вход по аккаунту


Elusiveness of CuIII Complexation; Preference for Trifluoromethyl Oxidation in the Formation of [CuI(CF3)4] Salts.

код для вставкиСкачать
Elusiveness of Cu"' Complexation;
Preference for Trifluoromethyl Oxidation
in the Formation of [Cu'(CF,),] - Salts**
F3C- -U
[Cat][Cu(CF,),]Cat-3. Cat
Bu,N, Ph,N, (Ph,P),N
The anion 3 is nearly square planar in the crystal; the four
CF, groups are alternately located slightly above and below the
mean molecular plane. The substances have been characterized
as Cu"' salts with a d8 configuration at the metal. The only other
putative copper(iI1) compound with a perfluoroalkyl substituent
(4)['*] shares with 3 a distorted square-planar structure and a
pair of nearly linear CCuS bonds.
Salts of 3 have been prepared by oxidation of Cu'CF, . Classification of 3 as a Cu"' species presupposes that the metal rather
than the ligand has relinquished its electrons. Let us presume the
opposite, namely that oxidation has resulted in one CF: moiety in the product along with three CF; units and Cu'(d'O).
Then, to a first approximation, [Cu(CF,)J can be considered
a normal cuprate [CF,-Cu-CF,]- in which the copper is further
coordinated to a trifluoromethyl anion and cation. Although
the bonding in 3 is highly ionic, the covalenL components can be
illuminated by a qualitative M O correlation diagram of the ideal
square-planar complex (Fig. 1). For simplicity only the Cu-C
bonds are considered.
On the left side of Figure 1 four symmetry-adapted molecular
orbitals are derivcd from the sp3 carbon hybrid orbitals available for bonding from the four trifluoromethyl groups. The
bonding and two nonbonding MOs are filled; the fully antibonding orbital is empty. On the right the copper 3d,,-y, and 4s
orbitals are depicted as occupied and virtual, respectively. Inter-
p] Prof. Dr. J. P. Snyder
IRBM P. Angeletti
Via Pontina Km 30.600
1-00040 Pomezia. Rome, (Italy)
1,4-Addition of organocuprates to sc,p-enones is a cornerstone carbon-carbon bond-forming reaction in synthetic chemistry." 31 Although the mechanism of delivery of the Cu-bound
moiety is not fully understood, the currently ascendent concept
postulates the preliminary and reversible formation of a n-olefin
complex (1),[4-'01 transformation to a copper(m) P-adduct
(2), and collapse to product^.^'. '. ' I . 12] Skepticism about the
viability of Cu"' intermediates
has been expressed.[l31 In addition, according to recent a b initio calculations using an effec0' tive core potential at copper, monoalkylcopper reagents
(RCu) may add 1,4 to enones
Scheme 1
through a six-membered transition state incorporating the carbony1 oxygen a t ~ m . [ ' ~ - ' 'By
implication, copper(I1r) intermediates are bypassed. In this context the very recent oxidative synthesis and X-ray structure determination of stable, colorless, and presumably diamagnetic
salts of the [Cu(CF,),]- ion (3)[",
along with the previously
described dithiocarbamate 4[181are of singular importance.
[ F3C-
James P. Snyder*
Telefax: Int. code (6)91093-654
I am grateful to Dr. Steven Bertz (Lonza Inc.) for stimulating our interest in
this problem and to Di- Gernot Frrnkiiig (Marburg) for many helpful discussions on the use of ah mitio calculations for treating copper complexes.
6 a93
Fig. 1 . [dealired MO correlation diagram for the Cu C bonds of 3, derived from
the symmetry-adapted MOs for four C sp3 centers (left) and Cu AOs (right). S and
A designations refer to the symmetry reflected across, respectively. vertical and
horizontal planes orthogonal to the plane of the paper.
action of SS, orbitals of Cu(4s) and CF, results in the fully
symmetric MO for 3 delocalized over the metal and the four
trifluoromethyl groups. A second bond arises from combination
of SS, orbitals, the metal d+,, orbital serving as a lone-pair
donor to satisfy the electron deficiency caused by CF:. Nonbonding MOs contain the third and fourth pair of electrons. The
MO filling scheme implies that [Cu(CF,)J salts are associated
covalently by only four bonding electrons. Further refinement
of the diagram would include secondary contributions from the
copper d orbitals and the fluorine atoms.
In an attempt to verify this picture, calculations have been
performed on the experimentally determined geometry of 3 using a 6-31G*(6d) basis set on CF,, the Hay- Wadt effective core
potential (ECP-2)[l9I for the ten core electrons of copper, a
double-< basis set for the remaining metal electrons ((9s5p5d)/
[3s4p2d], split (441/211 1/41)),[20*211
and MP2 electron correlation.[221This prescription has proved to be effective for geometries and energies of both Ti and Cu' complexes.[20, 2 1 . 231The
resulting Hartree-Fock (SCF) and correlated (MP2) wavefunctions were subsequently subjected to a natural population
(NPA) and bond orbital (NBO) analysis.[" 301 To test the predicted structure in the present case, anion 3 has been optimized
at the SCF level without symmetry constraints. A symmetrical
structure with r(Cu-C) = 1.967 A, r(C-F) = 1.338 and
1.343 A, O(Cu-C-F) = 113.5' and O(C-Cu-C) = 170.0' results
in excellent agreement with the X-ray crystal structure (e.g.,
r(Cu-C) =1.949-1.983 A, B(C-Cu-C) = 165.0 and 172.6").rt71
At the experimentally determined structure of anion 3, calculated atomic charges are high (MP2 (SCF): Cu 0.71 ( + l . l ) ,
C + 0.74 ( + 0.81), F -0.39 ( - 0.44)) and Wiberg Cu-C bond
orders moderate (0.29 (0.38)). A considerable degree of ionic
character is further reflected by the diminutive 19 % contribution of copper to the localized CuC bonding orbitals at the S C F
level. It falls away altogether when MP2 correlation is included.
According to NPA analysis the copper d orbitals are populated
as follows: dxJ,,d,,, d,,, and d,, 1.96-1.98, d,,-,, 1.77 (1.38)
electrons. I n addition 10.3 (9.88) electrons reside in copper's
outer shells: 4s (0.53 (0.49)). 3d (9.7 (9.4)). 4p (0.04 (0.02)). 4d
(0.06 (0.02)).A straightforward interpretation of the data is that
copper functions its a d-base through the 3d,,
orbital, retrieving lost electronic charge by means of backbonding to the 4s, 4p,
and 3d orbitals. The analysis places 3.3 (3.1) and 7.4 (7.5) electrons a t carbon and fluorine atoms, respectively. Carbon is thus
depleted by + 0.7 ( 0.9) units of charge, while fluorine gains
-0.4 ( - 0 . 5 ) . The CF, groups carry 25.4 (25.5) electrons. preciselq the average of three CF, groups and one CFT group.
Charge neutralization in anion 3 by an ammonium cation to
give ;I contact ion-pair causes no change in the electron distributi on.
Attemprs were made to force the SCF wavefunction to converge o n the Cu"' state by switching filled copper d orbitals with
the L U M O . the high energy counterpart of 3's bonding SS2MO
in Figure I . The numerical experiment led to a state with 1.992.00 electrons in all d orbitals except d.,v. which had 1.1 electrons. The total energy of this state is 250 kcalmol-' above the
full) relaxed d.,y orbital with 1.97 electrons. This is consistent
with the SC'F result that the largely localized lone pair Cu d
orbitals i n 3 are 193-218 kcalmol-I lower in energy than the
HOMO. I n the nwked Cu' monocation the same orbitals are
degenerate and lie at the frontier orbital gap. An empirical argument that might be made against the interpretation of 3 as a CU'
complex is the appearance of a quasi square-planar structure in
the solid state. the common geometry for d8 metal complexes.
Though tetracoordinate d ' " Cu' -1igand association ordinarily
produces tetrahedral structures. square-planar approximations
are not unknown.[-" - 3 3 1
The pictiire that emerges for 3 is one in which the bonding
between ( ' t i t and the average of (CF;), and CF: (i.e.
(CF,".'),) i b principally ionic. The covalent component derived
by the full cdculation is in complete accord with Figure 1 , Two
delocalized tivc-center. two-electron bonds are accompanied by
two nonbonding electron pairs. The description strongly implies
that oxithtion of CuCF, has effectively removed electrons from
a CF, Iigmd. and the charge depletion is then shared equally by
the four trilluoromethyl substituents in product 3. Furthermore.
the electron depletion at CF, depicted in Figure 1 suggests CF,CF, t o be ; i n important thermal decomposition product, as has
been observcd (D. Naumann and T. Roy. private communication). Similar calculations show that analogous concepts apply
t o 4.1'XI
The apparent willingness of a powerful electronegative moiety such i i j CF.; to undergo oxidation in preference to Cu'
underscore> both the importance of the role of ligand in assessing oxidation pathways in metal complexes and the resistance of
copper to ii5suine the Cu"' state. The principles outlined herein
can be expected to apply to the mechanism of cuprate addition
;is well a s t o contribute to a deeper understanding of other
metal5 assigned high oxidation states in complexes.
Received: May 13, 1994
ReLised: August 4. 1994 [Z6933IE]
German yci-sion: A n y ~ i i .Cliciii. 1995. 107. I I2
Keywords: a b initio calculations * copper compounds * cuprates
electronic structure
[ I ] D E R~n\itcr,N . M Swinglc. C ' h i i . R c d i . 1992. 412. 771
121 13. 11 I , i p h u t / . S.Scngupta. 01-2.Rcrici. 1992. 41. 135.
[3] Y i i C ' l ' ' S. II . Bertr. E. H. Fuirchild iii E i i c d ~ i / ~ d ; ii/
i i K q q v i r s /or Orpnic
.Ti i i / I n , \ i \ I I d . L Paquette). Wiley. Chichester. in press.
[4] C i tiallneinn. T Olswn. C Ullcnius. .I ~ r g O l 7 0 i l l P / C. ' / I ~ 1985.
282. 133- 144.
45. 523- 534.
IS] H. ('liri\lciiwii, T. O l s o n . C Ullenius. P i , - ~ i l i e r / r ~ 1989.
161 S. H. Bertr. R. A J. Smith. J A m . C / i i , i i i . Soc 1989. i l l . 8276 X277.
[7] E. J. Core).. N. W. Boar. f i ~ r r u h r i / r n nLcir 1984. 25. ( 2 9 ) . 3063- 3066.
[ X I E. J. Corey, N . W Boar. P.rruhr~/r~.on
LiJf/.1985. 26. (4Y), 6035 6018.
[9] S. Sharma, A. C. Oehlschlager. li,rruh(vhviii 1991. 47. I177 - 1184.
[lo] N. Krnuse. R. Wagner. A. Ger-old. .I A n / . C'hcm. Sot. 1994. //6,381 -3X2.
[ l l ] S. R. Krauss. S . G Smith, J Am. Chrnt S(J<,.
1981. iOi. 141 148.
[12] C . R. Johnson, '2. A. Dutra. J. A m . C/iiwi. S i c . 1973. Y5. 7783.
[I31 J. P. Collman. L. S. Hegedus. Pr;iiciph mid ilpp/iuii/oii.\ 0 / Orjiirno/riiii.~it,on
M c w l C/im~i.srri~.
Universitv Science Books. Mill Vallcy. CA. 1980.
[I41 A. E. Dorigo. K. Morokuma, J An/. U i c w . Sol.. 1989. / / I . 4635 4643.
[I51 A . E. Dorigo. K. Morokumd, J ,4117. C/iciii. S o . 1989. I / / , 6524-6536
[I61 H . 0. House. W L. Respess. G. M. Whitesides, .I Or,? C / i e n i . 1966, 31. 3128.
1171 D. Naumann. T. Roy. K.-F. Tehhe. W. Crump. A i i , q r i ~ . C / i m 1993. 105. 15551556: Aiijieii Chcwi. 0 1 1 GI. €ng/. 1993. 32. 14X2-14X3 Memwhile. a superconducting salt uith [Cu(CF,)J as anion has been reportcd- J A. Schleuter.
U . Geiser. J. M. Williams. H. H. Wang. W.-K Kwoi. J A Fendrich. K. D.
C;irlson. C . A Achcnbach, J. D. Dudek. D. Nauinann. r. Roy. J. E. Scliirber.
W. R. Bayless. J ( ' / i e i i i . Soc. C/r~wi,C ' o i n n i i i i i . 1994. 15UY 1600.
[I81 M. A. Willert Porada. D. J. Burton. N . C . Baenngei-. J Chiwi. Sol.., C ' h r w .
Cniiiiiiiiii. 1989. 1633.
[19] P. J. Hay, W. R. Wad[. J C ' / i u ~ Phi.,\.
1985. X2. 299.
[20] V. Jonas. G. Frenking, M. T Reetr. J. C'omnpiir. Chcin 1992. 13, (X), 919 934.
[?I] M. Bohme. G. Frenking, M. T. Reetz. Or~unonirio//mr5. submitted.
(221 M . J. Frisch, G. W. Trucks. M . Heiid-Gordon. P. M W. Gill. M. W. Wong.
J. B. Foresman. B G. Johnson. H. B. Schlegel. M. A. Robb. E S. Replogle. R.
Goniperts. J. 1.Andrea. K. Raghavachari. J. S. Binklcq. C'. Gonzalez. R. L.
Martin. D. J. Fox. D. J. Defrees. J. Baker. J. J. P. Stewai-t. .I.A. Pople. Guiissirrri
92. Revision A ed . Gaussian. Inc., Pittshurg. PA. 1992.
1231 The ECP-XMP2 combination used here provides result\ comparable with allelectron calculations. For C u - F we find r(Cu- F ) = 1.747 8, and )[ = 5.21 D.
A recent near-Hartree-Pock limit plus MP2 ciilculation gave 1.786 and
6.12 D [24] (vs. expenmentiil values of 1.745 A and 5.77 D [25]). For Cu -CH,
wc obtain r(Cu C) = 1.913 A and !I = 2.29 D. An all-elcctron triple-< optimization with correlation yielded 1.936 A and 2.41 D. respectively [26].
[24] P. Schwerdtfeger. P. D. W. Boyd. G. A Bowmaker. I.. P Aldridge. Sfriicr.
C/icw. 1990, 1. 405 -415.
[25] K . P. Huber. G. Herzherg, Mo/i,cu/ur Spei,/r[iund Moliv i i / w S r r i r m i r c . Con\ / i i i i / s of Diotoiiiii .Mo/i~i~~i/i~.c.
Van Nostrand. New Yori. 1979.
(261 C. W. Bauschlicher Jr.. S. R. Langhoff, H. Partidge. I.. A. Barnes. J. < h n .
P$T. 1989. 91. 399-2411.
[27] A. E. Reed. R. B. Weinstock. F. Weinhold. J C/iiwi. Phi..\. 1985. 83. 735.
[28] A. E. Reed. F. Weinhold, J Clieiii. Pli?.~.1985. 83. 1736.
129) A. E. Reed. F. Weinhold, J. Anr. Chcwi. Soc. 1986. 108. 3586 - 3593 and references therein.
[30] A. E. Reed. L. A. Curtis. F. Weinhold. C/iiwi. R P ~1988.
XX. 899.
[31] B. Chiari. U. Piwesana, T. Tarantelli. P. t' Zanazi.i. I n r q C'/wni 1984, 23.
342 2547.
[32] J. C . Dyason. P. C Healy. L. M. Engelhardt. C. Pakauatchai, V. A. Patrick. A.
H. White. J. C7ienr. SIK. Dulroii Trans. 1985. 4. 839 844.
[33] M. Hakanason. S. Jagner. M. Nilsson. J. Orq~noiiii~r.
C / i c n i . 1987. 336. 279285.
1,3,5,7-Tetraphosphabarrelene: A New Cyclic
Phosphaalkyne Tetramer**
Paul Binger,* Gerald Glaser, Barbara Gabor, a n d
Richard Mynott
The chemistry of phosphaalkynes (RC=P) has gained considerable momentum through the use of transition-metal templates
in cyclooIigomerization~.['~
The cyclodimerization of ~ t butylphosphaacetylene 5 to give the cobalt[21and rhodium[31
q4-l ,3-diphosphacyclobutadiene complexes 1. the unexpected
formation of the tricyclic zirconium and hafnium complexes 2
[*] Prof. Dr. P. Binger, Dr. G. Glaser, B. Gabor. Dr R. Mynott
Max-Planck Institiit fur Kohlenforschung
Kaiser-Wilhelm-Plat/ I . D-45470 Miilheim iiii der Ruhr (Germany)
Telefax Int. code (208)306-2980
This work was supported by the Volkswagen-Stiftiin:! and the Deutsche
t'orschungsgemeinschaft (Graduiertenkolleg. phosphorus a s 11 connecting link
between different chemical disciplines)
Без категории
Размер файла
270 Кб
salt, preference, oxidation, complexation, cui, formation, cuii, trifluoromethyl, cf3, elusiveness
Пожаловаться на содержимое документа