close

Вход

Забыли?

вход по аккаунту

?

Clusters from Vertex- and Face-Sharing Adamantane-Like Units A New Topology for Multinuclear Complexes.

код для вставкиСкачать
COMMUNICATIONS
Clusters from Vertex- and Face-Sharing
Adamantane-Like Units: A New Topology
for Multinuclear Complexes**
Euan K. Brechin, Steven G. Harris, Simon Parsons,
and Richard E. P. Winpenny*
The discovery of novel metal polyhedra remains one of the
chief goals in the synthesis of multinuclear complexes, as new
structures may lead to novel properties. The adamantane-like
structure featuring a central M,O, core has been reported for
the 3d-metals from Ti to Fe;“ -91however, it remains comparatively rare, and only a dozen or so examples are known.
Compounds containing this core have never been reported
for later 3d-metals, and only one example exists where adamantane units are linked into higher nuclearity clusters; this is
a hexairon complex reported by Nair and Hagen[”] in which
two adamantanes share a common edge. Here we report the
first nickel and cobalt complexes that contain M,O, units,
which are also the first examples of vertex- and face-sharing
adamantanes.
Reaction of “Ni(OH),” with 2.1 equivalents of 6-chloro-2hydroxypyridine (Hchp) at 130 “C under N, gives a green paste,
which, after drying in vacuo, can be crystallized from MeCN/
MeOH to give green crystals. Structural determination“ reveals a heptanuclear nickel complex [Ni,(chp),,Cl,(MeOH),]
(1, Figure 1 ) . The central Ni site (Nil) lies on a crystallographic
inversion center and is the shared vertex of two Ni,O, adamantanes; the 0 atoms are provided by the chp ligands. The central
Ni atom is bound to six oxygen atoms derived from chp ligands
that form chelates with the “basal” Ni sites in the structure.
Each basal Ni atom is chelated to one further chp and bound to
a bridging 0-donor from the chp chelating to a neighboring
basal nickel atom. These three chelating chp ligands lie virtually
coplanar with the Ni, base. Thus chp donors occupy five coordination sites on each basal nickel; the sixth site is occupied by
a molecule of MeOH in each case. These three terminal MeOH
ligands form strong H-bonds to chloride ions [d(O.. . C1) =
3.11(4) .%.I stemming from the nickel chloride used to prepare
“Ni(OH),”.
If the same procedure is utilized, but the nickel chloride is
treated with sodium methoxide prior to reaction to produce
“Ni(OMe),”, another heptanuclear complex, [Ni,(OH),(chp),,(MeOH),] (Z),can be crystallized. Structure determination[“] shows that a very similar cluster has formed. Again a
central Ni atom, disposed on an inversion center and bound to
six 0 donors derived from chp ligands, is the shared vertex of
two Ni, cages; however, in 2 the ideal adamantane cages are
distorted by the presence of a p3-hydroxo ligand (01H) linking
the three Ni sites in the basal plane. These three Ni sites are now
dissimilar. Ni2 is bound to two chelating chp ligands, the p 3 OH, and a chp oxygen atom that is p,-bridging to Ni3; the
pyridine-N donor of this chp ligand forms a hydrogen bond to
a MeOH ligand. Ni3 is bound to one chelating chp, the p,-OH,
p,-oxygen atoms from two chp ligands, and a terminal MeOH
ligand. Ni4 is bound to one chelating chp, the p,-OH, one p,O(chp), and two terminal MeOH ligands.
To remove the influence of MeOH from this chemistry, a
similar product was crystallized from MeCN only (see Experimental Section). Structure determination“ ‘I revealed a nonanuclear cage [N~,(OH),(C~~),,(M~CN)~]
(3, Figure 2), in which a
C17
Figure 2. The structure of 3 in the crystal. Bond length ranges [A]: N i - 0 2.011 2.198, Ni-N 2.053-2.109 (esd0.005). Bond angle ranges: cis at Nil 85.1-95.3,
fransatNil 175.8-177 7,cisatother Ni 62.7-109.4, transat other Ni 148.8-169.6
(esd 0.2)
Figure 1. The structure of 1 in the crystal. The structure of 2 is extremely similar in
the crystal Bond length ranges [A] (in square brackets data for 2): Nil -0 2.3372.369, other N i - 0 2.002 ~-2.145[ 1.976-2.1 651, Ni -N 2.066-2.105 [2.050-2.1161,
cis at
(av. estimated standard deviation (esd): 0.007 [O.OlO]). Bond angle ranges
Nil 89.4-90.6, cis at other Ni 63.2-108.5 [85.8-94.21, lrans at other Ni 152.6169.3 (157 0-171.01 (av. esd0.3 [0.4]).
r1:
[*] Dr R. E. P. Winpenny, E. K. Brechin, S . G. Harris, Dr. S Parsons
Department of Chemistry
The University of Edinburgh
West Mains Road, Edinburgh, EH9 3JJ (UK)
Fax. Int. code +(131)667-4743
E-mail: repwOl (o,tattoo.ed.ac.uk
[**I This work was supported by the EPSRC (UK).
Angen Chvm Int Ed Engl 1997.36,N o 18
central six-coordinate nickel atom (Nil) again lay on a symmetry element-in this case a twofold axis. A very similar heptanuclear cage to that in 2 is present, in which a p3-OH ligand bridges
the three Ni sites, Ni2, Ni3, and Ni4, within the basal plane.
However, the absence of MeOH ligands has led to a slightly
larger cage, and an additional [Ni(chp),]- unit containing Ni5
is attached to one edge of the triangular base.
The three Ni sites in the base are again distinct. Ni2 and Ni4
are identical, bound to two chelating chp ligands, the p3-OH
ligand, and the p2-0-atom that attaches the [Ni(chp),]- unit to
the cage. Ni3 is unique, bound to one chelating chp, the p,-OH
ligand, two p,-0-donors from chp ligands chelating to Ni2 and
Ni4, and a terminal MeCN ligand.
0 WILEY-VCH Verlag GmbH, D-69451 Weinhelm, 1997
0570-0833/97/3618-1967 $17 50+ 50:O
1967
COMMUNICATIONS
The reaction that produced 3 was repeated with cobalt in
place of nickel. The nonanuclear cage compound [Co,(chp),,]
(4) crystallized from EtOAc in a form suitable for structure
determination" (Figure 3). The cage has crystallographic D,,
d
Figure 3. The structure of 4 in the crystal. Bond length ranges [A]: C o l - 0 2.316,
2.026-2.211, Co2-N 2.122-2.126, C O 3 - 0 2.185, CO3-N 2.126
(esd 0.007). Bond angles: trans at Col 175.6, cis at Col 85.3 and 91.6, (runs at C02
165.3 and 167.2, cis at C02 61.4-132.3, trans at co3 151.9, cis at Co3 62.1-104.3
(esd 0.3).
C02-0
symmetry, and the central cobalt, Col, lies on both a threefold
axis (which also passes through co3) and a twofold axis. The
cage containing four adamantane units is clearly related to the
three nickel cages. Two of the adamantane units share Col as a
common vertex; two additional units share the basal planes
(C02 and symmetry equivalent atoms) with c o 3 (and symmetry
equivalent) as vertex. The three crystallographically unique Co
sites are chemically distinct. Col is bound to six 0-donors in an
identical manner to the central metal sites in 1-3. C02 is bound
to two chelating chp ligands, one p 2 - 0atom from a further chp
ligand within the basal plane, and one p 2 - 0 donor that attaches
the c o 3 vertex to the cage. c o 3 is chelated to three chp ligands
infac geometry; each 0-donor bridges to the Co atoms of the
basal plane. A comparison of 4 with 1 shows that the three
terminal MeOH ligands and the H-bonded chloride ion have
been replaced by a [Co(chp),]- moiety, leaving the connectivity
of the heptanuclear metal core unchanged.
Metal coordination geometries in 1-4 can be related to an
octahedron, with one exception. In each cage the central metal
site has a regular geometry (range for trans angles 175.8-180.0";
for cis angles 85.1-95.3"), whereas the other metal sites are
distorted by the narrow bite angle of the chelating chp ligands
(range for Ni cages: trans angles 148.8-171.0'; for cis angles
62.5 to 109.7'). For 4 the distortion at C02 is so large (only two
angles approaching a typical trans value at about 166") that an
octahedra1 description becomes difficult to maintain and the
geometry is best described as irregular.
The metal-to-metal distances within the adamantane units
vary from 3.808 to 4.143 A for the nickel cages and 3.742 to
4.1 16 A for 4. The metal-to-ligand bond lengths vary greatly
depending on the metal site. The M - 0 lengths to the central
metal in structures 1 and 4 are long, varying from 2.316 to
2.369 A. The length of these bonds led us to examine the possibility that this site was occupied by Na rather than Ni or Co;
however, refinement of the structure with a Na atom in place of
Nil or Col produced physically meaningless results. Analytical
data strongly support the formulae given, as does mass spec1968
0 WILEY-VCH Verlag GmbH, D-69451 Weinhelm, 1997
trometry. Fragments of the three Ni cages are observed in fast
atom bombardment mass spectrometry (FAB-MS) which are
very close to the molecular ion in mass, and for 4 peaks are seen
for fragments containing five Co centers, which would not be
feasible if the central metal site were not a cobalt atom. There is
no straightforward explanation for the observed bond lengths at
this site; steric crowding can be ruled out as the equivalent sites
in 2 and 3 have normal bond lengths, as do all other metaI
sites in the four structures (range for Ni-N 2.050-2.116; for
N i - 0 1.976-2.198; for Co-13 2.026-2.211; for Co-N 2.1222.245 A).
We have previously stated that the coordinative flexibility of
pyridonates and the presence of heteroleptic ligand sets are important factors in the formation of high nuclearity arrays of
these ligands;"'] however, 4 is a fascinating exception to this
rule in that it is homoleptic, and each pyridonate ligand in the
cluster has an identical coordination mode. The homoleptic chp
complex of the heavier Group 9 metal rhodium has been reported by Cotton et al.,['31 and the dimer [Rh,(chp),] contains a
Rh-Rh single bond. Compound 4 is a beautiful example of the
complicated arrays that 3d-metals must adopt to avoid forming
metal-metal bonds, and certainly this is a further factor contributing to the novel cages formed by 3d-metals with simple
bridging ligands.
Experimental Section
1: Hydrated nickel chloride (1 g, 4.2 mmol) was treated with NaOH (2 equiv) in
water (20 mL) to give a green precipitate of nickel hydroxide, which was dried in
vacuum. Hchp (1.088 g, 8.4 mmol) was added, and the mixture heated to 130°C
under N, for 2 h to produce a paste. This paste was dried in vacuum for 1 h before
being dissolved in 1: 1 MeCNjMeOH. Green crystals grew over 2 days. Yield: 30%.
Elemental analysis calcd (found) for C6,H,,C1,,Nl,Ni,0,,:
C 35.7 (35.6), H 2.71
(2.50), N 7.58 (7.40), Ni 18.5 (17.8)%. FAB-MS: m / z : 2181 [Ni,(chp),,CI(MeOH),I+, 1921 [Ni,(chp),,(MeOH),]',
1862 [Ni,(chp),,(MeOH),]. 1830
Pi,(chp), ,(MeOH),] , 1734 [Ni,(chp),(MeOH),] , 1605 [Ni,(chp),(MeOH),] +,
1509 [Ni,(chp),(MeOH),]+, 1477 [Ni,(chp),(MeOH),]+, 1413 [Ni,(chp),(MeOH)]', 1348 [Ni,(chp),(MeOH),I+, 1284 [Ni,(chp),(MeOH)], 1205 [Ni,(chp),]+,
889 [Ni,(chp),]+, 573 [Ni(chp),]+, 503 [Ni,(chp),]+, 375 [Ni,(chp),]+.
2 : Synthesis as for 1, using NaOMe in place of NaOH and MeOH in place of H,O.
Crystals grew after 2 days. Yield: 25%. Elemental analysis calcd (found) for
C,,H,,CI,,N,,Ni,O,,:
C 36.3 (36.0). H 2.84 (2.71), N 7.71 (7.63)%. FAB-MS: as
for 1 except peak at mjz 2180 assigned as ~i,(~hp)~,(0H),(MeOH)~]+.
3: Synthesis as for 1, except MeCN alone was used for crystallization. Crystals grew
after 1 day. Yield: 80%. Elemental analysis calcd (found) for C81Hs6C116N18NiY0,,: C 37.3 (37.3), H 2.07 (2.00), N 9.33 (9.30)%. FAB-MS: mjr: 2490
~Ni,(chp),,(OH),lf, 2174 ~Ni,(chp),,(OH),]+, 1602 [Ni,(chp),(OH),1+, 1526
[Ni,(chp),(OH)]+, 1288 [Ni,(chp),(OH)]', 1205 [Ni,(chp),]+, 889 [Ni,(~hp)~]+.
572 [Ni(chp),]+.
4: Synthesis as for 1, except that cobalt was used in place of nickel, and crystallization was from EtOAc and gave pink crystals after 1 week. Yield: 9%. Elemental
C 38.0 (37.7). H 1.90 (1.89), N
analysis calcd (found) for C,,H,,CI,,CO~N~~O,~:
8.86 (8.80), Co 18.7 (17.8)Yo. FAB-MS: mjz: 1522 [Co,(chp),,]+, 1323
[Co,(chp),]+, 1264 [Co,(chp),l+, 1195 [Co,(chp),]+, 1136 [Co,(chp),]+, 820
[Co,(chp),l+, 632 [CoAchp),l+, 573 [Co(chp),l+, 504 [Co,(chp),l+, 445
[Co(chp),l+ 3 16 [Co(chp),I +
+
+
.
Received: March 3,1997 [Z10189IE]
German version: Angew. Chem. 1997,109,2055-2057
-
Keywords: clusters cobalt
- multinuclear complexes
*
nickel
[I] K. Wieghardt, D. Ventur, Y. H. Tsai, C. Kruger, Inorg. Chim. Acta 1985, 99,
L25.
[2] L. M. Babcock, V. W. Day, W. G. Klemperer, J. Chem. SOC,Chem. Commun.
1987,858.
[ 3 ] F. Bottomley, C. P. Magill, B. Zhao, Organometallics 1990,9, 1700.
[4] D. Wormsbacher, K. M. Nicholas, A. L. Rheingold, J Chem. Soc. Chem. Commun. 1985,721.
[S] J. Glerup, H. Weihe, P. A. Goodson, D. J. Hodgson, Inorg. Chim. Acra 1993,
212,281.
[6] K. Wieghardt, U. Bossek, B. Nuber, J. Weiss, J. Bonvoisin, M. Corbella, S. E.
Vitols. J. J. Girerd, J. Am. Chem. Soc. 1988, 110, 7398.
0570-0833/97/3618-1968S 17 SO+ 50/0
Angew Chem
Int
Ed Engl 1997,36, No 18
COMMUNICATIONS
[7] K. S. Hagen, T D. Westmoreland, W. J. Scott, W. H. Armstrong, J. Am. Chem.
SOC.
1989, I l l . 1907.
[S] B. P. Murch, F. C. Bradley, P. D. Boyle, V. Papefthymiou, L. Que, Jr., J. Am.
Chem. Soc. 1987, 109, 7993.
[9] S. Drueke, K. Wieghardt, B. Nuber, J. Weiss, E. L. Bominaar, A. Sawaryn, H.
Winkler, A. X. Trautwein, Inorg. Chem. 1989,28,4477.
1101 V. S. Nair, K. S. Hagen, Inorg. Chem. 1992,31, 4048.
[ll] Crystal data for 1.1.7MeCN: C,, 4H6s
lCI,~N~3,,Ni,018,
M,= 2310, triclinic, PT. a=13.706(7), b=15.216(9), c=15.431(8)A, a=60.68(5), p =
67.15(4), = 69.33(4)", V = 2532 A', Z = 1 (the molecule lies on an inversion
=1.515 gem-', T = 220.0(2) K, crystai size 0.51 ~ 0 . 1 9 ~
center), pcSlrd
0.16 mm. ~(CU,,)= 5.33 rnm-'. Crystal data for 2: C,6H6,CI,,N,,Ni,0,0,
M,= 2180, monoclinic, P2,/n, a =12.502(4), b =19.917(6), c =17.601(4)A,
fl = 106.2312)'. V = 4208 A', Z = 2 (the molecule lies on an inversion center),
peslcd= 1.720 gcmT = 220.0(2) K, crystal size 0.31 x 0.27 x 0.19 mm,
p(MoK,) = 1.99 mm-'. Crystal data for3: C84H,6C116N,8Ni,0,8,M , = 2701,
monoclinic, C2,c, a = 31.570(3), b =12.784(2), c = 25.371(3) A,
=
101.160(10)". C'=10046A3, Z = 4 (the molecule lies on a twofold axis),
pESICd
= 1.786 gcm-',
T = 220.0(2) K, crystal size 0.23 x0.16xO.16 mm,
~(CU,.) = 6 33 mm-I. Crystal data for 4.3EtOAc: C,,,H,,CI,8C~gN,8024,
M , = 3108, rhombohedral, R ~ c , a = 18.298(2), c = 67.570(8) A, V =
19593 A', 2 = 6 (the molecule lies on 32 site), pCsrca
=1.581 g ~ m - ~ ,
T = 220.0(2) K, crystal size 0.16x0.16x0.04mm, ~(CU,.) =12.695mm-'.
Diffraction was very weak, and Cu, radiation was used for higher intensity.
Data were collected with a Stoe Stadi-4 diffractometer equipped with an Oxford Cryosystems low-temperature device. Absorption corrections were applied with Y-scan data for 2 and 3 (min./max. transmission: for 2 0.413/0.594;
for 3 0.193/0.343), with SHELXA for 1(min./max. transmission: 0.165/0.637),
and with DIFABS (N. Walker, D. Stuart, Acta Crysta/logr. Secl. A 1983, 39,
158) for 4 (min./max. transmission: 0.176/0.604). All structures were solved by
direct methods (SHELXTL, G. M. Sheldrick, Universitat Gottingen, 1993)
and completed by iterative cycles of AFsyntheses and full-matrix, least-squares
refinement against F' (SHELXTL). Hydrogen atoms were included in all
structures in calculated positions, riding on parent C atoms, with
U(H) = 1 2 U J C ) for chp H atoms and U(H) = 1.5 U,,(C) for methyl-H atoms.
For 1 there were partially occupied, diffuse solvent regions, which were modeled in the manner described by van der Sluis and Spek (Acta Crystallogr. Sect.
A 1990.46, 194). All non-H atoms were refined with anisotropic displacement
parameters: for 1, 529 parameters, wR2 = 0.2729 for 7366 unique data
(205 120'), R1 = 0.0890 for 4284 observed reflections, F,>4u(F), largest
residual difference peak and hole were, respectively, 0.984 and - 1 . 4 3 7 e k 3 ;
for 2 535 parameters, wR2 = 0.2220 for 5474 unique data (20545"),
R1 = 0.0827 for 2899 observed reflections, F,>4u(F), largest residual difference peak and hole were, respectively, 0.612 and - 0 . 8 9 9 e k 3 ; for 3 655
parameters, wR2 = 0.1524 for 7213 unique data (20c120"), R1 = 0.0602 for
4810 observed reflections, F,>4u(F), largest residual difference peak and hole
were. respectively, 0.496 and -0.504eA--'; for 4 285 parameters, wR2 =
0 2496 for 2729 unique data ( 2 0 1 120") and 305 restraints, R1 = 0.0771 for
1404 observed reflections, F, > 4u(F), largest residual difference peak and hole
were respectively, 0.700 and - 1 . 3 1 7 e k 3 . Crystallographic data (excluding
structure factors) have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication no. CCDC 100219. Copies of this
data can be obtained free of charge on application to The Director, CCDC, 12
Union Road, GB-Cambridge CBZIEZ, UK (fax: int. code +(1223)336-033;
e-mail : deposit(b,chemcrys.cam.ac.uk.
1121 S Parsons, R. E. P. Winpenny, Acc. Chem. Res. 1997,30, 89.
[I31 F. A. Cotton, T. R. Felthouse, Inorg. Chem. 1981, 20, 584.
',
a
Synthesis, Structure, and Oxidation of
Donor-Stabilized Gallium(1) Iodide:
Ga818*6PEt,**
Clemens U. Doriat, Markus Friesen, Elke Baum,
Achim Ecker, and Hansgeorg Schnockel*
Dedicated to Professor Wolfgang Beck
on the occasion of his 65th birthday
Gallium halides with low-valent gallium have been the subject
of intensive study in the fields of solid-state chemistry and
molecular compounds."] The readily available gallium(i1)
halides-which are present as gallium(I/w) halides as solids[21
and in donor-free solvents[3]but as genuine gallium(I1) species[41
with Ga-Ga bonds (Ga2X;2D) in donor-containing solvents
D-occupy a special position among the gallium halides because they can be used, for example, as starting materials in the
synthesis of organic gallium(1) compounds.[51The mixed-valent
Ga,Cl,. 5 OEt, is a peculiarity in which the Ga centers are formally present in the oxidation state
A binary Ga' halide, that is, a compound that contains no
Ga"' ions, has not yet been structurally characterized. Although
solid Gat could be prepared some years ago and its stoichiometric composition substantiated with subsequent reactions,['] a
structure determination of a solid GaX compound remained
elusive. As we have been studying donor-stabilized AIX species[*]prepared by cocondensation methods for some time, and
were successful in confirming the presence of both AIBrL9]and
AII['ol with planar Al, rings (A14X4.4NEt3),it was reasonable
to consider experiments with analogous GaX systems. We report here the synthesis and structure determination of the first
donor-stabilized Gat compound.
Condensing gaseous Gat (31 mmol, prepared from Ga and
HI at about 900°C)["1 with PEt, (48 mmol) and toluene
(1.05 mol) at - 196 "C and then thawing the matrix results in an
orange-red solution, from which an orange solid precipitates
at - 50 "C. The 31P NMR spectrum of the solution (300 MHz,
external standard 85 YO H,P04) shows a downfield-shifted
signal at 6 = -7.2 in addition to that for free PEt,
(6 = -19.7).['21 Analysis of the solid (yield about 65O/0), which
decomposes above 133 "C under disproportionation to metallic
gallium and GaI,.PEt,,['31 gives a Ga:I:PEt, ratio of approximately 1 :1 :0.75. Recrystallization from benzene provides yellow, hexagonal crystals that are suitable for a crystal structure
analysis (Figure 1). [ I 4 ]
The compound, Ga,I,.6 PEt, (l),has a planar eight-membered ring of Ga atoms that is bridged by two I atoms to form
a doubly transannular GaJ, ring. Thus, two of the eight Ga
atoms reach the coordination number 4; the remaining six Ga
atoms reach this coordination number by binding to one I atom
and one PEt, donor ligand. The eight Ga-Ga distances
(2.47 A) are of almost equal length. This value seems reasonable
when compared to known Ga-Ga distances in compounds of
the type Ga2X,-2D (2.39-2.42 A for X = C1, Br),C4] because
the radius of the Ga' ion should be larger than that of the Ga"
[*I
Prof. Dr. H. Schnockel, Dip].-Chem. C U. Doriat, Dip1.-Chem. M. Friesen,
Dr. E. Baum, Dr. A. Ecker
Institut fur Anorganische Chemie der Universitat
Engesserstrasse, Geb.-Nr. 30.45, D-76131 Karlsruhe (Germany)
Fax: Int. code +(721)608-4854
e-mail : hg@achpc9.chemie.uni-karlsruhe.de
p*] This work was supported by the Deutsche Forschungsgemeinschaft and the
Fonds der Chemischen Industrie. We especially thank Mrs. E. Mollhausen for
performing the X-ray diffraction experiments.
Angew. Chem. Int. Ed. Engl. 1997,36, No. 18
0 WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1997
0570-0833/9713618-1969$ 17.50+ SOjO
1969
Документ
Категория
Без категории
Просмотров
1
Размер файла
412 Кб
Теги
like, unit, multinucleate, sharing, clusters, facer, vertep, topology, complexes, new, adamantane
1/--страниц
Пожаловаться на содержимое документа