Electronic Control of the Asymmetry in Heteroatom Monosubstituted Olefin Bonding to the Cyclopentadienyl Iron(II) Dicarbonyl Cation. A Hammett Correlation Study

Article abstract of DOI:10.1021/om0305566

Nucleophilic substitution of CpFe(CO)₂(η²-H ₂CC(H)OEt)⁺PF₆^- with para-substituted anilines was used to prepare a series (4-10) of η ²-vinyl aniline Fp⁺ complexes of the general formula CpFe(CO)₂[η²-CH₂C(H)NH(p-C ₆H₄X)]⁺PF₆^-, where X = OMe (4), Me (5), H, (6), Br (7), COMe (8), CN (9), and NO₂ (10). Correlation of the Hammett σ para parameters with the ¹³C NMR shifts of the metal-coordinated vinyl carbons demonstrated the ability to control the position of the Fp⁺ moiety along the olefin face. As the electron-donating character of the para substituent was increased, the Fp ⁺ moiety was displaced away from the nitrogen-bearing carbon, increasing the asymmetry of the metal-olefin bonding. These conclusions were further supported by determining the X-ray crystal structure of the p-Me (5) and the p-acetyl (8) derivatives.

Full text of DOI:10.1021/om0305566

Organometallics 2003, 22, 5047-5053
5047
Electr on ic Con tr ol of th e Asym m etr y in Heter oa tom
Mon osu bstitu ted Olefin Bon d in g to th e Cyclop en ta d ien yl
Ir on (II) Dica r bon yl Ca tion . A Ha m m ett Cor r ela tion
Stu d y
Stephen A. Matchett,* Brenda R. Schmiege-Boyle,‡ J ohn Cooper,‡
David Fratterelli,‡ Kirk Olson,‡ J ustin Roberts,‡ J ames Thommen,‡
Dean Tigelaar,‡ and Frederique Winkler‡
Department of Chemistry, Grand Valley State University, Allendale, Michigan 49401
Received J uly 28, 2003
-
Nucleophilic substitution of CpFe(CO)2(η2-H2CC(H)OEt)+PF6 with para-substituted
anilines was used to prepare a series (4-10) of η2-vinyl aniline Fp+ complexes of the general
formula CpFe(CO)2[η2-CH2C(H)NH(p-C6H4X)]+PF6-, where X ) OMe (4), Me (5), H, (6), Br
(7), COMe (8), CN (9), and NO2 (10). Correlation of the Hammett σ para parameters with
the 13C NMR shifts of the metal-coordinated vinyl carbons demonstrated the ability to control
the position of the Fp+ moiety along the olefin face. As the electron-donating character of
the para substituent was increased, the Fp+ moiety was displaced away from the nitrogen-
bearing carbon, increasing the asymmetry of the metal-olefin bonding. These conclusions
were further supported by determining the X-ray crystal structure of the p-Me (5) and the
p-acetyl (8) derivatives.
In tr od u ction
distortion of the symmetric structure wherein the metal
fragment slides along the face of the olefin to favor one
side (hereafter referred to as the R carbon) results in a
stabilization and localization of a carbon-centered LUMO
on the opposite (â) carbon. This displacement of the
metal makes this developing LUMO available for over-
lap with an incoming nucleophile. They quantified the
extent of the displacement in their calculations by the
parameter ∆, defined as the horizontal displacement of
the metal away from the center of the olefin face.
Interest in the nature of metal-olefin bonding and
its effect on olefin reactivity finds relevance in organ-
otransition metal mediated organic synthesis as well as
polymer synthesis.1 The Chatt-Dewar-Duncanson
model2 describes the metal-olefin bond as a combina-
tion of σ donation from the olefin to the metal and a π
back-bond from the metal into the π* orbital on the
olefin. Metal-olefin complexes range in structure from
purely symmetric to highly distorted structures that
demonstrate significant reduction in the carbon-carbon
bond order.
In cases where the metal acts predominately as a σ
acid, conventional olefin chemistry can be reversed,
activating the olefin to nucleophilic attack.3 In the early
1980s, calculations on a series of metal ethylene com-
plexes by Hoffmann and Eisenstein4 sought to explain
the source of this activation. They concluded that a
They concluded that as ∆ approached a distance
reflecting a loss of the η2 olefin structure in favor of an
η1 σ interaction of the metal and the R carbon (0.69 Å),
the olefin became activated to nucleophilic attack at the
â carbon.
* To whom correspondence should be addressed. E-mail: matchets@
gvsu.edu.
‡ Contributed as undergraduate researchers at Grand Valley State
University.
Displacement of the metal along the olefin face can
also occur in the ground state structure as a result of
the olefin substituents. Rosenblum and co-workers5
reported a series of iron olefin complexes that explored
the structural consequences of heteroatom substitution
on the coordinated olefin. Comparison of the structures
of 1 ([(η5-C5H5)Fe(CO)2(η2-C2H4)]+PF6-), 2 ([(η5-C5H5)Fe-
(CO)2(η2-CH2CHOCH3)]+PF6- ), and 3 ([(η5-C5H5)Fe(CO)2-
(1) Numerous references can be traced through the following texts.
(a) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; Univerity
Science Books: Mill Valley, CA, 1987; Chapters 11, 17, 19, 20. (b)
Hegedus, L. S. Transition Metals in the Synthesis of Complex Organic
Molecules; University Science Books, Mill Valley, CA, 1994; Chapters
7 and 9.
(2) Dewar, M. J . S. Bull. Soc. Chim. Fr. 1951, 18, C71-C79; Annu.
Rep. Prog. Chem. 1951, 48, 112-135. Chatt, J .; Duncanson, L. A. J .
Chem. Soc. 1953, 2239-2947.
(3) Examples are numerous, including: (a) Green, M. L. H.; Naagy,
P. L. I. J . Organomet. Chem. 1963, 1, 58. And specific to Fp+olefin
complexes: (b) Chang, T. C. T.; Rosenblum, M. Isr. J . Chem. 1984,
24, 99. (c) Lennon, P.; Madhavarao, M.; Rosan, A.; Rosenblum, M. J .
Organomet. Chem. 1976, 108, 93. (d) Marsi, M.; Rosenblum, M. J . Am.
Chem. Soc. 1984, 106, 7264. (e) Chang, T. C. T.; Rosenblum, M.;
Samuels, S. B. J . Am. Chem. Soc. 1980, 102, 5931.
(4) (a) Eisenstein, O.; Hoffmann, R. J . Am. Chem. Soc. 1980, 102,
6148. (b) Eisenstein, O.; Hoffmann, R. J . Am. Chem. Soc. 1981, 103,
4308.
(5) Chang, T. C. T.; Foxman, B. M.; Rosenblum, M.; Stockman, C.
J . Am. Chem. Soc. 1981, 103, 7361.
10.1021/om0305566 CCC: $25.00 © 2003 American Chemical Society
Publication on Web 10/28/2003

5048 Organometallics, Vol. 22, No. 24, 2003
Matchett et al.
(η2-CH2CHNMe2)]+PF6-) revealed an increasing ground
state displacement of the iron toward the R carbon
across the series.
Herein, we report the synthesis and characterization
of a series of seven complexes, analogous to 3, in which
the NMe2 group is replaced by various para-substituted
anilines. The X-ray crystal structure of two of these
complexes is reported to place these compounds in the
context of the known complexes 1-3. Finally, a correla-
tion between Hammett σpara parameters and the 13C
NMR shifts of the carbons of the complexed olefin is
established. This correlation experimentally supports
the proposed influence of the heteroatom lone pair on
metal-olefin bonding. Thus, the position along the
continuum between resonance structures I and II can
be tuned by the functionality of the para substituent,
allowing for systematic control over the metal olefin
structure and potentially over its chemistry.
Rosenblum proposed5 that donation from the hetero-
atom lone pair could be stabilizing the â carbon LUMO
predicted by Hoffmann and Eisenstein,4 yielding the
displacements seen in their series.6 It was reasoned that
the greater displacement seen in 3 was due to better
overlap between the lone pair on nitrogen (vs oxygen)
with the potential LUMO on the â carbon. As such, the
electronic structure of these complexes can be thought
of as points along a continuum between the following
two resonance structures (I and II) where X is a
heteroatom.
Resu lts a n d Discu ssion
Using the known susceptibility of cyclopentadienyl
iron dicarbonyl (hereafter referred to as Fp) vinyl ether
complexes to nucleophilic substitution,10 we prepared
a series of para-substitued aniline analogues of com-
pound 3 according to the reaction shown (where X was
OMe, Me, Br, H, COMe, CN, and NO2). Each derivative
forms readily in CH2Cl2 solution by treatment of the
ethyl vinyl ether compound [CpFe(CO)2(CH2CHOEt)+-
- 11
PF6
]
with a slight molar excess of the para-substi-
tuted aniline as the nucleophile. The products are
orange to orange-brown solids that precipitate from
solution as the reaction proceeds. Reaction time varies
with the nucleophilicity of the aniline derivative, but
even reactions involving the para nitro derivative are
essentially complete within 3-4 h at room temperature.
The position along the continuum should therefore be
a function of the energy match between the heteroatom
lone pair and the developing â carbon-centered LUMO.
The presence of a heteroatom lone pair is not enough
to ensure a displacement of the metal along the olefin
face. If the metal fragment acts as a good π base, these
distortions of the metal-olefin bonding are reduced or
not observed. For example, neither the η2-vinyl ether7
+2
or the η2-2,5-dimethyl pyrrole complex8 of Os(NH3)5
shows any significant distortion from symmetrical olefin
binding.9
The chemical consequences of this displacement are
manifest in the relative reactivity of the complexed
olefin to nucleophilic attack. Rosenblum suggested that
some stabilization of the LUMO by the heteroatom lone
pair greatly increases the rate of nucleophilic attack (2
was reported5 to be 530 times more reactive than 1 to
the same nucleophile). If this stabilization becomes too
great (i.e., the displacement of the Fe is too large), the
reactivity toward nucleophiles passes through a maxi-
mum and then greatly decreases (3 was virtually
unreactive5). This suggested to us that one could control
the relative energy of the heteroatom lone pair through
appropriate substitution and hence tune the chemistry
of the olefin.
Cr ysta llogr a p h ic Da ta on 5 a n d 8. A single-crystal
X-ray diffraction structure was obtained for both com-
plex 5 and complex 8. Crystals of both complexes (dark
red triclinic crystals of 5 and orange-red orthorhombic
crystals of 8) were grown from CH2Cl2 solutions layered
with diethyl ether. The molecular structures of 5 and 8
are shown in Figures 1 and 2, respectively. Table 1
compares selected bond lengths in these complexes with
their analogues in 2 and 3. The solution of the structure
of 5 gave two molecules in the asymmetric unit. These
structures differ in some regions of the molecule due to
hydrogen bonding and crystal-packing forces in the solid
state.12 Bond lengths for each unique structure are given
(6) Participation of the heteroatom lone pair has also been invoked
to explain the distortion in the metal-olefin bonding seen in a Pt vinyl
alcohol complex. Cotton, F. A.; Francis, J . N.; Frenz, B. A.; Tsutsui,
M. J . Am. Chem. Soc. 1973, 95, 2483.
(7) Chen, H.; Harman, W. D. J . Am. Chem. Soc. 1996, 118, 5672.
Although parts of the X-ray crystal structure were reported to be badly
disordered, the geometry around the metal-olefin bonding was clearly
resolved.
(8) Myers, W. H.; Sabat, M.; Harman, W. D. J . Am. Chem. Soc. 1991,
113, 6682.
(9) This is reflected in a reversal in the chemistry with these Os
complexes activating the olefin to electrophilic, not nucleophilic attack.
See ref 7.
(10) (a) Rosenblum, M. Acc. Chem. Res. 1974, 7, 122. (b) Chang, T.
C. T.; Rosenblum, M.; Samuels, S. B. J . Am. Chem. Soc. 1980, 102,
5931. (c) Marsi, M.; Rosenblum, M. J . Am. Chem. Soc. 1984, 106, 7266.
(d) Rosenblum, M.; Bucheister, A.; Chang, T. C. T.; Cohen, M.; Marsi,
M.; Samuels, S. B.: Scheck, D.; Sofen, N.; Watkins, J . C. Pure Appl.
Chem. 1984, 56, 129, and references therein.
(11) Cutler, A.; Raghu, S.; Rosenblum, M. J . Organomet. Chem.
1974, 77, 381.

Electronic Control of Asymmetry in Olefin Bonding
Organometallics, Vol. 22, No. 24, 2003 5049
Ta ble 2. Com p a r ison of Selected Bon d An glesa in
Com p lexes 5 a n d 8
5a
5b
8
Fe-CR-Câ
Câ-N-Cring
92.0(3)°
av ) 94.1(3)°
128.5(5)°
av ) 127.9(5)°
-165.9°
96.1(3)° 87.3(2)°
127.3 (5)° 127.2(2)°
-160.2° -178.3°
dihedral angle between imine
and plane of the aniline ring:
Câ-N-C-C
dihedral angle between
olefin and aniline:
av ) -163.1°
-170.5°
-175.5° 170.1°
CR-Câ-N-Cring
av ) -173.0°
a
The dihedral angles were determined by importing the pdb
files for each structure into MacSpartan Plus (v 1.04) molecular
modeling software and measuring in the molecule editor.
F igu r e 1. ORTEP drawing showing the molecular struc-
ture of 5. The thermal ellipsoids were set at the 30%
probability level. For clarity, only one (5a ) of the two
molecules in the asymmetric unit is shown. ORTEP draw-
ings and full data sets for both molecules are available as
Supporting Information. The PF6- counterion was removed
for further clarity. Selected bond lengths and angles are
given in Tables 1 and 2.
listed in Table 1, which is equal to the Fe-Câ distance
minus the Fe-CR distance. The δ parameter is used in
place of the Hoffmann ∆ because the former is more
easily determined directly from the X-ray crystal struc-
ture data. The value of δ equals zero for a symmetric
olefin and gets larger as the iron atom is displaced
toward the R carbon. Comparing δ values for 2 and 3,
the extent of this displacement is more pronounced for
the amine-substituted olefin, in keeping with the pro-
posed5 influence of the nitrogen heteroatom. Both 5 and
8 show that the Fp+ fragment is strongly displaced
toward the R carbon,13 supporting an enhanced contri-
bution from resonance structure IV in the solid state.
Comparing 5 and 8 to the previously reported struc-
tures of 2 and 3 shows that the extent of iron displace-
ment along the face of the olefin lies between the
methoxy olefin complex 2 and that of the dimethyl
amino olefin complex 3. Of note is the relative placement
of 5 and 8 between the other two. The electron-
withdrawing effect of the para acetyl substituent on 8
gives a δ value that is larger than that of 2, yet shows
a much less distorted structure than seen in 3. Consid-
ering 8, 5, and 3 as a series with increasing electron
density at nitrogen, it appears that this increase acts
to displace the Fp+ fragment toward the R carbon by
stabilizing contributions from resonance structure IV.
Such a displacement is also demonstrated by an opening
of the Fe-CR-Câ angle from 87.3° in 8 to an average
of 94.1° in 5. While these structures demonstrate
significant displacement of the iron atom toward CR,
the observed Fe-CR-Câ angles demonstrate that there
remains a significant interaction with the olefin π
system (particularly in the case of 8). Hence, the R
F igu r e 2. ORTEP drawing showing the molecular struc-
ture of 8. The thermal ellipsoids were set at the 30%
probability level. Selected bond lengths and angles are
given in Tables 1 and 2.
Ta ble 1. Com p a r ison of Selected Bon d Len gth s (Å)
Releva n t to th e Disp la cem en t of Ir on a lon g th e
Olefin F a ce in Com p lexes 2, 3, 5, a n d 8
2
3
5a
5b
8
CR-Câ
olefin bond
Fe-CR
1.42(2) 1.408(8) 1.394(6)
1.405(7) 1.385(4)
2.121(5) 2.164(4)
av ) 1.400(7)
2.09(2) 2.121(5) 2.126(5)
av ) 2.123(5)
2.32(2) 2.819(5) 2.582
av ) 2.624
0.456
av ) 0.501
1.34(2) 1.295(7) 1.327(6)
av ) 1.322(6)
1.418(6)
av ) 1.420(6)
Fe-Câ
2.667
0.546
2.518(3)
0.354
δ value
[(Fe-Câ)-(FeCR)]
heteroatom-Câ
0.23a
0.70
1.317 (6) 1.335(4)
1.422 (6) 1.412(3)
N-ring C
NA
NA
(12) Structure 5a in Table 1 shows a hydrogen bonding distance of
2.215 Å from the N-H to the F of the PF6- counterion. This interaction
is not observed in 5b, where the closest F to the N-H bond was 2.352
Å. This interaction and other crystal-packing forces result in a slight
twisting of the aromatic ring, as well as the slight movement of the
iron fragment. An overlay projection of the two structures is available
in the Supporting Information.
(13) Comparisons given for 5 are based on the average bond lengths
and angles for the two molecules in the asymmetric unit. The values
in Table 1 clearly show that even the least distorted of the two
molecules still is distinct from the structure of 8.
in Table 1, along with their average, which may be more
representative of the solution phase structure. Table 2
provides a similar analysis of selected bond angles again
with the multiple entries for 5.
Structures of 5 and 8 provide a direct illustration of
the influence of the para substituent on the displace-
ment of the iron along the olefin face. To provide a basis
of comparison, we use the Rosenblum δ parameter5

5050 Organometallics, Vol. 22, No. 24, 2003
Matchett et al.
carbon in both 5 and 8 is not a fully developed sp3
center. The analogous angle in 35 has opened to 104.2°,
approaching that expected for a σ complex.
While the structures of both 5 and 8 show strong
contributions from resonance structure IV, their relative
position along the resonance continuum is expressed in
geometry differences throughout the molecule. Given
the trend in the δ parameters, it seems odd that the
olefin bond length in both 5 and 8 should be equal to or
shorter than that in 2. Reported disorders14 in the
structure of 2 (particularly around CR) cast some doubt
on its reported olefin bond length. It is not surprising
that the more electron-rich NMe2 complex, 3, displays
a slightly longer olefin bond. This is supported by the
fact that 3 has a shorter Câ-N bond favoring an
iminium structure. Comparison with structures 5 and
8 demonstrates that the para substituent directly influ-
ences the olefin bond length and the development of
iminium character in the Câ-N bond. Decreasing the
electron donation from the para position upon moving
from 5 to 8 results in a slight shortening of the olefin
bond and a simultaneous lengthening of the Câ-N
interaction. This latter point supports the idea that the
para substituent is directly influencing the interaction
between the nitrogen lone pair and the â carbon of the
olefin.
F igu r e 3. Side view of the structure of 8 showing the
planar nature of the organic portion of the molecule. The
PF6- counterion and the hydrogens have been removed for
clarity.
Ta ble 3. 13C Sh ifts (p p m in CD3NO2) of th e Olefin
Ca r bon s for Com p lexes 2-10, Rela tin g Olefin
Su bstitu tion to Disp la cem en t of th e F p + a cr oss th e
Olefin
As the interaction between the nitrogen and the â
carbon increases, one expects increasing planarity
around nitrogen. The structures of 5 and 8 are strongly
planar in this region. The bond angle (see Table 2)
around nitrogen in 8 (C2-N-C21) is 127.2(2)° and has
an average value of 127.9(5)° for the equivalent region
of 5, both revealing significant sp2 character at nitrogen.
In the structure of 8, the dihedral angle between the
olefin and nitrogen to aromatic ring bond (labeled C1-
C2-N-C21 in the atom-numbering scheme of 8) is
-170.1°, while the average dihedral angle in the equiva-
lent region of 5 is -173.0°. Worthy of note is the overall
molecular planarity of 8 seen in Figure 3 (the hydrogen
atoms have been omitted for greater clarity). This is
reflected by the dihedral angle of -178.3° between the
imine and the plane of the ring. In 5, the average
equivalent angle is -163.1°, showing a slight rotation
of the bond between nitrogen and the aromatic ring. It
would appear that the influence of the para acetyl group
provides 8 with more extensive conjugation across the
bulk of the molecule.
complex (substituent)
CR
Câ
δ
3a (NMe2)
0.10
8.94
9.88
10.89
12.36
14.70
16.34
17.93
27.30
177.6
163.8
161.9
160.0
156.5
151.2
147.4
146.3
146.2
0.70 Å
4 (para OMe aniline)
5 (para Me aniline)
6 (para H aniline)
7 (para Br aniline)
8 (para COMe aniline)
9 (para CN aniline)
10 (para NO2 aniline)
2a (OMe)
0.50 Å
0.35 Å
0.23 Å
a
See ref 16.
remarkable feature. At room temperature, the olefinic
proton resonances all show evidence of rotation about
the olefin carbon-carbon bond. In place of the expected
ABX system (cis, trans coupling across the double bond),
the olefin protons form an A2X spin system, giving a
high-field doublet for the R protons and a low-field
doublet of triplets (doublet splitting from coupling to
adjacent N-H) for the â protons. The equivalency of the
R protons is the result of carbon-carbon bond rotation
facilitated by the rearrangement toward resonance
structure IV. A facilitated bond rotation of this type was
previously reported for 2.5 A description of the dynamic
NMR properties of these complexes as well as a thor-
ough measurement of the ∆H of activation for olefin
bond rotation in several of these complexes is the subject
of a subsequent publication.15
The structures of 5 and 8 argue that increasing the
electron density on nitrogen allows for better overlap
between the nitrogen lone pair and the â carbon and is
directly related to the displacement of the Fp+ fragment
along the olefin face. This supports our contention that
the position of the iron along the olefin face (and
potentially the olefin reactivity) can be tuned by the
controlling the relative energy of the heteroatom lone
pair.
Carbon-13 NMR shifts for the olefinic carbons of this
series are clearly sensitive to the electronic properties
of the para substituent (see Table 3). Clearly the
crystallography of complexes 3, 5, and 8 shows that
nitrogen-substituted olefin complexes of Fp+ are not
symmetrically bound to the iron center. The sym-
1
NMR An a lysis a n d 13C Ha m m ett Stu d y. The H
NMR data for the complexes reported here all show one
(14) The crystal structure for complex 2 is disordered with respect
to a mirror plane relating the R carbon to a carbonyl carbon. The
authors report uncertainty in the Fe-CR distance (see ref 5), which
no doubt effects the reported C-C bond length. The authors note that
this uncertainty would act to set a lower limit on the reported δ value
of 0.12 Å, which would only strengthen the arguments presented above
in the discussion of displacement of the Fp+ fragment across the olefin
face in response to electron density.
metrically bound ethylene in CpFe(CO)2(C2H4)+PF6
-
shows one olefin carbon resonance at 56.7 ppm.16 The
presence of the heteroatom and presumedly the inter-
action with its lone pair cause the two carbons to
(15) Matchett, S. A.; Zhang, G. R. Manuscript in preparation.

Electronic Control of Asymmetry in Olefin Bonding
Organometallics, Vol. 22, No. 24, 2003 5051
F igu r e 5. Correlation of the 13C chemical shift of the R
carbon vs Hammett σ para parameters for complexes 4-10.
F igu r e 4. Correlation plot for 13C chemical shift of olefin
â carbon vs Hammett σ-para parameters for complexes
4-10.
correlation between NMR chemical shifts and Hammett
σ values.18a The value of F equals -16.72, corresponding
to an increase in the chemical shift of the â carbon as
the electron-releasing ability of the para group in-
creases. As the electron-donating character of the para
substituent increases, we feel there is greater overlap
of the nitrogen lone pair with the â carbon LUMO,
favoring displacement of the iron fragment toward the
R carbon. This results in an increased contribution from
resonance structure IV. When the para substituent is
electron withdrawing, the overlap decreases and the
structure reflects an increased contribution from reso-
nance structure III.
experience very different environments. While one could
dismiss the changes in the 13C shift of the â carbon from
56.7 ppm to be purely due to the inductive influence of
an electronegative heteroatom, such a rationalization
would not explain why 3 (N substituted) shows a larger
downfield shift for the â carbon than does 2 (O substi-
tuted). We contend that it is the availability of the
heteroatom lone pair for overlap with the â carbon that
determines the chemical shift of the olefin carbons. The
olefin 13C chemical shifts for 4-10 fall between those
seen for 2 and 3. More significantly, changes in the para
substituent on the ring directly influence the chemical
shift of both carbons of the olefin. Included for compari-
son in Table 3 is the δ value from the known crystal
structures as a measure of the displacement of the Fp+
moiety along the olefin face. We feel that the electronic
influence of the para substituent is changing the relative
energy of the nitrogen atom lone pair and hence its
ability to stabilize the shift of the iron toward the R
carbon. As overlap between the heteroatom lone pair
and the olefin fragment increases (increasing electron
donation from the para substituent), the R carbon is
shifted upfield while the â carbon resonance moves
downfield. This offers additional experimental support
for Rosenblum’s argument originally based on the
structures of 2 and 3. It should be noted that the para
nitro derivative, 10, shows 13C shifts for the olefin
carbons that are approaching those of the OMe deriva-
tive, 2.
An obvious choice in a study of this type is the para
NH2 derivative made from phenylene diamine. While
this derivative forms readily, the 13C shift for the â
carbon of 165.5 ppm lies well below the correlation line
given in Figure 4. The 1H NMR for this derivative varies
with the concentration of the solution. At increasing
concentrations, the spectrum shows no downfield dou-
blet for the N-H peak on the olefin. Instead, a broad
upfield resonance of variable chemical shift integrates
to approximately three protons. We attribute this to an
acidity of the olefin-bound N-H caused by a significant
contribution from resonance structure IV in the pres-
ence of the strong para electron-donating NH2 group.
In addition to favoring resonance structure IV, the
presence of the para NH2 group provides a convenient
proton acceptor site. The proton transfer is not complete
+
since the correlation with the σpara for an NH3 group
(σpara ) 0.60) still lies well off the correlation line. The
13C shift for the â carbon of this derivative is less than
that expected for a para NH2 group and much larger
The usefulness of Hammett17 σ constants to explore
the correlation between electron structure and chemical
structure/reactivity is well established. Specifically,
correlation of σ parameters to NMR chemical shifts is
also precedented.18 When the Hammett σpara coefficient
for each of the para substituents is plotted against
the13C chemical shifts of the â carbon, the plot (Figure
4) is seen to be linear with a correlation coefficient of
0.992 for seven data points. Correlation coefficients
greater than 0.95 are considered to show a significant
+
than expected for a para NH3 group. The observed
resonance in the 1H NMR represents the time-averaged
exchange between the acid and base sites. The collapse
of the downfield doublet for the olefinic N-H with
increasing concentration may imply a bimolecular pro-
cess, although a kinetic analysis has not yet been
performed.19
A similar plot (Figure 5) can be constructed comparing
the Hammett σpara coefficients to the 13C R carbon shifts.
A linear correlation to the electronic nature of the para
substituent is again observed; however, the fit is not
quite as good (R2 ) 0.988 for 7 data points). The positive
value of F (+8.19) shows a decrease in chemical shift as
(16) Substituted Olefin Complexes of Cyclopentadienyldicarbonyl-
rion: Structure and Spectroscopic Properites: Stockman, C. T. Ph.D.
Thesis, Brandeis University, Waltham, MA, 1986, and the data therein.
(17) Hammett, L. P. J . Am. Chem. Soc. 1937, 59, 96.
(18) (a) Ewing, D. F. Correlation of NMR Chemical Shifts with
Hammettt σ Values and Analogous Parameters. In Correlation Analy-
sis in Chemistry: Recent Advances; Chapman, N. B., Shorter, J ., Eds.;
Plenum Press: New York, 1978. (b) Malet, R.; Marcial, M. M.; Parella,
T.; Pleixats, R. Organometallics 1995, 14, 2463.
(19) Work to determine the ∆H of activation for proton transfer
using peak width at half-height techniques is ongoing in our lab.

5052 Organometallics, Vol. 22, No. 24, 2003
Matchett et al.
temperature for 2-4 h. The length of time was increased as
the nucleophilicity of the aniline species decreased. As the
reaction proceeded, the ionic product began to precipitate from
solution as a dark red-orange powder. The solvent from the
reaction mixture was then reduced under vacuum, and diethyl
ether was added to complete precipitation. The residue was
then recrystallized from dichloromethane and diethyl ether.
The resulting powder was isolated by filtration and dried
under vacuum to give respectively 4 (43.0% isolated yield), 5
(70.3% isolated yield), 6 (62.1% isolated yield), 7 (50.2%
isolated yield), 8 (59.7% isolated yield), 9 (50.1% isolated yield),
and 10 (67.1% isolated yield).
the electron-releasing properties of the para substituent
increase. This is consistent with the proposed displace-
ment of the Fp+ fragment toward the R carbon and the
developing sp3 character there.
Attempts to correlate 13C shifts for the R, â, or
carbonyl carbons with either σ+ or σ- values failed to
produce significant linear agreement. Failure of the σ+
value, despite the positive charge in the molecule is
reasonable given that the dominance of resonance
structure IV should place the majority of the plus charge
on the nitrogen atom.
Ch a r a cter iza tion Da ta for p-OMe Der iva tive 4. 1H
NMR (acetone-d6, 300.53 MHz): 2.49 (d, 2 H, J ) 8.5 Hz, olefin
CH2), 5.30 (s, 5 H, Cp), 7.06 (d, 2 H, J ) 7.7 Hz, aromatic),
This study demonstrates that changes in the electron-
donating/-withdrawing properties of the para substitu-
ent can be used to control the displacement of the Fp+
fragment along the olefin face. Since it has been
theorized that this displacement is directly related to
the reactivity toward nucleophilic substitution in these
complexes, such control should also extend to reactivity
in the series. It was previously demonstrated that small
ground state displacements strongly favored substitu-
tion, while large displacements effectively prevented the
reaction. Since reactivity cannot be thought of in terms
of an on/off switch, there must exist a transition between
these extremes as the Fp+ is displaced along the olefin
face. Work in our group is underway to assess the
relative reactivity of each of these complexes with a
common nucleophile in an effort to correlate this
reactivity to the 13C shift of the â carbon of the
coordinated olefin. While many factors determine the
final 13C shift in the NMR, such a correlation could
develop into a useful tool for predicting relative suscep-
tibility to nucleophilic attack in metal olefin complexes.
7.38 (d, 2 H, J ) 7.7 Hz, aromatic), 8.43 (d of t, 1 H, J H-NH
)
14.6 Hz J ) 9.1 Hz, olefin CHN), 9.30 (br d, 1H, J ) 12.0 Hz,
NH). 13C{1H} (nitromethane-d3, 75.58 MHz): δ 8.9 (olefin CH2),
87.4 (CHCp), 116.6, 121.5 (CH aromatic), 131.9, 160.6 (C
aromatic), 163.8 (olefin CHN), 214.3 (CO) ppm. IR (KBr): 2002,
2042 cm-1. Anal. Calcd for C16H16NO3F6FeP: C, 40.791; H,
3.423; N, 2.973. Found: C, 40.76; H, 3.35; N, 3.04.
Ch a r a cter iza tion Da ta for p-Me Der iva tive 5. 1H NMR
(acetone-d6, 300.53 MHz): δ 2.54 (d, 2 H, J ) 9.1 Hz, olefin
CH2), 5.32 (s, 5 H, Cp), 7.34 (s, 4 H, aromatic), 8.48 (d of t, 1
H, J H-NH ) 15.0 Hz J ) 8.6 Hz, olefin CHN), 9.28 (br d, 1H,
J ) 13.7 Hz, NH). 13C{1H} (nitromethane-d3, 75.58 MHz): δ
9.9 (olefin CH2), 87.4 (CHCp), 119.7, 132.0, (CH aromatic),
136.4, 139.5 (C aromatic), 161.9 (olefin CHN), 214.1 (CO) ppm.
IR (KBr): 1992, 2048 cm-1
.
1
Ch a r a cter iza tion Da ta for p-H Der iva tive 6. H NMR
(acetone-d6, 300.53 MHz): δ 2.57 (d, 2 H, J ) 9.1 Hz, olefin
CH2), 5.33 (s, 5 H, Cp), 7.33-7.60 (mult., 4 H, aromatic), 8.49
(d of t, 1 H, J H-NH ) 15.1 Hz J ) 8.9 Hz, olefin CHN), 9.13 (br
d, 1H, J ) 14.0 Hz, NH). 13C{1H} (nitromethane-d3, 75.58
MHz): δ 10.9 (olefin CH2), 87.5 (CHCp), 119.7, 131.6 (CH
aromatic), 128.7, 138.9 (C aromatic), 160.0 (olefin CHN), 213.9
Exp er im en ta l Section
(CO) ppm. IR (KBr): 2001, 2039 cm-1. Anal. Calcd for C15H14
-
Gen er a l P r oced u r es. All reactions were carried out under
N2 using standard Schlenk line techniques. Solvents were
distilled under a N2 atmosphere off of appropriate drying and/
or deoxygenating reagents (CH2Cl2, CaH2; Et2O, Na/benzophe-
none). CD3NO2 was dried on P2O5 and distilled by vacuum
transfer prior to use. CD3NO2 NMR tubes were prepared in
an inert atmosphere glovebox. CD3COCD3 was dried on 3 Å
molecular sieves and vacuum transferred directly into the
NMR tube. NMR spectra were recorded on a J EOL Eclipse
NO2F6FeP: C, 40.85; H, 3.20; N, 3.17. Found: C, 40.81; H,
3.20; N, 3.18.
1
Ch a r a cter iza tion Da ta for p-Br Der iva tive 7. H NMR
(acetone-d6, 300 MHz): δ 2.69 (d, J ) 9.1 Hz, olefin CH2), 5.55
(s, 5 H, Cp), 7.51 (d, 2 H, J ) 9.0 Hz, aromatic), 7.69 (d, 2 H,
J ) 9.0 Hz, aromatic), 8.87 (d of t, 1 H, J H-NH ) 14.3 Hz J )
9.1 Hz, olefin CHN), 10.63 (br d, 1H, J ) 11.3 Hz, NH). 13C-
{1H} (nitromethane-d3, 75.58 MHz): δ 12.4 (olefin CH2), 87.6
(CHCp), 121.3, 131.6, (CH aromatic), 121.0, 134.4 (C aromatic),
156.5 (olefin CHN), 213.9 (CO) ppm. IR (KBr): 2002, 2052
cm-1. Anal. Calcd for C15H13NO2FeBrPF6: C, 34.65; H, 2.52;
N, 2.69. Found: C, 34.32; H, 2.45; N, 2.65.
1
300 at 300.52 MHz for H and 75.57 MHz for 13C. All chemical
shifts were referenced to the residual protons in the deuterated
solvents. Vinylic carbon 13C shift assignments used in the
Hammett plots were confirmed using an HMQC NMR experi-
ment. Values for the various σ parameters were obtained from
Ch a r a cter iza tion Da ta for p-COMe Der iva tive 8. 1H
NMR (acetone-d6, 300 MHz): δ 2.78 (br d, J ) 8.26 Hz, olefin
CH2), 5.59 (s, 5 H, Cp), 7.63 (d, 2 H, J ) 8.8 Hz, aromatic),
the text Advanced Organic Chemistry.20 Samples of the ethyl
-
vinyl ether starting material, CpFe(CO)2(CH2CHOEt)+PF6
,
were prepared by reported literature methods.7 All para-
substituted anilines were sublimed prior to use. Elemental
analyses were performed by Galbraith Laboratories, Knoxville,
TN. As cations, these Fp+ salts tend to retain solvent (par-
ticularly diethyl ether), making elemental analysis more
difficult. As complexes 4-10 represent an analogous series,
successful analyses for 4, 6, 7, and 10 were taken as confirma-
tion for the series.
8.11 (d, 2 H, J ) 8.8 Hz, aromatic), 8.91 (d of t, 1 H, J H-NH
)
14.3 Hz, J ) 9.2 Hz, olefin CHN), 10.51 (br d, 1H, J ) 13.8
Hz, NH). 13C{1H} (nitromethane-d3, 75.58 MHz): δ 14.7 (olefin
CH2), 26.8 (CH3), 87.7 (CHCp), 118.9, 131.8 (CH aromatic),
136.3, 142.8 (C aromatic), 151.2 (olefin CHN), 198.8 (CO of
acetyl), 213.1 (CO) ppm. IR (KBr): 2000, 2043 cm-1
.
Ch a r a cter iza tion Da ta for p-CN Der iva tive 9. 1H NMR
(acetone-d6, 300.53 MHz): δ 2.76 (d, 2 H, J ) 9.4 Hz, olefin
CH2), 5.42 (s, 5 H, Cp), 7.52 (d, 2 H, J ) 8.8 Hz, aromatic),
Typ ica l P r oced u r e for t h e P r ep a r a t ion of p a r a -
+
Su bstitu ted An ilin e Com p lexes of th e Cp F e(CO)2 Ca t-
ion . To a solution of CpFe(CO)2(CH2CHOEt)+PF6 (0.200 g,
-
7.84 (d, 2 H, J ) 8.5 Hz, aromatic), 8.38 (d of t, 1 H, J H-NH
)
14.3 Hz, J ) 9.4 Hz, olefin CHN), 8.71 (br d, 1H, J ) 14.0 Hz,
NH). 13C{1H} (nitromethane-d3, 75.58 MHz): δ 16.3 (olefin
CH2), 87.9 (CHCp), 119.4, 135.6, (CH aromatic), 110.5, 143.0
(C aromatic), 147.4 (olefin CHN) ppm. CN carbon not located.21
Fe-CO peak(s) not located due to dynamic exchange.22 IR
0.508 mmol) in CH2Cl2 (10 mL) was added a 10 mol % excess
of the para-substituted aniline species as a solution in 4 mL
of CH2Cl2. The resulting orange solution was stirred at room
(20) March, J . Advanced Organic Chemistry; Wiley and Sons: New
York, 1985; p 244.
(KBr): 2009, 2052, 2234 cm-1
.

Electronic Control of Asymmetry in Olefin Bonding
Organometallics, Vol. 22, No. 24, 2003 5053
Ta ble 4. Exp er im en ta l Da ta for th e Cr ysta l
Str u ctu r es of 5 a n d 8
in each group, and the exposure time was 5 s per frame. Among
them, three groups of images were collected at 2θ ) -40°, and
the other three groups were at 2θ ) -80°, which makes the
2θmax ) 133.07°. The data reduction program SAINT+ version
6.22 (Bruker 2001) determined the Laue group was 2/m, and
a total of 4124 unique reflections were integrated for structure
solution and refinements. The space group was P21/m. The
structure was solved by direct methods using SHELXS version
6.12 (Bruker, 2001). The trial solution obtained 54 non-
hydrogen atoms in the asymmetrical unit. Least-squares
refinement included all non-hydrogen atomic coordinates and
anisotropic thermal parameters. With two molecules in the
asymmetric unit, the final refinement cycle gave, with 3045
reflections with intensities > 2σ(I): R ) 0.0505, Rw2 ) 0.1245;
with all 4121 reflections: R ) 0.0710. The hydrogen atoms
were placed in calculated positions. Key crystallographic
constants and parameters are given in Table 4.
5
8
empirical formula
fw
C16H16F6FeNO2P C17H16F6FeNO3P
455.12
483.13
temperature
wavelength
cryst syst
space group
unit cell dimens
120 K
1.54178 Å
triclinic
P21/n
a ) 17.9949(6),
R ) 90°
233 K
0.71073 Å
orthorhrombic
Pna21
a ) 15.095(3),
R ) 90°
b ) 10.5648(3),
â ) 95.033(2)°
c ) 19.1257(7),
γ ) 90°
b ) 7.2282(17),
â ) 90°
c ) 17.975(4),
γ ) 90°
volume (Å3)
Z
density(calc)
abs coeff
3622.0(2)
8
1961.2(8)
4
1.669 Mg/m3
8.194 mm-1
1840
1.636 Mg/m3
0.924 mm-1
976
Cr ysta llogr a p h y on 8.24 X-ray diffraction data were col-
lected on a Bruker P4 diffractometer equipped with a SMART
CCD detector, and crystal data, data collection, and refinement
parameters are summarized in Table 4. A total of 4332 unique
reflections were integrated for structure solution and refine-
ment. The structures were solved using direct methods and
standard difference map techniques and were refined by full-
matrix least-squares procedures on F2 with SHELXTL (Version
6.10). The final refinement cycle, with 4332 reflections with
intensities >2σ, gave R ) 0.0311, Rw2 ) 0.0637 and R ) 0.0465
with all reflections. The hydrogen atoms were placed in
calculated positions. Key crystallographic constants and pa-
rameters are given in Table 4.
F(000)
θ range for collection
no. of reflns collected
no. of indep reflns
data, I > 2σ(I)
3.23 to 59.21°
12071
4124
2.27 to 28.24°
12898
4332
3045
4332
no. of params/restraints 490/0
4332/343
0.0465, 0.0677
0.0311, 0.0637
1.061
R1, wR2 (all data)
R1, wR2 [I > 2σ(I)]
GOF on F2
0.0710, 0.1395
0.0505, 0.1245
1.039
Ch a r a cter iza tion Da ta for p-NO2 Der iva tive 10. 1H
NMR (acetone-d6, 300 MHz): δ 2.83 (d, 2 H, J ) 6.9 Hz, olefin
CH2), 5.44 (s, 5 H, Cp), 7.55 (d, 2 H, J ) 6.3 Hz, aromatic)
8.34, (d with broad base, 3H, J ) 6.3 Hz, aromatic with 1 olefin
CH overlapping base), 8.62 (br d, 1H, J ) 10.2 Hz, NH). 13C-
{1H} (nitromethane-d3, 75.58 MHz): δ 17.9 (olefin CH2), 88.0
(CHCp), 118.9, 143.8, (CH aromatic), 127.2, 143.8 (C aromatic),
146.3 (olefin CHN) ppm. Fe-CO peak(s) not located due to
dynamic exchange.19 IR (KBr): 2015, 2060 cm-1. Anal. Calcd
for C15H14N2O4F6FeP: C, 37.06; H, 2.70; N, 5.76. Found: C,
37.11; H, 2.67; N, 5.76.
Cr ysta llogr a p h y on 5.23 The data were collected on a
Bruker SMART 6K CCD X-ray area detector with a window
diameter of 13.5 cm. It was controlled by a Windows 2000
based PC computer with SMART version 5.625 software
(Bruker, 2001), at low temperature (120 K), with graphite-
monochromatized Cu KR radiation [λ(Cu KR) ) 1.54184 Å].
All reflections were measured in image groups with 606 frames
Ack n ow led gm en t. Thanks are expressed to the
Michigan Space Grant Consortium, the Council on
Undergraduate Research, and the Grand Valley State
University Summer Undergraduate Research Program
for their financial support of this work. Special thanks
to the laboratories of Dr. Gerald Parkin (Columbia
University, New York) for performing the X-ray crystal
structure determination of 8 and to the laboratories of
Pfizer Corporation (Kalamazoo, MI) and in particular
Garold Bryant, J r. for performing the X-ray crystal
structure determination of 5.
Su p p or tin g In for m a tion Ava ila ble: Listings of the
atomic coordinates, bond lengths, bond angles, anisotropic
displacement parameters, hydrogen coordinates, and aniso-
tropic displacements for the crystal structures of 5 and 8 are
available. In addition, an overlay projection of the two mol-
ecules in the asymmetric unit of 5 is available. This material
can be obtained free of charge via the Internet at http://
pubs.acs.org.
(21) 13C run with up to a 6 s relaxation delay. Peak expected at about
118 or 119 ppm based on aromatic nitriles and may lay beneath the
peak at 119.4 ppm.
(22) Work on the dynamic behavior of these molecules by our lab is
demonstrating that as the electron-withdrawing properties of the para
substituent increase, the dynamic process that equilibrates the two
metal carbonyls slows down. At room temperature, these peaks for
complexes 9 and 10 appear to have coalesced. Manuscript in prepara-
tion.
OM0305566
(23) Structure performed as a courtesy of Garold L. Byrant at the
X-ray structure facilities of Pfizer in Kalamazoo, MI. See acknowledg-
ments.
(24) The structure was performed as a courtesy by the research
group of Gerald Parkin, Department of Chemistry, Columbia Univer-
sity, New York. See acknowledgments.