close

Вход

Забыли?

вход по аккаунту

?

Complexes Containing Phosphorus and Arsenic as Terminal Ligands.

код для вставкиСкачать
COMMUNICATIONS
Complexes Containing Phosphorus and Arsenic
as Terminal Ligands**
Manfred Scheer,* Jan Miiller, and Marco Haser
Alkali metal bis(trimethylsilyl)phosphanides react with different halogenophosphorus compounds to form multiple P-P
bonds.['] The alkali metal phosphanide serves as a reagent for
metalation of silylphosphorus functionalities with the elimination of P(SiMe,), and as a source for the central phosphorus
atom in the resulting tri- and isotetraphosphanes. Our
interests have focused on the extension of the above concept by
the use of transition metal halides and the heavier homologues
[E(SiMe,),]-(E = As, Sb, Bi) .Iz1
The paramagnetic W'" halogenide [ (N,N)WC1][31 1 b [(N,N) = N(CH,CH,NSiMe,),]
serves as an ideal starting material in both respects. With this
species and the corresponding molybdenum compound 1 a,14]
Schrock et al. were able to synthesize the first examples of compounds containing a metal-phosphorus triple bond (2) [Eq. (a)]
that were structurally characterized.['] At the same time Cummins et al. reported the synthesis of a complex containing
a Mo-P triple bond by using [Mo(NAr'R),] (Ar' = 3,5C,H,Me,, R = C(CD,),CH,) and white phosphorus as starting
CI
dyne complexes, react under cycloaddition with the phosphaalkyne to form four-membered ring derivatives. A subsequent 1,3-OtBu shift yielded the corresponding phosphaalkoxysubstituted compounds 4 and 5. Due to decomposition of the
alkoxides on the column material, chromatographic separation
of the reaction mixture has not been possible, nor have we been
able to obtain single crystals of 3 suitable for structural analysis.
It has been possible, however, to enrich solutions of 3 by fractional crystallization.
In an effort to find an explanation for the different 183W,3'P
coupling constants, we decided to synthesize 2 b, starting with
lithium bis(trimethylsily1)phosphanide and subsequent complexation with [W(CO),(thf)]. Furthermore, it should be possible to use bis(trimethylsi1yl)pnictides [E(SiMe,),] - (E = As,
Sb, Bi) to synthesize complexes containing the heavier homologues of phosphorus.is1
The reaction of 1b with [P(SiMe,),]- at 80 "C for two days
afforded the phosphido complex 2b in 65% yield [Eq. (c)]. To
P
prepare the corresponding arsenido complex 6 temperatures of
110 "C are needed. Whereas Schrock et al. observed and isolated
la: M = Mo
2a: M = Mo
the phosphanido complex [(N,N)W-PPh(H)] as an intermedilb:M=W
2b:M=W
ate in reaction (a), no corresponding Me,Si-substituted phosphanido
derivative could be detected by monitoring reaction (c)
The 183W,31Pcoupling constant of 2 b (138 Hz) differs signifby
31P
NMR
~ p e c t r o s c o p yInstead,
. ~ ~ ~ with growing amounts of
icantly from the coupling constants determined for the stabi2b an increase in the amount of P(SiMe,), is detected. Evidentlized phosphido complexes 3a,b (3a: 536Hz, 3 b : 554Hz),
ly, the rate-determining step of reaction (c) is the formation of
which we synthesized according to the methathesis reaction
the
phosphanido complex [(N,N)W-P(SiMe,),j, which is subgiven in Equation (b).'71Limited amounts of 3, as well as alkylisequently metalated by a second equivalent of LiP(SiMe,), with
the elimination of P(SiMe,), , forming the phosphido complex
[(/BuO)~WGCtBu]
[Wz(OtBu)s]+ tBuC s P/ [M(COk(thf)]
+
(b) 2 b, presumably, by the metal-catalyzed elimination of LiSiMe, .
These conclusions are in accordance with our results on the
[(tBuO),W -P -M(CO)J
reaction pathway of multiple P-P bond formation[" as well as
3a: M = Cr
with findings by Schrock et al., who emphasize the need for a
3b.M=W
base (LiPPhH or LiR) in the transformation of phosphanido to
phosphido complexes 2 by the elimination of LiPh.'"]
The phosphido complex 2 b first reacts with [M(CO),(thf)]
(M = Cr, W) to form complexes 7, although a second substitution in the trans-position is preferred, leading to the M(CO),
complexes 8, which can be isolated in 68 to 74% yield
[Eq. (d)].["] Complexes 6 and 8 are yellow and red crystalline
materials, respectively, which are moderately soluble in n-hexane, although well soluble in toluene and THF. These compounds were completely characterized by spectroscopic meth['I Prof. Dr. M. Scheer, DipLChem. J. Muller
Institut fur Anorganische Chemie der Universitat
ods.rlzlThe mass spectrum of 6 reveals the molecular peak,
D-76128 Karlsruhe (Germany)
whereas for 8 the peak of the [(N,N)W=P + MIi fragment is
Fax: Int. code +(721)661921
the highest one observed. In the range of CO vibrations, IR
e-mail: mascheer(a'achibm6.chemie.uni-karisruhe.de
spectra show a strong band at 1912 (8a) and 1896 cm-' (8 b),
Priv.-Doz. Dr. M. Haser
respectively, corresponding to local D,, symmetry of the
Institut fur Physikalische Chemie der Universitat
M(CO), unit. A discussion of the W s E vibrations in the Raman
D-76128 Karlsruhe (Germany)
spectra of 2a and 6,can be found in a publication by Schrock et
Fax: Int. code +(721)6084856
a].J131 who synthesized 6 and the corresponding MO derivative
e-mail: marco(a>tchibm3.chemie.uni-karlsruhe.de
[**I Thls work was supported by the Fonds der Chemischen Industrie and the
by reaction (a) starting with LiAsPhH.
Deutsche Forschungsgemeinschaft. We thank Prof. Dr. R. Ahlrichs for valuconfirm
The crystal structure analyses of 1 b and 6 (Fig.
able comments and his coworkers K Eichkorn and F. Weigend for unpublished
their
isostructural
relationships
with
2
b.
All
compounds
exhibit
basis sets.
2492
Q VCH VeriugsxeseltschuJt mbH. 0-69451 Wernheim. 1996
0570-0833/96/352/-2492$15.00 + ,2510
Angew. Cllem. Inl. Ed. Engl. 1996. 35. No. 21
.COMMUNICATIONS
a linear N,,-W-E axis with a distorted trigonal-bipyramidal geometry at the central W atom.
The W-As distance in 6 is
2.290(1)
and isthan
therefore
13 pm Alonger
the
M(c0h
cryMe:
Me3Si
SiMe,
t
Me3SiC L Ajlj S M
'Sue,e ,
i
Me3Si
SMe3
< p y i M e 3
+ 2b
[M(C0)5(tht)L
oc
i ...co
analogous W-P distance in 2 b
(2.16 A). The corresponding
Mo- As bond length is reported
SWe3
to be 2.25 A.[lol The distortion
Me3Si
< j + y M e 3
of the trigonal bipyramid is
most pronounced for the As
derivative 6 (E-W-N2 angle in
N
Table 1 and 2) and lessens for
the phosphido complex 2 b and
8a. M = Cr
8b:M=W
still further for the chloro
derivative 1 b. Whereas the WN2 distances are almost not influenced by the substituent E, the extent of intramolecular
N1 + W coordination is strongly dependent on it. In accordance with the higher electronegativity, this distance in the
chloro-substituted compound 1b (2.182(6) A) is shorter than
that in the As derivative 6 (2.336(6) A), which is almost identical
to the distance found for the phosphido complex 2 b.
Complex 8 b (Fig. 2)[14] exhibits a linear N,,-W-P-W-P-WN,, axis. The W1 atoms are coordinated in a trigonal-bipyramidal fashion, the W2 atom resides in an octahedral environment.
In comparison with 2 b the coordination of the phosphido ligands at the Wo center causes a lengthening of the W-P triple
bond by 4 pm to 2.202(2) (Table 1 and 2). The mean W2-P
distance is 2.460 A and is therefore shorter than that observed
in [(CO)5W(PMe,)]['s1 (2.516 A), although it is within the same
region as that in trans-[(CO),W(PR,),] (2.458 A).[161n Back
donation of the central Wo atom to the phosphido ligand weakens the triple bond.
N
2b
P
I,...
(d)
OC+
i
7a: M = Cr
7b:M=W
b
W
Fig. I . Molecular wucture of 6 (without H atoms)
Table I . Comparison of selected experimental and calculated bond lengths
[A]and angles ['I
of complexes 1b, 2b, 6, and related compounds (see Fig. 1 for atom labeling).
Compd
W-E
W-Nl,,
W-N2,,
E-W-N2
N2-W-N2
Nl-W-N2
1 b (expt)
I b (calcdl
[(N ,N)W =N] (calcd)
2b (expt)[51
2 b (cdlcd)
2.39912)
2.434
1.724
2.162(4)
2.181
2.2903(11)
2.290
2.514
2.175
2.182(6)
2.434
2.524
2.34(1)
2.527
2.336(6)
2.521
2.530
2.391
1.985(11)
2.028
2 021
1.975(6)
2.019
1.989(4)
2.020
2.020
2.025
98.89(11)
99.8
102.9
101.9(2)
103.5
102.1 5(11)
103.6
104.0
102.1
117 66(6)
117.2
115.2
115.8(11)
114.7
115.69(8)
114 7
114.4
115.7
81 ll(11)
80.3
77.1
78.1(2)
16.5
?7.85(11)
76 4
76.0
77.9
6 (expt)
6 (cdkd)
[ (N N)W ESbl (calcd)
[(N,N)W =P- BH,] (calcd) [a]
[a] The P B distance IS 1 943
A.
Table 2. Comparison of selected experimental and calculated bond lengths
[A]and angles ['I
Compd
Wl -E
W1 -Nax
Wt -Ncq
P-
7 b (Cdkd)
2.198
2.457
N1; 2.020
N2. 2.022
N3- 2.024
8 b (expt)
2.202(2)
2.290(6)
8 b (cdicd)
2.205
2.457
Angels. Clicvn. Inl. Ed. E q l . 1996. 35. No. 21
w2
of complexes 7 b and 8 b (see Fig. 2 for atom labeling).
E-WI -Neq
N,,-Wl-N,,
N,,-Wl-N,,
2.568
N1: 102.7
N2: 103 5
N3: 103.8
N1. N2: 114.3
N1, N3: 114.9
N2, N3- 115.4
N1: 76.8
N2: 76.6
N3: 76.8
N1: 1.985(7)
N2: 1.997(8)
N3: 2.001(8)
2.459(2)
2.460(2)
N1: 101.7(2)
N2: 102.1(2)
N3: 102.8(2)
N1, N2: 115.9(4)
N1, N3: 114.5(3)
N2. N3. 116.6(4)
N1: 78.2(3)
N2: 77.5(3)
N3 ; 77 6(3)
N1: 2.024
N2: 2.028
N3: 2.031
2.501
N1- 102.7
N2: 103.3
N3: 103.9
N1, N2- 114.1
N1, N3. 114.4
N2, N3: 116.1
N I : 76.8
N2: 16.5
CI VCH V e r l a ~ . ~ ~ e s e l l . ~mhH.
~ l l u l i0-69451 Weinheim.1996
0570-0833l96i3S21-2493 $ 15.00i.25 0
N3; 77.8
2493
COMMUNICATIONS
cri2
Fig. 2. Molecular structure of 8b (without H atoms).
The comparison of the 31P NMR data['2] of 2b, 7, and 8
indicates that the additional coordination of a metal center by
the phosphido substituent leads to a high-field shift and to a
considerable increase in the 1J(183W,31P)
values.['21 Thus, 31P
NMR chemical shifts and the 'J('83W,3'P) values in 3 (3a:
6 = 595.4, J = 535.8 Hz; 3b: 6 = 544.6, J = 554, 163 Hz) are
within the range of those found for complexes 7 and 8.
The following points were investigated by theoretical methods: 1) For which ligands E is reaction (c) thermodynamically
favored? 2) Why does 2b react with [M(CO),(thf)] to afford 8
instead of 7?3) Which ligating properties does 2 b have? 4) What
causes the considerable change in the 1J(183W,31P)value on
linear coordination of tungsten complexes with terminal phosphido ligands to Lewis acids? In order to discuss these points,
equilibrium structures of all compounds in question were calculated by applying the B-P86/SVP density functional approximation (see Tables 1 and 2).["] There is good agreement between
the calculated and experimentally observed structural parameters, with exception that the distance between W and the axial
nitrogen atom i s calculated to be 0.17-0.25 8, longer (the corresponding bond is rather weak) .['
Electronic energies, calculated with the B-P86/SVP approximation, show reaction (c) to be slightly exothermic (exception:
E = N ; cf. Table 3). These calculated reaction energies are
Table 3. Calculated reaction energies (without correction of zero point vibrations)
of reaction (c) in kJmol-' as a function of the terminal ligands E.
~
are shifted by +42 2 kJmol-' into the endothermic region.
The reason for this systematic difference is the different spin
multiplicity of starting materials and products: compound 1b
exhibits a triplet spin state. A more accurate calculation of the
spin coupling energies for molecules of this size is currently not
feasible. Nevertheless, the fallowing conclusions can be drawn:
A) Reaction (c) is equally favored from a thermodynamic point
of view for E = P, As, or Sb, but not for E = N. This is surprising since compounds containing nitrido ligands have been
known for a long time. The probable explanation lies in the high
stability of the amide LiN(SiMe,), , which would be a starting
material in reaction (c). B) Reaction (c) is most exothermic
when E = As. C) From a purely thermodynamic point of view,
since reaction (c) had been carried out successfully for E = P, it
should work for E = Sb as well.
To establish why 2b and [W(CO),(thf)] form 7b only as an
intermediate, which then preferentially yields 8 b, reaction energies for the reactions 2 b w(CO),(thf)] + 7 b +THF, and 2 b
+ 7 b + 8 b + CO were calculated. In the B-P86/SVP (B3-LYP/
SVP) approximation these reaction energies are - 51
( - 40) kJmol-', and +66 (+ 58) kJmol-', respectively. Evidently, CO is a more strongly binding ligand than 2b. The
apparent contradiction with experimental findings can be resolved if the reaction [W(CO),(thf)] + CO + [W(CO),] + THF
is taken into account. With the formation of w(CO),] acting
as a driving force,[241 the overall reaction 2b + 7 b +
[W(CO),(thf)] + 8 b + [W(CO),] THF has a reaction energy
of -19 (-14) kJmol-'. Further reaction of 8b with elimination of CO is not feasible for steric reasons.
The characteristics of bonding in 2b and in the other complexes were investigated by PESHO analysis[2s1(PESHO = pair
of electron sharing hybrid orbitals).[261The W=P bond in 2 b
largely consists of one do-p, bond and two d,-p, bonds
(Table 4, Fig. 3); there are only small contributions from the 6s
+
+
Table 4. Bonding in 2 b and its complexes characterized by PESHOs. For further
explanations about the information given in the Table see reference [25]. The )Ih
values associated with the radii R,, R , are recorded in columns 5-8.
Compound
A
B
h
60
R , , RB [pml
80
100 120
2b
W
P
[(N,N)WP-BH,]
W
P
P
B[d]
1
2
3
4
1
2
3
1
2
3
0.09
0.07
0.07
0.02
0.09
0.07
0.07
0.11
0.02
0.02
0.24
0.20
0.20
0.04
0.22
0.20
0.20
0.30
0.06
0.06
W1
P
1
2
0.08
0.07
0.07
0.06
0.02
0.02
0.21
0 19
~
Method
E
N
B-P86/SVP
B3-LYPISVP
+ 61
+ 104
+ 26
- 14
- 30
- 25
P
As
Sb
8b
+13
f19
P
W2
3
1
2
3
based upon the assumption that all lithium compounds appear
in a partially aggregated state: LiE(SiMe,), as a dimer with D,
symmetry,['g1 LiCl as a tetramer,["I and LiSiMe, as a hexamer.[211Interactions with the solvent (toluene) and with donor
compounds (E(SiMe,), , T H F ) that are present in low concentration are ignored!221 Reaction (c) is supposed to be favored by
entropy due to an increase in the number of particles.
Reaction energies were also determined by the B3-LYP/SVP
density functional approximation[23]at fixed B-P86/SVP equilibrium geometries (Table 3). This yields reaction energies that
2494
0 VCH
Verlagsgesell.~chaftmhH. 0-69451 Weinheim, 1996
0.19
0.16
0.06
0.06
0.46
0.37
0.37
0.07
0.40
0.37
0.37
0.55
0.11
0.11
0.72
0.54
0.54
0.12
0.67
0 54
0.54
0.84
0.20
0.20
0.39
0.62
0.52
0.51
0.44
0.19
0.18
0.36
0.35
0.28
0.12
0.12
[a]
[a] Dominant electron sharing mechanism. [b] See Figure 3. [c] The fourth PESHO
with its significantly smaller shared electron number qh signals that electron sharing
between W and P IS qualitatively described by a triple bond (as opposed to a
for A = P and B = B cannot be directly
"quadruple bond"). [d] qfi(A,RA;B,R8)
compared to tfh(A,R,;B,R,) for A = P and B = W2, because typical valence radii
of W and B are different (PESHO analysis does not rely on empirical data).
[el Electron sharing between P and Bin [(N ,N)WP BH,] is dominantly of CT type;
A contributions are smaller by a factor of 5. [f] IS and A contributions to electron
sharlng between P and W2 in 8 b differ by a factor of 2.5. A comparison of radial
electron densities pp(r.l,rn)(not shown) reveals that s and p, denslttes at phosphorus
in 2 b are larger than in 8b, while the reverse holds f o r p z densities. This may be
interpreted as IS donation and A acceptance of 2 b in 8b.
0570-0833/96/3521-24948 15.00f .25/0
-
Angew. Chem. Int. Ed. Ennl. 1996, 35. No. 21
COMMUNICATIONS
Keywords: arsenido complexes . complexes with elements of
group 15 phosphido complexes tungsten compounds
-
P
I
'\
I
I1
111
Fig. 3. PESHOs JA,R,;%) and IB.R,,%) with A = W and B = P; R, = R, =
100 pm (radii of the circles drawn) for compounds 2 b (I: h = 2; 11: h =1) and
[(N,N)W-P BH,](III:h = l ) . TheamplitudesofthePESHOs weremultiplied by
qb (cf. Table 4) and subsequently plotted as contour lines. Bold lines correspond to
nodal lines (zero amplitude): other contour lines represent amplitudes of kO.02.
The centers of the circular frames are occupied by the
k0.04. 2 0 OX. etc (in
corresponding nuclei (not drawn). their vertical distances are to scale. Since an
effective core potential has been used at W. its valence orbitals exhibit few if any
inner nodes. Otherwise, I corresponds to a typical p.--d, bond between P and W.
Comparison of I1 and 111 shows !he increasing contribution from the phosphorus
3s orbital to the p, ~ d bond
,
upon coordination to BH,.
orbital of tungsten and the 3s orbital of phosphorus to the d,-p,
bond (Fig. 3). Coordination of 2 b to an (3 acceptor, for example
BH,, leads to formation of a weak o bond between P and B
(Table 4). with simultaneous shortening of the W = P bond (cf.
calculated values in Tables 1 and 2). This effect can beexplained
by increased participation of the 3s orbital of phosphorus in
covalent interactions with its partners (Fig. 3). Theenergy of the
reaction [(N,N)WP] + H,B(thf) + [(N,N)WP-BH,]
THF
is calculated to be -22 ( - 2) kJ mol-' by using the B-P86/SVP
approximation (B3-LYP/SVP approximation). Accordingly, a
possible borane adduct of 2 b would be much less stable than the
complex with [W(CO),], which compares with our experimental
observations. This can probably be explained by the 7~ acceptor
ability of ligand 2b, which is shown, for example, by PESHO
analysis of 8 b (Table4). Since the W r P - 7 ~bond is thereby
weakened, the W = P distance in 8b is 2.4 pm (calculated) or
4.0 pm (observed) greater than in 2 b. Thus, the phosphido complex 2 b is characterized by similar, but less pronounced, ligating
properties than CO.
The increase of the 183W,31Pcoupling constant upon linear
coordination of 2 b to a 0 acceptor may be rationalized by the
stronger phosphorus 3s orbital involvement in W-P bonding
upon complex formation (Fermi contact term).
+
E.uperinieii/ul Procedure
M. Scheer, S. Gremler, E. Herrmann. U. Griinhagen. M. Dargatz, E. Kleinpeter, 2. Anorg. A&. Chem. 1991, 600. 203-210; M. Scheer. F. Uhlig. T. T.
Nam. M. Dargatz. H.-D. Schidler. E. Herrmann. ;hid. 1990.585. 177- 188;M.
Scheer. S. Gremler, E. Hernnann, P. G. Jones, J: Orgunomer. Chem. 1991,414,
337-349; M. Scheer, S. Gremler, 2. Anorg. A&. Chem. 1993,619. 471 -475.
K. Schuster, Diplomarbeit, Universitat Halle-Wittenberg, 1991, M. Scheer. K.
Schuster. J. Muller. E. Matern, unpublished results.
K.-Y. Shih, K . Totland, S. W. Seidel, R. R. Schrock, J. A m . Chem. Soc. 1994,
116, 12103.
Z. Duan. J. G. Verkade, Inorg. Chem. 1995.34, 1576-1578.
N. Zanetti, R. R. Schrock, W. M. Davis, Angen. Chon. 1995. 107.2184-2186;
Angew. Chem. I n / . Ed. Engl. 1995.34. 2044-2046.
C. E. Laplaza. W. M. Davis. C. C. Cummins. Angew. Chem. 1995, 107, 2181 2183; Angew. Chem. l n t . Ed. Engl. 1995,34, 2042-2043.
M. Scheer, K . Schuster. T. A. Budzichowski, M. H. Chisholm, W. E. Streib,
J. Chem. So< Chem. Commun. 1995, 1671-1672.
Highlight: M. Scheer. Angew. Chem. 1995. 107.2151 - 2153, Angen. Chem. Int.
Ed. Engl. 1995,34,1997-1999; some of the results were presented in a lecture:
M. Scheer, lecture at the 21 1th ACS Meeting, New Orleans 1996. see INOR 222
in the book of abstracts.
In a sealed NMR tube. reaction (c) was monitored for 48 h by "P N M R
spectroscopy in [DJtoluene starting at ambient temperature and heating t o 60
and 80 'C. Measurements were taken every 2 h in the beginning, later every 8 h.
Small amounts of 2 b and P(SiMe,), could already be detected after 2 h. The
increase in the amount of 2 b formed is accompanied by a decrease in the
amount of LiP(SiMe,), and a n increase in that of P(StMe,),. The trace
amounts of HP(SiMe,),, evidently caused by hydrolysis at the inner wall of the
tube. remain constant.
R. R. Schrock. lecture at the 21 1th ACS Meeting, New Orleans 1996, see INOR
51 7 in book of abstracts.
The reaction was monitored by "P N M R spectroscopy between -40°C
and ambient temperature Intensities of the products showed that the complexes 7 are only intermediates; complexes 8 are formed as the main
products with an increase in temperature and time. After 48 h 8 is formed
u uantitativelv.
[I21 IR (CH,CI,): 8a: 1912 (s). 8 b : 1896 (s)cm-'; ' H NMR(250.133 MHz, C,D,,
298K,TMS)6:6=3.54(t,6H.CH,),1.55(t.6H,CH,)079(~,27H,CH,);
8 b : 6 = 3.14 (t. 6 H , CH,). 1.60 (t, 6 H , CH,) 0.64 (s); 'C N M R (62.896 MHz,
toluene-[DJTHF, 298 K). 6: 6 = 55.3 (s. CH,), 52.3 (s. CH,). 6.9 (s. CH,);
31P('H) N M R (101.256MHz. C,D,, 298K. 85% &PO, ext.); 2b.
6 =1077.6. iJ('83W,31P)=136 Hz; 7 a : 6 =708.1. 'J(18JW.'iP) = 442. 7b:
6 = 662.6, iJ(1*3W,31P)= 450. 135; 8 a : 6 =728.1, iJ(i83W.31P)= 413 Hz;
= 426. 151; EI-MS (70 eV; 100 C) 6 : mi: ("A).
8 b : 6 = 679.8. 'J('"W."P)
618 (16) [ M + ] . 603 (100) [ M + -CH,]; 8 a (140'C): m:: ( O h ) : 611 (6)
[(N,N)W=P + G I + , 574 (7) [(N,N)W=P]+, 559 (8) [(N,N)W=P-CH,]+; 8 b
(130 'C): mi; (Yo):758 (6) [(N,N)W=P -* W]', 574 (4) [ ( N , N ) W s P ] + .559(6)
[(N,N)W=P - CH,]'.
[13] R. R. Schrock, personal communication; J. A. Johnson-Carr, N. Zanetti,
R. R. Schrock. M. D. Hopkins, J. Am. Chenz. Soc. in press.
[14] Crystal data and details of refinement: data were collected on a Stoe IPDS
diffractometer using Mo,, (1. = 0.71069 A) radiation. T = 200(1) K. no absorption corrections were performed; the structures were solved by direct
methods, method of least-squares refinement, all non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were fixed in caicuiated positions and
were refined isotropically. l b : C,,H,,CIN,Si,W, M , = 579.07, crystal dimensions 0.30 x 0.30 x 0.18 mm3. cubic, space group Pu3 (no. 205); u =
i7.220(2) A,
= 8, v = 5i06.2(io)A3, pellid=i.507 Mgm-',
/[(MO~=
J
47.76 cm-I. 1687 independent reflections (20,,, = 52 ), 1494 observed with
F0>4u(F,); 76 parameters, R, = 0.0340. w R , = 0.0958; 6. C,,H,,AsN,Si,W,
M , = 618.54. crystal dimensions 0.38 x 0.38 x 0.15 mm3, cubic, space group
P U ~ (no. 205);
=i7.255(2) A,
= 8, v = 5137.4(10)A3. pcalCd
=
1.599 M g m - , p(Mo,.) = 59.21 cm-'. 2080 independent reflections (20,., =
56;). 1785 observed with Fo24u(Fo); 76 parameters, R , = 0.0402,
wR, = 0.1063; 8 b : C.,,H,,O,P,Si,N,W,
x2C,H8. M , = 1629.34. crystal dimensions 0.2 x 0.08 x 0.08 mm', monoclinic. space group P 2 ( l ) j n (no. 14);
u=11.587(2), h=16.700(3). c=16.971(3)A, 8 = 9 7 . 2 1 ( 3 ) . 2 = 2 , V =
3258(1)81', pralir=1.661 Mgm-3. ~(Mo,.) = 54.87 c m - ' . 6088 independent
reflections (20,,, = 52.). 4939 observed with Fo>40(F,). 294 parameters,
R , = 0.0487, wR, = 0.1278. All calculations were performed by using the
SHELXS-86 [29]. SHELXL-93 [29] and Schakal-92 programs. Further details
of the crystal structure investigations may be obtained from the Fachinformationszentrum Karlsruhe. D-76344, Eggenstein-Leopoldshafen. on quoting !he
depository numbers CSD-405518, CSD-405519, and CSD-405520.
[15] F. A. Cotton. D. J. Darensbourg. B. W. S. Kolthammer. Innrg. Chem. 1981.20,
4440-4442
[I61 E. Fluck. P. Kuhn. A. Miiller, H. Bogge, Z. Anorg AIIg. Chem 1988, 567,
13-22.
z
z
2b.6: 1 b(103 mg.0.18 mmo1)and LiE(SiMe3),.2THF(0.37 mmol)(E = P[27],As
[28]) in toluene (10 mL) were stirred for 48 h a t 80 and 110 C, respectively. The deep
red suspension was reduced to dryness in vacuum and the residue was extracted
twice with hexane (40 mL). The solvent was removed in vacuum until the onset of
crystalliration. Compounds 2 band 6 crystallized at - 20 'C as yellow-orange cubes.
The compounds could he recrystallized from Et,O (2b: 67 mg, 65%; 6: 53 mg,
48 Y o )
8: To a solution of 2 b (104mg. 0.18mmol) dissolved in toluene (lOmL),
[M(CO),(thf)] ( M = Cr, W) (9 mL of a 0 . 0 2 ~solution in THF). prepared by
irradiating M(CO), m a UV-apparatus, was added and stirred for 60 h a t ambient
temperatiire. The solvent was removed in vacuum until crystals began to form. At
- 2 0 C Sa (X1 mg. 68%). 8b (96 mg, 74%), respectively, crystallized as red rhombuses from deep red solutrons.
Received: June 18, 1996 [Z92381E]
German version. Angel<. Chem. 1996, IOX. 2637 2641
~
Anget*,. Chlwr. l n l . Ed Engf. 1996. 35. s o . 21
-
g]
V C H Verlugsgesellschuft mbH. 0-69451 Weinheim. 1996
0570-083319613521-24958 15.00+ .25 0
2495
COMMUNICATIONS
[17] Structure optimizations were performed by using the TURBOMOLE set of
programs with the RI-lapproximation (K. Eichkorn. 0. Treutler, H. Ohm, M.
Hiiser, R. Ahlrichs, Chem. Phy.v Leu. 1995. 242, 652-660). For definition of
the B-P86 density functional see A. D. Becke, Pl7.v~.Reif.A 1988,38, 3098; J. P.
Perdew, Phys. Rev. B 1986.33.8822; hid., 1986.34. 7046. The acronym SVP
refers to TURBOMOLE split valence basis sets, augmented by a shell of polarization functions, cf. A. Schifer, H Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97,
2571. Quasi-relativistic pseudopotentials were used for the elements Wand Sb
(D. Andrae, U. Hiussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. A m
1990, 77,123; A. Bergner, M. Dolg, W Kuchle, H. Stoll, H. Preuss, Mol. Phys
1993,80, 1431). The corresponding SVP basis sets optimized for W and Sb by
F. Weigend, K. Eichkorn and R. Ahlrichs are unpublished.
[18] Results are not improved upon switching to the MP2/SVP approximation
(SiMe, in 2 b replaced by Me).
I191 D,-symmetric (LiE(SiMe,),), corresponds to the crystal structures of
[LiP(SiMe,),(thf)], (cf. G. Becker, H.-M. Hartmann, W. Schwarz, Z . Anorg.
Allg. Chem. 1989, 577, 9-22) and [LiAs(SiMe,),(dme)], (cf. G. Becker, C.
Witthauer, 2. Anorg. Allg. Chem. 1982,492,28-36). The analogous antimony
compound prefers a slightly different mode of aggregation (cf. G. Becker, A.
Munch, C. Witthauer, 2. Anorg. Allg Chem. 1982,492, 15-27).
[20] In the experiment, LiCl is formed as a solid in reaction (c).
[21] The difference in energy between the hexamer and the tetramer was calculated
to be 4.5 kJ per mol of LiSiMe, (B-P86/SVP). The crystal structure of the
hexamer is known: T. F.Schaaf, W. Butler, M. D. Glick, J. P. Oliver, J. Am.
Chrm. Soc. 1974,96,7593-7594; W. H. Ilsley, T. F. Schaaf. M. D. Glick, J. P.
Oliver, ihid. 1980. 102, 3769-3774.
[22] Calculated reaction energies will be correct as long as solvation and aggregation of the lithium compounds are considered in a similar approximation for
the starting materials and for the products. Relative reaction energies amongst
pnictido ligands E are supposed to be very reliable, since they are not influenced by errors associated with lithium aggregation.
[23] For a concise definition of the B3-LYP density functional see C. W.
Bduschlicher, H. Partridge, Chem. Phys. Leu. 1994,231, 277. The calculations
were performed by using theTURBOMOLE set of programs (R. Ahlrichs. M.
Bar, M. Haser, H. Horn, C. Kolmel. Chem. PI7ys. Lerr. 1989. f62, 165; 0.
Treutler, R. Ahlrichs, J Chem. P h i s 1995, 102, 346).
1241 M. Scheer, C. Troitzsch. P. G. Jones, Angeu. Chem. 1992, 104, 1395-1397;
Angen. Chem. Inr. E d Engi. 1992, 31, 1377-1379; M. Scheer, U. Becker,
Phosphorus. Sulfur. Silicon. 1994.93-94,391-392; M. Scheer, C. Troitzsch, L.
Hilfert, M. Dargatz. E. Kleinpeter. P. G. Jones, J. Sieler, Chem. Eer. 1995. 128,
251-257; M. Scheer, U. Becker, J. C. Huffman. M. H. Chisholm, J.
Organomel. Chem. 1993,461, C1-C3.
[25] M. Hiiser, J. A m . Chem. Sac. 1996. tf8,731 1-7325. A PESHO is a pair of
atomic hybrid orbitals I A , R 4 ; h ) and lB,RB;h); here "atomic" means
lA,RA;h) and IB,R,;h) are contained within spherical neighborhoods of the
nuclei A and B, respectively, with radii given by R , and R E , respectively; the
PESHOs are obtained by requiring the closest description of electronic delocalization between the atomic neighborhoods of A and B (thus approximating
the oneelectron density operator) for any pre-set number of PESHOs
...). EachPESHO(lA.R,;6), IB,R,;h))ischaracterized byanumber O<q,(A,R,;B,R,)I2
of electrons shared between l A . R A ; T ) and
lB,R,;K). The number of PESHOs with values qh(A,R,;B,Rn) significantly
different from zero relates to the number of covalent bonds between A and B,
while q k itself refers to the extent of the corresponding electron sharing phenomenon. Entries in Table 4 are values of q,(A.R,;B,R,) for R, = R, = 60.80,
100. and 120 pm.
[26] The analysis of bonding was based on Kohn-Sham orbitals from the B-P86/
SVP calculations. Conclusions regarding electronic deloca~izationremain valid
only in as far as electronic correlations are of minor importance.
[27] F. Uhlig, S. Gremler, M. Dargatz, M. Scheer, E. Herrmann. 2. Anorg. A l k .
Chem. 1991,606. 105-108.
[28] G. Becker, G. Gutekunst. H.-J. Wessely, Z . Anorg. Allg. Chem. 1980, 462.
1 13- 129.
[29] G. M. Sheldrick, SHELXS-86. Universitat Gottingen, 1986: SHELXL-93.
Universitat Gottingen, 1993.
Self-Assembling Covalently Linked
Supramolecular Arrays of Defined Structure :
The Remarkable Redox Reactivity of 15-mesoSubstituted 5-Oxyporphyrins**
Richard G. Khoury, Laurent Jaquinod, Daniel J. Nurco,
Ravindra K. Pandey, Mathias 0. Senge, and
Kevin M. Smith*
The synthesis and chemistry of organized supramolecular assemblies have received considerable attention recently, particularly with regard to self-assembly of macromolecules to give
supramolecular arrays possessing, or potentially possessing, biological or other catalytic activity-l'. z] It is hoped that such
biofunctional assemblies might show unique biomimetic function, demonstrate molecular discrimination enabling unique
catalytic activity, or act as selective transport agents. In the
present paper we show that, based on their redox properties,
certain carefully designed oxyporphyrins (oxophlorins) can be
induced to reversibly self-assemble, by way of covalent carboncarbon bonds, to give supramolecuIar assemblies with defined
structures.
Iron complexes of oxophlorins 1 have been postulated as
intermediates in the catabolism of heme leading to bile pigm e n t ~ . [ Oxophlorins
~]
are porphyrinoids that have a meso carbony1 group, and exist in the keto form 1 rather than in the enol
form 2 in neutral solution.[41In acidic solution or in the presence
15
1
2
of divalent metals, the enol structure predominates, but with
trivalent metal ions such as iron(Irr), either the keto or enol form
can predominate, depending upon pH.14' Based on the known
radical chemistry of metallooxophlorins (due to their low oxidation potentials)"] and the tendency of the radicals to dimerize to
give unique structures,[61we chose to synthesize a number of
rneso-substituted oxophlorins and then investigate their chemical oxidation to provide regio- and stereochemically pure
oligomers. All known radical dimerizations of 5-oxophlorins''. 61 involve the 15-position, and thus the largest oligomer
that can be obtained is a dimer. In order to investigate access to
oligomers (and therefore supramolecular assemblies), we targeted 15-substituted 5-oxophlorins in the hope that radical
oligomerization reactions of these substrates would take place
['
[*I
Prof. Dr. K. M. Smith, R. G. Khoury, Dr. L. Jaquinod, D. J. Nurco,
Dr. R. K. Pandey,'+' Dr. M. 0. Senge" '1
Department of Chemistry, University of California
DdVlS, CA 95616-5295 (USA)
Fax. Int. code +(916)752-8995
e-mail: smith@chem.ucdavis.edu
['I
Permanent address: Department of Radiation Biology
Roswell Park Cancer Institute, Buffalo, NY 14263 (USA)
Permanent address: Institut fur Organische Chemie (WE02)
Freie Universitdt Berlin, Takustrasse 3, D-14195 Berlin (Germany)
'1
[**I This work was supported by the National Science Foundation (CHE-9305577). by the National Institutes of Health (HL-22252). and by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen Industrie.
2496
(Q VCH Verlugsgeseilschafl mbH, 0-69451 Weinheim. 1996
OS70-0833/96/35Zr-2496$ lS.OO+ .2SlO
Angeu. Chem. h i . Ed. En$
1996, 35, No. 21
Документ
Категория
Без категории
Просмотров
0
Размер файла
626 Кб
Теги
containing, terminal, arsenic, complexes, ligand, phosphorus
1/--страниц
Пожаловаться на содержимое документа