Synthesis, Computational Studies, Inelastic Neutron Scattering, Infrared and Raman Spectroscopy of Ruthenocene

Article abstract of DOI:10.1002/ejic.201800914

In this work we report a new and simplified synthesis of ruthenocene that prevents the formation of unwanted by-products. We have also revisited the vibrational spectroscopy of this iconic molecule and in addition to infrared and variable temperature Raman spectra, we present the first inelastic neutron scattering (INS) spectrum, a technique that is little used in inorganic chemistry. The Raman spectra in the low energy range (< 200 cm^–1) clearly show that the mode assigned as the ring–Ru–ring torsion is a librational mode. By generating the INS spectra predicted by previous assignment schemes we are able to show that they are all, at least partially, wrong because they fail to correctly predict the experimental INS spectrum. In the case of ruthenocene, the addition of INS data, in combination with periodic-DFT calculations, has enabled the first correct assignment of the internal modes of ruthenocene. This straightforward means to test proposed assignments is one of the great strengths of vibrational spectroscopy with neutrons.

Full text of DOI:10.1002/ejic.201800914

A Journal of
Accepted Article
Title: Synthesis, computational studies, inelastic neutron scattering,
infrared and Raman spectroscopy of ruthenocene
Authors: Stewart F. Parker and Ian R Butler
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201800914
Link to VoR: http://dx.doi.org/10.1002/ejic.201800914

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
Synthesis, computational studies, inelastic neutron scattering,
infrared and Raman spectroscopy of ruthenocene
[a]
[b]
Stewart F. Parker,* and Ian R. Butler
Abstract: In this work we report a new and simplified synthesis of
ruthenocene that prevents the formation of unwanted by-products. We
have also revisited the vibrational spectroscopy of this iconic molecule
and in addition to infrared and variable temperature Raman spectra,
we present the first inelastic neutron scattering (INS) spectrum, a
technique that is little used in inorganic chemistry. The Raman spectra
in the low energy range (<200 cm ) clearly show that the mode
assigned as the ring–Ru–ring torsion is a librational mode. By
generating the INS spectra predicted by previous assignment
schemes, we are able to show that they are all, at least partially, wrong
because they fail to correctly predict the experimental INS spectrum.
In the case of ruthenocene, the addition of INS data, in combination
with periodic-DFT calculations, has enabled the first correct
assignment of the internal modes of ruthenocene. This straightforward
means to test proposed assignments is one of the great strengths of
vibrational spectroscopy with neutrons.
1.[7] To date, only ferrocene has been investigated by INS
[
8]
spectroscopy.
Literature methods for the synthesis of ruthenocene are
[
3]
complex involving either a Grignard reagent or the preparation
[
9]
of sodium cyclopentadienyl. In this work, we present a new and
much simpler synthesis of ruthenocene together with a detailed
spectroscopic characterization of the solid state spectra by
infrared, Raman and INS spectroscopies. A complete assignment
is made via periodic density functional theory (periodic-DFT). All
previous work has relied on isolated molecule calculations[ that
neglect the effect of the solid state.
-
1
10]
Results and Discussion
Synthesis
As mentioned, there are many synthetic routes to ruthenocene
which essentially use the reaction of a metal or metalloid
Introduction
cyclopentadienide with
a
suitable ruthenium (II) or (III)
.
[3,11]
complex
When Ru (III) is used, a reducing agent is also
The discovery of ferrocene[1] and the recognition[2] of its
unprecedented bis-η mode of coordination marked the beginning
of modern organometallic chemistry. It was very quickly realised
that ruthenium and osmium should form analogous compounds
required. In our labs the method of choice is one developed in-
house in 1981 which is essentially a variant of one of the general
5
[
3]
[
12]
published synthetic methods. The method is shown in Scheme
. This uses hydrated ruthenium trichloride, freshly cracked
1
5
and the preparation of bis-η -cyclopentadienyl ruthenium
dicyclopentadiene (to give cyclopentadiene) and powdered zinc,
as the reducing agent. The difference of our method is that the
reaction is performed in ethanol at low temperature. What is of
interest in this synthetic route is that the deep blue color attributed
to an initially formed ruthenocenium complex, is discharged
[
3]
5
(
(
ruthenocene )
and
bis-η -cyclopentadienyl
osmium
[
4]
osmocene ) soon followed.
The vibrational spectroscopy of all three compounds has been
characterised, e.g.[ with ferrocene being the most studied.
However, the spectroscopy is complicated by selection rules.
These are irrelevant for inelastic neutron scattering (INS)
5]
o
rapidly, (< 0.1 s) on warming from -60 C to room temperature to
leave a clear solution with only the grey colloidal zinc remaining.
Thus, the reduction step is clearly observable. The workup is
[
6]
spectroscopy, a technique that is little used in inorganic and
organometallic chemistry. It has the advantage that, in principle,
all the modes are observable. In practice, the spectra are
dominated by modes involving hydrogen motion and, for
instrumental reasons, the optimal spectral range is 0 – 1600 cm-
[12a]
essentially the same as the method used by Porri et al. , which
involves crystallization of the colorless iridescent ruthenocene
[
a]
b]
Dr Stewart F. Parker
ISIS Facility
STFC Rutherford Appleton Laboratory
Chilton, Didcot, Oxon OX11 0QX, UK
E-mail: stewart.parker@stfc.ac.uk
Dr Ian. R Butler
[
School of Chemistry
Bangor University
Bangor, Gwynedd, Wales, LL57 2UW, UK
Scheme 1. Synthesis of ruthenocene. The intermediate ruthenocenium ion is
not isolated.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
from ethanol. By removing the need for heating the solution,
unwanted by-products are not formed (see Figures S1 and S2)
and yields are higher. The reaction is carried out typically on a 5-
observed and calculated INS spectra is a very stringent test of an
[
6]
assignment scheme. We have previously shown that while the
transition energies are very dependent on the quality of the
calculation, the atomic displacements of each atom in the mode
are relatively insensitive, so long as the calculated geometry is
10 g scale of ruthenium trichloride trihydrate but is fully scaleable
to the ten’s of gram scale.
[
16]
reasonably correct. This means that the INS spectrum can be
simulated by replacing the calculated transition energies with
those from the literature on a mode-by-mode basis. Figure 3
shows the results for the literature assignments in Table 3. It can
be seen that none of the calculated spectra match the
experimental data. All fail to reproduce the very strong feature at
Vibrational Spectroscopy
Figure 1 shows the infrared, Raman and INS spectra of
ruthenocene. The complementarity of the three techniques is
readily apparent and emphasizes the need for all three in order to
observe all of the modes. Our infrared and Raman data are in
-1
~900 cm , this is probably because there are only very weak
features at this energy in the Raman and infrared spectra, Figure
1b,c, so were not considered to be fundamentals. In contrast, the
[
13]
complete agreement with that of Adams and Fernando and of
Bodenheimer and Low.[14] The INS spectrum has not been
-1
-1
previously reported. The complete (0–3200 cm ) Raman spectra
as a function of temperature are shown in Figures S3 and S4 of
the Supporting Information.
intensity in the 1200 cm region is overestimated because weak
modes in this region were assigned as fundamentals. Figure 1a
Figure 2. Variable temperature 532 nm Raman spectra of ruthenocene in the
low energy region. Note the y-axis is logarithmic.
Figure 1. Vibrational spectra of ruthenocene: (a) infrared at room temperature,
(b) Raman at 10 K and (c) INS at 10 K.
One of the most problematic modes to assign has been the
-1 [13,14]
ring–Ru–ring torsion. This was assigned at 134 cm ,
although it was recognized that the temperature dependence was
anomalous: unlike all the other internal modes this undergoes a
marked shift on cooling. The low energy region is shown in detail
-1
in Figure 2 and the shift in the mode from 106 cm at 300 K to
-
1
1
31 cm at 10 K is clearly seen. However, a similar temperature
-1
dependence is seen for the lattice modes at 40 – 100 cm
consistent with the shrinkage in the unit cell volume , (818 Å at
00 K and 789 Å at 101 K), and we assign the 131 cm mode to
,
[
15]
3
3
-1
3
a libration. The detailed assignment of the internal modes of
ruthenocene is considered in the next section.
Assignment of the Spectra
Figure 3. Comparison of INS spectra of ruthenocene: (a) experimental and
There have been several previous assignments of the
vibrational spectra of ruthenocene[5,10,13,14] and these are
summarized in columns 4 – 7 of Table 1. It is apparent that there
is no consensus between the assignments. Comparison of
[
10a]
those calculated from the assignment of (b) Prosenc et al,
(c) Adams and
[13]
[14]
Fernando and (d) Bodenheimer and Low
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
shows that there are only overtones and combinations at this
energy. As shown in the previous section, the assignment of the
torsional mode to the feature at ~130 cm- is clearly erroneous.
From Figure 3, it is apparent that assignments based on neither
a much better match. The only area of serious disagreement is
-1
around 800 cm , where a doublet rather a triplet is observed. By
slightly shifting the calculated transition energies to match the
experimental ones, Figure 4c is obtained. It can be seen that for
the internal modes the agreement is almost exact. Column 7 of
Table 1 lists the average values of the transitions from Table S1.
It can be seen that the “best fit” values of column 8 of Table 1 are
in close agreement with those from the periodic-DFT calculation,
column 7. Note that column 8 contains experimental values.
Inspection of Table 1, shows that our best fit values are in
disagreement with previous assignments in several respects. The
reasons for some of these have already been discussed: the
bands at 900 cm- were not considered to be fundamentals, the
INS spectrum shows unambiguously that this is the case.
Conversely, the bands around 1200 cm- assigned as
fundamentals are not. This is seen in the INS spectrum and also
in the infrared and Raman spectra calculated by CASTEP, see
Figures 5 and 6 respectively. These only include fundamentals
(since the calculation assumes the harmonic approximation and
overtones and combinations result from anharmonicity) and there
1
[13,14]
empirical correlations
or isolated molecule DFT
[10a]
calculations
are able to reproduce the data successfully. To
address this problem, we have carried out a periodic-DFT
calculation of the complete orthorhombic unit cell, (space group
[
15]
Pnma ). The isolated molecule has D5h symmetry, however,
fivefold symmetry is incompatible with long range order and in the
crystal, the molecule occupies a site of C symmetry. The
s
correlation scheme (reproduced from [14]) is shown in Figure S5
of the Supporting Information. The calculated transition energies
for the complete unit cell, together with the mode number and
symmetry of the “parent” mode under D5h symmetry, are given in
Table S1 of the Supporting Information. In most cases, the factor
1
1
-1
group splitting is small, ~10 cm or so, the only exception is mode
4, an out-of-plane C–H deformation, which shows a very large
1
-1
splitting of 50 cm . We have no explanation for this anomaly.
In Figure 4, we compare the results of the periodic-DFT
calculation of the complete orthorhombic unit cell, 4b, with the
experimental spectrum, 4a. It is immediately obvious that there is
-1
is no intensity at 1200 cm .
-1
Table 1. Mode description, symmetry and transition energies (cm ) for ruthenocene.
Description[a]
D
5h symmetry
Mode no[ .
Adams
13]
Bodenheimer
[14]
Prosenc
[10a]
37
This work
CASTEP Pnma
b]
[
D
5h optimised
85
Torsion
δ(ring–Ru–ring)
ν(Ru–ring)
ν(Ru–ring)
Ring tilt
Ring tilt
π(CCC)
π(CCC)
π(C–H)
π(C–H)
δ(CCC)
δ(CCC)
π(C–H)
π(C–H)
π(C–H)
π(C–H)
δ(C–H)
δ(C–H)
δ(C–H)
δ(C–H)
ν(C–C)
ν(C–C)
δ(C–H)
δ(C–H)
ν(C–C)
ν(C–C)
v(C–C)
v(C–C)
ν(C–H)
ν(C–H)
ν(C–H)
ν(C–H)
ν(C–H)
ν(C–H)
A
1
"
6
22
4
11
16
21
28
34
9
134
168
333
450
400
381
604
604
808
818
868
902
839
130
170
333
380
400
450
603
600
808
818
868
900
842
834
1050
1065
1025
1000
1205
1188
1100
1095
1250
1255
1360
1360
1410
1410
3084
3076
3092
3100
3104
3112
E
A
1
'
438
322
375
373
147
577
570
791
803
856
823
794
784
821
866
990
160/168
323
167/177
335
1
'
A
E
2
"
"
373
389
400
455
595
602
808
818
820
838
840
866
904
904
994
1007
1053
1064
1100
1100
1253
1253
1327
1341
1411
1411
3080
3080
3095
3100
3105
3111
1
382/387
441/446
595/598
583/586
789
E
E
1
'
'
2
E
A
2
"
"
2
A
1
'
2
814
E
E
2
"
33
27
14
19
31
25
18
13
24
30
3
798/801
825/829
805
2
'
E
E
1
"
1
'
829
837/845
882/888
894/900
990/995
982/986
1043/1056
1037/1040
1092
E
E
E
2
"
1049
1063
1000
1001
1187
1206
1099
1099
1251
1251
1342
1361
1406
1410
3080
3080
3095
3100
3105
3111
2
'
'
1
E1"
979
E
2
'
1043
1033
1093
1096
1235
1234
1330
1352
1394
1394
3162
3162
3147
3173
3148
3172
E
2
"
A
1
'
A
2
"
10
7
5
1094
1244
1236
A
2
'
A
E
1
"
"
2
32
26
20
15
12
17
29
8
1323/1330
1344/1352
1398/1399
1389/1394
3159/3164
3164/3168
3138/3139
3170
E
E
2
'
'
1
E
E
1
"
"
1
E
1
'
E
A
2
"
"
2
E
A
2
'
'
23
1
3170
3176
1
[a]ν = stretch, δ = in-plane bend, π = out-of-plane bend
[b]The mode numbering is that used by refs. [11] and [12].
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
Figure 6. Comparison of Raman spectra of ruthenocene: (a) experimental and
(b) calculated by periodic-DFT calculation of the complete unit.
Figure 4. Comparison of INS spectra of ruthenocene: (a) experimental and
those calculated by periodic-DFT calculation of the complete unit cell (b) and (c)
after scaling the transition energies.
and the reverse for the Ru–ring tilt mode. Inspection of the atomic
displacements shows that for the mode at ~170 cm , the C–Ru–
-1
C bond angles change by ± 3° with only a ± 0.01 Å change in the
-1
Ru–C bond length. While for the mode at ~450 cm , the C–Ru–C
bond angles change by <± 0.5° and the Ru–C bond length by ±
1
0
.05 Å. Thus the 170 cm- mode must be the ring–Ru–ring bend
') and the 450 cm- mode the Ru–ring tilt (ν21, E
1
'). The
(
ν
22, E
1
1
other ambiguity has been between the asymmetric Ru–ring
stretch (ν11, A ") and the Ru–ring tilt (ν21, E '). Inspection of the
2
1
mode visualizations confirms that the asymmetric Ru–ring stretch
is at 389 cm- and, as previously discussed, that the Ru–ring tilt
1
-1
(
ν21, E
1
') is at 450 cm . This is also apparent from the simulation
of the predicted spectra, Figure 3. We agree with the assignments
[
14]
of Bodenheimer and Low in this region and it can be seen that
Figure 3d is the only one that agrees with the experimental
spectrum, Figure 3a.
The only remaining uncertainty is the location of the torsional
mode. The mode visualizations show that the four factor group
-1
components are at 42, 58, 110 and 114 cm , Table S1. The
parent mode, ν , A " is infrared and Raman forbidden under D5h
Figure 5. Comparison of infrared spectra of ruthenocene: (a) experimental and
(b) calculated by periodic-DFT calculation of the complete unit. The derivative-
6
1
like linehape in (a) is a particle size effect.
symmetry, in the crystal it is allowed, but the induced activity is
very small.
All previous work has implicitly used an isolated molecule
model, whereby to satisfy the invariance of the center-of-mass of
the system undergoing the motion (a condition for a normal mode),
motion in one ring generates motion in the second ring. In the
solid state, this requirement is more subtle: it is that the center-of-
mass of the unit cell is invariant for an optic mode. Thus for a
system that has more than one molecule in the primitive cell, such
as ruthenocene (Z = 4), motion in one entity can be compensated
by motion in another. For ruthenocene this means that only one
ring of a molecule may be active in a given mode. Visualisation of
the ~130 cm- mode, confirms that this is the case here and that
it is a libration.
The most contentious assignments are for the metal–ring
vibrations. There are five of these: symmetric and asymmetric
(
with respect to the horizontal mirror plane in the isolated
molecule) Ru–ring stretch (ν , A ' and ν11, A ") and Ru–ring tilt (ν16
" and ν21, E ') and the ring–Ru–ring bend (ν22, E '). The
symmetric Ru–ring stretch (ν ') at 335 cm is unambiguously
assigned by its polarization characteristics in both solution and
4
1
2
,
E
1
1
1
-1
4
, A
1
[
13]
[10a]
the solid. It has been claimed
that the Ru–ring tilt (ν21, E
1
')
and the ring–Ru–ring bend (ν22, E
1
') are so mixed that the
1
distinction between them is arbitrary. This is not correct. The ring–
Ru–ring bend mode involves predominantly changes in the
C–Ru–C bond angles with little change in the Ru–C bond lengths
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
Conclusions
provided by STFC's e-Science facility. Dr dan Evans (Bangor
University) is thanked for measuring the nmr spectra.
In this work we report a simplified synthesis of ruthenocene and
we have also revisited the vibrational spectroscopy of this iconic
molecule. The addition of INS data in combination with periodic-
DFT calculations has enabled the first correct assignment of the
internal modes of ruthenocene. By generating the INS spectra
predicted by previous assignment schemes we are able to show
that they all fail to correctly predict the experimental INS spectrum.
This straightforward means to test proposed assignments is one
of the great strengths of vibrational spectroscopy with neutrons.
Keywords: Ruthenocene • Inelastic neutron scattering
spectroscopy • Infrared spectroscopy • Raman spectroscopy •
Density functional theory
[
1]
2]
a) T. J. Kealy, P. L. Pauson, Nature, 1951, 168, 1039; b) S. A. Miller, J.
A. Tebboth J. F. Tremaine, J . Chem. Soc. Resumed 1952, 632-635.
G. Wilkinson, M. Rosenblum, M. C. Whiting, R. B. Woodward, J. Amer.
Chem. Soc. 1952, 74, 2125-2126.
[
[3]
[4]
[5]
G. Wilkinson, J. Amer. Chem. Soc. 1952, 74, 6146-6147.
E. O. Fischer, H. Grumbert, Chem. Ber. 1959, 92, 2302-2309.
B. V. Lokshin, V. T. Aleksanian, E. B. Rusach, J. Organomet. Chem.
Experimental Section
1975, 86, 253-256.
[6]
P. C. H. Mitchell, S. F. Parker, A. J. Ramirez-Cuesta, J. Tomkinson,
Vibrational spectroscopy with neutrons, with applications in chemistry,
biology, materials science and catalysis, World Scientific, Singapore,
Materials: The synthesis of ruthenocene is described in the main text.
Ruthenium trichloride hydrate was obtained from Johnson-Matthey and
the zinc dust used was standard laboratory grade.
2005.
[7]
S. F. Parker, D. Lennon. P.W. Albers, Appl. Spec. 2011, 65, 1325-
1341.
Vibrational spectroscopy: Variable temperature, 10 – 300 K, Raman
spectra were recorded using 532 nm excitation with a modified Renishaw
[8]
a) E. Kemner, I. M. de Schepper, G. J. Kearley, U. A. Jayasooriya, J.
Chem. Phys. 2000, 112, 10926-10929; b) U. A. Jayasooriya, S. A.
Malone, A. I. Chumakov,R. Rüffer, A. R. Overweg, C. R. Nicklin,
ChemPhysChem 2001, 2, 177-180.
[
17]
InVia system that has been previously described. Infrared spectra were
recorded using a Bruker Vertex70 FTIR spectrometer, over the range 100
to 4000 cm- at 4 cm resolution with a DLaTGS detector using 64 scans
and the Bruker Diamond ATR. The use of the ultra-wide range beamsplitter
enabled the entire spectral range to be recorded without the need to
change beamsplitters. The spectra have been corrected for the
wavelength-dependent variation in path length using the Bruker software.
1
-1
[9]
D. E. Bublitz, W. E. McEwen, and J. Kleinberg, Org. Synth. 1961, 41,
96-99.
[10]
a) M. H. Prosenc, H. Reddmann, H.-D. Amberger, Spectrochimica Acta
Part A, 2012, 87, 126-134; b) C. Latouche, F. Palazzetti, D. Skouteris,
V. Barone, J. Chem. Theory Comput. 2014, 10, 4565−4573.
a) M. D. Rausch, E. O. Fischer, H. Grubert, J. Amer. Chem. Soc.,1960,
82, 76-82. b) G.J. Gauthier, J. Chem Soc., D: Chem. Comm., 1969,
[
18]
[19]
The INS spectrum was recorded at <20 K using TOSCA at ISIS. The
spectrum is available at the INS database:
http://wwwisis2.isis.rl.ac.uk/INSdatabase/ .
[11]
[12]
13, 690. c) D. C. Liles, A. Shaver, E. Singleton, Eric; M. B., J.
Computational studies: The plane wave pseudopotential based program
Organomet. Chem. 1985, 288, C33-C36.d) M. O. Albers, D.C, Liles,
D.J. Robinson, A. Shaver,E. Singleton,M. B. Wiege, J. C. A. Boeyens,
D.C. Levendis, Organometallics, 1986, 5, 2321-7.e) C. H. Winter, S.
Pirzad, D, H, Cao, J. Chem. Soc., Chem. Comm., 1991, 1026-.7.
a) P. Pertici, G. Vitulli,,L. Porri, J. Chem. Soc., Chem. Comm, 1975,
846. b) A. Z. Rubezhov, A. S. Ivanov, A. A. Bezrukova, Izv. Akad.
Nauk SSSR, Ser. Khim., 1979, 1606-8.
CASTEP was used for the calculation of the vibrational transition energies
[
20,21]
and their intensities.
(
Dispersion corrected density functional theory
DFT-D) calculations using the Tkatchenko and Scheffler[22] correction
were carried out with the generalised gradient approximation (GGA)
Perdew-Burke-Ernzerhof (PBE) functional in conjunction with optimised
norm-conserving pseudopotentials. All of the calculations were converged
-
1
to better than |0.0075| eV Å . After geometry optimisation, the vibrational
spectra were calculated in the harmonic approximation using density-
[13]
[14]
D. M. Adams, W. S. Fernando, J. Chem. Soc. Dalton 1972, 2507-2511.
J. S. Bodenheimer, W. Low, Spectrochimica Acta Part A 1973, 29,
1733-1743.
[
23]
functional perturbation-theory. This procedure generates the vibrational
eigenvalues and eigenvectors, which allows visualisation of the modes
within Materials Studio[24] and is also the information needed to calculate
[15]
a) G. L. Hardgrove, D. H. Templeton, Acta Cryst. 1959, 12, 28-32;
b) P. Seiler, J. D. Dunitz, Acta Cryst. 1980, B36, 2946-2950; c) A. O.
Borissova, M. Yu. Antipin, D. S. Perekalin, K. A. Lyssenko,
CrystEngComm 2008, 10, 827–832; d) Y Miyamoto, S. Takamizawa,
Dalton Trans. 2015, 44, 5688–5691.
[
25]
the INS spectrum using the program ACLIMAX. The initial structure for
[
15c]
the optimisation was that determined at 100 K.
This consistently
resulted in one imaginary mode, irrespective of the convergence criteria
used. Allowing both the lattice and the geometry to optimise resulted in a
[16]
[17]
[18]
J. Tomkinson, S. F. Parker, Spectrochimica Acta Part A 2011, 79,
2017-2019.
3
3
structure that was <1% larger (796.3 Å vs 789.8 Å ) and with no imaginary
modes anywhere in the Brillouin zone, Figure S4. An isolated molecule
calculation was generated by taking the optimised structure and removing
three molecules from it. The eigenvectors from this were then used with
the transition energies from Table 1 to generate Figures 3b,c,d and 4c.
M. A. Adams, S.F. Parker, F. Fernandez-Alonso, D.J. Cutler, C.
Hodges, A. King, Appl. Spec. 2009, 63, 727-732.
S. F. Parker, F. Fernandez-Alonso, A. J. Ramirez-Cuesta, J.
Tomkinson, S. Rudic, R. S. Pinna, G. Gorini, J. Fernández Castañon,
J. Phys. Conf. Series 2014, 554, 012003.
[
19]
20]
http://www.isis.stfc.ac.uk/
[
S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. J. Probert, K.
Refson, M. C. Payne, Z. Kristallographie 2005, 220, 567-570.
K. Refson, S. J. Clark, P. R. Tulip, Phys. Rev. B 2006, 73, 155114.
A. Tkatchenko and M. Scheffler, Phys. Rev. Lett. 2009, 102, 073005.
V. Milman, A. Perlov, K. Refson, S. J. Clark, J. Gavartin, B. Winkler. J.
Phys.: Condens. Matter 2009, 21, 485404.
Acknowledgements
[21]
The STFC Rutherford Appleton Laboratory is thanked for access
to neutron beam facilities. Computing resources (time on the
SCARF compute cluster for the CASTEP calculations) were
[22]
23]
[
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
[24]
http://accelrys.com/products/collaborative-science/biovia-materials-
studio/
[25]
A. J. Ramirez-Cuesta, Comp. Phys. Comm. 2004, 157, 226-238.
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201800914
FULL PAPER
Entry for the Table of Contents
FULL PAPER
In this work we report a simplified
synthesis of ruthenocene and the
vibrational spectroscopy of this iconic
molecule. The addition of inelastic
neutron scattering data in combination
with periodic-DFT calculations has
enabled the first correct assignment of
the internal modes of ruthenocene.
Synthesis and spectroscopy of
ruthenocene
Stewart F. Parker, Ian R. Butler
Page No. – Page No.
Synthesis and inelastic neutron
scattering spectroscopy of
ruthenocene
This article is protected by copyright. All rights reserved.