close

Вход

Забыли?

вход по аккаунту

?

Four-Center Two-Electron Bonding in a Tetrahedral Topology.

код для вставкиСкачать
COMMUNICATIONS
Attractive ideas based on qualitative molecular orbital
considerations often have not survived or have given disappointing results when subjected to experimental test.
Homoaromaticity in neutral and carbanionic systems is a
case in point.['] Hence, such ideas are best tested calculationally at an adequate level of quantitative theory before
investing the much greater effort needed for an experimental investigation.
When this project was begun a decade
MIND0/3l4I was employed to evaluate the energies of the
isodesmic reactions (a) and (b). The former predicts the
Four-Center Two-Electron Bonding in a
Tetrahedral Topology. Experimental Realization of
Three-Dimensional Homoaromaticity in the
1,3-Dehydro-5,7-adamantanediylDication**
By Matthias Bremer. Paul von Rague Schleyer.*
Karl Schotz, Michael Kausch, and Michael Schindler
Dedicated to Professor George A . Olah
on the occasion of his 60th birthday
Four orbitals placed at the bridgeheads of adamantane
extend towards the center of the cage. Of the four possible
combinations of phases, only the nondegenerate one, I, is
I
1'
1
bonding. The occupation of this molecular orbital by two
electrons (Fig. 1)-e.g., as in the dication l/l'-should result in considerable stability, provided the overlap is sufficient.
Fig. I . Jorgensen plot of the highest occupied molecular orbital of I . (The
STO-3G basis set employed underestimates the overlap of the bridgehead
orbitals. The Jorgensen program cannot deal with higher basis sets.)
[*] Prof. Dr. P. von R. Schleyer, Dip1.-Chem. M. Ererner, Dr. K. Schotz,
Dr. M. Kausch
Institut fur Organische Chemie der Universitat Erlangen-Niimberg
Henkestrasse 42, D-8520 Erlangen (FRG)
Dr. M. Schindler
Lehrstuhl fur Theoretische Chemie der Universitat
Postfach 102148, D-4630 Eochum 1 (FRG)
[**I
We thank the Deutsche Forschungsgemeinschaft, the Fonds der Chemischen Industrie, ARDEC, Dover, NJ, and the CONVEX Corporation
(the most recent calculations were carried out on a CONVEX C1 computer) for support, Dr. Alan Reed (Erlangen) for discussions, and Dr. W.
Bauer (Erlangen) for the NMR spectra.
Angew Chem In1 Ed Engl 26 11987) No 8
3
L
2
trishomoaromatic 1,3-dehydro-5-adamantyl cation 2, an
intermediate which has been implicated in solvolysis reactions,['] to be stabilized by about 25 kcal/mol relative to
the 1-adamantyl cation 3 and 1,3-dehydroadamantane 4.
Equation (b) indicates that the extension of the two-electron delocalization from three centers (as in 2) to all four
bridgeheads nearly doubles the stabilization. While 1 is
tetratrishornoaromatic, this large resonance energy, along
with the delocalized electronic structure, qualifies it for the
designation "aromatic," as well. While Olah et al. have not
succeeded in preparing the 1,3-adamantanediyl dication
5,l6] nor, indeed, any other 1,3-carbodication in solution,131
the high degree of stabilization indicated for 1 by Equation (b) encouraged us to seek experimental verification.
The problem of obtaining a suitable precursor for dication 1 was complicated by the known instability of 1,3-dehydroadamantane derivatives, e.g., towards reaction with
oxygen and polymerization.[s1 After a large number of
preliminary experiments,['] a simple route was found.
When reacted with mercury(ir) fluoride, 1,3,5,7-tetrai ~ d o a d a m a n t a n e [gives
~ ] a trisubstituted product, 1,3,5-trifluoro-7-iodoadamantane 6, almost exclusively. This obviates the need for tedious chromatographic separations of
polyhaloadamantane mixtures. With butyllithium in ether/
pentane solution at -8O"C, 6 gave (presumably via I-Li
exchange and LiF elimination) 1,3-dehydro-5,7-difluoroadamantane 7 directly. As 7 only is stable at room temperature in solution (the solid polymerizes with near explosive
exothermicity above O T ) , workup had to be carried out
below - 30°C. After crystallization from pentane at
-7O"C, 7 shows I3C- and 'H-NMR spectra typical of dehydroadamantanes.['. 17] Chemical characterization was
achieved by reaction of 7 with iodine in hexane; 1,3difluoro-5,7-diiodoadamantane8 is formed in over 80%
yield.['71
0 YCH Verlagsgeselischafi mbH. 0-6940 Weinberm. 1987
0570-0833/87/0608-0761$ 02 SO/O
76 1
'WF
7
6
SbF5
S02CIF
1
8
T < -80°C
Kept cold, solid 7 was added to a solution of SbF, in
S 0 2 C l F(ca. 5 M) at temperatures below - 80°C in order to
prepare the NMR sample. The resuIting dication 1 appears to be rather stable in superacid media, as spectra
could be taken successfully at temperatures u p to 0°C. The
N M R parameters of 1 are remarkable, and constitute a
proof of the nonclassical structure. The I3C-NMR chemical shift of the bridgehead carbons, which share the positive charge at least in a formal sense, comes upfield (6=6.6
at - 7 I "c,with internal [D,]acetone/TMS capillary) of the
methylene carbon resonances (6= 35.6), despite the presence of the two positive charges. Shielded values are characteristic of hypercoordinate carbocation centers. A good
analogy is provided by the trishomocyclopropenyl cation
9, which exhibits similar N M R parameters: 6 ~ 4 . 9(CH)
and 17.6 (CH2).I8l
&
These chemical shifts can be analyzed in several ways. If
1 were a classical, rapidly equilibrating dication (an unlikely possibility!), the bridgehead carbon chemical shift
should be an average of the values found for the carbocation center in the I-adamantyl cation 3 (6=300)161and a
cyclopropane bridgehead in 1,3-dehydroadamantane 4
(6=37.3).'5' The chemical shift (6=169), predicted for the
bridgehead carbons by the classical model on this basis,
deviates from the experimental value by over 160 ppm! Alternatively, the sums of all the I3C chemical shifts for l
(240 ppm) and for 1,3-dehydroadarnantane 4 (454 ppm)
may be used as a classical-nonclassical classification criterion."" For typical classical carbodications, the dicationneutral difference is on the order of 800 ppm (i.e., about
twice that found on the same basis for monocarbocations).
For 1 , the difference is negative, -214 ppm! Such negative
values (but not as large) are found, e.g., for the related
trishomocyclopropenyl cation 9 (-48 ppm),["] and for
the 7-norbornadienyl cation ( - 96 ppm).l'O1
Clearly, the I3C-NMR data for 1 cannot be reconciled
with a classical, rapidly equilibrating model. But how well
d o the experimental shifts compare with those predicted
by a nonclassical model? The IGLO (individual gauge for
localized molecular orbitals) method has been demonstrated to give good to excellent agreement between calculated and experimental chemical shifts for both classical
and nonclassical carbocations.l"l For example, the IGLO
(DZ basis set) and experimental (in parentheses) I3C-NMR
chemical shifts for the I-adamantyl cation 3 are: 6=337.7
(300.0), 57.7 (66.6), 67.8 (87.6), and 37.2 (34.6). When applied to 1 (without prior knowledge of the experimental
results), bridgehead (6=7) and methylene carbon (6=33)
chemical shifts were calculated (DZ basis on the 3-21G
geometry). The remarkable agreement with the experimental values (6=6.6 and 35.6) speaks for itself. The calculated proton chemical shift, 6=3.0, is lower than experiment (6=3.8), but the latter, measured from an internal
standard, depends on the magnetic anisotropy of the medium and is somewhat uncertain.
As indicated by population analysis at various theoretical levels (MNDO, MIND0/3, STO-3G, 3-21G), the
bonding and charge distribution in 1 are revealing. The
positive charge resides not only on the bridgehead carbons
but also on the 12 hydrogens; the methylene carbons are
neutral or even negatively charged. The overlap among the
bridgehead carbons at 3-21G//3-21G (0.38 vs. about 0.7
for a normal C-C single bond),"61 is significant, consistent
with the dotted-line representation. Each bridgehead carbon is involved with one electron, and each has a total
bond order of 0.5. Charge delocalization in a nearly spherical topology, as well as partial bridging among the bridgehead positions, is responsible for the stability of 1.
The dehydroadamantanediyl dication 1 now takes its
place along with pyramidal
e.g., those based
on (CH)? and (CH);@, as purely "organic" examples of
three-dimensional aromaticity"51which exhibit large resonance energies. Hiickelk 4n + 2 rule, involving interstitial instead of x electrons, can easily be extended to such systems.['''
9
The triplet splitting of the CH2 group in 1, J(C.H)=
165
Hz, in line with the trend found for adamantane (125 Hz)19'
and the 1-adamantyl cation 3 (148 Hz),[~'confirms not
only the formation of dication 1 but also the strong deformation of the C-CH2-C angles predicted by the calculations. Figure 2 gives the 3-21G ab initio geometry. The
J(C.H)
value, 168.4 Hz, for 9 is similar.'81 The single, sharp
' H chemical shift of 1 , 6=3.8, also is unusual in comparison with 6=4.2[,] for the a-CH, groups of the 1-adamantyl
cation, and 6 = 1.74191for the adamantane methylenes.
+
Received: February 27, 1987 [Z 2116 1E]
Publication delayed at the authors' request
German version: Angew. Chem. 99 (1987) 795
Fig. 2. The 3-21G-optimized structure 0 1 the dehydroadamantanediyl dication 1 !distances in A). The 6-31G*-optmized data are similar: C - C = 1.504,
2.052A,C-C-C=86.0, 118.2". H-C-H=114.2".
762
0 YCH Verlagsgesellscha~mbH. 0.6940 Weinheim. 1987
[ I ] For recent references, see P. van R. Schleyer, E. Kaufmann, A. J. Kos,
H. Mayr, J. Chandrasekhar, J . Chem. SOC.Chem. Commun. 1986. 1583;
A. McEwen, P. van R. Schleyer, J . Org. Chem. 51 (1986) 4357. A recent
claim 10 have demonstrated 4.5 kcal/mol of "homoaromaticity" in triquinacene (J. F. Liebman, L. A. Paquelte, J. R. Peterson, D. W. Rogers,
0044-8249/87/0808-0762 $ 02.50/0
Angew. Chem. Int. Ed. Engl. 26 (1987) No. 8
J. Am. Chem. Soc. 108 (1986) 8267). is not “unequivocal” as no attempt
was made to dissect electronic from strain effects. “Homoconjugation”
may be a bet!er term to describe stabilizdtlons of such small magnitude,
if present.
[2] We thank Michael Waeber. Herbert Gareis. and Roland Hacker, Erlangen, for preliminary experimental investigations, and E. M. Engler for
early STO-3G calculations. Further work is described in the Diplomarbeit of M. B.. Universitat Erlangen-Nurnberg 1986.
[3] See a review on stable carbodications: G. K. Surya Prakash, T. N. Rowdah, G. A Olah, Angew. Chem. 95 (1983) 356; Angew. Chem. In!. E d .
C i y l . 22 (1983) 390. Also see: G. K. Surya Prakash, M. Formia, S.
Keyonian, G. A. Olah, H. J. Kuhn, K. Schaffner, J . Am. Chem. SOC.109
(1987)911
141 a ) R. C. Bingham, M. J. S . Dewar, D. H. Lo, J . Am. Chem. Soc. 97(1975)
1285; b) for reviews, see T. Clark: A Handbook ofComputationa1 Chemi.7rr-i.. Wiley, New York 1986; c) D. F. V. Lewis, Chem. Revs. 86 (1986)
IIII
IS] a) R. E. Pincock, J. Schmidt, W. B. Scott, E. J. Torupka, Can. J . Chem.
50 (1972) 3958; b) C. S . Gibbons, J . Trotter, ibid. 51 (1973) 87; c) R. E.
Pincock, F.-N. Fung, Tetrahedron Left. 21 (1980) 19; d) W. B. Scott, R.
E. Pincock, J Am. Chem. Soc. 95 (1973) 2040; e) R. E. Pincock, E. J.
Torupka, ihid. 91 (1969) 4593.
161 G. A. Olah. G. K. Surya Prakash, J. G. Shih, V. V. Krishnamurthy, G. D.
Mateescu, G. Liang, G. Sipos, V. Buss, T. M. Gund, P. von R. Schleyer.
J Am. Chem SOC.107 (1985) 2764. The pagodane dication is of interest
in this context: G. K. S . Prakash, V. V. Krishnamurthy, R. Herges, R.
Bau, H. Yuan, G. A. Olah, W.-D. Fessner, H. Prinzbach, J . Am. Chem
Soc. 108 (1986) 836.
171 G. P. Sollott, E. E. Gilbert, 1.Org. Chem. 45 (1980) 5405, and reference\
cited therein. A modified procedure using CSI and AII’, prepared in
situ. was employed here.
[S] a) S. Masamune, M. Sahai, A. V. K. Jones, T. Nakashima, Can. J. Chem.
52 (1974) 8 5 5 ; b) G. A. Olah, G. K. Surya Prakash, D. Whittaker, J. C.
Rees, J. Am. Chem. SOC. 101 (1979) 3935; c) G. K. Surya Prakash, M.
Arvanaghi, G. A. Olah, ibid. 107 (1985) 6017; d ) also compare the pentacyclononyl cation reported by R. M. Coates, E . R. Fretz, J . Am. Chem.
Soc. 97 (1975) 2538.
191 R. C. Fort, P. von R. Schleyer, J . Org. Chem. 30 (1965) 789.
[lo] P. von R. Schleyer, D. Lenoir, P. Mison, G. Liang, G. K. Surya Prakash,
G. A. Olah, J. Am. Chem. SOC.102 (1980) 683.
[ I I] M. Schindler, J. Am. Chem. SOC.109 (1987) 1020.
1121 Reviews on pyramidal carbocations: H. Schwartz, Angew. Chem. 93
(1981) 1046; Angew. Chem. I n t . Ed. Engl. 20(1981)991; P. Vogel: Carbocatlon Chemistry. Elsevier, Amsterdam 1985.
1131 S . Masamune, M. Sahai, H. Ona, A. J. Jones, J. Am. Chem. Soc. 94
(1972) 8956; A. V. Kemp-Jones, N. Nakamura, S . Masamune, J . Chem.
Soc. Chem. Commun. 1974. 109, and references cited therein.
[I41 H. Hogeveen, P. W. Kwant, Acc. Chem. Res. 8 (1975) 413.
[I51 E. D. Jemmis. P. von R. Schleyer, J . Am. Chem. Soc. I04 (1982) 4781,
and earlier papers in the same series. E. D. Jemmis, ibid. 104 (1982)
7017: J. Chandrasekhar, E. D. Jemmis, P. von R. Schleyer, in preparation.
1161 Natural population analysis (A. E. Reed, R. B. Weinstock, F. Weinhold,
J . Chem. Phys. 83 (1985) 735) based on natural localized molecular
orbitals (A. E. Reed, F. Weinhold, ibid. 83 (1985) 1736) and the
ST0-3G//STOO-3Gand the 3-21G//3-21C wave functions (W. J. Hehre,
L. Radom, P. von R. Schleyer, J. A. Pople: Ab Initio Molecular Orbital
77wory. Wiley, New York 1986) were employed.
1171 Spectroscopic data of 6 , 7,and 8. (The multiplicities of the NMR signals are taken from ‘H-decoupled ”C-NMR spectra; ‘Jle.,,l coupling
constants from “gated” spectra): 6 , m.p. =72-73”C; ‘H-NMR (400
MHz, CDC13): 6=2.10, 2.19 (2mc, 6 H , CF-CHZ-CF), 2.50 ( s , 6 H ,
CF-CHI-CI); “C-NMR (100 MHz, CDCI,): 6=25.5 (4, ‘ J l c F ) = 14
Hz. CI), 45.7 (t, ’ J l c e,=20 Hz, CF-CHZ-CF, ’ J l c . ~ ) 134
= Hz), 53.6 (d,
‘J(<> , = 2 0 Hz, CF-CHI-CI,
‘ J , c H ) = 136 Hz), 90.2 (dt, ‘J,C,F,= 195 Hz,
‘Jl<,,= 15 Hz. CF); IR (KBr): 3=2920, 2810, 1440, 1430, 1310, 1295,
1230, 1205, 1130, 1025. 1005,970,940, 885,860,780, 700cm-’.-7, ‘HNMR (400 MHz, [D,ltetrahydrofuran): 6 = 1.36 (d, 4 H , ’ J I H
10 Hz,
CF-CHI-C),
1.64 (d, 2 H , ‘ J , e ~ ) = 8 Hz, C-CHZ-C), 2.01 (d, 4 H .
is)=
10 HZ, CF-CHI-C), 2.22 (t. 2 H , ‘ J 1 t ~ 1 = Hz,
4 CF-CHI-CF);
“C-NMR (100 MHz, [D&etrahydrofuran): 6= 18.9 (t, ‘J,cFl=9 Hz.
CH:-C-CH?), 37.2 (t, ‘Jlc 6 1-4 Hz, C-CHz-C, ‘Jlc..HI= 159 Hz), 45.3
(dd, ‘J(c
15 Hz, “ J , c , , = 3 Hz, CF-CHZ-C, ‘ J , c H l = 136 Hz), 46.0(t,
‘J,<g , = I8 Hz, CF-CHz-CF, ‘ J , C , ~
132
~=
Hz), 98.4 (dd, ‘J1c,,,=226
Hz, ‘Jtc, , = 5 Hz, CF).-S, m.p.=132-134”C; ‘ H - N M R (400 MHz,
CDCli): 6=2.18 (t. ZH, ‘ J l b * , , = 5 Hz, CF-CHZ-CF), 2.52 (d, 4 H ,
‘JCtC
to= I I
Hz, CF-Ctf-CI),
2.62 (d. 4 H , ‘J(H.H)= 11 Hz,
“C-NMR (100 MHz,
CF-CHZ-CI), 2.91 (s, 2 H , CI-CH?-CI);
CDCI,): 6=28.8 (t. ‘J<c.11=12 Hz, CI), 45.3 (t, ‘Jlc.+)=20 Hz,
136 Hz), 53.1 (dd, ‘ J l c . F I = 18 Hz, ‘JiC-F,=3 Hz,
CF-CHI-CF, ’J,<..,~I=
CF-CH?-CI, ‘ J , < . H I =135 Hz), 60.2 ( s , CI-CHI-CI, ‘ J ( c H 1 = 136 Hz),
89.7 (dd, ‘ J t c t i = 198 Hz, ‘Ji<.,,=
14 Hz, CF); IR (KBr): C=2910, 2820,
1430, 1310. 1290, 1250, 1215, 1205, 1095, 1015, 995, 960, 945, 930, 870,
815, 725, 675 cm ~ I .
Angen, Chem. I n ! . Ed. Engl. 26 11987) No. 8
Dimerization of 1,6-Dithiacyclodeca-3,8-diyne
in the Presence of Cobalt Complexes. A Simple
Synthesis of a [2.2](2,5)Thiophenophane Derivative**
By Rolf Gleiter,* Michael Karcher. Bernhard Nuber. and
Manfred L. Ziegler
Dedicated to Professor Hermann Schildknecht
on the occasion of his 65th birthday
Recently, we showed that cyclodeca- 1,6-diyne 1 undergoes dimerization in the presence of [CpCo(CO),] o r
[CpCo(C8HI2)1(Cp =q5-C5H5)to give a superphane of cyclobutadiene, 3, doubly capped by CpCo units.“’
CP
co
CO
CP
3
2 (x=s)
4
To test the scope and limitations of this new reaction, we
have investigated the reaction of 1,6-dithiacyclodeca-3,8diyne 2IZ1with [CpCo(CO),] and [CpCo(C8H ,?)I. Heating 2
(1.0 g, 5.95 mmol) with [ C ~ C O ( C O )(1.07
~ ] g, 5.95 mmol) in
n-octane yielded a colorless product (120 mg) which contained no CpCo unit. The analytical data of the product
(see Table 1) showed a molecular weight of 336, and the
IR, UV,“’ and NMR data indicated the presence of thiophene units and two nonequivalent C H 2 bridges. When 2
was heated with less than the equimoiar amount (10%) of
the Co complexes, we obtained the same product. Three
structural possibilities, the anti and syn isomers 4 and 5 ,
respectively, and the cage compound 6, are compatible
with the spectroscopic data.
5
6
7
From the observation that the most intense peak in the
mass spectrum of the product is at m / z 168 ( M 0 / 2 ) , we
favored 4 and 5 over 6.
[*] Prof. Dr. R. Gleiter, Dipl.-Chem. M. Karcher
[**I
Institut fur Organische Chemie der Universitat
Im Neuenheimer Feld 270, D-6900 Heidelberg (FRG)
Prof. Dr. M. L. Ziegler, Dr. B. Nuber
lnstitut fur Anorganische Chemie der Universitat
I m Neuenheimer Feld 234, D-6900 Heidelberg (FRG)
This work was supported by the Deutsche Forschungsgemeinschaft, the
Fonds der Chemischen Industrie, BASF AG, and Degussa AG.
0 V C H Verlagsgesellschaji m b H . 0-6940 Wernheim, 1987
0044-8249/87/0808-0763 $ 02.50/0
763
Документ
Категория
Без категории
Просмотров
1
Размер файла
426 Кб
Теги
bonding, two, four, topology, electro, center, tetrahedral
1/--страниц
Пожаловаться на содержимое документа