, 2H, CH₂), 4.92 (bs, 2H, H5,H5′), 6.78 (d, 1H, J ) 5.51

Hz, H3), 7.69 (d, 1H, J ) 5.51 Hz, H4).

An aliquot of the product was dissolved into a few milliliters

of CH₂Cl₂, and the resulting solution was stored at -20 °C for

1

a few days: crystals were isolated and analyzed by H NMR:

¹H NMR (CD₂Cl₂) δ 1.28 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃],

1.96 (bs, 6H, CH₃), 3.06 [q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃],

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](2-eth yl-

1H-p yr r ol-1-yl)bor a te (5a ). A solution of triethylamine

(0.2288 g, 2.25 mmol) in 6 mL of dichloromethane was added

dropwise at room temperature under nitrogen atmosphere to

a solution of 5 (1.3518 g, 2.23 mmol) in 10 mL of dichlo-

romethane in a 25 mL Schlenk flask. Exothermicity was not

observed. The orange solution was stirred for 4 h at room

temperature, and then the solvent was removed in vacuo to

give an orange powder as product (1.58 g, 100%): mp 79.7-

1

5.68 (bs, 3H, H3, H4, NH); H NMR (C₆D₆) δ 0.38 [t, 9H, J )

7.24 Hz, N(CH₂CH₃)₃], 1.87 [q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃],

2.13 (bs, 6H, CH₃), 5.55 (bs, 3H, H3, H4, NH); ¹³C NMR (CD₂-

Cl₂) δ 8.84 [N(CH₂CH₃)₃], 16.63 (m, CH₃), 47.35 [N(CH₂CH₃)₃],

104.69 (C3, C4), 137.29 (C2, C5).

N-[Tr is(p en ta flu or op h en yl)bor a n e]-2,4-Dim eth yl-5H-

p yr r ole (4). A yellow-orange solution of 2,4-dimethylpyrrole

(0.564 g, 5.75 mmol) in 5 mL of dichloromethane was added

at room temperature under nitrogen atmosphere to a light

yellow solution of tris(pentafluorophenyl)borane (3.267 g, 6.34

mmol) in 20 mL of dichloromethane in a 50 mL Schlenk flask.

Exothermicity was not observed. The yellow reaction mixture

1

81.6 °C; H NMR (CD₂Cl₂) δ 0.94 (t, 3H, J ) 8.02 Hz, CH₃),

1.28 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 1.94-2.80 (bm, 2H,

CH₂), 3.07 [q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃], 5.89 (bs, 2H,

H3, H4), 6.17 (bs, 1H, NH), 6.74 (bs, 1H, H5); ¹³C NMR (CD₂-

Cl₂) δ 8.74 [N(CH₂CH₃)₃], 13.67 (CH₃), 21.24 (CH₂), 47.33

[N(CH₂CH₃)₃], 102.24 (C3 or C4), 103.74 (C4 or C3), 126.73

(C5), 142.91 (C2).

1

was stirred for 20 h at room temperature and analyzed by H

NMR at different times. The final yellow solution was dried

in vacuo, giving a dark yellow powder as product (yield )

100%): mp 209.2-211.8 °C; ¹H NMR (CD₂Cl₂) δ 2.20 (t, 3H, J

) 2.74 Hz, CH₃in 2), 2.29 (d, 3H, J ) 1.57 Hz, CH₃in 4), 4.82

(broad AB system, 2H, H5,H5′), 6.41 (q, 1H, J ) 1.57 Hz, H3);

¹H NMR (C₆D₆) δ 1.14 (d, 3H, J ) 1.47 Hz, CH₃in 4), 1.41 (t,

3H, J ) 2.74 Hz, CH₃in 2), 4.20 (bs, 2H, H5,H5′), 5.06 (bq,

1H, J ) 1.47 Hz, H3); ¹³C NMR (CD₂Cl₂) δ 14.56 (CH₃in 4),

18.40 (CH₃in 2), 70.32 (C5), 128.65 (C3), 169.60 (C4), 185.40

(C2); NOESY (CD₂Cl₂) δ¹H/δ¹H 4.82/2.29 (H5/CH₃in 4), 2.29/

6.41 (CH₃in 4/H3), 6.41/2.20 (H3/CH₃in 2).

1

¹H NMR analysis at variable temperatures: T ) 238 K, H

NMR (CD₂Cl₂) δ 0.94 (t, 3H, J ) 7.27 Hz, CH₃), 1.25 [t, 9H, J

) 7.24 Hz, N(CH₂CH₃)₃], 1.95-2.70 (m, 2H, CH₂), 3.05 [bm,

6H, N(CH₂CH₃)₃], 5.89 (bs, 1H, H3), 5.92 (m, 1H, H4), 6.75

(bs, 1H, H5), 6.86 (bs, 1H, NH).

N-[Tr is(pen taflu or oph en yl)bor an e]-2H-4,5,6,7-Tetr ah y-

d r oin d ole (6). A solution of 4,5,6,7-tetrahydroindole (0.65 g,

5.25 mmol) in 3 mL of dichloromethane was added at room

temperature under nitrogen atmosphere to a solution of tris-

(pentafluorophenyl)borane (2.69 g, 5.25 mmol) in 15 mL of

dichloromethane in a 25 mL Schlenk flask. A light exother-

micity was observed. The reaction mixture was stirred for 30

NMR analyses at variable temperatures: T ) 239 K, ¹H

NMR (CD₂Cl₂) δ 2.16 (m, 3H, CH₃in 2), 2.25 (d, 3H, J ) 1.22

Hz, CH₃in 4), 4.64 (AB system, 1H, J ) 24.26 Hz, J ) 2.54

5458 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

min at room temperature, and then the solvent was evaporated

in vacuo to give a brick red powder as product (yield ) 100%):

mp 108.8-111.8 °C; ¹H NMR (CD₂Cl₂) δ 1.02-3.42 (bs, 8H,

H4,H4′, H5,H5′, H6,H6′, H7,H7′), 4.85 (broad AB system, 2H,

H2,H2′), 7.34 (bm, 1H, H3); ¹³C NMR (CD₂Cl₂) δ 21.74 (C5

and C6), 23.87 (C4), 29.76 (C7), 66.39 (d, C2, J _CF) 10.4 Hz),

140.78 (C3a), 147.02 (C3), 186.25 (C7a).

¹H NMR analyses at variable temperatures: T ) 238 K, ¹H

NMR (CD₂Cl₂) δ 1.26-3.16 (bm, 8H, H4,H4′, H5,H5′, H6,H6′,

H7,H7′), 4.80 (AB system, 2H, J ) 26.02 Hz, H2,H2′), 7.34

(bs, 1H, H3).

6.93 (m, 2H, CH), 7.06-7.10 (m, 1H, CH), 7.36-7.41 (m, 1H,

CH); ¹³C NMR (CD₂Cl₂) δ 8.80 [N(CH₂CH₃)₃], 16.41 (dd, J _CF

)

6.10 Hz, J _CF) 3.70 Hz, CH₃), 47.58 [N(CH₂CH₃)₃], 101.09 (CH),

114.87 (dd, J _CF) 6.10 Hz, J _CF) 1.20 Hz, CH), 117.46 (CH),

117.56 (CH), 118.83 (CH), 130.09 (C), 142.96 (C), 146.34 (C).

1

¹H NMR analyses at variable temperatures: T ) 238 K, H

NMR (CD₂Cl₂) δ 0.92 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 2.17

(bs, 3H, CH₃), 2.64 [q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃], 4.20

(bs, 1H, NH), 6.15 (bs, 1H, H3), 6.74-6.91 (m, 2H, CH), 7.02

(d, 1H, J ) 8.31 Hz, CH), 7.37 (dd, 1H, J ) 7.34 Hz, J ) 0.98

Hz, CH).

N-[Tr is(p en t a flu or op h en yl)b or a n e]-2-Met h yl-3H-in -

d ole (7). (a ) Exp er im en t in CH₂Cl₂. A solution of 2-meth-

ylindole (0.67 g, 5.01 mmol) in 10 mL of dichloromethane was

added at room temperature under nitrogen atmosphere to a

solution of tris(pentafluorophenyl)borane (2.60 g, 5.05 mmol)

in 15 mL of dichloromethane in a 50 mL Schlenk flask.

Exothermicity was not observed. During the addition the color

of the solution turned from light orange to orange. A ¹H NMR

analysis in CD₂Cl₂showed quantitative conversion of the

starting 2-methylindole after 1 h of stirring at room temper-

ature. The reaction mixture became a light pink suspension

after 4 h of stirring at room temperature. The stirring was

continued overnight, and then the suspension was filtered on

a G3 frit. The residue on the frit was a white solid and was

N-[Tr is(p en t a flu or op h en yl)bor a n e]-3-Met h yl-3H-in -

d ole (8). A solution of 3-methylindole (0.92 g, 6.87 mmol) in

10 mL of dichloromethane was added at room temperature

under nitrogen atmosphere to a solution of tris(pentafluo-

rophenyl)borane (3.53 g, 6.85 mmol) in 15 mL of dichlo-

romethane in a 50 mL Schlenk flask. Exothermicity was not

observed. During the addition the color of the solution turned

from light yellow to yellow. After 30 min of stirring at room

1

temperature, a H NMR analysis showed the presence of traces

of unreacted 3-methylindole. Then 0.23 g (0.45 mmol) of tris-

(pentafluorophenyl)borane was added to complete the reaction.

After overnight stirring, the solvent was removed in vacuo to

give a white powder as product (yield ) 100%): mp 114.2-

117.5 °C; ¹H NMR (CD₂Cl₂) δ 1.61 (bs, 3H, CH₃), 4.31 (bs, 1H,

H3), 7.35-7.67 (m, 4H, Ar), 8.69 (d, 1H, J _HF) 5.3 Hz, H2); ¹H

NMR (C₆D₆) δ 0.65 (bs, 3H, CH₃), 2.74 (bs, 1H, H3), 6.62-

6.84 (m, 3H, Ar), 7.53-7.62 (m, 1H, Ar), 7.91 (bs, 1H, H2, first

diastereoisomer), 7.97 (bs, 1H, H2, second diastereoisomer);

¹³C NMR (C₆D₆) δ 11.72 (CH₃), 46.97 (C3), 111.18 (CH), 117.99

(C7), 123.76 (CH), 128.97 (CH), 138.32 (C3a), 146.52 (C7a),

179.29 (C2). The complex 8 cleanly shows two diastereoisomers

at 291 K in CD₂Cl₂. The ratio between the two diastereoiso-

mers is 55(first):45(second) at 283 K in CD₂Cl₂and does not

change at lower temperature.

1

determined to be the desired product by H NMR analysis in

C₆D₆(2.16 g, 67.0%), mp 204.3-204.5 °C. The final complex

is not fully soluble in CD₂Cl₂, whereas it is fully soluble in

C₆D₆: ¹H NMR (C₆D₆) δ 1.70 (m, 3H, CH₃), 2.46 (AB system,

2H, J ) 25.63 Hz, H3,H3′), 6.64-6.83 (m, 3H, Ar), 7.61-7.69

(m, 1H, Ar); ¹³C NMR (C₆D₆) δ 18.77 (dd, J _CF) 9.20 Hz, J _CF

2.50 Hz, CH₃), 46.88 (C3), 117.74 (dd, J _CF) 7.66 Hz, J _CF

)

1.84 Hz, C7), 123.83 (Ar), 127.75 (Ar), 128.15 (Ar), 130.79

(C3a), 150.44 (d, J _CF) 3.98 Hz, C7a), 189.36 (C2); COSY (C₆D₆)

δ¹H/δ¹H 2.46/1.70 (H3/ CH₃).

¹H NMR analyses at variable temperatures: T ) 297 K, ¹H

NMR (C₆D₅CD₃) δ 1.84 (m, 3H, CH₃), 2.59 (AB system, 2H, J

) 25.43 Hz, H3,H3′), 6.71-6.88 (m, 3H, Ar), 7.60-7.66 (m,

¹H NMR analyses at variable temperatures: T ) 173 K, ¹H

NMR (CD₂Cl₂) δ 1.49 (d, 3H, J ) 6.85 Hz, CH₃, first diaste-

reoisomer), 1.61 (d, 3H, J ) 6.85 Hz, CH₃, second diastereoi-

somer), 4.24 (q, 1H, J ) 6.85 Hz, H3, second diastereoisomer),

4.38 (q, 1H, J ) 6.85 Hz, H3, first diastereoisomer), 7.31-

1

1H, Ar); T ) 354 K, H NMR (C₆D₅CD₃) δ 1.95 (bs, 3H, CH₃),

2.85 (broad AB system, 2H, H3,H3′), 6.80-6.91 (m, 3H, Ar),

1

7.53-7.61 (m, 1H, Ar); T ) 381 K, H NMR (C₆D₅CD₃) δ 1.98

7.64 (m, 4H, Ar), 8.38-8.73 (m, 1H, H2); T ) 203 K, H NMR

(bs, 3H, CH₃), 2.93 (bs, 2H, H3,H3′), 6.86-6.98 (bm, 3H, Ar),

(CD₂Cl₂) δ 1.50 (d, 3H, J ) 7.9 Hz, CH₃, first diastereoisomer),

1.62 (d, 3H, J ) 7.9 Hz, CH₃, second diastereoisomer), 4.24

(q, 1H, J ) 7.9 Hz, H3, second diastereoisomer), 4.36 (q, 1H,

J ) 7.9 Hz, H3, first diastereoisomer), 7.32-7.64 (m, 4H, Ar),

7.53-7.58 (bm, 1H, Ar).

(b) Exp er im en t in Tolu en e. A pink solution of 2-meth-

ylindole (0.3104 g, 2.32 mmol) in 10 mL of toluene was added

at room temperature under nitrogen atmosphere to a solution

of tris(pentafluorophenyl)borane (1.2029 g, 2.34 mmol) in 15

mL of toluene in a 50 mL Schlenk flask. Exothermicity was

not observed. A ¹H NMR analysis in toluene-d₈showed

quantitative conversion of the starting 2-methylindole after

10 min of stirring at room temperature. Then the orange

solution was evaporated in vacuo to give a light pink powder,

1

8.64-8.71 (m, 1H, H2); T ) 239 K, H NMR (CD₂Cl₂) δ 1.53

(d, 3H, J ) 8.02 Hz, CH₃, first diastereoisomer), 1.65 (d, 3H,

J ) 8.02 Hz, CH₃, second diastereoisomer), 4.23 (q, 1H, J )

8.02 Hz, H3, second diastereoisomer), 4.36 (q, 1H, J ) 8.02

Hz, H3, first diastereoisomer), 7.34-7.65 (m, 4H, Ar), 8.66-

8.71 (m, 1H, H2); T ) 239 K, ¹H NMR (C₆D₅CD₃) δ 0.69 (d,

3H, J ) 7.92 Hz, CH₃, first diastereoisomer), 0.75 (d, 3H, J )

7.92 Hz, CH₃, second diastereoisomer), 2.49 (q, 1H, J ) 7.92

Hz, H3, first diastereoisomer), 2.78 (q, 1H, J ) 7.92 Hz, H3,

second diastereoisomer), 6.61-6.85 (m, 3H, Ar), 7.52-7.68 (m,

2H, H2 for the first diastereoisomer and 1 Ar), 8.05 (t, 1H, J

1

which was determined to be the desired product by H NMR

analysis in toluene-d₈(1.44 g, 96.5%).

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](2-m eth -

yl-1H-in d ol-1-yl)bor a te (7a ). A solution of triethylamine

(0.1339 g, 1.32 mmol) in 10 mL of dichloromethane was added

dropwise at room temperature under nitrogen atmosphere to

a white suspension of 7 (0.8476 g, 1.32 mmol) in 10 mL of

dichloromethane in a 25 mL Schlenk flask. During the addition

the white suspension became a light yellow solution. Exother-

micity was not observed. A ¹H NMR analysis in C₆D₆showed

complete conversion after 1 h of stirring at room temperature.

Then after 2 h, the solvent was removed in vacuo to give a

white solid, which was determined to be the desired product

1

) 4.79 Hz, H2 for the second diastereoisomer); T ) 283 K, H

NMR (CD₂Cl₂) δ 1.54 (d, 3H, J ) 8.22 Hz, CH₃, first diaste-

reoisomer), 1.67 (d, 3H, J ) 8.22 Hz, CH₃, second diastereoi-

somer), 4.22 (q, 1H, J ) 8.22 Hz, H3, second diastereoisomer),

4.34 (q, 1H, J ) 8.22 Hz, H3, first diastereoisomer), 7.34-

7.65 (m, 4H, Ar), 8.68 (two overlapping doublets, 1H, H2); T

1

) 297 K, H NMR (CD₂Cl₂) δ 1.61 (bs, 3H, CH₃), 4.28 (bs, 1H,

H3), 7.36-7.66 (m, 4H, Ar), 8.68 (d, 1H, J _HF) 5.38 Hz, H2).

¹³C NMR analyses at variable temperatures: T ) 239 K,

¹³C NMR (CD₂Cl₂) δ 12.23 (CH₃, first diastereoisomer), 12.38

(CH₃, second diastereoisomer), 47.64 (C3, first diastereoiso-

mer), 47.72 (C3, second diastereoisomer), 117.88 (Ar), 124.15

(Ar), 128.95 (2 Ar), 138.56 (C3a), 146.52 (C7a), 179.81 (C2,

second diastereoisomer), 180.19 (C2, first diastereoisomer).

1

by ¹H NMR analysis (0.858 g, 87.3%): mp 87.0-89.2 °C; H

NMR (C₆D₆) δ 0.20 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 1.64 [q,

6H, J ) 7.24 Hz, N(CH₂CH₃)₃], 2.39 (bs, 3H, CH₃), 6.61-6.78

(m, 2H, CH), 7.04-7.09 (m, 1H, CH), 7.14 (m, 1H, CH), 7.47-

1

7.51 (m, 1H, CH); H NMR (CD₂Cl₂) δ 0.99 [t, 9H, J ) 7.24

Hz, N(CH₂CH₃)₃], 2.23 (bs, 3H, CH₃), 2.66 [q, 6H, J ) 7.24

Hz, N(CH₂CH₃)₃], 4.20 (bs, 1H, NH), 6.17 (bs, 1H, H3), 6.76-

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](3-m eth -

yl-1H-in d ol-1-yl)bor a te (8a ). A solution of triethylamine

J . Org. Chem, Vol. 68, No. 14, 2003 5459

Guidotti et al.

(0.36 g, 3.56 mmol) in 10 mL of dichloromethane was added

dropwise at room temperature under nitrogen atmosphere to

a solution of 8 (2.2671 g, 3.52 mmol) in 10 mL of dichlo-

romethane in a 25 mL Schlenk flask. During the addition the

solution turned from light yellow to yellow. Exothermicity was

not observed. After overnight stirring at room temperature,

the final pink reaction mixture was evaporated in vacuo to

give a pink solid, which was determined to be the desired

product by ¹H NMR analysis in CD₂Cl₂(2.630 g, 100%), mp

181.4-182.1 °C. An aliquot of the product was dissolved into

10 mL of CH₂Cl₂, and the resulting solution was stored at 5

°C for 1 day and then at -20 °C for 2 days. Crystals were

isolated and analyzed by H NMR in CD₂Cl₂. H NMR (CD₂-

Cl₂) δ 1.04 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 2.31 (d, 3H, J )

0.78 Hz, CH₃), 2.73 [q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃], 3.61

(bs, 1H, NH), 6.87-6.96 (m, 2H, CH), 7.09 (bs, 1H, CH), 7.21-

7.29 (m, 1H, CH), 7.40-7.48 (m, 1H, CH); ¹³C NMR (CD₂Cl₂)

δ 8.61 [N(CH₂CH₃)₃], 9.99 (CH₃), 47.35 [N(CH₂CH₃)₃], 107.11

(C3), 111.25 (C3a or C7a), 114.02, 116.74, 117.52, 119.42 (C4,

C5, C6, C7), 131.00 (C7a or C3a), 133.07 (C2).

50 mL Schlenk flask. Exothermicity was not observed. An

orange solution was obtained; after 1 h of stirring, the solvent

was removed in vacuo to give a white powder as product (2.75

g, 100%): ¹H NMR (CD₂Cl₂) δ 4.21 (broad AB system, 2H, H3,

H3′), 5.11 (s, 2H, OCH₂), 7.02 (dd, 1H, J ) 9.05 Hz, J ) 2.45

Hz, H6), 7.29 (d, 1H, J ) 2.45 Hz, H4), 7.35-7.52 (m, 5H, Ar),

7.61 (d, 1H, J ) 9.05 Hz, H7), 8.65 (d, 1H, J ) 4.89 Hz, H2);

¹³C NMR (CD₂Cl₂) δ 42.04 (C3), 71.17 (OCH₂), 111.31 (C4),

116.11 (C6), 119.09 (C7), 128.08 (2 ortho- or meta-CH), 128.76

(para-CH), 129.14 (2 meta- or ortho-CH), 135.29 (C), 136.73

(C), 141.77 (C7a), 160.05 (C), 173.19 (C2).

N-[Tr is(p en t a flu or op h en yl)b or a n e]-5-Ch lor o-3H -in -

d ole (11). A solution of 5-chloroindole (0.66 g, 4.27 mmol) in

10 mL of CH₂Cl₂was added at room temperature under

nitrogen atmosphere to a solution of tris(pentafluorophenyl)-

borane (2.32 g, 4.50 mmol) in 20 mL of CH₂Cl₂in a 50 mL

Schlenk flask. Exothermicity was not observed. A colorless

solution was obtained; after 2 h of stirring, the solvent was

removed in vacuo to give a white powder as product (yield )

100%): ¹H NMR (CD₂Cl₂) δ 4.30 (broad AB system, 2H, H3,

H3′), 7.42 (dd, 1H, J ) 9.05 Hz, J ) 2.20 Hz, H6), 7.53 (d, 1H,

J ) 9.05 Hz, H7), 7.68 (dm, 1H, J ) 2.20 Hz, H4), 8.80 (d, 1H,

J _HF) 4.89 Hz, H2); ¹³C NMR (CD₂Cl₂) δ 42.11 (C3), 119.23

(C7), 125.66 (C4), 129.72 (C6), 134.79 (C3a or C5), 135.55 (C5

or C3a), 146.47 (C7a), 175.46 (C2); NOESY (CD₂Cl₂) δ¹H/δ¹H

8.80/4.30 (H2/ H3), 4.30/7.68 (H3/ H4).

1

¹H NMR analyses at variable temperatures: T ) 238 K, ¹H

NMR (CD₂Cl₂) δ 0.98 [t, 9H, J ) 7.24 Hz, N(CH₂CH₃)₃], 2.27

(d, 3H, J ) 0.78 Hz, CH₃), 2.70 [q, 6H, J ) 7.24 Hz, N(CH₂-

CH₃)₃], 4.11 (bs, 1H, NH), 6.85-6.94 (m, 2H, CH), 7.09 (bt,

1H, J ) 3.67 Hz, CH), 7.16-7.23 (bm, 1H, CH), 7.38-7.47 (m,

1H, CH).

N-[Tr is(p en ta flu or op h en yl)bor a n e]-5-Meth oxy-3H-in -

d ole (9). A solution of 5-methoxyindole (0.78 g, 5.25 mmol) in

10 mL of CH₂Cl₂was added at 0 °C under nitrogen atmosphere

to a solution of tris(pentafluorophenyl)borane (2.98 g, 5.79

mmol) in 20 mL of CH₂Cl₂in a 50 mL Schlenk flask. During

the addition, the color of the solution turned immediately from

light yellow to yellow. The reaction mixture was allowed to

warm to room temperature and stirred for 2 h, and then the

solvent was removed in vacuo to give a white powder as

product (yield ) 100%): ¹H NMR (CD₂Cl₂) δ 3.85 (s, 3H,

OCH₃), 4.21 (broad AB system, 2H, H3, H3′), 6.94 (dd, 1H, J

) 9.11 Hz, J ) 2.69 Hz, H6), 7.18 (d, 1H, J ) 2.69 Hz, H4),

7.50 (d, 1H, J ) 9.11 Hz, H7), 8.66 (d, 1H, J ) 4.83 Hz, H2);

¹³C NMR (CD₂Cl₂) δ 41.99 (C3), 56.20 (OCH₃), 110.45 (C4),

115.12 (C6), 118.94 (C7), 135.17 (C3a), 141.43 (C7a), 160.82

(C5), 172.87 (C2); NOESY (CD₂Cl₂) δ¹H/δ¹H 4.21/7.18 (H3/

H4), 4.21/8.66 (H3/ H2), 3.85/7.18 (OCH₃/H4), 3.85/6.94 (OCH₃/

H6), 7.50/6.94 (H7/ H6).

N-[Tr is(p en t a flu or op h en yl)b or a n e]-2-Met h yl-6,10b -

d ih yd r oin d en o[2,1-b]in d ole (12). 2-Methyl-5,6-dihydroin-

deno[2,1-b]indole (1.77 g, 8.1 mmol) was dissolved in 10 mL

of CH₂Cl₂and charged into a 50 mL Schlenk under nitrogen

atmosphere. A solution of tris(pentafluorophenyl)borane (4.14

g, 8.1 mmol) in 25 mL of CH₂Cl₂was added at room temper-

ature under stirring. During the addition, the color of the

solution turned immediately from green to dark brown;

exothermicity was not observed. The reaction mixture was

stirred at room temperature for 1 h, and then the solvent was

removed in vacuo to give a brown solid as product (5.90 g,

1

100%): mp 160.5-166.1 °C; H NMR (CD₂Cl₂) δ 2.46 (s, 3H,

CH₃), 3.78 (d, 1H, J ) 20.1 Hz, CH₂, H6b), 4.23 (dd, 1H, J )

20.1 Hz, J ) 3.0 Hz, CH₂, H6a), 5.86 (s, 1H, H10b), 7.18 (d,

1H, J ) 8.9 Hz, Ar, H3), 7.30-7.36 (m, 1H, Ar, H7), 7.33-

7.43 (m, 2H, Ar, H8, H9), 7.45-7.53 (m, 1H, Ar, H4), 7.60-

7.66 (m, 1H, Ar, H10), 7.68 (s, 1H, Ar, H1); ¹³C NMR (CD₂Cl₂)

δ 21.33 (CH₃), 35.72 (d, CH₂, J _CF) 10.8 Hz), 62.88 (CH_10b),

117.88 (m), 124.04, 125.31, 125.80, 129.18, 129.48, 129.98,

133.20, 134.07, 139.25, 141.19, 149.24 (d, J _CF) 4.2 Hz), 200.03

(peak assigned by a DEPT experiment); NOESY (CD₂Cl₂) δ¹H/

δ¹H 5.86/4.23 (H10b/ H6a), 5.86/7.68 (H10b/ H1), 7.60-7.66/

5.86 (H10/ H10b), 7.68/2.46 (H1/ CH₃), 7.18/2.46 (H3/ CH₃),

7.30-7.36/3.78 (H7/ H6b); COSY (CD₂Cl₂) δ¹H/δ¹H 7.68/7.18

(H1/ H3), 7.18/7.45-7.53 (H3/ H4), 7.45-7.53/2.46 (H4/ CH₃).

1

¹H NMR analysis at variable temperatures: T ) 238 K, H

NMR (CD₂Cl₂) δ 3.82 (s, 3H, OCH₃), 4.25 (AB system, 2H, J

) 26.16 Hz, H3, H3′), 6.90 (dd, 1H, J ) 9.11 Hz, J ) 2.32 Hz,

H6), 7.16 (d, 1H, J ) 2.32 Hz, H4), 7.46 (d, 1H, J ) 9.11 Hz,

H7), 8.65 (t, 1H, J ) 4.50 Hz, H2).

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](5-m eth -

oxy-1H-in d ol-1-yl)bor a te (9a ). A solution of triethylamine

(0.17 g, 1.70 mmol) in 10 mL of dichloromethane was added

dropwise at room temperature under nitrogen atmosphere to

a solution of 9 (1.12 g, 1.70 mmol) in 15 mL of dichloromethane

in a 50 mL Schlenk flask. During the addition the solution

turned from yellow to pink. Exothermicity was not observed.

After 1 h of stirring at room temperature, the solvent was

removed in vacuo to give a pink powder, which was determined

to be the desired product by NMR analysis in CD₂Cl₂(yield )

100%); mp 81.7-82.9 °C; ¹H NMR (CD₂Cl₂) δ 1.06 [t, 9H, J )

7.58 Hz, N(CH₂CH₃)₃], 2.85 [q, 6H, J ) 7.58 Hz, N(CH₂CH₃)₃],

3.80 (s, 1H, OCH₃), 4.93 (bs, 1H, NH), 6.26 (bs,1H, H3), 6.59

(dd, 1H, J ) 9.05 Hz, J ) 2.69 Hz, H6), 7.00 (d, 1H, J ) 2.69

Hz, H4), 7.19 (d, 1H, J ) 9.05 Hz, H7), 7.31 (bs, 1H, H2); ¹³C

NMR (CD₂Cl₂) δ 8.89 [N(CH₂CH₃)₃], 47.89 (N(CH₂CH₃)₃), 56.79

(OCH₃), 97.78 (C3), 102.31 (C4), 109.60 (C6), 114.85 (C7),

130.81 (C3a), 135.89 (C2), 136.78 (C7a), 153.06 (C5).

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](2-m eth -

yl-5,6-d ih yd r oin d en o[2,1-b]in d ol-5-yl)bor a te (12a ). Tri-

ethylamine (0.23 mL, 1.67 mmol) was added dropwise at room

temperature under nitrogen atmosphere to a suspension of 12

(1.22 g, 1.67 mmol) in 15 mL of dichloromethane in a 25 mL

Schlenk flask. During the addition exothermicity was not

observed. After 1 h of stirring at room temperature, the solvent

was removed in vacuo to give a brown solid, which was

1

determined to be the desired product by H NMR analysis in

CD₂Cl₂(1.38 g, 100%): mp 75.3-80.4 °C; ¹H NMR (CD₂Cl₂) δ

0.90 [t, 9H, J ) 7.3 Hz, N(CH₂CH₃)₃], 2.44 (s, 3H, CH₃), 2.63

[q, 6H, J ) 7.3 Hz, N(CH₂CH₃)₃], 3.36 (s, 2H, CH₂), 4.40 (bs,

1H, NH), 6.78-7.51 (7H, m, Ar); ¹³C NMR (CD₂Cl₂) δ 8.69

(NCH₂CH₃), 21.39 (CH₃), 34.09 (d, J _CF) 7.7 Hz, CH₂), 47.69

(NCH₂CH₃), 114.85 (m), 116.77, 117.49, 119.06, 121.72, 121.83,

123.28, 124.93, 127.20, 128.18, 140.30, 143.65, 144.80 (d, J _CF

) 2.5 Hz), 155.60.

N-[Tr is(p en ta flu or op h en yl)bor a n e]-5-Ben zyloxy-3H-

in d ole (10). A solution of 5-benzyloxyindole (0.88 g, 3.74

mmol) in 10 mL of CH₂Cl₂was added at room temperature

under nitrogen atmosphere to a solution of tris(pentafluo-

rophenyl)borane (1.93 g, 3.75 mmol) in 20 mL of CH₂Cl₂in a

N-(Tr ich lor obor a n e)-3H-In d ole (13). A solution of in-

dole (1.79 g, 15.13 mmol) in 20 mL of dichloromethane was

added in 5 min at -20 °C under nitrogen atmosphere to a

solution of boron trichloride (1 M in heptane, 15 mL, 15.0

5460 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

mmol) in 15 mL of dichloromethane in a 100 mL Schlenk flask.

At the end of the addition a yellow suspension was formed.

The reaction mixture was kept at -20 °C for 15 min and then

allowed to warm to room temperature. The color of the

suspension turned slowly from yellow to pink. A ¹H NMR

analysis showed that the reaction was already complete after

1 h of stirring at room temperature. After 4 h of stirring at

room temperature, the suspension was filtered on a G4 frit

and the residue dried to give a pink powder, which was

solution turned from colorless to yellow and a light exother-

micity was observed. The reaction mixture was stirred for 3.5

h at room temperature with final formation of a yellow ochre

suspension. The latter was neutralized by addition of a sodium

hydroxide aqueous solution (2.24 g of NaOH in 50 mL of

water). Additional water (∼50 mL) was added, and after 1 h

of stirring at room temperature, the resulting suspension was

filtered. The filtrate was discarded, whereas the residue was

dried in vacuo to give a yellowish powder as product (19.64 g,

purity ) 93.8%, yield ) 93.0%): mp 156.4-158.3 °C; ¹H NMR

(CDCl₃) δ 2.21 (s, 6H, CH₃), 6.03 (s, 1H, CHPh), 7.12-7.62

(m, 15H, Ar and NH); ¹³C NMR (CDCl₃) δ 8.48 (CH₃), 40.80

(CHPh), 108.64 (C), 110.79 (CH), 118.43 (CH), 119.41 (CH),

121.58 (CH), 127.25 (CH), 128.41 (CH), 128.92 (CH), 129.46

(C), 133.30 (C), 135.19 (C), 139.98 (C) (peak assigned by a

DEPT experiment); m/z (%) 351 (17) [M⁺+ 1], 350 (69) [M⁺],

349 (15), 335 (12), 273 (12), 257 (18), 256 (22), 221 (16), 220

(100), 219 (50), 218 (71), 217 (21), 207 (12), 206 (10), 205 (17),

204 (45), 175 (14), 144 (8), 131 (15), 130 (19), 129 (15), 128

(13), 77 (16).

N,N′-[Bis(tr ich lor obor an e)]-3,3′-Dih ydr o-3,3′-dim eth yl-

2,2′-d iin d olylm eth a n e (14). A solution of boron trichloride

(1 M in heptane, 14 mL, 14.0 mmol) was added at 0 °C in 5

min under nitrogen atmosphere to a solution of phenyl-3,3′-

dimethyl-2,2′-diindolylmethane (2.43 g, 6.9 mmol) in 27 mL

of dichloromethane in a 100 mL Schlenk flask. At the end of

the addition a dark green suspension was formed. The reaction

mixture was kept at 0 °C for 15 min, then allowed to warm to

room temperature, and stirred for 16 h. The solvent was

evaporated in vacuo to give a dark green powder as product

(3.51 g, 87.0%): ¹H NMR (CDCl₃) δ 0.96 (d, 6H, J ) 7.58 Hz,

CH₃), 4.24 (q, 2H, J ) 7.58 Hz, CH), 7.13-7.56 (m, 14H, CHPh

and Ar); ¹³C NMR (CDCl₃) δ 15.97 (CH₃), 43.42 (CH), 118.21

(CH), 122.63 (2CH), 124.76 (CH), 128.04 (CH), 128.38 (CH),

129.68 (CH), 130.52 (CH), 134.43 (C), 136.24 (C), 144.39 (C),

169.46 (C) (peak assigned by a DEPT experiment).

1

determined to be the desired product by H NMR analysis in

1

CD₂Cl₂(2.79 g, 79.4%): mp 184.8-185.6 °C; H NMR (CD₂-

Cl₂) δ 4.27 (bs, 2H, H3,H3′), 7.42-7.81 (m, 3H, Ar), 8.37-8.41

(m, 1H, Ar), 9.44-9.48 (m, 1H, H2); ¹H NMR (C₂D₂Cl₄) δ 4.19

(s, 2H, H3,H3′), 7.29-7.72 (m, 3H, Ar), 8.35-8.41 (m, 1H, Ar),

9.38-9.48 (m, 1H, H2).

The sample is not fully soluble in CD₂Cl₂or C₂D₂Cl₄.

Because of the low solubility of the complex, also at 120 °C in

C₂D₂Cl₄, attempts to acquire a ¹³C NMR failed. Nevertheless,

the complex is stable at high temperature for a few hours.

The synthesis of N-(trichloroborane)-3H-indole was carried

out also by using the same conditions reported above but by

adding the boron trichloride solution in heptane to the indole

solution in dichloromethane, obtaining the same results.

Tr iet h yla m m on iu m

(1H -In d ol-1-yl)t r ich lor ob or a t e

(13a ). Triethylamine (0.86 mL, 6.14 mmol) was added drop-

wise at room temperature under nitrogen atmosphere to a

suspension of 13 (1.44 g, 6.14 mmol) in 25 mL of dichlo-

romethane in a 50 mL Schlenk flask. During the addition a

light exothermicity was observed, and the starting pink

suspension became an orange solution. After 2 h of stirring at

room temperature, the solvent was removed in vacuo to give

a red-orange sticky solid, which was determined to be the

desired product by ¹H NMR analysis in CD₂Cl₂(1.98 g,

1

96.1%): mp 176.7-178.1 °C; H NMR (CD₂Cl₂) δ 1.27 [t, 9H,

J ) 7.34 Hz, N(CH₂CH₃)₃], 2.95 [m, 6H, J ) 7.34 Hz, N(CH₂-

CH₃)₃], 6.39 (d, 1H, J ) 2.35 Hz, H3 or H2), 7.04 (t, 1H, J )

7.43 Hz, H5 or H6), 7.11 (t, 1H, J ) 7.43 Hz, H6 or H5), 7.54

(d, 1H, J ) 7.43 Hz, H4 or H7), 7.73 (d, 1H, J ) 2.35 Hz, H2

or H3), 8.06 (d, 1H, J ) 7.43 Hz, H7 or H4), 9.51 (bs, 1H, NH).

Because of the low solubility of the complex in different

deuterated solvents, attempts to acquire a ¹³C NMR spectrum

failed.

Attem p ted Syn th esis of N,N′-[Tr is(p en ta flu or op h e-

n yl)b or a n e]-3,3′-Dih yd r o-3,3′-d im et h yl-2,2′-d iin d olyl-

m eth a n e. A solution of phenyl-3,3′-dimethyl-2,2′-diindolyl-

methane (1.01 g, 2.74 mmol) in 10 mL of dichloromethane was

added dropwise at 0 °C under nitrogen atmosphere to a

solution of tris(pentafluorophenyl)borane (2.84 g, 5.51 mmol)

in 15 mL of CH₂Cl₂in a 50 mL Schlenk flask. At the end of

the addition a black solution was formed. The reaction mixture

was kept at 0 °C for 15 min, then allowed to warm to room

temperature, and stirred for 20 h. Subsequent NMR analyses

showed that the reaction did not afford any adduct: the

starting phenyl-3,3′-dimethyl-2,2′-diindolylmethane was the

Attem p ted Syn th esis of N-(Tr ip h en ylbor a n e)-3H-In -

d ole. Triphenylborane (3.43 g, 14.2 mmol) was dissolved under

nitrogen atmosphere into 48 mL of toluene in a 100 mL

Schlenk flask equipped with a cooler. The resulting solution

was cooled to 0 °C and slowly added to a solution of indole

(1.68 g, 14.2 mmol) in 16 mL of toluene. No color change was

observed during addition. The reaction mixture was allowed

to warm to room temperature and stirred for 21 h. A ¹H NMR

analysis of the reaction mixture showed exclusively the

presence of starting materials. Then the reaction mixture was

1

only compound detected by H NMR analysis after 20 h.

N-[Tr is(p en t a flu or op h en yl)bor a n e]-7-Met h yl-3H-in -

d ole (15). A solution of 7-methylindole (0.45 g, 3.33 mmol) in

10 mL of dichloromethane was added at 0 °C under nitrogen

atmosphere to a solution of tris(pentafluorophenyl)borane (1.70

g, 3.30 mmol) in 16 mL of dichloromethane in a 50 mL Schlenk

flask. During the addition a color change from light yellow to

deep yellow was observed. Exothermicity was not observed.

At the end of the addition the solution was allowed to warm

to room temperature and stirred for 5 days. During this time

the reaction mixture color changed from dark yellow to orange

and finally became red. After 5 min after mixing at room

temperature, a mixture of 1:1.4 (% mol, calculated by ¹H NMR)

of the starting 7-methylindole and the expected complex was

formed together with lower amounts of byproducts. The

subsequent ¹H NMR analyses showed both a slow conversion

of 7-methylindole and a decomposition of 15 to a complex

mixture of byproducts, which were neither isolated nor char-

acterized. ¹H NMR (CD₂Cl₂) δ 2.35 (bs, 3H, CH₃), 4.21 (AB

system, 2H, J ) 26.41 Hz, H3,H3′), 6.91-7.59 (m, 3H, Ar),

8.97 (t, 1H, J _HF) 4.89 Hz, H2).

1

heated at 80 °C for 4 h. A second H NMR analysis of the crude

showed mainly unreacted indole and BPh₃.

Attem p ted Syn th esis of N-(Tr iflu or obor a n e)-3H-In -

d ole. Boron trifluoride diethyl etherate (5 mL, 5.60 g, 39.5

mmol) was added at room temperature to a solution of indole

(4.63 g, 39.5 mmol) in 30 mL of dichloromethane under

nitrogen atmosphere in a 50 mL Schlenk flask. At the end of

the addition a white suspension was formed. After 30 min of

stirring at room temperature, the reaction mixture was dried

in vacuo to give a white powder, which became pink upon

standing at room temperature. The color change indicates

possible decomposition of the final product. Because the

product was not soluble in CD₂Cl₂or other different deuterated

solvents, NMR characterization was not possible.

P h en yl-3,3′-d im eth yl-2,2′-d iin d olylm eth a n e.^13b3-Meth-

ylindole (15.12 g, 112.96 mmol) and benzaldehyde (6.07 g,

56.63 mmol) were dissolved under magnetic stirring into 100

mL of methanol in a 250 mL round-bottom flask. Then

hydrochloric acid (37% aq, 4.65 mL, 55.96 mmol) was added

dropwise at room temperature. During the addition the

N-[Tr is(p en ta flu or op h en yl)bor a n e]-2-P h en yl-3H-in -

d ole (16). (a ) Exp er im en t in CH₂Cl₂. A solution of 2-phe-

nylindole (0.99 g, 4.87 mmol) in 10 mL of dichloromethane was

J . Org. Chem, Vol. 68, No. 14, 2003 5461

Guidotti et al.

added at 0 °C under nitrogen atmosphere to a solution of tris-

(pentafluorophenyl)borane (2.50 g, 4.85 mmol) in 20 mL of

dichloromethane in a 50 mL Schlenk flask. At the end of the

addition the yellow solution was allowed to warm to room

temperature, stirred for 24 h, and followed by ¹H NMR

analysis in CD₂Cl₂. After 1 h of stirring at room temperature,

J ) 7.34 Hz, N(CH₂CH₃)₃], 1.73 [q, 6H, J ) 7.34 Hz, N(CH₂-

CH₃)₃], 6.28 (m, 1H, H4 or H5), 6.89 (bs, 1H, H5 or H4), 7.44

(bs, 1H, H2), 13.30 (bs, 1H, NH); NOESY (C₆D₆) δ¹H/δ¹H 13.30/

7.44 (NH/H2), 6.28/6.89 (H4/H5); ¹³C NMR (C₆D₆) δ 7.99

(N(CH₂CH₃)₃), 44.63 (N(CH₂CH₃)₃), 122.39 (C4 or C5), 125.26

(C5 or C4), 138.83 (C2).

1

a 1:1 (% mol, calculated by H NMR) mixture of the starting

2-[Tr is(p en ta flu or op h en yl)bor a n e]-1H-P yr a zole (18).

Pyrazole (0.70 g, 10.08 mmol) was dissolved at room temper-

ature under nitrogen atmosphere into 20 mL of dichlo-

romethane in a 50 mL Schlenk flask. A light yellow solution

of tris(pentafluorophenyl)borane (5.53 g, 10.74 mmol) in 20 mL

of dichloromethane was added at room temperature in sub-

sequent four steps (0.25 equiv each with respect to starting

pyrazole). After every addition, the reaction mixture was

stirred at room temperature for 15-20 min and followed by

¹H NMR analysis. At the end of the last addition the resulting

white suspension was dried in vacuo to give a white powder

as product (yield ) 100%): ¹H NMR (CD₂Cl₂) δ 6.70 (q, 1H, J

) 2.54 Hz, H4), 7.93 (d, 1H, J ) 2.54 Hz, H3), 7.94 (d, 1H, J

) 2.54 Hz, H5), 10.52 (bs, 1H, NH); ¹³C NMR (CD₂Cl₂) δ 108.26

2-phenylindole and the expected complex was formed. The

ratio between starting material and product did not change

during the 24 h. Attempts to shift the equilibrium to the

desired product by the addition of a second equivalent of tris-

(pentafluorophenyl)borane failed. ¹H NMR (CD₂Cl₂) δ 4.47

(broad AB system, 2H, J ) 23.87 Hz, H3,H3′), 7.10-7.78 (m,

9H, Ar).

(b) Exp er im en t in Et₂O. The reaction was carried out by

using the same conditions described in the previous experi-

ment, but diethyl ether was used as solvent instead of

dichloromethane. After 1 h of stirring at room temperature, a

1

2.4:1 (% mol, calculated by H NMR) mixture of the starting

2-phenylindole and the expected complex was obtained. The

ratio between starting material and product changed from

2.4:1 after 1 h to 2.8:1 after 24 h.

1

(C4), 133.62 (C5), 140.38 (C3); H NMR (C₆D₆) δ 5.41 (q, 1H,

J ) 2.69 Hz, H4), 6.16 (t, 1H, J ) 2.69 Hz, H5), 7.00 (bs, 1H,

H3), 8.58 (bs, 1H, NH); ¹³C NMR (C₆D₆) δ 106.79 (C4), 132.66

(C5), 139.59 (d, J _CF) 1.72 Hz, C3). The H3 and H5 protons in

the ¹H NMR spectra were assigned by using the HSQC

analysis. Indeed, the H3 proton is closer to B(C₆F₅)₃than H5,

and its intensity is lower due to the coupling with fluorine

atoms of B(C₆F₅)₃.

3-[Tr is(pen taflu or oph en yl)bor an e]-1H-Im idazole (17).

A colorless solution of imidazole (0.31 g, 4.50 mmol) in 5 mL

of dichloromethane was added at room temperature under

nitrogen atmosphere to a light yellow solution of tris(pen-

tafluorophenyl)borane (2.33 g, 4.55 mmol) in 13 mL of dichlo-

romethane in a 25 mL Schlenk flask. During the addition the

color of the solution turned from light yellow to yellow.

Exothermicity was not observed. The reaction mixture was

stirred for 1 h at room temperature, and then the solvent was

evaporated in vacuo to give a white powder as product (2.60

By decoupling the NH, it appeared that the multiplicity in

C₆D₆of the H4 proton changed from a quartet to a triplet and

that the multiplicity of the H5 proton changed from a triplet

to a doublet; thus, both H4 and H5 couple with NH: ¹H NMR

(C₆D₆) δ 5.41 (t, 1H, J ) 2.69 Hz, H4), 6.16 (d, 1H, J ) 2.69

Hz, H5), 7.00 (bs, 1H, H3), 8.58 (bs, 1H, NH).

1

g, 99.6%): mp 214.9-217.8 °C; H NMR (C₆D₆) δ 5.49 (t, 1H,

J ) 1.76 Hz, H5), 6.30 (bs, 1H, H4), 6.66 (bs, 1H, H2), 7.24

1

¹H NMR analysis at variable temperatures: T ) 215 K, H

(bs, 1H, NH); NOESY (C₆D₆) δ¹H/δ¹H 7.24/6.66 (NH/H2), 7.24/

NMR (CD₂Cl₂) δ 6.70 (q, 1H, J ) 2.45 Hz, H4), 7.92 (bs, 1H,

H5), 7.97 (t, 1H, J ) 2.45 Hz, H3), 10.61 (m, 1H, NH). Attempts

to link a second molecule of B(C₆F₅)₃to the complex pyrazole-

B(C₆F₅)₃failed.

1

5.49 (NH/H5), 6.30/5.49 (H4/H5); H NMR (CD₂Cl₂) δ 7.18-

7.24 (m, 2H, H4 and H5), 8.08 (bs, 1H, H2), 10.05 (bs, 1H,

NH); ¹³C NMR (CD₂Cl₂) δ 117.81 (C5), 126.66 (C4), 136.22 (C2).

Tr ieth yla m m on iu m [Tr is(p en ta flu or op h en yl)](1H-im -

id a zol-1-yl)bor a te (17a ). A solution of triethylamine (0.231

g, 2.27 mmol) in 5 mL of dichloromethane was added dropwise

at room temperature under nitrogen atmosphere to a solution

of 17 (1.254 g, 2.16 mmol) in 15 mL of dichloromethane in a

25 mL Schlenk flask. During the addition the color of the

solution turned from light yellow to yellow, and exothermicity

was not observed. After overnight stirring at room tempera-

ture, a ¹H NMR analysis in CD₂Cl₂showed complete conver-

sion of the starting material, and then the solvent was removed

in vacuo to give a white powder as product (1.35 g, 90.8%,

purity ) 99%): ¹H NMR (CD₂Cl₂) δ 1.27 [t, 9H, J ) 7.24 Hz,

N(CH₂CH₃)₃], 3.01 [q, 6H, J ) 7.24 Hz, N(CH₂CH₃)₃], 6.93 (t,

1H, J ) 1.47 Hz, H4 or H5), 7.00 (bs, 1H, H5 or H4), 7.62 (m,

1H, H2), 13.76 (m, 1H, NH); ¹³C NMR (CD₂Cl₂) δ 9.02

[N(CH₂CH₃)₃], 45.82 [N(CH₂CH₃)₃], 121.92 (C4 or C5), 125.44

(C5 or C4), 138.89 (C2).

Tr ieth ylam m on iu m [Tr is(pen taflu or oph en yl)](1H-pyr a-

zol-1-yl)bor a te (18a ). A solution of triethylamine (0.264 g,

2.60 mmol) in 5 mL of dichloromethane was added dropwise

at room temperature under nitrogen atmosphere to a solution

of 18 (1.51 g, 2.60 mmol) in 15 mL of dichloromethane in a 25

mL Schlenk flask. A white suspension was obtained. After 1

1

h of stirring at room temperature, a H NMR analysis in C₆D₆

showed complete conversion of the starting material, and then

the solvent was removed in vacuo to give a white powder as

product (1.74 g, 98.2%): ¹H NMR (C₆D₆) δ 0.18 [t, 9H, J )

7.30 Hz, N(CH₂CH₃)₃], 1.80 [q, 6H, J ) 7.30 Hz, N(CH₂CH₃)₃],

5.93 (t, 1H, J ) 1.96 Hz, H4), 6.60 (d, 1H, J ) 1.96 Hz, H3 or

H5), 7.53 (bs, 1H, H5 or H3), 9.67 (bs, 1H, NH); ¹H NMR (CD₂-

Cl₂) δ 1.14 [t, 9H, J ) 7.34 Hz, N(CH₂CH₃)₃], 2.98 [q, 6H, J )

7.34 Hz, N(CH₂CH₃)₃], 6.30 (t, 1H, J ) 2.20 Hz, H4), 7.48 (d,

1H, J ) 2.20 Hz, H3), 7.66 (bs, 1H, H5), 10.81 (bs, 1H, NH);

¹³C NMR (CD₂Cl₂) δ 8.19 [N(CH₂CH₃)₃], 46.49 [N(CH₂CH₃)₃],

104.26 (C4), 137.12 (C5), 138.66 (C3); NOESY (CD₂Cl₂) δ¹H/

δ¹H 6.30/7.48 (H4/H3), 6.30/7.66 (H4/H5), 7.48/10.81 (H3/NH),

7.48/1.14 [H3/N(CH₂CH₃)₃], 7.48/2.98 [H3/N(CH₂CH₃)₃].

N-[Tr is(p en ta flu or op h en yl)bor a n e]p yr r olid in e (19). A

solution of pyrrolidine (0.34 g, 4.78 mmol) in 3 mL of dichlo-

romethane was added at room temperature under nitrogen

atmosphere to a solution of tris(pentafluorophenyl)borane (2.44

g, 4.77 mmol) in 15 mL of dichloromethane in a 25 mL Schlenk

flask. A light exothermicity was observed. The reaction

mixture was stirred for 30 min at room temperature, and then

the solvent was evaporated in vacuo to give a white powder

as product (yield ) 100%): mp 205.0-206.9 °C; ¹H NMR (CD₂-

Cl₂) δ 1.84-2.09 (m, 4H, H3, H3′, H4, H4′), 2.68-2.86 (m, 2H,

H2 and H5), 3.44-3.54 (m, 2H, H2′ and H5′), 6.30 (bs, 1H,

NH); ¹³C NMR (CD₂Cl₂) δ 23.86 (C3 and C4), 50.37 (C2 and

C5).

(a ) Exp er im en t in On e Step . A solution of imidazole

(0.1619 g, 2.38 mmol) in 5 mL of dichloromethane was added

dropwise at room temperature under nitrogen atmosphere to

a solution of tris(pentafluorophenyl)borane (1.3175 g, 2.56

mmol) in 15 mL of dichloromethane in a 50 mL Schlenk flask.

1

After 1 h of stirring at room temperature, a H NMR analysis

in CD₂Cl₂showed complete conversion of the starting material

to the imidazole-B(C₆F₅)₃complex. Then a solution of triethy-

lamine (0.2604 g, 2.56 mmol) in 5 mL of dichloromethane was

added at room temperature. After overnight stirring at the

same temperature, the solvent was removed in vacuo to give

a white powder as product (1.45 g, 88.5%, purity ) 99%): mp

1

62.5-63.3 °C; H NMR (CD₂Cl₂) δ 1.28 [t, 9H, J ) 7.34 Hz,

N(CH₂CH₃)₃], 3.02 [q, 6H, J ) 7.34 Hz, N(CH₂CH₃)₃], 6.92 (t,

1H, J ) 1.22 Hz, H4 or H5), 6.99 (bs, 1H, H5 or H4), 7.58 (m,

1H, H2), 14.21 (bs, 1H, NH); ¹H NMR (C₆D₆) δ 0.37 [t, 9H,

5462 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

N-[Tr is(p en ta flu or op h en yl)bor a n e]in d olin e (20). In-

doline (0.57 g, 4.70 mmol) was added at room temperature

under nitrogen atmosphere to a solution of tris(pentafluo-

rophenyl)borane (2.41 g, 4.70 mmol) in 17 mL of dichlo-

romethane in a 25 mL Schlenk flask. During the addition the

color of the solution turned from light yellow to yellow. The

reaction mixture was stirred for 1 h at room temperature, and

then the solvent was evaporated in vacuo to give a whitish

powder as product (2.95 g, 99.4%), mp 231.3-235.0 °C.

In a repeated experiment, a solution of indoline (0.60 g, 5.07

mmol) in 3 mL of dichloromethane was added at room

temperature under nitrogen atmosphere to a solution of tris-

(pentafluorophenyl)borane (2.60 g, 5.07 mmol) in 15 mL of

dichloromethane in a 25 mL Schlenk flask. A light exother-

micity was observed. The reaction mixture was stirred for 30

min at room temperature, and then the solvent was evaporated

in vacuo to give a white powder as product (yield ) 100%):

¹H NMR (CD₂Cl₂) δ 2.16-2.34 (m, 1H, H3), 2.91-3.06 (m, 1H,

H3′), 4.05-4.20 (m, 2H, H2,H2′), 7.10-7.40 (m, 4H, Ar), 8.15

(bs, 1H, NH); ¹³C NMR (CD₂Cl₂) δ 29.29 (C3), 52.68 (t, C2,

Syn th esis of (C₆F ₅)₃B(OH₂).⁵A solution of tris(pentafluo-

rophenyl)borane (0.3196 g, 0.6 mmol) in 15 mL of pentane was

treated at room temperature with H₂O (0.0325 g, 1.8 mmol)

in a 50 mL Schlenk flask. Neither a color change nor exother-

micity was observed. The resulting suspension was stirred for

1 h at room temperature, and then an additional 2 equiv of

B(C₆F₅)₃(0.6181 g, 1.2 mmol) suspended in 10 mL of pentane

was added. After 3 h of stirring at room temperature, the

solvent was evaporated in vacuo, giving (C₆F₅)₃B(OH₂) as a

white solid (yield ) 100%): ¹H NMR (C₆D₆) δ 4.56 (s, 2H, H₂O).

Rea ction of In d ole w ith (C₆F ₅)₃B(OH₂). A solution of

(C₆F₅)₃B(OH₂) (0.9064 g, 1.7 mmol) in 12 mL of toluene was

added at room temperature to a solution of indole (0.2026 g,

1.7 mmol) in 8 mL of toluene in a 25 mL Schlenk flask. During

the addition the solution became yellow, but exothermicity was

not observed. The reaction mixture was stirred for 3 days at

room temperature, and its path was constantly followed by

¹H NMR analysis. At the end the solvent was evaporated in

vacuo to give 2 as a whitish solid (0.4841 g, yield ) 45.3%).

The low yield is due to the many aliquots taken for NMR

analysis.

J _CF) 4.6 Hz), 120.64 (C7), 125.94 (C4), 128.57 (C5 or C6),

129.86 (C6 or C5), 135.51 (C3a or C7a), 140.35 (C7a or C3a).

A possible intermediate species was observed in the NMR

spectra made immediately after the mixing of the starting

materials: ¹H NMR (CD₂Cl₂) δ 3.92 (dd, 2H, J ) 8.61 Hz, J )

3.91 Hz), 5.87 (t, 1H, J ) 8.61 Hz), 7.15-7.83 (m, 4H), 8.68

(bs, 1H); ¹³C NMR (CD₂Cl₂) δ 35.33, 61.64, 113.03-132.42.

The progressive disappearance of NMR signals of the

intermediate species matched with the increase of NMR

signals of 2.

N-[Tr is(p en t a flu or op h en yl)b or a n e]-2-Met h yl-1-p yr -

r olin e (21). A solution of 2-methyl-1-pyrroline (0.39 g, 4.41

mmol) in 3 mL of dichloromethane was added at room

temperature under nitrogen atmosphere to a solution of tris-

(pentafluorophenyl)borane (2.26 g, 4.41 mmol) in 15 mL of

dichloromethane in a 25 mL Schlenk flask. A light exother-

micity was observed. The reaction mixture was stirred for 30

min at room temperature, and then the solvent was evaporated

in vacuo to give a white-gray powder as product (yield )

Rea ction of In d ole w ith (C₆F ₅)₃B(OH₂) a n d (C₆F ₅)₃B.

A solution of indole (0.2081 g, 1.8 mmol) in 10 mL of

dichloromethane was added at room temperature to a mixture

of (C₆F₅)₃B(OH₂) (0.2812 g, 0.5 mmol) and (C₆F₅)₃B (0.6365 g,

1.2 mmol) in 15 mL of dichloromethane in a 50 mL Schlenk

flask. During the addition the solution became yellow. The

reaction mixture was stirred for 1 day at room temperature,

and its path was constantly followed by ¹H NMR analysis.

Then the solvent was evaporated in vacuo to give 2 as a

whitish solid (0.9566 g, yield 84.6% with respect to starting

indole).

1

100%): mp 200.3-201.9 °C; H NMR (CD₂Cl₂) δ 2.13 (s, 3H,

CH₃), 1.94-2.36 (m, 2H, H4, H4′), 3.00-3.08 (m, 2H, H3, H3′),

4.00-4.58 (bm, 2H, H5, H5′); ¹³C NMR (CD₂Cl₂) δ 19.65 (CH₃),

21.09 (C4), 42.64 (C3), 62.05 (C5), 191.95 (C2).

The complex 22 was analyzed at variable temperatures by

¹H NMR in CD₂Cl₂. By cooling the temperature from 298 to

238 K, the H4 and H4′ protons became a more complex

multiplet, whereas the H5 or H5′ proton became a broad

triplet.

At t em p t ed Syn t h esis of Wa t er -F r ee B(C₆F ₅)₃. (a )

B(C₆F₅)₃(22.0 g) was suspended at room temperature under

nitrogen atmosphere into 100 mL of dry hexane (<10 ppm of

water) in a 250 mL Schlenk flask. The temperature was slowly

incresead to 60 °C with an oil bath. After 15 min of stirring at

60 °C, the borane was almost totally dissolved into the solvent.

The mixture was then filtered on a G4 frit, keeping the

temperature at 60 °C. The light yellow-beige filtrate was

collected into a jacketed reactor, and then the temperature was

cooled from 60 to 25 °C in 2 h. The crystallization of the

product started at 50 °C under stirring. At the end the product

was collected by filtration on a G4 frit under nitrogen at room

temperature. After drying in vacuo, 10.0 g of B(C₆F₅)₃was

obtained as a white powder. The borane contained 4870 ppm

of water by Karl Fischer analysis (∼0.5 wt %) and had an mp

of 112.8-113.9 °C.

(b) B(C₆F₅)₃(3.68 g, containing 0.2 wt % of water by Karl

Fischer analysis corresponding to 0.41 mmol of water) was

suspended at room temperature under nitrogen atmosphere

into 100 mL of dry pentane (<10 ppm of water) in a 250 mL

Schlenk flask. Almost all of the product was dissolved into the

solvent. Then 24.0 mg of dry methylalumoxane (MAO, 0.41

mmol) was added at room temperature. After 10 min of stirring

at the same temperature, a gray precipitate was formed. The

mixture was decanted for 20 min and the supernatant liquid

filtered on a G4 frit. The decanted gray precipitate was washed

with an additional 50 mL of pentane, and all of the washings

were passed through the frit. The different filtrates were

collected and dried in vacuo to give a white powder as product.

The latter contained <20 ppm of water by Karl Fischer

analysis and <2 ppm of aluminum.

X-r a y Cr ysta l Str u ctu r e An a lysis. Suitable crystals of

2, 2a , and 20 were mounted in air on a glass fiber tip onto a

goniometer head. Single-crystal X-ray diffraction data were

collected on a Siemens SMART CCD area detector diffracto-

meter for 2 and 2a and on an Enraf-Nonius CAD-4 diffracto-

meter for 20, using graphite-monochromated Mo KR radiation

(λ ) 0.71073 Å) at room temperature [295(2) K].

For 2 and 2a unit cell parameters were initially obtained

from ∼100 reflections taken from 45 frames collected in three

different ω regions and eventually refined against ∼3000

reflections with 3° e θ e 25°, whereas for 20 the setting angles

of 25 randomly distributed intense reflections with 10° e θ e

14° were processed by least-squares fitting.

For 2 and 2a a full sphere of reciprocal space was scanned

by 0.3° ω step, collecting 1800 frames (the exposure times were

40 and 20 s for 2 and 2a , respectively). Intensity decay was

monitored by re-collecting the initial 50 frames at the end of

data collection and analyzing the duplicate reflections. After

integration, an empirical absorption correction was made on

the basis of the symmetry-equivalent reflection intensities

measured (for 2, 4745 reflections, with an average redundancy

of 2.0; for 2a , 11594 reflections, with an average redundancy

of 3.8).²⁸

For 20, data collection was performed by the ω scan method

with variable scan speed (maximum time per reflection of 60

s) and variable scan range (0.90 + 0.35 tan θ). Crystal stability

under diffraction was checked by monitoring three standard

reflections every 180 min. The measured intensities were

corrected for Lorentz, polarization, and background effects and

reduced to F_o². An empirical absorption correction was applied

J . Org. Chem, Vol. 68, No. 14, 2003 5463

Guidotti et al.

TABLE 5. Su m m a r y of Cr ysta l Da ta , Da ta Collection , a n d Str u ctu r e Refin em en t P a r a m eter s for 2, 2a , a n d 20

2

2a

20

formula

formula wt

crystal system

space group

a (Å)

C

₂₆H₇BF₁₅

N

C₃₂H₂₂BF₁₅N₂

730.33

monoclinic

P2₁/n (no. 14)

12.058(5)

C₂₆H₉BF₁₅

631.15

monoclinic

P2₁/n (no. 14)

13.586(1)

N

629.14

triclinic

P1h (no. 2)

9.906(5)

10.969(6)

11.670(6)

107.17(1)

93.02(1)

101.54(1)

1178.6(11)

2

b (Å)

c (Å)

18.061(8)

14.683(7)

10.882(1)

16.261(2)

R (deg)

â (deg)

93.760(10)

98.06(1)

γ (deg)

V (Å³)

3191(2)

4

2380.3(4)

4

Z

F(000)

620

1.773

0.187

colorless plate

0.18 × 0.10 × 0.04

1.8 e θ e 25.0

-11 e h e 11

-13 e k e 13

-13 e l e 13

none

1472

1.520

0.150

colorless block

0.26 × 0.26 × 0.24

1.8 e θ e 25.0

-14 e h e 14

-21 e k e 21

-17 e l e 17

none

1248

1.761

0.185

colorless block

0.30 × 0.28 × 0.28

3.0 e θ e 25.0

-16 e h e 15

0 e k e 12

0 e l e 19

none

density (g cm^-3

)

absorption coeff (mm^-1

crystal description

crystal size (mm)

θ range (deg)

)

index ranges

intensity decay (%)

transmission factors (min, max)

no. of measured reflns

0.870, 0.993

8464

0.868, 0.965

26047

0.936, 0.950

4311

no. of ind reflns

4135

0.0249, 0.0381

3272

3272/388

0.045, 0.3

1.039

0.0366

0.0900

0.211, -0.181

5640

0.0296, 0.0245

3896

3896/388

0.05, 1.8

1.107

0.0492

0.1235

0.260, -0.200

4165

0.0146, 0.0116

3447

3447/388

0.05, 0.6

1.047

0.0333

0.0861

0.210, -0.211

a

R

_int, R_σ

no. of reflns with I > 2σ(I)

no. of data/parameters

weights (a, b)^b

goodness-of-fit S(F²)^c

R(F)^d

wR(F²)^e

largest diff peak, hole (e Å^-3

)

a

2

b

2

R_int) ∑|F_o- F_mean|/∑|F_o|; R_σ) ∑|σ(F_o²)|/∑|F_o|. w ) 1/[σ²(F_o²) + (aP)²+ bP], where P ) (F_o+ 2F_c²)/3. ^cS ) [∑w(F_o- F_c²)²/(n -

p)]^1/2, where n is the number of reflections and p is the number of refined parameters. R(F) ) ∑||F_o| - |F_c||/∑|F_o|. ^ewR(F²) ) [∑w(F_o^2-F_c²)²/

d

4

1/2

∑wF_o

] .

using ψ scan of three suitable reflections having ø values close

to 90°.²⁹Crystal data and data collection parameters are

summarized in Table 5.

the fluorine atoms of the pentafluorophenyl substituents. Final

conventional agreement indices and other structure refinement

parameters are listed in Table 5.

The structures were solved by direct methods (SIR97³⁰) and

subsequent Fourier synthesis; they were refined by full-matrix

least-squares on F²(SHELX97³¹) using reflections with I >

2σ(I). Scattering factors for neutral atoms and anomalous

dispersion corrections were taken from the internal library of

SHELX97. Weights were assigned to individual observations

according to the formula w ) 1/[σ²(F_o²) + (aP)²+ bP], where

Com p u ta tion a l Deta ils. All of the reported structures

represent stationary points on the potential energy surface.

They were calculated with the Amsterdam Density Functional

(ADF) program system,³⁴developed by Baerends et al.^35,36The

electronic configurations of the molecular systems were de-

scribed by sets for B, C, N, O, F (2s,2p) augmented with a

single 3d function and a double-ú STO basis set for hydrogen

(1s), augmented with a single 2p function.^37,38The frozen-core

approximation was applied up to the [He] core for all of the

non-hydrogen atoms. Energetics and geometries were evalu-

ated by using the local exchange-correlation potential by Vosko

et al.,³⁹augmented in a self-consistent manner with Becke’s⁴⁰

exchange gradient correction and Perdew’s^41,42correlation

gradient correction (BP86). Geometry optimizations were

terminated if the largest component of the Cartesian gradient

was <0.003 au. Transition state geometry of the proton

transfer from H₂O-B(C₆F₅)₃to C₂of pyrrole was approached

by a linear-transit procedure, whereas all other degrees of

2

P ) (F_o+ 2F_c²)/3; a and b were chosen to give a flat analysis

of variance in terms of F_o². Anisotropic parameters were

assigned to all non-hydrogen atoms. All hydrogen atoms were

clearly evidenced from difference Fourier maps; they were

placed in idealized position and refined riding on their parent

atom with an isotropic displacement parameter 1.2 times that

of the pertinent parent atom.

The structure of 2a is affected by a disorder of the triethyl-

ammonium cation. Because it was difficult to refine a consis-

tent disordered model (possibly because of the large volumes

252 Å³sof the cavity occupied by the cation), its contribution

was subtracted from the observed structure factor according

to the BYPASS procedure,³²as implemented in PLATON.³³

The final difference electron density map showed no features

of chemical significance, with the largest peaks lying close to

(34) ADF 2.3.0; Vrije Universiteit Amsterdam: Amsterdam, The

Netherlands, 1996.

(35) Baerends, E. J .; Ellis, D. E.; Ros, P. J . Chem. Phys. 1973, 2,

41.

(36) te Velde, B.; Baerends, E. J . J . Comput. Phys. 1992, 99, 84.

(37) Snijders, J . G.; Baerends, E. J .; Vernooijs, P. At. Nucl. Data

Tables 1982, 26, 483.

(38) Vernoijs, P.; Snijders, J . G.; Baerends, E. J . Department of

Theoretical Chemistry, Free University Amsterdam, Amsterdam, The

Netherlands, 1981.

(29) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.

1968, A24, 351.

(30) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;

Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,

R. J . Appl. Crystallogr. 1999, 32, 115.

(31) Sheldrick, G. M. SHELX97; Universita¨t Go¨ttingen, Go¨ttingen,

Germany, 1997.

(32) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A46, 194.

(33) Spek, A. L. Acta Crystallogr. 1990, A46, C34.

(39) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J . Phys. 1980, 1200.

(40) Becke, A. Phys. Rev. A 1988, 38, 3098.

(41) Perdew, J . P.; Zunger, A. Phys. Rev. B 1986, 33, 8822.

(42) Perdew, J . P. Phys. Rev. B 1986, 34, 7406.

5464 J . Org. Chem., Vol. 68, No. 14, 2003

Generation of Highly Acidic sp³Carbons in Pyrroles and Indoles

freedom were optimized. A full transition state search was

started from the geometry corresponding to the maximum

along the linear-transit curve and terminated if the largest

component of the Cartesian gradient was <0.002 au. Atomic

coordinates for the optimized structures of 22_{in d}and 22_{p yr r}are

reported in the Supporting Information.

Su p p or tin g In for m a tion Ava ila ble: Variable-tempera-

ture H NMR spectra of compounds 4, 5, 7, and 8; 2D spectra

1

of compounds 1-5, 7, 9, 11, 12, 17, 17a , 18, and 18a ; tables

of crystal data, structure solution and refinement, atomic

coordinates, bond lengths and angles, and anisotropic dis-

placement parameters for compounds 2, 2a , and 20; and

atomic coordinates of the computed structures of compounds

22_{in d}and 22_{p yr r}. This material is available free of charge via

the Internet at http://pubs.acs.org.

Ack n ow led gm en t. We thank M. Colonnesi for the

proton NMR spectra and Dr. Moskau (Bruker AG, CH)

for the 2D ¹⁹F-¹H experiments. We also thank Prof. F.

Gasparrini, Universita` La Sapienza, Roma, for the PC

version of the DNMR-6 program.

J O020647X

J . Org. Chem, Vol. 68, No. 14, 2003 5465

Article Doi

Synthesis and reactivity of (C6F5)3B-N-heterocycle complexes. 1. Generation of highly acidic sp3 carbons in pyrroles and indoles

DOI: 10.1021/jo020647x

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo020647x

Full text of DOI:10.1021/jo020647x

Products guided by the article

R&D Labs maybe for 333334-62-6

Relevant to this article

Hot Product