Determination of structural parameters for ferrocenecarboxaldehyde using Fourier transform microwave spectroscopy

Article abstract of DOI:10.1063/1.1993593

Gas-phase structural parameters for ferrocenecarboxaldehyde have been determined using Fourier transform microwave spectroscopy. Rotational transitions due to a -, b -, and c -type dipole moments were measured. Eighteen rotational constants were determined by fitting the measured transitions of various isotopomers using a rigid rotor Hamiltonian with centrifugal distortion constants. Least-squares fit and Kraitchman analyses have been used to determine the gas-phase structural parameters and the atomic coordinates of the molecule using the rotational constants for various isotopomers. Structural parameters determined from the least-squares fit are the Fe-C bond lengths to the cyclopentadienyl rings, r (Fe-C) =2.047 (4) A, and the distance between the carbon atoms of the cyclopentadienyl rings, r (C-C) =1.430 (2) A and r (C1 - C 1′) =1.46 (1) A of ring carbon and aldehyde carbon atom. Structural parameters were also obtained using density-functional theory calculations, and these were quite helpful in resolving ambiguities in the structural fit analysis, and providing some fixed parameters for the structural analysis. The results of the least squares and the calculations indicate that the carbon atoms of the Cp groups for ferrocenecarboxaldehyde are in an eclipsed conformation in the ground vibrational state.

Full text of DOI:10.1063/1.1993593

Determination of structural parameters for ferrocenecarboxaldehyde using
Fourier transform microwave spectroscopy
Ranga Subramanian, Chandana Karunatilaka, Riley O. Schock, Brian J. Drouin, Paul A. Cassak et al.
Citation: J. Chem. Phys. 123, 054317 (2005); doi: 10.1063/1.1993593
View online: http://dx.doi.org/10.1063/1.1993593
View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v123/i5
Published by the American Institute of Physics.
Additional information on J. Chem. Phys.
Journal Homepage: http://jcp.aip.org/
Journal Information: http://jcp.aip.org/about/about_the_journal
Top downloads: http://jcp.aip.org/features/most_downloaded
Information for Authors: http://jcp.aip.org/authors
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

THE JOURNAL OF CHEMICAL PHYSICS 123, 054317 ͑2005͒
Determination of structural parameters for ferrocenecarboxaldehyde
using Fourier transform microwave spectroscopy
Ranga Subramanian, Chandana Karunatilaka, Riley O. Schock, Brian J. Drouin,a͒
Paul A. Cassak,b͒ and Stephen G. Kukolichc͒
Department of Chemistry, University of Arizona, Tucson, Arizona 85721
͑Received 23 March 2005; accepted 13 June 2005; published online 9 August 2005͒
Gas-phase structural parameters for ferrocenecarboxaldehyde have been determined using Fourier
transform microwave spectroscopy. Rotational transitions due to a-, b-, and c-type dipole moments
were measured. Eighteen rotational constants were determined by fitting the measured transitions of
various isotopomers using a rigid rotor Hamiltonian with centrifugal distortion constants.
Least-squares fit and Kraitchman analyses have been used to determine the gas-phase structural
parameters and the atomic coordinates of the molecule using the rotational constants for various
isotopomers. Structural parameters determined from the least-squares fit are the Fe–C bond lengths
to the cyclopentadienyl rings, r͑Fe–C͒=2.047͑4͒ Å, and the distance between the carbon atoms of
the cyclopentadienyl rings, r͑C–C͒=1.430͑2͒ Å and r͑C1–C1 ͒=1.46͑1͒ Å of ring carbon and
Ј
aldehyde carbon atom. Structural parameters were also obtained using density-functional theory
calculations, and these were quite helpful in resolving ambiguities in the structural fit analysis, and
providing some fixed parameters for the structural analysis. The results of the least squares and the
calculations indicate that the carbon atoms of the Cp groups for ferrocenecarboxaldehyde are in an
eclipsed conformation in the ground vibrational state. © 2005 American Institute of Physics.
͓DOI: 10.1063/1.1993593͔
I. INTRODUCTION
correlated the effects of substituents to the redox behavior of
the ferrocene moiety by changing the energy level of the
highest occupied molecular orbital ͑HOMO͒. In the case of
FcCHO, the electron-withdrawing group ͑such as –CHO
group͒ lowers the energy level of the HOMO, making it
easier to reduce.13 To have a better understanding of struc-
tural, chemical, and electronic properties, we believe that it
is important to have accurate and detailed structural data.
The x-ray diffraction studies provide structural parameters;
however, the crystal-packing effects present in the solid
phase distort the “true structure” of the molecule. Since these
effects are absent in the gas phase, the structural determina-
tion of this molecule using Fourier transform gas-phase mi-
crowave spectroscopy was undertaken. We have previously
been successful in determining the structures of various sub-
stituted ferrocene compounds.14–18 More recently, we have
correlated the effect of substituents with a structural param-
eter, namely, the distance between the Fe atom and center of
the cyclopentadienyl ͑Cp͒ ring.18
Ferrocene-based systems have played an important role
in the development of organometallic chemistry and continue
to find new applications in the fields of biological, inorganic,
and physical chemistry. Ferrocenecarboxaldehyde ͑FcCHO͒
has been used as a redox sensor to monitor substrate
binding,1 for incorporation in large proteins to act as redox
relays2 and as a DNA probe for study of DNA damage,3
illustrating the versatile nature of the ferrocene systems.
Widespread use of substituted ferrocenes has sustained the
interest in the study of different properties of various fer-
rocene systems since the time of the initial characterization
by Wilkinson,4 and Wilkinson et al.5
FcCHO is a particularly interesting ferrocene derivative,
as it exists in two phases, a plastic crystalline phase ͑phase I͒
between temperatures of 44 and 123.5 °C and a low-
temperature amorphous phase ͑phase II͒. These two phases
have been studied using x-ray diffraction6,7 and the dynamics
of molecular reorientations in phases I and II were investi-
gated by incoherent neutron-scattering techniques.8 FcCHO
has also been studied using 57Fe-NMR,9 13C-NMR,10 Möss-
bauer spectroscopy,7,11 He͑I͒ photoelectron spectroscopy,12
and electrochemical studies in solution.12 These studies pro-
vide us with a better understanding of the electronic proper-
ties of the molecule. There have been various studies that
The structure of FcCHO with the labeling scheme for the
carbon atoms is shown in Fig. 1. From measurements using
the “normal” distribution of isotopomers for FcCHO mol-
ecule, the spectra for 56Fe, 54Fe, 57Fe, and four unique 13C
isotopomers were recorded. For the deuterium-enriched
sample ͑ferrocene–2H–carboxaldehyde, FcCDO͒, which was
2
synthesized, only the 56Fe isotopomer with the H substitu-
a͒
tion at the Hald position ͑see Fig. 1͒ was measured. All spec-
tra were measured in the 4–12 GHz frequency range. Rota-
tional constants were determined by least-squares fitting of
the assigned transitions. From these rotational constants,
structural parameters of the molecule were obtained using a
Present address: California Institute Of Technology, Jet Propulsion Labo-
ratory, 4800 Oak Grove Drive, Pasadena, CA 91109.
b͒
Present address: Department of Physics, University of Maryland, College
Park, MD 20742.
c͒
Author to whom correspondence should be addressed. Electronic mail:
kukolich@u.arizona.edu
0021-9606/2005/123͑5͒/054317/9/$22.50
123, 054317-1
© 2005 American Institute of Physics
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-2
Subramanian et al.
J. Chem. Phys. 123, 054317 ͑2005͒
to 80 °C to obtain sufficient vapor pressure, and the pressure
of the neon carrier gas was maintained at 0.5 atm. This mix-
ture was pulsed into the Fabry-Perot cavity at a constant
pulse rate of 2 Hz. A low-noise liquid-nitrogen-cooled am-
plifier was placed in the rigid coax line attached to the an-
tenna in the cavity to increase the sensitivity by improving
the signal-to-noise ratio. Typical improvements in S/N were
a factor of 2, with some variation with frequency, since the
amplifier was not attached to the mirror nor directly con-
nected to the antenna. Since the factor-of-2 improvement in
S/N reduces averaging time by a factor of 4 to achieve
equivalent sensitivity, this facilitated measuring transitions of
the different isotopomers in natural abundance. Ninety-three
transitions were recorded for the 56Fe isotopomer and 15
transitions were observed for 54Fe ͑6% natural abundance͒,
six transitions were observed for 57Fe ͑2% natural abun-
dance͒ and five each for the two 13C isotopomers ͑1% natural
abundance͒. In the case of FcCDO, 24 transitions were re-
corded for 56Fe. These measured transitions for the various
isotopomers are listed in Tables I–III.
FIG. 1. Structure of the ferrocenecarboxaldehyde FcCHO with the atoms of
the molecule labeled. The values of the parameters that were determined
from the fits are indicated. The units for the distances are Å and the units of
the angle are degrees. The aldehyde ͑–CHO͒ group is tilted by only 0.3͑1͒°
out of the cyclopentadienyl plane, in the direction away from the Fe atom.
structural least-squares analysis and also from Kraitchman
analyses. Comparisons of the theoretical and x-ray data with
the gas-phase microwave data provide a more complete un-
derstanding of the structure of FcCHO.
III. THEORETICAL METHODS
Density functional theory ͑DFT͒ calculations were used
to optimize the geometry of FcCHO. Becke’s three-
parameter hybrid exchange potentials ͑B3͒ with the Perdew-
21–23
Wang ͑PW91͒
correlation functional have been known
II. EXPERIMENT
to produce accurate values for structural parameters for the
first-row transition metals. In the present work, we have used
two basis sets for the iron atom, namely, the Hay and Wadt
͑n+1͒ basis set with effective core potential24 and the Stutt-
gart relativistic, small-core effective core potential ͑ECP͒ ba-
sis set ͑Stuttgart RSC1997 ECP͒.25,26 In each case, we used
split-valence plus polarization SVP27,28 basis sets for the car-
bon, hydrogen, and oxygen atoms.
The FcCHO was purchased from Aldrich ͑catalog num-
ber: 122459͒ and used as received. The preparation of the
ferrocene–2H–carboxaldehyde was based on a modification
of the synthesis of FcCHO.19 2.0 g of ferrocene and a stir bar
were placed in a two-neck round-bottom flask. The round-
bottom flask was placed in −20 °C ice-methanol slush and a
stream of nitrogen gas passed through it. With the nitrogen
gas continuously flowing, 30 ml of dry tetrahydrofuran
͑THF͒ was added and the mixture was stirred for 15 min.
The red crystals of ferrocene dissolved to give a deep orange
solution. To this solution, 9.2 ml of 1.7M t-butyl lithium
͑tBuLi͒ in pentane was added over a 15-min period. The
solution was then stirred for 40 min and the temperature was
warmed to −10 °C. In the dry box a solution of 1-ml dry
THF and 1.7 ml of N, N-dimethylformamide-1–2H was
made, removed from the dry box, and added to the solution
containing ferrocene and tBuLi. The mixture was left to react
for 40 min, and no color change was observed. The ice-
methanol slush was removed and 0.25 ml of 1M HCl was
added, resulting in the formation of the deep red precipitate.
The solvent was removed at 55 °C with a slow decrease in
pressure from 1 atm to 0.5 torr. After being brought back to
atmospheric pressure by adding nitrogen, a cold-water finger
was inserted and the solid was sublimed at 0.5 torr and
75 °C. The dark red crystals were removed from the cold
finger in the dry box and were used with no further purifica-
tion. This synthesis yields FcCDO, with deuterium substitu-
tion at the aldehyde hydrogen site, Hald ͑see Fig. 1͒.
The combination of the B3PW91 method with the Hay
and Wadt ͑n+1͒ basis sets with the effective core potential
for iron and the SVP basis sets for the carbon, hydrogen, and
oxygen atoms gave structural parameters close to the experi-
mental parameters. However, this method and basis set com-
bination did not characterize the “tilt angle” and the devia-
tion of Hald from the plane of the Cp ring, which was
observed in the FcCHO molecule. The tilt angle is defined to
be the angle that the substituent ͑in our case, the bond
C1–C1 ͒ makes with the plane of the substituted Cp ring. The
Ј
deviation of the Hald from the plane of the Cp ring along the
z axis is represented as ⌬z͑Hald͒. To more accurately charac-
terize the tilt angle and ⌬z͑Hald͒, the combination of
rB3PW91 method with Stuttgart RSC1997 ECP for Fe atom
and the SVP basis sets for the carbon, hydrogen, and oxygen
atoms were used.
The success of the B3PW91 calculations to produce
near-experimental quality results is consistent with the usage
of the “best” exchange-correlation potentials for these
molecules.29 An energy optimization starting from an initial
staggered conformation of the molecule, using any of the
above basis sets, resulted in rotation to the eclipsed con-
former. This indicates that the eclipsed conformer is the
lower-energy conformation.
The pulsed-jet Fourier transform microwave spectrom-
eter used for the current experiment is described elsewhere in
detail.20 Both compounds, FcCHO and FcCDO, were heated
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-3
Structure of ferrocenecarboxaldehyde
J. Chem. Phys. 123, 054317 ͑2005͒
TABLE I. Observed transitions for FcCHO ͑56Fe͒ with fit residuals. The observed frequencies are given in
MHz. The average error in the observed values is 0.0005 MHz for all the transitions. The listed observed
frequencies are an average of four to six measurements taken at different stimulating frequencies.
J
Ka
Kc
J
KaЈ
KcЈ
Observed frequencies
O–C ͑kHz͒
Ј
4
4
4
3
4
4
4
3
4
4
3
3
5
5
5
5
5
4
3
3
3
3
4
5
5
5
5
4
5
5
5
4
6
6
6
5
6
6
7
4
4
4
3
4
7
6
4
4
6
6
6
6
6
6
7
3
5
6
1
0
1
2
2
3
3
2
1
2
2
2
1
0
1
0
1
2
3
3
3
3
1
2
4
3
4
2
3
1
2
2
0
1
0
2
1
1
1
2
3
3
3
2
2
2
3
3
5
5
3
4
1
4
2
3
2
3
4
4
4
2
3
2
1
1
3
2
2
1
4
5
5
5
5
3
1
0
1
0
3
4
2
3
1
2
2
4
3
3
6
6
6
4
5
6
6
2
2
1
0
2
5
5
2
1
2
1
4
3
5
2
5
1
3
3
3
1
0
0
1
2
3
3
1
1
2
1
1
2
1
1
0
0
1
2
2
2
2
0
2
4
3
4
1
3
1
2
1
1
1
0
1
2
0
2
1
2
2
1
0
3
2
2
2
5
5
3
4
1
4
3
1
1
3
3
3
3
1
2
1
0
1
2
1
2
2
3
4
4
4
4
2
0
0
1
1
3
3
1
2
0
2
1
3
2
3
5
5
5
3
4
5
4
3
1
1
1
3
3
4
2
2
1
0
3
2
4
1
4
2
3
2
5530.9236
5632.2061
5752.4220
5847.4059
5870.0252
5949.0948
5974.1313
6000.5630
6127.4693
6132.7385
6313.6044
6466.7604
6613.3339
6817.9444
6876.4005
6938.1616
6996.6167
7085.5298
7172.6270
7174.1945
7204.9092
7209.2116
7277.0527
7297.7690
7440.7335
7442.7455
7444.4522
7501.4024
7524.6701
7571.3947
7725.1143
8013.6190
8183.7942
8210.1967
8242.2543
8255.8321
8265.4302
8268.6520
8405.3807
8429.4875
8579.2856
8608.6308
8635.0564
8650.9869
8688.5989
8701.9979
8732.4449
8761.7849
8925.2805
8925.7363
8927.6012
8947.9450
8949.8635
8964.0758
8992.0689
9096.9539
9099.0473
9119.8115
0
0
3
3
2
3
3
3
2
3
3
2
2
4
4
4
4
4
3
2
2
2
2
3
4
4
4
4
3
4
4
4
3
5
5
5
4
5
5
6
3
3
3
2
3
6
5
3
3
5
5
5
5
5
5
6
2
4
5
0
0
0
0
0
0
1
0
2
1
0
−1
−1
0
−1
−1
−2
−1
−2
−1
−1
0
−2
−2
1
1
−1
−1
−1
1
2
1
5
1
1
0
−1
−1
−3
4
−2
1
0
2
0
1
3
−4
0
0
−1
−2
−2
0
−1
1
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-4
Subramanian et al.
TABLE I. ͑Continued.͒
J. Chem. Phys. 123, 054317 ͑2005͒
J
Ka
Kc
J
KaЈ
KcЈ
Observed frequencies
O–C ͑kHz͒
Ј
5
6
6
3
7
7
7
7
4
4
4
4
5
7
5
4
5
7
7
5
7
5
7
7
7
7
7
5
7
7
6
8
8
8
8
1
2
2
3
0
1
0
1
4
4
4
4
2
1
3
3
3
2
1
3
3
3
5
5
4
4
2
2
3
2
2
0
1
0
1
4
4
5
1
7
7
7
7
1
0
1
0
4
6
3
1
2
6
6
3
5
2
3
2
4
3
6
3
4
5
4
8
8
8
8
4
0
2
1
0
1
1
0
0
3
3
3
3
1
2
2
1
2
2
1
2
3
2
5
5
4
4
1
1
3
2
1
1
1
0
0
4
3
4
2
6
6
6
6
0
0
1
1
4
5
2
2
2
5
5
3
4
3
2
1
3
2
5
4
3
4
4
7
7
7
7
9216.2464
9286.6486
9386.4318
9449.1218
9525.1421
9536.4898
9551.5436
9562.8934
9768.8864
9769.3626
9773.1896
9773.6654
9780.4617
9833.4977
9889.2983
9977.2894
10 000.5574
10 082.4885
10270.064
10 305.1697
10395.747
10 416.4304
10 432.9767
10 435.4656
10 459.0807
10 509.6208
10 519.0545
10 623.6773
10747.393
10 797.2045
10 814.3033
10 853.8665
10 858.5824
10 865.2253
10 869.9337
3
−1
1
5
5
2
6
6
6
6
3
3
3
3
4
6
4
3
4
6
6
4
6
4
6
6
6
6
6
4
6
6
5
7
7
7
7
2
1
0
−1
0
−2
1
0
2
−1
1
1
−1
−3
0
1
2
0
0
1
1
−5
5
0
−2
1
−1
2
−7
−1
3
2
All the theoretical calculations were performed on a HP/
Compaq Alpha supercomputer ͑AURA͒ using GAUSSIAN 03
programs30 at the University of Arizona.
from the normal isotopomer fit. For the synthesized com-
pound, FcCDO, the transitions were fitted using five adjust-
able parameters: the three rotational constants ͑A, B, and C͒
along with two distortion constants ͑⌬J and ⌬K͒. The values
obtained for these parameters are given in Table IV.
For an accurate determination of the rotational constants
for different low-abundance isotopomers, four transitions
were measured and fitted. A prediction resulting from this
least-squares fit was used to find and measure the fifth tran-
sition, which was then added to the least-squares fit. This
procedure helped to verify that the initial assignments were
correct. Even though the molecule has three dipole moment
components, a, b, and c, transitions due to the b and c dipole
moments are weaker. Hence, only the strong a-type dipole
transitions were observed for the low-abundance isoto-
pomers.
IV. DATA ANALYSIS
The observed transitions for the various isotopomers
were fitted using a weighted least-square fitting program,
SPFIT,31 which employs a standard rigid rotor Hamiltonian
with centrifugal distortion constants. Due to a uniform un-
derestimation of the measurement uncertainties during the fit
procedure, a program, PIFORM,32 was used, for convenience,
to correct the fitted parameter uncertainties to obtain values
of the various constants with conventional standard errors.
For
FcCHO,
with
the
normal
isotopes
12
͑ C , 1H10, 16O1, 56Fe͒, transitions were fitted using eight
11
adjustable parameters: the three rotational constants ͑A, B,
and C͒, and five distortion constants ͑⌬J, ⌬K, ⌬JK, ␦J, and
␦k͒. The observed transitions for isotopomers of the molecule
͑54Fe, 57Fe, 13C1, 13C ͒ were fitted using the rotational con-
stants as adjustable parameters, while holding the distortion
constants ⌬J, ⌬K, ⌬JK, ␦J, and ␦k fixed to the values obtained
V. STRUCTURE DETERMINATION
A. Least-squares fit
1
Ј
The least-squares structural fits were accomplished by
adjusting the structural parameters to obtain the best fit to the
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-5
Structure of ferrocenecarboxaldehyde
J. Chem. Phys. 123, 054317 ͑2005͒
TABLE II. Observed transitions for FcCDO ͑56Fe2H͒ with fit residuals. The observed frequencies are given in
MHz. The numbers within the parentheses are the errors in the observed values obtained by averaging four to
six subsequent spectra.
J
Ka
Kc
J
KaЈ
KcЈ
Observed frequencies
O–C ͑kHz͒
Ј
4
4
4
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
7
7
7
0
2
1
1
0
2
3
3
1
2
1
0
2
5
5
3
4
1
4
3
2
1
0
1
4
2
3
5
5
4
3
2
4
3
6
6
5
2
1
4
3
5
2
3
4
7
7
6
3
0
2
1
1
0
2
3
3
1
2
1
0
2
5
5
3
4
1
4
3
2
1
0
1
3
1
2
4
4
3
2
1
3
2
5
5
4
1
0
3
2
4
1
2
3
6
6
5
5574.1185͑3͒
6051.5508͑3͒
6052.1776͑5͒
6805.3642͑4͒
6868.2139͑6͒
7213.6725͑9͒
7351.5415͑3͒
7427.4602͑3͒
7482.4944͑5͒
7623.4049͑6͒
8126.477͑3͒
8159.6646͑5͒
8603.9092͑4͒
8815.2571͑6͒
8815.6572͑8͒
8819.0781͑6͒
8836.9596͑7͒
8850.5052͑4͒
8851.5059͑4͒
8998.3549͑8͒
9167.2119͑9͒
9440.1607͑5͒
9456.0000͑5͒
10 161.3214͑9͒
−4
9
3
3
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
6
6
6
11
4
−5
−8
4
5
2
6
−6
−4
−8
7
1
−7
−6
−2
−3
−8
−3
9
7
3
18 rotational constants from the various isotopomers. For the
least-squares fit, a coordinate system is defined with the ori-
gin at the Fe atom, and such that the z axis passes through
the center of the two Cp rings. The x axis is perpendicular to
the z axis, passes through the Fe atom, and lies parallel to a
line joining the center of the Cp ring and C1 carbon atom
͑see Fig. 1͒. The y axis is perpendicular to the x and z axes.
It is also assumed that the two Cp rings have C5 symmetry
and the planes of these Cp rings are parallel. The assump-
tions that deviations from C5 symmetry for the Cp rings and
the angle between the Cp planes ͑C5H4CHO and C5H5͒ are
small enough to be neglected are supported by the DFT cal-
culations. A detailed study was done for the case of
chloroferrocene,15 where the effects of isotopic substitution
TABLE III. Observed transitions for FcCHO 54Fe, 57Fe, 13C1, and 13C isotopomers with fit residuals. The observed frequencies are given in MHz. The
1Ј
numbers within the parentheses are the errors in the observed values.
54Fe
57Fe
13C
13C1
1Ј
O–C
O–C
O–C
O–C
J
Ka Kc
J
Ј
KЈ KcЈ Observed frequencies ͑kHz͒ Observed frequencies ͑kHz͒ Observed frequencies ͑kHz͒ Observed frequencies ͑kHz͒
a
4
4
4
5
5
3
3
3
3
4
5
5
5
5
4
4
1
0
1
1
0
3
3
3
3
1
3
4
1
2
2
3
4
4
3
5
5
1
0
1
0
3
3
1
4
3
2
2
3
1
0
1
1
0
2
2
2
2
0
3
4
1
2
1
2
3
3
2
4
4
0
0
1
1
3
2
0
3
2
3
1
5533.7223͑7͒
5634.9570͑6͒
6130.5451͑6͒
6879.8672͑4͒
6941.5635͑4͒
7174.1945͑2͒
7178.5062͑2͒
7206.5183͑1͒
7210.8291͑1͒
7280.3853͑2͒
7446.5959͑3͒
7448.3291͑3͒
7575.1122͑4͒
1
0
5529.5404͑6͒
5630.8512͑7͒
6125.9539͑5͒
6874.6869͑6͒
6936.4805͑7͒
0
6
3
3
4
4
2
2
2
2
3
4
4
4
4
3
3
5583.3349͑5͒
6066.9440͑9͒
6
3
6
−1
6104.1582͑4͒
6852.4554͑6͒
6915.1812͑15͒
−4
−2
2
−2
0
−3
1
6878.4947͑6͒
−4
−1
0
0
1
−2
0
−1
1
7569.5521͑4͒
−6
7500.1742͑10͒
7642.0903͑9͒
0
1
7544.0983͑6͒
7692.8711͑3͒
1
1
8432.7883͑3͒
8581.5281͑1͒
0
0
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-6
Subramanian et al.
J. Chem. Phys. 123, 054317 ͑2005͒
TABLE IV. The spectral parameters obtained for the different isotopomers of FcCHO ͑56Fe; 54Fe; 57Fe; 13C
least-square fits. The error limits on all the values are 1␴.
,
13C1͒ and FcCDO ͑56Fe2H͒ determined using
1Ј
Parameter
56Fe
54Fe
57Fe
13C
13C1
56Fe2H
1Ј
A͑MHz͒
B͑MHz͒
C͑MHz͒
⌬J͑Hz͒
1289.908 85͑2͒
814.627 05͑1͒
659.227 67͑1͒
101.4͑1͒
−292.4͑5͒
364͑1͒
32.82͑5͒
−65͑1͒
1.8
1290.1502͑5͒
815.0455͑3͒
659.5647͑3͒
1289.79͑2͒
814.419͑1͒
659.0611͑6͒
1286.97͑2͒
805.8324͑6͒
653.4004͑9͒
1289.48͑1͒
811.2310͑6͒
656.9553͑6͒
1277.713͑6͒
803.4413͑6͒
652.8231͑7͒
a
a
a
a
͓101.4͔
͓−292.4͔
͓364͔
͓32.82͔
͓101.4͔
͓−292.4͔
͓364͔
͓32.82͔
͓101.4͔
͓101.4͑1͔͒
86͑8͒
¯
a
a
a
a
⌬K͑Hz͒
͓−292.4͔
͓−292.4͑5͔͒
a
a
a
a
⌬
͑Hz͒
͓364͔
͓364͔
−32͑2͒
¯
JK
a
a
a
a
␦J͑Hz͒
␦k͑Hz͒
␴͑fit͒͑kHz͒
Number of lines
͓32.82͔
͓−65͔
͓32.82͔
a
a
a
a
͓−65͔
͓−65͔
͓−65͔
¯
1.2
4.4
3.5
2.1
6.2
93
15
6
5
5
24
aHeld fixed during the fit.
were carefully studied in order to measure deviations from
planarity and deviations from C5 symmetry ͑for C atoms͒ for
the C5H4Cl ligand. Even though some small distortions were
detected, the resulting values for these distortions were
barely larger than the uncertainties. That experimental study
along with the present DFT calculations supports the above
assumptions used in the present analysis.
C–CHO moiety constrained to be planar, but the standard
deviations of the fit were larger ͑␴=0.14 MHz versus 0.07
MHz͒ and a much larger ͑unreasonable͒ value was obtained
for r͑C1–C1 ͒ ͑1.6 Å͒.
Ј
The structural fit obtained as described above yielded a
standard deviation of 0.07 MHz. This is considered to be a
very good fit since the rotational constants are over 1000
MHz. The coordinates ͑in angstroms͒ obtained from the
least-squares fit for the principal axis system of the molecule
͑a,b,c͒ and in the Cartesian coordinate system derived from
the unsubstituted ferrocene principal axes ͑x,y,z͒ are shown
in Table V. The parameters derived from this least-square-fit
structure are summarized in Table VII. As stated, the bond
distances r͑C1 –Hald͒ and r͑C1 –O͒ were manually adjusted
Many different structural fits were done to decide which
structural parameters could be unambiguously determined.
The DFT calculations were also used as a guide to reason-
able initial structures. From the analysis of the fit results, a
particular set of five structural parameters was found to be
the best determinable set. These five parameters are ͑1͒ the
distance from the Fe atom to the center of either the Cp ring
r͑Fe–Cp͒, ͑2͒ the ring radius of both the Cp rings r͑Cp͒
͓which can be used to directly calculate r͑C–C͒, the carbon-
carbon bond distances for the Cp rings͔, ͑3͒ the angle that the
Ј
Ј
and fixed at the values of 1.116 and 1.243 Å, respectively.
These values are very close to values obtained from the DFT
calculations. Several test fits were run where the aldehyde
͑–CHO͒ group was rotated with respect to the plane of the
Cp ring and the standard deviation was found to be greater
than 0.7 MHz. The resulting value of a͑Cp–CHO͒ indicates
C1–C1 makes with the Cp ring ͑the tilt angle͒ a͑Cp–CHO͒,
Ј
͑4͒ the C1–C1 bond length r͑C1–C1 ͒, and ͑5͒ z displace-
Ј
Ј
ment of the Hald with respect to the plane of the Cp ring,
⌬z͑Hald͒. The best-fit values obtained are ͑1͒ r͑Fe–Cp͒
=1.647͑4͒, ͑2͒ r͑Cp͒=1.2164͑2͒, ͑3͒ tilt angle a͑Cp–CHO͒
that the C1–C1 deviates out of the plane of the Cp ring, in a
Ј
direction away from the Fe atom only by 0.3͑1͒°, so it is
almost in the plane of the Cp ring. The Hald atom is moved
upwards, towards the Fe atom by 0.29͑4͒ Å. From the struc-
tural fit, it was also found that C1 and C6 are eclipsed ͑see
Fig. 1͒ and any attempt to fit a structure starting from a
staggered conformation did not converge.
=0.3͑2͒°, ͑4͒ r͑C1–C1 ͒=1.46͑1͒ Å, and ͑5͒ ⌬z͑Hald
͒
Ј
=0.29͑4͒ Å. The values of the parameters are shown in Fig.
1, which also shows the structure of the complex. As indi-
cated above, the tilt angle was found to be very small ͓only
0.3͑1͒°͔ and moves C1 away from the Fe atom. The dis-
Ј
tances from the Fe atom to the centers of the two Cp rings
were assumed to be the same. This was verified by a test fit,
where the distances were independently varied and the re-
sulting values for the two r͑Fe-Cp͒ distances came out to be
the same, within the uncertainties. The initial starting struc-
ture for the least-squares fit was based on the results from the
DFT calculations. Since deuterium substitutions were not
made on the Cp ring, the C–H bond distances of the Cp ring
were all fixed at 1.088 Å and the radial symmetry of the C–H
B. Kraitchman analysis
Since we have not measured all of the 13C and
deuterium-substituted isotopomers, a complete substitution
structure could not be obtained. Nevertheless, the Kraitch-
man analysis was very helpful in the structural analysis be-
cause the Kraitchman-derived coordinates are not subjected
to the same “correlation errors” that the least-squares-fit pa-
rameters are. The present substitution structural coordinates
͑see Table VI͒ obtained are close to the least-squares-fit val-
ues and provide support for the accuracy of the least-squares-
fit parameters. The various structural parameters derived
from the Kraitchman analysis are given in Table VII.
bonds was maintained during the fit. Both C1 –Hald and
Ј
C1 –O distances were manually adjusted by small increments
Ј
in the fits. We also found that the CHO moiety with respect
to the Cp plane is quite nonplanar in our best-fit results.
Relative to the C1–C1 bond, the C1 HO group is “bent” up
Ј
Ј
͑toward Fe͒ by nearly 30°. This result can readily be seen in
Since the Kraitchman analysis uses the changes in the
moments of inertia due to the monosubstitution of different
the last lines of Table V. Fits were also done with the
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-7
Structure of ferrocenecarboxaldehyde
J. Chem. Phys. 123, 054317 ͑2005͒
TABLE V. The coordinates ͑in Å͒ for FcCHO obtained from a least-squares fit, given in the principal axis
system ͑a,b,c͒ and in the Cartesian system ͑x,y,z͒ as explained in the least-square-fitting section of the text.
The estimated uncertainties are 0.01 Å.
Atom
a
b
c
x
y
z
Fe
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
H2
H3
H4
H5
H6
H7
H8
H9
H10
Ϫ0.424
1.612
Ϫ0.197
Ϫ0.331
Ϫ0.807
Ϫ1.896
Ϫ2.093
1.126
0.009
Ϫ0.166
1.114
0.000
1.216
0.000
0.000
0.000
1.647
1.188
0.376
1.157
1.647
0.286
0.904
Ϫ0.984
Ϫ0.984
0.376
0.715
1.647
0.152
Ϫ0.505
Ϫ1.167
Ϫ0.220
1.060
Ϫ0.715
Ϫ1.157
0.000
1.647
Ϫ0.971
Ϫ0.919
Ϫ1.343
Ϫ2.245
Ϫ2.379
Ϫ1.560
1.498
1.647
1.776
1.216
Ϫ1.647
Ϫ1.647
Ϫ1.647
Ϫ1.647
Ϫ1.647
1.647
1.300
0.376
1.157
0.211
0.851
Ϫ0.984
Ϫ0.984
0.376
0.715
0.014
Ϫ0.559
Ϫ1.221
2.079
Ϫ0.715
Ϫ1.157
2.192
0.981
Ϫ0.410
Ϫ2.473
Ϫ2.847
Ϫ1.014
2.599
0.712
Ϫ0.212
Ϫ0.466
1.087
1.681
Ϫ1.864
Ϫ1.864
0.712
1.355
1.647
Ϫ0.990
Ϫ2.243
Ϫ0.400
2.025
Ϫ1.355
Ϫ2.192
0.000
1.647
1.647
Ϫ0.230
Ϫ1.033
Ϫ2.742
Ϫ2.997
Ϫ1.444
2.539
2.305
Ϫ1.647
Ϫ1.647
Ϫ1.647
Ϫ1.647
Ϫ1.647
1.654
1.697
0.712
2.192
Ϫ0.366
Ϫ0.740
1.093
1.627
Ϫ1.864
Ϫ1.864
0.712
1.355
Ϫ1.044
Ϫ2.297
Ϫ0.458
Ϫ1.500
0.404
Ϫ1.355
Ϫ2.192
Ϫ0.050
Ϫ1.026
0.945
C
0.764
2.679
1Ј
Hald
O
2.502
1.161
3.130
1.355
2.853
1.602
3.369
1.374
atoms in the molecule and that the moments of inertia are
related to the square of the coordinates of the substituted
atoms, one can obtain only the absolute values of these co-
ordinates. The absolute values of these coordinates deter-
mined in the principal axis system of the parent molecule are
given in Table VI.
bond distances of the Cp rings are 1.430͑2͒ Å. The uncer-
tainties given in the parentheses are larger than the statistical
error contributions to the fit, because C5 symmetry was as-
sumed and these parameters are correlated with other param-
eters, which are not fitted directly.
The Fe parameters derived from the Kraitchman analysis
͑see Table VII͒ are obtained by using 54Fe as, bs, and cs
coordinates with the most abundant 56Fe isotopomer as the
“parent” molecule. A larger number of transitions was mea-
sured for the 54Fe isotopomer than for the 57Fe isotopomer,
so this is the more appropriate parent molecule.
TABLE VII. Structural parameters obtained for the FcCHO molecule. The
approximate ro structure is obtained from the least-squares fitting. The rs
structure is obtained using Kraitchman analysis. The rdft structure is the
output of the theoretical calculation ͑rB3PW91/Stuttgart RSC 1997 ECP͒
and, since vibrational corrections were not included, is a re structure. The
rx-ray structure is obtained from the x-ray studies ͑Ref. 6͒. The listed error
limits on the parameters are 1␴.
VI. DISCUSSION
Bond length
rs͑Å͒
ro͑Å͒
rx-ray͑Å͒ rdft͑Å͒
Gas-phase structural parameters for FcCHO have been
determined from microwave spectroscopic measurements.
Using the least-squares fit, with some parameters fixed, we
find that the bond distances between the Fe atom and the C
atoms of the two Cp rings are 2.047͑4͒ Å, and that the C–C
r͑C–C͒
r͑Fe–Cp͒
r͑Fe–C1͒
¯
¯
1.430͑2͒
1.647͑4͒
1.40
1.65
2.03
1.48
1.07
1.05
1.429
1.648
2.037
1.468
1.122
1.210
1.99͑9͒ 2.047͑4͒
1.44͑1͒ 1.46͑1͒
1.03͑4͒ 1.116a
r͑C1–C
͒
1Ј
r͑C –Hald
͒
1Ј
r͑C –O͒
¯
1.243a
1Ј
TABLE VI. The rs coordinates ͑in Å͒ obtained from Kraitchman analyses in
the principal axis system of the parent molecule, FcCHO
Bond angle
Єs͑°͒
¯
Єo͑°͒
108a
116a
122a
18a
Єx-ray͑°͒ Єdft͑°͒
1H1016O56Fe͒.
12
͑
C
Cp ring C–C–C
109.5
111
136
1
108
114
125
−6
11
C1–C –Hald interbond angle
¯
1Ј
Atom
͉as͉
͉bs͉
͉cs͉
C1–C –O interbond angle
¯
1Ј
C5–C1–C –Hald dihedral angle
¯
54Fe
57Fe
Hald
13C1
0.398
0.397
2.475
1.600
2.519
0.191
0.190
1.150
0.307
0.702
0.007
0.017
1.577
0.196
0.649
1Ј
C2–C1–C –O dihedral angle
¯
−16a
7
4.3c
0
2.8c
1Ј
b
–C HaldO tilt angle
1Ј
¯
0.3͑2͒
aHeld fixed during the structural fit. See text for further details.
bAway from the Fe atom.
13C
1Ј
cTowards the Fe atom.
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-8
Subramanian et al.
J. Chem. Phys. 123, 054317 ͑2005͒
We note that the agreement of the structural parameters
for Kraitchman analysis, DFT calculations, and least-squares
fit are well within the experimental error limits. The value
from the Fe atom to the center of the Cp ring. Using the
result for r͑Fe–Cp͒ from our earlier paper,18 r͑Fe–Cp͒
=1.6585−0.0821⌺␴I, and substituting the value for ⌺␴
I
=0.35 for the –CHO group,11 one determines the r͑Fe-Cp͒
distance to be 1.630 Å. This value falls within 1% of the
experimental gas-phase microwave value of 1.647͑2͒ Å and
adds further support for the validity of this correlation analy-
sis. This study also confirms that our previous correlations of
structural trends with substituent properties are reasonably
quantitative and fit experimental results within the experi-
mental uncertainties.
obtained for C1–C1 bond distance from Kraitchman is
Ј
slightly lower than the DFT and structural fit values, whereas
x-ray diffraction6 studies overestimate this bond length. The
difference between the Kraitchman and least-squares-fit val-
ues may be an indication of correlation effects in the fit, and
in this case we would favor the Kraitchman-derived value.
The experimental and DFT-calculated structures ob-
tained for the FcCHO molecule show that the Cp rings are in
an eclipsed conformation. However, in the case of ferrocene,
the Cp rings are staggered in the crystal33 and eclipsed in the
gas phase.34 Attempts to perform a least-squares fit using a
staggered conformation as a starting point did not converge,
and this is consistent with the prediction from the theoretical
calculations that the staggered conformation has a much
higher energy when compared to the eclipsed conformer. A
single-point energy calculation predicts the difference in en-
ergy between eclipsed and staggered conformer to be
4 kJ/mol, with the eclipsed conformer being more stable.
This calculation was done using the rB3PW91 method and
using the Stuttgart RSC1997 ECP for iron and the SVP basis
set for the carbon, oxygen, and hydrogen atoms.
In conclusion, the near-complete gas-phase structure of
FcCHO has been obtained using microwave spectroscopy.
The present work also indicates that this compound exists in
the gas phase as an eclipsed conformer, with a small tilt of
the aldehyde group away from the Fe atom of only 0.3͑2͒°
and Hald is displaced towards the Fe atom by 0.29͑4͒ Å. The
present work also supports the previously obtained coherent
picture of the effect of substituent effects on the structure of
ferrocene.
ACKNOWLEDGMENTS
This material is based on the work supported by the
National Science Foundation under Grant No. CHE-
0304969. This support from the National Science Foundation
is gratefully acknowledged. One of the authors ͑P.A.C.͒
gratefully acknowledges support from a NASA Arizona
Space Grant internship during the course of this project. We
thank Oliver Indris, Jennifer Dannemiller, and Kirsten Fields
for help in obtaining data on the normal isotopomer.
The tilt angle is defined to be the angle that the substitu-
ent ͑in our case, the bond C1–C1 ͒ makes with the carbon
Ј
plane of the substituted Cp ring. From previous studies in
this lab, this angle has been consistently shown to have a
nonzero value, in the direction away from the Fe atom. In the
case of chloroferrocene,15 the value of the tilt angle is
2.7͑6͒°, for dimethylferrocene,16 the value of this angle is
2.66͑2͒°, and for ethynylferrocene,18 the value of this angle
is 2.75͑6͒°. According to the present best-fit results for Fc-
CHO, we found that the tilt angle determined from the least-
squares fit is 0.3͑2͒°, in the direction away the Fe atom. This
1 P. D. Beer, D. B. Crowe, M. I. Odgen, M. G. B. Drew, and B. Main, J.
Chem. Soc. Dalton Trans. 14, 2107 ͑1993͒.
2 A. Heller, Acc. Chem. Res. 23, 128 ͑1990͒.
3 C. Xu, H. Cai, P. He, and Y. Fang, Fresenius’ J. Anal. Chem. 367, 593
͑2000͒.
4 G. Wilkinson, J. Organomet. Chem. 100, 273 ͑1975͒.
5 G. Wilkinson, M. Rosenblum, M. C. Whiting, and R. B. Woodward, J.
Am. Chem. Soc. 74, 5764 ͑1952͒.
indicates that the substituent ͑in this case C1 ͒ almost lies in
Ј
the plane of the substituted Cp ring. However, the ab initio
calculation using the method B3PW91 and basis set Stuttgart
RSC 1997 ECP described in Sec. III predicts the tilt angle to
be 2.8° in the direction towards the Fe atom and the x-ray
diffraction6 study indicates the tilt angle is 4° in the direction
towards the Fe atom ͑see Table VII͒. Nevertheless, according
to our previous results for chloroferrocene ͓a substituted fer-
6 M. F. Daniel, A. J. Leadbetter, and M. A. Mazid, J. Chem. Soc., Faraday
Trans. 2 77, 1837 ͑1981͒.
7 M. F. Daniel, A. J. Leadbetter, R. E. Meads, and W. G. Parker, J. Chem.
Soc., Faraday Trans. 2 74, 456 ͑1978͒.
8 M. F. Daniel, A. J. Leadbetter, and R. M. Richardson, J. Chem. Soc.,
Faraday Trans. 2 77, 1851 ͑1981͒.
9 B. Wrackmeyer, O. L. Tok, A. Ayazi, H. E. Maizel, and M. Herberhold,
Magn. Reson. Chem. 42, 827 ͑2004͒.
rocene with
a
strongly electron-withdrawing group,
10 B. Wrackmeyer, O. L. Tok, and A. A. Koridze, Magn. Reson. Chem. 42,
750 ͑2004͒.
tilt angle=2.7͑6͒° away from Fe atom͔, we believe that the
tilt away from the Fe atom is a reasonable result. We also
found that the Hald atom of the aldehyde group ͑–CHO͒ is
displaced towards the Fe atom by 0.29͑4͒ Å.
11 R. M. G. Roberts and J. Silver, J. Organomet. Chem. 263, 235 ͑1984͒.
12 T. Matsumura-Inoue, K. Kuroda, Y. Umezawa, and Y. Achiba, J. Chem.
Soc., Faraday Trans. 2 85, 857 ͑1989͒.
13 J. C. Green, Struct. Bonding ͑Berlin͒ 43, 37 ͑1981͒.
14 B. J. Drouin, T. G. Lavaty, P. A. Cassak, and S. G. Kukolich, J. Chem.
Phys. 107, 6541 ͑1997͒.
The linear correlation between the distance between the
Fe atom and the center of the Cp, r͑Fe–Cp͒, and the Ham-
15 B. J. Drouin, J. J. Dannemiller, and S. G. Kukolich, J. Chem. Phys. 112,
747 ͑2000͒.
mett parameter ⌺␴ was presented in an earlier paper by our
I
group.18 We ascribe this linear correlation to the electron-
withdrawing effects of the substituent group relative to the H
atom, which alters the electron density of the Cp ring. There
is a decrease in electron density of the Cp ring as the elec-
tronegativity of the substituent group increases. This affects
the binding of the Fe-␲-Cp ligand and is observed as an
increase in the binding energy35 and ionization energy of the
complex, and structurally as the variation of the distance
16 D. S. Margolis, C. Tanjaroon, and S. G. Kukolich, J. Chem. Phys. 117,
3741 ͑2002͒.
17 C. Tanjaroon, K. S. Keck, and S. G. Kukolich, J. Am. Chem. Soc. 126,
844 ͑2004͒.
18 R. Subramanian, C. Karunatilaka, K. S. Keck, and S. G. Kukolich, Inorg.
Chem. 44, 3137 ͑2005͒.
19 U. T. Mueller-Westerhoff, Z. Yang, and G. Ingram, J. Organomet. Chem.
463, 163 ͑1993͒.
20 R. E. Bumgarner and S. G. Kukolich, J. Chem. Phys. 86, 1083 ͑1987͒.
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions

054317-9
Structure of ferrocenecarboxaldehyde
J. Chem. Phys. 123, 054317 ͑2005͒
21 A. D. Becke, J. Chem. Phys. 98, 5648 ͑1993͒.
22 J. P. Perdew and Y. Wang, Phys. Rev. B 45, 13244 ͑1992͒.
23 C. Lee, W. Yang, and R. G. Parr, Phys. Rev. B 37, 785 ͑1987͒.
24 P. J. Hay and W. R. Wadt, J. Chem. Phys. 82, 299 ͑1985͒.
25 A. Bergner, M. Dolg, W. Kuechle, H. Stoll, and H. Preuss, Mol. Phys.
80, 1431 ͑1993͒.
30 M. J. Frisch, G. W. Trucks, H. B. Schlegel et al., GAUSSIAN 03, REV. B.05,
Gaussian, Inc., Pittsburgh, PA, 2003.
31 H. M. Pickett, J. Mol. Spectrosc. 148, 371 ͑1991͒.
32 This program is authored by Dr. Zbigniew Kisiel and can be obtained
along with the instructions for using the program from the website http://
info.ifpan.edu.pl/~kisiel/prospe.htm
26 M. Dolg, H. Stoll, H. Preuss, and R. M. Pitzer, J. Phys. Chem. 97, 5852
͑1993͒.
33 W. L. Baun, Anal. Chem. 31, 1308 ͑1959͒.
34 A. Haaland and J. E. Nilsson, Acta Chem. Scand. ͑1947-1973͒ 22, 2653
͑1968͒.
27 A. Schaefer, H. Horn, and R. Ahlrichs, J. Chem. Phys. 97, 2571 ͑1992͒.
28 A. Schaefer, C. Huber, and R. Ahlrichs, J. Chem. Phys. 100, 5829
͑1994͒.
35 P. G. Gassman, D. W. Macomber, and J. W. Hershberger, Organometal-
lics 2, 1470 ͑1983͒.
29 M. Swart and J. G. Snijders, Theor. Chem. Acc. 110, 34 ͑2003͒.
Downloaded 12 Mar 2013 to 141.161.91.14. Redistribution subject to AIP license or copyright; see http://jcp.aip.org/about/rights_and_permissions