Protonation of ferrous dinitrogen complexes containing a diphosphine ligand with a pendent amine

Article abstract of DOI:10.1021/ic4000704

The addition of acids to ferrous dinitrogen complexes [FeX(N ₂)(P^EtN^MeP^Et)(dmpm)]⁺ (X = H, Cl, or Br; P^EtN^MeP^Et = Et ₂PCH₂N(Me)CH₂PEt₂; and dmpm = Me₂PCH₂PMe₂) gives protonation at the pendent amine of the diphosphine ligand rather than at the dinitrogen ligand. This protonation increased the ν_N2 band of the complex by 25 cm ^-1 and shifted the Fe(II/I) couple by 0.33 V to a more positive potential. A similar IR shift and a slightly smaller shift of the Fe(II/I) couple (0.23 V) was observed for the related carbonyl complex [FeH(CO)(P ^EtN^MeP^Et)(dmpm)]⁺. [FeH(P ^EtN^MeP^Et)(dmpm)]⁺ was found to bind N₂ about three times more strongly than NH₃. Computational analysis showed that coordination of N₂ to Fe(II) centers increases the basicity of N₂ (vs free N₂) by 13 and 20 pK _a units for the trans halides and hydrides, respectively. Although the iron center increases the basicity of the bound N₂ ligand, the coordinated N₂ is not sufficiently basic to be protonated. In the case of ferrous dinitrogen complexes containing a pendent methylamine, the amine site was determined to be the most basic site by 30 pK_a units compared to the N₂ ligand. The chemical reduction of these ferrous dinitrogen complexes was performed in an attempt to increase the basicity of the N₂ ligand enough to promote proton transfer from the pendent amine to the N₂ ligand. Instead of isolating a reduced Fe(0)-N₂ complex, the reduction resulted in isolation and characterization of HFe(Et ₂PC(H)N(Me)CH₂PEt₂)(P^EtN ^MeP^Et), the product of oxidative addition of the methylene C-H bond of the P^EtN^MeP^Et ligand to Fe.

Full text of DOI:10.1021/ic4000704

Article
pubs.acs.org/IC
Protonation of Ferrous Dinitrogen Complexes Containing a
Diphosphine Ligand with a Pendent Amine
Zachariah M. Heiden,‡ Shentan Chen,‡ Michael T. Mock,‡,* William G. Dougherty,† W. Scott Kassel,†
Roger Rousseau,‡ and R. Morris Bullock‡
‡Center for Molecular Electrocatalysis, Physical Sciences Division, Pacific Northwest National Laboratory, Richland, Washington
99352, United States
†Department of Chemistry, Villanova University, Villanova, Pennsylvania 19085, United States
S
* Supporting Information
ABSTRACT: The addition of acids to ferrous dinitrogen
complexes [FeX(N2)(PEtNMePEt)(dmpm)]+ (X = H, Cl, or Br;
PEtNMePEt = Et2PCH2N(Me)CH2PEt2; and dmpm =
Me2PCH2PMe2) gives protonation at the pendent amine of the
diphosphine ligand rather than at the dinitrogen ligand. This
protonation increased the νN2 band of the complex by 25 cm−1
and shifted the Fe(II/I) couple by 0.33 V to a more positive
potential. A similar IR shift and a slightly smaller shift of the
Fe(II/I) couple (0.23 V) was observed for the related carbonyl
complex [FeH(CO)(PEtNMePEt)(dmpm)]+. [FeH(PEtNMePEt)-
(dmpm)]+ was found to bind N2 about three times more strongly
than NH3. Computational analysis showed that coordination of
N2 to Fe(II) centers increases the basicity of N2 (vs free N2) by 13 and 20 pKa units for the trans halides and hydrides,
respectively. Although the iron center increases the basicity of the bound N2 ligand, the coordinated N2 is not sufficiently basic to
be protonated. In the case of ferrous dinitrogen complexes containing a pendent methylamine, the amine site was determined to
be the most basic site by 30 pKa units compared to the N2 ligand. The chemical reduction of these ferrous dinitrogen complexes
was performed in an attempt to increase the basicity of the N2 ligand enough to promote proton transfer from the pendent amine
to the N2 ligand. Instead of isolating a reduced Fe(0)−N2 complex, the reduction resulted in isolation and characterization of
HFe(Et2PC(H)N(Me)CH2PEt2)(PEtNMePEt), the product of oxidative addition of the methylene C−H bond of the PEtNMePEt
ligand to Fe.
The nitrogenase metalloenzyme reduces N2 using a
combination of protons and electrons at ambient temperature
and pressure to generate ammonia.5 The N2 reduction pathway
requires eight protons and eight electrons, producing one
hydrogen molecule and two ammonia molecules; a compre-
hensive understanding of the N2 reduction mechanism has been
sought for over three decades.6 Recent spectroscopic studies on
the FeMo cofactor from the groups of Hoffman and Seefeldt
suggest N2 reduction occurs by an “alternating” reduction
pathway (symmetric pathway in Scheme 1), where N2 is
reduced stepwise, with protonation occurring first at the distal
N atom followed by protonation at the proximal N atom,
possibly involving multiple iron centers in oxidation states
common to biological systems (e.g., FeII and/or FeIII; Scheme
1).6b,7 Importantly, these findings encourage the development
of synthetic systems to both replicate and understand the
factors that govern N2 binding, elucidate the mechanism of
reduction pathways, and examine the protonated N2
intermediates along the reduction pathway.
INTRODUCTION
■
The development of homogeneous transition metal complexes
for the reduction and/or functionalization of N2 has been an
area of research for more than 40 years.1 Iron complexes have
been targeted2 for N2 reduction studies because iron is an
abundant, inexpensive,3 and biorelevant transition metal found
in the active site of the nitrogenase metalloenzyme. The active
site of nitrogenase contains two Fe4S3 cubane units connected
through a carbide and three sulfide linkers (Figure 1), where
one of the metal atoms (designated M in Figure 1) can be Mo,
V, or Fe.4
Figure 1. Proposed active site of nitrogenase, where M can be Fe, Mo,
or V.
Received: January 11, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Possible N2 Reduction Pathways with Iron
RESULTS AND DISCUSSION
Synthesis and Characterization of Ferrous N2 Com-
■
plexes. To assess whether the introduction of a proton relay in
the ligand of a ferrous dinitrogen complex aids in N2 binding
and subsequent reduction to ammonia, we synthesized
complexes of the type [FeX(N2)(PEtNMePEt)(dmpm)]+ (X =
H, Cl, or Br; PEtNMePEt = Et2PCH2N(Me)CH2PEt2; and dmpm
= Me2PCH2PMe2; see Figure 2). The PEtNMePEt ligand was
To date, two types of coordination complexes have been
shown to catalytically reduce N2 to NH3 with protons and
electrons. In contrast to nitrogenase, these coordination
complexes are proposed to undergo an asymmetric reduction
of N2 to NH3 (Scheme 1), where protonation occurs first at the
distal nitrogen atom, followed by NH3 expulsion, prior to
protonation of the proximal nitrogen atom. Schrock’s Mo
triamidoamine catalyst has been shown to produce eight
equivalents of NH3 (four turnovers).8 In a recent example by
Nishibayashi et al., a Mo catalyst supported by a phosphine-
based pincer ligand was shown to produce 23 equivalents of
NH3 (12 catalyst turnovers).9 Although these two Mo-based
complexes catalyze the reduction of N2 to ammonia, only
stoichiometric ammonia formation has been demonstrated with
iron-based complexes.10 The stoichiometric production of
ammonia and hydrazine in yields ranging from 4 to 82% has
been reported using FeII,11 FeI,12 and Fe0 dinitrogen13 and iron-
nitride complexes.14 Peters and co-workers expanded upon this
previous work by demonstrating that upon the addition of an
external reductant, N2H4 generation can be increased from 17
to 47%.15
Figure 2. Ferrous complexes containing pendent amines used in the
synthesis of Fe−N2 complexes.
chosen to allow for the direct comparison of N2 chemistry to
the previously reported H2 chemistry, [FeH(H2)(PEtNMePEt)-
(dmpm)]+.22 Incorporation of only one pendent amine allowed
for an easier comparison to previously reported ferrous
dinitrogen complexes.2 The dinitrogen complex was prepared
as shown in eq 1. The iron chloride complexes
Biological enzymes hydrogenase and nitrogenase have been
found to employ pendent bases (histidine-195 in the case of
nitrogenase),7e,16 or proton channels,5a,17 to promote the
activation/production of H2 and N2, through proton-coupled-
electron-transfer (PCET) reactions.18 One way to design a
molecular catalyst to promote PCET reactions is by the
incorporation of pendent amines that function as proton relays,
in the second coordination sphere of the ligand. This concept
has been previously employed in the electrochemical reduction
of protons to H2 in diphosphine supported Ni catalysts,19 and
in the reduction of dioxygen.20 Recognizing that protonation of
the pendent amine groups can modulate the redox potential of
the complex and appreciating the importance of controlling
proton movement in these multiproton, multielectron reac-
tions, we sought to explore the effect of pendent amines as
proton relays on the coordination of N2 and its subsequent
protonation,21 with the ultimate goal of catalytic reduction of
N2 to NH3.
Herein, we report the synthesis of ferrous dinitrogen
complexes of the type [FeX(N2)(PEtNMePEt)(dmpm)]+ (X =
H, Cl, or Br; PEtNMePEt = Et2PCH2N(Me)CH2PEt2; and dmpm
= Me2PCH2PMe2), which introduces a basic pendent amine
site in the phosphine ligand backbone. In this investigation, we
study the reaction of the Fe(N2) complexes with acid and
examine the effects of protonation with regard to N2 ligand
binding and redox behavior. We complement these exper-
imental results with a computational study to provide a
quantitative comparison of the basicity of the Fe-bound N2 and
the pendent amine. In addition, N2 ligand substitution by NH3
and H2 is examined, and initial efforts to prepare Fe0-N2
complexes with the PEtNMePEt ligand are described.
FeCl2(PEtNMePEt)(dmpm) (1(Cl)) and FeHCl(PEtNMePEt)-
(dmpm) (1(H)) have been previously reported by DuBois
and co-workers,22a and iron bromide FeBr2(PEtNMePEt)-
(dmpm) (2(Br)) and FeHBr(PEtNMePEt)(dmpm) (2(H)) are
reported herein. FeHBr(PEtNMePEt)(dmpm) was prepared
following the synthetic method reported for FeHCl-
(PEtNMePEt)(dmpm),22a by the addition of Bu4NBH4 to
FeBr2(PEtNMePEt)(dmpm) in THF at −35 °C, and was isolated
in 48% yield.
Ferrous dinitrogen complexes were synthesized via abstrac-
tion of a halide from the parent dihalide and hydrido-halide
complexes. The N2-hydride complex trans-[FeH(N2)-
(PEtNMePEt)(dmpm)]+ ([3(N2)]+) can be prepared by halide
abstraction from FeHCl(PEtNMePEt)(dmpm) (1(H)) or
FeHBr(PEtNMePEt)(dmpm) (2(H)) using salts containing
noncoordinating anions: NaBPh4, KB(C6F5)4, or NaBArF
4
(ArF = 3,5-bis(trifluoromethyl)phenyl) in 60−75% yield (eq
1). The reaction of 1(H) or 2(H) with NaBPh4 in THF yielded
[3(N2)]+ in ∼2.5 h as monitored by in situ IR spectroscopy
(Figure 3) and 31P NMR spectroscopy. [3(N2)]BPh4 exhibits a
νNN band at 2095 cm−1 (KBr; 2103 cm−1 in THF) and a νFeH
band at 1886 cm−1 (KBr). The measured vibrational
frequencies of [3(N2)]BPh4 are similar to the values previously
reported for trans-[FeH(N2)(dmpe)2]BPh4 (dmpe =
Me2PCH2CH2PMe2; νNN = 2094 cm−1, νFeH = 1841 cm−1
(ethanol))23 and [FeH(N2)((P(C3H6OMe)2CH2)2)2]BPh4
B
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
angle of the PEtNMePEt ligand is 90.73(14)°. The average Fe−P
bond length for the PEtNMePEt ligand is 2.211(4) Å, and the
average Fe−P bond length for the dmpm ligand is slightly
longer at 2.221(4) Å. The hydride ligand trans to the N2 ligand
was located from the difference map and exhibits a H−Fe−
N(2) angle of 178.9(7)°. The Fe−H and Fe−N2 bond lengths
are 1.46(2) Å and 1.827(1) Å, respectively. Values involving the
Fe−H bond length should be interpreted cautiously since
hydrogen atoms are not located with high accuracy by X-ray
diffraction studies.25 The N(2)−N(3) bond length of the
coordinated N2 ligand is 1.109(2) Å. For comparison,
[FeH(N2)(dmpe)2]BPh4 and [FeH(N2)(depe)2]BPh4 have
N−N bond lengths of 1.13(3) Å and 1.070(12) Å,
respectively.23,26 In all of these examples, the N2 ligand is
considered to be only weakly activated with respect to free N2
(1.0975 Å)2a based on N−N bond length and IR stretching
frequency, described above. The six-membered ring formed by
coordination of the PEtNMePEt ligand to the Fe center resides in
the chair conformation with the pendent amine positioned exo
with respect to the N2 ligand (Figure 4).
Figure 3. In situ infrared spectra versus time, monitoring the
formation of [3(N2)]+ (νNN = 2103 cm−1) from the halide abstraction
of 1(H) (νFeH = 1816 cm−1) with NaBPh4 in THF over 4 h.
(νNN = 2088 cm−1 (KBr)).24 The 31P NMR spectrum of
[3(N2)]+ contains two multiplets of an AA′XX′ pattern at δ
38.9 and −0.1, assigned as the PEtNMePEt and dmpm
resonances, respectively. The hydride resonance in the 1H
NMR spectrum appears as a multiplet at δ −14.6, which is
about 13 ppm downfield from the hydride resonance of 1(H)
or 2(H). Reactions to prepare [3(N2)]+ performed in
Halide abstraction from 1(Cl) or 2(Br) with KB(C6F5)4 in
fluorobenzene under an N2 atmosphere resulted in the
formation of [FeCl(N2)(PEtNMePEt)(dmpm)]+ ([1(N2)]+) or
[FeBr(N2)(PEtNMePEt)(dmpm)]+ ([2(N2)]+), which contains a
νNN band at 2110 and 2111 cm−1, respectively (eq 2). In
fluorobenzene, employing KB(C6F5)4 or NaBArF as a halide
4
abstraction agent, were considerably faster than in THF and
reached completion within seconds. Experiments performed
using 15N2 resulted in the expected shift of the νNN band to 60
cm−1 lower frequency, appearing at 2054 cm−1 (fluoroben-
zene). The 15N NMR spectrum (THF-d8) exhibits two
resonances at δ −41.8 and δ −59.8 vs CH3NO2, assigned as
the distal and proximal nitrogen atoms, respectively. Dis-
solution of this complex in fluorobenzene under ambient 14N2
atmosphere resulted in complete exchange of the 15N2 label for
14N2 over the course of ∼2 h.
Vapor diffusion of pentane into a fluorobenzene solution of
[3(N2)]B(C6F5)4 produced yellow X-ray quality crystals in
60% yield. The molecular structure, shown in Figure 4, reveals a
slightly distorted octahedral Fe(II) cation with the PEtNMePEt
and dmpm ligands in the equatorial plane. The P3−Fe−P4
angle of the dmpm ligand is 73.85(14)°, while the P1−Fe−P2
contrast to [3(N2)]+, the N2 ligand was found to be highly
labile.27 For example, no νNN band was found in the IR
spectrum of the solid product collected in KBr after removing
the solvent from [2(N2)]+ under a vacuum, or if the reaction
was undertaken in THF. Crystallization of the reddish-orange
solution under an N2 atmosphere yielded a red crystalline
product for which the elemental analysis is consistent with a
five-coordinate complex without an N2 ligand, [FeBr-
(PEtNMePEt)(dmpm)]+. Although an X-ray diffraction study of
these crystals did not yield a structure of publishable quality,
the refined data clearly showed connectivity in the first
coordination sphere in which a five-coordinate iron center
coordinated to four phosphines and a bromide could be
observed. To our knowledge, only two examples of a 5-
coordinate Fe complex of this type have been structurally
characterized, reported by Tyler et al.24 and Tuczek et al.28
[1(N2)]+ can also be prepared from 1(H), by reaction with
either Ph3C+ or B(C6F5)3 in chlorobenzene-d5. However,
further attempts to obtain better crystallographic data of
[1(N2)]+ or the five-coordinate iron complex were unsuccess-
ful.
Protonation Studies of [FeH(N2)(PEtNMePEt)(dmpm)]+
([3(N2)]+). The addition of acids to [3(N2)]+ was investigated
by a combination of 1H, 31P, and 15N NMR spectroscopy and in
situ IR experiments. Accordingly, the lability of N2, subsequent
decomposition pathways, and IR spectroscopic signature of N2
bound to Fe in the protonated complex were investigated. For
the NMR studies, 15N2 was incorporated into [3(N2)]+,
([FeH(15N2)(PEtNMePEt)(dmpm)]BPh4), to conveniently ac-
quire 15N NMR spectral data. The addition of one equivalent of
triflic acid (HOTf) to [3(15N2)]BPh4 at −35 °C in THF-d8
Figure 4. Molecular structure of the cation [FeH(N2)(PEtNMePEt)-
(dmpm)]+ ([3(N2)]+). Thermal ellipsoids are drawn at 30%
probability. The hydrogen atoms (except for the Fe−H) are omitted
for clarity. Selected bond distances (Å) and angles (deg): P(1)−Fe(1)
= 2.214(1), P(2)−Fe(1) = 2.208(1), P(3)−Fe(1) = 2.222(1), P(4)−
Fe(1) = 2.219 (1), Fe(1)−N(2) = 1.827(1), Fe(1)−H(1A) =
1.457(18), N(2)−N(3) = 1.109(2), P(2)−Fe(1)−P(1) = 90.73(1),
P(3)−Fe(1)−P(4) = 73.85(1), N(2)−Fe(1)−H(1A) = 178.9(7),
Fe(1)−N(2)−N(3) = 178.67(12).
C
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
resulted in the protonation of the pendent amine of the
PEtNMePEt ligand and the formation of two protonated isomers
(either endo or exo with respect to the N2 ligand; [3(15N2)-
H]2+), see eq 3 and Figure 5. Two pairs of multiplets of an
of [H(OEt2)2][B(C6F5)4] to a solution of [3(N2)]+ at −40 °C
resulted in an immediate increase in the stretching frequency of
the νN2 band by 25 cm−1 to 2139 cm−1, as shown in Figure 6.
Figure 6. In situ IR plot recorded at −40 °C from the reaction of
[3(N2)]B(C6F5)4 with one equivalent of [H(OEt2)2][B(C6F5)4] in
fluorobenzene followed by the addition of one equivalent of Et3N. The
total reaction time was 50 min (each increment is 1 min).
Figure 5. 15N{1H} NMR spectra at −35 °C in THF-d8 of (bottom)
[3(15N2)]+ and (top) [3(15N2)]+ with one equivalent of HOTf, where
Nα = proximal nitrogen atom and Nβ = distal nitrogen.
The complex resulting from protonation of [3(N2)]+, ([3(N2)-
H]2+), is stable in solution for over 24 h at −40 °C, and the
protonation of [3(N2)]+ is reversible upon the addition of a
base (Et3N). This is a unique example of a ferrous-N2 complex,
where the shift in the stretching frequency of the νNN band was
a result of directly protonating the ancillary ligand. The
observed increase in νN2 stretching frequency indicates that N2
is not protonated, as a lower energy N2 band would be expected
if that were the case.30
AA′XX′ pattern are observed in the 31P NMR spectrum at δ
47.9, −1.2 and δ 52.8, −0.2 in a ∼3:1 ratio, assigned as the
PEtNMePEt and dmpm ligands, respectively. Notable features in
1
the corresponding H NMR spectrum include two pentets
corresponding to the hydride ligand at δ −14.62 (1JHP = 47.6
Hz) and −14.92 (1JHP = 48.2 Hz) and two broad peaks at δ
9.54 and 9.35 assigned to the NH of the protonated pendent
amine for the major and minor product, respectively (see
Supporting Information for NMR spectral data). A cross-peak
in the 1H−15N correlation experiment (HSQC) confirmed this
assignment as the protonated pendent amine with the
corresponding 15N NMR shifts at δ −337.3 and −335.9 for
the major and minor isomers, respectively. The observed 15N
chemical shifts are consistent with values reported for
protonated alkyl amines.29 15N NMR spectral data also contain
two pairs of singlets (four total resonances) in a ∼3:1 ratio. The
major isomer, at δ −43.3 and −66.5, and the minor species
located at δ −44.0 and −69.2 (Figure 5), correspond to the
distal (Nβ) and proximal (Nα) nitrogen atoms of 15N2,
respectively. While NMR studies have not established which
protonated isomer, exo-[3(N2)H]2+ or endo-[3(N2)H]2+, is the
major product (NOESY experiments were inconclusive), it is
probable that the major isomer contains the proton on the
pendent amine positioned exo with respect to the N2 ligand (eq
3). This assignment has been corroborated by DFT calculations
(see computational section below).
To explain the origin of the increase in the νN2 band, two
possibilities are considered: a through bond inductive effect
(diminished back-bonding to the N2 ligand) or an electrostatic
effect (due to increased positive charge of the complex). In a
related example by Schrock and co-workers, the addition of one
equivalent of acid to the molybdenum complex Mo(N2)-
(HIPTN3N) (HIPT = 3,5-(2,4,6-iPr3C6H2)2C6H3), supported
by the triamidoamine ligand, generates a protonated product in
which the νN2 band shifts 67 cm−1, from 1990 to 2057 cm−1.31
Their studies suggest that direct protonation of one of three
amido groups of the ligand resulted in decreased back-bonding
to the dinitrogen ligand, increasing the frequency of the νNN
band. Studies by Sellmann and Sutter reveal an analogous result
in the shift of the νCO band stretching frequency used to probe
the electronic effect induced by protonation at the sulfur atom
of a bound thiolate donor of the bis-benzene-dithiolate ethylene
bridged ligand in [Fe(CO)2(MeSC6H4SC2H4SC6H4SMe)] or
the bis-benzene-dithiolate bis-ethylene-amine bridged ligand in
[Fe(CO)(SC6H4SC2H4NHC2H4SC6H4S)]. In this example,
protonation of the thiolate directly bound to the Fe center
resulted in an increase of the νCO bands by ca. 40 cm−1.5c The
greater magnitude of this shift in stretching frequency observed
by Sellmann and Sutter, compared to the smaller difference
between [3(N2)]+ and [3(N2)H]2+, can be attributed to the
difference in the through bond proximity of the protonated site
to the metal center. In complexes structurally similar to
[3(N2)]+, where the proximity of the protonated group to the
The protonation of [3(N2)]+ was investigated further by IR
spectroscopy. For this experiment, complex [3(N2)]B(C6F5)4
was conveniently generated in situ from 2(H) and one
equivalent of KB(C6F5)4 in fluorobenzene. The reaction,
monitored by in situ IR spectroscopy, yielded a νNN band of
[3(N2)]B(C6F5)4 at 2114 cm−1. The addition of one equivalent
D
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
metal center is comparable, smaller carbonyl frequency shifts
have been observed. For example, the addition of protons to
azadithiolate [FeFe]-hydrogenase model complexes results in a
νCO shift of 15 cm−1 independent of the nitrogen substituents
or number of phosphines present on the iron centers.32
For the protonation of [3(N2)]+, the shift of the νN2 band is
ascribed to be electrostatic in origin. This assessment is made
on the basis of the increase in positive charge of the molecule
from a monocation to a dication and is supported by a shift of
similar magnitude in the protonation of the CO analogue and
by DFT calculations, both described below.
hydride ligands, which dramatically affects the lability of the N2
ligand (vide infra). The νCO bands for [1(CO)]+ and
[2(CO)]+, 1936 and 1947 cm−1, respectively, match the
previously reported data.22a,33 X-ray quality crystals of [1-
(CO)]+ and [2(CO)]+ were grown from vapor diffusion of
pentane into THF solutions, and details of the structural
determinations are provided in the Supporting Information.
1
On the basis of H and 15N NMR spectroscopic data, the
protonation reactions of [3(N2)]+ and [2(N2)]+ occur at the
pendent amine. To further probe the origin and magnitude of
the shift in IR frequency of the protonated species, and to
validate our spectroscopic results, we examined the protonation
chemistry of the CO containing complexes, [1(CO)]+,
[2(CO)]+, and [3(CO)]+.22a Generation of [1(CO)]+ and
[2(CO)]+ from the addition of CO to [1(N2)]+ and [2(N2)]+
resulted in a νCO band at 1943 and 1947 cm−1, respectively.
The addition of one equivalent of [H(OEt2)2][B(C6F5)4] in
fluorobenzene to [1(CO)]+ and [2(CO)]+ resulted in a shift of
the νCO band by 25 (1943 to 1968) and 24 (1947 to 1971)
cm−1, respectively (Figure 7). The protonation of [1(CO)]+
Protonation reactions of [3(N2)]+ performed at 25 °C reflect
a remarkable difference in the lability of the N2 ligand and
stability of [3(N2)H]2+. Following the procedure described
above, the addition of one equivalent of [H(OEt2)2][B(C6F5)4]
in fluorobenzene at 25 °C resulted in the formation of
[3(N2)H]2+ as observed by an immediate shift of the νN2 band
to 2139 cm−1, but [3(N2)H]2+ is thermally unstable at 25 °C.
In situ IR spectra collected at 15 s intervals reveal the
intermediacy of the νNN band and the instability of [3(N2)H]2+,
as the νNN band first shifts from 2114 to 2139 cm−1 and then
disappears within 30 s (see Supporting Information for IR
1
traces). In addition, H and 15N NMR spectral analysis shows
that upon warming [3(N2)H]2+ to 25 °C, in addition to the
free phosphine ligand and an unidentified diamagnetic product,
15N2 is ejected from the complex and H2 is produced. The
generation of H2 and loss of N2 is irreversible, and the addition
of Et3N to the resulting product did not regenerate [3(N2)]+.
Substitution Reactions of Ferrous N2 Compounds. To
examine the lability of the N2 ligand in [3(N2)]+ and [1(N2)]+,
ligand substitution reactions with CO were investigated. For
[3(N2)]+, the displacement of N2 by CO (1 atm) in
fluorobenzene occurred slowly over the course of ∼2 h as
monitored by in situ IR spectroscopy, affording [FeH(CO)-
(PEtNMePEt)(dmpm)]+ ([3(CO)]+) (eq 4). The νCO band at
Figure 7. In situ IR plot recorded at 25 °C in fluorobenzene from the
reaction of (A) [2(N2)]+ with CO to form [2(CO)]+, (B) the addition
of one equivalent of [H(OEt2)2][B(C6F5)4] forming [2(CO)H]2+, and
(C) the addition of one equivalent of Et3N, generating [2(CO)]+.
Each product was stirred for 15 min before addition of the subsequent
acid or base (total reaction time = 45 min).
(pKa = 9.6 in MeCN for [1(CO)H]2+) was previously shown
to occur at the pendent amine and has been only characterized
by NMR spectroscopy.22a The addition of 1.5 equivalent of
Et3N resulted in a return of the CO band back to the original
position for [1(CO)]+ and [2(CO)]+. Treatment of [3(CO)]+
with acid resulted in a 22 cm−1 shift, and [3(CO)]+ could be
regenerated by adding an equivalent of Et3N to [3(CO)H]2+
(see Supporting Information). The protonation of [3(CO)]+
(pKa = 10.5 in MeCN for [3(CO)H]2+) was also previously
shown to occur at the pendent amine and was only
characterized by NMR spectroscopy.22a
1938 cm−1 matched that of the previously reported complex.22a
A similar rate of substitution was observed in the exchange
reaction of 15N2 from [3(15N2)]+ for 14N2, suggesting that N2
dissociation is the rate-limiting step.
In an analogous reaction monitored by in situ IR spectros-
copy, fluorobenzene solutions of [1(N2)]B(C6F5)4 rapidly lost
N2 (<15 s) upon exposure to CO (1 atm), resulting in the
previously reported complex [FeCl(CO)(PEtNMePEt)(dmpm)]+
([1(CO)]+) (eq 5).22a Identical results were obtained for
The formation of both dihydrogen and ammonia by
nitrogenase enzymes prompted us to explore the displacement
of dinitrogen by dihydrogen and ammonia. Exposure of a
degassed THF-d8 solution of [3(N2)]B(C6F5)4 to one
atmosphere of H2 resulted in complete conversion to trans-
[FeH(H2)(PEtNMePEt)(dmpm)]+ ([3(H2)]+),22a over the
course of 48 h as monitored by 31P NMR, with multiplets at
δ 49.9 and 3.4 and a very broad peak at δ −8.7 in the 1H NMR
spectrum (eq 6). Removal of the dihydrogen atmosphere and
introduction of an atmosphere of 15N2 resulted in complete
conversion back to [3(15N2)]B(C6F5)4 over the course of 3 h,
[2(N2)]+, generating [FeBr(CO)(PEtNMePEt)(dmpm)]+ ([2-
(CO)]+), indicating that the trans halide ligands are less
effective at stabilizing the coordinated N2 compared to the
hydride. Halides are poor σ-donors and better π-donors than
E
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Crystallization of [3(15NH3)]B(C6F5)4 by the vapor
diffusion of pentane into fluorobenzene under an argon
atmosphere resulted in X-ray quality crystals. Crystallographic
analysis of [3(15NH3)]B(C6F5)4 confirmed the presence of the
Fe bound NH3 ligand, with an Fe−NH3 bond distance of
2.111(3) Å (Figure 8). The observed Fe−NH3 bond distance in
indicating that the binding of both N2 and H2 is reversible. The
increased rate of binding dinitrogen suggests a higher lability of
the dihydrogen adduct and that [FeH(PEtNMePEt)(dmpm)]+
has a slight preference for dinitrogen binding.
Treatment of a pale yellow, degassed THF-d8 solution of
[3(N2)]B(C6F5)4 with one atmosphere of 15NH3 resulted in a
change to dark yellow after 30 min at 25 °C (eq 7). Complete
Figure 8. Molecular structure of the cation [FeH(15NH3)(PEtNMePEt)-
(dmpm)]+ ([3(15NH3)]+). Thermal ellipsoids are drawn at 30%
probability. The hydrogen atoms (except for the Fe−H and NH3) and
−
conversion to [FeH(15NH3)(PEtNMePEt)(dmpm)]+ ([3-
B(C6F5)4 anion are omitted for clarity. Selected bond distances (Å)
1
(15NH3)]+) required about 5 h as determined by 31P and H
and angles (deg): P(1)−Fe(1) = 2.209(1), P(2)−Fe(1) = 2.195(1),
P(3)−Fe(1) = 2.210(1), P(4)−Fe(1) = 2.207 (1), Fe(1)−N(2) =
2.111(3), Fe(1)−H(1A) = 1.42(3), P(2)−Fe(1)−P(1) = 90.87(4),
P(3)−Fe(1)−P(4) = 73.69(1), N(2)−Fe(1)−H(1A) = 178.9(13).
NMR spectroscopy. Prolonged exposure (beyond 12 h) to an
excess of NH3 resulted in dissociation of the phosphine ligands
from the Fe center. 31P NMR resonances shifted from δ 38.8
and −0.8 for [3(N2)]B(C6F5)4 to 47.9 and 2.6 for [3(15NH3)]-
B(C6F5)4, and a sharp resonance in the 15N NMR spectrum at
−433.1 confirmed the presence of a bound 15NH3 ligand.
Additionally, a concomitant shift of the hydride resonance in
the 1H NMR spectrum from δ −14.5 to −25.5 indicated ligand
exchange had occurred. Protons of the bound amine ligand
were anticipated to resonate in the upfield region around δ
−0.1. For example, in previously reported Fe-NH3 complexes
of this type, the NH3 protons were assigned as a broad singlet
at δ −1.61 and −0.09 for trans-[Fe(H)(NH3)(dmpe)2]-
[C13H9]34 and [Fe(H)(NH3)(dmpe)2][BPh4]35 (both in
THF-d8), respectively, and −0.86 for trans-[Fe(H)(NH3)((P-
(C3H6OMe)2CH2)2)2][BPh4] in C6D6.36 However, for [3-
(15NH3)]B(C6F5)4, no resonances were found in this region of
[3(15NH3)]+ is in accord with the Fe−NH3 distance reported
for [FeH(NH3)(dmpe)2]OH (2.103 Å)34 and is within the
range of other previously reported Fe−NH3 complexes
(2.008−2.190 Å).37 In addition, the six-membered ring of the
PEtNMePEt ligand is in the chair configuration with the pendent
amine positioned exo with respect to the NH3. Thus, in the
solid state there does not appear to be an interaction between
the pendent methylamine group and the protons of the Fe
bound NH3. However, in the crystal lattice a hydrogen bonding
interaction (∼3.2 Å) was observed between the fluorine
substituents of the B(C6F5)4 anion and the protons of the Fe
bound NH3 group (see Supporting Information). The
remaining bond lengths and angles of [3(15NH3)]B(C6F5)4
were found to be similar to those of [3(N2)]B(C6F5)4 (see
above).
1
the H NMR spectrum. Further analysis of the sample by a
1H−15N HSQC correlation experiment located the 1H
resonance of the bound amine ligand at δ 1.61, correlating to
the 15N signal at δ −433.1. A similar downfield resonance for
the amine protons was also observed for [3(15NH3)]BPh4,
which eliminated a hydrogen bonding interaction with the
fluorine atoms in the anion (see description of solid state
structure below) as the cause of the unusually downfield
resonance for the amine protons. It is unclear if an interaction
with a pendent amine group in the PEtNMePEt ligand is
responsible for this observation.
Electrochemistry of Ferrous-N2 Complexes. The
electrochemistry of [3(N2)]BPh4 was investigated and
compared to [FeH(N2)(dmpe)2]BPh4,23 to further understand
the effect of pendent amines on the proton transfer chemistry
of the N2 reduction. [FeH(N2)(dmpe)2]BPh4 has been
previously shown to produce NH3 in 4% yield when treated
with H2SO4.11a The cyclic voltammogram of [FeH(N2)-
(dmpe)2]BPh4 exhibits an irreversible reduction wave assigned
as the FeII/I couple at −1.89 V in CH2Cl2. All reduction
potentials are referenced to Cp2Fe0/+ at 0 V using Cp*2Fe0/+
(E1/2 = −0.53 V) as a reference. An irreversible peak
corresponding to the oxidation of the iron hydride is observed
at −0.60 V in CH2Cl2. For comparison, the irreversible FeII/I
reduction wave of [3(N2)]BPh4 at −24 °C (−2.58 V) was
observed at a 0.25 V more negative potential than that of
[FeH(N2)(dmpe)2]BPh4 (see the Supporting Information for
the voltammogram).
In contrast to the rapid ligand exchange for [3(15NH3)]B-
(C6F5)4, Tyler and co-workers reported that the treatment of a
THF solution of trans-[FeH(N2)((P(C3H6OMe)2CH2)2)2]+
with a saturated NH3 solution required two days to obtain
full conversion to trans-[FeH(NH3)((P(C3H6OMe)2CH2)2)2]+
and that an argon atmosphere is required for this reaction to
occur.36 Investigating the ligand exchange reaction in the
opposite direction, exposure of a THF-d8 solution of isolated
[3(15NH3)]B(C6F5)4 to 1 atm of 15N2 resulted in a mixture of
[3(15NH3)]B(C6F5)4 and [3(N2)]B(C6F5)4 with an equili-
brium constant at 25 °C of ∼2.5.
To study the effect of the added acid on the FeII/I redox
couple, cyclic voltammetry of [FeH(N2)(dmpe)2]BPh4 and
[3(N2)]BPh4 was performed in the presence of acid. The
F
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
addition of one equivalent of [H(OEt2)2][B(C6F5)4] to
methylene chloride solutions of [FeH(N2)(dmpe)2]BPh4
resulted in a proton reduction wave that is attributed to the
direct reduction of H+ to H2 by the electrode (Figure 9), with
in redox potential. Similar results were observed by Sellmann
and Sutter where protonation or alkylation of the thiolate of
[Fe(CO)(SC6H4SC2H4NHC2H4SC6H4S)] was found to shift
the redox potential by 0.50−0.70 V per protonation or
alkylation.5c The larger shift in redox potential observed in
Sellmann and Sutter’s complexes is attributed to the closer
proximity of the proton/alkyl group to the iron center.
Protonation Reactions to Form NH3. Reports of
ammonia production by protonation of ferrous-N2 complexes
are rare, and yields of ammonia are low (<4%, with respect to
the metal, in the case of the reaction of trans-[FeH(N2)-
(dmpe)2]+ with H2SO4).11a Rather than stoichiometric
reduction of N2 to ammonia, more common observations
include liberation of N2 and H2, or no reaction.38 Higher yields
(up to 82% with respect to the metal) of ammonia upon
protonolysis with strong acids such as HCl, H2SO4, or triflic
acid have been reported for low-valent Fe0 dinitrogen
complexes supported by a variety of chelating diphosphine
ligands such as Fe(N2)(dmpe)2,2 where the electrons are
supplied by the low valent iron center.
To probe whether the ferrous-N2 complexes supported by
diphosphine ligands containing a pendent amine could generate
NH3, we examined the reaction of [3(N2)]+ with excess strong
acid. A 0.01 M [3(N2)]BPh4 in THF solution was treated with
50 equivalents of concentrated H2SO4, resulting in a lighter
yellow colored solution and the formation of a white solid.
1
After 3 h, a H NMR spectrum of the reaction products in
DMSO-d6 showed no ammonium (7.0 ppm).13 It was reported
that [FeH(N2)(dmpe)2]BPh4 gave 4% ammonia under similar
conditions,11a but in our hands, treatment of [FeH(N2)-
(dmpe)2]BPh4 with 50 equivalents of H2SO4 as described
+
above yielded only protonated dmpe; NH4 was not detected
by NMR spectroscopy. A similar result was observed by
Henderson, where only a protonated ligand was observed
following the addition of HCl to [FeH(N2)(depe)2]BPh4.38a
Attempts to Synthesize Fe(0)-N2 Complexes. Due to
the absence of reactivity toward H+ at the N2 ligand in the FeII
complexes, we examined the reduction of these complexes, with
the targets being the Fe0−N2 complexes Fe(N2)(dmpm)-
(PEtNMePEt) or Fe(N2)(PEtNMePEt)2. Reduction of green
solutions of 1(Cl) or 2(Br) with Mg or KC8 resulted in the
formation of an orange product. 31P NMR spectra (THF-d8) of
the reaction mixture showed the presence of free dmpm and
four multiplets at δ 14.9, 40.6, 53.2, and 72.6, consistent with
Figure 9. Cyclic voltammograms showing the effect of acid on (top)
[FeH(N2)(dmpe)2]BPh4 (25 °C) and (bottom) [3(N2)]BPh4 (−24
°C), in CH2Cl2. Acid = [H(OEt2)2][B(C6F5)4]. Electrolyte = 0.1 M
[Bu4N][B(C6F5)4]; scan rate = 100 mV/s.
1
four inequivalent phosphorus atoms. The H NMR spectrum
an onset of the proton reduction occurring at a potential of
−2.26 V. As indicated above, when one equivalent of acid is
added to [3(N2)]BPh4, the protonation occurs on the pendent
amine as opposed to the N2 ligand, and through cyclic
voltammetry experiments we aimed to probe the effect of this
protonation on the reduction potential of the FeII/I couple. The
protonation reaction was undertaken at −24 °C because
[3(N2)H]2+ is more stable at lower temperatures. The addition
of one equivalent of [H(OEt2)2][B(C6F5)4] to [3(N2)]BPh4
resulted in a shift of the onset potential of the reduction wave
corresponding to the FeII/I couple to shift about 0.33 V to more
positive potentials (Figure 9). To compare the effect of
protonation of the pendent amine on the redox potential of the
iron center, we also examined the redox chemistry of
[3(CO)]+. Unlike the IR studies described above that resulted
in almost identical vibrational frequency shifts upon proto-
nation of the pendent amine for both [3(N2)]+ and [3(CO)]+,
the redox potential for the FeII/I couple was found to differ by
0.10 V, where the protonation of [3(N2)]+ exhibits a larger shift
contains a multiplet at δ −13.86, indicating the presence of a
hydride ligand bound to the Fe center, and a medium intensity
νFeH band was observed at 1900 cm−1 (KBr) in the IR
spectrum. Dehydrohalogenation reactions by the addition of
one or more equivalents of KOtBu to 1(H) and 2(H) resulted
in the formation of the same Fe−H product. Due to the
expected trigonal bipyramidal coordination geometry of the
five-coordinate Fe0−N2 complex, it is not surprising that dmpm
(bite angle ∼73° in 1(Cl)) dissociates from the metal center
upon the change in geometry. The reduction of FeBr2 with Mg
in the presence of two equivalents of PEtNMePEt (bite angle
∼92° in in 1(Cl)) was attempted as a synthetic route to
Fe(N2)(PEtNMePEt)2. However, instead of detecting the
expected intense νNN band of this orange product in the IR
spectrum, the same νFeH band was observed at 1900 cm−1, and
31P and 1H NMR spectra matched the spectroscopic data of the
reduction and dehydrohalogenation product of 2(Cl) and
2(H), respectively (Scheme 2). The presence of the Fe−H in
G
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
reversible, and attempts to avoid C−H activation of the
methylene group of the PEtNMePEt ligand by performing the
reduction of FeBr2 with Mg and two equivalents of PEtNMePEt
under elevated N2 pressures (up to 40 atm) were unsuccessful,
resulting in the production of 4.
Scheme 2. Reactions That Form Complex
HFe(Et2PC(H)N(Me)CH2PEt2)(PEtNMePEt) (4)
Computational Analysis. Computational analysis employ-
ing density functional theory (DFT) based electronic structure
methods was used to provide insight into N2 binding, acid
reactivity, and redox potentials of the Fe complexes. We first
examined the binding affinity of N2 in [3(N2)]+ and [1(N2)]+
as well as the protonated species [3(N2)H′]2+ and [1(N2)H′]2+
(Figure 11). The calculated free energy binding affinity of N2 in
the product suggests oxidative addition of the C−H bond of the
PEtNMePEt ligand to Fe occurred as did ligand redistribution in
the FeX2(PEtNMePEt)(dmpm) and FeHX(PEtNMePEt)(dmpm)
reactions which contain free dmpm. The reaction could be
envisioned to occur by deprotonation at the methylene position
followed by metal coordination, or by deprotonation of the
hydride, followed by oxidative addition of the C−H bond.
Crystallographic analysis of the orange product described
above confirmed the molecular structure, as shown in Figure
10. Oxidative addition of the C−H of a methylene group in the
Figure 11. Possible products from protonation of [3(N2)]+ and
[1(N2)]+.
Table 1. Binding Affinity of N2 and CO to Iron Complexes
Containing a Protonated and Non-Protonated Pendent
Amine
Figure 10. Molecular structure of complex HFe(Et2PC(H)N(Me)-
CH2PEt2)(PEtNMePEt), 4. Thermal ellipsoids are shown at 30%
probability. Selected bond distances (Å) and angles (deg): P(1)−
Fe(1) = 2.152(1), P(2)−Fe(1) = 2.183(1), P(3)−Fe(1) = 2.149(1),
P(4)−Fe(1) = 2.171(1), Fe(1)−C(7) = 2.068(1), Fe(1)−H(1F) =
1.479(18), C(7)−Fe(1)−P(2) = 49.69(4), C(7)−Fe(1)−P(1) =
83.09(4), P(2)−Fe(1)−P(1) = 124.68(18), P(3)−Fe(1)−P(4) =
91.41(12).
binding affinity of ligand to Fe complexes (kcal/mol)
[3(N2)]+/
[1(N2)]+/
[3(N2)H′]2+/
[1(N2)H′]2+/
ligand
[3(CO)]+
[1(CO)]+
[3(CO)H′]2+
[1(CO)H′]2+
N2
10.1
36.0
7.4
8.4
6.4
CO
33.2
33.9
32.1
PEtNMePEt ligand backbone produced the neutral iron hydride
complex HFe(Et2PC(H)N(Me)CH2PEt2)(PEtNMePEt) (4).
Coordination of the methylene carbon to the iron center
forms a three-membered ring, where the methylene carbon
bound to the Fe center is trans to a phosphorus atom from the
other PEtNMePEt ligand. The unmodified PEtNMePEt ligand in
this complex has a bite angle of 91.41(12)°. Similar C−H
activation of the methylene group in a phosphine based
ancillary ligand was identified by Karsch and co-workers, in
which an Fe0 complex containing two (Me2PCH2)2PMe ligands
undergoes intramolecular oxidative addition, forming a
hydrido−FeII complex.39 Furthermore, in a related example
describing the preparation of an Fe0−N2 complex, reduction of
FeCl2(depe)2 with sodium-naphthalene under an argon
atmosphere formed Fe(depe)2, which undergoes oxidative
addition of a C−H bond of the ethyl group to form the
hydrido−FeII complex. This process was reversible, as this
product was readily converted to Fe(N2)(depe)2 when placed
[3(N2)]+ and [1(N2)]+ is 7.4 and 10.1 kcal/mol (Table 1),
respectively. The small N2 binding affinity indicates that N2 is
weakly coordinated. For comparison, the binding affinity of CO
for the corresponding species, [3(CO)]+, [1(CO)]+, [3(CO)-
H′]2+, and [1(CO)H′]2+ was also determined (Table 1).
Compared to N2, the binding affinity of CO in [3(CO)]+ and
[1(CO)]+ is 36.0 and 33.2 kcal/mol (Table 1) respectively,
which is more favorable by 26 kcal/mol than N2 binding. The
much larger binding affinity of CO vs N2 and a larger binding
affinity of N2 in [3(N2)]+ than in [1(N2)]+ verify the
experimental observations that CO was found to displace N2
from [3(N2)]+ and [1(N2)]+ but with a slower exchange rate in
[3(N2)]+ than in [1(N2)]+.
To address the relative binding/ligand strengths of N2 and
CO, we follow a similar methodology as our previous work on
Cr−N2 complexes, where an NBO based fragment analysis is
employed.21b This approach avoids well-known pitfalls of
population analysis performed with large basis sets and
40
under N2. In the present case, the C−H activation was not
H
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
additionally affords energy values for the various orbital
interactions to provide a quantitative description of the
bonding.41 The FeII(d)−N2(σ) interaction is found to be
about 150 kcal/mol for complexes [3(N2)]+ and [1(N2)]+,
whereas the FeII(d)−N2(π*) interaction is found to be
appreciably smaller, ca. 10−12 kcal/mol. Thus, N2 can be
viewed as mainly a σ-donor but exhibits negligible π-acceptor
capability for these FeII species. In comparison, CO is found to
be a stronger σ-donor (FeII(d)−CO(σ) interaction energy
about 220 kcal/mol) and a stronger π* acceptor (FeII(d)−
CO(π*) interaction energy about 32 kcal/mol). A similar
analysis on hypothetical Fe0 complexes Fe(L)(PEtNMePEt)-
(dmpm) (L = N2, CO) further confirms our bonding
assessment. In the Fe0 complexes, the σ-bonding interaction
energy decreases to 99 and 215 kcal/mol for N2 and CO,
respectively, in accord with the expectation that Fe0 will be a
weaker acceptor than FeII. Likewise, the Fe0(d)−L(π*)
interaction energy increased to 37 and 63 kcal/mol for N2
and CO, in accord with the expectation that Fe0 will be a
stronger π-donor. This description of binding is in qualitative
accord with the preference for CO to bind to the Fe complexes
described above, suggesting CO is also a poor π-acceptor for
FeII species and an appreciably stronger π-acceptor on Fe0. In
contrast, N2 has negligible π-acceptor strength on FeII and acts
as only a moderate π-acceptor on Fe0.
Table 2. Comparison of Experimental and Computational IR
Shifts for Protonation of Complexes [1(N2)]+, [3(N2)]+,
[1(CO)]+, and [3(CO)]+
In accord with this bonding picture, there is negligible charge
transfer (≪ 0.1e−) from the metal to the N2 in [3(N2)]+. Thus,
to explain the blue shift of the IR frequency and observed
lengthening on the N−N bond, we resort to a qualitative
picture where these observations are attributed to the mixing of
π/π* states of N2 upon binding to an electron-rich metal
center. As in our work on Cr−N2 complexes,21b it was found
that a simple electrostatic model where a negative charge on the
metal/phosphine core polarizes the bound N2 results in a
lowering of N−N bond order (see Supporting Information for
an analysis of N−N bond order, bond length increase, and IR
frequency shift from an electrostatic model). Interpretation of
our results on N2 and CO binding in this context is more
appropriate to explain our observations than the more
traditional approach based on metal-π* back-bonding, which,
as outlined above, has negligible energetic contribution in the
FeII species.
a
This IR shift corresponds to alkylation of a sulfur atom with Et+.5c
b
This corresponds to the IR shift of protonation of the metal center.30
c
This corresponds to the IR shift of protonation of the N2 ligand.30
WBI of the C−O increased from 2.05 to 2.10 while the WBI of
the Fe−C decreased from 0.74 to 0.70 upon protonation. The
small change in bond order is consistent with the small shift in
the IR bands and the small decrease in binding affinity of N2
and CO. Further electron density analysis (Figure 12) shows
that the change in electron density upon protonation mainly
occurred at the local region where the proton resides (pendent
For the protonated species, [3(N2)H′]2+ and [1(N2)H′]2+,
where the protons reside on the pendent amine, the calculated
binding affinity of N2 was found to be ∼1−2 kcal/mol lower
than in nonprotonated complexes. Similar results were found
for the corresponding CO analogues, [3(CO)H′]2+ and
[1(CO)H′]2+. Harmonic vibrational analysis showed that
when protons were added to complexes [3(N2)]+ and [1(N2)]+
and complexes [3(CO)]+ and [1(CO)]+, the νNN and νCO
bands increased by 20 and 22 and by 23 and 25 cm−1,
respectively (Table 2). As a comparison to the IR shifts of
related N2 and CO containing Fe complexes, protonation of the
thiolate sulfur in [Fe(CO)(SC6H4SC2H4NHC2H4SC6H4S)] or
to either the Fe center or N2 ligand in Fe(N2)(dmpe)2 resulted
in a more dramatic increase in IR shifts of 395c and 120 and a
decrease in IR shift of 395 cm−1,30 respectively, Table 2.
Further analysis of N2 and CO ligands in [3(N2)]+ and
[3(CO)]+ by evaluating the Wiberg bond index (WBI),
obtained from a NBO analysis, indicated the N−N bond
order increased from 2.70 to 2.74 upon protonation while the
WBI of Fe−N decreased from 0.48 to 0.44 upon protonation.
Similar trends are observed for the CO analogues, where the
Figure 12. The electron density difference (isovalue = 0.0012)
between the nonprotonated species and with a proton at the pendent
amine. (a) [3(N2)]+ and (b) [3(CO)]+. Depletion of electron density
is represented by red, while increased electron density is represented
by blue. Methyl and methylene protons are removed for clarity.
Phosphorus = orange, nitrogen = light blue, carbon = gray, hydrogen =
white, iron = green, and oxygen = red.
I
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
amine). There is a small change in electron density of the N2
and CO ligands, where the electron density decreased slightly
at distal N and O while it slightly increased at the proximal N
and C. From these observations, coupled with our above
analysis of the Fe−N2 and Fe−CO bonding, the interaction of
N2 with Fe can be understood as an electrostatic effect, where
protonation at the pendent amine depolarizes the N2 ligand and
thus strengthens the N−N π bond and weakens the overall
interaction between the N2 and Fe center.
move from the pendent amine to the iron center/N2 ligand
and to the chloride ligand, respectively (Figure 13). The larger
degree of electron density present on the N2 ligand upon
protonation of [3(N2)]+ is attributed to the increased basicity
of the N2 ligand of [3(N2)H′]+ vs [1(N2)H′]+ (see below and
Supporting Information). This also indicates that by increasing
the electron density at the metal center by changing the ligand
trans to dinitrogen that the electron density can be increased at
the N2 ligand, to encourage proton transfer from the pendent
amine to the dinitrogen ligand.
Modulating the basicity of the iron-bound N2 ligand and the
pendent amine in complexes similar to [3(N2)]+ is important in
controlling the reactivity toward protons. Therefore, a
quantitative assessment of the preferred protonation sites in
[3(N2)]+ is valuable for interpreting the reactivity and
spectroscopic data. To quantify the basicity of the N2 ligand
in the ferrous iron complexes and to determine the conditions
required for proton transfer to the N2 ligand from the pendent
amine, we calculated the proton affinity of nitrogen on the
pendent amine and the distal nitrogen atom of coordinated N2.
The proton affinity of the pendent amine of [3(N2)]+ was
calculated to be 43.5 kcal/mol higher than that of the N2 ligand
(31.9 pKa units more basic for [3(N2)H′]2+ versus [3(N2)-
H″]2+). In the case of [1(N2)]+, the proton affinity of the
pendent amine is lower than that of [3(N2)]+ (0.9 kcal/mol or
0.7 pKa units more acidic for [1(N2)H′]2+ versus [3(N2)-
H′]2+), and the pendent amine was calculated to be 52.8 kcal/
mol higher (or 38.7 pKa units more basic for [1(N2)H′]2+
versus [3(N2)H″]2+) than the N2 ligand. Therefore, it is not
surprising that the pendent amine was protonated when
protons were added to these complexes. Replacement of the
trans chloride ligand by hydride increases the proton affinity
(basicity) of pendent amine and N2 ligands. This indicates that
the introduction of a better electron donating ligand results in
an increase of the electron density at the metal center, which in
terms of our picture of Fe−N2 bonding based on an inductive
effect would increase the polarization of the bound N2 and
render it more basic.
The high acidities of the protonated N2 ligands of
[3(N2)H″]2+ and [1(N2)H″]2+ indicate that PCET will be
required to promote the reduction of N2 to NH3. To probe the
effect of PCET on [3(N2)]+, we calculated the redox potential
for the FeII/I couple of [3(N2)]+ and [3(N2)H′]2+. The
calculated FeII/I redox potential of [3(N2)H′]2+ shifts to a more
positive value by 0.44 V, which is comparable to the
experimental value of 0.33 V. For comparison, we also
calculated the redox potential for the FeII/I couple of [3(CO)]+
and [3(CO)H′]2+. The calculated FeII/I redox potential of
[3(CO)H′]2+ shifts to a more positive value 0.11 V less than
observed with [3(N2)H′]2+, which is almost identical to the
values measured experimentally.
To promote proton transfer to the N2 ligand from the
pendent amine, the low valent complex, Fe0(N2)(PEtNMePEt)-
(dmpm), was probed computationally. The pKa of [Fe0(N2)-
(PEtNMeHPEt)(dmpm)]+ increased by 11 pKa units vs [3(N2)-
H′]2+, while the pKa of [Fe0(HN2)(PEtNMePEt)(dmpm)]+
dramatically increased by about 43 pKa units. The reduction
of FeII to Fe0 increases the basicity of the N2 ligand to the same
level as that of pendent amine, which would allow a facile
proton transfer to the N2 ligand from the pendent amine.
However, as described above, reduction of 1(Cl) or
dehydrohalogenation of 1(H) led to C−H activation of the
methylene group of the PEtNMePEt ligand resulting in complex
4. Computationally, oxidative addition of the methylene group
was found to be 4 kcal/mol more favorable than N2 binding to
Fe(PEtNMePEt)2, thus explaining the experimental results where
C−H activation was observed at elevated N2 pressure.
To further examine the effect of the trans ligand on the
basicity of the N2 ligand upon protonation, the frontier
molecular orbitals of [3(N2)]+ and [1(N2)]+ were determined
and compared to the frontier molecular orbitals of [3(N2)H′]2+
and [1(N2)H′]2+. It was found that the HOMO resided on the
pendent amine for [3(N2)]+ and [1(N2)]+ (Figure 13 and
Supporting Information). Upon protonation of [3(N2)]+ and
[1(N2)]+ at the pendent amine, resulting in complexes
[3(N2)H′]2+ and [1(N2)H′]2+, respectively, the HOMOs
CONCLUSIONS
■
Ferrous dinitrogen complexes supported by the diphosphine
ligand PEtNMePEt containing a pendent amine base have been
synthesized and characterized. These complexes undergo
reactions with acid, resulting in the protonation of the pendent
amine with subsequent loss of N2 rather than protonation at
the N2 ligand. Protonation of these ferrous-dinitrogen
complexes at low temperature results in stabilization of the
protonated complexes, and the addition of base deprotonates
the pendent amine, regenerating the starting FeII(N2) complex.
Similar results were observed for the related carbonyl
complexes, except that the protonated complexes were stable
at 25 °C. Although the incorporation of pendent amines in
ferrous-dinitrogen complexes does not lead to proton transfer
to the N2 ligand, protonation of the pendent amine shifts the
reduction potential of the FeII/I couple to more positive
potentials. Protonation of the pendent amine resulted in
strengthening of the N−N bond (higher νN≡N by 25 cm−1) of
the dinitrogen ligand. This finding has implications in regards
to the nature of the recently proposed N2 binding site of
nitrogenases,7e as protonation of a pendent base results in a
decrease of the N2 binding affinity. Thus, the electrostatic
effects of nearby protons and the role of PCET in the
Figure 13. Comparison of the HOMOs of (left) [3(N2)]+ and (right)
[3(N2)H′]2+. Methyl and methylene protons are removed for clarity.
Phosphorus = orange, nitrogen = light blue, carbon = gray, hydrogen =
white, iron = green.
J
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
reactions contain ∼0.03 M iron complex in a volume of 1−2 mL
solvent. IR spectra were collected in intervals of 15 s in the normal
collection mode or in “rapid collect” mode at a rate of five scans per
second. Complexes [3(CO)]+ and [1(CO)]+ were prepared according
to literature procedures.22a
stabilization of N2 binding and reduction to NH3 is worthy of
consideration in the mechanistic discussion of nitrogenase and
in the design of molecular catalysts for N2 reduction. Chemical
reduction of the reported ferrous complexes resulted in
oxidative addition of the C−H bond of the methylene carbon
of the PEtNMePEt ligand as opposed to forming an Fe0−N2
complex. These experimental results were validated computa-
tionally, and it was shown that ferrous-dinitrogen complexes are
incapable of aiding in the protonation of coordinated N2, as the
pKa determination of protonation at the coordinated N2 was
found to be highly unfavorable; less basic pendent amines are
needed. Computational results also show that incorporation of
a strong electron donating ligand trans to the dinitrogen ligand
or a more electron-rich metal can be used to tune the basicity of
the dinitrogen ligand. Further investigation of the tuning of
pendent amines to facilitate proton transfer to low-valent
dinitrogen complexes is currently under intense investigation in
our laboratory.
FeBr2(PEtNMePEt)(dmpm) (2(Br)). A similar procedure to the
previously reported procedure for FeCl2(PEtNMePEt)(dmpm) was
used, where FeBr2 was used in place of FeCl2.22a 2(Br) was isolated as
yellowish-green colored crystals (see Supporting Information for
structural data). Yield: 82%. The paramagnetism of 2(Br) resulted in
broad resonances in the 1H NMR spectrum, which were
uninformative, and there were no resonances in the 31P NMR
spectrum. Anal. Calcd for C16H41Br2FeNP4: C, 32.73; H, 7.04; N, 2.39.
Found C, 32.78; H, 6.95; N, 2.44.
FeHBr(PEtNMePEt)(dmpm) (2(H)). A similar procedure to the
previously reported procedure for FeHCl(PEtNMePEt)(dmpm) was
used, where FeBr2(PEtNMePEt)(dmpm) was used in place of
FeCl2(PEtNMePEt)(dmpm).22a 2(H) was isolated as feathery orange
colored crystals (see Supporting Information for structural data).
1
2
Yield: 48%. H NMR (500 MHz, CD2Cl2): −27.03 (pentet, JHP
=
48.2 Hz, 1H), 1.00 (m, 6H, dmpm-P(CH3)2), 1.07 (m, 6H, dmpm-
P(CH3)2), 1.35 (m, 4H, P(CH2CH3)2), 1.43 (m, 6H, PEtNMePEt-
P(CH2CH3)2), 1.63 (m, 6H, PEtNMePEt-P(CH2CH3)2), 1.67 (m, 4H,
PEtNMePEt-P(CH2CH3)2), 1.95 (m, 2H, Et2PCH2NMeCH2PEt2), 2.34
(s, 3H, N−CH3), 2.68 (m, 2H, Et2PCH2NMeCH2PEt2), 3.01 (m, 1H,
Me2PCH2PMe2), 3.31 (m, 1H, Me2PCH2PMe2). 31P{1H} NMR (202
MHz, THF-d8): 0.61 (m, 2P, dmpm), 45.44 (m, 2P, PEtNMePEt). Anal.
Calcd for C16H42BrFeNP4·1/4OC4H8: C, 38.80; H, 8.43; N, 2.66.
Found C, 38.68; H, 8.44; N, 2.83.
EXPERIMENTAL SECTION
■
Synthesis and Materials. All synthetic procedures were
performed under an atmosphere of N2 using standard Schlenk or
glovebox techniques. Unless described otherwise, all reagents were
purchased from commercial sources and were used as received.
Solvents were dried by passage through activated alumina in an
Innovative Technology, Inc., PureSolv solvent purification system.
Deuterated NMR solvents THF-d8, toluene-d8, and benzene-d6 were
purchased from Cambridge Isotope Laboratories, dried over NaK, and
vacuum transferred before use. K[B(C6F5)4] was purchased from
Boulder Scientific. Ferrous chloride, ferrous bromide, and 1,2-
bis(dimethylphosphino)methane were purchased from Strem Chem-
icals, Inc. and used as received. Magnesium powder, paraformaldehyde
(95%), methylamine in ethanol (33 wt %), and HCl (2 M in ether)
were purchased from Sigma-Aldrich. FeCl2(PEtNMePEt)(dmpm),22a
FeHCl(PEtNMePEt)(dmpm),22a PEtNMePEt,42 and [NBu4][B(C6F5)4]43
were synthesized according to literature preparations. [H(OEt2)2][B-
(C6F5)4] was prepared according to a modified literature procedure,44
see below. NMR spectra were recorded on a Varian Inova or NMR S
[Fe(N2)Cl(PEtNMePEt)(dmpm)]B(C6F5)4 ([1(N2)]B(C6F5)4). A 1 mL
fluorobenzene solution of 10 mg (0.020 mmol) of FeCl2(PEtNMePEt)-
(dmpm) was treated with 14.5 mg (0.020 mmol) of KB(C6F5)4 in 0.25
mL of fluorobenzene. A pinkish-red solution resulted after stirring for
2 min. These solutions were used subsequently in protonation studies.
IR (fluorobenzene): 2111 cm−1 (Fe−N2).
[FeBr(PEtNMePEt)(dmpm)]B(C6F5)4. A 5 mL fluorobenzene
solution of 30 mg (0.06 mmol) of FeBr2(PEtNMePEt)(dmpm) was
treated with 44.0 mg (0.06 mmol) of KB(C6F5)4 in 5 mL of
fluorobenzene. A pinkish-red solution resulted after stirring for 2 h.
The reaction mixture was filtered and concentrated to about 1 mL.
Red crystals were grown by vapor diffusion of pentane into a
fluorobenzene solution. Yield: 42 mg (63%). Anal. Calcd for
C40H41BBrF20FeNP4·C6H5F: C, 43.09; H, 3.62; N, 1.09. Found C,
43.44; H, 3.46; N, 1.32.
1
500 MHz spectrometer. H chemical shifts are referenced to residual
protio impurity in deuterated solvent. 31P chemical shifts are proton
decoupled unless otherwise noted and referenced to H3PO4 as an
external reference. 15N NMR chemical shifts were externally
referenced to CH315NO2 (δ = 0). Infrared spectra were recorded on
a Thermo Scientific Nicolet iS10 FT-IR spectrometer at ambient
temperature and under a purge stream of nitrogen gas. Solid-state FT-
IR samples were prepared as KBr pellets. Elemental analysis was
performed by Atlantic Microlabs, Norcross, GA. Cyclic voltammetry
was performed in a Vacuum Atmospheres Nexus II glovebox under a
N2 atmosphere using a CH Instruments model 620D or 660C
potentiostat in THF using 0.2 M [NBu4][B(C6F5)4] as the supporting
electrolyte. Measurements were performed using a standard three-
electrode cell containing a 1 mm PEEK-encased glassy carbon working
electrode, Cypress Systems EE040, a 3 mm glassy carbon rod (Alfa) as
the counter electrode, and a silver wire suspended in electrolyte
solution and separated from the analyte solution by a Vycor frit (CH
Instruments 112) as the pseudoreference electrode. Prior to the
acquisition of each voltammogram, the working electrode was polished
using 0.1 μm γ-alumina (BAS CF-1050) and rinsed with THF.
Decamethylferrocene was used as an internal reference, and all
potentials are reported versus the ferrocenium/ferrocene couple at 0.0
V. Low-temperature voltammetry experiments were performed in an o-
xylene (−24 °C) slush bath in a glovebox coldwell. In situ IR
experiments were recorded on a Mettler-Toledo ReactIR 15 FTIR
spectrometer equipped with an MCT detector, connected to a 1.5 m
AgX Fiber DS series (9.5 mm × 203 mm) probe with a silicon sensor.
Experiments were performed in a 5 mL two-neck pear-shaped flask
under a dinitrogen atmosphere using Schlenk line techniques. Typical
[Fe(N2)Br(PEtNMePEt)(dmpm)]B(C6F5)4 ([2(N2)]B(C6F5)4). A 1 mL
fluorobenzene solution of 10 mg (0.017 mmol) of FeBr2(PEtNMePEt)-
(dmpm) was treated with 12.3 mg (0.017 mmol) of KB(C6F5)4 in 0.25
mL of fluorobenzene. A pinkish-red solution resulted after stirring for
2 min. These solutions were used subsequently in protonation studies.
IR (fluorobenzene): 2110 cm−1 (Fe−N2).
[FeCl(CO)(PEtNMePEt)(dmpm)]BPh4 ([1(CO)]B(C6F5)4). A 5 mL
THF solution of 50 mg (0.10 mmol) of FeCl2(PEtNMePEt)(dmpm)
was treated with 36 mg (0.11 mmol) of NaBPh4 in 5 mL THF. The
resulting orange-red solution was treated with a light bubbling of CO
(1 atm), immediately resulting in a yellow solution. The solution was
filtered, and the solvent was removed under reduced pressure. X-ray
quality crystals were grown from vapor diffusion of pentane into a
THF solution (see Supporting Information for structural data). The IR
characterization data were found to match [FeCl(CO)(PEtNMePEt)-
(dmpm)]2FeCl4, which was previously reported by DuBois and co-
workers.22a
[FeBr(CO)(PEtNMePEt)(dmpm)]BPh4 ([2(CO)]B(C6F5)4). A 5 mL
THF solution of 40 mg (0.07 mmol) of FeBr2(PEtNMePEt)(dmpm)
was treated with 24 mg (0.07 mmol) of NaBPh4 in 5 mL of THF. The
resulting yellow solution was treated with a light bubbling of a CO
atmosphere, immediately resulting in a yellowish-orange color. The
solution was filtered, and the solvent was removed under reduced
pressure. X-ray quality yellow crystals were grown from vapor diffusion
of pentane into a THF solution (see Supporting Information for
1
structural data). Yield: 36 mg (60%). H NMR (500 MHz, CD2Cl2):
K
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1.00−1.29 (m, 12H, dmpm-P(CH3)2), 1.34−2.22 (m, 20H, P-
(CH2CH3)2), 1.52 (s, 3H, N−CH3), 2.43 (s, 3H, Fe−NH3), 2.89
( m , 2 H , E t 2 P C H 2 N M e C H 2 P E t 2 ) , 3 . 1 7 ( m , 2 H ,
Et2PCH2NMeCH2PEt2), 3.49 (m, 2H, Me2PCH2PMe2), 7.30−7.47
(m, 20H, BPh4). 31P NMR (202 MHz, CD2Cl2): −12.2 (m, 2P,
dmpm), 25.8 (m, 2P, PEtNMePEt). IR (fluorobenzene) 1946 cm−1.
[FeH(N2)(PEtNMePEt)(dmpm)]BPh4 ([3(N2)]BPh4). A 5 mL THF
solution of 31 mg (0.065 mmol) of FeHCl(PEtNMePEt)(dmpm) was
treated with 22 mg (0.063 mmol) of NaBPh4 in 5 mL of THF. A
yellow solution resulted after stirring for 20 h. The cloudy yellow
solution was filtered and recrystallized from a vapor diffusion of
pentane into a THF solution, resulting in yellow crystals. Yield: 37 mg
FeH). IR (KBr): 1900 cm−1 (Fe−H). Anal. Calcd for
C22H54FeN2P4·Me3SiOSiMe3: C, 48.78; H, 10.31; N, 4.55. Found C,
48.33; H, 9.90; N, 5.10.
[H(OEt2)2][B(C6F5)4]. Ether (400 mL) was added to KB(C6F5)4
(25.29 g, 35.22 mmol), giving a cloudy off-white suspension. HCl (2
M in diethyl ether, 20 mL) was added via syringe over 5 min, and the
mixture was stirred overnight. Volatile components were removed
under reduced pressure, affording an oily solid. Ether (50 mL) was
added, and the volatiles were again removed, giving a clumpy off-white
solid. The solids were extracted with fluorobenzene (300 mL), then
filtered through Celite, and then through a syringe filter to give a clear,
pale brown-yellow solution. Pentane (300 mL) was added, and the
resulting off-white precipitate was collected with a medium porosity
frit and extracted with fluorobenzene (70 mL). The resulting pale
brown-yellow solution was layered with pentane (250 mL), affording
clear/colorless crystals after 24 h. The crystals were collected, rinsed
with pentane, and dried under a vacuum. Yield: 18.19 g (62%).
Spectroscopic analysis matched the previously reported data.44
Theoretical Calculations. All structures were fully optimized
without symmetry constraints using the B3P8645 functional as
implemented in Gaussian 09.46 The Stuttgart basis set with effective
core potential (ECP)47 was used for the Fe atom, and the 6-31G**
basis set48 was used for other nonmetal atoms. Each stationary point
was confirmed by frequency calculation at the same level of theory to
be a real local minimum on the potential energy surface without
imaginary frequency. All reported free energies are for THF solution at
the standard state (T = 298 K, P = 1 atm of N2, 1 mol/L concentration
of all species in THF) as modeled by a polarized continuum model
(C-PCM)49 with standard correction for (harmonic) vibrational,
rotational, and translational thermal free energy contributions. All
calculated proton affinities and pKa values are for THF solutions and
are calculated relative to the value of Et3NH+ (pKa = 12.5),50 which is
assigned its experimental value to anchor the calculated pKa scale.
X-Ray Diffraction Studies. X-ray diffraction data were collected
on a Bruker-AXS Kappa APEX II CCD diffractometer with 0.71073 Å
Mo Kα radiation. Selected crystals were mounted using NVH
immersion oil onto a nylon fiber and cooled to the data collection
temperature of 100−120 K. Unit cell parameters were obtained from
60 data frames, 0.5° Φ, from three different sections of the Ewald
sphere. Cell parameters were retrieved using APEX II software51 and
refined using SAINT+52 on all observed reflections. Each data set was
treated with SADABS53 absorption corrections based on redundant
multiscan data. The structure was solved by direct methods and
refined with the least-squares method on F2 using the SHELXTL
program package.54 Crystallographic data for each structure and details
regarding specific solution refinement for each compound are provided
in the Supporting Information.
1
2
(75%). H NMR (500 MHz, THF-d8): −14.53 (dt, JPH = 95.7, 47.9
Hz, 1H, Fe-H), 1.15 (m, 12H, dmpm-P(CH3)2), 1.43 (m, 4H,
P(CH2CH3)2), 1.60 (m, 6H, P(CH2CH3)2), 1.72 (m, 6H, P-
(CH2CH3)2), 1.83 (m, 4H, P(CH2CH3)2), 2.41 (s, 3H, N−CH3),
2.47 (m, 2H, Et2 PCH2 NMeCH2 PEt2 ), 3.00 (m, 2H,
Et2PCH2NMeCH2PEt2), 3.22 (m, 2H, dmpm-Me2PCH2PMe2), 6.71
3
3
(t, 4H, JHH = 6.8 Hz, p-phenyl protons of BPh4), 6.86 (t, 8H, JHH
=
7.8 Hz, o-phenyl protons of BPh4), 7.28 (m, 8H, m-phenyl protons of
BPh4). 31P{1H} NMR (202 MHz, THF-d8): 4.55 (m, 2P, dmpm),
44.88 (m, 2P, PEtNMePEt). 15N{1H} NMR (51 MHz, THF-d8) −41.75
2
(s, Fe−Nα≡Nβ), −59.80 (s, Fe-Nα≡Nβ), −361.82 (t, JNP = 13.5 Hz,
PEtNMePEt). Anal. Calcd for C40H62BFeN3P4: C, 61.95; H, 8.06; N,
5.42. Found C, 61.73; H, 7.87; N, 5.23.
In Situ Protonation of [1(N2)]B(C6F5)4, [2(N2)]B(C6F5)4, and
[3(N2)]B(C6F5)4. A 1.0 mL fluorobenzene solution of 10 mg (∼ 0.015
mmol) of 1(Cl), 2(Br), or 1(H) was treated with 11 mg (∼ 0.015
mol) of KB(C6F5)4 in 1.0 mL of fluorobenzene. A yellow (complex
[3(N2)]B(C6F5)4) or pinkish-red (complexes [1(N2)]B(C6F5)4 and
[2(N2)]B(C6F5)4) solution resulted after stirring for 2 min. After this
time, the ReactIR probe was inserted into the reaction mixture. The
solution was cooled in an acetonitrile slush bath (−45 °C) and stirred
for 30 min for the temperature to equilibrate. The addition of 0.5 mL
of 0.045 M [H(OEt2)2][B(C6F5)4] in fluorobenzene resulted in a
slight darkening of the solution. The solution was stirred for 30 min;
then 10 μL (0.045 mmol) Et3N was added and the solution was stirred
for 30 min.
In Situ Protonation of [1(CO)]B(C6F5)4, [2(CO)]B(C6F5)4,
[3(CO)]B(C6F5)4. The same procedure as described for the in situ
protonation of [1(N2)]B(C6F5)4, [2(N2)]B(C6F5)4, and [3(N2)]B-
(C6F5)4 was used, except that the solutions were treated with CO (1
atm) for 30 min before the acid addition. The acid and base additions
were undertaken at 25 °C, as the protonated CO complexes were not
temperature-sensitive, unlike the N2 complexes described above.
[FeH(15NH3)(PEtNMePEt)(dmpm)]B(C6F5)4 ([3(15NH3)]B(C6F5)4).
A 1.0 mL THF-d8 solution of 10 mg (∼ 0.015 mmol) of
[3(N2)]B(C6F5)4 was treated with 1 atm of 15NH3. A bright yellow
solution resulted after 30 min. Complete conversion to [3(NH3)]+ was
ASSOCIATED CONTENT
* Supporting Information
■
1
observed to occur after 5 h. H NMR (500 MHz, THF-d8): −25.0
S
2
(pentet, JPH = 46.2 Hz, 1H, Fe-H), 1.00−1.25 (m, 12H, dmpm-
ReactIR plots showing the reactivity of [3(N2)]B(C6F5)4 with
CO and [1(CO)]BPh4 and [2(CO)]BPh4 with protons;
molecular structures of 1(Cl), 2(Br), 2(H), [1(CO)]BPh4,
[2(CO)]BPh4, and [3(NH3)]B(C6F5)4; experimental details of
X-ray diffraction studies; 15N NMR of [3(N2)]+; multinuclear
NMR spectra of acid addition to [3(N2)]+; scan rate
dependence on the CV of [3(N2)]BPh4; frontier molecular
orbitals of [3(N2)]+ and [1(N2)]+; NBO analysis of Fe
complexes; charge model for N2 bond polarization; and
coordinates for optimized structures. This material is available
free of charge via the Internet at http://pubs.acs.org.
P(CH3)2), 1.40−2.1 (m, 20H, P(CH2CH3)2), 1.52 (s, 3H, N−CH3),
1.60 (br s, 3H, Fe−NH3), 2.39 (m, 2H, Et2PCH2NMeCH2PEt2), 2.69
(m, 2H, Et2PCH2NMeCH2PEt2), 3.04 (m, 1H, Me2PCH2PMe2), 3.25
(m, 1H, Me2PCH2PMe2). 31P{1H} NMR (202 MHz, THF-d8): 2.6
(m, 2P, dmpm), 47.9 (m, 2P, PEtNMePEt). 15N{1H} NMR (51 MHz,
2
THF-d8) −356.2 (t, JNP = 12.9 Hz, PEtNMePEt), −433.1 (s, NH3).
HFe(Et2PC(H)N(Me)CH2PEt2)(PEtNMePEt) (4). A 15 mL THF
slurry of 265 mg (1.23 mmol) of FeBr2 and 575 mg (2.44 mmol)
of PEtNMePEt was treated with 550 mg (22.6 mmol) of Mg powder.
The blackish-gray slurry was stirred for 20 h. The solvent was removed
under reduced pressure, and the residue was extracted with pentane.
X-ray quality crystals were grown from hexamethyldisiloxane solutions
1
at −30 °C. Yield: 249 mg (39%). H NMR (500 MHz, THF-d8):
AUTHOR INFORMATION
Corresponding Author
*E-mail: Michael.Mock@pnnl.gov.
−13.86 (m, 1H, Fe-H), 0.91−1.94 (40H, PEt), 2.21−2.99 (6H,
Et2PCH2NMeCH2PEt2), 2.25 (s, 3H, N−CH3), 2.52 (s, 3H, N−CH3),
3.43 (br s, 1H, Fe−CH). 31P{1H} NMR (202 MHz, THF-d8): 14.9
(m, 1P, P trans to hydride), 40.6 (dt, 1P, 2JPP = 22, 57 Hz, P adjacent
to CH activated methylene), 53.2 (dd, 1P, 2JPP = 36, 57 Hz, P trans to
■
Notes
2
CH activated methylene), 72.6 (dd, 1P, JPP = 22, 36 Hz, P closest to
The authors declare no competing financial interest.
L
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(15) Mankad, N. P.; Whited, M. T.; Peters, J. C. Angew. Chem., Int.
Ed. 2007, 46, 5768−5771.
ACKNOWLEDGMENTS
■
This work was supported as part of the Center for Molecular
Electrocatalysis, an Energy Frontier Research Center funded by
the U.S. Department of Energy Office of Science, Office of
Basic Energy Sciences. Computational resources were provided
by the National Energy Research Scientific Computing Center
(NERSC) at Lawrence Berkeley National Laboratory. Pacific
Northwest National Laboratory is operated by Battelle for
DOE. We thank Dr. Daniel DuBois for helpful discussions.
(16) (a) Dilworth, M. J.; Fisher, K.; Kim, C.-H.; Newton, W. E.
Biochemistry 1998, 37, 17495−17505. (b) Barton, B. E.; Olsen, M. T.;
Rauchfuss, T. B. J. Am. Chem. Soc. 2008, 130, 16834−16835.
(c) Silakov, A.; Wenk, B.; Reijerse, E.; Lubitz, W. Phys. Chem. Chem.
̈
Phys. 2009, 11, 6592−6599. (d) Erdem, O. F.; Schwartz, L.; Stein, M.;
Silakov, A.; Kaur-Ghumaan, S.; Huang, P.; Ott, S.; Reijerse, E. J.;
Lubitz, W. Angew. Chem., Int. Ed. Engl. 2011, 50, 1439−1443.
(17) Dance, I. Dalton Trans. 2008, 37, 5992−5998.
(18) (a) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.;
Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.;
Meyer, T. J. Chem. Rev. 2012, 112, 4016−4093. (b) Huynh, M. H. V.;
Meyer, T. J. Chem. Rev. 2007, 107, 5004−5064. (c) Warren, J. J.;
Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110, 6961−7001.
(19) (a) DuBois, D. L.; Bullock, R. M. Eur. J. Inorg. Chem. 2011,
2011, 1017−1027. (b) Kilgore, U. J.; Stewart, M. P.; Helm, M. L.;
Dougherty, W. G.; Kassel, W. S.; DuBois, M. R.; DuBois, D. L.;
Bullock, R. M. Inorg. Chem. 2011, 50, 10908−10918. (c) Kilgore, U. J.;
Roberts, J. A. S.; Pool, D. H.; Appel, A. M.; Stewart, M. P.; DuBois, M.
R.; Dougherty, W. G.; Kassel, W. S.; Bullock, R. M.; DuBois, D. L. J.
Am. Chem. Soc. 2011, 133, 5861−5872.
REFERENCES
■
(1) (a) Hinrichsen, S.; Broda, H.; Gradert, C.; Soncksen, L.; Tuczek,
F. Annu. Rep. Prog. Chem., Sect. A 2012, 108, 17−47. (b) Nishibayashi,
Y. Dalton Trans. 2012, 41, 7447−7453. (c) MacKay, B. A.; Fryzuk, M.
D. Chem. Rev. 2004, 104, 385−401. (d) Fryzuk, M. D.; Johnson, S. A.
Coord. Chem. Rev. 2000, 200−202, 379−409.
(2) (a) Hazari, N. Chem. Soc. Rev. 2010, 39, 4044−4056.
(b) Crossland, J. L.; Tyler, D. R. Coord. Chem. Rev. 2010, 254,
1883−1894.
(3) Enthaler, S.; Junge, K.; Beller, M. Angew. Chem., Int. Ed. Engl.
2008, 47, 3317−3321.
(4) (a) Lancaster, K. M.; Roemelt, M.; Ettenhuber, P.; Hu, Y.; Ribbe,
M. W.; Neese, F.; Bergmann, U.; DeBeer, S. Science 2011, 334, 974−
977. (b) Spatzal, T.; Aksoyoglu, M.; Zhang, L.; Andrade, S. L. A.;
Schleicher, E.; Weber, S.; Rees, D. C.; Einsle, O. Science 2011, 334,
940−940. (c) Ramaswamy, S. Science 2011, 334, 914−915. (d) Eady,
R. R. Chem. Rev. 1996, 96, 3013−3030. (e) Burgess, B. K.; Lowe, D. J.
Chem. Rev. 1996, 96, 2983−3011.
(5) (a) Dance, I. Dalton Trans. 2012, 41, 7647−7659. (b) Sellmann,
D.; Sutter, J. Acc. Chem. Res. 1997, 30, 460−469. (c) Sellmann, D.;
Sutter, J. J. Biol. Inorg. Chem. 1996, 1, 587−593.
(6) (a) Dance, I. Chem. Commun. 1997, 165−166. (b) Dance, I.
Dalton Trans. 2008, 5977−5991. (c) Chatt, J.; Dilworth, J. R.;
Richards, R. L. Chem. Rev. 1978, 78, 589−625.
(7) (a) Lukoyanov, D.; Yang, Z.-Y.; Barney, B. M.; Dean, D. R.;
Seefeldt, L. C.; Hoffman, B. M. Proc. Natl. Acad. Sci. U. S. A. 2012, 109,
5583−5587. (b) Lukoyanov, D.; Dikanov, S. A.; Yang, Z.-Y.; Barney,
B. M.; Samoilova, R. I.; Narasimhulu, K. V.; Dean, D. R.; Seefeldt, L.
C.; Hoffman, B. M. J. Am. Chem. Soc. 2011, 133, 11655−11664.
(c) Seefeldt, L. C.; Hoffman, B. M.; Dean, D. R. Annu. Rev. Biochem.
2009, 78, 701−722. (d) Barney, B. M.; Lee, H.-I.; Dos, S. P. C.;
Hoffman, B. M.; Dean, D. R.; Seefeldt, L. C. Dalton Trans. 2006,
2277−2284. (e) Hoffman, B. M.; Lukoyanov, D.; Dean, D. R.;
Seefeldt, L. C. Acc. Chem. Res. 2013, 46, 587−595.
(8) (a) Schrock, R. R. Acc. Chem. Res. 2005, 38, 955−962.
(b) Schrock, R. R. Catalytic reduction of dinitrogen to ammonia by
molybdenum. In Catalysis without Precious Metals; Bullock, R. M., Ed.;
Wiley-Blackwell: Weinheim, Germany, 2010; pp 25−50. (c) Yandulov,
D. V.; Schrock, R. R. Science 2003, 301, 76−78.
(9) (a) Arashiba, K.; Miyake, Y.; Nishibayashi, Y. Nat. Chem. 2011, 3,
120−125. (b) Kinoshita, E.; Arashiba, K.; Kuriyama, S.; Miyake, Y.;
Shimazaki, R.; Nakanishi, H.; Nishibayashi, Y. Organometallics 2012,
31, 8437−8443.
(10) Yuki, M.; Tanaka, H.; Sasaki, K.; Miyake, Y.; Yoshizawa, K.;
Nishibayashi, Y. Nat. Commun. 2012, 3, 1254.
(11) (a) Leigh, G. J.; Jimenez-Tenorio, M. J. Am. Chem. Soc. 1991,
113, 5862−5863. (b) Hills, A.; Hughes, D. L.; Jimenez-Tenorio, M.;
Leigh, G. J.; Rowley, A. T. J. Chem. Soc., Dalton Trans. 1993, 3041−
3049.
(20) (a) Tronic, T. A.; Kaminsky, W.; Coggins, M. K.; Mayer, J. M.
Inorg. Chem. 2012, 51, 10916−10928. (b) Matson, B. D.; Carver, C.
T.; Von Ruden, A.; Yang, J. Y.; Raugei, S.; Mayer, J. M. Chem.
Commun. 2012, 48, 11100−11102. (c) Carver, C. T.; Matson, B. D.;
Mayer, J. M. J. Am. Chem. Soc. 2012, 134, 5444−5447.
(21) (a) Weiss, C. J.; Groves, A. N.; Mock, M. T.; Dougherty, W. G.;
Kassel, W. S.; Helm, M. L.; DuBois, D. L.; Bullock, R. M. Dalton Trans.
2012, 41, 4517−4529. (b) Mock, M. T.; Chen, S.; Rousseau, R.;
O’Hagan, M. J.; Dougherty, W. G.; Kassel, W. S.; DuBois, D. L.;
Bullock, R. M. Chem. Commun. 2011, 47, 12212−12214.
(22) (a) Henry, R. M.; Shoemaker, R. K.; Newell, R. H.; Jacobsen, G.
M.; DuBois, D. L.; DuBois, M. R. Organometallics 2005, 24, 2481−
2491. (b) Henry, R. M.; Shoemaker, R. K.; DuBois, D. L.; DuBois, M.
R. J. Am. Chem. Soc. 2006, 128, 3002−3010. (c) Attempts to
synthesize FeCl2(PEtNMePEt)(dmpe) (dmpe = Me2PCH2CH2PMe2),
resulted in only FeCl2(dmpe)2, FeCl2(PEtNMePEt), and PEtNMePEt; thus
dmpm was chosen as the second bidentate phosphine.
(23) Hills, A.; Hughes, D. L.; Jimenez-Tenorio, M.; Leigh, G. J. J.
Organomet. Chem. 1990, 391, C41−C44.
(24) Crossland, J. L.; Young, D. M.; Zakharov, L. N.; Tyler, D. R.
Dalton Trans. 2009, 9253−9259.
(25) Bau, R.; Teller, R. G.; Kirtley, S. W.; Koetzle, T. F. Acc. Chem.
Res. 1979, 12, 176−183.
(26) Buys, I. E.; Field, L. D.; Hambley, T. W.; McQueen, A. E. D.
Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1993, C49, 1056−
1059.
(27) Wiesler, B. E.; Lehnert, N.; Tuczek, F.; Neuhausen, J.; Tremel,
W. Angew. Chem., Int. Ed. 1998, 37, 815−817.
(28) Franke, O.; Wiesler, B. E.; Lehnert, N.; Nather, C.;
Ksenofontov, V.; Neuhausen, J.; Tuczek, F. Inorg. Chem. 2002, 41,
3491−3499.
(29) Mason, J. Chem. Rev. 1981, 81, 205−227.
(30) Yelle, R. B.; Crossland, J. L.; Szymczak, N. K.; Tyler, D. R. Inorg.
Chem. 2009, 48, 861−871.
(31) Kinney, R. A.; McNaughton, R. L.; Chin, J. M.; Schrock, R. R.;
Hoffman, B. M. Inorg. Chem. 2011, 50, 418−420.
(32) (a) Stanley, J. L.; Heiden, Z. M.; Rauchfuss, T. B.; Wilson, S. R.
Organometallics 2008, 27, 119−125. (b) Carroll, M. E.; Barton, B. E.;
Rauchfuss, T. B.; Carroll, P. J. J. Am. Chem. Soc. 2012, 134, 18843−
18852. (c) Schwartz, L.; Eilers, G.; Eriksson, L.; Gogoll, A.; Lomoth,
R.; Ott, S. Chem. Commun. 2006, 520−522. (d) Ott, S.; Kritikos, M.;
Akermark, B.; Sun, L.; Lomoth, R. Angew. Chem.; Int. Ed. Engl. 2004,
43, 1006−1009. (e) Lawrence, J. D.; Li, H.; Rauchfuss, T. B.; Benard,
M.; Rohmer, M.-M. Angew. Chem., Int. Ed. Engl. 2001, 40, 1768−1771.
(33) Gao, Y.; Holah, D. G.; Hughes, A. N.; Spivak, G. J.; Havighurst,
M. D.; Magnuson, V. R.; Polyakov, V. Polyhedron 1997, 16, 2797−
2807.
(12) Rodriguez, M. M.; Bill, E.; Brennessel, W. W.; Holland, P. L.
Science 2011, 334, 780−783.
(13) Gilbertson, J. D.; Szymczak, N. K.; Tyler, D. R. J. Am. Chem. Soc.
2005, 127, 10184−10185.
(14) (a) Smith, J. M.; Subedi, D. Dalton Trans. 2012, 41, 1423−1429.
(b) Scepaniak, J. J.; Young, J. A.; Bontchev, R. P.; Smith, J. M. Angew.
Chem., Int. Ed. Engl. 2009, 48, 3158−3160.
M
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(34) Fox, D. J.; Bergman, R. G. Organometallics 2004, 23, 1656−
1670.
(35) Field, L. D.; Li, H. L.; Dalgarno, S. J.; Jensen, P.; McIntosh, R.
D. Inorg. Chem. 2011, 50, 5468−5476.
(36) Crossland, J. L.; Balesdent, C. G.; Tyler, D. R. Inorg. Chem.
2012, 51, 439−445.
(37) (a) Sellmann, D.; Soglowek, W.; Knoch, F.; Ritter, G.; Dengler,
́
J. Inorg. Chem. 1992, 31, 3711−3717. (b) Chetcuti, P. A.; Liegard, A.;
Rihs, G.; Rist, G.; Schweiger, A. Helv. Chim. Acta 1991, 74, 1591−
1599. (c) Bowman, A. C.; Bart, S. C.; Heinemann, F. W.; Meyer, K.;
Chirik, P. J. Inorg. Chem. 2009, 48, 5587−5589. (d) Vela, J.; Stoian, S.;
Flaschenriem, C. J.; Munck, E.; Holland, P. L. J. Am. Chem. Soc. 2004,
̈
126, 4522−4523. (e) Lee, Y.; Mankad, N. P.; Peters, J. C. Nat. Chem.
2010, 2, 558−565.
(38) (a) Henderson, R. A. J. Chem. Soc., Dalton Trans. 1988, 515.
(b) Franke, O.; Wiesler, B. E.; Lehnert, N.; Peters, G.; Burger, P.;
Tuczek, F. Z. Anorg. Allg. Chem. 2006, 632, 1247−1256. (c) Tuczek,
F.; Lehnert, N. Angew. Chem., Int. Ed. Engl. 1998, 37, 2636−2638.
(39) Karsch, H. H.; Neugebauer, D. Angew. Chem., Int. Ed. 1982, 21,
312−313.
(40) Hirano, M.; Akita, M.; Morikita, T.; Kubo, H.; Fukuoka, A.;
Komiya, S. J. Chem. Soc., Dalton Trans. 1997, 3453−3458.
(41) (a) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988,
88, 899−926. (b) Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980,
102, 7211−7218.
(42) Curtis, C. J.; Miedaner, A.; Ciancanelli, R.; Ellis, W. W.; Noll, B.
C.; Rakowski DuBois, M.; DuBois, D. L. Inorg. Chem. 2003, 42, 216−
227.
(43) LeSuer, R. J.; Geiger, W. E. Angew. Chem., Int. Ed. Engl. 2000,
39, 248−250.
(44) Rach, S. F.; Herdtweck, E.; Kuhn, F. E. J. Organomet. Chem.
̈
2011, 696, 1817−1823.
(45) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(b) Perdew, J. P. Phys. Rev. B 1986, 33, 8822−8824.
(46) 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, N. J.; 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.; 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 A.1; Gaussian, Inc.: Wallingford, CT, 2009.
(47) Andrae, D.; Haußermann, U.; Dolg, M.; Stoll, H.; Preuß, H.
̈
Theor. Chem. Acc. 1990, 77, 123−141.
(48) (a) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.;
Gordon, M. S.; DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77,
3654−3665. (b) Rassolov, V. A.; Ratner, M. A.; Pople, J. A.; Redfern,
P. C.; Curtiss, L. A. J. Comput. Chem. 2001, 22, 976−984.
(49) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995−2001.
(50) (a) Abdur-Rashid, K.; Fong, T. P.; Greaves, B.; Gusev, D. G.;
Hinman, J. G.; Landau, S. E.; Lough, A. J.; Morris, R. H. J. Am. Chem.
Soc. 2000, 122, 9155−9171. (b) Rodima, T.; Kaljurand, I.; Pihl, A.;
Maemets, V.; Leito, I.; Koppel, I. J. Org. Chem. 2002, 67, 1873−1881.
(51) APEX II, v. 2009.3; Bruker AXS: Madison, WI, 2009.
(52) SAINT+, v. 7.56A; Bruker AXS: Madison, WI, 2009.
(53) SADABS, v. 2008/1; Bruker AXS Inc.: Madison, WI, 2008.
(54) Sheldrick, G. M. SHELXTL, v. 2008; Bruker AXS Inc.: Madison,
WI, 2008.
N
dx.doi.org/10.1021/ic4000704 | Inorg. Chem. XXXX, XXX, XXX−XXX