Toward Frameworks with Multiple Aligned and Interactive Fe(CO)3Rotators: Syntheses and Structures of Diiron Complexes Linked by Two trans-Diaxial α,ω-Diphosphine Ligands Ar2P(CH2)nPAr2

Article abstract of DOI:10.1021/acs.inorgchem.0c03737

Reactions of (η4-benzylideneacetone)Fe(CO)3 and the α,ω-diphosphines Ar2P(CH2)nPAr2 afford the trigonal bipyramidal diiron tetraphosphorus complexes trans,trans-(CO)3Fe[Ar2P(CH2)nPAr2]2Fe(CO)3 (n/Ar = 3/Ph 3, 4/Ph 4a, 4/p-tol 4b; 56-19%). Crystal structures establish essentially parallel P-Fe-P axes, iron-iron distances of 5.894(9)-5.782(1) ? (3) and 6.403(1)-6.466(1) ? (4a,b), and van der Waals radii of 4.45 ? for the Fe(CO)3 rotators, the planes of which are offset by 0.029-1.665 ?. Analogous reactions of Ph2P(CH2)6PPh2 yield the square pyramidal monoiron complex trans-(CO)3Fe[Ph2P(CH2)6PPh2] (6′, 31%), a rare case where a diphosphine spans trans basal positions (

Full text of DOI:10.1021/acs.inorgchem.0c03737

pubs.acs.org/IC
Article
Toward Frameworks with Multiple Aligned and Interactive Fe(CO)3
Rotators: Syntheses and Structures of Diiron Complexes Linked by
Two trans-Diaxial α,ω-Diphosphine Ligands Ar2P(CH2)nPAr2
Samuel R. Zarcone, Gong M. Chu, Andreas Ehnbom, Ashley Cardenal, Tobias Fiedler,
Nattamai Bhuvanesh, Michael B. Hall,* and John A. Gladysz*
Cite This: Inorg. Chem. 2021, 60, 3314−3330
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Reactions of (η4-benzylideneacetone)Fe(CO)3 and the α,ω-diphosphines
Ar2P(CH2)nPAr2 afford the trigonal bipyramidal diiron tetraphosphorus complexes
trans,trans-(CO)3Fe[Ar2P(CH2)nPAr2]2Fe(CO)3 (n/Ar = 3/Ph 3, 4/Ph 4a, 4/p-tol 4b;
56−19%). Crystal structures establish essentially parallel P−Fe−P axes, iron−iron
distances of 5.894(9)−5.782(1) Å (3) and 6.403(1)−6.466(1) Å (4a,b), and van der
Waals radii of 4.45 Å for the Fe(CO)3 rotators, the planes of which are offset by 0.029−
1.665 Å. Analogous reactions of Ph2P(CH2)6PPh2 yield the square pyramidal monoiron
complex trans-(CO)3Fe[Ph2P(CH2)6PPh2] (6′, 31%), a rare case where a diphosphine
spans trans basal positions (∠P−Fe−P 147.4(2)°). Both 3 and 6′ exhibit two CO 13C
NMR signals at room temperature, indicating slow exchange on the NMR time scale,
which in the former could entail Fe(CO)3/Fe(CO)3 gearing. Under analogous
conditions, 4a,b exhibit one signal. Previously reported adducts of Fe(CO)3 and
Ph2P(CH2)nPPh2 are surveyed (1:1, n = 1−5; 2:2, n = 5), and the IR νCO band
patterns and energies of all complexes analyzed with the aid of DFT calculations. The diiron complexes are preferred
thermodynamically. Attention is given to limiting types of Fe(CO)3/Fe(CO)3 interactions in the diiron complexes.
our results with II have involved trigonal bipyramidal iron
complexes with the rotators Fe(CO)3, Fe(CO)2(NO)+, and
Fe(CO)(NO)(X) (X = Cl, Br, I, and CN).7 Hence, for initial
exploratory studies, a minimalist approach involving reactions
of α,ω-diphosphines Ar2P(CH2)nPAr2 and Fe(CO)3 sources
was selected, as outlined in Scheme 2. We were particularly
interested in accessing diiron tetraphosphorus species V with n
= 3 and 4. For these systems, molecular models suggested
sterically interactive rotators. Indeed, as described below, this
simple strategy does afford the target complexes.
During the later stages of this work, we became aware of a
similar synthesis of species V with n/Ar = 5/Ph by Liu and
Gau.9 Furthermore, as shown in Scheme 2, reactions of
diphosphines Ar2P(CH2)nPAr2 and Fe(CO)3 sources are also
capable, in principle, of giving monoiron diphosphorus species
such as VI, as well as oligomers or other adducts. Indeed, there
are several previous reports of VI (Ar = Ph) with n ranging
from 1 to 5.10−14 As outlined below, these can exhibit different
INTRODUCTION
■
There has been much recent interest in molecular rotors, a
broad class of compounds that can be dissected into two
components, rotators and stators.1 In particular, assemblies
with multiple rotators that are capable of coupled interactions
hold promise for various types of molecular machines.2−4 Such
phenomena have been most intensively studied in the context
of sterically driven gearing.2,3 Purely electronic rotator/rotator
interactions have also received attention.4
We have been engaged in the synthesis of what we term
gyroscope like transition metal complexes.5−8 These are most
commonly accessed as shown in Scheme 1 (top). Complexes
with two trans-trialkylphosphine ligands of the formula
P((CH2)mCHCH2)3 (I) are first prepared. Then 3-fold
intramolecular ring closing alkene metatheses are carried out.
Subsequent hydrogenations afford target complexes II with
triply trans-spanning dibridgehead diphosphine ligands in
surprisingly high overall yields. Depending upon the sizes of
the ligands on the metal (Ly) and the lengths of the (CH2)n
segments (n = 2m + 2), the MLy moieties in II may (or may
not) be able to rotate within the cage like diphosphine.5−8
As sketched in Scheme 1 (bottom), we sought to extend this
effort to related bimetallic complexes with sterically coupled
MLy rotators, both with (IV) and without (III) protective
trans-spanning polyphosphine ligands. A significant portion of
Received: December 21, 2020
Published: February 11, 2021
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
© 2021 American Chemical Society
Inorg. Chem. 2021, 60, 3314−3330
3314

Inorganic Chemistry
pubs.acs.org/IC
Article
Scheme 1. Syntheses of Gyroscope-Like Complexes (II) and
New Targets with Two MLy Rotators with Parallel Axes (III
and IV)
Scheme 3. Syntheses of New Diiron and Monoiron
Complexes
Scheme 2. Some Possible Products Derived from Equimolar
Amounts of Fe(CO)3 Sources and α,ω-Diphosphines
Ar2P(CH2)nPAr2 (V and VI)
limiting geometries,11 and as a collateral result of this study, a
rare type of isomer has been isolated in the case of n = 6. To
provide additional support for the structural assignments made
herein and in earlier reports,9−14 as well as a road map to guide
future research, the IR νCO patterns are computed by DFT
methods and compared to those found experimentally. These
calculations furthermore yield free energies that show the
equilibria 2VI ⇌ V to be exergonic.
(CO)3Fe[Ph2P(CH2)nPPh2]2Fe(CO)3 (n = 3 (3) and 4 (4a))
as air-sensitive yellow or pale yellow solids in 44 and 19%
yields, respectively.
These and all other new complexes below were characterized
by microanalyses, IR spectroscopy (ATR and powder film) and
NMR spectroscopy (1H, 13C{1H}, and 31P{1H}). The IR νCO
bands are depicted in Figure 1 and summarized in Table 1.
The remaining data are given in the Experimental Section.
Both 3 and 4a were soluble in toluene, THF, and DMSO and
insoluble in hexane. Although 3 was insoluble in CH2Cl2, 4a
was soluble in both CH2Cl2 and CDCl3.
RESULTS
■
Syntheses of Title Complexes. The η4-benzylideneace-
15
tone complex (BDA)Fe(CO)3 has seen extensive use as a
functional equivalent of the Fe(CO)3 fragment.7a,15,16 The
phenyl substituted α,ω-diphosphines Ph2P(CH2)nPPh2 (n = 3,
4, and 6) were obtained from commercial sources, and the p-
tolyl analog (p-tol)2P(CH2)4P(p-tol)2 was isolated in 47%
yield from the reaction of the phosphido anion (p-tol)2P− Li+
and the corresponding α,ω-dibromide. This 1,4-diphosphine
had been previously synthesized from Cl2P(CH2)4PCl2 and the
Grignard reagent p-tolMgBr.17
Accordingly, (BDA)Fe(CO)3 was treated with 1.2 equiv of
Ph2P(CH2)3PPh2 (1,3-bis(diphenylphosphino)propane) or
Ph2P(CH2)4PPh2 (1,4-bis(diphenylphosphino)butane) in
THF. After 48 h, the solvent was removed. As shown in
Scheme 3, workups gave the target complexes trans,trans-
The spectroscopic properties of 3 and 4a were consistent
with the proposed structures. In particular, the IR spectra
showed one strong νCO band (1871 or 1864 cm−1), and a
second much weaker band at higher frequency (1969 or 1964
cm−1). As can be derived from group theory,18 this pattern is
characteristic of axially disubstituted trigonal bipyramidal
complexes of the formula trans-Fe(CO)3L2. However, there
are many literature reports (L = phosphine) in which the
weaker band is missing or has possibly been overlooked.7a,16b
This includes the higher homologue derived from 1,5-
bis(diphenylphosphino)pentane, trans,trans-(CO)3Fe[Ph2P-
(CH2)5PPh2]2Fe(CO)3 (5; 1877 cm−1),9 which is depicted
in Scheme 3 (bottom; see the Introduction).19
3315
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Table 1. Experimental and Computed IR νCO Bands for
Monoiron (1′−6′) and Diiron (3, 4a,b, 5, and 6) Complexes
νCO (cm−1
)
νCO (cm−1
)
a
b
complex
n
geometry
experimental
medium
computed
c
d
1′
1
sq pyr cis 1984 s, 1911 m, CH2Cl2
1968 s, 1915 s,
1911 s
1970 s, 1919 m,
1901 s
1969 s, 1910 m,
1890 s
1966 s, 1907 m,
1882 s
1901 s
e
d
2′
2
3
4
4
5
6
sq pyr cis 1982 s, 1913 m, CH2Cl2
1892 s
f
3′
tbp cis
tbp cis
tbp cis
tbp cis
1982 s, 1909 m, CH2Cl2
1881 s
1981 s, 1908 m, CH2Cl2
1879 s
g
4′a
h
4′b
1964 s, 1904 m,
1880 s
i
5′
1983 s, 1912 s, CH2Cl2
1883 s
1964 s, 1906 m,
1882 s
j
k
6′
sq pyr
trans
1965 w, 1863 s powder
1941 w, 1892
film
m, 1865 s
j
k
3
3
4
4
5
6
tbp trans
tbp trans
tbp trans
tbp trans
tbp trans
1969 w, 1892
sh, 1871 s
1964 w, 1864 s powder
powder
1939 w, 1889
Figure 1. IR spectra (ATR/powder film, νCO region) of the new
diiron complexes V with principal features highlighted.
film
sh, 1876 s
j
k
4a
1933 w, 1871
film
powder
film
sh, 1867 s
j
k
4b
1965 w, 1889
sh, 1870 s
1877 s
1930 w, 1871
sh, 1862 s
Of course, the diiron complexes 3 and 4a and the trigonal
bipyramidal monoiron complexes trans-Fe(CO)3L2 have
different molecular symmetries, and this will affect the
vibrational spectra. Also, a shoulder was evident on the strong
νCO band of 3 (1892 cm−1; Figure 1). Thus, the IR νCO
bands of all iron carbonyl complexes prepared or discussed in
this paper have been computed using DFT, as summarized in
Table 1 and described and analyzed below.
The NMR spectra also supported the structures of 3 and 4a.
With 3, two CO 13C{1H} signals could be observed (214.7 and
212.4 ppm, both apparent m), consistent with sterically
restricted Fe(CO)3 rotators. To our knowledge, all monoiron
diphosphorus adducts VI (Scheme 2) examined to date afford
a single CO signal at ambient probe temperatures. In the case
of 4a, only a very weak CO signal (214.0 ppm) could be
tentatively assigned. The 31P{1H} NMR chemical shifts of 3
and 4a (75.12 and 73.96 ppm in DMSO-d6) were also quite
close to that of the analogous Ph2P(CH2)5PPh2 adduct 5
(75.53 ppm in CDCl3).9 The signal of the corresponding
i
k
5
CH2Cl2
1935 w, 1873
sh, 1865 s
h
k
6
1960 w, 1897
s, 1893 s
a
These correspond to the crystal structures for 1′−3′, 5′, 6′, 3, 4a,b,
and 5 and are replicated by the computed structures in all cases.
b
c
Determined by DFT as described in the text. The experimental
values are from ref 14b; similar data have been reported in refs 14a
(1989 s, 1919 ms, 1909 s; CHCl3) and 14c (1989, 1913, 1907; THF,
no intensity information). These complexes exhibit intermediate sq
pyr/tbp geometries but are considered predominantly sq pyr; see refs
10−12. The experimental values are from ref 14b; similar data have
been reported in ref 14c (1984, 1914, 1888; THF, no intensity
d
e
f
information). The experimental values are from ref 14b; similar data
have been reported in refs 13 (1983 s, 1908 ms, 1882 s; CH2Cl2/0.1
M Bu4NPF6), 14c (1984, 1914, 1888; THF, no intensity
information), and 14d (1986 s, 1914 ms, 1984 vs; CH2Cl2 or 1980
s, 1908 s, 1885 vs; Nujol). The experimental values are from ref 14b;
similar data have been reported in ref 14c (1983, 1912, 1886; THF,
no intensity information). This complex has not yet been crystallo-
g
h
monoiron adduct cis-(CO)3Fe[Ph2P(CH2)5PPh2] (5′)20 is
graphically characterized. This complex has not yet been synthesized.
i
j
k
The experimental values are from ref 9. This work. These
much further upfield (55.46 ppm).
absorptions are predicted to be very weak or undetectable.
The preceding reactions were extended in two ways. First,
with the idea of procuring more soluble complexes with
additional NMR “handles”, (BDA)Fe(CO)3 was similarly
combined with the p-tolyl substituted diphosphine (p-tol)2P-
(CH2)4P(p-tol)2 (0.8 equiv; Scheme 3). Workup gave the
target complex trans,trans-(CO)3Fe[(p-tol)2P(CH2)4P(p-
tol)2]2Fe(CO)3 (4b) in 56% yield, which was soluble in the
same solvents as 4a and partially soluble in hexane. The IR
spectrum showed two main νCO bands analogous to those of
3 and 4a (1870 s and 1965 w cm−1), and a shoulder as noted
with 3 (1889 cm−1; Figure 1 and Table 1). A high-resolution
mass spectrum (ESI) gave an envelope of intense ions with the
masses and isotope distribution expected for 4b + H. The
13C{1H} NMR spectrum also showed a single CO signal that
was coupled to phosphorus (220.7 ppm, t, JCP = 8.7 Hz),
consistent with rapid Fe(CO)3 rotation on the NMR time
scale. Although 1H NMR spectra suggested a purity of ca. 90%,
a crystalline sample was easily obtained as described below.
Second, (BDA)Fe(CO)3 was similarly combined with the
hexamethylene diphosphine Ph2P(CH2)6PPh2 (1.2 equiv) in
THF (Scheme 3). A chromatographic workup of the red
solution gave a yellow solid (31%) that was initially thought to
be the diiron complex trans,trans-(CO)3Fe[Ph2P-
(CH2)6PPh2]2Fe(CO)3 (6). However, a high-resolution mass
spectrum (ESI) suggested a monoiron diphosphorus species.
The IR spectrum showed a strong νCO band at 1863 cm−1
and a second weak one at 1965 cm−1. Also, the 13C{1H} NMR
spectrum exhibited two CO signals (ca. 2:1). These data were
best reconciled with the structural assignment trans-(CO)3Fe-
[Ph2P(CH2)6PPh2] (6′),20 as further supported in the
following section.
The isolation of 6′ prompted us to check whether related
monoiron species (3′, 4′a, and 4′b) might have been
generated under the conditions used to access 3 and 4a,b.
As further detailed in the Experimental Section, species with
plausible 31P{1H} NMR chemical shifts for such adducts were
sometimes observed in the crude reaction mixtures but were
3316
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Table 2. General Crystallographic Data
a
3·(CH2Cl2)4
C64H60Cl8Fe2O6P4
1444.30
110(2)
4a·(toluene)6
C104H104Fe2O6P4
1685.45
110(2)
4b
6′·(toluene)0.5
C36.5H36FeO3P2
640.44
empirical formula
formula weight
temperature [K]
diffractometer
wavelength [Å]
crystal system
space group
unit cell dimensions
a [Å]
C70H72Fe2O6P4
1244.86
110(2)
120(2)
Bruker Gadds
1.5418
triclinic
Bruker Gadds
1.5418
triclinic
Bruker Quest
0.71073
monoclinic
P21/c
Bruker Quest
0.71073
monoclinic
P21/n
P1
̅
P1
̅
12.5070(5)
12.5333(5)
22.0792(8)
88.807(3)
73.868(2)
77.784(2)
3246.8(2)
2
1.477
11.2478(5)
15.2495(6)
26.8362(12)
106.296(3)
99.572(3)
94.533(3)
3246.8(2)
2
12.2470(16)
15.299(2)
22.763(3)
90
94.774(2)
90
4250.2(10)
2
10.1433(5)
29.2030(14)
10.7720(5)
90
92.699(10)
90
3187.3(3)
4
1.335
b [Å]
c [Å]
α [deg]
β [deg]
γ [deg]
V [Å3]
Z
ρ calc [Mg/m3]
1.296
0.973
μ [mm−1
F(000)
]
7.950
1480
3.833
1776
0.455
1304
0.609
1340
crystal size [mm3]
0.11 × 0.07 × 0.04
2.08−60.00
−13, 14; −4, 14; −24, 24
76986
9463 [R(int) = 0.0701]
0.7416 and 0.4751
9463/3/765
1.031
0.05 × 0.04 × 0.02
1.749−60.968
−12, 12; −17, 17; −30, 30
92070
12925 [R(int) = 0.0701]
0.7227 and 0.5462
12925/346/976
1.106
0.10 × 0.08 × 0.05
2.261−22.545
−13, 13; −16, 16; −24, 24
60697
5547 [R(int) = 0.0488]
0.4267 and 0.3776
5547/0/374
1.074
0.20 × 0.18 × 0.11
2.790−28.428
−13, 13; −39, 38; −14, 14
43668
7998 [R(int) = 0.0372]
0.4311 and 0.3852
7998/456/481
1.120
Θ limit [deg]
index range (h, k, l)
reflections collected
independent reflections
max. and min. transmission
data/restraints/parameters
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R1
0.0606
0.1508
0.0945
0.1772
0.0998
0.2644
0.0331
0.0674
wR2
R indices (all data)
R1
0.0691
0.1262
0.1055
0.0470
wR2
0.1563
1.881 and −1.073
0.1937
0.707 and −0.675
0.2683
1.025 and −0.661
0.0766
0.397 and −0.342
largest diff. peak/hole [e Å−3
a
]
The crystal contained disordered solvent molecules, which were treated as described in the Experimental Section. As a result, the density is
underestimated.
removed upon the prescribed workup. When concentrations
were varied, these sometimes became the major products.
Structural Data. Crystals of solvates of 3, 4a, 4b, and 6′
could be grown, and the X-ray structures were determined as
outlined in Table 2 and the Experimental Section. Thermal
ellipsoid plots are depicted in Figures 2 and 3. Key metrical
parameters are listed in Tables 3 and 4. The latter emphasizes
features relevant to nonbonded interactions in the diiron
complexes, and includes the previously published structure for
59 and DFT results for 6 described below. These two
structures are depicted in Figure 4. For all complexes, the
atoms have been renumbered from those in the CCDC files to
achieve a uniform pattern that facilitates data comparison.
For both 3 and 4a, two independent molecules were found
in the unit cell, each with an inversion center. These are
denoted as 3(1), 3(2), 4a(1), and 4a(2) in the graphics and
tables. The structure of 4b also exhibits an inversion center,
and 5 features a C2 axis that is perpendicular to the Fe2P4
plane. The radii of the Fe(CO)3 rotators fall into a narrow
range centered at 4.45 Å (Table 4), as computed by taking the
average iron−oxygen distance and adding the van der Waals
radius of an oxygen atom (1.52 Å).21 Unsurprisingly, these
closely agree with values found earlier.7a
Measures of the relationships between the two P−Fe−P axes
of the diiron complexes were sought. Three nearly collinear
points, each with independent experimental error, do not very
accurately define a plane. Thus, least-squares planes derived
from each P−Fe−P axis and the distal iron atom were
calculated (P−Fe−P/Fe′). As summarized in Table 4, the
complexes with inversion centers (3 and 4a,b) gave plane/
plane angles of 0°, indicating rigorously parallel axes. The
others (5 and 6) gave plane/plane angles of 7.6−12.0°. The
relationships between the axes are also reflected by the four P−
Fe−Fe′−P torsion angles, which as noted in Table 4 are always
close to 0° or 180°.
Measures of the relationships between the planes of the
rotators were also sought. For this purpose, the four-atom
least-squares Fe(C)3 planes were used (as opposed to the
seven-atom Fe(CO)3 planes). The inversion centers in 3 and
4a,b render the two Fe(C)3 planes as parallel, so it is a simple
matter to measure the offsets. In both independent molecules
of 3, the two planes are nearly coplanar (miniscule offsets of
0.087 and 0.029 Å). In the two independent molecules of 4a,
the separations increase significantly and are more highly
differentiated (0.985 and 1.217 Å). In 4b, the offset further
3317
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 2. Thermal ellipsoid plots (50% probability) of the molecular structures of 3 (two independent molecules (1 and 2)), 4a (two independent
molecules (1 and 2)), and 4b. Solvent molecules in the lattices have been omitted.
3318
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
on the potential for Fe(CO)3/Fe(CO)3 interactions, which are
further analyzed in the Discussion.
The IR spectrum of crystalline monoiron complex 6′ was
identical with that isolated from Scheme 3. The P−Fe−P bond
angle, 147.4(2)° (Table 3 and Figure 3), represents a dramatic
reduction from the 180° expected for an idealized trigonal
bipyramid with trans-diaxial phosphine ligands, and the values
found for 3 and 4a,b (172.8(1)−178.4(1)°). Thus, 6′ is best
regarded as trans square pyramidal, with the carbonyl ligand
that extends to the left in Figure 3 representing the apical
position. These relationships are sketched in Scheme 3.
Accordingly, the OC−Fe−CO bond angles are no longer ca.
120° as in the other complexes, but rather 151.8(1)° (for the
two basal CO ligands) or 104.7(1)−103.5(1)° (for the apical
CO ligand). The same conclusions are reached using other
geometric features.11
In 6′, the iron−carbon distances for the two CH2 carbon
atoms closest to the plane of the Fe(CO)3 rotator (see Figure
3) are 3.939 and 4.518 Å. When the van der Waals radius of an
sp3 carbon atom21 is subtracted, “clearances” of 2.24−2.82 Å
are obtained. Since these values are well within the radius of
the Fe(CO)3 rotator (4.45 Å), a 360° rotation should be
impossible (although oscillations of a few degrees may take
place).
DFT Computations. Given the established relationship
between IR νCO band patterns and structure,18 and the
various ways in which experimental data might be misleading
(impurities, mixtures of isomers, crystal vs solution structures,
etc.), DFT was utilized to calculate the absorptions for all of
the preceding diiron and monoiron complexes. Techniques
and functionals that have proved reliable in past studies were
employed,13,22 as detailed further in the Experimental Section
and the Supporting Information. In all cases, initial energy
minimizations afforded structures very close to those obtained
crystallographically, as illustrated in Figures S5 and S6. The IR
results are summarized in Table 1 and Figure 6. The IR
spectrum of 3′ has been similarly computed in an earlier
study.13
Figure 3. Thermal ellipsoid plots (50% probability) of the molecular
structure of 6′ (left) and a relevant complex from the literature, 6″
(right; see the Discussion). Solvent molecules in the lattices have
been omitted.
increases (1.677 Å). These relationships are illustrated in
Figure 5.
In 5 and 6, the Fe(C)3 planes are no longer parallel;
therefore, the rotators do not have a uniform separation. We
have not been able to devise a meaningful geometric parameter
that quantifies the offset. However, as can be “eyeballed” in
Figure 5, the rotators are more closely spaced than those in
4a,b. The situation with the two P−Fe−P axes is similar, for as
noted above these are rigorously parallel for 3 and 4a,b but not
5 and 6. Accordingly, fixed distances can be assigned with the
former group but not the latter.
As illustrated in Figure 5, the iron−iron distances in 3 are
5.894(9)−5.782(1) Å (independent molecules 1 and 2),
whereas those in 4a,b (6.466(1)−6.403(1) Å), 5 (7.601(3) Å),
and 6 (8.517 Å) are progressively longer, reflecting the
increasing (CH2)n tether lengths. However, it should be
emphasized that due to the offset of the Fe(C)3 planes, the
distances between the P−Fe−P axes are shorter. For
complexes 3 and 4a,b where this can be quantified, the trend
is not monotonic. Rather, 3 (n = 3) and 4b (n = 4) give
essentially identical values (∼5.8 Å). All of these factors bear
As shown in Figure 6 (bottom), the title complexes 3, 4a,b,
5,9 and 6 (only characterized computationally given the
Table 3. Key Interatomic Distances (Å) and Bond Angles (deg) for Crystallographically Characterized Complexes
ab
,
ab
,
a c
,
3·(CH2Cl2)4
4a·(toluene)6
4b
6′·(toluene)0.5
Fe1−P1
Fe1−P2
Fe1−C1
Fe1−C2
Fe1−C3
C1−O1
C2−O2
C3−O3
Fe1−O1
Fe1−O2
Fe1−O3
2.211(1)/2.200(1)
2.211(2)/2.212(1)
1.772(5)/1.778(6)
1.773(4)/1.763(6)
1.776(4)/1.774(4)
1.160(6)/1.148(7)
1.160(5)/1.164(7)
1.157(5)/1.157(5)
2.931(4)/2.926(4)
2.933(3)/2.926(5)
2.933(3)/2.931(3)
2.213(2)/2.214(2)
2.209(2)/2.207(2)
1.781(8)/1.760(8)
1.752(8)/1.781(7)
1.767(7)/1.765(6)
1.156(8)/1.170(9)
1.166(8)/1.163(8)
1.163(7)/1.162(7)
2.938(6)/2.930(5)
2.919(6)/2.944(5)
2.931(4)/2.927(4)
2.209(2)
2.208(2)
1.784(9)
1.765(9)
1.784(9)
1.195(1)
1.162(1)
1.144(1)
2.928(6)
2.926(7)
2.933(6)
2.214(1)
2.217(1)
1.769(2)
1.775(2)
1.769(2)
1.155(2)
1.156(2)
1.152(2)
2.933(1)
2.931(1)
2.921(1)
C1−Fe1−C2
C1−Fe1−C3
C2−Fe1−C3
P1−Fe1−P2
115.7(2)/118.7(2)
125.0(2)/122.1(2)
119.3(2)/119.2(2)
174.6(1)/178.4(1)
122.3(3)/122.0(3)
118.3(3)/117.7(3)
119.4(3)/120.3(3)
172.8(1)/174.3(1)
122.2(4)
120.1(4)
117.7(4)
176.8(1)
151.8(1)
103.5(1)
104.7(1)
147.4(2)
a
b
The diiron complexes feature inversion centers that exchange primed and unprimed atoms (see Figure 2). Values are given for each of the two
c
independent molecules (1,2) in the unit cell. See Experimental Section for information on solvent in the lattice.
3319
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
a
Table 4. Additional Structural Features of Crystallographically or Computationally Characterized Diiron Complexes
b c
,
b c
,
bd
,
e
complex (method)
3·(CH2Cl2)4 (XRD)
4a·(toluene)6 (XRD)
4b (XRD)
5·(CH2Cl2)2 (XRD)
6 (DFT)
4.47
8.517
4.455
8.452
5.531
7.738
f
radius Fe(CO)3 rotator
4.45/4.45
4.45/4.45
4.45
4.45
g
Fe1···Fe1′
5.894(9)/5.782(1)
4.417(2)/4.412(2)
5.703(1)/5.722(2)
3.308(5)/3.007(5)
5.284(5)/5.954(5)
6.403(1)/6.466(1)
4.414(2)/4.416(2)
6.597(2)/6.599(2)
3.677(8)/3.610(7)
4.546(7)/5.285(7)
6.151(2)
4.415(3)
6.128(3)
3.432(1)
6.531(9)
7.601(3)
4.422(3)
7.539(4)
6.723(1)
6.913(1)
g
P1···P2
P1···P2′
g
g
h
O1···O3′
O3···O3′
g
h
i
P1−Fe1−P2/Fe1′ vs P1′−Fe1′−P2′/Fe1
0.00/0.00
0.00/0.00
0.00
7.67
12.00
j
P2−Fe1−Fe1′−P1′
P2−Fe1−Fe1′−P2′
P1−Fe1−Fe1′−P2′
2.1/−0.2
180.0/−180.0
−2.1/0.2
5.3/4.7
−3.3
180.0
3.3
−7.7
171.0
−7.7
173.7
−10.8
167.8
−13.3
168.2
j
−180.0/180.0
−5.3/−4.7
−180.0/180.0
j
j
P1′−Fe1′−Fe1−P1
−180.0/180.0
−180.0
j
Fe1−P2−C4−C5
P2−C4−C5−C6
C4−C5−C6−C7
C5−C6−C7−C8
57.4/−42.3
−177.6/166.3
−70.3/−70.0
111.9/104.8
178.3/−179.7
80.1
−136.2
74.1
60.3
173.9
66.3
63.4/68.6
159.2/−169.7
65.2/67.5
76.0/170.5
63.2/−63.0
155.31/−137.2
64.42/74.05
j
j
j
164.5
j
C6−C7−C8−C9
jk
Cω−2−Cω−1−Cω−P1′ ,
162.8/−175.3
−46.2/50.9
−174.6/−167.1
−70.5/−71.6
163.1
−175.1
−174.9
54.5
jl
Cω−1−Cω−P1′−Fe1′ ,
m
P1−Fe1−P2/P1′−Fe1′−P2′ axis/axis distance
5.884/5.780
6.166/6.400
0.980/1.209
5.853
n
n
n
m
offset of rotators Fe1(C)3/Fe1′(C′)3
0.087/0.029
1.665
n
b
a
Quantities not available from the crystallographic output were calculated using standard programs such as Mercury. These complexes feature
inversion centers that exchange primed and unprimed atoms (see Figure 2). Values are given for each of the two independent molecules (1, 2) in
c
d
e
f
the unit cell. See the Experimental Section for information on solvent in the lattice. From ref 9. Calculated by averaging the Fe−O distances and
g
h
adding the van der Waals radius of oxygen (1.52 Å). Nonbonded distance (Å). These values are for O1···O1′ and O3···O1, respectively (the
i
crystallographic C2 symmetry of this compound necessitates minor changes in atom numbering). Angle between two four-atom least-squares
j
k
l
planes (deg). Torsion angle (deg). This torsion angle is the counterpart to P2−C4−C5−C6; ω = 6, 7, 7, 8, and 9 for the respective entries. This
m
torsion angle is the counterpart to Fe1−P2−C4−C5; ω = 6, 7, 7, 8, and 9 for the respective entries. Due to the inversion centers in 3·(CH2Cl2)4,
n
4a·(toluene)6, and 4b, the two axes or planes are parallel with a constant intervening distance (Å). Due to the absence of inversion centers in 5
and 6, the P−Fe−P axes and Fe(C)3 planes are not parallel, and unique separations cannot be assigned. Figure 5 allows visual estimates.
Figure 4. Thermal ellipsoid plots (50% probability) of the previously reported molecular structure of 5 (top)9 and a plot of the computed structure
for the yet unknown higher homologue 6 (bottom). The primed atoms in the latter (employed for uniformity with the other structures) are
misleading as there are no symmetry elements.
3320
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 5. Space-filling side views of the Fe(CO)3 rotators in 3 (two independent molecules (1 and 2)), 4a (two independent molecules (1 and 2)),
and 4b, 5, and 6 (computed structure) illustrating the offset between the two Fe(CO)3 planes and iron−iron and (when constant) P−Fe−P axis−
axis distances.
isolation of 6′) are predicted to exhibit three vibrational modes
(a−c), each with the Fe(CO)3 units strongly coupled. The
lowest frequency band (corresponding to c) is always the most
intense, and shifts randomly within a 26 cm−1 range as n
increases from 3 to 6 (1876, 1867/1862, 1865, 1893 cm−1).
These bands are furthermore predicted to exhibit shoulders at
slightly higher frequencies (4−13 cm−1, corresponding to b).
However, in computations the peaks can be narrowed or
broadened per the line width selected, as illustrated in Figure 6
(compare the inset with 8 cm−1 line widths to the other spectra
with 4 cm−1 line widths). Thus, these shoulders may or may
not be visible experimentally (they are seen for 3 and 4b).
There is no dipole moment change associated with the third
highest frequency vibrational mode (a). Hence, these are
formally IR-inactive. Nonetheless, weak absorptions are
observed experimentally for 3 and 4a,b (1964−1969 cm−1).
Next, we focus on the spectra of monoiron complexes 3′,
4′a, 5′, and 6′ depicted in Figure 6 (top). These feature
different coordination geometries as summarized in Table 1
and represented in Figure 6 as VI′ (tbp cis), VI″ (sq pyr cis),
and VI′′′ (sq pyr trans). Complex 6′, new to this study and
possessing the rare geometry VI′′′, is predicted to exhibit three
IR νCO bands. However, the two lowest frequency bands
(1865 s, 1892 m cm−1) apparently merge into a single band
experimentally (1863 cm−1). In any case, the weak highest
frequency band is reproduced (1941 vs 1965 cm−1),
is that an impurity of this type should be apparent in the IR
spectrum of a diiron complex. For example, the highest
frequency bands are intense in the monoiron complexes
(1984−1981 cm−1) but very weak in the diiron complexes
(1969−1965 cm−1).
In the course of computing the IR spectra, total energies
corresponding to the most stable form of each iron complex
were obtained (both gas and solution phase). These allow the
free energies for the monoiron/diiron complex equilibria
shown in Scheme 4 to be calculated. Interestingly, the diiron
complexes are always the more stable, at least for the range of n
examined.
DISCUSSION
■
Syntheses. The preceding results, together with literature
data, indicate that the Fe(CO)3 source and conditions
associated with Scheme 2 have a profound influence upon
the type of product obtained. In our first cycle of experiments
with the diphosphines Ar2P(CH2)nPAr2, only the target diiron
tetraphosphorus complexes (V) were detected. However, as
these reactions were investigated by additional co-workers,
with the usual minor variances in conditions, it became
apparent that monoiron diphosphorus complexes VI could also
form. From a different direction, several other groups have
reported reactions of Fe(CO)3 sources and the diphosphine
Ph2P(CH2)3PPh2.13,14b−d In contrast to our results, only the
monoiron complex 3′ was isolated.
The IR νCO bands computed for the remaining monoiron
complexes are in excellent agreement with those found
experimentally. Furthermore, coordination modes VI′ and
VI′′ give quite similar patterns, with three bands of strong to
medium intensities. However, the most important conclusion
So far, only Liu and Goh, in their studies with Ph2P-
(CH2)5PPh2, have been able to optimize conditions for either
type of product.9 In refluxing THF, the maleic acid (MA)
adduct (MA)Fe(CO)4 and the amine oxide Me3N+−O− afford
3321
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 6. Computed IR νCO bands for monoiron and diiron carbonyl complexes.
3322
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
conformations as well as others. The FeP(CH2)nPFe
conformations of 5 and 6 are not analyzed in similar detail,
but a casual inspection of Figure 4 shows increasingly complex
torsion angle patterns that at some value of n would promote
P−Fe−P axes that are distinctly nonparallel.
Scheme 4. Computed ΔG Values for the Hypothetical
Equilibrium between Monoiron and Diiron Complexes
With regard to steric interactions between the Fe(CO)3
rotators in 3, 4a,b, 5, and 6, the starting point for analysis is the
van der Waals radii of the rotators (4.45 Å; Table 4). In order
to ensure that two parallel and coplanar rotators never
sterically interact, the P−Fe−P axes must be spaced by twice
this amount, or 8.90 Å. Of all of the diiron complexes
considered in this study, only 6, with an iron−iron distance of
8.517 Å, approaches this limit.
Top views of the Fe(CO)3 rotators in 3 and 4a,b are given
in Figure 7. Interestingly, only in one case are there van der
Waals contacts. In viewing such projections, it should be kept
in mind that per Figure 5, none of the neighboring Fe(CO)3
units are strictly coplanar with respect to each other. Thus, as
noted above, the distances between the parallel P−Fe−P axes
are shorter than the iron−iron distances.
Next, we consider the arbitrary Fe(CO)3/Fe(CO)3
orientation in panel A of Figure 8, which has no crystallo-
graphic counterpart in Figure 7. This can be viewed as a head-
to-tail arrangement of two rotators with respect to two Fe−CO
vectors. In these analyses, the Fe(CO)3 units are constrained
to be coplanar, so the iron−iron and P−Fe−P axis separations
are identical. These distances are reduced until there is van der
Waals contact between the two rotators, which occurs at an
iron−iron separation of 6.67 Å. Interestingly, the approach is
limited not by the van der Waals radii of the iron atom or
carbonyl oxygen, but rather the carbonyl carbon (see inset).
In panel B of Figure 8, the iron−iron distance in panel A is
maintained, but the Fe(CO)3 rotator on the left is rotated by
the diiron complex 5 in 45% yield, as depicted in Scheme 3
(bottom). In contrast, the cyclooctene (COE) adduct
(COE)2Fe(CO)3 gives the monoiron product 5′ in 93%
yield (−60 °C to RT, THF). There is presently no rationale for
this dichotomy. The higher temperature reaction affords the
more stable product, 5, according to the computational data in
Scheme 4. However, 5′ is recovered unchanged after 2 days
when subjected to these conditions.9
Steric Interactions in Diiron Complexes. With regard to
the main objectives of this study, the most important structural
features are those that affect correlated Fe(CO)3/Fe(CO)3
rotation or gearing. Thus, it is encouraging to see parallel or
nearly parallel P−Fe−P axes maintained throughout the series
3, 4a,b, 5, and 6 in Figures 2 and 4. However, as the lengths of
the (CH2)n tethers increase, potential structural motifs
multiply.
For example, conformations with nonparallel axes or twisted
endgroups are likely to become increasingly accessible.
Alternatively, the planes of the Fe(CO)3 rotators can become
increasingly offset, such that they no longer efficiently engage.
In this study, the maximum separation is found with 4b (1.67
Å; Table 4 and Figure 5), which is comparable to the van der
Waals radii of carbon and oxygen atoms (1.70 Å, 1.52 Å).
When the offset reaches the van der Waals diameters of the CO
ligands, the two rotators can no longer sterically engage.
The offsets have their origins in the conformations of the
FeP(CH2)nPFe segments. In both independent molecules of 3,
the two FePCC units have gauche conformations (torsion
angles 42−57°; see Table 4), and the two PCCC units have
anti conformations (torsion angles 178−166°), as easily
visualized in Figure 2. Alternative gauche/anti patterns are
highly unlikely. However, the longer FeP(CH2)4PFe segments
in 4a,b allow two different motifs. In both molecules of 4a, the
two FePCC units remain gauche (torsion angles −70° to
−72°). The CCCC units and one PCCC unit are anti ( 167−
180°), whereas the other PCCC unit adopts an intermediate
conformation (105−112°). In contrast, with 4b, one FePCC
unit becomes anti (−175°), while the other remains gauche
(80°). The former catapults a methylene group much further
from the rotator plane (Figure 2, bottom), rendering a greater
offset more likely. The adjacent PCCC unit is approximately
anti (−136°), and to close the ring the CCCC unit must adopt
a gauche conformation (74°).
In summary, the anti FePCC unit in 4b can be viewed as
introducing a “stair step” that increases the likelihood of
significantly offset Fe(CO)3 rotators. In solution, one could
logically expect 4a and 4b to sample both types of
Figure 7. Space-filling “top” views of the Fe(CO)3 rotators of 3 (top,
both independent molecules), 4a (middle, both independent
molecules), and 4b (bottom) at van der Waals radii.
3323
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 8. Analyses of interactions in hypothetical conformations of complexes with two coplanar Fe(CO)3 rotators (see text).
Figure 9. Minimum iron/iron spacing required (6.79 Å) for rotation of a Fe(CO)3 rotator with no van der Waals (vdW) interactions with a
neighboring Fe(CO)3 rotator.
180° to give a tail-to-tail orientation of the two Fe−CO
vectors, as seen in the crystal structures of 3(2) and 4b. This
opens up some void space between the rotators, which at the
narrowest is 0.64 Å (CO···OC). As shown in panel C, a variant
is then considered in which both rotators have been rotated
clockwise by 25°, approximating other orientations in Figure 7
(3(1), 4a(1), and 4a(2)). Now the void space narrows to 0.73
Å (CO···OC) and 0.65 Å (CO···C).
Next, the Fe(CO)3 conformations in panels B and C were
maintained but the iron−iron distances were compressed until
the onset of van der Waals contacts. As shown in panels D and
E, separations of 6.03−6.01 Å could be realized. We suggest
that these represent realistic “minimum contacts” for coplanar
Fe(CO)3 rotators. Of course, at such distances gearing would
likely be frozen. It is more challenging to define an iron−iron
distance at which gearing would be optimal.
3324
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Scheme 5. A Triphosphine That Templates the Synthesis of Trinickel Complexes with Three Approximately Parallel Rotators
One approach is to define additional limiting values. For
example, what is the minimum iron−iron distance at which
one Fe(CO)3 rotator can experience sterically barrierless
rotation (i.e., no van der Waals contacts) with respect to the
other? This is addressed in Figure 9, where the Fe(CO)3 on
the right (Fe2) is kept static, with the CO ligands as far
removed from the Fe(CO)3 on the left (Fe1) as possible. For
example, rotating Fe2 slightly clockwise or counterclockwise
from the orientation depicted would force one of the two CO
oxygen atoms directed toward Fe1 still closer to Fe1.
trigonal bipyramidal adducts. Indeed, α,ω-diphosphines Ph2P-
(CH2)nPPh2 have also been used to template the synthesis of
analogous square planar adducts trans,trans-L′LM[Ar2P-
(CH2)nPAr2]2MLL′, for example, with M/n = Rh/1−4.26
There is also grounds for cautious optimism with regard to
extending this synthetic strategy to arrays with increased
numbers of rotators. As shown in Scheme 5, combination of
the triphosphine CH3C(CH2CH2PPh2)3 (8) with NiCl2
affords the 2:3 trinickel hexaphosphorus adduct 9, with three
roughly parallel P−Ni−P axes.27 Here the phosphorus atoms
are connected by five-carbon bridges. In principle, the degree
of steric communication between the rotators could be fine-
tuned by adjusting this spacing or varying the chloride ligands.
However, as detailed in the preceding analyses, the
conformational flexibility of these types of diphosphine or
triphosphine ligands present challenges with regard to precisely
controlling rotator/rotator interactions. As illustrated by 4a
and 4b in Figures 2 and 5, even when the (CH2)n tethers are of
equal lengths, the iron−iron distances, P−Fe−P axis
separations, and distances between the planes of the two
rotators can vary considerably. A second-generation approach
would involve more rigid diphosphines with fewer degrees of
freedom, as might be realized with a cyclohexane containing
ligand with two 1,3-diaxial PAr2 moieties. Such building blocks
would seemingly enforce coplanar Fe(CO)3 rotors and parallel
P−Fe−P axes.
Accordingly, Fe1 and the proximal CO oxygen atom from
Fe2 must have a steric clearance that matches the sum of the
radius of the Fe(CO)3 rotator and the van der Waals radius of
the CO oxygen atom, 5.97 Å (4.45 + 1.52 Å). Applying
standard trigonometric relationships, this corresponds to a
Fe1−Fe2 separation of 6.79 Å. Said differently, this is the
closest iron−iron approach for which one Fe(CO)3 can rotate
independently of the other (although van der Waals contacts
will ensue with any other Fe2 conformation). One take-home
message from Figures 8 and 9 is that the “sweet spot” for
gearing involves a narrow range of iron−iron distances,
perhaps on the order of 6.70−6.50 Å. Other approaches to
defining optimal spacings for gears derived from trigonal
rotators have been described by Baldridge and Siegal, who
have defined concepts such as gearing fidelity.3a They also
emphasize issues with treating atoms as “hard spheres”,
possible roles of dispersion forces, and a host of other factors.
Related Complexes and Directions. Several attributes of
new square pyramidal monoiron complex 6′ also merit note.
First, it features the shortest methylene chain capable of
bridging two trans phosphorus donor atoms for any
coordination geometry.23,24 Interestingly, a similar complex,
6″, with a diphosphine derived from 2,2′-dimethylbiphenyl has
been reported by Casey, the crystal structure of which is shown
in Figure 3 (right).25 As with 6′, six carbon atoms tether the
trans phosphorus donor atoms, but most are sp2 hybridized.
The P−Fe−P and basal CO−Fe−CO bond angles in 6″
(152.0(1) and 141.8(4)°), are close to those of 6′ (147.4(2)
and 151.8(1)°). As with 6′, the IR spectrum shows one strong
νCO band (1882 cm−1) and a second weaker band at a higher
frequency (1970 cm−1).25 However, with 6″ only a single CO
13C{1H} NMR signal is observed at −80 °C, in contrast to the
two found at ambient probe temperature for 6′.25 As noted
above, the close Fe···(CH2) contacts in 6′ seemingly block
Fe(CO)3 rotation, and the crystallographic data for 6″ deliver
an analogous prediction. Other mechanisms for 13C NMR
signal coalescence would include rapid and reversible CO
dissociation.
CONCLUSION
■
This study has shown that it is possible to access diiron
tetraphosphorus systems of the type V directly from Fe(CO)3
sources and diphosphines Ar2P(CH2)nPAr2 with n = 3 and 4
(Schemes 2 and 3). These results augment an earlier report of
an adduct with n/Ar = 5/Ph.9 The short Fe(CO)3/Fe(CO)3
distances in these systems lead to steric interactions. The
current NMR data suggest very high rotational barriers in 3
and lower barriers likely involving correlated rotation in 4a,b.
However, preliminary experiments show that a rigorous
interrogation of these processes is best approached with
13CO labeled substrates or perhaps via lower symmetry
isosteric systems that feature one or more CO/NO+
substitutions.7 Thus, no additional data regarding dynamic
properties are presented at this time. Additionally, as noted in
the preceding section, other types of diphosphines with fewer
degrees of conformational freedom may afford adducts with
more efficient rotator/rotator coupling.
The computational data further show that the diiron
tetraphosphorus complexes V are thermodynamically more
stable than all of the possible coordination geometries of the
monoiron diphosphorus complexes VI. They also establish
diagnostic IR νCO band patterns for the diiron complexes that
With regard to bimetallic complexes with sterically coupled
rotors, as embodied by III in Scheme 1, the synthetic strategy
applied in Scheme 2 is not limited to Fe(CO)3 sources or
3325
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
to P), 128.4 (br s, w1/2 34 Hz, m to P), 33.8 (apparent m, PCH2),
21.5 (s, PCH2CH2). 31P{1H} (202 MHz) 75.12 (s). IR (powder film,
cm−1): 3068 (w), 2959 (w), 1969 (w, νCO), 1892 (sh, νCO), 1871
(s, νCO), 1433 (m), 1092 (m), 957 (m), 795 (m), 745 (m), 692 (s),
633 (s). HRMS (ESI, m/z): calcd for C60H52Fe2O6P4 [M + H]+:
1105.1441. Found: 1105.1449.
allow structures to be verified for crystalline and noncrystalline
samples as well as in solution where dynamic NMR properties
may be probed. Although other potential applications for the
new diiron and monoiron carbonyl compounds should not be
overlooked,28 this work has provided a detailed foundation for
further studies of interactive Fe(CO)3/Fe(CO)3 rotators.
Finally, extensions of these themes to triiron, tetrairon, and
higher systems are easily envisioned, where there are intriguing
possibilities for “water wheels” and other types of molecular
machines.
trans,trans-(CO)3Fe[Ph2P(CH2)4PPh2]2Fe(CO)3 (4a). (BDA)Fe-
(CO)3 (0.500 g, 1.75 mmol), Ph2P(CH2)4PPh2 (0.611 g, 1.43 mmol),
and THF (19 mL) were combined in a procedure analogous to that
for 3. An identical workup gave 4a as a yellow solid (0.153 g, 0.135
mmol, 19% of theory (1.43/2 mmol)) that decomposed to a black
solid at ≥118 °C. Anal. Calcd (%) for C62H56Fe2O6P4 (1076.24): C,
65.74; H, 4.98. Found: C, 65.14; H, 5.20.31
EXPERIMENTAL SECTION
1
■
NMR (CDCl3, δ in ppm): H (500 MHz) 7.69−7.52 (br m, 16H,
General. All reactions and workups were conducted under inert
atmospheres using oven-dried glassware. Solvents were purified using
a Glass Contour system; CDCl3 and DMSO-d6 (2 × Cambridge
Isotopes) were freeze−pump−thaw degassed. The diphosphines
Ph2P(CH2)nPPh2 (n = 3, TCI; n = 4 and 6, Alfa-Aesar), P(p-tol)3
(Strem), n-BuLi (2.5 M in hexane, Sigma-Aldrich), 1,4-dibromobu-
tane (Alfa-Aesar), and silica (Silicycle, 40−63 μm, 230−400 mesh),
were used as received. (BDA)Fe(CO)3 was prepared by a literature
procedure.15
C6H5), 7.46−7.30 (br m, 24H, C6H5), 2.67 (br apparent s, 8H,
PCH2), 1.81 (br apparent s, 8H, PCH2CH2). 13C{1H} (125
MHz)30,32 214.0 (very weak s, tentative CO), 137.5−137.1 (apparent
2
m, i to P), 131.9 (virtual t, JCP = 4.8 Hz, o to P), 129.4 (s, p to P),
3
128.0 (virtual t, JCP = 4.6 Hz, m to P), 29.5 (apparent m, PCH2),
25.6 (apparent m, PCH2CH2). 31P{1H} (202 MHz) 73.96 (s). IR
(powder film, cm−1): 3054 (w), 2965 (w), 1964 (w, νCO), 1864 (s,
νCO), 1434 (m), 1093 (m), 1014 (m), 794 (m), 743 (m), 694 (s),
633 (s). HRMS (ESI, m/z): calcd for C62H56Fe2O6P4 [M + H]+,
1133.1754. Found, 1133.1775.
NMR spectra were recorded on standard FT instruments at
1
ambient probe temperatures and referenced as follows (δ, ppm): H,
trans,trans-(CO)3Fe[p-tol)2P(CH2)4P(p-tol)2]2Fe(CO)3 (4b). A
Schlenk flask was charged with (BDA)Fe(CO)3 (0.500 g, 1.75 mmol),
(p-tol)2P(CH2)4P(p-tol)2 (0.690 g, 1.43 mmol), and THF (19 mL)
with stirring and covered with aluminum foil. After 48 h, the solvent
was removed from the red solution by oil-pump vacuum. Then, 2:1 v/
v hexane/CH2Cl2 was added, and the mixture was filtered through a
pad of silica (3 × 5 cm). The solvent was removed from the filtrate by
oil-pump vacuum to give 4b as a pale yellow solid (0.498 g, 0.400
mmol, 56% of theory (1.43/2 mmol)) that decomposed to a black
solid at >146 °C. Although the 13C {1H} NMR spectrum suggested a
residual internal CHCl3 (7.27) or DMSO-d5 (2.50); 13C {1H},
internal CDCl3 (77.00) or DMSO-d6 (39.52); 31P{1H}, external 85%
H3PO4 (0.00). IR spectra were recorded using a Shimadzu IRAffinity-
1 spectrometer with a Pike MIRacle ATR system (diamond/ZnSe
crystal). Microanalyses were conducted by Atlantic Microlab.
Electrospray ionization mass spectrometry (ESI-MS) was carried
out with a Thermo Scientific Q Exactive Focus instrument.
(p-tol)2P(CH2)4P(p-tol)2. A Schlenk flask was charged with a
solution of P(p-tol)3 (2.0 g, 6.7 mmol) in THF (17 mL) and cooled
to 0 °C. Then n-BuLi (2.7 mL, 6.67 mmol, 2.5 M in hexane) was
added dropwise with stirring. After 1 h, the cold bath was removed,
and the bright red mixture transferred by cannula to a solution of 1,4-
dibromobutane (0.723 g, 3.35 mmol) in THF (13 mL). After 16 h,
the solvent was removed by oil-pump vacuum. Hexane was added,
and the sample filtered through a pad of silica (3 × 5 cm) that was
rinsed with 2:1 v/v hexane/CH2Cl2 and then CH2Cl2. The solvent
was removed from the CH2Cl2 rinses by oil-pump vacuum to give (p-
tol)2P(CH2)4P(p-tol)2 as a white solid (0.759 g, 1.57 mmol, 47%).
Mp 120.8−122.9 °C. Anal. Calcd (%) for C32H36P2 (482.58): C,
79.64; H, 7.52. Found: C, 79.92; H, 7.53.
1
high purity (≥96%), the H NMR spectrum indicated a somewhat
lower value (ca. 90%).33
1
NMR (CDCl3, δ in ppm): H (500 MHz) 7.56−7.39 (br m, 16H,
C6H4), 7.19 (br apparent s, 16H, C6H4), 2.47 (br apparent s, 24H,
CH3), 2.38 (br apparent s, 8H, PCH2), 1.73 (br apparent s, 8H,
32
PCH2CH2). 13C{1H} (125 MHz) 220.7 (t, JCP = 8.7 Hz, CO),
2
1
139.3 (s, p to P), 135.8 (virtual t, JCP = 21.9 Hz, i to P), 131.9
(virtual t, 2JCP = 5.2 Hz, o to P), 128.9 (virtual t, JCP = 4.7 Hz, m to
3
1
P), 32.3 (virtual t, JCP = 11.9 Hz, PCH2), 23.6 (s, PCH2CH2), 21.2
(s, CH3). 31P{1H} (202 MHz) 71.34. IR (powder film, cm−1): 3019
(w), 2923 (w), 2858 (w), 1965 (w, νCO), 1889 (sh, νCO), 1870 (s,
νCO), 1453 (m), 1096 (m), 803 (m), 644 (s). HRMS (ESI, m/z):
calcd for C70H73Fe2O6P4 [M + H]+: 1245.3051. Found: 1245.2973.
NMR (CDCl3, δ in ppm):29 1H (500 MHz) 7.33−7.27 (m, 8H, o
to P), 7.17−7.13 (m, 8H, m to P), 2.36 (s, 12H, CH3), 2.00 (br t,
3JHH = 6.9 Hz, 4H, PCH2) 1.59−1.51 (m, 4H, CCH2C); 13C{1H}
(125 MHz) 138.3 (s, p to P), 135.4 (d, 1JCP = 12.3 Hz. i to P), 132.6
(d, 2JCP = 18.8 Hz, o to P), 129.1 (d, 3JCP = 7.1 Hz, m to P), 28.0 (d,
trans-(CO)3Fe[Ph2P(CH2)6PPh2] (6′). (BDA)Fe(CO)3 (0.370 g,
1.29 mmol), Ph2P(CH2)6PPh2 (0.482 g, 1.06 mmol), and THF (14
mL) were combined in a procedure analogous to that for 4b. An
identical workup gave 6′ as a pale yellow solid (0.195 g, 0.328 mmol,
31%) that decomposed to a black solid at ≥116 °C. Anal. Calcd (%)
for C33H32FeO3P2 (594.40): C, 66.68; H, 5.43. Found: C, 66.17; H,
5.64.
2
1JCP = 10.7 Hz, PCH2), 27.7 (t, JCP = 14.8 Hz, PCH2CH2), 21.3 (s,
CH3); 31P{1H} (162 MHz) −18.2 (s). IR (powder film, cm−1): 2963
(w), 1558 (w), 1258 (m), 1089 (m), 1013 (s), 793 (s).
trans,trans-(CO)3Fe[Ph2P(CH2)3PPh2]2Fe(CO)3 (3). A Schlenk
flask was charged with (BDA)Fe(CO)3 (0.200 g, 0.699 mmol),
Ph2P(CH2)3PPh2 (0.236 g, 0.573 mmol), and THF (7.4 mL) with
stirring and covered with aluminum foil. A yellow precipitate formed.
After 48 h, the solvent was removed by oil-pump vacuum, and cold
THF was added. The mixture was filtered through a pad of silica (3 ×
5 cm) that was rinsed first with cold THF and then with toluene. The
solvent was removed from the yellow toluene rinses by oil-pump
vacuum to give 3 as a pale yellow solid (0.139 g, 0.126 mmol, 44% of
theory (0.573/2 mmol)) that decomposed to a black solid at >153
°C. Anal. Calcd (%) for C60H52Fe2O6P4 (1048.21): C, 65.24; H, 4.75.
Found: C, 65.86; H, 4.73.
NMR (CDCl3, δ in ppm): 1H (500 MHz) 7.76 (br m, 8H, C6H5),
7.40 (br m, 12H, C6H5), 2.47 (br m, 4H, PCH2), 2.06 (br m, 4H,
PCH2CH2), 1.65 (br m, 4H, PCH2CH2CH2). 13C{1H} (125 MHz)30
218.9 (m, 2CO), 213.2 (m, CO), 138.5 (m, i to P), 131.6 (br
apparent s, o to P), 129.4 (br apparent s, p to P), 128.2 (br apparent s,
m to P), 29.3 (br m, PCH2), 27.3 (br m, PCH2CH2), 20.4 (br m,
PCH2CH2CH2). 31P{1H} (202 MHz) 73.13 (s). IR (powder film,
cm−1): 3051 (w), 2914 (w), 1965 (w, νCO), 1863 (s, νCO), 1433
(m), 1092 (m), 787 (m), 739 (s), 690 (s), 640 (s). HRMS (ESI, m/
z): calcd for C33H32FeO3P2 [M]+, 594.1176. Found 594.1164; calcd
for C32H32FeO2P2 [M − CO]+, 566.1227. Found 566.12119; calcd for
C31H32FeOP2 [M − 2CO]+, 538.1278. Found 538.1267; calcd for
C30H32FeP2 [M − 3CO]+, 510.1329. Found 510.1266.
1
NMR (DMSO-d6, δ in ppm): H (500 MHz) 7.71−7.65 (br m,
16H, C6H5), 7.52−7.44 (m, 24H, C6H5), 2.75−2.65 (br m, 8H,
PCH2), 2.39−2.28 (br m, 4H, PCH2CH2). 13C{1H} (125 MHz,
cryoprobe)30 214.7 (apparent m, CO), 212.4 (apparent m, CO),
136.7 (apparent m, i to P), 131.2 (br s, w1/2 37 Hz, o to P), 129.9 (s, p
Crystallography. A. A CH2Cl2 solution of 3 was kept at −40 °C.
After 4 days, colorless blocks were collected, and data were obtained
3326
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
as outlined in Table 2. Cell parameters were obtained from 180 data
frames taken at widths of 0.5° and refined with 9463 reflections. The
unit cell was determined using the program Cell Now.34 Integrated
intensity information for each reflection was obtained by reduction of
the data frames with APEX2.35 Data were corrected for Lorentz and
polarization factors, crystal decay effects, and absorption effects (using
SADABS).36 The structure was solved using SHELXTL.37 Hydrogen
atoms were placed in idealized positions and refined using a riding
model. All non-hydrogen atoms were refined with anisotropic thermal
parameters. Two independent molecules were found in the unit cell,
each with an inversion center. Four molecules of CH2Cl2 were present
for every molecule of 3. Two were disordered but could be modeled
successfully. However, the irregular thermal ellipsoids and residual
electron density close to the chlorine atoms suggested further
disorder.
B. A toluene solution of 4a was kept at −40 °C. After 2 weeks,
colorless blocks were collected, and data were obtained as outlined in
Table 2. The structure was solved as in A (180 frames, 0.5° scan,
12 925 reflections). Two independent molecules were found in the
unit cell, each with an inversion center. Six toluene molecules were
present for every molecule of 4a. Residual electron densities close to
four toluene molecules indicated disorder, and each was modeled
between two positions. Some of the carbon atoms exhibited irregular
thermal ellipsoids, and restraints were used to keep them reasonable.
C. CHCl3 vapor was allowed to slowly diffuse into a toluene
solution of 4b at 0 °C. After 2 days, yellow blocks were collected, and
data were obtained as outlined in Table 2. Cell parameters were
obtained from 45 data frames taken at widths of 1°. Integrated
intensity information for each reflection was obtained by reduction of
the data frames with APEX3.37 All data were corrected for Lorentz
and polarization factors, crystal decay effects, and (using SADABS)36
absorption effects. The structure was solved using XT/XS in
APEX3.35,37 Solvent molecules were found. However, they were
disordered, and/or the sites were partially occupied, and could not be
modeled successfully. Hydrogen atoms were placed in idealized
positions and refined using a riding model. All non-hydrogen atoms
were refined with anisotropic thermal parameters. The solvent
molecules were MASKed using OLEX238 during the final least-
squares refinement cycles. The absence of additional symmetry and
voids was confirmed using PLATON (ADDYSM).39 The structure
was refined (weighted least-squares refinement on F2) to
convergence.37,38
D. Pentane vapor was allowed to slowly diffuse into a toluene
solution of 6′ at 0 °C. After 2 days, yellow blocks were collected, and
data were obtained as outlined in Table 2. The structure was solved as
in A (45 frames, 1° scan, 7998 reflections). A half molecule of toluene
was present for every molecule of 6′; the elongated ellipsoids
suggested disorder, which was successfully modeled between two
positions with a 0.66:0.34 occupancy ratio.
DFT Computations. An all-electron triple ζ quality basis set was
used on all atoms including iron (def2-TZVP).40 Thus, no
pseudopotential was used on heavier elements. This basis set was
paired with the TPSS41 functional, which has been shown to
reproduce experimental IR spectra of iron carbonyl complexes.13,22 All
computations were performed using the Gaussian09 program suite,
employing an ultrafine grid (99 590) to enhance accuracy.42 First, all
species were optimized and then subjected to frequency calculations
to perform a vibrational analysis. The IR spectra were computed for
all mono- and diiron species. These calculations were separated from
the optimizations, and additional memory (150 GB) was necessary to
reach convergence. The vibrational frequencies were inspected and all
structures were deemed to be local minima. The scaling factor
(0.9863) was close to values that proved optimal in earlier studies.43
Additional details of the DFT computations (PDF)
Accession Codes
CCDC 2050436−2050439 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
■
John A. Gladysz − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States; orcid.org/0000-0002-7012-4872;
Email: gladysz@mail.chem.tamu.edu
Michael B. Hall − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States; orcid.org/0000-0003-3263-3219;
Email: mbhall@tamu.edu
Authors
Samuel R. Zarcone − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States
Gong M. Chu − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States
Andreas Ehnbom − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States; orcid.org/0000-0002-7044-1712
Ashley Cardenal − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United
States; orcid.org/0000-0002-0639-8596
‡Tobias Fiedler − Department of Chemistry, Texas A&M
University, College Station, Texas 77842-3012, United States
Nattamai Bhuvanesh − Department of Chemistry, Texas
A&M University, College Station, Texas 77842-3012,
United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.inorgchem.0c03737
Notes
The authors declare no competing financial interest.
‡T.F. is deceased as of 28 October 2020.
ACKNOWLEDGMENTS
■
The authors thank the US National Science Foundation
(J.A.G.: CHE-1566601, CHE-1900549; M.B.H.: CHE-
1664866) and the Welch Foundation (M.B.H.: A-0648) for
support. The authors also thank Dr. D. J. Darensbourg for
helpful discussions and the Laboratory for Molecular
Simulation and Texas A&M High Performance Research
Computing for computational resources.
REFERENCES
■
(1) Kottas, G. S.; Clarke, L. I.; Horinek, D.; Michl, J. Artificial
Molecular Rotors. Chem. Rev. 2005, 105, 1281−1376.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.inorgchem.0c03737.
(2) Reviews of this extensive literature: (a) Iwamura, H.; Mislow, K.
Stereochemical Consequences of Dynamic Gearing. Acc. Chem. Res.
1988, 21, 175−182. (b) Goswami, A.; Saha, S.; Biswas, P. K.;
Schmittel, M. (Nano)mechanical Motion Triggered by Metal
Coordination: from Functional Devices to Networked Multi-
component Catalytic Machinery. Chem. Rev. 2020, 120, 125−199.
■
sı
Optimized geometries for computed structures (XYZ)
3327
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
(see especially section 2.2). (c) Dattler, D.; Fuks, G.; Heiser, J.;
Moulin, E.; Perrot, A.; Yao, X.; Giuseppone, N. Design of Collective
Motions from Synthetic Molecular Switches, Rotors, and Motors.
Chem. Rev. 2020, 120, 310−433. (see especially section 4.2).
(d) Liepuoniute, I.; Jellen, M. J.; Garcia-Garibay, M. A. Correlated
motion and mechanical gearing in amphidynamic crystalline
molecular machines. Chem. Sci. 2020, 11, 12994−13007.
Dibridgehead Diphosphine Stators: Syntheses, Substitution Reactions,
Structures, and Dynamic Properties. Inorg. Chem. 2017, 56, 7454−
7469.
( 9 ) S y n t h e s e s a n d c r y s t a l s t r u c t u r e s o f c i s -
(CO) 3 Fe[Ph2 P(C H2 )5 PP h2 ] (5 ′) and t r a n s, t r a n s-
(CO)3Fe[Ph2P(CH2)5PPh2]2Fe(CO)3 (5): Liu, L.-K.; Luh, L.-S.;
Gau, H.-M. Preparation and Structure Studies of Monomeric and
Dimeric [Bis(diphenylphosphino)pentane]tricarbonyliron(0). Inorg.
Chem. 1992, 31, 3434−3437.
(3) Some representative primary literature: (a) Frantz, D. K.;
Linden, A.; Baldridge, K. K.; Siegel, J. S. Molecular Spur Gears
Comprising Triptycene Rotators and Bibenzimidazole-Based Stators.
J. Am. Chem. Soc. 2012, 134, 1528−1535. (b) Lemouchi, C.;
Iliopoulos, K.; Zorina, L.; Simonov, S.; Wzietek, P.; Cauchy, T.;
Rodríguez-Fortea, A.; Canadell, E.; Kaleta, J.; Michl, J.; Gindre, D.;
Chrysos, M.; Batail, P. Crystalline Arrays of Pairs of Molecular Rotors:
Correlated Motion, Rotational Barriers, and Space-Inversion
Symmetry Breaking Due to Conformational Mutations. J. Am.
Chem. Soc. 2013, 135, 9366−9376. (c) Pérez-Estrada, S.;
Rodríguez-Molina, B.; Maverick, E. F.; Khan, S. I.; Garcia-Garibay,
M. A. Throwing in a Monkey Wrench to Test and Determine Geared
Motion in the Dynamics of a Crystalline One-Dimensional (1D)
Columnar Rotor Array. J. Am. Chem. Soc. 2019, 141, 2413−2420.
(4) (a) de Jonge, J. J.; Ratner, M. A.; de Leeuw, S. W. Local Field
Controlled Switching in a One-Dimensional Dipolar Array. J. Phys.
Chem. C 2007, 111, 3770−3777. (b) Zhao, K.; Dron, P. I.; Kaleta, J.;
Rogers, C. T.; Michl, J. Arrays of Dipolar Molecular Rotors in Tris(o-
phenylenedioxy)cyclotriphosphazene. Top. Curr. Chem. 2014, 354,
163−211. (c) Sharma, K.; Friedrich, B. Pair-eigenstates and mutual
alignment of coupled molecular rotors in a magnetic field. Phys. Chem.
Chem. Phys. 2016, 18, 13467−13477. (d) Abolins, B. P.; Zillich, R. E.;
Whaley, K. B. Quantum phases of dipolar rotors on two-dimensional
lattices. J. Chem. Phys. 2018, 148, 102338.
(10) Crystal structure of cis-(CO)3Fe[Ph2PCH2PPh2] (1′): Cotton,
F. A.; Hardcastle, K. I.; Rusholme, G. A. An X-ray Structural Study of
the Fluxional Molecule [bis(diphenylphosphino)methane]-
tricarbonyliron. J. Coord. Chem. 1973, 2, 217−223.
(11) The energies of trigonal bipyramidal and square planar isomers
can be quite close, with modest intervening barriers, and intermediate
structures are sometimes found.10,12 To help differentiate borderline
cases, a “tau” parameter has been devised: Crans, D. C.; Tarlton, M.
L.; McLauchlan, C. C. Trigonal Bipyramidal or Square Pyramidal
Coordination Geometry? Investigating the Most Potent Geometry for
Vanadium Phosphatase Inhibitors. Eur. J. Inorg. Chem. 2014, 2014,
4450−4468. and earlier papers cited therein; 1 = tbp, 0 = sq pyr.
Those associated with 1′, 2′, 3′, 5′, and 6′ are 0.76, 0.33, 0.93, 0.74,
and 0.10, respectively. Those for the diiron complexes range from
0.92 to 0.84.
( 1 2 ) S y n t h e s i s a n d c r y s t a l s t r u c t u r e o f c i s -
(CO)3Fe[Ph2P(CH2)2PPh2] (2′): Battaglia, L. P.; Delledonne, D.;
Nardelli, M.; Pelizzi, C.; Predieri, G.; Chiusoli, G. P. The crystal and
molecular structure of [bis(diphenylphosphino) ethane] tricarbonylir-
on(0). J. Organomet. Chem. 1987, 330, 101−113.
( 1 3 ) S y n t h e s i s a n d c r y s t a l s t r u c t u r e o f c i s -
(5) Ehnbom, A.; Gladysz, J. A. Gyroscopes and the Chemical
Literature, 2002−2020: Approaches to a Nascent Family of Molecular
Devices. Chem. Rev. 2021, DOI: 10.1021/acs.chemrev.0c01001.
(6) Lead references, square planar complexes: (a) Nawara-Hultzsch,
A. J.; Stollenz, M.; Barbasiewicz, M.; Szafert, S.; Lis, T.; Hampel, F.;
Bhuvanesh, N.; Gladysz, J. A. Gyroscope-Like Molecules Consisting
of PdX2/PtX2Rotators within Three-Spoke Dibridgehead Diphos-
phine Stators: Syntheses, Substitution Reactions, Structures, and
Dynamic Properties. Chem. - Eur. J. 2014, 20, 4617−4637. (b) Kharel,
S.; Joshi, H.; Bhuvanesh, N.; Gladysz, J. A. Syntheses, Structures, and
Thermal Properties of Gyroscope-like Complexes Consisting of PtCl2
Rotators Encased in Macrocyclic Dibridgehead Diphosphines P-
((CH2)n)3P with Extended Methylene Chains (n = 20/22/30) and
Isomers Thereof. Organometallics 2018, 37, 2991−3000. (c) Estrada,
(CO)3Fe[Ph2P(CH2)3PPh2] (3′): Ringenberg, M. R.; Wittkamp, F.;
Apfel, U.-P.; Kaim, W. Redox Induced Configurational Isomerization
of Bisphosphine-Tricarbonyliron(I) Complexes and the Difference a
Ferrocene Makes. Inorg. Chem. 2017, 56, 7501−7511.
( 1 4 ) O t h e r w o r k s r e p o r t i n g s y n t h e s e s o f c i s -
(CO)3Fe[Ph2P(CH2)nPPh2]: (a) n = 1 (1′): Wegner, P. A.; Evans,
L. F.; Haddock, J. Bimetallic Iron Carbonyl Derivatives of Bidentate
Phosphine Ligands. Inorg. Chem. 1975, 14, 192−194. (b) n = 1−4
(1′−4′): Sowa, J. R. Jr.; Zanotti, V.; Facchin, G.; Angelici, R. J.
Calorimetric Studies of the Heats of Protonation of the Metal in
Fe(CO)3(bidentate phosphine, arsine) Complexes: Effects of Chelate
Ligands on Metal Basicity. J. Am. Chem. Soc. 1992, 114, 160−165. (c)
n = 1−4 (1′−4′): Luo, L.; Nolan, S. P. Enthalpies of Reaction of
(Diene)- and (Enone)iron Tricarbonyl Complexes with Monodentate
and Bidentate Ligands. Solution Thermochemical Study of Ligand
Substitution in the L2Fe(CO)3 Complexes. Inorg. Chem. 1993, 32,
2410−2415. (d) n = 3 (3′): Langford, G. R.; Akhtar, M.; Ellis, P. D.;
MacDiarmid, A. G.; Odom, J. D. Synthesis and Multinuclear Magnetic
Resonance Study of Stereochemically Nonrigid Derivatives of Iron
Pentacarbonyl Containing Bidentate Ligands. Inorg. Chem. 1975, 14,
2937−2941.
(15) (a) Howell, J. A. S.; Johnson, B. F. G.; Josty, P. L.; Lewis, J.
Synthesis and reactions of tetracarbonyl- and tricarbonyliron
complexes of α,β-unsaturated ketones. J. Organomet. Chem. 1972,
39, 329−333. (b) Brookhart, M.; Nelson, G. O. The reaction of
benzylideneacetoneiron tricarbonyl with dienes; measurement of
relative reactivities using competition experiments. J. Organomet.
Chem. 1979, 164, 193−202.
A. L.; Jia, T.; Bhuvanesh, N.; Blumel, J.; Gladysz, J. A. Substitution
̈
and Catalytic Chemistry of Gyroscope-Like Complexes Derived from
Cl−Rh−CO Rotators and Triply trans Spanning Di-
(trialkylphosphine) Ligands. Eur. J. Inorg. Chem. 2015, 2015, 5318−
5321.
(7) Lead references, trigonal bipyramidal complexes: (a) Lang, G.
M.; Shima, T.; Wang, L.; Cluff, K. J.; Skopek, K.; Hampel, F.; Blumel,
̈
J.; Gladysz, J. A. Gyroscope-Like Complexes Based on Dibridgehead
Diphosphine Cages That Are Accessed by Three-Fold Intramolecular
Ring Closing Metatheses and Encase Fe(CO)3, Fe(CO)2(NO)+, and
Fe(CO)3(H)+ Rotators. J. Am. Chem. Soc. 2016, 138, 7649−7663.
(b) Lang, G. M.; Skaper, D.; Hampel, F.; Gladysz, J. A. Synthesis,
reactivity, structures, and dynamic properties of gyroscope like iron
complexes with dibridgehead diphosphine cages: pre- vs. post-
metathesis substitutions as routes to adducts with neutral dipolar
Fe(CO)(NO)(X) rotors. Dalton Trans. 2016, 45, 16190−16204.
(8) Lead references, octahedral complexes: (a) Fiedler, T.;
Bhuvanesh, N.; Hampel, F.; Reibenspies, J. H.; Gladysz, J. A.
Gyroscope like molecules consisting of trigonal or square planar
osmium rotators within three-spoked dibridgehead diphosphine
stators: syntheses, substitution reactions, structures, and dynamic
properties. Dalton Trans. 2016, 45, 7131−7147. (b) Hess, G. D.;
Fiedler, T.; Hampel, F.; Gladysz, J. A. Octahedral Gyroscope-like
Molecules Consisting of Rhenium Rotators within Cage-like
(16) (a) Skopek, K.; Gladysz, J. A. Syntheses of gyroscope-like
molecules via three-fold ring closing metatheses of bis(phosphine)
complexes trans-LyM(P((CH2)nCHCH2)3)2, and extensions to
bis(phosphite) complexes trans-Fe(CO)3(P(O(CH2)nCHCH2)3)2.
J. Organomet. Chem. 2008, 693, 857−866. (b) Lang, G. M.; Skaper,
D.; Shima, T.; Otto, M.; Wang, L.; Gladysz, J. A. Syntheses of Iron(0)
Complexes of Symmetrical Trialkylphosphines with Three Terminal
Vinyl Groups, P((CH2)mCHCH2)3. Aust. J. Chem. 2015, 68, 1342−
1351. (c) Steigleder, E.; Shima, T.; Lang, G. M.; Ehnbom, A.;
3328
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
1
Hampel, F.; Gladysz, J. A. Partially Shielded Fe(CO)3 Rotors:
Syntheses, Structures, and Dynamic Properties of Complexes with
(29) The H and 13C{1H} NMR signal assignments are made by
analogy to those rigorously verified for (p-tol)2P(CH2)14P(p-tol)2
earlier. See de Quadras, L.; Stahl, J.; Zhuravlev, F.; Gladysz, J. A.
Monophosphine and diphosphine ligands for diplatinum polyynediyl
complexes: Efficient syntheses of new functionality-containing
systems and model compounds. J. Organomet. Chem. 2007, 692,
1859−1870. and endnote 27 therein.
Doubly trans Spanning Diphosphines, trans-Fe(CO)3(PhP-
((CH2)n)2PPh). Organometallics 2017, 36, 2891−2901.
(17) Leitner, W.; Bu
Dinjus, E.; Kessler, M.; Kru
in Complexes Bearing Chelating Bidentate Phosphine Ligands.
Organometallics 1999, 18, 1196−1206.
(18) Adams, D. M. Metal−Ligand and Related Vibrations. St.
Martin’s Press: New York, 1968; Chapter 3.4.
̈
hl, M.; Fornika, R.; Six, C.; Baumann, W.;
̈
ger, C.; Rufinska, A. 103Rh Chemical Shifts
́
(30) Of the phenyl 13C{1H} NMR signals, that with the chemical
shift closest to benzene (128.4 ppm) is attributed to the meta carbon
atom, and the other signal of comparable intensity (and phosphorus
coupling) is attributed to the ortho carbon atom. See Mann, B. E. The
Carbon-13 and Phosphorus-31 Nuclear Magnetic Resonance Spectra
of Some Tertiary Phosphines. J. Chem. Soc., Perkin Trans. 2 1972, 30−
34.
(19) The IR spectrum of 4a (Figure 1) exhibits some weak features
that do not have counterparts in the spectra of 3 or 4b. These do not
arise from any 4′a contaminant.
(31) The 31P{1H} NMR spectra of the crude reaction mixture
suggest that some 4′a14b may be formed. However, this species is
removed upon workup.
(20) Throughout this manuscript, primed compound numbers
indicate monoiron complexes (CO)3Fe[Ph2P(CH2)nPPh2] regardless
of coordination geometry or stereochemistry. All trigonal bipyramidal
complexes presented have cis phosphorus donor atoms (3′, 4′a, 5′;
axial/equatorial), but square pyramidal complexes can be either cis
(1′, 2′) or trans (6′, 6′′; basal/basal in all cases).
(32) For virtual triplets, the xJCP values represent apparent couplings
between adjacent peaks that take place through a minimum of x
bonds. Hersh, W. H. False AA’X Spin-Spin Coupling Systems in 13C
NMR: Examples Involving Phosphorus and a 20-Year-Old Mystery in
Off-Resonance Decoupling. J. Chem. Educ. 1997, 74, 1485−1488.
(33) All reported syntheses have been carried out by at least two co-
workers. In this case, the second found that pure 4b was best obtained
by scaling down the solvent volume (THF, 3.3 mL; (BDA)Fe(CO)3,
0.200 g/0.414 mmol; (p-tol)2P(CH2)4P(p-tol)2, 0.145 g/0.506
mmol). Otherwise what was presumed to be 4′b (based upon the
chemical shift relationship exhibited by 5′ and 5)9 dominated after
workup.
(21) (a) Bondi, A. van der Waals Volumes and Radii. J. Phys. Chem.
1964, 68, 441−451. (b) Mantina, M.; Chamberlin, A. C.; Valero, R.;
Cramer, C. J.; Truhlar, D. G. Consistent van der Waals Radii for the
Whole Main Group. J. Phys. Chem. A 2009, 113, 5806−5812.
(22) Representative studies validating the use of the TPSS functional
and/or def2-TZVP basis set for IR calculations in related systems:
(a) Brothers, S. M.; Darensbourg, M. Y.; Hall, M. B. Modeling
Structures and Vibrational Frequencies for Dinitrosyl Iron Complexes
(DNICs) with Density Functional Theory. Inorg. Chem. 2011, 50,
8532−8540. (b) Meng, L.; Liu, S.; Qin, Q.; Zeng, B.; Chi, C. Infrared
Photodissociation Spectroscopy of Heteronuclear Arsenic-Iron
Carbonyl Cluster Anions. J. Phys. Chem. A 2020, 124, 1158−1166.
(c) Speelman, A. L.; Zhang, B.; Silakov, A.; Skodje, K. M.; Alp, E. E.;
Zhao, J.; Hu, M. Y.; Kim, E.; Krebs, C.; Lehnert, N. Unusual Synthetic
Pathway for an {Fe(NO)2}9 Dinitrosyl Iron Complex (DNIC) and
Insight into DNIC Electronic Structure via Nuclear Resonance
Vibrational Spectroscopy. Inorg. Chem. 2016, 55, 5485−5501.
(d) Pan, S.; Zhao, L.; Dias, H. V. R.; Frenking, G. Bonding in
Binuclear Carbonyl Complexes M2(CO)9 (M = Fe, Ru, Os). Inorg.
Chem. 2018, 57, 7780−7791. (e) Kerpal, C.; Harding, D. J.; Rayner,
D. M.; Fielicke, A. Small Platinum Cluster Hydrides in the Gas Phase.
J. Phys. Chem. A 2013, 117, 8230−8237.
(34) Sheldrick, G. M. Cell_Now: Program for Obtaining Unit Cell
Constants from Single Crystal Data, version 2008/1; University of
Gottingen, 2007.
(35) APEX2/3; Bruker AXS Inc.: Madison, WI, 2016.
(36) Sheldrick, G. M. SADABS; Bruker AXS Inc.: Madison, WI,
2001.
(37) (a) Sheldrick, G. M. A short history of SHELX. Acta
Crystallogr., Sect. A: Found. Crystallogr. 2008, A64, 112−122.
(b) Sheldrick, G. M. SHELXT − Integrated space-group and crystal
structure determination. Acta Crystallogr., Sect. A: Found. Adv. 2015,
71, 3−8. (c) Sheldrick, G. M. Crystal structure refinement with
SHELXL. Acta Crystallogr., Sect. C: Struct. Chem. 2015, C71, 3−8.
(d) XT, XS; Bruker AXS Inc.: Madison, WI, 2016.
(38) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.
K.; Puschmann, H. OLEX2: A Complete Structure Solution,
Refinement and Analysis Program. J. Appl. Crystallogr. 2009, 42,
339−341.
(39) Spek, A. L. Single-crystal structure validation with the program.
J. Appl. Crystallogr. 2003, 36, 7−13.
(40) Weigend, F.; Ahlrichs, R. Balanced basis sets of split valence,
triple zeta valence and quadruple zeta valence quality for H to Rn:
Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(41) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E.
Climbing the Density Functional Ladder: Nonempirical Meta-
Generalized Gradient Approximation Designed for Molecules and
Solids. Phys. Rev. Lett. 2003, 91, 146401.
(42) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.;
Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi,
R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar,
S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox,
J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
̈
(23) Bessel, C. A.; Aggarwal, P.; Marschilok, A. C.; Takeuchi, K. J.
Transition-Metal Complexes Containing trans-Spanning Diphosphine
Ligands. Chem. Rev. 2001, 101, 1031−1066.
(24) A report of a square planar nickel complex with a trans-
spanning pentamethylene diphosphine Cy2P(CH2)5PCy2 has been
questioned:23 Issleib, V. K.; Hohlfeld, G. Beitrage zur Komplexchemie
̈
der Phosphine und Phosphinoxyde. X. Schwermetallkomplexe
̈
ditertiarer Phosphine. Z. Anorg. Allg. Chem. 1961, 312, 169−179.
(25) Casey, C. P.; Whiteker, G. T.; Campana, C. F.; Powell, D. R.
Pentacoordinate Fe(CO)3 Complexes of Diphosphine Ligands with
Bite Angles Greater Than 120°. Inorg. Chem. 1990, 29, 3376−3381.
(26) See the following lead references, and earlier work cited
therein: (a) Balch, A. L.; Tulyathan, B. Interactions between
rhodium(I) centers in dimeric complexes. Inorg. Chem. 1977, 16,
2840−2845. (b) Cowie, M.; Dwight, S. K. Unusual structural and
chemical trends within a series of binuclear rhodium carbonyl halide
complexes and the structure of one member of this series, trans-
[Rh2Cl2(CO)2((C6H5)2PCH2P(C6H5)2)2]. Inorg. Chem. 1980, 19,
2500−2507.
(27) Cecconi, F.; Midollini, S.; Orlandini, A.; Sacconi, L. Trimeric
nickel(II) complexes with bridging triphosphine ligand 1,1,1-tris-
(diphenylphosphinoethyl)ethane (etp3). X-ray structure of the
complex [Ni3Cl6(etp3)2]·0.7CHCl3. Inorg. Chim. Acta 1980, 42,
59−63.
(28) Jiang, X.; Xiao, Z.; Zhong, W.; Liu, X. Brief survey of diiron and
monoiron carbonyl complexes and their potentials as CO-releasing
molecules (CORMs). Coord. Chem. Rev. 2021, 429, No. 213634.
3329
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330

Inorganic Chemistry
pubs.acs.org/IC
Article
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
̈
D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(43) (a) Kesharwani, M. K.; Brauer, B.; Martin, J. M. L. Frequency
and Zero-Point Vibrational Energy Scale Factors for Double-Hybrid
Density Functionals (and Other Selected Methods): Can Anharmonic
Force Fields Be Avoided? J. Phys. Chem. A 2015, 119, 1701−1714.
(b) Assefa, M. K.; Devera, J. L.; Brathwaite, A. D.; Mosley, J. D.;
Duncan, M. A. Vibrational scaling factors for transition metal
carbonyls. Chem. Phys. Lett. 2015, 640, 175−179.
3330
https://dx.doi.org/10.1021/acs.inorgchem.0c03737
Inorg. Chem. 2021, 60, 3314−3330