close

Вход

Забыли?

вход по аккаунту

?

Ligands as УCompass NeedlesФ How Orientations of Alkene Alkyne and Alkylidene Ligands Reveal -Bonding Features in Tetrahedral Transition Metal Complexes.

код для вставкиСкачать
REVIEWS
Ligands as “Compass Needles” : How Orientations of Alkene, Alkyne, and
Alkylidene Ligands Reveal n-Bonding Features in Tetrahedral
Transition Metal Complexes
Vernon C. Gibson*
This article outlines the strikingly simple
structural pattern to which transition
metal complexes containing x-bonding
ligands conform, and suggests simple
rules (and ligand classifications) by
which the pattern can be understood.
and by which individual structures can
be rationalized or predicted. Alkenes,
alkynes. and alkylidene ligands adopt
distinct orientations with respect to oth-
er ligands attached to a transition metal
because of competition between the various ligand types for metal-ligand n
bonds, whatever the coordination geometry. However, a particularly interesting feature of the tetrahedral geometry is that “single-faced” x ligands can
function as “sensors” of relative ligand
x-donor capacities for all other groups
attached to the metal. Here, using
simple “triad representations’’ of tetrahedral complexes, a simple qualitative
analysis is presented that builds on calculations by Hoffmann et al. and which
allows preferred hgand orientations and
relative ligand x-bond strengths to be
deduced for a wide variety of these transition metal complexes.
,
1. Introduction
Ligands that are capable of forming only a single x interaction with a metal center often adopt distinct orientations with
respect to other ligands attached to a metal center. Provided
there is no steric conflict, such orientation effects are a consequence of the competition that arises between the different ligands for bonding to the metal and are relatively well understood for octahedral complexes where the alignment of the
ligands along the coordinate axes allows a clear distinction between the d orbitals responsible for forming x bonds (the t,, set)
and o bonds (the eg set). The situation is much less clear for
structures where the orbitals are no longer directed along, or
perpendicular to. the metal-ligand axes. A good example is the
tetrahedral geometry whose traditional
representation within a cube (with the s,
y. and :axes emerging from the centers
of the faces) is shown in Figure 1.
Competitive effects are. therefore, not
so readily assessed for tetrahedral species
and it is often necessary to resort to
v
quantum chemical calculations. HowFig. 1. Rcl:ltlonship of‘
ever. complexes with tetrahedral geomeIigiintt positions with
try are particularly attractive for the
rcipcc, ,,, , , rlxeS i n
study of competitive ligand n bonding
ii tetraIicdr;iI complex.
~,
[*I
~
Prol: V. C‘ Gihstrn
Department of C’liemistry. University Science Laboratories.
S o u t h Road. D u r h a m DHl 3LE ( U K )
Teelef:~~.
Int. code + (Y1)386-1127
due to the ability of metal centers with this coordination geometry to accommodate up to four multiply bonded ligands all of
which necessarily engage in competition for the available p and
d metal orbitals capable of n bonding.”]
Here, we shall examine a series of tetrahedral and pseudotetrahedral complexes containing either alkene. alkyne, or
alkylidene ligands and exploit their orientation preferences to
evaluate the relative x-donor capacities of the other attendant
ligands. This task is facilitated by many recent additions to
the family of pseudo-tetrahedral four-coordinate transition
metal complexes, particularly examples containing relatively
hard. strong x donors such as 0x0 and imido ligands, and these
will be highlighted during the subsequent discussion. We shall
also include within the definition of tetrahedral complexes those
containing polyene or polyenyl ligands which are considered to
occupy a single coordination site. Such systems were the basis of
extended Huckel calculations by Hoffmann et al.I2.31 which
first laid the foundations for an understanding of the factors
influencing the preferred orientations of carbene. alkene,
alkyne, and ally1 ligands L with respect to the ancillary Cp
ligands in these pseudo-tetrahedral environments. We shall
not dwell on the details of MO calculations here but rather
call on their qualitative findings to draw together these
seemingly quite different yet structurally related systems, to assess the influence of different n-bonded ligands on the structural patterns observed. and to develop a simple unified approach to predicting ligand orientations and relative rr-bond
strengths in these tetrahedral and pseudo-tetrahedral environments.
V. C. Gibson
REVIEWS
2. A Triad Representation of Tetrahedral and
Pseudo-Tetrahedral Complexes
To assist the analysis, a simple triad representation of a tetrahedral complex will be used (related to Newman projections for
organic systems). If we consider a tetrahedral complex containing the four ligands A,B, C. D, and the metal atom M , and view
the molecule along the A-M axis, then the three ligands B, C.
and D present a triad which is conveniently represented by the
triangle notation shown in Figure 2a. Here, the metal atom lies
Note that the metal orbitals have been shown aligned (for
convenience) with respect to the M - C axis. but it is evident that
the orbitals could also be orientated with respect to either the
M-B or M - D axes. Since the n bonding ligands compete for
the available metal orbitals with dn symmetry in a tetrahedral
environment. the orientation of the available n symmetry orbitals for ligand A will in effect depend on which of ligands B.
C, or D can form the strongest TI bonds.
3. Ligand Types
Before proceeding further, we first need to consider the electronic and orbital characteristics of the types of ligands. Tdbk 1
shows the frontier orbitals of a variety of ligands, classified as o,
x. o r 6 according to their potential for bonding to the metal; the
n symmetry orbitals are assigned h or c subscripts depending on
whether they align with the “horizontal” or “vertical” metal n
orbitals i n the diagrams shown in Figure 2 b.
The ligands are grouped according to the number of orbitals
with n symmetry available for bonding to the metal atom:
A
C
C
b)
B
D
D
Fix. 2 A m a d representation of a tetrahedral complex. a ) View along the A M
axis. b) The metal orbitals with rc symmetry ( K , and nh) which are available for
interaction with the ligand A.
above the plane defined by B, C. and D and directly beneath the
ligand A which lies at the center of the triangle along the normal
to the BCD plane. If the ligand A is capable of n bonding to the
metal center. it could interact with either (or both) of the metal
orbitals with n symmetry (n, or nJ shown in Figure 2 b which
illustrate the MBCD pyramid as viewed from A. Note that the
metal orbitals x,,and n,, are represented as p-type orbitals Iying
parallel to the BCD plane, but they could equally be forwardprojecting lobes of dn orbitals or, as is more likely the case
hybrids of both p and d orbitals. This representation captures
the essence of MO calculations[2,31 in which the authors aligned
the z axis along the A-M vector and where ligand C is Cp and
B and D are carbonyl or nitrosyl ligands.
1. n,-type ligands; these have only one n: symmetry orbital and
therefore can form at most a double bond to the metal atom
consisting of a sigma and pi interaction (lo. I n ) . Such ligands have. on occasion. been referred to as “single-faced” n
ligands.
2. n,- and n2,-type ligands; these possess two n symmetry orbitals and thereby can form two equivalent n bonds potentially giving rise to a triple bond to the metal (lo, 2n).
It is helpful to make a distinction between ligands with
degenerate n levels and those with n orbitals differing in
energy, since for the latter, one of the n orbitals becomes
energetically more accessible to the metal. These latter ligands with non-degenerate n orbitals we shall denote a s n,.
ligands.
Figure 3 shows the principal s and p (or cr and n) ligand
orbitals for each ligand type with accompanying energy level
diagrams to emphasize the similarities and differences within,
and between, categories. Thus, for the n , ligands. the singlet
carbene has a vacant p orbital and so is able to accept n-electron
density from the metal, while the ainido ligand has three electrons distributed between its two available orbitals and so will
Vernon C. Gilnon horn in 1958 it7 Grantham. Englarzd, studied chemistry at the Universitjs of
Sheffield hejbre moving to the University of O.xffbrd kvhere he was avbwded (1 D . Phil, in 1983
, f . r work on the coordination und organornetallic chemistry qfthe earlj. transition metals carried
out in the group of M . L. H . Green. He then spent two y a r s as a NATO post-doctoral
researcher at the Californiu Institute of Technolog?, kvith J. E. Bercaw bqfore returning to
England to take up u lectureship in chemi.vtrj, at Durham University in 1986, where he is
now a ,full Profi.ssor of‘ Chemistry. His interests lie in sjzthetic, structural, and mechanistic
inorganic and organometallic chemistrj,, with a particular emphasis on transition-nietal compounds contaivririg metal- ligund miiltil,lr~bond7 and their applications in polymer synthesis.
1566
. 4 1 i , ~ wClieni
.
1171.Ed EnKi. 1994.
33, 1565-1572
Ligand Orientations and n-Bonding Systems
REVIEWS
Table I . C labsification of ligands by the number and type of their frontier orbitals.
~
Ligiind A
Electrons
available
Frontier orbitals
-
2
0
0
0
a
2
0
a
x+
0
a
-IT
+,
n+
4t-u
4t
CR
NR
(7
+,
IT+
-tt-
(1
OR
0
0
0
0
0
(7
0
(J
3
Y
4
0
5
72
Fig. 3. Frontier orbitals of some representative hgands. The two orbitals indicated
by # :ire the degenerate pair of filled alkyne C - C n-bonding MOa. but are shown
as non-degenerate because of their different orientations with respect to the metal
lipand axis. The MOs of n, ligands are represented by a four-electron alkyne
Iigand.
6%
1J
0
(J
4
%-J
n
nh
n:
6“
act as a net n donor to the metal. Simple energy level diagrams
can be derived for the cyclic El2 Iigands, C,H,>, by standing
the ring on an apex. I t can be seen that the n levels are
degenerate[*’ in each case and therefore directly comparable
with CR, N R , and OR. By contrast, it can be seen that the n
levels for n,. ligands are non-degenerate.
Of particular interest to this study are the El, ligands such as
alkylidene (CHR), amido (NHR). alkene (C,R,), and the “twoelectron” alkyne ligand (C,R,), which have available only one
n orbital for interaction with a metal center and so an orientation preference should be observed depending upon which of the
[*I
It should be noted that a Jahn--Tellerdistortion would remove the degeneracy
for an isolated fragment. Thc view shown in Figure 3 is adopted to assist comparisons between the different ligand types and for the purpme of electron
”book-keeping”.
1567
V. C. Gibson
REVINYS
metal nh or nv orbitals is the most accessible. In the following
sections, we shall examine the effect of either one or two of the
three ligands in the triad dominating the x-donor bonding.
IQI
2ui
4. Complexes Possessing Two Dominant H,-Donor
Ligands
Let us first consider triad representations for complexes containing cyclopentadienyl (Cp) or imido (NR) groups in combination with ancillary alkene ligands (Fig. 4). Figure 4 a shows a
Fig. 5. The frontier wbitals of the bent [Cp2M] metallocene fragment illustrating
their shapes and 01-icntations with respect to the 1, and :axes.
Fig. 6. A triad representation illustrating the orientation or the rr-symmetry orbital
available for bonding to a fourth ligand in a [Cp,ML] fragment. The rr, orbital is
vertical w#ithrespect to the ligand L.
- . ;pc
RN
RN
Fig. 4. Triad representations of complexes containing cyclopentadhenyl and irnido
ligands in combination with alkenc ligands: a) bent metallocene derivatives; b)
half-aandwich imido complexes: c) his(imido) complexes
typical triad representation for a zirconocene alkene complex of
the type [Cp,Zr(H,C=CHR)(PMe,)].141Here. the C=C bond
of the alkene ligand orientates in the plane that contains the
Zr-P bond, the so-called equatorial binding plane or binding
wedge of the metallocene fragment. This is, of course, the wellestablished bent metallocene geometry for which the electronic
constraints are well-documented both through their extensive
derivative chemistry1'] and through detailed quantum chemical
calculations.["]The frontier orbitals of the bent [Cp,M] unit are
shown in Figure 5. which illustrates their shapes and projections
in the J: plane. Thus, if we consider a triad representation for a
[Cp2ML]fragment (Fig. 6), we find that it is only the n, orbital
(with respect to L) that is accessible to a fourth n-bonding ligand. It is also worth noting that, since this ligand approaches i n
the J: plane but herit~rcnthe J' and z axes, the n interaction will
largely consist of a "side-on" interaction with the "d,.2" l t r ,
orbital of the metal atom (Fig. 7).
According to this description the n,, orbital is virtually inaccessible to a fourth n-bonding ligand since it is heavily involved
in n bonding to the Cp rings, and therefore any interaction with
the n,,orbital would necessarily weaken the n bonds between the
C p ligands and the metal center. The orientation preference
1568
would be expected to be lost only if the n,
orbital is filled (i.e. a d' metal center) and
the n, ligand is a n donor such as an
amide.
More recently, a related series of half-
tt
''Y
- =-u
sandwich niobium imido complexes has
been ~ynthesized;['~
these show a similar
Fig, 7, The side.on rr
interaction between a
alignment of the alkene C=C bond, orientated towards the phosphane ligand
when viewed as a triad representation (See
and orbital
Fig. 4 b). Precisely the same "metallocene-like" orientation is found for the
M~,)~[~]
alkene ligand in [ M ~ ( N ~ B U ) ~ ( H ~ C = C H M ~ ) ( P (see
Fig. 4c). At first sight the close relationship apparent between
bent metallocenes, half-sandwich imido. and bis(imido) systems
may not seem at all obvious. However, when the symmetry
properties of the frontier orbitals for the C p and N R units are
taken into consideration. the situation becomes much clearer,
since both of these fragments are seen to bind to a metal using
a combination of lo and 2n interactions (see Table I ) ; the C p
group of course also has two &symmetry acceptor orbitals
available for back-donation but provided these &bonding interactions are weak, they will have minimal impact on the metalligand bonding. The alkene ligand. then. effectively orientates
towards the o-bonded phosphane group so as to avoid direct
competition with the dominant n,-donor groups for metal-ligand x bonds.
An analogous situation is found for the alkylidene derivatives
[Cp2Ta(CHPh)(CH2Ph)][91and [Cp*Nb{N(2.6-iPr2C,H,))(CHPh)(PMe,)]["] (Cp* = C5Me,) (Fig. 8). Here, the p orbital
of the benzylidene ligands is orientated with one lobe pointing
in the direction of the o-bonded group and with the other lobe
bisecting the n,-donor groups (the orientations of the alkylidene substituents are indicated by wedge lines); this again
avoids competition with the C p and N R ligands for n bonds.
~;~~~-5'' ~' et~~~~!
A I I , ? ~Chrln.
~ . hu. Ed. E q l . 1994, 53. 1565-1572
6
REVIEWS
R
N
siv
I
RN
Fig. X. Triad representations showing alkylidene hgand orientations in [Cp,Ta.
(CHPh )(CH,Ph)] and [Cp*N b(NR)(CHPh)(PMe,)].
Thus the alignments of the alkene double bond and the alkylidene substituents differ by 90".
The OR ligand can also act as a n, donor (see Table 1 ) and
therefore. as far as the symmetries of its frontier orbitals are
concerned, it can be regarded as being pseudo-isolobal with the
C p ligand. It should come as little surprise, then, that the two
substituted phenolato complexes, [Ta{O(2,6-tBu,C6H,)},(CHSiMe,)(CHzSiMe,),1['
and
[Ti{O(2,6-Ph,C6H,)j
(C,H,)(PMe,)]['21 shown in Figure 9 also display related orien-
,-
Fig. 10. Examples of complexes possessing a single doininant n,-donor ligand: a)
[M(N(2,6-iPr,C,H,):(CHCMe,)(OtBu),l ( M = Mo. W); b) [CpV(PMe,),(C,H,)I:
c) [Cp*Ta(CHPh)(CH,Ph)J
veals that the alkylidene ligand now bisects the R'O-OR' edge
of the triangle and aligns in the direction of the N R group.
This orientation is consistent with the imido ligand now
dominating the n-donor bonding, a point the alkylidene
ligand neatly emphasizes by pointing towards it in each of the
complexes. A related alkene complex of the type [Mo(NR)(H,C=CH,)(OR'),], not to our knowledge synthesized to date,
would be anticipated to align its alkene
R
n-acceptor orbital with the nh orbital of
the metal atom, resulting in the alkene orientation illustrated in Figure 1 1 .
An analogous orientation effect is
found in the cyclopentadienyl vanadium
RQ
OR'
complex [CpV(PMe,),(C,H,)][' 51 (see
Fig. 11. The anticiFig. l o b ) . Here. since the PMe, lieands
pated o r ~ e n t a c ~ oof
n
cannot function as n donors, the cyd,~'~~~~~c
clopentadienyl ligand dominates the
of the type
.. rMo(NR).
(HL=CHd(OR')il.
bonding, a point the alkene ligand helpfully "underlines" by its orientation. The
dominance of the cyclopentadienyl ligand is also apparent in the
complex [Cp*Ta(CHPh)(CH,Ph),1["
(see Fig. 1Oc) in which
the plane of the alkylidene substituents aligns with the metalring centroid vector, so pointing to the Cp ligand as the doniinant n-bonding ligand. If the R system in the ligand is extended,
for example, in [CpTiCl,{ =N=C(~BU),)],["~ the preferred
orientation of the substituents changes by a 90' twist, as expected in this complex for alignment of the nitrogen p orbital
with the n,, orbital. Similar "twists" are found in complexes
containing cumulene-type ligands such as vinylidene and allenylidene.
The origin of the orientation effect in "tetrahedral" complexes possessing a single dominant n,-donor ligand can be seen by
considering the d orbital splitting diagram in Figure 12: for
convenience and clarity this is derived unconventionally with
the z axis aligned along the n,-ligand-metal axis. Here, instead
of the usual "3 over 3" pattern, the degeneracy of the upper
triplet is split into an e and a , set. There are now two sets of
A
i.3
v
Fig. 9. Triad representations showing alkene and alkylidene ligand orlentations in
substituted bis(phenolat0) complexes.
tation preferences to those seen in the cyclopentadienyl and
imido systems.
5. Complexes Possessing One Dominant II,-Donor
Ligand
Now. if we bear in mind the orientation preferences we have
just witnessed for bis(pheno1ato) complexes in Section 4, we can
then examine the effect of placing a very strong NR donor group
in competition with the two alkoxo ligands through the four-coordinate alkylidene(imido) complexes [M{N(2,6-iPr,C6H,))(CHCMe,)(OtBu),] (M = Mo, W)[I3] and [Mo(NtBu)(CHCMe,){OCH(CF,),j
The triad representation of a
structurally characterized derivative shown in Figure 10 a reAnpiw
C.liuii.
Int. Ed. Eiipl. 1994. 33, 1565 1572
~
_
.
v
1569
~~:~
REVIEWS
V. C. Gibson
doubly degenerate orbitals, the upper set (d,,, d,J allowing n-Interactions with the n donor to be maximized, while the lower P
set (d,,-,,, dxy)aligns orthogonal to the M - n , ligand bond
thereby minimizing interactions with the TI, ligand. Since we are
now viewing the molecule side-on to the M
i (Fig. 12). it
is apparent that d,, effectively correspon
ie nh orbital
viewed from A, while both d,, and d,l can
interactions
with A in the vertical plane (n,). Note that d,, is already engaged
in n bonding with the “axial” n,-ligand and that the d,, orbital
is o bonding with respect to the n2ligand whilst also being
capable of forming a n interaction with the ligand A of the type
illustrated in Figure 7. This provides a simple pictorial illustration of the competition for ligand-metal n bonds in “tetrahedral” complexes.
R
N
R
N
AC1
AC
c1
c1
I
I1
Fig. 13. A triad representation of half-sandwich imido complexes of the type
[CpNb(NR)CII]: a) orientation of the Cp ligand; b) R interactions between the
orbitals of the Cp ligands and the il,and ilhorbitals of the metal atom.
*
r
Fig 13b). Consequently. the filled n-level of the Cp ligand in I1
will remain largely ligand centered and give rise to the allyl-ene
distortion.
6. Complexes Containing 112, Ligands
*
n2
Fig. 12. Splitting diagram for the d orbitals of the metal atom in a tetrahedral
complex possessing a single dominant n,-donor ligand (derived with the :axis
aligned in the direction of the n,-ligand).
An interesting situation arises when the strongly n-donating
N R unit is in competition with a strong polyenyl K donor such
as the cyclopentadienyl ligand. A series of half-sandwich niobium,[”l m ~ l y b d e n u m , “and
~ ~ rhenium[201compounds of general
formula [CpM(NR)CI,] shows a significant ring-slippage along
with an orientation preference for the C,R, ring such that one
of the ring carbon atoms eclipses the M-NR bond (the ring is
orientated so as to “point” to the N R group). This is illustrated
in Figure 13a for [CpNb(NR)CI,] by using the triad notation
viewed down the ring centroid-metal axis.
The ring-slippage has been attributed in general terms to the
strong trans influence of the imido ligand leading to a weakening of the two metal-carbon bonds truns to the N R group and
a consequent adjustment towards an allyl-ene Cp bonding situation.[’8-201However, the ring distortion also follows from our
previous observations for complexes with a “single dominant”
N R n-donor ligand. For the two n-type interactions of the
cyclopentadienyl group, it is expected that the interaction perpendicular to the M-N bond (I in Fig. 13 b), that is the one with
the nh orbital, will be stronger than that with the metal
n-symmetry orbital (TI.,) nearly parallel to the M - N bond (11 in
1570
Orientation effects are also often seen for n,. ligands in competition with strong n2 donors. The strongest n interaction
might again be expected to be with the nh orbital, but the orientation of the n2,ligand is less easily predicted since their frontier
n-symmetry orbitals are non-degenerate (see Fig. 3) and either
of the n interactions could dominate. Just which will dominate
will depend on a number of factors including i) the relative
energies of the n and n* levels of the ligand orbitals relative to
nh and n> orbitals of the metal fragment, and ii) the directional
properties of the ligand n orbitals (with respect to the metal). We
shall not elaborate on these factors here, but the structures of
the acetylene complex [Cp*Ta(PhC=CPh)C1,][2‘1 and the benzyne complex [Cp*Ta(C,H,)Me,][221 should be noted (Fig. 14).
In both complexes the alkyne ligand functions as a four-electron
donor. The orientation of the diphenylacetylene is the same as
that expected for a n, ligand, indicating that the n bonding may
be dominated by alignment of the n*-acceptor orbital with the
nh orbital. However, steric factors have also been suggested to
play a role since the less bulky benzyne ligand is found to align
in a perpendicular orientation. The alignment of the butadiene
ligand in [CpTa(b~tadiene)Cl,][’~~~
(Fig. 14c) is consistent with
the butadiene n-donor orbital interacting with the nh orbital, the
n*-acceptor orbital then aligning with the n, orbital.
A useful illustration of how the n-bonding capacities of n,
and n,. ligands can be compared in a complex is provided by
the example, [CpNb(q4-butadiene)(q2-PhC=CMe)(PMe3)][z3b1
which, in addition to possesing a n, C p ligand, has two different
r12,ligand systems (the butadiene and alkyne groups) attached
to the metal center (Fig. 14d). Without knowing anything about
the detailed M O description of the complex, the orientation of
the alkyne indicates that the n-donor bonding is dominated by
the C p and butadiene ligands, a point reinforced by low-frequency 13CN M R shifts for the acetylenic carbon atoms which
reflect little or no donation of electron density from the alkyne’s
orthogonal n system.
A n g r ~ Chrm.
.
In1 Ed. Engl. 1994, 33, 1565-1572
REVlRNS
Ligand Orientations and x-Bonding Systems
A
co
com
co
C
0
I
[CpCr(NO)(CO)(=CPh,)l
[CpRe(NO)(PPh3)(=CHPh)l'
co
A
CP
PPh?
NO
CP
NO
1v
Ill
Fig. 15. Triad representations of some low-valent complexes containing n I ligands.
Fig. 14. Triad representations for complexes possessing I12. ligands in competition
with 112 donors: a ) [Cp*Ta(PhC=CPh)CI,]; b) [Cp*Ta(C6H,)Me,]; c) [CpTa(C,H,)CI,]: d) [CpNb(C,H,)(PhC=CMe)(PMe,)].
7. Low-Valent Systems
Although the analysis presented so far has largely focused on
high-valent complexes (i.e. those with metal centers in high oxidation states) where n-donor interactions are of overriding significance, the first detailed MO studies of the factors influencing
the orientations of n, ligands were carried out by Hoffmann et
al.[2.31 on low-valent complexes of the type [CpM(CO),L] and
[CpM(CO)(NO)(L)] (where L is a n, ligand such as carbene,
alkene, or alkyne). For the non-symmetrical nitrosyl case, the
orientation preference of the carbene ligand in, for example,
[CpCr(CO)(NO)(CPh2)]rZ41
has been interpreted in terms of the
relative n-acidities of the CO and NO+ groups.
How does this description compare with the treatment we
have adopted for the complexes with n-donor ligands and metal
atoms in high oxidation states described in the previous sections? Qualitatively, we would expect very little difference: the
MO descriptions of the fragments [CpM(CO),] and [CpM(CO)(NO)] also apply to high-valent transition metal systems,
since irrespective of whether the ligands are n donors or n acceptors (e.g. NR, CI, vs. CO etc.) they bind to the metal using
similar combinations of o and n symmetry orbitals (see Table 1).
It is only the relative energies of the orbitals and the localization
of the electrons ultimately in ligand- or metal-centered MOs
that distinguish the situations in high- and low-valent systems.
Thus, in accord with a n-donor approach, CO and NO could
be considered to lie at the weak end of a n-donor scale. It then
An,q,ii.
('him Inr Ed. EngI. 1994, 33. 1565-1572
follows that the symmetrical dicarbonyl complexes [CpMn(CO),(L)J (L = n, ligand, e.g. carbene, alkene, alkyne etc.)
would be expected to show an alignment of their Il ligands with
respect to the single dominant n-donor, the cyclopentadienyl
group. This is indeed the case as exemplified by complexes I[251
and 11[261in Figure 15.
For the mixed carbonyl(nitrosy1) complex [CpCr(CO)(NO)(=CPh2)],1241(111) and the related rhenium phosphane
complex [CpRe(NO)(PPh,)(=CHPh)][PF6][271
(IV) the relative
donor capacities of CO in comparison to those of NO are seen
to be important. If NO is considered as a neutral ligand (rather
than as the nitrosonium ion), it may be regarded as a better
n-donor than CO and the alignment of the p orbital of the
carbene ligand with the weaker n donor ligands CO and PPh,
can be ascribed to the Cp and NO ligands dominating the ndonor bonding resulting in an orientation preference reminiscent of the type seen in bent metallocene derivatives.
Finally, perhaps one of the best illustrations of a ni ligand
acting as a molecular sensor of n-donor capacity can be found
in the complexes [CpMo(O)(CF,C-CCF,)(SC,F,)I and [CpMO(CO)(CF~C=CCF,)(SC,F,)][~~~
(Fig. 16). In the 0x0
derivative, an 18 electron count is achieved with the acetylene
acting as a two-electron (n,)ligand and the acetylene is directed
towards the SC,F, group showing that the cyclopentadienyl and
0x0 ligands dominate the n-donor bonding. However, when the
0x0 ligand is replaced by a carbonyl group, the Cp and SC,F,
ligands now dominate the n-donor bonding, and the acetylene
responds like a compass needle. reorientating towards the CO
ligand. Thus, here the n,-acetylene is clearly acting as an indica-
/4&
CP
0
/&
CP
co
Fig. 16. Triad representations of [CpMo(O)(CF,C-CCF,)(SC,F,)I and [CpMo(C0)(CF,C=CCF,)(SC,F5)] illustrating how the alkyne ligand acts as a sensor of
n-donor capacity.
1571
REVIEWS
V. C. Gibson
Professor Ken Wade is thanked,for many helpful and stinzulattor of relative ligand a-donor strength. Analogous orientation
ing discussions.
effects are also seen in the tungsten complexes [CpW(O)Received: November 16. 1993 [A37IE]
(PhC=CPh)(Ph)]1291and [CpW(0)(PhC-CH)(CH2C0,Et)].1301
German version: Angen. Chem. 1994, /Mi. 1640
where the acetylene aligns with the Ph and CH,CO,Et groups,
W. A. Nugent, J. M. Mayer, Mrtul-Ligund Mulriple Borid~,Wiley-Interscience,
respectively, and even in heterobimetallic systems such as
New York. 1988. Chapter 2
[C~(O)(P~C=CP~)MOR~(CO),C~][~'~
where the molybdeB. E. R. Schilling. R. Hoffmnnn. D. L. Lichtenberger. J. An]. Chrm. Soc. 1979,
num-bound acetylene aligns with the Mo-Ru bond.
lor. 585.
8. Summary
Ligands capable of forming two strong n bonds to a metal
atom dominate the bonding in tetrahedral and pseudo-tetrahedral complexes allowing ligands capable of forming only one a
interaction to be exploited as sensors to identify the dominant
a-bonding groups. These observations lead to some general
rules that (in the absence of overriding steric influences) govern
the preferred orientations of H I ligands in the presence of strong
n2donor groups:
For "tetrahedral" species containing a single dominant n2donor, if a triad notation is employed, the substituents of an
ancillary carbene o r alkylidene ligand will point to the
strongest a donor A, while an alkene or alkyne ligand will
doubly or triply "underline" it (B and C, respectively),
B
A
C
For complexes containing two strong II,-donor ligands, the
alkene and alkyne groups point to the weakest a donor (E
and F, respectively), the alkylidene substituents "underline"
it (or provide it with a bow tie!) (D).
D
E
F
By comparing a series of complexes, it is thereby possible to
derive a qualitative ordering of ligand a-bond strengths. For
example, from the limited range of examples outlined here the
imido group is seen to TI bond more effectively than OR, Cp, and
CI ligands, whereas the Cp ligand a bonds more strongly than
CI and SR ligands. All of these ligands of course dominate the
solely a-bonding groups such as PMe, and alkyl and n-acceptors such as CO. CHR, and alkenes.
We have found it useful to highlight relationships between
complexes containing ligands of a similar class (n,,Tl,, IT,.
etc.). Such relationships should not be regarded as rigid. but
rather as an aid for understanding stoichiometry and structural
patterns in new and seemingly unrelated systems, while at the
same time providing a useful basis for evaluating their likely
rea~tions.[~'1
1512
B. E. R. Schilling. R. Hoffmann, J. W. Faller, J. Ani. Ch(,rii. S U C 1979.
.
101,
592.
H. G . Alt, C. E. Denner, U. Thewalt, M. D. Rausch. J Orgunornet. Chow..
1988, 356. C83: P. Binger. P. Miiller. R. Benn. A . Rufinska. B. Gabor. C.
Kriiger. P. Betz. Chmi. Ber. 1989, 122, 1035.
For leading reSei.ences see: Comprrhm.sirc Orgunoi~irtullwChwni.srr:i.. (Eds: C;.
Wilkinson, F. G . A. Stone. E. W. Abel) Pergamon, New York. 1982; Principle\
und Appht.urion.! n/ Orgunorruiisitioii Mrrul Chetnisrry, (Eds.: J. P. Collman. L
S. Hegedus. J. R . Norton. R. G . Finke). University Science Books. Mill Valley,
C'A. 1987.
1. W! Lauher, R. Hoffmann. J Ani. ( ' h ~ mSoc.
.
1976. 98. 1729: L. Zhu. N M
Kosric. J Or,qmonirr. Chziii. 1987. 335. 395.
A D. Poole. V. C Gibson. W. Clegg. J. C h m . Soc. C h m . Cnmrnuii
1992. 237.
P. W Dyer. V. C. Gibson. J. A. K. Howard, B. Whittle. C. Wilson. J. C h m . Soc.
Cheni. Coiiimuii. 1992, 1666.
R. R. Schrock, L. W. Messerle, C. D Wood. L. J. Guggenberger. J Am. Chcm
Sot.. 1978. 100. 3793.
1. K. Cockcroft. V. C . Gibson, J. A. K. Howard, A. D. Poole. U. Siemeling. C
Wilson, J ('hen?. S i c . Cheni C-oiiiniun. 1992. 1668.
L. Chamberlain. I. P. Rothwell. J. C . Hoffman, 1 Am. C h m . Soc. 1982. i04.
7338.
I. E. Hill, P. E. Fanwick, I . P. Rothwell. Orpnoinrtulhcs. 1992. / / .
1771.
C. J. Schaverien. J C. Dewan. R. R. Schrock. J A m . Chrm. Sol. 1986. 10K.
2771: J. S. Murdzek. R. R. Schrock. Oraunomctul1ic.s 1987. 6. 1373.
[I41 G . Schoettel. J. Kress. J. A . Osborn. J Chrin. So( , Choii. C'oininim. 1989.
1062.
[15] B. Hessen. A. Meetsma, J. H . Teuben. J. Am. Chern. So<.1988, 110, 4860, B
Hessen. A. Meetsma, F. van Bolhuis. J. H . Teuben. G . Helgesson. S. Jagner.
#rgunomera//irs 1990. Y. 1925.
1161 L W Messerle. P. Jennische. R. R Schrock. 0 . Stucky, J Am. Cheiii. So(
1978. IOO. 3793.
1171 I A. Latham, G. J. Leigh. G . Huttner. 1. Jibril, J. C'hem Soc. Duirow Truii.<.
1986, 377.
[18] V. C. Gibson. D. N. Williams. W. Clegg, D . C . R. Hockless, Po/v/i&nn. 1989.
8. 1819: D. N. Williams. J. P. Mitchell. A. D. Poolc. U. Siemeling, W. Clegg.
D. C. R. Hockless, P. A. O'Neil. V. C. Gibson. 1 Chrm. SO(,.Dulron Tra n s ,
1992, 739.
[19] M. L. H. Gieen. P. C. Konidaris, P. Mountford. S. 1. Simpyon. .I Chmi. SJC.
Clrem. Coniinun. 1992. 256.
[?0] W. A. Herrmann. G. Weichselbaumer. R. A. Paciello, R. A. Fischer. E.
Herdtweck. J. Okuda. D. W. Marr. #r,quiioiiirruNirs 1990. 9. 489.
1211 G . Smith. R. R. Schrock, M. R. Churchill, W J. Youngs, I n o r x . Chem. 1981.20.
387.
[22] S.J. McLain. R. R. Schrock. P. R. Sharp, M. R. Churchill, W. J. Youngs. J Am.
C h i . Soc. 1979. fO/. 263.
[23] a ) H. Yasuda, K. Tatsumi. T. Okamoto. K. Mashima, K. Lee. A. Nakamura.
Y. Kai. N. Kanehisa. N. Kasai. J Air?. Chem. Sot.. 1985. 107. 2410; b) G E
Herherich. U. Englert. K. Linn.. P. Roos, J. Runsink, Chem. Bcw.. 1991, 124.
975.
[24] W. A. Herrmann. I. L. Hubbard. I. Bernal. J D. Korp. B L. Haymore. G. L
Hillhouse. Inorx. Chm!. 1984. 23. 2978.
[25] A. D. Redhouse. J Orgunonirr. Chcvii. 1975. 55. C29.
[26] 1. B. Benson, S. A. R Knox, R. D. Stansfield. P Woodward. J Chem. Soc.
Chein. Coinniim. 1977. 404.
1271 W. A Kiel. G-Yu Lin. A G Constable. F. B. McCormick. C. E. Strouse. 0.
Eisenstein. J. A. Gladysr. J A m C h ~ mSo(
.
1982. 104. 4865.
1281 P. S. Braterman. 1. L. Davidson. D. W. A. Sharp, J Chmi. Sot.. Dulron
7kn.s. 1976, 241: J. A. K Howard. R. F. D. Stansfield. P. Woodward. ihid
1976, 246.
[29] N. G . Bokiy. Yu. V. Gatilov. Yu. T. Struchkov. N A. Ustynyuk, J. #rgunoniL,r.
Chmi. 1973. 54. 233.
[30] E. R. Burkhardt. J. J. Done);, R. G. Bergman. C. H. Heathcock. J A m . Cheiti
Soc. 1987, 1 0 9 , 2022.
[31] J. S. Drdge. M. Tilset. P. C. Vollhardt. T. W Weidman. #r~unomcrullics1984.
3. 812.
(321 For a more detailed outline of the arguments presented here see V. C. Gibson.
J Chew. Soc. Dulroii Truns. 1994. 1607.
Angeu
Chem. l i i r . Ed. Engl 1994, 33. 1565-1572
Документ
Категория
Без категории
Просмотров
1
Размер файла
730 Кб
Теги
bonding, features, alkynes, transitional, complexes, reveal, alkenes, ligand, tetrahedral, orientation, alkylidene, уcompass, metali, needlesф
1/--страниц
Пожаловаться на содержимое документа