Light Alters the NH
3
vs N
2
H
4
Product Profile in Iron-catalyzed
Nitrogen Reduction via Dual Reactivity from an Iron Hydrazido
(Fe=NNH
2
) Intermediate
Pablo Garrido-Barros
+,a
,
Matthew J. Chalkley
+,a
,
Jonas C. Peters
a
[a]
Division of Chemistry and Chemical Engineering, California Institute of Technology (Caltech),
1200 E California Blvd, Pasadena, CA 91125
Abstract
Whereas synthetically catalyzed nitrogen reduction (N
2
R) to produce ammonia is widely studied,
catalysis to instead produce hydrazine (N
2
H
4
) has received less attention despite its considerable
mechanistic interest. Herein, we disclose that irradiation of a trisphosphine-borane (P
3
B
) Fe
catalyst, P
3
B
Fe
+
, significantly alters its product profile to increase N
2
H
4
versus NH
3
; P
3
B
Fe is
otherwise known to be highly selective for NH
3
. We posit a key terminal hydrazido intermediate,
P
3
B
Fe=NNH
2
, as selectivity-determining. Whereas its singlet ground state undergoes protonation
to liberate NH
3
, a low-lying triplet excited state leads to reactivity at N
α
and formation of N
2
H
4
.
Associated electrochemical and spectroscopic studies establish that N
2
H
4
lies along a unique
product pathway; NH
3
is not produced from N
2
H
4
. Our findings are distinct from the canonical
mechanism for hydrazine formation, which proceeds via a diazene (HN=NH) intermediate and
showcase light as a tool to tailor selectivity.
Entry for the Table of Contents
jpeters@caltech.edu .
[+]
These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of the document
HHS Public Access
Author manuscript
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Published in final edited form as:
Angew Chem Int Ed Engl
. 2023 February 20; 62(9): e202216693. doi:10.1002/anie.202216693.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Light irradiation of a tris(phosphine)borate iron (P
3
B
Fe) catalyst system during nitrogen reduction
(N
2
R) catalysis alters its product selectivity profile, producing significant N
2
H
4
alongside NH
3
.
Only trace N
2
H
4
is otherwise produced. This study provides insight into the dominant factors
determining the selectivity of this catalysis and implicates dual reactivity from a singlet ground
versus a triplet excited state hydrazido (Fe=NNH
2
) intermediate.
Social Media:
Pablo Garrido-Barros (@PGarridoBarros)
Matthew J. Chalkley (@chalkley_talk)
Caltech Chemistry and Chemical Engineering (@CaltechCCE)
Keywords
nitrogen fixation; light-controlled selectivity; hydrazine synthesis; Fe-mediated N
2
R; hydrazido
intermediate
Elucidating factors that predict product selectivity underpins mechanistic research in
catalysis. Biocatalytic transformations of O
2
and N
2
are illustrative. In cytochrome P450
enzymes,
[
1
−
7
]
the standard pathway of O
2
reduction leads to distal (O
β
) protonation of an
Fe
III
–OOH hydroperoxo intermediate, with concomitant O–O bond scission to yield water
(H
2
O) and (Por
•+
)Fe
IV
=O (Cpd1; Por = Porphyrin); the latter further promotes productive
Garrido-Barros et al.
Page 2
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
C–H functionalization or undergoes reduction to liberate a second H
2
O molecule in a net 4
H
+
/ 4 e
−
process. Alternatively, an uncoupling pathway leads to proximal (O
α
) protonation
and the liberation of (Por
•+
)Fe
III
and hydrogen peroxide (H
2
O
2
) in an overall 2 H
+
/ 2 e
−
process (Figure 1a).
Mechanistically less well understood, biocatalytic N
2
reduction (N
2
R) at nitrogenase
cofactors generates ammonia (NH
3
) via N–N scission by an as yet unidentified pathway(s)
in a net 6 H
+
/ 6 e
−
process (excluding the 2 H
+
/ 2 e
−
for presumed obligatory H
2
).
[
8
]
Limiting mechanistic scenarios are either a distal (early N–N cleavage and implicating
formal N
3-
or NH
2-
intermediates) or an alternating mechanism (implicating a diimide
(HN=NH) intermediate).
[
9
]
Depending on the type of N
2
ase (MoFe, VFe, or Fe-only) and
the specific conditions studied, hydrazine (N
2
H
4
) can also be liberated as a minor product
in a net 4 H
+
/ 4 e
−
reductive process.
[
10
−
12
]
Diimide is also a substrate for nitrogenase
and is converted to NH
3.
[
13
−
16
]
These observations have contributed to an emphasis on an
alternating mechanism during biocatalytic N
2
R, where H-atoms are alternatively delivered to
the distal and proximal N-atom of a bound N
2
intermediate (Figure 1b).
Iron model compounds have shown parallels to this dual selectivity for nitrogen fixation
(Figure 1c).
[
17
,
18
]
For instance, our lab demonstrated the generation of N
2
H
4
from a
P
3
Si
Fe(N
2
) complex (P
3
Si
is a tris(phosphino)silyl ligand, Figure 1c) via addition of H
+
/
e
−
equivalents (CrCl
2
/HBF
4
) in ~47% yield,
[
19
]
whereas the same system is a (modest)
catalyst for NH
3
formation when a different acid and reductant are used.
[
20
]
Using a related
iron compound, P
3
B
Fe
+
, (P
3
B
= tris(o-diisopropylphosphinophenyl)borane), Fe-mediated
N
2
-to-NH
3
catalysis occurs, but only trace N
2
H
4
is also observed.
[
21
,
22
]
Nishibayashi
and coworkers have reported on iron catalysts such as (PNP)Fe(N
2
) (Figure 1c) and
(PCP)Fe(N
2
) that generate varying amounts of NH
3
and N
2
H
4
depending on the reaction
conditions.
[
23
−
25
]
Perhaps most strikingly, Ashley and coworkers described (depe)
2
Fe(N
2
) as
a catalyst that is highly selective for N
2
-to-N
2
H
4
conversion.
[
26
]
Of ongoing interest is to understand how such iron systems mechanistically discriminate
between NH
3
and N
2
H
4
. Does NH
3
generation proceed along the same pathway as N
2
H
4
formation? Or are NH
3
and N
2
H
4
generated via a branching point from a common
intermediate, for example via Fe=NNH
2
, akin to the Fe
III
-OOH intermediate in cytochrome
P450 (Figure 1a)?
Related to these issues, we showed several years ago that low temperature reduction of
an iron hydrazido intermediate, P
3
Si
Fe(NNH
2
)
+
, in the absence of added acid, results
in the formation of a 1:1 ratio of P
3
Si
Fe(N
2
H
4
)
+
and P
3
Si
Fe(N
2
) (Eq 1), implicating a
distal-to-alternating hybrid pathway to hydrazine, which offers one plausible mechanism
for the general scheme shown in Figure 1b.
[
27
]
Herein we show, using the structurally
related P
3
B
Fe
+
system, that irradiation with blue light (440 nm emission maximum) alters
the selectivity profile of catalytic N
2
R, leading to competitive and even dominant N
2
H
4
formation alongside NH
3
production. Associated data allow us to posit Fe=NNH
2
as the
discriminatory intermediate between these two products, and that the N
2
H
4
produced lies
along a pathway (distal-to-alternating) that is distinct from the NH
3
production pathway
(distal).
Garrido-Barros et al.
Page 3
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
[Eq.(1)]
The lead observation is as follows. Blue LED irradiation of a solution of [P
3
B
Fe][BAr
F
4
]
(BAr
F
4
= tetrakis-(3,5-bis(trifluoromethyl)phenylborate)) in the presence of excess Cp*
2
Co
(54 equiv) and [Ph
2
NH
2
][OTf] (108 equiv) leads to catalytic yields of a mixture of NH
3
(7.1
± 0.6 equiv) and N
2
H
4
(3.1 ± 1.0 equiv) (Table 1, entry 1). Likewise, Hg lamp irradiation
yields similar results (NH
3
: 8.4 ± 0.3, N
2
H
4
: 3.0 ± 0.1; entry 2). In contrast, very little
N
2
H
4
is produced (0.4 ± 0.2 equiv) in the absence of irradiation (entry 3). The total yield of
fixed-N is similar in each case, and the dark reaction yield compares well with a previous
study.
[
28
]
Irradiation in the absence of P
3
B
Fe
+
does not afford any fixed product (< 0.1
equiv). Freeze-quench Mössbauer spectroscopy performed after 30 minutes of blue LED
irradiation with
57
Fe-labeled P
3
B
Fe
+
revealed a P
3
B
Fe-speciation closely resembling that
which is obtained in the absence of irradiation (Figure S12).
[
22
]
We have previously experimentally and computationally interrogated the more stable
and hence isolable, methylated hydrazido, P
3
B
Fe(NNMe
2
), and shown that it is a good
analog for the catalytically relevant parent hydrazido, P
3
B
Fe(NNH
2
).
[
29
]
These hydrazido
complexes feature singlet ground states but low-lying
S
= 1 excited states with increased
spin at N
α
resulting from increased occupation of an Fe-N
π
anti-bonding orbital (d
xz
and
N-N
π
*) and worsened overlap in the Fe-N
π
-bonding orbitals (Figure 2).
[
29
,
30
]
Moreover,
we have performed time-dependent DFT (TD-DFT) calculations (see SI) that are consistent
with the notion that visible light excitation of
S
= 0 P
3
B
Fe(NNH
2
) similarly increases the
population of the Fe-N
π
anti-bonding orbital (d
xz
and N-N
π
*; see SI). This contrasts
directly with the ground state electronic structure, in which the HOMO is Fe-localized
(d
x
2-
y
2
), in an orbital orthogonal to the Fe-N vector. The structural consequence of the
increased spin on N
α
in the low-lying excited states (both
S
= 0 and
S
= 1) is partial bending
of the hydrazido ligand, potentially priming it for protonation (or perhaps HAT reactivity)
at N
α
,
[
31
]
which in turn could lead to N
2
H
4
formation. On the other hand, it is known that
protonation at N
β
, presumed to proceed from the ground state, leads to NH
3
release and
generation of P
3
B
Fe(N)
+
(Figure 1a).
[
29
]
Given our hypothesis that a protonation-step is selectivity determining, we evaluated the
consequence of using a weaker acid ([
2,6-Me
PhH
3
][OTf], p
K
a of 6.8 in THF, versus 3.2 for
[Ph
2
NH
2
][OTf]),
[
32
]
and we found it increases the relative yield of N
2
H
4
vs NH
3
under
irradiation, albeit with diminished yield of fixed-N overall (Table 1, entry 4). By contrast,
a stronger acid ([
2,4,6-Cl
PhNH
3
][OTf], pK
a
of 2.1 in THF) leads to a similar ratio of N
2
H
4
versus NH
3
(entry 5), as did [Ph
2
NH
2
][OTf].
[
33
]
We have also studied previous conditions using KC
8
and HBAr
F
4
as the reductant and
acid source, respectively, but now under irradiation.
[
17
,
19
,
21
]
For internal consistency we
Garrido-Barros et al.
Page 4
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
(re)confirmed that NH
3
is primarily produced using KC
8
and HBAr
F
4
(6.1 equiv NH
3
versus
0.8 equiv of N
2
H
4
, Table 1, entry 6).
[
21
,
30
,
34
]
In contrast, with Hg-lamp irradiation a greater
than 1:1 N
2
H
4
/NH
3
ratio is produced (4.7 equiv N
2
H
4
versus 4.2 equiv of NH
3
; Table 1,
entry 7). Selectivity for hydrazine is even more favored with blue LED irradiation using
this same cocktail (5.1 equiv of N
2
H
4
versus 2.5 equiv of NH
3
). The difference in relative
selectivity likely arises due to a previously observed enhancement in N
2
R efficiency via
Hg-lamp irradiation of KC
8
/HBAr
F
4
-driven catalysis, resulting from light-induced reductive
elimination of H
2
from an off-path hydride-borohydride resting state of the iron catalyst (to
regenerate on-path P
3
B
Fe(N
2
)).
[
20
,
35
]
This idea is consistent with our observation of highly
similar NH
3
/N
2
H
4
selectivity in Cp*
2
Co/[Ph
2
NH
2
][OTf]-driven N
2
R catalysis, irrespective
of blue LED or Hg-lamp irradiation; we have shown previously that the hydride-borohydride
resting state does not form when using Cp*
2
Co/[Ph
2
NH
2
][OTf].
[
22
]
The first excited state of P
3
B
Fe(NNH
2
) is estimated (via DFT) to be sufficiently low-lying
at ~2 kcal·mol
−1
to be thermally accessible, calibrated against an experimentally measured
value of 3.7 ± 0.1 kcal·mol
−1
for P
3
B
Fe(NNMe
2
).
[
29
]
Temperature might therefore also alter
product selectivity. Accordingly, a reaction performed at −45 °C in the dark also produced
more hydrazine (1.0 ± 0.3 equiv) compared with −78 °C (0.4 ± 0.1 equiv), providing a
N
2
H
4
/NH
3
ratio of 0.12 versus 0.03 respectively. The total fixed-N yield drops from 67.4
± 0.3 % to 56.4 ± 0.9 % between −78 °C and −45 °C. Running the catalysis instead
at −10 °C gives yet higher selectivity for N
2
H
4
(the N
2
H
4
/NH
3
ratio is now 0.17), but
with a substantially lower total yield of fixed-N products (due to competing H
2
evolution).
Curiously, this behavior parallels observations that have been made for VFe-nitrogenase,
which shows increasing (albeit still low) N
2
H
4
yields at higher temperatures.
[
10
,
11
]
The generation of N
2
H
4
, presumably via thermal population of the low-lying excited
state of P
3
B
Fe(NNH
2
), is also evidenced under electrochemical conditions via the use
of rotating ring disk electrode (RRDE) techniques (Figure 3).
[
36
]
In this experiment, a
linear sweep voltammogram of a 1 mM solution of P
3
B
Fe
+
in the presence of slight
excess of [Ph
2
NH
2
][OTf] (10 equiv) was performed at a slow scan rate (20 mV·s
−1
) on
a glassy carbon disk electrode at room temperature. Under these conditions, P
3
B
Fe
+
shows
a single electrocatalytic wave with an onset potential of −2.0 V vs Fc
+/0
, corresponding
to the reduction of in situ formed P
3
B
Fe(OTf). Such a reduction generates P
3
B
Fe and
then P
3
B
Fe(N
2
)
−
; the latter enters the N
2
R catalytic cycle via protonation. Simultaneously,
the ring Pt electrode was held at a constant potential of –0.5 V vs Fc
+/0
, where N
2
H
4
(but not NH
3
) can be oxidized (Figure S14). Therefore, the generation of N
2
H
4
in this
electrochemical experiment is evidenced by an increase in the oxidative current at the Pt
electrode at potentials below −2.0 V vs Fc
+/0
, corresponding with N
2
R activity by P
3
B
Fe
+
.
Whereas NH
3
can be generated via protonation of P
3
B
Fe(NNH
2
), in principle it can
be alternatively generated via a hydrazine-bound intermediate. Indeed, a number of
catalysts are known to mediate the conversion of N
2
H
4
to NH
3
.
[37−40]
To determine
whether N
2
H
4
can lead to NH
3
production in this system we next evaluated the
behavior of P
3
B
Fe(N
2
H
4
)
+/0
, under catalytically relevant conditions, via spectroscopic and
electrochemical techniques.
Garrido-Barros et al.
Page 5
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
The hydrazine complex, P
3
B
Fe(N
2
H
4
)
+
,
[
41
]
can be generated in situ by addition of excess
N
2
H
4
(10 equiv) to P
3
B
Fe
+
at −78 °C (Figure 4b). Subsequent addition of either [Ph
2
NH
2
]
[OTf] or [Ph
2
NH
2
][BAr
F
4
] results in the release of free N
2
H
4
(via its protonated form,
N
2
H
5
+
) and either P
3
B
Fe(OTf) or P
3
B
Fe
+
, respectively (Figure 4a and S18). HBAr
F
4
reacts
similarly. Thus, hydrazine is readily released from P
3
B
Fe
+
under catalytic conditions where
excess acid is present.
Addition of one equivalent of Cp*
2
Co to in situ formed P
3
B
Fe(N
2
H
4
)
+
under an N
2
atmosphere produces P
3
B
Fe(N
2
) (determined via UV-vis, Figure 4b). This behavior mirrors
prior results we have disclosed for P
3
Si
Fe(N
2
H
4
)
+
upon reduction under an N
2
atmosphere.
[
42
]
Reduction of P
3
B
Fe(N
2
H
4
)
+
generates a formal Fe(0) state primed for substitution of
N
2
H
4
by N
2
owing to a more favorable
π
-donor/acceptor interaction. Under argon, reduction
of P
3
B
Fe(N
2
H
4
)
+
leads instead to the same featureless UV-vis spectrum that is obtained via
the reduction of P
3
B
Fe
+
under argon (see SI). Further addition of N
2
H
4
does not alter this
spectrum; N
2
H
4
has a very low affinity for zerovalent (ignoring sigma backbond) P
3
B
Fe.
We next discuss the redox chemistry of P
3
B
Fe
+
under N
2
and N
2
H
4
(Figure 5a). CVs of
a 1 mM P
3
B
Fe
+
solution in diethyl ether under N
2
display two redox processes (−1.4 V
and −2.1 V vs Fc
+/0
), corresponding to reduction of P
3
B
Fe
+
to P
3
B
Fe, the latter of which
coordinates N
2
, and then further reduction to P
3
B
Fe(N
2
)
−
(Figure 5b).
[
19
,
43
]
For CVs taken
in the presence of 1 equiv of N
2
H
4
, the redox wave at ~ −1.4 V splits into two peaks when
N
2
H
4
is present, indicative of equilibrium binding of N
2
H
4
at the Fe(I) state; the intensity
of the P
3
B
Fe
+/0
redox couple correspondingly decreases and a new feature at more cathodic
potentials, attributed to the reduction of P
3
B
Fe(N
2
H
4
)
+
, appears. The close match of the
potential of the returning anodic wave to that of the P
3
B
Fe(N
2
)/P
3
B
Fe
+
couple corroborates
the low-binding affinity of N
2
H
4
to the formally Fe(0) state discussed above.
Addition of excess N
2
H
4
to the solution leads to P
3
B
Fe(N
2
H
4
)
+
becoming the predominant
Fe(I) species. CV’s now reveal behavior typical of an EC (electron transfer-chemical
step) process in a pure kinetic regime (Fig. 5a and SI).
[
44
]
One electron reduction of
P
3
B
Fe(N
2
H
4
)
+
is followed by a chemical step (N
2
H
4
substitution by N
2
) leading to
P
3
B
Fe(N
2
). Therefore, the potential of the reduction process is determined by the
K
eq
of
the chemical step. The CE nature of the anodic feature is evidenced by a broader oxidation
wave, corresponding to oxidation of P
3
B
Fe(N
2
H
4
) following ligand exchange of N
2
by
N
2
H
4
. Because of the small equilibrium for binding N
2
H
4
in the Fe(0) state and the fast
kinetics of N
2
H
4
release full reversibility is never reached;
[
44
]
nonetheless, analysis of these
CV’s using the corresponding Nernst equation (Eq. 2) allows us to estimate a
K
eq
for the
exchange reaction between N
2
, N
2
H
4
, P
3
B
Fe(N
2
) and P
3
B
Fe(N
2
H
4
), which is on the order
of 10
−3
-10
−4
, disfavoring hydrazine coordination to P
3
B
Fe (see SI for details). This estimate
is consistent with analysis of CV’s that only go through the first redox event (see SI, Figure
S26).
A final observation is that catalytic runs using P
3
B
Fe
+
with added N
2
H
4
as a possible
substrate in the presence of chemical reductant and acid sources (see SI), but in the absence
of N
2
, do not furnish any NH
3
.
Garrido-Barros et al.
Page 6
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
E
red
= Eº
1/2
– RT/F · ln
K
eq
N
2
/
N
2
H
4
[Eq. (2)]
The results and discussion provided above, combined with observations from prior studies,
lead us to formulate the scheme outlined in Figure 6 to account for the catalytic production
of either NH
3
or N
2
H
4
using this P
3
B
Fe-system. In accord with prior studies, a series
of steps comprising the net addition of N
2
and 2 H
+
/ 3 e
−
to P
3
B
Fe
+
generates the
key hydrazido intermediate P
3
B
Fe(NNH
2
). Protonation of the singlet ground state of this
intermediate releases NH
3
and generates P
3
B
Fe(N)
+
.
[
30
]
Further reductive protonation of
the nitride leads to formation of amide P
3
B
Fe(NH
2
) and then P
3
B
Fe(NH
3
)
+
, the latter of
which is reduced to liberate NH
3
with rebinding of N
2
closing the catalytic cycle.
[
18
]
An
additional minor pathway leads to N
2
H
4
, which we posit follows from protonation (or
HAT) at N
α
of P
3
B
Fe(NNH
2
) (presumably via an as yet uncharacterized P
3
B
Fe(NHNH
2
)
+/0
intermediate), likely due to thermal occupation of a low-lying
S
= 1 P
3
B
Fe(NNH
2
) excited
state. Subsequent reductive protonation (2 H
+
/ 1 e
−
from P
3
B
Fe(NNH
2
) overall) leads to
P
3
B
Fe(N
2
H
4
)
+
, which is labile and can release N
2
H
4
to reform P
3
B
Fe
+
to reinitiate the cycle.
Under irradiation, this minor pathway is enhanced due to an increase in the excited state
population of the P
3
B
Fe(NNH
2
) intermediate.
For well characterized synthetic catalyst systems to date, terminal M(N
2
) adducts are twice
functionalized at N
β
to generate a hydrazido (M(NNH
2
)) intermediate.
[
18
]
The electronic
structure of this intermediate can determine its selectivity for the NH
3
versus N
2
H
4
pathway.
A linear hydrazido(2−) ligand that features strong metal-to-ligand multiple bonding and lone
pair character at N
β
leads to ammonia via protonation at N
β
(distal path). On the other hand,
formation of a “bent” hydrazido intermediate, with attenuated metal-to-ligand pi-bonding
and increased nucleophilic (or radical) character at N
α
can instead lead to formation of
N
2
H
4
via a distal-to-alternating hybrid mechanism.
N
2
R catalysts featuring 4d (or 5d) metals (i.e., Mo) tend to favor linear hydrazido complexes
owing to strong covalency; thus far they are only known to produce NH
3
. By contrast, N
2
R
catalysts featuring 3d metals (i.e., Fe) are more likely to access high- or intermediate-spin
electronic structures that weaken metal-to-ligand covalency; they produce (at least some)
N
2
H
4
in all cases reported to date. The observations discussed herein, where irradiation
favors a hydrazido excited state that is primed for an N
2
H
4
pathway, alongside the other data
discussed, provides evidence for this selectivity model.
Supplementary Material
Refer to Web version on PubMed Central for supplementary material.
Acknowledgements
We thank the National Institutes of Health (R01 GM-075757) for support of this work; the Dow Next Generation
Educator Funds and Instrumentation Grants for their support of the NMR facility at Caltech; the Resnick Water and
Environment Laboratory at Caltech for the use of instrumentation. P.G.B. thanks the Ramón Areces Foundation for
a postdoctoral fellowship. M.J.C. thanks the Resnick Sustainability Institute for a graduate fellowship.
Garrido-Barros et al.
Page 7
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
References
[1]. Albertolle ME, Guengerich FP, Inorg J. Biochem 2018, 186, 228–234.
[2]. Guengerich FP, ACS Catal. 2018, 8, 10964–10976. [PubMed: 31105987]
[3]. Krest CM, Onderko EL, Yosca TH, Calixto JC, Karp RF, Livada J, Rittle J, Green MT, J. Biol.
Chem, 2013, 288, 17074–17081. [PubMed: 23632017]
[4]. Meunier B, de Visser SP, Shaik S, Chem. Rev 2004, 104, 3947–3980. [PubMed: 15352783]
[5]. Poulos TL, Chem. Rev 2014, 114, 3919–3962. [PubMed: 24400737]
[6]. Makris TM, Denisov I, Schlichting I, Sligar SG, In Cytochrome P450: Structure, Mechanism, and
Biochemistry; Montellano de Ortiz, R. P, Ed.; Springer US: Boston, MA, 2005; 149−182.
[7]. Brezny AC, Johnson SI, Raugei S, Mayer JM, J. Am. Chem. Soc. 2020, 142, 4108–4113.
[PubMed: 32064870]
[8]. Einsle O, Rees DC, Chem. Rev 2020, 120, 4969–5004. [PubMed: 32538623]
[9]. Hoffman BM, Lukoyanov D, Dean DR, Seefeldt LC, Acc. Chem. Res. 2013, 46, 587–595.
[PubMed: 23289741]
[10]. Barney BM, Laryukhin M, Igarashi RY, Lee H-I, Dos Santos PC, Yang T-C, Hoffman BM, Dean
DR, Seefeldt LC, Biochemistry 2005, 44, 8030–8037. [PubMed: 15924422]
[11]. Dilworth MJ, Eady RR, Biochem. J. 1991, 277, 465–468. [PubMed: 1859374]
[12]. Dilworth MJ, Eldridge ME, Eady RR, Biochem. J. 1993, 289, 395–400. [PubMed: 8424785]
[13]. McKenna CE, Simeonov AM, Eran H, Bravo-Leerabhandh M, Biochemistry 1996, 35, 4502–
4514. [PubMed: 8605200]
[14]. Malinak SM, Simeonov AM, Mosier PE, McKenna CE, Coucouvanis D, J. Am. Chem. Soc 1997,
119, 1662–1667.
[15]. Barney BM, Lukoyanov D, C Yang T, Dean DR, Hoffman BM, Seefeldt LC, Proc. Natl. Acad.
Sci 2006, 103, 17113–17118. [PubMed: 17088552]
[16]. Barney BM, McClead J, Lukoyanov D, Laryukhin M, Yang T-C, Dean DR, Hoffman BM,
Seefeldt LC, Biochemistry 2007, 46, 6784–6794. [PubMed: 17508723]
[17]. Hazari N. Chem. Soc. Rev 2010, 39, 4044–4056. [PubMed: 20571678]
[18]. Chalkley MJ, Drover MW, Peters JC. Chem. Rev 2020, 120 (12), 5582–5636. [PubMed:
32352271]
[19]. Mankad NP, Whited MT, Peters JC, Angew. Chem., Int. Ed 2007, 46, 5768–5771.
[20]. Del Castillo TJ, Thompson NB, Peters JC, J. Am. Chem. Soc 2016, 138, 5341–5350. [PubMed:
27026402]
[21]. Anderson JS, Rittle J, Peters JC, Nature 2013, 501, 84–87. [PubMed: 24005414]
[22]. Chalkley MJ, Del Castillo TJ, Matson BD, Roddy JP, Peters JC, ACS Cent. Sci 2017, 3, 217–223.
[PubMed: 28386599]
[23]. Kuriyama S, Arashiba K, Nakajima K, Matsuo Y, Tanaka H, Ishii K, Yoshizawa K, Nishibayashi
Y, Nat. Commun 2016, 7, 12181. [PubMed: 27435503]
[24]. Sekiguchi Y, Kuriyama S, Eizawa A, Arashiba K, Nakajima K, Nishibayashi Y, Chem. Commun
2017, 53, 12040–12043.
[25]. Kuriyama S, Kato T, Tanaka H, Konomi A, Yoshizawa K, Nishibayashi Y, Bull. Chem. Soc. Jpn
2022, 95, 683–692.
[26]. Hill PJ, Doyle LR, Crawford AD, Myers WK, Ashley AE, J. Am. Chem. Soc 2016, 138 (41),
13521–13524. [PubMed: 27700079]
[27]. Rittle J, Peters JC, J. Am. Chem. Soc 2016, 138, 4243–4248. [PubMed: 26937584]
[28]. We have adapted a slightly different work-up procedure (see SI) herein that better quantifies
N2H4 but slightly underestimates the total fixed-N yield (see SI).
[29]. Thompson NB, Oyala PH, Dong HT, Chalkley MJ, Zhao J, Alp EE, Hu M, Lehnert N, Peters JC,
Inorg. Chem 2019, 58, 3535–3549. [PubMed: 30762355]
[30]. Thompson NB, Green MT, Peters JC, J. Am. Chem. Soc 2017, 139, 15312–15315. [PubMed:
28992418]
Garrido-Barros et al.
Page 8
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
[31]. Previously, the reactivity priming effect of excited state-induced bending has been discussed
in the context of nitrogen fixation for bridged N2 complexes: Reiher M, Kirchner B, Hutter J,
Sellmann D, Hess BA, Chem. Eur. J 2004, 10, 4443–4453. [PubMed: 15378622]
[32]. In our previous study, [2,6-MePhH3][OTf] was the acid with the highest pKa for which the yield
of fixed-N product still predominates vs H2 evolution and does not interfere with the N2H4
quantification method. See reference 43 and SI.
[33]. Study of further anilinium acids are unfortunately constrained by the colorimetric methods used
to quantitate NH3 and N2H4, which are only compatible with at least 2,6-disubstitution of the
phenyl-ring.
[34]. Anderson JS, Cutsail GE, Rittle J, Connor BA, Gunderson WA, Zhang L, Hoffman BM, Peters
JC, J. Am. Chem. Soc 2015, 137, 7803–7809. [PubMed: 26000443]
[35]. Buscagan TM, Oyala PH, Peters JC. Angew. Chem. Int. Ed, 2017, 56, 6921–6926.
[36]. Saha P, Amanullah S, Dei A. J. Am. Chem. Soc 2020, 142, 17312–17317. [PubMed: 33006899]
[37]. Gu NX, Ung G, Peters JC. Chem. Commun 2019, 55, 5363–5366.
[38]. Su T, Xu S, Yang D, Su L, Wang B, J. Qu. Eur. J. Inorg. Chem 2020, 4263–4269.
[39]. Wu B, Gramigna KM, Bezpalko MW, Foxman BM,Thomas CM, Inorg. Chem 2015, 54, 10909–
10917. [PubMed: 26492046]
[40]. Umehara K, Kuwata S, Ikariya T, J. Am. Chem. Soc 2013, 135, 6754–6757. [PubMed:
23611139]
[41]. Anderson JS, Moret M-E, Peters JC, J. Am. Chem. Soc 2013, 135, 534–537. [PubMed:
23259776]
[42]. Lee Y, Mankad NP, Peters JC, Nat. Chem 2010, 2, 558–565. [PubMed: 20571574]
[43]. Chalkley MJ, Del Castillo TJ, Matson BD, Peters JC, J. Am. Chem. Soc 2018, 140, 6122–6129.
[PubMed: 29669205]
[44]. Costentin C, Savéant J-M, Elements of Molecular and Biomolecular Electrochemistry: An
Electrochemical Approach to Electron Transfer Chemistry (2nd ed.). (John Wiley & Sons,
Hoboken, 2006).
Garrido-Barros et al.
Page 9
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 1.
(a) Conceptual comparison of selectivity-determining iron intermediates of O
2
reduction
in Cytochrome P450 and N
2
reduction in a synthetic iron catalyst system. (b) Primary
mechanisms considered for the catalytic reduction of N
2
to N
2
H
4
. (c) Representative
synthetic iron systems that generate varying amounts of hydrazine from N
2
.
Garrido-Barros et al.
Page 10
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 2.
Representation of the calculated HOMO and SOMO for the
S
= 0 and S = 1 state of
P
3
B
Fe(NNH
2
), respectively; isosurface value of 0.1 au.
Garrido-Barros et al.
Page 11
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 3.
Rotating ring disk electrode experiment for hydrazine detection in a 0.1 M [Na][BAr
F
4
]
solution in Et
2
O at RT with (dotted trace) and without (solid black trace) P
3
B
Fe
+
(1 mM).
The top graph shows the current collected at the glassy carbon disk electrode during a
linear sweep voltammetry at 20 mV·s
-1
. The bottom graph shows the current simultaneously
collected in the Pt ring electrode upon applying a constant potential of –0.5 V vs Fc
+/0
(for
hydrazine oxidation). A Pt-wire counter electrode, a 5 mM Ag/AgOTf reference electrode
and a rotation speed of 1600 rpm were employed.
Garrido-Barros et al.
Page 12
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 4.
(a) Reactivity at –78 °C of P
3
B
Fe(N
2
H
4
)
+
with reductant and acids relevant to catalysis. (b)
UV-vis spectra of P
3
B
Fe
+
(1 mM, black trace), P
3
B
Fe(N
2
H
4
)
+
formed via addition of 10
equiv of N
2
H
4
(green trace), and the consumption of P
3
B
Fe(N
2
H
4
)
+
via addition of 10 equiv
of [Ph
2
NH
2
][BAr
F
4
] (blue trace) or 10 equiv of Cp*
2
Co (red trace), in Et
2
O solution at –78
°C.
Garrido-Barros et al.
Page 13
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Figure 5.
(a) Scheme representing the reactivity of P
3
B
Fe
+
in the presence of N
2
H
4
upon one-electron
reduction. (b) Cyclic voltammetry of P
3
B
Fe
+
(1 mM) in the absence (black trace) or
presence of 1 equiv of N
2
H
4
(red trace) in a 0.1 M [Na][BAr
F
4
] solution in Et
2
O at RT.
Set-up used a glassy carbon working electrode, a Pt-wire counter electrode, and 5 mM
Ag/AgOTf reference electrode; the scan rate was 100 mV·s
-1
.
Garrido-Barros et al.
Page 14
Angew Chem Int Ed Engl
. Author manuscript; available in PMC 2024 February 20.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript