close

Вход

Забыли?

вход по аккаунту

?

Free or Not That Is the Question Silyl and Germyl Cations in Condensed Phases.

код для вставкиСкачать
HIGHLIGHTS
Free or Not, That Is the Question:
Silyl and Germyl Cations in Condensed Phases
Johannes B e h e r *
“Even after a very short experience, it was evident that the
corresponding derivatives of the two elements showed very considerable differences in their chemical properties. . .’’.[’] The two
elements that F. S. Kipping, one of the fathers of organosilicon
chemistry, was talking about are carbon and silicon. One notable difference is, for instance, the pronounced instability of compounds with triply substituted silicon. With the synthesis of
persistent disilenes and silenes in 1981, which showed that compounds with triply bonded silicon may well be sufficiently stable
to allow their isolation, the question arose whether this would
also be true of silyl cations.”] Everything seemed favorable:
silicon is more electropositive than carbon and should therefore
be better able to stabilize a positive charge. Ab initio calculations confirm this: the reaction enthalpy of the isodesmic reaction (1) for R = His -57.4 kcalmol-‘ at the MP2(FC)/6-31G*
R S I H ~+RCH,
H
---f
RSiH,-H +RCHT
(1)
level of theory ; this shows that [SiH,]+ is thermodynamically
significantly more stable than the methyl
In agreement with these findings, it is easy to generate silyl cations in the
gas phase, where the chemistry of these species has been explored thoroughly over the years.[41But what about silyl cations
in condensed phases, that is, in solution or in the solid state?
Silyl cations were postulated early on as short-lived intermediates in the reactions of organosilicon compounds, but conclusive evidence for their existence was missing.f51More recent
studies especially kinetics have, however, provided evidence for
a positively charged silicon intermediate in solution;161nevertheless, the structures of these intermediates could not be deduced from the experimental results. Not surprisingly, many
efforts were aimed at synthesizing and characterizing stable silicon analogues of carbenium ions in solution and in the solid
state. Kipping’s aforementioned differences between silicon and
carbon compounds again became apparent :I1] all reactions and
techniques successfully used for the generation and isolation of
stable carbenium ions[’] failed for silyl cations. Why?
The principle difference between silicon and carbon in their
compounds is the tendency of silicon to react with Lewis bases
with expansion of its coordination sphere. If neutral silicon
[*] Dr. J. Beizner
Institut fur Organische Chemie der Universitlt
Tammannstrasse 2, D-37077 Gottingen (Germany)
Fax: Int. code +(551)39-9475
e-mail: jbelzneio gwdg.de
A q e w Ctuw. In!. Ed. EnxI. 1997. 34, No. 12
already displays a tendency for higher coordination numbers,
what is to be expected of electron-deficient compounds like silyl
cations? Their proverbial “voracious appetite”[*] for nucleophiles of any kind has been not exaggerated in view of the
computed stability of coordination complexes of silyl cations
with noble gases.f3a1This “appetite for electrons” is also reflected in the existence of numerous stable highly coordinated silyl
cations. The structure of such a compound, pyridine adduct 1,
was determined in 1983 for the first time; the dative Si-N bond
causes pyramidalization of the otherwise planar silyl cation.[”]
Even the isolation of doubly charged cations like 21gb1or 3”01 is
possible, provided they are stabilized by coordination to a base.
Me
i
S
‘
,
-N
s
]
r
1-
2+
Me
2 c1-
Me 1
Me
1
Me
2
2 c1-
3
The observed 29Si NMR shift to relatively high field showed
clearly that the positive charge is not localized on silicon, but is
transferred to the coordinated group. These species may be
more appropriately called silyl ammonium ions. The formation
of the 2-silanorbornyl cation 5, which is stable in toluene for
several weeks after formation by hydride abstraction from
silane 4 [Eq. (2)], shows that the C-C double bond can also
satisfy the appetite of the silyl cation for electrons.[’ I 1
VCH Verlugsgesellxhuft mbH, 0.49451 Wrinheim.1997
OS70-0833/97i3612-1277$1750+.5010
1277
HIGHLIGHTS
191 pm. This, together with the observation that
the geometry of the toluene molecule is only slightly distorted, led Lambert et al. to call this a com[Ph3CIi[B(C&dJ
lB(ceF5)41 (2)
pound a silyl cation with a weakly coordinated
- Ph,CH
toluene molecule. Many objected immediately :I1
L
the pronounced pyramidalization of the Et,Si
4
5
fragment was completely inconsistent with the description of a silyl cation. The solid-state structure,
How should one proceed experimentally to avoid nuclehowever, correlated very well with a computed structure of the
ophilic attack on a newly formed silyl cation? Three compleo-Wheland complex 6, the intermediate for electrophilic attack
mentary strategies of proven value for the generation of stable
of [Et,Sij+ on t ~ l u e n e [ ’ ~ ~ , ~ ] ;
carbenium ions are available: 1) The use of counterions and
alternatively, an q‘-bonded
solvents of very low nucleophilicity to reduce the chance of
7~ complex was suggested.
nucleophilic attack on the cationic silicon center. 2) Sterically
The main argument against
M
e
m
i
IB(C&AI demanding substituents at the silicon center should hamper the
a silyl cation is that the
approach of a nucleophile. The problem here is that silicon is
experimentally determined
6
larger than carbon, so that substituents which effectively shield
29Si NMR chemical shift
a carbon center from nucleophiles of a certain size may allow the
(6 = 81.8 in toluene), which
same nucleophile to reach a silicon center. 3) A third possibility
is shifted very far to high field relative to the calculated shift of
is to stabilize the silyl cation thermodynamically by substitution
[Me,Si]+ (6 = 355), is close to the computed chemical shift of
with NR, or SR groups, the electron lone pairs of which should,
6 = 60r19b1for a [Me,Si.-. toluene]+ complex (6 = 82.1 for
in principle, be capable of back-bonding into the empty silicon
~ ,was
~ ~ )argued
.
that the coordinative
[Et,Si .. . t ~ l u e n e ] + [ ’ ~ It
p orbital. However, the corresponding stabilization energies,
interaction between the silyl cation and toluene is stronger than
calculated for Equation (1) for different substituents R, are
initially assumed by Lambert et al.,[19c1and that the positive
much smaller than for the analogous carbenium ions.[3.I z l
charge was mostly transferred from the silicon center onto
In spite of extensive utilization of these three strategies, espetoluene. Based on the 2ySiNMR chemical shift as a criterion for
cially in the groups of J. Lambert, G. A. Olah, and C. Reed, no
the “freedom of a silyl cation”, there is no doubt that the comcomplete success was achieved for a long time in the search for
pounds [R,Si]+[CB, lHa16H6]- (R,Si = Et,Si, iPr,Si, tBu,Si,
persistent silyl cations in condensed phases. Lambert et al., for
and tBu,MeSi), all of which display ”Si NMR resonances in
instance, reported that hydride abstraction from silanes R,SiH
the range of 6 = 105-115, are not “free” silyl cations. X-Ray
(R = Ph, MeS, EtS, zPrS) by trityl perchlorate leads to silyl
structure analyses also showed that in these compounds the
perchlorates for which NMR data, molecular mass determinaanion is coordinated through its halogen atom to the silicon
tion, and conductivity measurements in sulfolane and acetocenter in the crystal. Based on the ’,Si NMR chemical shifts and
nitrile seem to imply the existence of ionic structures.[’31 In
geometric parameters such as the C-Si-C bond angle, the cationcontrast, Olah et al. showed by using 35Cl and 29Si NMR
ic character was estimated to be around 55-70%.r’7b*c1At
spectroscopy that-at
least at concentrations around
times the concept of a free silyl cation seemed to be a reason for
0.12 mol L-’-the perchlorate group in Ph,SiOClO, is bound
a quarrel over terminology; nowadays there is a consensus that
covalently to silicon through an oxygen atom.“4a1Furthermore,
the transition between coordinated and free silyl cations is conthey suggested that Lambert et al.’s results, derived from meatinuous. The crux of the matter is how to define and how to
surements in highly diluted solutions, were corrupted by inrealize such a free silyl cation, whatever its definition may be.
evitable hydrolysis at these concentrations.[14b1Subsequently,
Three ethyl groups, as used by Lambert et al., certainly do not
Lambert et al. tried to refute these objections,[’51and the quarrel
provide optimal shielding for the silicon center. It seemed obviover the question of whether water was really present in these
ous to use kinetically stabilizing substituents such as mesityl
systems, and if so how much, would have continued if the introgroups whose ortho-methyl substituents should hinder the apor the
duction of less nucleophilic anions such as [B(C6FJ4]proach of, at least, the large nucleophiles. However, the otherhalogenated carboranes [CB,Br,H,]-[’
and [CB, ,Ha16H6]wise so useful hydride abstraction method (i.e., the reaction of
had not advanced the chemistry of silyl
(Hal = C1, Br, I)r17b*cJ
a silane with a trityl salt) could then become difficult. Since the
cations.
silicon center in the starting material is sterically shielded by the
There was great excitement in 1993 when Lambert et al. remesityl substituents, it may be difficult for the trityl cation to
ported the isolation of a silyl cation “with no coordination to
remove hydride from the silane. This is indeed the case: trimeanion and distant coordination to solvent”.~’slUndoubtedly,
sitylsilane does not react with trityl salts, even on prolonged
the anion, and especially its fluorine substituent, was so remote
heating. Lambert et a1.‘’O1 have now used a surprisingly simple
from the silicon center in the solid-state structure of
trick to solve this problem. The allyltrimesitylsilane 7 operates
[Et,Si]+[B(C,F,),]- that a bonding interaction could be excludas if the ally1 substituent were a “rip cord” [Eq. (3)] that aids in
ed with certainty. However, two toluene molecules per unit cell
liberating the silyl cation. Electrophile 8 reacts with the allylic
were incorporated into the crystal on recrystallization. The
double bond, which is located in the accessible outer region of
shortest distance between the silicon center and the aromatic
the sterically crowded silane 7. As a consequence, the Si-C
para-carbon atom was 218 pm, which is significantly longer
bond p to the carbenium ion 9,already weakened by hyperconjugation, breaks. The larger the angle at the silicon center due to
than the average Si-C bond length, typically between 186 and
i siEt31
1278
0 VCH Verlugsgesellschuf! mbH, 0-69451 Weinheim, 1997
0570-0833]97/3612-1278 $17.50+ .SO10
Angeii. Chem. Int. Ed. Engl. 1997.36, N o . 12
HIGHLIGHTS
[Mes3Si] [B(C6F5)J +
Mes,Si
r
1
10
7
__
+
(3)
+
L
1
9
[Et3SiCHzCPhzl +p(C,F,),l
8
steric hindrance, the faster the bond cleavage. Compound 10,
obtained by this technique, has a 29SiNMR signal at 6 = 225.5
in benzene. This record value, which agrees well with the computed chemical shift of [Ph,Si]+ (6 = 208.9), shows that the
positive charge is mostly if not completely localized on the silicon center. The chemical shift is practically the same in other
solvents. Upon addition of acetonitrile, the ”Si NMR signal
moves to higher field (6 = 37.0); thus this nucleophile is small
enough to bypass the mesityl substituents and reach the silicon
center. In any event, the N M R data have to be interpreted as an
indication that no solvent interaction occurs in aromatic solvents. A weak coordination with the anion cannot be excluded
completely but seems unlikely.
In contrast to the (almost) never-ending story of silyl cations,
very little has been reported on the experimental generation of
stable germyl cations. Cation 11, which contains an alkyl group
and two transition metal fragments as substituents, is stabilized
by coordination with a Lewis base.[’’] The short Ge- W bond in
complex 12 demonstrates that the positive charge at the triply
l+
11
12
coordinated germanium center is delocalized by back-bonding
to the tungsten atom.[221The combination of steric shielding
and electronic stabilization has now made it possible to isolate
another germyl ati ion.''^] The Ge, moiety in 13 is a slightly
distorted triangle whose
edges are between a GeGe double and a Ge-Ge
Si(fBu),
single bond length in the
Ge-Ge,
][Bph41solid state. These struc(iBU),SI ’
sl(m)3
tural parameters suggest
that the positive charge is
13
delocalized over the threemembered ring. This delocalization, together with steric shielding by the large silyl substituents, suppresses nucleophilic attack at the cation. Compound 13 is best described as a germanium analogue of the
smallest Hiickel aromatic system, the cyclopropenylium ion.
This confirms the results of ab initio studies which predicted
that the [Ge,H,]+ ion is a local minimum. Its aromatic stabilization energy of 31.9 kcal mol- I , however, is considerably smaller
(by 26.8 kcal mol- ’) than that of the cyclopropenylium
[
k,
Angrit. Clirm. lnf Ed. Engl. 1997, 36, No. 12
the cyclopropenylium ion.[241Thus, despite all differences, germanium cannot deny its relationship to carbon: the aromaticity
principle is valid in both cases, albeit to varying degrees.
German version: Angew. Chem. 1997. 109, 1331-1334
Keywords: cations * germanium
silicon * structure elucidation
-
nucleophilic substitution
-
111 F. S Kipping, Proc. R. SOC.London. Ser. A 1937, 159, 139 -148.
[2] Just as difficult as the search for a stable silyl cation in condensed phases ISthe
search for a proper name for this class of compounds, acceptable to all researchers involved While silyl cations were named siliconium tons (which was
consistent since carbenium ions were designated as carbonium ions at that
time) in what is probably the first publication postulating a silicon analogue of
a carbenium ion as an intermediate [F. C. Whitmore, L. H Sommer, J. Gold,
J Am. Chem. SOC.1947,60,1976-19771, this name is now commonly used for
cationic silicon complexes with a coordination number higher than four. Since
then the terms silylenium ion, silicenium ion, or silylium ion have been used for
triply coordinated positively charged silicon compounds. The latter name is
also suggested by IUPAC IG. J. Leigh, Nomenclature of Inorganic Chembrry,
Blackwell Scientific Publishing, Oxford, 1990, p. 1061; alternatively, the term
silyl cation is suggested.
131 a) C. Maerker, J. Kapp, P. von R. Schleyer in OrganosilironChemisrrj,II(Eds.:
N. Auner, J. Weis), VCH, Weinheim, 1996, 329-359; b) L Olsson, C.-H.
Ottosson, D. Cremer, J Am. Chem. SOC.1995, 117, 7460~7479.
[4] a) V. D. Nefedov, T. A. Kochina, E. N. Sinotova, Russ. Chem. Rev. 1986. 55,
426-438; b) H. Schwarz in The Chemisfryoforganic Silicon compounds (Eds.:
S. Patai, Z. Rappoport), Wiley, Chichester, 1989, 4 4 - 5 1 0 ; c) J. Kapp, P. R.
1996. I I N , 12154-12158.
Schreiner, P. von R. Schleyer, J Am. Chem. SOC.
[5] a) R. J. P Corriu, M. Henner, J Orgonomet. Chenz. 1974. 74, 1-28; b) J.
Chojnowski, W. Stanczyk, Adv. Organome!. Chem. 1990. 30.243-307.
[6] See, for example: a) J. Chojnowski, W Fortuniak, W. Stanczyk, J Am. Chem.
So?. 1987, 109,7776-7781; b) H. Mayr, N. Basso, G. Hagen, ibid. 1992,114,
3060-3066
171 G. A. Olah, Angew. Chem. 1995,107, 1519-1532; Angen (‘krm. Int. Ed. Engl.
1995,34, 1393-1405.
[8] K . N. Houk, Chemtrucis: Org. Chem. 1993, 215, 360-363.
[9] a) K. Hensen, T. Zengerly, P. Pickel, G. Klebe, Angew. Chem. 1983, 95, 739139; Angen. Chem. In/ Ed. Engl. 1983,22,725; b) K. Hensen, T. Zengerly. T.
Miiller, P Pickel, Z . Anorg. Allg. Chem. 1988, 558. 21 -27.
[lo] F. Carre. C. Chuit, R. J. P. Corriu, A. Mehdi, C. Reye. Angen. Chem. 1994,
f06,1152-1154; Angew,. Chem. Int. Ed. Engl. 1994, 33. 1097-1099.
Ill]
Steinberger, T Miiller, N. Auner, C. Maerker. 1.’ von R. Schleyer,
Angen. Chem. 1997, 109, 667-669; Angew. Chem. Int Ed. Engl. 1997, 36,
626 - 628.
[12] U. Pidun, M. Stahl, G. Frenking, Chem. Eur J. 1996. 2. X69- 876.
[13] a) J. B. Lambert, W J. Schulz, Jr., J Am. Chem. Soc. 1983, 105. 1671-1672;
b) I B. Lambert. J. A. McConnell, W. J. Schulz, Jr., ihid. 1986, 108,2482-2484;
c) J. B. Lambert. W. J. Schulz, Jr., J A. McCohnell. W. Schilf. ihrd. 1988, 110,
2201 -2210.
[14] a) G. K. S. Prakash. S. Keyaniyan, R. Anzisfeld, L. Heiliger, G. A. Olah. R. C.
Stevens. H.-K. Choi, R. Bau. J: Am. Chem. Soc. 1987. 109. 5123-5126; b)
G. A. Olah, L. Heiliger, X:Y. Li, G . K. S. Prakash, rhrd. 1990, 112, 59915995.
1151 a) J. B. Lambert, L. Kania, W Schilf, Jr.. J. A. McConnell. Orgnnometa//rcs
1991, IO,2578-2584; for summaries, see b) P. D. Lickiss. J Chem. SOC.Dalton
Trans. 1992, 1333-1338; c ) J B. Lambert. L. Kania. S Zhang, Chem. Re,,.
1995, 95. 1191-1201.
[16] J. B. Lambert, S. Zhang, J Chem. SOC.Chrm. Commun 1993, 383-384.
[17] a) 2. Xie, D J. Liston, T. Jelinek, V. Mitro, R. Bau, C. A. Reed, J Chem. Soc.
Chem. Commun. 1993. 384-386; b) Z. Xie, R. Bau, A. Benesi. C. A. Reed,
Organomelollic.r 1995. 14, 3933-3941 ; c) Z. Xie, J. Mannig, R. W. Reed, R.
Mathur, P D. W. Boyd, A. Benesi, C. A. Reed, J. Am. Chem. Soc. 1996. 118,
2922-2928.
C VCH Verlagsgesell.rchu/tmbH. 0-69451
Weinheim, 1997
0570-0833i97i3612-127Y $ 17.50+ 50/0
1279
HIGHLIGHTS
[18] J. B. Lambert, S. Zhang, C. L. Stern, J. C. Huffrnan, Science 1993,260, 19171918.
[19) a) L. Olsson, D. Cremer, Chem Phys. Left 1993, 215, 433-443; b) P. von R.
Schleyer, P. Buzek, T.Miiller, Y Apeloig, H.-U. Siehl, Angew. Chem. 1993, 105,
1558-1561; AngeM.. Chem. Int. Ed. Engl. 1993,32,1471; c) L. Pauling, Science
1994.263,983: d) G. A. Olah, G. Rasul, X.-Y Li, H. A. Buchholz, G. Sandford, G. K. S. Prakash, ibid. 1994, 263, 983-984; e) G. A. Olah, G. Rasul,
X : Y Li, H. A. Buchholz, G. K. S. Prakash, Bull. Soc. Chim. Fr. 1995, 132.
569-574.
[20] J. B. Lambert, Y. Zhao. Angew. Chem. 1997,109,389-391; Angew. Chem. Int.
Ed. Engl. 1997, 36, 400-401.
[21] J. Fujlta, Y Kawano, H. Tobita, H. Ogino. Chem. Lett. 1994, 13531356.
[22] L. K. Figge, P. J. Carroll, D. H. Berry, Angen’. Chem. 1996, 108. 465-467;
Angew. Chem. I n t . Ed. Engl. 1996, 35, 435-437.
[23] A. Sekiguchi, M. Tsukamoto, M. Ichinohe, Scrence 1997. 275, 60-61.
[24] E. D. Jernmis. G. N. Srinivas. J. Leszczynski, J. Kapp, A. A. Korkin, P. von
R. Schleyer,J. Am. Chem. SOC.1995, 117, 11361-11362.
Combinatorial Chemistry for the Synthesis of Carbohydrate Libraries
Prabhat Arya* and Robert N. Ben*
The use of combinatorial libraries in the identification and
elucidation of structure-activity relationships has become a
powerful tool in the pharmaceutical sector.[’] Traditionally,
novel lead compounds were obtained as natural products from
a number of sources including extracts from plants, animals,
insects, or microorganisms. When an extract shows a desired
biological activity, the active compound is identified, isolated,
and then subjected to further biological testing. Optimization of
the chemical structure to enhance biological activity is a laborintensive, time-consuming process, which dictates that each new
structure be independently synthesized. This overall approach
has made the development of new therapeutics a very lengthy
and expensive process.
In contrast, combinatorial chemistry has provided an attractive alternative to these traditional synthetic approaches since it
allows for the synthesis of a large number of structurally diverse
compounds within a short period of time. The approach utilizes
a large array of building blocks that are systematically assembled in such a way that all possible combinations are represented. Typically, a solution or solid-phase approach may be used in
conjunction with either a “split” or “parallel” synthetic strategy. Although the technology required to assemble a small molecule library is not new, combinatorial chemistry was not fully
exploited until recently, since efficient methods for screening such
libraries were virtually nonexistent. Many of these screening
strategies, as well as technical aspects of combinatorial chemistry, have been summarized in several well-written reviews.[’]
Unlike for protein -protein and nucleotide- protein interactions, progress in understanding the role of cell-surface carbohydrates in biological and pathological processes has been
Although comparatively little is known about these
weak, noncovalent interactions between cell-surface carbo[*I
Dr. P. Arya, Dr. R. Ben
Steacie Institute for Molecular Sciences
National Research Council of Canada
100 Sussex Drive. Ottawa KlAOR6 (Canada)
Fax: Int. code +(613)952-0068
e-rnail: prabhat arya(u?nrc.ca and rben(u.ned1 .sims.nrc.ca
1280
Q VCH Verlagsgesellschaft mbH, 0-69451 Weinheim. 1997
hydrate ligands and various protein receptors, they form the
basis of recognition events that are fundamental to a vastly
diverse range of biological and pathological processes. For instance, interactions of this nature have been implicated in cellto-cell communication, bacterial and viral infections, chronic
inflammation, cancer/metastasis formation, and rheumatoid
arthritis.[j]
Oligosaccharides are very complex and diverse, which makes
their synthesis both labor-intensive and expensive. As a result,
the discovery of new biologically active oligosaccharide ligands
is a complicated problem. This aside, even when a promising
compound has been identified, optimization to enhance activity
is difficult and time-consuming. The synthesis of oligosaccharide libraries by a combinatorial approach offers a feasible solution to these problems.
In contrast to peptide and nucleotide libraries, preparation of
an oligosaccharide library is not a facile process. It is complicated by the issues of stereochemistry at the anomeric position and
the fact that multiple hydroxyl groups are present. Traditionally, these groups would be dealt with using a less than elegant
orthogonal protection-deprotection strategy. As an alternative, Hindsgaul et al[4a1demonstrated a random glycosylation
approach for forming small di- and trisaccharide libraries. This
strategy utilizes a glycosyl donor, which is protected with only
one type of protecting group, and a glycosyl acceptor in which
all hydroxyl groups are unprotected (Scheme 1). Hindsgaul
et al. coupled the benzylated glycosyl donor 1 to the disaccharide 2, which has six free hydroxyl groups. After three hours at
room temperature, a complex mixture was obtained in which
about 30% of acceptor 2 was fucosylated. Separation of the
mixture with reverse-phase chromatography furnished individual trisaccharides, which were analyzed by NMR spectroscopy.
Analysis confirmed that all six expected products were present
in yields of 8-23 YO.Ideally, a statistical mixture would contain
17 YOof each product.
In an alternate solution-phase approach, a latent - active glycosylation method was developed by Boons and co-workers
(Scheme2).14blThis strategy uses a glycosyl donor (4) and a
o570-o83319713612-128o$ 1 7 SO+ S O j O
Angew Chem. Int. Ed. Engl. 1997, 36. No. 12
Документ
Категория
Без категории
Просмотров
2
Размер файла
475 Кб
Теги
free, condensed, sily, germyl, questions, cation, phase
1/--страниц
Пожаловаться на содержимое документа