ARTICLE

DOI: 10.1038/s41467-018-03795-8 OPEN

Heterogeneous Fe3 single-cluster catalyst for

ammonia synthesis via an associative mechanism
Jin-Cheng Liu 1, Xue-Lu Ma1, Yong Li1, Yang-Gang Wang 1, Hai Xiao 1 & Jun Li 1

The current industrial ammonia synthesis relies on Haber–Bosch process that is initiated by
the dissociative mechanism, in which the adsorbed N2 dissociates directly, and thus is limited
1234567890():,;

by Brønsted–Evans–Polanyi (BEP) relation. Here we propose a new strategy that an anchored

Fe3 cluster on the θ-Al2O3(010) surface as a heterogeneous catalyst for ammonia synthesis
from first-principles theoretical study and microkinetic analysis. We have studied the whole
catalytic mechanism for conversion of N2 to NH3 on Fe3/θ-Al2O3(010), and find that an
associative mechanism, in which the adsorbed N2 is first hydrogenated to NNH, dominates
over the dissociative mechanism, which we attribute to the large spin polarization, low
oxidation state of iron, and multi-step redox capability of Fe3 cluster. The associative
mechanism liberates the turnover frequency (TOF) for ammonia production from the lim-
itation due to the BEP relation, and the calculated TOF on Fe3/θ-Al2O3(010) is comparable to
Ru B5 site.

1 Departmentof Chemistry and Key Laboratory of Organic Optoelectronics & Molecular Engineering of Ministry of Education, Tsinghua University, Beijing
100084, China. Correspondence and requests for materials should be addressed to J.L. (email: [email protected])

NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8

A
mmonia synthesis is one of the most important industrial Several molecular catalysts and naturally occurring nitrogenase
catalytic reactions, which is based on the Haber–Bosch enzymes are capable of reducing N2 under ambient conditions17–
process (N2 + H2 → NH3) since the World War II, and it 21, and the associated mechanisms are likely initiated by the
plays a key role in the growth of human population1,2. Although associative adsorption or hydrogenation of N2, rather than the
the process using Fe and Ru metal-based catalysts with promoters dissociative adsorption that is the key to the Haber–Bosch pro-
has been developed for more than one hundred years, it still cess. Recently, there were some theoretical indications showing
requires high pressure (~100 bar) and moderately high tem- that in the electrochemical ammonia synthesis the N2 molecule
perature (~700 K)3, which is dictated by the compromise between did not dissociate upon adsorption, but was hydrogenated to
thermodynamic equilibrium and kinetics4. Ammonia synthesis *NNH instead22–29. Even for the thermal catalytic process, Skú-
on Fe and Ru metal surfaces is widely regarded as a classical lason et al.25 showed that the proportion of associative process
example of correlating the experimentally observable turnover (i.e., *N2 + *H→*NNH + *) is underestimated based on Bayesian
frequency (TOF) with the predicted atomistic mechanism5, where statistics. For heterogeneous catalysis, O2 and CO can be
the N2 activation is confirmed to be a direct N≡N dissociation hydrogenated via an associative mechanism to *OOH and *HCO,
process6–8. The performance of promoted Fe- and Ru-based respectively, which have been proved to be key intermediates for
catalysts clearly shows a site dependence of the N2 dissociation O2 and CO activation30–36. When the N2 hydrogenation becomes
and NHx desorption9. N2 molecules are firstly dissociated on the dominating process, the N–N bond is much weakened, and
specific active sites, such as the B5 site of Ru(0001) step and the consequently the dissociation barrier no longer obeys the BEP
C7 site of Fe(111) or Fe(211) surface5,10,11, then the adsorbed *N relation. Thus, designing a catalyst with surface active centers that
is hydrogenated step by step to produce *NH3. This is the well- hydrogenate N2 first can be a new strategy to accomplish
studied dissociative mechanism. ammonia synthesis at low temperature and pressure.
Previous theoretical studies discussed the possibility of The remarkable recent development of surface single-atom
ammonia synthesis at low temperature and low pressure12,13, but catalyst (SAC) and single-cluster catalyst (SCC) demonstrates the
the TOF was limited, due to the Brønsted–Evans–Polanyi (BEP) possibility to build homogeneous catalytic active centers on het-
relation14,15. The BEP relation regulates that the dissociation erogeneous solid surfaces37–41. Inspired by nitrogenase, in which
barrier of N2 and the desorption energies of NHx scale linearly the FeMoco is responsible for N2 activation and ammonia
with the adsorption energy of N atom12,14,16. Stronger adsorption synthesis42–45, Holland and co-workers designed a series of
of N atom implies lower N2 dissociation barrier but higher NHx multinuclear iron complexes to mimic the nitrogenase, and
desorption energies, such as on Re, Mo, Fe metal surfaces; while showed that the formally Fe(I) and Fe(0) complexes can weaken
weaker adsorption of N atom indicates higher N2 dissociation or even break N2 triple bond at or below room tempera-
barrier and lower NHx desorption energies, such as on Pd, Co, Ni ture18,19,46. Hosono and co-workers also reported a series of
metal surfaces. Thus a good metal catalyst for ammonia synthesis stable electrides as efficient electron donor for loaded Ru
must have a moderate atomic N adsorption energy, around where metal, which is proved to be more reactive than the bare metal
the top of volcano plot locates15,16. surface47.

a b Al
O O
O
Al Al
Fe Fe
HN NH
Fe N N
O O
NH HN
Fe
Al Al
Fe N N Fe
NH O O
HN O
Al Al

c d μ3−η2:η1:η1 e μ3−η1:η1:η1 f N2 on Fe(211)

Fe3

γ-Al2O3(010)

1.26 Å/ –1.13 |e| 1.26 Å 1.33 Å/ –1.55 |e|

–1.19 |e|

Fig. 1 Homogeneous and heterogeneous Fe3 cluster with N2 adsorption. a Schematic representation of N2 coordinated with three Fe(I)-ion homogeneous
complexes in the side-on/end-on/end-on (μ3−η2:η1η1) configuration; b schematic representation of N2 coordinated with heterogeneous Fe3/θ-Al2O3(010)
in the same configuration; c optimized Fe3 cluster on θ-Al2O3(010); d, e optimized configurations of N2 adsorption on Fe3/θ-Al2O3(010); f N2 adsorption
configuration on the C7 site of Fe(211) surface

2 NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8 ARTICLE

1 A2 ts A3
N-N dissociation

0.25
A1 NNH
0 –0.10 1.33 Å 1.72 Å
dissociation 2.78 Å

DFT energy /eV

A2
–1 –0.99
–0.60
–1.31 A4 ts A5
B1 B2
–1.13
–1.78
–2 –2.45
A3 B3
–2.4
A5

–3 –2.92 –2.48
A4

B1 ts B2 ts B3

1.33 Å 1.35 Å 1.40 Å 1.74 Å

Fig. 2 The energy diagram for the first two steps of NH3 synthesis on Fe(211) C7 site. The black line is for the dissociative mechanism, where *N2
dissociates directly, and the blue line for the associative mechanism, where the N–N bond breaks only after hydrogenation to the *NNH species

In this work, we propose an active center of Fe3 cluster that is the most stable clusters on Al2O3 substrate, with the Fe3 cluster
anchored on the θ-Al2O3(010) surface, and we predict that the kinetically stable against aggregation.
direct dissociation of *N2 on this center is difficult (dissociative Here we focus on the catalytic performance of Fe3 on ammonia
mechanism), but the *N2 is easily hydrogenated to form the synthesis, while we have also proved that the pyramid Fe4 cluster,
*NNH species (associative mechanism), which has a much lower consisting of four triangular planes, exhibits similar catalytic
N–N bond dissociation barrier than that of *N2. We further activity to that of Fe3 cluster (Supplementary Fig. 3). Fe3/θ-
reveal that the large spin polarization of Fe3 is responsible for N2 Al2O3(010) with magnetic moment of 10 μB is the most stable,
activation, and the low oxidation state iron atom works as an and the state of 8 μB is only 0.08 eV less stable (Supplementary
electron reservoir, regulating the charge variation of the whole Fig. 4), which is consistent with the ab initio molecular dynamics
process. The surface-anchored Fe3 SCC renders a robust multi- (AIMD) simulations in which the magnetic moment oscillates
step redox capability necessary for ammonia synthesis from between 10 and 8 μB (Supplementary Fig. 5g). Such large spin
dinitrogen. polarization on Fe sites is thought to be one of the key factors to
activate N2 on FoMoco and Fe complexes19,43,56.
Results Two of the most stable configurations of N2 adsorption on the
Fe3 site are shown in Fig. 1d and e, which are denoted as
N2 adsorption on supported Fe3 cluster. The FeMoco site of
nitrogenase has been considered responsible for biological μ3−η2ːη1ːη1 (side-on/end-on/end-on) and μ3−η1ːη1ːη1 (end-on/
end-on/end-on), respectively. These two configurations are also
nitrogen fixation. Its performance in activating N2 molecule
originates from the highly efficient redox cycle between Fe(II) and observed in the AIMD simulations (Supplementary Fig. 5). N2
Fe(III) of the Mo-Fe-S-C cluster of FeMoco. Based on the model can also adsorb at single or double Fe sites as μ2−η2ːη1, μ1−η1,
of FeMoco, various Fe containing complexes are designed to and μ1−η2 configurations57,58, but these are less stable here
mimic the N2 activation process on FeMoco19,20,44,48–51. A gen- (Supplementary Fig. 6). The Bader charge on N2 decreases to
eral implication for designing such complex is to keep the Fe −1.13 |e| with its bond length significantly stretched from 1.10 Å
center at a reduced state to facilitate electron donation to the N2 in gas phase to 1.26 Å, close to that in N2F2. In comparison, the
molecule. N–N bond length on C7 site of Fe(221) surface with high surface
In recent years, embedded Fe clusters with low oxidation state, energy is 1.33 Å with Bader charge of −1.55 |e| (Fig. 1f).
such as three diketiminate-bound Fe(I) ions in Fig. 1a, are shown
to synergistically facilitate N2 reduction18,19,46,48,52. Inspired by Competition between dissociative and associative hydrogena-
this finding, we conceive that the small Fe clusters supported by tion. Lots of efforts have been made to investigate the mechanism
θ-Al2O3(010) surface should be capable of efficient N2 activation of ammonia synthesis on metal surfaces. It is found that the N2
(Fig. 1b, c), because the inert support has little electronic molecule dissociates directly at specific sites of the surface, such as
interaction with the Fe cluster and thus retains Fe in an even the B5 site of Ru(0001) step surface and the C7 site of Fe(111) or
more reduced state for metal-metal bonded Fe clusters3,53. This Fe(211) surface5,10,59. The barrier of N2 direct dissociation over
type of Fe clusters on alumina surface may be experimentally close-packed Ru(0001) surface is as high as 1.9 eV but lowers to
prepared by soft-landing cluster method or thermal treatment of 0.4 eV on the step site6. Different from on the Fe(C7) site, the
ligated tri-iron cluster (e.g., Fe3(CO)n)54,55. By testing the binding binding energy of *NHx on Ru surface is lower. According to the
energies for Fen clusters (n = 1–5) (Supplementary Fig. 1 and 2), BEP relation, the dissociation barrier of N2 (ΔG≠(N2)) and des-
we have shown that the triangular Fe3 and the pyramidal Fe4 are orption energies of *NHx (ΔG(*NHx)) scale linearly with the

NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8

a b c a2 ts a3
1 NN break 1
N-N dissociation Distal I
Alternative I Distal II
a1 a1
0 NNH break Alternative II 0
Alternative III
HNNH break

DFT energy /eV

DFT energy /eV

–1 a2
–1 a2 *HNNH2
H2NNH break
b2 a1 b2 a1
c1
–2 b1 a3 a4 –2 b1 e1
e4
d1 d2 e2 d2 b2 ts b3
b3
a5 e5
–3 a13 –3 e3 a13
a6 a12 a12
*N/NH3 e6
a9
–4 a7 –4 a9
a8 a8
a11 a11

–5 a10 –5 a10

d –0.42(1.89) –0.44 –0.88(0.45) –0.43(1.41) –0.61 –0.28(1.32) 0.21(1.22) –1.02

H H H H H
H H H H H H H H H H H H H H
N N N N N N N N N N N N N
Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe
Fe Fe Fe Fe Fe Fe
Fe Fe Fe Fe Fe Fe Fe
Fe N N N H H Fe H
N
a2 a3 a4 a5 N a6 a7 a8 H a9 a10 H
–1.15(0.95)
–0.75 –0.62(1.11) –0.96(0.61)
0.35(0.98) 0.05(0.45) –1.20(0.45)
H H
H H H
–1.01
N N N N N N
N N H
Fe –0.61
Fe Fe Fe Fe Fe Fe Fe Fe Fe
H Fe H Fe H Fe
H Fe H Fe H N H N
b2' b3 H a11
b1 b2
0.40(1.42)
–0.23(1.16) –1.54(0.80) 1.01
0.38(1.32)
Fe Fe H H H H
H H H H
Fe N N N N H
N Fe Fe
Fe Fe N
a1 Fe Fe Fe
Fe Fe H Fe Fe
c1 a6 a12 a13 Fe
–0.60 0.02(1.24) –2.04(0.64)
–0.25(1.31) 0.99
H H H H H H H H
N N 0.29(0.91)
N N N N Fe Fe
Fe Fe Fe Fe Fe Fe Fe
Fe Fe Fe
H H H H
d1 d2 a8

0.03(1.24)
H H H H
H H H
H H N N
N N N N N N N
Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe Fe
Fe Fe Fe Fe Fe Fe
H e2 H H H H H
e1 e3 e4 e5 e6

–0.60 –0.98(0.95) 0.92 –0.81(0.84) –0.35

Fig. 3 Energy diagrams for ammonia synthesis. a The dissociative mechanism, and three pathways of associative mechanism with N–N bond dissociation at
*NNH, *HNNH, and *HNNH2 intermediates by the alternating hydrogenation route. b Two pathways of associative mechanism by the distal hydrogenation
route. c Initial, transition, and final states of *N2 and *NNH dissociation step. d Schematic depiction of the six reaction pathways for conversion of N2 to
NH3 catalyzed by Fe3/θ-Al2O3(010). Reaction energies are shown for every step and barriers are enclosed in brackets. The dissociation steps of *N2,
*NNH, *HNNH, and *HNNH2 intermediates are enclosed with dashed lines

adsorption energy of N atom (ΔG(*N)) on the same type of 1.38 eV, which is much higher than the *N2 direct dissociation
sites16, so the dependence of ΔG≠(N2) and ΔG(*NHx) on ΔG(*N) barrier. Thus, it is the high hydrogenation barrier initiating the
can be fitted into lines. It is found that the intercepts of such lines associative mechanism that makes the dissociative mechanism
for step sites are lower than those for terrace sites, but the slopes dominate on Fe(211) surface.
are similar for different kinds of sites16. Thus, with the N2 dis- Based on the mechanism on metal catalysts, we investigate the
sociative mechanism on metal surfaces (the slopes of ΔG≠(N2) reaction pathways of N2 activation on single-cluster catalyst Fe3/
and |ΔG(*NHx)| against |ΔG(*N)| are of opposite signs), high θ-Al2O3(010), as shown in Fig. 3 and Supplementary Fig. 8. The
temperature and high pressure are intrinsically needed to over- calculated barrier of *N2 direct dissociation is 1.89 eV (Fig. 3c, d),
come ΔG≠(N2) and ensure the desorption of NHx but the associative hydrogenation of *N2 to *NNH only has a
simultaneously12,14. barrier of 0.98 eV, which can be driven over with relatively low
When N2 adsorbs on Fe(211) surface, the anti-bonding π* temperature. Note that the H2 dissociative adsorption barrier is as
orbitals of N2 interact strongly with Fe metal surface, and each N low as 0.05 eV (Supplementary Fig. 9), which can be neglected
is coordinated with four surface Fe, so N2 cannot be comparably. The *N2 dissociation is a structure-sensitive process,
hydrogenated without reconstruction. As shown in Fig. 2, the which usually needs more than 5 surface metal atoms’
*NNH species (B2) has to rotate from the original *N2 (B1), with cooperation. The geometry of Fe3 cluster is similar to a close-
the N–N bond length increased to 1.40 Å. Although the packed Fe(111) surface, where the dissociation of *N2 is not
dissociation barrier of *NNH (i.e., from B2 to B3) is only 0.32 favored. However, the electronic structure of supported Fe3
eV, the hydrogenation barrier of *N2 (from B1 to B2) is as high as cluster is distinct from that of metal surface, which will be

4 NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8 ARTICLE

discussed in the electronic structures section. On metal surface 1.5

4
sites, such as Ru B5 site, the barrier of *N2 hydrogenation is about

2.97

2.98
2.61
2.45

3.64
3.38
0.2 eV higher than that of *N2 dissociation, which leads to around

1.74
1.72

2.89

1.63
2.56
three orders of magnitude difference between the rate con- 2 1.4

1.40 Å
stants25. On the contrary, on Fe3/θ-Al2O3(010), the barrier of *N2

0.59

1.82

0.59
0.53

N-N bond length /Å

Bader charge /|e|

1.36 Å
*N2H2
hydrogenation is 0.91 eV lower than that of *N2 dissociation, and 0 Bader charge of Fe3
the calculated rate constants are 9.8 × 105 s−1 and 5.2 × 10−1 s−1

*N2
a1
a2 b1 1.3

Fe3
at 700 K and 100 bar, respectively.

*NH/NH/H2
b2

1.26 Å
–2

1.26 Å

*NH2/NH2/H2
b2’

*NH2/NH/H
*NNH/H

*NH3/NH2/H
*NH2/NH2
*NNH/H
b3 1.2
Reaction mechanisms. As is shown earlier (see also the movie file

*NH2/H
*N/NH/H
–4 a6 a7
in the supplementary material), N2 is first activated on Fe3 active

1.10 Å

*NH/NH
a8

*NH3
a9 a10

Fe3
center followed by attacking by dissociated H atom, which differs a11 a12
1.1
–6
from the procedure in electrochemical condition where proton N-N bond length a13 a14
Bader charge of dinitrogen
attacks the adsorbed N2 in solution. With thermochemical con- +H2 N-N break +H2 +H2
dition, after the first N2 hydrogenation step, it is more favorable
for *NNH to dissociate into *N and *NH rather than further Fig. 4 Changes of N–N bond length and charges of Fe3 and the adsorbates.
hydrogenation to *HNNH or *NNH2. Such associative process is Black, blue, and red curves represent N–N bond lengths, Bader charges of
believed to occur in the homogeneous catalysis and enzymatic Fe3 cluster, and Bader charges of dinitrogen at every step of the catalytic
mechanism44,50,60,61. As shown in Fig. 3d, the *NNH can migrate cycle with the associative mechanism
from Fe3 to the Fe3/θ-Al2O3(010) interface (i.e., b2 to b2′) with a
barrier of 0.45 eV, where the N-end coordinates with two Fe polarization and low oxidation state metal provides a heterogeneous
atoms and the NH-end coordinates with one Fe and one substrate single-cluster catalyst design to mimic the FeMoco in nature.
Al ion. The dissociation barrier of *NNH in b2′ is only 0.45 eV We further tracked the Bader charges of Fe3 cluster and the N-
with an exothermic reaction energy of −1.20 eV (Fig. 3c, d). containing adsorbates along the reaction pathway with associative
Afterwards, the upper *N atom moves to the Fe3 3-fold site, while mechanism, as shown in Fig. 4. The highly reducing Fe3 cluster is
the lower *NH is anchored at the Fe3/θ-Al2O3(010) interface by strongly oxidized during the whole process. When N2 and H2 are
coordinating with one surface Al ion and two Fe atoms. adsorbed on Fe3 cluster, the Bader charge of Fe3 increases from
The *NNH can also be hydrogenated further via the alternating 0.59 to 2.97 |e|, while N2 and H2 are reduced to *N2– and 2*H–
or the distal pathway49,51,56. In the alternating pathway, *NNH is adsorbates. The bond order of N2 is reduced to 2.5, with its bond
hydrogenated to *HNNH43 with a barrier of 1.16 eV (b2 to c1), length of 1.26 Å and stretching vibrational frequency of 1384 cm
and the N–N bond length is elongated to 1.42 Å with the N–N −1. Then, one *H− combines with *N – to form a *NNH– species,
2
stretching frequency of 1012 cm−1, suggesting a single bond with one electron released back to Fe3 simultaneously. The N–N
between N atoms. The formed *HNNH species can either bond length is then lengthened to 1.36 Å with stretching
dissociate into two *NH group or be hydrogenated to *H2NNH frequency of 1099 cm−1, and the N–N bond order becomes 2.0
that can again dissociate into *NH2 and *NH. The *NHx (x = with around two electrons occupying its π* orbitals. Bader
1–2) species can be ultimately hydrogenated to *NH3. charges on Fe3 and *NNH turn to 2.45 and −1.13 |e|, respectively.
In the distal pathway, *NNH is hydrogenated to *NNH2 where When *NNH migrates from the Fe3 cluster to the Fe3/θ-
both H atoms are on the same end of N–N bond. This process Al2O3(010) interface, the N–N bond length is further stretched
requires a barrier of 1.31 eV, slightly higher than that of the to 1.40 Å with Bader charge of −1.47 |e|. As shown in Fig. 4, Fe3
alternating pathway. Next, *NNH2 can be further hydrogenated is a bifunctional multi-step redox active center for donating
to *HNNH2 or *N + *NH3. The latter step features spontaneous electrons at the adsorption steps and accepting electrons at the
N–N bond dissociation with a barrier of 0.95 eV, while the former hydrogenation steps. Thus, the surface-anchored Fe3 single
step drives the distal pathway back to the alternating one with a cluster serves as an electron reservoir that regulates the charge
barrier of 1.24 eV. variation of the whole process.
Overall, we find that the most hydrogenation steps experience
barriers ranged from 1.0 to 1.3 eV, which are close to those on
metal surfaces5,6, but the indirect N–N bond breaking via *NNH Electronic structures. To elucidate the bonding nature of the
has only a barrier of 0.45 eV, much lower than the direct *N2 species involved in the associative mechanism, we investigate the
dissociation. As a result, the N–N bond breaking step is not the densities of states (DOS) of N2 adsorption on Fe3/θ-Al2O3(010)
rate-determining step (RDS) of ammonia synthesis on surface- (with the most stable configuration μ3−η2ːη1ːη1) and C7 site of
supported Fe3 cluster, distinct from on the traditional metal Fe(211) (Fig. 5a and Supplementary Fig. 10) for comparison. The
catalyst surfaces. energy levels of Fe3 minority β-spin d orbitals and N2 π* orbitals
The N–N bond dissociation via *NNH bypasses the direct are well matched, leading to partial occupation of the formed d-π*
dissociation of *N2 that is one of the severe limitations for the orbitals. While the strong spin polarization provides large
Haber–Bosch process. On the Sabatier volcano curve14, the *N2 exchange stabilization energy for the majority α-spin orbitals,
dissociation is replaced by the *NNH dissociation, which has a leading to that the energy levels of Fe3’s α-spin d orbitals are
lower barrier that can be driven over with moderate thermo- about 2.5 eV lower than the π* orbitals of N2, and thus no obvious
dynamic conditions. Therefore, the RDS is no longer the interaction of α orbitals is observed. This indicates that only the β
dissociation of N2 but desorption of NHx species. Such an π* orbitals of *N2 are partially occupied, and thus forcing *N2 to
associative mechanism for N2 activation is what nitrogenase does be of radical nature (with unpaired electron, Fig. 5c) that is active
in nature43, and has been reproduced by metal complexes for for hydrogenation. Therefore, the large spin polarization on Fe3
homogeneous catalysis, where the N2 triple bond is first weakened cluster is responsible for the activation of N2. When *N2 is
by single-metal or multiple-metal center, and then hydrogenated by hydrogenated to *NNH, one electron transfers from hydrogen to
protons and electrons toward NH3 formation42. The surface- the α π* orbitals of *N2, and now both α and β DOS of Fe3’s d
anchored Fe3 center with multi-step redox capability, large spin orbitals overlap with NNH’s π* orbitals, which further weakens

NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8

Spin density
a Fe3 Fe3N2 N2 Fe3 Fe3N2 N2 c
α β α β α β α β α β N N
5
d-π*(unocc) Fe Fe Fe
π*(unocc)
π* π* π*
0
d-π*

Energy /eV
d Charge density
d orbitals difference
d-π*(occ)
d orbitals –5 σ
σ N N
π
π Fe Fe Fe
–10

b Fe3 Fe3NNH N2 Fe3 Fe3NNH N2 e

α β α β α β α β α β Fe3 Fe3-N2 N2
5
d-π*
d orbitals (unocc)

Energy of β MOs /eV

d-π*(unocc) π* π*
0 πt
d-π*
Energy /eV

d-π*(occ) d-π* π*
σt
d-π*(occ)
d orbitals –5 σ
σ
π
π
–10

Fig. 5 Electronic structure analysis. a Projected electronic densities of states (pDOS) and schematic illustrations of 3d orbitals of Fe3 cluster on θ-
Al2O3(010), 2p-orbitals of the N2 gas molecule, and their interaction within Fe3N2/θ-Al2O3(010) of μ3−η2:η1η1 configuration. b pDOS and schematic
illustrations for the Fe3NNH/θ-Al2O3(010) case. c Spin density of Fe3N2/θ-Al2O3(010). (yellow stands for spin up and cyan for spin down) d Charge
density differences (δρ = ρA+B−ρA−ρB) of N2 adsorption on Fe3/θ-Al2O3(010) (cyan stands for holes and yellow for electrons). e The major interactions
and energy levels of the scalar relativistic Kohn–Sham β-spin MOs of isolated Fe3N2 with correlation to the orbitals from Fe3 and N2 fragments

the N–N bond. Therefore, the more interaction and thus larger increase of temperature, the concentrations of NHx decrease and
occupation of the π* orbitals in NNH leads to a lower N–N bond that of bare site increases, since the entropy of free gas molecules
order, and is thus responsible for the lower dissociation barrier becomes dominant. The calculated TOF of ammonia synthesis on
than that of *N2. Fe3/Al2O3 is 1.4 × 10−2 s−1 site−1 at 100 bar and 700 K, which is
Fragment orbital analysis is further performed to provide more comparable to that on the Ru B5 site12. As shown in Fig. 6b, the
details of such interaction between isolated Fe3 and N2 (Fig. 5e, contribution from the associative mechanism is six orders of
Supplementary Fig. 11 and Supplementary Table 1). The main magnitude larger than from the dissociative mechanism. While
contribution to the interaction is from the tangential σt, πt on the Ru step surface, the TOF of the associative mechanism is
molecular orbitals of Fe3 and N2 π* orbitals, which lead to two calculated to be three orders of magnitude lower than that of the
bonding orbitals and two anti-bonding orbitals. This bonding dissociative mechanism25.
model shows that the electrons from metal d orbitals partially Based on the above microkinetic analysis of the ammonia
transfer to the empty π* orbitals of N2, which is consistent with synthesis on Fe3/θ-Al2O3(010), we further compare it with the
the electron transfer shown in the charge density difference widely reported cases on Fe C7 site and Ru B5 site (Fig. 6c)5,6,12,62.
(Fig. 5d) and the DOS analysis. Remarkably, in homogeneous The coverages and free energy diagrams for surface species are
catalysis, the N2 activation process is also initiated with the shown in Supplementary Fig. 13 and 14. For Fe C7 site, although
electron transfer from the electron-rich metal center to the empty the transition state energy of *N2 dissociation is only −0.10 eV
π* orbitals of N2, which lowers the bond order in N2 and with respect to the gas phase N2 and clean surface, the adsorption
facilitates the hydrogenation process. energy of two *N is as high as −2.4 eV (Fig. 2), which results in
dominant coverage of *N on the active center (Supplementary
Microkinetic simulations. To explore the reactive performance Fig. 14f). Thus, the RDS on Fe C7 site is the desorption of *NHx
of ammonia synthesis under realistic conditions, we performed species. For Ru B5 site, the transition state energy of *N2
comparative kinetic analysis of Fe3/Al2O3, Fe surface, and Ru dissociation is 0.4 eV, with two *N adsorption energy of −0.8 eV.
surface based on the free energy calculations. (Supplementary At low (<450 K) and high (>450 K) temperature, the Ru B5 site is
Fig. 12 and 13) The TOF map (Fig. 6a) is calculated under the covered by *NH2 and *H, respectively. Thus, Ru B5 site is the
pressure range of 1~100 bar and the temperature range of closest to the top of volcano curve, which balances the *N2
300–1000 K. The TOF of ammonia production on Fe3/Al2O3 is dissociation and *NHx desorption processes.
less than 10−10 s−1 site−1 below 400 K because of too stable For our single-cluster catalyst Fe3/θ-Al2O3(010), the Fe3 active
adsorption of NHx species (Supplementary Fig. 14a). At high site is covered by *NH2 and *NH3 at low temperature, and at
temperature and low pressure, the conversion of NH3 is lower constant temperature or pressure (Supplementary Fig. 14), the
than 10%, and the decomposition of ammonia occurs. With the TOF on Fe3/θ-Al2O3(010) is comparable to that on Ru B5 site,

6 NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8 ARTICLE

a 100 b
700 K 100 bar 0 100
*N2 associative
–2 mechanism
80 10–2
–4

log(TOF) /s–1site–1

TOF /s–1site–1
10–4

Pressure /bar
–6
60
–8
10–6
*N2 dissociation
–10 mechanism
40 10–8
–12
–14 10–10 100 bar, 700 K
20 NH3=10%, NH3=10%,
–16 N2:H2 = 1:3 N2:H2 = 1:3
–18 10–12

300
400
500
600
700
800
900
0

20

40

60

80

100
1
400 600 800 1000
Temperature /K Pressure /bar
Temperature /K

c Fe(211) C7 site d 10–1

e
700 K Ru(0001) step Metal steps Pd
100 bar 3 K-promoted metal steps
10–2
TOF /s–1site–1

Fe3/θ-Al2O3(010)
2

EN-N /eV
10–3 Fe3/θ-Al2O3(010)
Ni

N2 dissociation Rh
Ru(0001) step B5 site
Fe(211) 1 Co
10–4 Ru

0 Fe
10–5 NNH dissociation

10–6 –1
0.0 0.2 0.4 0.6 0.8 1.0 –1.5 –1.0 –0.5 0.0 0.5 1.0
pN2/(pN2 + pH2) EN /eV

Fig. 6 Microkinetic simulations. a TOF per site of ammonia synthesis on Fe3/θ-Al2O3(010) mapped with pressure (1–100 bar) and temperature
(300–1000 K). H2:N2 ratio is fixed at 3, and the conversion ratio of NH3 is 10%. b TOF contributions from the associative mechanism (black curve) and the
dissociative mechanism (red curve) at constant pressure of 100 bar and constant temperature of 700 K, respectively. c Structures of B5 site on Ru(0001)
step surface and C7 site on Fe(211) surface. d TOFs per site of ammonia synthesis over the three catalysts as a function of N2 partial pressure at 700 K and
100 bar. e Calculated transition state energies (EN–N) for N2 dissociation or NNH dissociation as a function of the nitrogen adsorption energy (EN) over step
sites on Fe, Ru, Rh, Co, Ni, and Pd surfaces. The solid blue line represents a least-squared interpolation between the points, the red line depicts the cases
with addition of alkali (potassium) promotion. Note that the energy data for the metal surfaces in e are taken from ref. [12]

and is about two orders of magnitude larger than on Fe C7 site. of the transition metal can also affect the N–NH bond breaking
(Supplementary Fig. 15) By plotting the TOF against the partial barrier. Especially, the procedure of N2 activation and hydro-
pressure of N2 as shown in Fig. 6d, the maxima for Fe C7, Fe3/θ- genation can be affected by moisture or solvents if existing.
Al2O3(010), and Ru B5 site are 0.06, 0.44, and 0.78, respectively. Further study of the optimal surface metal clusters and support
As the Fe C7 site is mainly covered by *N, lowering the N2 partial for N2-to-NH3 conversion under different chemical conditions
pressure from 25% to 6% increases the TOF. On the contrary, for will be interesting. The nature of the surface SCCs will likely offer
Ru B5 site, the main surface species is *H at 700 K and 100 bar, high selectivity like single-atom catalysts.
and thus increasing the proportion of N2 accelerates the reaction.
For Fe3/θ-Al2O3(010), to reach a high TOF, one should balance
the partial pressure of N2:H2, because the associative mechanism Discussion
requires the co-adsorption of N2 and H2 to form the *NNH We propose a surface-anchored Fe3 single-cluster active center
species, which is the key intermediate as discussed above. for ammonia synthesis by first-principle calculations and
As shown in Fig. 6e, for most active sites on metal surfaces, microkinetic analysis. This Fe3 cluster can be stably anchored on
there is a linear relation (i.e., BEP relation) between the the θ-Al2O3(010) surface, and its multi-step redox capacity, large
adsorption energies of N atom (EN) and the transition state spin polarization and low oxidation state metal enable efficient N2
energies (EN–N, which is equivalent to the apparent activation activation, due to the spin-polarized charge transfer from Fe’s 3d
barrier) for N2 dissociation. On Fe3/θ-Al2O3(010), although the orbitals to N2 π* orbitals. The partial occupation of N2’s β-spin π*
EN–N is above the line for metal cases, the transition state energy orbitals both lowers the N–N bond order and also grants *N2 a
of NNH dissociation is even much lower than the line for novel radical nature that leads to an associative mechanism for N2
potassium-promoted metal cases. It is thus possible to bypass the activation, which mimics the initiation process in the nitrogenase
BEP relation with the associative mechanism, and the limitation as well as artificial metal complexes for homogeneous catalysis
underlying one side of the volcano curve is then removed. As a involving nitrogen fixation.
counterpart, very recently Chen et al. showed63,64 very interesting We predict the whole catalytic mechanism for ammonia
cases that the other scaling relation between EN and desorption synthesis on Fe3/θ-Al2O3(010) that is distinctly different from on
energies of NHx can also be broken by nitrogen transfer from the industrially used Fe and Ru metal surfaces. The dissociative
metal to LiH, which separates the N2 dissociation site from the mechanism dominates the ammonia synthesis on metal surfaces,
hydrogenation and desorption site. With these strategies of where *N2 dissociates directly. Thus, the TOF of ammonia
breaking the BEP relation, ammonia synthesis with ambient synthesis on metal surfaces obeys the BEP relation that demands
conditions does not seem to be impossible. While this study focus the balance between N2 dissociation and NHx desorption. How-
on the Fe3/Al2O3 system, it is conceivable that change of support ever, we find that in our associative mechanism at the single-
can influence the charge state of the metal cluster and alteration cluster catalyst the first hydrogenation of N2 to *NNH is much

NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications 7

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8

faster than the dissociative mechanism on Fe3/θ-Al2O3(010), and Data avalability. All other data supporting the findings of this study are available
the following dissociation of *NNH only has a barrier of 0.45 eV. in the article and its Supplementary Information files and from the corresponding
author on request.
Remarkably, the associative mechanism bypasses the BEP relation
and thus the limitation underlying one side of the volcano curve.
Such surface-anchored Fe3 center represents a class of new cat- Received: 26 October 2017 Accepted: 13 March 2018
alyst—single-cluster catalyst, which features identical yet isolated
active centers on support and thus bridges the gap between
heterogeneous and homogeneous catalysis, serving as a hetero-
geneous catalyst design that enables the associative mechanism
for ammonia synthesis from dinitrogen.
The calculated TOF of ammonia synthesis on Fe3/θ- References
Al2O3(010) is comparable to that on Ru B5 site, which is known 1. Haber, F. & van Oordt, G. Z. Über die Bildung von Ammoniak den
as the most active metal catalyst, and is two orders of magnitude Elementen. Anorg. Chem. 44, 341–378 (1905).
2. Ertl, G. Reactions at Surfaces: From Atoms to Complexity (Nobel Lecture).
faster than Fe C7 site. Thus, the anchored Fe3 center is a pro- Angew. Chem. Int. Ed. 47, 3524–3535 (2008).
mising candidate heterogeneous catalyst for highly selective 3. Cherkasov, N., Ibhadon, A. O. & Fitzpatrick, P. A review of the existing and
ammonia synthesis via the associative mechanism. In the future alternative methods for greener nitrogen fixation. Chem. Eng. Process Process
work, we will conduct extensive investigation of various surface- Intensif. 90, 24–33 (2015).
anchored metal trimer and polynuclear clusters, in order to find 4. Hammer, B. & Nørskov, J. K. Theoretical surface science and catalysis—
calculations and concepts. Adv. Catal. 45, 71–129 (2000).
the optimum of such design for ammonia synthesis toward high 5. Honkala, K. et al. Ammonia synthesis from first-principles calculations.
TOF at low temperature and low pressure. Highly stable single- Science 307, 555–558 (2005).
cluster catalysts with well-accommodating support hold promises 6. Logadóttir, Á. & Nørskov, J. K. Ammonia synthesis over a Ru(0001) surface
for rational design of highly selective and active catalysts for studied by density functional calculations. J. Catal. 220, 273–279 (2003).
complicated catalytic reactions such as N2-to-NH3 conversion. 7. Ertl, G., Lee, S. & Weiss, M. Kinetics of nitrogen adsorption on Fe (111). Surf.
Sci. 114, 515–526 (1982).
8. Bozso, F., Ertl, G., Grunze, M. & Weiss, M. Interaction of nitrogen with iron
Methods surfaces: I. Fe (100) and Fe (111). J. Catal. 49, 18–41 (1977).
DFT parameters. Energetics calculations for reaction mechanisms were carried out 9. Egeberg, R. et al. N 2 dissociation on Fe (110) and Fe/Ru (0001): what is the
by using spin-polarized density functional theory (DFT) with Perdew-Burke- role of steps? Surf. Sci. 491, 183–194 (2001).
Ernzerhof (PBE)65 generalized gradient approximation as implemented in VASP 10. Strongin, D., Carrazza, J., Bare, S. R. & Somorjai, G. The importance of
5.3.566. The cutoff energy of plane-wave basis set is 400 eV and single gamma-point C7 sites and surface roughness in the ammonia synthesis reaction over iron. J.
grid sampling was used for Brillouin zone integration. Atomic positions were Catal. 103, 213–215 (1987).
optimized until the forces were less than 0.02 eV/Å. Transition states were searched
11. Reuter, K., Plaisance, C. P., Oberhofer, H. & Andersen, M. Perspective: On the
by climbing image nudged elastic-band method (CI-NEB) and further confirmed
active site model in computational catalyst screening. J. Chem. Phys. 146,
by vibrational frequency analysis67.
040901 (2017).
12. Vojvodic, A. et al. Exploring the limits: A low-pressure, low-temperature
Computational model. A θ-Al2O3(010)-p(2 × 4) surface slab was used to model Haber–Bosch process. Chem. Phys. Lett. 598, 108–112 (2014).
the substrate with cell parameters of a = 11.24 Å, b = 11.79 Å, and c = 25.00 Å. The 13. Rod, T. H., Logadottir, A. & Nørskov, J. K. Ammonia synthesis at low
slab consists of seven O−Al layers, where the bottom two O−Al layer were frozen temperatures. J. Chem. Phys. 112, 5343 (2000).
while the remaining layers were allowed to relax. The slab lattice parameters were 14. Medford, A. J. et al. From the Sabatier principle to a predictive theory of
fixed to the optimized cell parameters of bulk θ-Al2O3, in order to mimic the transition-metal heterogeneous catalysis. J. Catal. 328, 36–42 (2015).
support bulk. The trinuclear Fe clusters were anchored by coordinating with sur- 15. Bligaard, T. et al. The Brønsted–Evans–Polanyi relation and the volcano curve
face oxygen as shown in Fig. 1c, and Supplementary Fig. 1. The formation energy
in heterogeneous catalysis. J. Catal. 224, 206–217 (2004).
of adsorbed iron clusters is defined as Ef = E(Fen/Al2O3)–E(Al2O3)–nE(Fe), where
16. Dahl, S., Logadottir, A., Jacobsen, C. J. & Nørskov, J. K. Electronic factors in
E(Fen/Al2O3) is the total energy of Al2O3 surface with Fen adsorbed, E(Al2O3) is the
catalysis: the volcano curve and the effect of promotion in catalytic ammonia
energy of pristine Al2O3 surface, and E(Fe) is the energy per atom of Fe metal. So
the formation energy includes both the formation energy of gas phase Fen cluster synthesis. Appl. Catal. A 222, 19–29 (2001).
and its binding energy on surface. Therefore, we can directly evaluate the stability 17. Sellmann, D. & Sutter, J. In quest of competitive catalysts for nitrogenases and
by Ef. The binding energy of Fe3 cluster is defined as Ebind = E(Fe3/Al2O3)–E other metal sulfur enzymes. Acc. Chem. Res. 30, 460–469 (1997).
(Al2O3)–E(Fe3), where E(Fe3) is the energy of gas phase Fe3 cluster. The adsorption 18. Rodriguez, M. M., Bill, E., Brennessel, W. W. & Holland, P. L. N2 reduction
energies of molecules are defined as Eads(X) = E(X/Al2O3)–E(Al2O3)–E(X), where and hydrogenation to ammonia by a molecular iron-potassium complex.
X is H2, N2, and NH3. The adsorption energy of N atom is defined as Eads(N) = 1/2 Science 334, 780–783 (2011).
(E(2 N/Al2O3)–E(Al2O3)–E(N2)), where E(2 N/Al2O3) is the total energy for N2 19. McWilliams, S. F. & Holland, P. L. Dinitrogen binding and cleavage by
dissociative adsorption. multinuclear iron complexes. Acc. Chem. Res. 48, 2059–2065 (2015).
20. Coric, I., Mercado, B. Q., Bill, E., Vinyard, D. J. & Holland, P. L. Binding of
dinitrogen to an iron-sulfur-carbon site. Nature 526, 96–99 (2015).
Chemical bonding analysis. Electronic structure analyses were performed using
21. Falcone, M., Chatelain, L., Scopelliti, R., Zivkovic, I. & Mazzanti, M. Nitrogen
spin-unrestricted DFT with PBE exchange-correlation functional and TZ2P Slater
basis sets as implemented in the Amsterdam Density Functional (ADF 2016.101) reduction and functionalization by a multimetallic uranium nitride complex.
program68. Frozen core approximations were applied to N [1 s2] and Fe[1s2–2p6]. Nature 547, 332–335 (2017).
The scalar relativistic (SR) effect was included by the zero-order-regular approx- 22. Montoya, J. H., Tsai, C., Vojvodic, A. & Nørskov, J. K. The challenge of
imation (ZORA)69. Fe3 and Fe3N2 were optimized under D3h and Cs point-group electrochemical ammonia synthesis: a new perspective on the role of nitrogen
symmetries, respectively, at all possible spin configurations. Molecular orbitals scaling relations. ChemSusChem 8, 2180–2186 (2015).
(MOs) of Fe3N2 and their corresponding contributions from Fe3 and N2 were 23. Skúlason, E. & Jónsson, H. Atomic scale simulations of heterogeneous
obtained from fragment MO analyses. electrocatalysis: recent advances. Adv. Phys. X 2, 481–495 (2017).
24. Skúlason, E. et al. A theoretical evaluation of possible transition metal electro-
catalysts for N2 reduction. Phys. Chem. Chem. Phys. 14, 1235–1245 (2012).
Microkinetic method. Microkinetic modeling was carried out using the CatMAP
25. Garden, A. L. & Skúlason, E. The mechanism of industrial ammonia synthesis
software package70. The model was constructed by numerically solving the dif-
ferential equations that describe the coverage of each surface intermediates under revisited: calculations of the role of the associative mechanism. J. Phys. Chem.
the steady-state approximation. The rate constant of each elementary step was C. 119, 26554–26559 (2015).
calculated by using harmonic transition state theory. The free energies for gas 26. Abghoui, Y. & Skúlason, E. Onset potentials for different reaction mechanisms
molecules were estimated using the ideal gas approximation considering the of nitrogen activation to ammonia on transition metal nitride electro-catalysts.
vibrational, rotational, and translational contributions to both entropy and Catal. Today 286, 69–77 (2017).
enthalpy, while the free energies for surface adsorbates were approximated using 27. Abghoui, Y., Garden, A. L., Howalt, J. G., Vegge, T. & Skúlason, E.
the harmonic approximation that treats all degrees of freedom as vibrational Electroreduction of N2 to Ammonia at Ambient Conditions on Mononitrides
modes. The steady-state TOFs were calculated based on the steady-state surface of Zr, Nb, Cr, and V: A DFT Guide for Experiments. ACS Catal. 6, 635–646
coverages. (2016).

8 NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-03795-8 ARTICLE

28. Abghoui, Y. et al. Enabling electrochemical reduction of nitrogen to ammonia 58. MacLachlan, E. A. & Fryzuk, M. D. Synthesis and reactivity of side-on-bound
at ambient conditions through rational catalyst design. Phys. Chem. Chem. dinitrogen metal complexes. Organometallics 25, 1530–1543 (2006).
Phys. 17, 4909–4918 (2015). 59. Somorjai, G. & Materer, N. Surface structures in ammonia synthesis. Top.
29. Zhao, J. & Chen, Z. Single Mo atom supported on defective boron nitride Catal. 1, 215–231 (1994).
monolayer as an efficient electrocatalyst for nitrogen fixation: A 60. Coric, I. & Holland, P. L. Insight into the iron-molybdenum cofactor of
Computational Study. J. Am. Chem. Soc. 139, 12480–12487 (2017). nitrogenase from synthetic iron complexes with sulfur, carbon, and hydride
30. Liu, J.-C., Tang, Y., Chang, C.-R., Wang, Y.-G. & Li, J. Mechanistic insights ligands. J. Am. Chem. Soc. 138, 7200–7211 (2016).
into propene epoxidation with O2–H2O mixture on Au7/α-Al2O3: a 61. Rao, L., Xu, X. & Adamo, C. Theoretical Investigation on the Role of the
hydroproxyl pathway from ab initio molecular dynamics simulations. ACS Central Carbon Atom and Close Protein Environment on the Nitrogen
Catal. 6, 2525–2535 (2016). Reduction in Mo Nitrogenase. ACS Catal. 6, 1567–1577 (2016).
31. Saavedra, J., Doan, H. A., Pursell, C. J., Grabow, L. C. & Chandler, B. D. The 62. Medford, A. J. et al. Assessing the reliability of calculated catalytic ammonia
critical role of water at the gold-titania interface in catalytic CO oxidation. synthesis rates. Science 345, 197–200 (2014).
Science 345, 1599–1602 (2014). 63. Wang, P. et al. Breaking scaling relations to achieve low-temperature
32. Chang, C.-R., Yang, X.-F., Long, B. & Li, J. A water-promoted mechanism of ammonia synthesis through LiH-mediated nitrogen transfer and
alcohol oxidation on a Au(111) surface: understanding the catalytic behavior hydrogenation. Nat. Chem. 9, 64–70 (2017).
of bulk gold. ACS Catal. 3, 1693–1699 (2013). 64. Wang, P. et al. The formation of surface lithium-iron ternary hydride and its
33. Fedoseev, G., Cuppen, H. M., Ioppolo, S., Lamberts, T. & Linnartz, H. function on catalytic ammonia synthesis at low temperatures. Angew. Chem.
Experimental evidence for glycolaldehyde and ethylene glycol formation by Int. Ed. Engl. 56, 8716–8720 (2017).
surface hydrogenation of CO molecules under dense molecular cloud 65. Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation
conditions. Mon. Not. R. Astron. Soc. 448, 1288–1297 (2015). made simple. Phys. Rev. Lett. 77, 3865 (1996).
34. Studt, F., Abild-Pedersen, F., Varley, J. B. & Nørskov, J. K. CO and CO2 66. Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio total-
hydrogenation to methanol calculated using the BEEF-vdW functional. Catal. energy calculations using a plane-wave basis set. Phys. Rev. B: Condens. Matter
Lett. 143, 71–73 (2013). Mater. Phys. 54, 11169 (1996).
35. Behrens, M. et al. The active site of methanol synthesis over Cu/ZnO/Al2O3 67. Henkelman, G., Uberuaga, B. P. & Jónsson, H. A climbing image nudged
industrial catalysts. Science 336, 893–897 (2012). elastic band method for finding saddle points and minimum energy paths. J.
36. Chang, C.-R., Wang, Y.-G. & Li, J. Theoretical investigations of the catalytic Chem. Phys. 113, 9901–9904 (2000).
role of water in propene epoxidation on gold nanoclusters: a hydroperoxyl- 68. te Velde, G. et al. Chemistry with ADF. J. Comput. Chem. 22, 931–967 (2001).
mediated pathway. Nano Res. 4, 131–142 (2011). 69. Lenthe, Ev, Baerends, E. J. & Snijders, J. G. Relativistic regular two‐component
37. Qiao, B. et al. Single-atom catalysis of CO oxidation using Pt1/FeOx. Nat. Hamiltonians. J. Chem. Phys. 99, 4597–4610 (1993).
Chem. 3, 634–641 (2011). 70. Medford, A. J. et al. CatMAP: a software package for descriptor-based
38. Zhang, S. et al. Catalysis on singly dispersed bimetallic sites. Nat. Commun. 6, microkinetic mapping of catalytic trends. Catal. Lett. 145, 794–807 (2015).
7938 (2015).
39. Yang, X.-F. et al. Single-atom catalysts: a new frontier in heterogeneous
catalysis. Acc. Chem. Res. 46, 1740–1748 (2013). Acknowledgements
40. Zhang, B. et al. Homogeneously dispersed multimetal oxygen-evolving This work is supported by National Natural Science Foundation of China (NSFC, grant
catalysts. Science 352, 333–337 (2016). nos. 21590792, 91645203, and 21521091), China Postdoctoral Science Foundation (No.
41. Ma, X.-L., Liu, J.-C., Xiao, H. & Li, J. Surface Cluster Catalyst for N2-to-NH3 2017M610863) and Beijing Natural Science Foundation (2184105). The calculations were
Thermal Conversion. J. Am. Chem. Soc. 140, 46–49 (2018). performed by using supercomputers at Tsinghua National Laboratory for Information
42. van der Ham, C. J., Koper, M. T. & Hetterscheid, D. G. Challenges in Science and Technology, the Supercomputer Center of the Computer Network Infor-
reduction of dinitrogen by proton and electron transfer. Chem. Soc. Rev. 43, mation Center at the Chinese Academy of Sciences, the National Supercomputer Centre
5183–5191 (2014). in Guangzhou (NSCC-GZ, Tianhe II), and the Computational Chemistry Laboratory of
43. Hoffman, B. M., Lukoyanov, D., Yang, Z. Y., Dean, D. R. & Seefeldt, L. C. Department of Chemistry at Tsinghua University, which is supported by Tsinghua
Mechanism of nitrogen fixation by nitrogenase: the next stage. Chem. Rev. Xuetang Talents Program.
114, 4041–4062 (2014).
44. Anderson, J. S., Rittle, J. & Peters, J. C. Catalytic conversion of nitrogen to Author contributions
ammonia by an iron model complex. Nature 501, 84–87 (2013). J.L. conceived and directed the research. J.-C.L. performed all the calculations. J.-C.L., X.
45. Musaev, D. G., Bobadova‐Parvanova, P. & Morokuma, K. in Computational M. and Y.L. designed the study, analyzed the data. J.-C.L., Y.-G.W., H.X. and J.L. co-
Modeling for Homogenous and Enzymatic Catalysis. (eds Morokuma K. & wrote the paper. All the authors discussed the results and commented on the manuscript.
Musaev D. G.) 83–108 (Wiley‐VCH., Weinheim, Germany, 2008).
46. Grubel, K., Brennessel, W. W., Mercado, B. Q. & Holland, P. L. Alkali metal
control over N–N cleavage in iron complexes. J. Am. Chem. Soc. 136, Additional information
16807–16816 (2014). Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467-
47. Kitano, M. et al. Ammonia synthesis using a stable electride as an electron 018-03795-8.
donor and reversible hydrogen store. Nat. Chem. 4, 934–940 (2012).
48. Figg, T. M., Holland, P. L. & Cundari, T. R. Cooperativity between low-valent Competing interests: The authors declare no competing interests.
iron and potassium promoters in dinitrogen fixation. Inorg. Chem. 51,
7546–7550 (2012). Reprints and permission information is available online at http://npg.nature.com/
49. Rittle, J. & Peters, J. C. An Fe-N2 complex that generates hydrazine and reprintsandpermissions/
ammonia via Fe·NNH2: demonstrating a hybrid distal-to-alternating pathway
for N2 reduction. J. Am. Chem. Soc. 138, 4243–4248 (2016). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in
50. Creutz, S. E. & Peters, J. C. Catalytic reduction of N2 to NH3 by an Fe-N2 published maps and institutional affiliations.
complex featuring a C-atom anchor. J. Am. Chem. Soc. 136, 1105–1115 (2014).
51. Anderson, J. S. et al. Characterization of an Fe identical with N-NH2 Open Access This article is licensed under a Creative Commons
Intermediate Relevant to Catalytic N2 Reduction to NH3. J. Am. Chem. Soc. Attribution 4.0 International License, which permits use, sharing,
137, 7803–7809 (2015).
adaptation, distribution and reproduction in any medium or format, as long as you give
52. Wucherer, E. J. & Vahrenkamp, H. Azoalkane‐nitrene cleavage on an Fe3‐
appropriate credit to the original author(s) and the source, provide a link to the Creative
cluster. Angew. Chem. Int. Ed. 26, 355–356 (1987).
Commons license, and indicate if changes were made. The images or other third party
53. Rayment, T., Schlögl, R., Thomas, J. & Ertl, G. Structure of the ammonia
material in this article are included in the article’s Creative Commons license, unless
synthesis catalyst. Nature 315, 311–313 (1985).
indicated otherwise in a credit line to the material. If material is not included in the
54. Ji, S. et al. Confined pyrolysis within metal-organic frameworks to form
article’s Creative Commons license and your intended use is not permitted by statutory
uniform Ru3 clusters for efficient oxidation of alcohols. J. Am. Chem. Soc. 139,
9795–9798 (2017). regulation or exceeds the permitted use, you will need to obtain permission directly from
55. Lei, Y. et al. Increased silver activity for direct propylene epoxidation via the copyright holder. To view a copy of this license, visit http://creativecommons.org/
subnanometer size effects. Science 328, 224–228 (2010). licenses/by/4.0/.
56. Li, X. F. et al. Conversion of dinitrogen to ammonia by FeN3-embedded
graphene. J. Am. Chem. Soc. 138, 8706–8709 (2016). © The Author(s) 2018
57. Holland, P. L. Metal-dioxygen and metal-dinitrogen complexes: where are the
electrons? Dalton. Trans. 39, 5415–5425 (2010).

NATURE COMMUNICATIONS | (2018)9:1610 | DOI: 10.1038/s41467-018-03795-8 | www.nature.com/naturecommunications 9