Ambident reactivity of the nitrite ion revisited

Article abstract of DOI:10.1002/anie.200501274

(Chemical Equation Presented) Lost control: The rationalization of the ambident reactivity of NO₂^- by the change between charge control to orbital control has to be revised. SN1-type reactions of carbocations with NO₂^- give kinetically controlled product mixtures only when these reactions proceed without activation energy (diffusion control). Activation-controlled S_N1 alkylations are reversible and lead to the thermodynamically more stable nitro compounds.

Full text of DOI:10.1002/anie.200501274

Angewandte
Chemie
Ambident Anions
This confusion prompted us to reinvestigate the ambident
reactivity of the nitrite ion with emphasis on absolute rate
constants.
Ambident Reactivity of the Nitrite Ion
Revisited**
We have recently reported that the knowledge of absolute
rate constants and of the diffusion control limit is a key to
understanding the ambident reactivity of the thiocyanate[4a]
and cyanide ions[4b] in alkylation reactions. These ambident
anions are preferentially attacked at the soft S or C termini by
hard and soft alkylating agents. Only when diffusion control is
reached, can the attack of carbocations at the hard nitrogen
compete with attack at sulfur (SCNꢀ) or carbon (CNꢀ),
respectively. We now report on the kinetics and selectivities of
alkylations of the nitrite ion.
Alexander A. Tishkov, Uli Schmidhammer, Stefan Roth,
Eberhard Riedle, and Herbert Mayr*
Dedicated to Professor Horst Kessler
on the occasion of his 65th birthday
Extensive investigations on the reactions of the nitrite ion
with alkylating agents[1] led Kornblum to the conclusion “The
greater the carbonium contribution to the transition state the
greater is the yield of nitrite ester and the smaller is the yield of
Reactions of the blue benzhydrylium salts (1a–d)-BF4
(see Table 1) with nBu4N+NO2 in anhydrous acetonitrile
ꢀ
[Eq. (5)], which gave colorless adducts, were followed photo-
nitroparaffin.”[1a] However, Pearsonꢀs specification of this
ꢀ
rule, “tC4H9Cl reacts with the hard oxygen atom of NO2
,
while the softer CH3I reacts with the softer nitrogen atom,”[2a]
which was expressed by Equation (1) in later theoretical
treatments of ambident reactivity,[2b–d] is not in line with
experimental findings, because CH3I and other primary
haloalkanes yield mixtures of alkyl nitrites and nitroalkanes
with either NaNO2 or AgNO2.[1,2e]
metrically using the stopped-flow technique described pre-
viously.[5] For [1a–d]0 = (1–10) ” 10ꢀ6 m and [NO2 ]0 = (1–
ꢀ
100) ” 10ꢀ5 m, pseudo-first-order conditions were employed
resulting in exponential decays of the benzhydrylium absorp-
tions, from which the pseudo-first-order rate constants k1Y
(sꢀ1) were derived. The second-order rate constants k2
(mꢀ1 sꢀ1) listed in Table 1 were obtained as the slopes of the
plots of k1Y vs. [NO2 ].[6] From the linearity of these
ꢀ
Flemingꢀs statement “Although silver nitrite does react
with alkyl halides to give nitrites, sodium nitrite gives more
nitroalkane than alkyl nitrite” [Eqs. (2) and (3)][3a] is contra-
dicted by Streitwieser, Heathcock, and Kosower, who com-
pare the reactions of iodoalkanes with NaNO2 and AgNO2
and conclude “Yields of nitroalkane are higher when silver
nitrite is used, but this added economy is tempered by the cost
of silver salt” [Eq. (4)].[3b]
correlations one can infer that ion pairing does not play a
role under these conditions.
The reactions with the highly electrophilic benzhydrylium
ions 1e–k (see Table 1) were too fast to be followed by
stopped-flow techniques. Therefore, these carbocations were
generated photolytically in the presence of the nitrite ion. For
this purpose, solutions of (1e–h)-NO2 (obtained according to
AgNO2 þ RBr ! RONO
NaNO2 þ RBr ! RNO2
ð2Þ
ð3Þ
Table 1: Rate constants of the reactions of nitrite ions with the benzhydrylium
cations 1a–k (208C, MeCN).
Benzhydrylium ion
E[a]
k2 [mꢀ1 sꢀ1
]
n=1
n=2
1a ꢀ10.04 1.58”105
CH3ðCH2Þ6CH2I þ AgNO2 ! CH3ðCH2Þ6CH2NO2 ð83 %Þ
þCH3ðCH2Þ6CH2ONO ð11 %Þ
ꢀ9.45 3.30”105
ð4Þ
1b
[*] Dr. A. A. Tishkov, Prof. Dr. H. Mayr
Department Chemie und Biochemie
n=1
n=2
1c
1d
ꢀ8.76 9.85”105
ꢀ8.22 2.76”106
Ludwig-Maximilians-Universität München
Butenandtstrasse 5–13 (Haus F), 81377 München (Germany)
Fax: (+49)89-2180-77717
E-mail: herbert.mayr@cup.uni-muenchen.de
R=NMe2
R=NMePh
R=NMe(CH2CF3) 1g
R=NPh(CH2CF3) 1h
R=OMe
R=Me
R=H
1e
1 f
ꢀ7.02 1.56”107
ꢀ5.89 1.49”108
ꢀ3.85 1.09”109
Dipl.-Phys. U. Schmidhammer, Dipl.-Phys. S. Roth, Prof. Dr. E. Riedle
Lehrstuhl für BioMolekulare Optik
Ludwig-Maximilians-Universität München
ꢀ3.14 3.41”109
Oettingenstrasse 67, 80538 München (Germany)
[b]
1i
1j
1k
0.00 –
[**] This research was supported by the Alexander von Humboldt
Foundation (stipend to A.A.T.) and the Deutsche Forschungsge-
meinschaft.
3.63 2.50”1010
5.90 2.2”1010 [c]
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
[a] Electrophilicity parameter as defined by Equation (6), see ref. [5a]. [b] Not
determined. [c] From ref. [6].
Angew. Chem. Int. Ed. 2005, 44, 4623 –4626
DOI: 10.1002/anie.200501274
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4623

Communications
Eq. (5)) or 1j-Cl and nBu4N+NO2 in acetonitrile were
irradiated by laser pulses (< 1 ps, ca. 1.5 mJ, center
wavelengths between 250 and 300 nm) to yield the
benzhydrylium ions 1e–j in less than 2 ps. The behavior
of these transients (lmax = 460–612 nm) was monitored
ꢀ
on different timescales, and we detected reactions with
[7]
1
10 ms > t = > 10 ns. As in the stopped-flow experi-
2
ments, these reactions were also performed under
pseudo-first-order conditions [Ar2CH+] ! [NO2 ], and
ꢀ
Scheme 1. Reactions of benzhydrylium tetrafluoroborates with nBu4N+NO2
(equilibrium constants K in CH3CN at 208C).
ꢀ
the second-order rate constants listed in Table 1 were
ꢀ
again derived from plots of k1Y vs. [NO2 ].
In previous work we have shown that the reactions of
carbocations with nucleophiles can be described by Equa-
tion (6), where k is the second-order rate constant, E is a
chemical shift of the nitro group (d = 3.8 ppm) in 1e-NO2.[10]
For the least reactive carbocations 1a–c an excess of
nBu4N+NO2 is needed to drive the reactions toward
ꢀ
log k20 o ¼ sðE þ NÞ
ð6Þ
completion. The equilibrium constants, which were deter-
mined by UV/Vis spectroscopy (see the Supporting Informa-
tion), are listed in Scheme 1.[11]
C
nucleophile-independent electrophilicity parameter, and N
and s are electrophile-independent nucleophilicity parame-
ters.[5,8] Figure 1 correlates the second-order rate constants
listed in Table 1 with the previously determined electro-
philicity parameters of benzhydrylium ions[5] and shows that
These observations clearly indicate that the photometri-
cally monitored process for the reactions of 1a–c with NO2ꢀ is
N attack: Since alkyl nitrites are thermodynamically less
stable than the corresponding nitro compounds,[12] the equi-
librium constants for the formation of (1a–c)-ONO must be
much smaller, and one can exclude that the observed rates of
carbocation decay are due to the formation of (1a–c)-ONO,
which rearrange to the observed nitro compounds (1a–c)-
NO2 in consecutive reactions.[13]
ꢀ
the reactions of carbocations 1a–f with NO2 follow Equa-
tion (6).
For the more electrophilic benzhydrylium ions 1d–h one
cannot a priori exclude the possibility that the UV/Vis
spectroscopically monitored process is the formation of
(1d–h)-ONO, which successively rearrange to the observed
nitro compounds (1d–h)-NO2. However, since the obtained
rate constants for the reactions of 1d–h with NO2ꢀ are on the
same correlation line as the rate constants for N attack at 1a–c
(Figure 1), the observed rate constants are compatible only
with exclusive N attack or with simultaneous N and O attacks
that occur with similar rates. In none of these cases could a
biexponential decay be observed as in the reactions of some
benzhydrylium ions with SCNꢀ, where the faster reversible S
attack was followed by the slower irreversible N attack.[4a]
Pure benzhydrylium nitrites 1i-ONO and 1j-ONO were
obtained by treatment of the corresponding benzhydrols with
NOCl/Et3N (Scheme 2), while analogous attempts to synthe-
Figure 1. Plot of lgk2 for the reactions of the nitrite ion with the benz-
hydrylium ions 1a–k versus their electrophilicity parameters E (208C,
MeCN).[5a]
From the linear part of this correlation (1a–f) one can
derive N = 17.2 and s = 0.72 for the nitrite ion in acetonitrile.
For the more electrophilic benzhydrylium ions 1g–k, a
flattening of the curve can be seen, and the observed diffusion
limit is in accord with the previously reported value of (2.2–
2.7) ” 1010 mꢀ1 sꢀ1 for reactions of carbocations with anions in
acetonitrile.[6]
Which reactions are responsible for the decay of the
benzhydrylium absorbances? Addition of an excess of
nBu4N+NO2 to solutions of benzhydrylium tetrafluorobo-
ꢀ
rates (1a–h)-BF4 in CD3CN gives rise to the exclusive
[9]
formation of the nitromethanes (1a–h)-NO2 (Scheme 1),
1
as revealed by the characteristic H NMR chemical shifts of
the benzhydryl protons (d = 6.5–6.8 ppm) and the 15N NMR
Scheme 2. i: Ar=4-MeOC6H4; j: Ar=4-MeC6H4.
4624
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 4623 –4626

Angewandte
Chemie
size (1a–h)-ONO failed.[14] Whereas pure 1j-ONO remained
unchanged when stored at room temperature for several days,
1i-ONO rearranged within two days at 08C into 1i-NO2,
which decomposed to 2i either spontaneously or upon
treatment with nBu4N+NO2ꢀ in MeCN (Scheme 2).[15]
Both benzhydrylium nitrites 1i-ONO and 1j-ONO were
in contrast to statements in the secondary literature (e.g., see
ref. [3a]) silver ions were reported not to exert a significant
influence on the O/N selectivity (Table 2, entries 1–4).[2e] On
the other hand, the methyl nitrite/nitromethane ratio is
changed significantly by variation of the solvent (Table 2,
entries 5–8).[17]
recovered unchanged when kept in a solution of nBu4N+NO2
We have now studied the reactions of nBu4N+NO2 with
ꢀ
ꢀ
in acetonitrile for 3 h. These experiments show that benzhy-
dryl nitrites 1i-ONO and 1j-ONO are kinetically stable under
the conditions of the experiment described in Scheme 3.
Treatment of the benzhydrylium halides 1i-Cl and 1j-Br with
soft (MeI) and hard (MeOSO2Me, MeOSO2CF3, and
Me3O+BF4 ) methylating agents in CDCl3. The yields were
ꢀ
determined by integration of the 1H NMR spectra with
MeOAc as the internal standard. Overall yields were close
to quantitative in all cases, and the methyl nitrite/nitro-
methane ratios (Table 2, entries 9–12) clearly indicate that the
effect of the leaving groups on the O/N selectivities is small.
Changing from the soft methylating agent MeI to the hard
Me3O+ changes the O/N selectivities by only a factor of 2.4,
while a larger difference is found among the hard methylating
agents MeOSO2Me and MeOSO2CF3.
In conclusion, neither SN1 nor SN2 alkylations of the
nitrite ion are satisfactorily described in the framework of
perturbation molecular orbital theory (PMO). The reactions
ꢀ
of carbocations with NO2 only yield kinetically controlled
product mixtures when k > 1 ” 109 mꢀ1 sꢀ1, that is, when they
proceed without an activation barrier. Therefore PMO
approaches, such as the Klopman–Salem model, which aim
Scheme 3. X=Cl, Br.
nBu4N+NO2ꢀ in MeCN gave 7:2 mixtures of
Table 2: O/N selectivities for methylations of nitrite ions.[a]
1i-ONO and benzophenone 2i, and 1j-
ONO and benzophenone 2j, respectively,
which were analyzed by comparison of the
NMR spectra of the product mixtures with
those of authentic compounds. Since 1i-
ONO and 1j-ONO are stable under the
conditions of the experiment, and the ben-
zophenones 2i and 2j are consecutive
products of the corresponding diarylnitro-
methanes 1i-NO2 and 1j-NO2, the product
ratios described in Scheme 3 imply a kO/kN
ratio of 3.4 for the diffusion-controlled
Entry
Nitrite
Methylation agent
Solvent
Me-ONO/Me-NO2
1
2
3
4
5
6
7
8
9
AgNO2
NaNO2
AgNO2
NaNO2
KNO2/[18]crown-6
KNO2/[18]crown-6
KNO2/[18]crown-6
KNO2/[18]crꢀown-6
nBu4N+NO2ꢀ
nBu4N+NO2ꢀ
nBu4N+NO2ꢀ
nBu4N+NO2
MeI
MeI
MeI
MeI
MeOSO2Me
MeOSO2Me
MeOSO2Me
MeOSO2Me
MeI
DMSO
DMSO
DMF
46:54[b]
54:46[b]
54:46[b]
54:46[b]
30:70[c]
50:50[c]
92:8[c]
DMF
EtOH
MeCN
THF
C6H6
85:15[c]
30:70[d]
32:67[d]
50:50[d]
59:41[d]
CDCl3
CDCl3
CDCl3
CDCl3
10
11
12
MeOSO2Me
ꢀ
MeOMe2+BF4
ꢀ
MeOSO2CF3
reactions of NO2 with the benzhydrylium
ions 1i and 1j.
[a] At room temperature. [b] Determined by LC with radioactivity detection for reactions of 11CH3I (see
ref. [2e]). [c] Determined by GC (see ref. [17]). [d] Determined by integration of H NMR spectra (this
work, see the Supporting Information).
1
We thus conclude that diffusion-control-
led reactions of benzhydrylium ions with
ꢀ
NO2 occur preferentially at the O termi-
nus.[16] Since Figure 1 shows that carboca-
tions with E > 0 (i.e., carbocations less stabilized than 1i)
at comparing relative activation energies cannot be applied.
Furthermore, O/N selectivities of the methylations of the
nitrite ion (Table 2) also do not correlate well with the
ꢀ
generally undergo diffusion-controlled reactions with NO2
,
it is evident that the reactions of NO2ꢀ with most carbocations
typically generated in SN1 processes (e.g., tert-alkyl, allyl, and
benzyl cations) proceed without an activation barrier.
ꢀ
hardness of the methylating agents. Thus, NO2 is the third
prototype of the ambident anions (after SCNꢀ and CNꢀ)[4] for
which the previously accepted rationalization by frontier
orbital theory has to be withdrawn.
ꢀ
Explanations of the ambident reactivity of NO2 based on
transition-state models, therefore, cannot apply. On the other
hand, those benzhydryl nitrites, which are formed by activa-
ꢀ
Received: April 12, 2005
tion-controlled reactions of benzhydrylium ions with NO2
,
are thermodynamically unstable and rapidly isomerize to the
corresponding nitro compounds.
Let us now turn to SN2 reactions with the nitrite ion. Silver
ꢀ
ion was proven to assist the cleavage of the C Hal bond
Keywords: HSAB principle · kinetics · Klopman–Salem
equation · Kornblum’s rule · nitro compounds
.
during the reactions of AgNO2 with alkyl halides.[1a] However,
Angew. Chem. Int. Ed. 2005, 44, 4623 –4626
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4625

Communications
N. L. Holy, R. C. Kerber, M. T. Musser, D. H. Snow, J. Am.
Chem. Soc. 1967, 89, 725 – 727.
[1] a) N. Kornblum, R. A. Smiley, R. K. Blackwood, D. C. Iffland, J.
Am. Chem. Soc. 1955, 77, 6269 – 6280; b) N. Kornblum, H. O.
Larson, R. K. Blackwood, D. D. Mooberry, E. P. Oliveto, G. E.
Graham, J. Am. Chem. Soc. 1956, 78, 1497 – 1501; c) N.
Kornblum, R. A. Smiley, H. E. Ungnade, A. M. White, B.
Taub, S. A. Herbert, Jr., J. Am. Chem. Soc. 1955, 77, 5528 –
5533; d) N. Kornblum, H. O. Larson, D. D. Mooberry, R. K.
Blackwood, E. P. Oliveto, G. E. Graham, Chem. Ind. 1955, 78,
443; e) N. Kornblum, Organic Reactions, Vol. 12 1962, 101 – 156;
f) N. Kornblum, L. Fishbein, R. A. Smiley, J. Am. Chem. Soc.
1955, 77, 6261 – 6266.
[12] a) J. F. Liebman, M. S. Campbell, S. W. Slayden in The Chemistry
of Amino, Nitroso, Nitro and Related Groups, Supplement F2
(Ed.: S. Patai), Wiley, Chichester, 1996, chap. 8, p. 340; b) B. M.
Rice, S. V. Pai, J. Hare, Combust. Flame 1999, 118, 445 – 458;
c) “Neutral Thermochemical Data”: H. Y. Afeefy, J. F. Liebman,
S. E. Stein in NIST Chemistry WebBook (Eds.: P. J. Linstrom,
W. G. Mallard), NIST Standard Reference Database Number 69,
National Institute of Standards and Technology, Gaithersburg
MD, 20899, 2003 (http://webbook.nist.gov).
[13] These results disagree with the claim ꢀthat malachite green ((4-
Me2NC6H4)2PhC+) reacts with NO2 in acetone and ethyl
acetate yielding (4-Me2NC6H4)2PhC-ONO as the kinetic product
followed by slow (for a UV/Vis spectrometer with a regular
mixing mode) irreversible formation of the more stable (4-
Me2NC6H4)2PhC-NO2: O. H. Abed, N. S. Isaacs, J. Chem. Soc.
Perkin Trans. 2 1983, 839 – 842. Since the electrophilicity
parameter of (4-Me2NC6H4)2PhC+ (E = ꢀ10.29, ref. [8b]) is
comparable to that of 1a (E = ꢀ10.04), the rate constant
reported by Isaacs for the reaction of (4-Me2NC6H4)2PhC+
[2] a) R. G. Pearson, J. Songstad, J. Am. Chem. Soc. 1967, 89, 1827 –
1836; b) G. Klopman, J. Am. Chem. Soc. 1968, 90, 223 – 234;
c) R. F. Hudson, Angew. Chem. 1973, 85, 63 – 84; Angew. Chem.
Int. Ed. Engl. 1973, 12, 36 – 57; d) S. J. Formosinho, L. G. Arnaut,
J. Chem. Soc. Perkin Trans. 21989, 1947 – 1952; e) K.-O. Schoeps,
C. Halldin, S. Stone-Elander, B. Långström, T. Greitz, J.
Labelled Compd. Radiopharm. 1988, 25, 749 – 758.
[3] a) I. Fleming, Frontier Orbitals and Organic Chemical Reactions,
Wiley, Chichester, 1976, p. 41 – 42; b) A. Streitwieser, C. H.
Heathcock, E. M. Kosower, Introduction to Organic Chemistry,
Macmillan, New York, 1992, p. 779; This statement is in line with
Kornblumꢀs report (p. 114 in ref. [1e]): “In the synthesis of
saturated primary nitro compounds, silver nitrite gives nitro-
paraffins in about 80% yield as against about 60% yields
obtained with sodium nitrite. While silver nitrite is, therefore, the
reagent of choice here, the lower cost and ready availability of
sodium nitrite, the not very large disparity in yield, and the shorter
reaction time all combine to make sodium nitrite an excellent
second choice.”
ꢀ
with NO2 in acetonitrile is more than six orders of magnitude
smaller than expected for the formation of either (4-
Me2NC6H4)2PhC-NO2 or (4-Me2NC6H4)2PhC-ONO.
[14] Addition of NOCl to solutions of benzhydrols (1a–h)-OH in the
presence of Et3N led to the immediate development of blue
color characteristic for the corresponding benzhydryl cations.
[15] a) Ph2CH-NO2 was reported to convert into benzophenone
spontaneously: H. Feuer, H. Friedman, J. Org. Chem. 1975, 40,
187 – 190; b) For the conversion of secondary nitro compounds
into ketones in the presence of NO2ꢀ, see: N. Kornblum, R. K.
Blackwood, D. D. Mooberry, J. Am. Chem. Soc. 1956, 78, 1501 –
1504; c) A. Gissot, S. NꢀGouela, C. Matt, A. Wagner, C.
Mioskowski, J. Org. Chem. 2004, 69, 8997 – 9001.
[16] Preferential or exclusive formation of nitrites was reported for
reactions of tert-alkyl halides as well as 4-methoxybenzyl
bromide and 1-phenylethyl chloride with AgNO2 in Et2O (see
refs. [1a,c,f]).
[17] H. Lemmetyinen, L. Lehtinen, J. Koskikallio, Finn. Chem. Lett.
1979, 6, 72 – 75.
[4] a) R. Loos, S. Kobayashi, H. Mayr, J. Am. Chem. Soc. 2003, 125,
14126 – 14132; b) A. A. Tishkov, H. Mayr, Angew. Chem. 2005,
117, 145 – 148; Angew. Chem. Int. Ed. 2005, 44, 142 – 145.
[5] a) H. Mayr, T. Bug, M. F. Gotta, N. Hering, B. Irrgang, B. Janker,
B. Kempf, R. Loos, A. R. Ofial, G. Remennikov, H. Schimmel, J.
Am. Chem. Soc. 2001, 123, 9500 – 9512; b) H. Mayr, B. Kempf,
A. R. Ofial, Acc. Chem. Res. 2003, 36, 66 – 77.
[6] J. Bartl, S. Steenken, H. Mayr, J. Am. Chem. Soc. 1991, 113,
7710 – 7716.
[7] The laser flash photolysis was performed using state-of-the-art
femtosecond pulses as excitation and light-emitting diodes
(LEDs) as sources of monitoring beams. New instrumentation
was developed that is compatible with standard pump probe
setups. Details of this method will be published separately: U.
Schmidhammer, S. Roth, E. Riedle, A. A. Tishkov, H. Mayr,
Rev. Sci. Instrum. 2005, submitted.
[8] a) S. Minegishi, R. Loos, S. Kobayashi, H. Mayr, J. Am. Chem.
Soc. 2005, 127, 2641 – 2649; b) S. Minegishi, H. Mayr, J. Am.
Chem. Soc. 2003, 125, 286 – 295; c) R. Lucius, R. Loos, H. Mayr,
Angew. Chem. 2002, 114, 97 – 102; Angew. Chem. Int. Ed. 2002,
41, 91 – 95; d) T. Bug, H. Mayr, J. Am. Chem. Soc. 2003, 125,
12980 – 12986; e) T. Bug, T. Lemek, H. Mayr, J. Org. Chem.
2004, 69, 7565 – 7576.
[9] Because of the ease of the transformation of diarylnitrome-
thanes into the corresponding benzophenones (see Scheme 2)
these reactions were performed in NMR tubes and analyzed
immediately after the reagents had been mixed.
[10] D. M. Grant, R. K. Harris, Encyclopedia of Nuclear Magnetic
Resonance, Vol. 5, Wiley, Chichester, 1996, p. 3232.
[11] The nonobservance of tetraarylethanes indicates that electron-
transfer processes do not occur under these conditions; see: a) N.
Kornblum, L. Cheng, T. M. Davies, G. W. Earl, N. L. Holy, R. C.
Kerber, M. M. Kestner, J. W. Manthey, M. T. Musser, H. W.
Pinnick, D. H. Snow, F. W. Stuchal, R. T. Swiger, J. Org. Chem.
1987, 52, 196 – 204; b) N. Kornblum, T. M. Davies, G. W. Earl,
4626
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 4623 –4626