Coupling of Carbon Dioxide with Epoxides Efficiently Catalyzed by Thioether-Triphenolate Bimetallic Iron(III) Complexes: Catalyst Structure–Reactivity Relationship and Mechanistic DFT Study

Article abstract of DOI:10.1002/adsc.201600621

A series of dinuclear iron(III)^Icomplexes supported by thioether-triphenolate ligands have been prepared to attain highly Lewis acidic catalysts. In combination with tetrabutylammonium bromide (TBAB) they are highly active catalysts in the synthesis of cyclic organic carbonates through the coupling of carbon dioxide to epoxides with the highest initial turnover frequencies reported to date for the conversion of propylene oxide to propylene carbonate for iron-based catalysts (5200 h^?1; 120 °C, 2 MPa, 1 h). In particular, these complexes are shown to be highly selective catalysts for the coupling of carbon dioxide to internal oxiranes affording the corresponding cyclic carbonates in good yield and with retention of the initial stereochemical configuration. A density functional theory (DFT) investigation provides a rational for the relative high activity found for these Fe(III) complexes, showing the fundamental role of the hemilabile sulfur atom in the ligand skeleton to promote reactivity. Notably, in spite of the dinuclear nature of the catalyst precursor only one metal center is involved in the catalytic cycle. (Figure presented.).

Full text of DOI:10.1002/adsc.201600621

FULL PAPERS
DOI: 10.1002/adsc.201600621
Coupling of Carbon Dioxide with Epoxides Efficiently Catalyzed
by Thioether-Triphenolate Bimetallic Iron(III) Complexes:
Catalyst Structure–Reactivity Relationship and Mechanistic DFT
Study
Francesco Della Monica,a Sai V. C. Vummaleti,b Antonio Buonerba,a,c
Assunta De Nisi,c,d Magda Monari,c,d Stefano Milione,a,c Alfonso Grassi,a,c
Luigi Cavallo,b,* and Carmine Capacchionea,c,*
a
Dipartimento di Chimica e Biologia “Adolfo Zambelli”, Universitꢀ degli Studi di Salerno, via Giovanni Paolo II, 84084
Fisciano (SA), Italy
E-mail: ccapacchione@unisa.it
KAUST Catalysis Center, King Abdullah University of Science and Technology, Thuwal 23955-6900, Saudi Arabia
b
E-mail: luigi.cavallo@kaust.edu.sa
Interuniversity Consortium Chemical Reactivity and Catalysis, CIRCC, via Celso Ulpiani 27, 70126 Bari, Italy
Dipartimento di Chimica G. Ciamician, Alma Mater Studiorum, Universitꢀ di Bologna, via Selmi 2, 40126 Bologna, Italy
c
d
Received: June 15, 2016; Revised: July 18, 2016; Published online: && &&, 0000
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201600621.
Abstract: A series of dinuclear iron(III)I complexes sponding cyclic carbonates in good yield and with re-
supported by thioether-triphenolate ligands have tention of the initial stereochemical configuration. A
been prepared to attain highly Lewis acidic catalysts. density functional theory (DFT) investigation pro-
In combination with tetrabutylammonium bromide vides a rational for the relative high activity found
(TBAB) they are highly active catalysts in the syn- for these Fe(III) complexes, showing the fundamen-
thesis of cyclic organic carbonates through the cou- tal role of the hemilabile sulfur atom in the ligand
pling of carbon dioxide to epoxides with the highest skeleton to promote reactivity. Notably, in spite of
initial turnover frequencies reported to date for the the dinuclear nature of the catalyst precursor only
conversion of propylene oxide to propylene carbon- one metal center is involved in the catalytic cycle.
ate for iron-based catalysts (5200 hÀ1; 1208C, 2 MPa,
1 h). In particular, these complexes are shown to be Keywords: carbon dioxide fixation; cycloaddition;
highly selective catalysts for the coupling of carbon density functional theory; homogeneous catalysis;
dioxide to internal oxiranes affording the corre- iron
Introduction
useful products under milder reaction conditions. In
this context, one of the most promising synthetic
Over the last two decades an increasing interest in routes is the development of efficient catalytic sys-
the use of carbon dioxide as chemical feedstock has tems able to promote the coupling of CO2 with oxir-
engendered intensive efforts, in both academia and in- anes giving the corresponding cyclic organic carbo-
dustry, in the search for new synthetic pathways to nates (COCs)[2] or polycarbonates.[3] In particular
useful compounds and polymers.[1] The main problem COCs have many applications, such as polar aprotic
slowing the development of a chemistry based on green solvents, electrolytes in lithium-ion batteries,
CO2 as C1 building block lies in the kinetic and ther- and as chemical intermediates in the synthesis of
modynamic stability of this molecule that renders its other small molecules or polymers. As a matter of
conversion into valuable chemicals energetically un- fact, a large repertoire of catalytic systems mainly
favorable. Similarly to other industrially relevant pro- based on transition metal complexes combined with
cesses, the key to overcome the kinetic inertness of a nucleophile (quaternary ammonium salts and phos-
carbon dioxide is the implementation of efficient cata- phonium salts) as co-catalyst has been developed and
lytic systems that allow the conversion of CO2 to shown to be highly active and selective for the cyclo-
Adv. Synth. Catal. 0000, 000, 0 – 0
1
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
addition of CO2 to epoxides.[2] It is worth noting that ic co-catalyst. Here we report the synthesis and com-
the most studied and active complexes are based on plete characterization of three novel iron(III) com-
metals such as chromium, cobalt, zinc, magnesium or plexes 4–6 bearing dithioether-triphenolate ligands
niobium.[2d] These are metals that have given rise to and their use as catalysts for the coupling of CO2 with
health and environmental concerns, and have also epoxides and compare their performances with the
been recently defined as endangered elements, i.e., known iron(III) complexes 1–3 under the same reac-
metals that will be not available anymore at afforda- tion conditions. In combination with TBAB and
ble prices in the near future.[4] This situation has trig- under optimized reaction conditions, they show high
gered the search for new, greener and inexpensive activity in the coupling of various terminal epoxides
catalytic systems able to efficiently promote this reac- with CO2. In addition, they also convert various inter-
tion.[2e]
nal epoxides affording the corresponding carbonates
The two most promising approaches to achieve this in good yields and with high stereoselectivity. Density
goal are: (i) the development of metal-free organoca- functional theory calculations (DFT) reveal that in
talytic systems[5] and (ii) the use of complexes based spite of the bimetallic nature of the catalyst precursor
on Earth-crust abundant metals (Fe, Al, Ti, Ca). In only one metal center is involved in the catalytic
spite of the growing number of organocatalysts, the cycle.
presence of a Lewis acidic metal center that activates
the opening of the oxirane ring still offers advantages
in terms of high activity, low catalyst loading and Results and Discussion
even milder reaction conditions. In this field alumi-
num-based catalysts have been extensively investigat- Synthesis and Characterization of the Iron(III)
ed.[6] In particular, the bimetallic aluminum (salen) Complexes 4–6
complex developed by North[6e,7] and co-workers and
the aluminum amino-tris(phenolate) complex report- The Lewis acidity of the metal center, according to
ed by Kleij[8] and co-workers have been shown to be the widely accepted reaction mechanism,[2d] should
highly active and versatile catalysts for the formation play a crucial role in the determination of the catalyt-
of cyclic carbonates. Conversely, notwithstanding the ic activity for the cycloaddition of carbon dioxide to
growing interest in the use of iron in catalysis,[9] the oxiranes. Indeed, with the aim to increase the Lewis
examples of iron-based catalysts for the formation of acidity of the iron atoms, we decided to modify the
cyclic carbonates from CO2 and epoxides are rather pro-ligand structure by introducing an electron-with-
scarce.[10] We have recently reported on new catalysts drawing substituent, such as chlorine, on the lateral
based on dinuclear Fe(III) complexes (complexes 1– (L4) and central (L5) phenol rings, in order to evalu-
3; Scheme 1), bearing dithioether-triphenolate-based ate the impact on the catalytic activity in the forma-
ligands in combination with tetrabutylammonium bro- tion of the cyclic carbonates. The methyl-substituted
mide (TBAB), which have proved to be highly active pro-ligand (L6) was also synthesized to have an alkyl-
and versatile in the synthesis of COCs.[11] These re- substituted complex (6) with a steric hindrance com-
sults have shown that it is possible to obtain efficient parable with that of the chlorine substituted one (4).
catalysts for the cycloaddition of CO2 to epoxides The pro-ligands L4–L6, were synthesized by a modifi-
based on iron complexes, and that the presence of the cation of the procedure previously reported
hemilabile Fe S bond in the catalyst precursor has (Scheme 1).[11] All the new pro-ligands were fully
À
a beneficial role in the subsequent epoxide coordina- characterized by nuclear magnetic resonance (1H, 13C
1
tion to the metal center, promoting the formation of and H-13C HSQC NMR), mass spectrometry (ESI-
the cyclic carbonate in the presence of the nucleophil- MS), elemental analysis (EA) and infrared spectros-
Scheme 1. Synthesis of the proligands L1–L6 and the corresponding iron(III) complexes 1--6.
Adv. Synth. Catal. 0000, 000, 0 – 0
2
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
copy (FT-IR) (see the Supporting Information). Com-
plexes 4-6, were obtained by reaction of an equimolar
amount of FeCl3 with the sodium salt of the corre-
sponding pro-ligand in THF, obtained from the reac-
tion of the ligand precursor with 3 equivalents of
sodium hydride. All the products were recovered in
good yields (from 78 to 85%), as deep-blue powders.
Complexes 4–6 were characterized by elemental anal-
ysis. Furthermore, by comparison of the FT-IR spec-
tra of complexes 4–6 with those of the corresponding
ligands, disappearance of the hydroxy group vibration
around 3300 cmÀ1 confirms the formation of the de-
sired complexes. Moreover, the shift of the vibration
bands diagnostic of the alkyl-sulfide moiety (see the
Supporting Information), indicates the coordination
of the sulfur atoms to the iron centers. Indeed the
UV-Vis spectra of the complexes only show a strong
ligand-to-metal charge transfer absorption and no d–d
transition as expected for two hexacoordinated HS
iron(III) centers. The ESI-MS analysis was consistent
with the formation of dinuclear species (Supporting
Information, Figures S14–S16–S18). In addition, the
values of the effective magnetic moments determined
using the Evans method[12] in toluene-d8 at 308C, are
comparable with the calculated value for two isolated
high spin (HS) iron(III) centers (8.37 mB).[13]
Notably, the magnetic moments measured in the
solid state with a Johnson–Matthey balance by using
the Faradayꢂs method[14] are in good agreement with
values obtained in solution confirming that the com-
Figure 1. Molecular structure of complex 6. a) The structure
is shown for the sake of clarity with the C atoms of one of
the two ligands represented in violet color (hydrogen atoms
À
have been omitted). Selected bond lengths (ꢃ): Fe(1) Fe(2)
À
À
plexes possess a dimeric structure both in the solid 3.331; Fe(1) O(1) 1.8894(14); Fe(1) O(4) 1.9030(13);
À
À
À
Fe(1) O(5) 2.0314(13); Fe(1) O(2) 2.0384(14); Fe(1) S(1)
state and in solution (see the Supporting Information,
Table S2). Finally addition of a large excess of
a strong coordinating species mimicking the epoxide
substrate, such as pyridine, to a toluene solution of 6
(50 equivalents with respect to 6) does not produce
any change of the UV-Vis spectrum profile, underly-
ing the stability of the dimeric structure in solution
(see the Supporting Information, Figures S21–23).
À
À
2.6102(5); Fe(1) S(3) 2.6466(5); Fe(2) O(6) 1.8943(14);
Fe(2) O(3) 1.8994(15); Fe(2)VO(5) 2.0345(13); Fe(2) O(2)
2.0514(14); Fe(2) S(2) 2.5914(5); Fe(2) S(4) 2.6585(5). b)
The coordination geometry of the two iron(III) centers is
highlighted.
À
À
À
À
Single crystals of 6 were grown from a saturated ace- Cycloaddition of CO2 to Epoxides Promoted by
tonitrile solution; the molecular structure of the com- Catalysts 1–6
plex is shown in Figure 1. It is worth noting that each
ligand is coordinated to both iron atoms with the cen- Aiming to optimize the reaction parameters in the cy-
tral phenolate moiety bridging the metal centers pre- cloaddition of CO2 to (Æ)-propylene oxide (PO) we
sumably to relieve steric strain of the ligand. The mol- used the complex 6 as benchmark under solvent-free
ecule adopts an idealized D2 symmetry one axis being conditions.
A screening of various co-catalysts,
coincident with the Fe–Fe vector. Each iron(III) in namely dimethylaminopyridine (DMAP), bis(triphe-
the dimeric complex exhibits a six-coordinated [O4S2] nylphosphine)iminium chloride (PPNCl) or quaterna-
environment generated by four phenoxo oxygens and ry ammonium salts such as tetrabutylammonium chlo-
two sulfur atoms. The Fe–O(bridging) bond lengths in ride (TBAC), iodide (TBAI) and bromide (TBAB) in
6 are longer than the Fe–O(non-bridging) ones. These combination with 0,5 equivalent of 6 (0.025 mol%)
Fe–O distances are slightly longer, for example, than was performed (see Figure 2a and entries S1–S5 of
the corresponding ones found in an iron(III) thioca- Table S1 in the Supporting Information).
lix[4]arene complex.[15] The Fe–S interactions are
Notably, by varying the amount of TBAB from
rather weak and fall in the range 2.5914–2.6585(5) ꢃ 0.05 mol% to 0.25 mol% an appreciable improvement
whereas the Fe–Fe distance of 3.331 ꢃ indicates that of the catalytic activity is observed (entries S5–S7,
no bonding interaction is present.
Table S1 in the Supporting Information). Increasing
Adv. Synth. Catal. 0000, 000, 0 – 0
3
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Table 1. CO2/propylene oxide coupling promoted by the
iron(III) complexes 1–6.
Entry[a]
Catalyst
Conversion[b,c] [mol%]
TOF[d] [hÀ1]
1
2
3
4
1
2
3
4
5
6
–
43.0
35.5
40.4
36.3
50.0
52.0
5
4300
3550
4040
3630
4990
5200
–
5
6
7[e]
[a]
Reaction conditions: PO=5 mL, 7.15ꢄ10À2 mol; cata-
lyst=1.43ꢄ10À5 mol (0.01 mol%); TBAB=1.43ꢄ10À4
mol (0.1 mol%); PO/TBAB/catalyst=10000/10/1; T=
1208C; PCO =2.0 MPa; reaction time=1 h.
2
Determined by 1H NMR using mesitylene as internal
standard.
[b]
[c]
The selectivity toward the formation of PC was always
found to be >99%.
Turnover frequency (molPC ꢄmolcatalystÀ1 ꢄreaction timeÀ1).
Control experiment in the absence of the iron catalyst.
[d]
[e]
equal to 5200 hÀ1 at 1208C and 2.0 MPa of CO2
(entry 6, Table 1). To the best of our knowledge, this
is the highest TOF for an iron-based catalyst reported
in literature and not far from those observed for the
most active catalysts based on magnesium or alumi-
num.[2d] Complex 5 also proved to be highly active,
with conversion near to that of 6 (compare entries 5
and 6, Table 1), suggesting that the electronic effect
of the substituent R2 does not play a relevant role in
Figure 2. Effect of the co-catalyst type and loading (a) and
of temperature (b) on the catalytic activity of 6 in PO/CO2
coupling.
the temperature from 80 to 1408C results in a linear the determination of the catalytic performance. On
increasing of the TOF (Figure 2b; entries S5 and S8– the other hand, the introduction of chlorine atoms in
S10 of Table S1 in the Supporting Information). A the R1 positions, as in complex 4, has a deleterious
control experiment using TBAB without the iron effect on the catalytic activity (compare entries 4 and
complex results in very low conversion (10%; entry 6, Table 1). In addition, the appreciable lower activity
S11, Table S1 in the Supporting Information), com- displayed by complex 2, bearing the more sterically
pared to the binary system under the same reaction demanding cumyl groups on the phenol rings, con-
conditions (80%; entry S7, Table S1 in the Supporting firms that the catalytic activity is strongly influenced
Information). The effect of the CO2 pressure on the by the bulkiness of the substituents on the external
catalytic activity was also investigated (compare en- phenol rings.
tries S7 and S12–S13, Table S1 in the Supporting In-
To further evaluate the role of the steric demand
formation); by reduction of the pressure from 2.0 to around the catalytic site on the catalytic performance,
0.5 MPa the activity does not drop dramatically, while the coupling of CO2 with a bulkier terminal epoxide
on increasing the pressure to 4.0 MPa an activity en- such as the 1,2-epoxyhexane was performed (Table 2).
hancement from 3300 to 4000 hÀ1, reaching complete
conversion of PO in only 1 hour, was observed.
Both 5 and 6 exhibit again the highest activities (re-
spectively, 3390 and 3450 hÀ1), with differences in the
Owing to these good results the catalyst loading order of the experimental error (entries 12 and 13,
was reduced from 0.025 to 0.01 mol%, in order to Table 2). Notably, the longer alkyl chain on the epox-
compare the catalytic activity of complexes 1–6 in the ide ring exalts differences in the activity due to the
coupling of CO2 with PO (Table 1). The best per- steric hindrance of the substituents in the R1 posi-
formance was achieved using complex 6, with a TOF tions, with complex 2 displaying a considerably lower
Adv. Synth. Catal. 0000, 000, 0 – 0
4
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Table 2. CO2/hexene oxide coupling promoted by iron(III)
complexes 1–6.
Entry[a]
Catalyst
Conversion[b,c] [mol%]
TOF[d] [hÀ1]
8
9
10
11
12
13
1
2
3
4
5
6
45.7
17.9
43.1
35.0
67.8
68.9
2290
900
2160
1750
3390
3450
[a]
Reaction conditions: hexene oxide=8.6 mL, 7.15ꢄ10À2
mol; complex=1.43ꢄ10À5 mol; TBAB=1.43ꢄ10À4 mol;
T=1208C; PCO =4.0 MPa; reaction time=1 h.
2
Determined by 1H NMR using mesitylene as internal
standard.
[b]
[c]
[d]
The selectivity toward the formation of hexene carbonate
(HC) was always found to be >99%.
Turnover
frequency
(molHC ꢄmolcatalystÀ1 ꢄreaction
timeÀ1).
Scheme 2. Proposed mechanism for the retention of stereo-
chemical configuration in the production of cis-4,5-dimethyl-
1,3-dioxolan-2-one from cis-2,3-epoxybutane and CO2 pro-
moted by catalyst 6 in the presence of TBAB.
activity (entry 9, Table 2). The less sterically demand-
ing complex 3 shows lower activity compared to 6 due
to solubility issues in the reaction medium as previ-
ously reported.[11b] In order to investigate the applica-
bility of the catalytic system 6/TBAB to a wider range Scheme 2). A complete stereoretention was also ob-
of substrates, we performed the coupling reaction of served in the case of trans-stilbene oxide (entry 22,
carbon dioxide with a series of mono-, di-substituted Table 3) and the exclusive formation of cis-cyclohex-
and internal epoxides. The results summarized in the ene carbonate (cis-CHC) (entry 18, Table 3) also sup-
Table 3 clearly show that the presence of an electron- ports the proposed reaction mechanism. In fact, the
withdrawing substituent in the substrate, such as thermodynamically disfavored formation of cis-CHC
a chlorine atom in epichlorohydrin, has a positive should necessarily arise from a mechanism involving
effect with a TOF of 7000 hÀ1 (entry 14, Table 3).
a double inversion of the starting CHO configura-
Conversely, the presence of an electron-releasing tion.[2d]
group, such as a phenoxy or a phenyl group results in
an attenuation of the reactivity (entries 15–17,
Table 3). Notably the more challenging substrate iso- Theoretical Calculations
butylene oxide (1,1-dimethyloxirane) was converted
in good yield to the corresponding carbonate Deeper insights into the reaction mechanism were ob-
(entry 19 of Table 3). Moreover the cis form of 1,2-di- tained by performing DFT on the reaction mechanism
methyloxirane was converted to the corresponding for the cycloaddition of CO2 to PO catalyzed by com-
carbonate not only in good yield but also with a high plex 6 in the presence of TBAB, the complete energy
degree of configuration retention (entry 20, Table 3).
profile is reported in Figure 3 (for clarity, we refer to
Intriguingly, complex 6 in combination with TBAB the two Fe(III) centers in 6 as Fe1 and Fe2). Coordina-
retains the configuration of the starting substrate re- tion of PO to the Fe1 center in 6 is only possible by
gardless of the cocatalyst/catalyst molar ratio (com- dissociation one of the two hemilabile S atoms [r(Fe1–
pare entries 20 and 21, Table 3) indicating that the S)=2.704 ꢃ].
attack of the bromide anion to the epoxide ring pro-
PO coordination is calculated to be endergonic and
ceeds selectively via two consecutive inversions of leads to intermediate 6-A, which lies 7.5 kcalmolÀ1
configuration (SN2) on the same stereocenter and above 6. From a structural point of view, in 6-A PO
that, differently to what was reported for other coordinates to the Fe1 center [r(Fe1–O)=2.412 ꢃ] in
iron(III)-based catalysts,[10f] this mechanism is not de- the coordination position previously occupied by the
pendent on the bromide concentration (see dissociated S atom. PO coordination results in a slight
Adv. Synth. Catal. 0000, 000, 0 – 0
5
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Table 3. CO2/epoxide coupling promoted by iron(III) complexes 6.
Entry[a] Substrate
Epoxide/TBAB/6 (molar Product
ratio)
Conversion[b,c]
[mol%]
cis:trans[d] (molar
ratio)
TOF[e]
[hÀ1]
14
15
16
10000/10/1
10000/10/1
10000/10/1
70.0
44.5
21.0
–
–
–
7000
4450
2100
17
10000/10/1
1000/10/1
1000/10/1
13.0
55.0
65.8
–
1300
550
18[f]
19[f,g]
>99:1
–
660
20[f,g]
1000/10/1
1000/2/1
67.1
12.6
97:3
98:2
670
125
21[f,g]
22[g,h]
1000/10/1
7.3
>1:99
70
[a]
Reaction conditions: epoxide=7.15ꢄ10À2 mol; T=1208C; PCO =2.0 MPa; reaction time=1 h.
Determined by H NMR using mesitylene as internal standard.
2
[b]
[c]
[d]
[e]
[f]
1
The selectivity toward the formation of cyclic carbonate was always found >99%.
1
Determined by H NMR.
Turnover frequency (molcarbonate ꢄmolcatalystÀ1 ꢄreaction timeÀ1).
Epoxide=1.0 mL.
[g]
[h]
PCO =4.0 MPa.
2
Stilbene oxide=1.0 g, MEK=2.0 mL.
decrease in the Fe1–Fe2 distance in 6-A when com- nario. As a matter of fact, the coordinating ability of
pared to the catalyst species 6 (3.271 vs. 3.364 ꢃ, re- the phenoxide O atoms should be reduced in 4 due to
spectively; see Figure 4).
the presence of the ortho Cl atom. This should result
To explore the eventual dissociation of one of the in increased Lewis acidity at the metal center and
À
two Fe1 S bonds prior to PO coordination in the should reduce the lability of the sulfur atoms, render-
starting complex 6, we removed PO from the opti- ing substitution of one sulfur moiety by PO less favor-
mized geometry of intermediate 6-A and we re-opti- able. This scenario is supported by our calculations on
À
mized the complex. The resulting complex with Fe1
S
complex 4, which presents slightly shorter Fe1–S dis-
bond being dissociated is lying only 5.8 kcalmolÀ1 tances when compared to complex 6 (2.66 ꢃ vs.
above species 6, in line with the expected lability of 2.70 ꢃ, see Figure S1 in the Supporting Information).
À
the S binding. The experimental evidence that the in- Consistently, dissociation of one Fe1 S bond from
troduction of the electron-withdrawing chlorine atoms complex 4 requires 8.6 kcalmolÀ1 versus 5.8 kcalmolÀ1
on the lateral phenol rings (complex 4) results in in 6, and the PO coordinated intermediate 4-A, is
a lowering of the catalytic activity supports this sce- slightly less stable than 6-A (8.7 vs. 7.5 kcalmolÀ1).
Adv. Synth. Catal. 0000, 000, 0 – 0
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Figure 3. Computed free energy surface for the cycloaddition of propylene oxide (PO) and CO2 catalyzed by species 6/
TBAB. The free energies in solution (PO as the solvent) are given in kcal·molÀ1 relative to the starting complex 6. The
energy values in red represent the energies of the transition states. * indicates decoordination of a hemilabile S atom.
Back to the main mechanism catalyzed by 6, the this case we observed a clear interaction between the
next step from the PO coordinated specie 6-A corre- Fe2 center and one of the O atoms of the CO2 mole-
sponds to the opening of the PO ring by a Br anion, cule [r(Fe2–CO2)=2.320 ꢃ], see Figure 5b. However,
leading to the formation of the more stable intermedi- transition state 6-TS-BC’ is placed 15.7 kcalmolÀ1
ate 6-B, 5.5 kcalmolÀ1 below 6. This step proceeds higher in energy relative to transition state 6-TS-BC.
through transition state 6-TS-AB and requires over- These observations suggest that though the catalyst
coming an overall barrier of 18.1 kcalmolÀ1 above 6. species 6 under investigation is bimetallic, catalysis es-
À
Then, we studied insertion of CO2 into the Fe O sentially occurs at a single Fe(III) center without co-
bond in the intermediate 6-B to give the hemicarbon- operation of the second Fe(III) center. This is clearly
ate intermediate 6-C, which lies 10.0 kcalmolÀ1 above different from the case of PO coupling with CO2 to
6-B. This CO2 insertion step requires the overcoming afford the cyclic carbonate promoted by monometal-
of a barrier (6-TS-BC) of 21.9 kcalmolÀ1 from the lic Nb complexes, where cooperativity between two
most stable intermediate 6-B. To our surprise, despite Nb centers occurs both in the homogeneous phase as
the relatively low calculated barrier we did not ob- well as after grafting of the Nb complex on silica.[16]
serve any cooperativity between the two Fe centers, Next, starting from intermediate 6-C, we studied the
since the CO2 molecule is quite a way away from the ring closing step leading to the formation of inter-
Fe2 center [r(Fe2–CO2)=3.772 ꢃ], see Figure 5a.
mediate 6-D with the desired cyclic carbonate product
As a further check on the possible cooperativity be- coordinated to the Fe metal center. According to cal-
tween the two Fe centers in the CO2 insertion step, culations, ring closing is predicted to be the rate-de-
we calculated an alternative transition state for CO2 termining step, with a barrier of 26.7 kcalmolÀ1 (6-TS-
insertion, 6-TS-BC’, by forcing decoordination of one CD) relative to the most stable intermediate 6-B. This
of the two hemilabile S atoms from the Fe2 center. In is clearly a high barrier, which is however consistent
Adv. Synth. Catal. 0000, 000, 0 – 0
7
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Figure 5. Geometry of transition states 6-TS-BC (a) 6-TS-
À
BC’ (b), for insertion of CO2 into the Fe O bond. Hydrogen
atoms omitted for clarity and the selected distances in ꢃ.
Conclusions
We have reported the synthesis and the complete
characterization of three novel bimetallic iron(III)
complexes (4–6) bearing thioether-triphenolate li-
gands. When activated by tetrabutylammonium bro-
mide the title iron(III) complexes prove to be very ef-
fective in the cycloaddition of CO2 to epoxides giving
the highest initial activity so far reported for an iron-
Figure 4. Geometry of the starting complex 6 (a) and of the
PO coordinated intermediate 6-A (b). Hydrogen atoms
omitted for clarity and the selected distances in ꢃ.
with the remarkably high temperature of 1208C based catalyst. In particular, complex 6 shows the best
needed experimentally to achieve a yield of at least catalytic performance in terms of activity and selectiv-
51% in 1 hour. Finally, intermediate 6-D would re- ity for several terminal epoxides with TOF values up
lease the carbonate product and a BrÀ anion, regener- to 5200 hÀ1 for the conversion of PO to PC. Further-
ating intermediate 6-A with recoordination of a free more, complex 6 also promotes the conversion to the
PO molecule, thus closing the catalytic cycle.
corresponding cyclic carbonates of internal epoxides
Adv. Synth. Catal. 0000, 000, 0 – 0
8
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
1 magnetic susceptibility balance by means of the Faraday
method based on the following equation: fs/fr =(cs·ms)/
(cr·mr) where fs, cs and ms are the magnetic force, the specif-
ic susceptibility and the mass of the sample, while fr, cr and
mr are the magnetic force, the specific susceptibility and the
mass of CuSO4·5H2O used as reference compound with
a value of cr =6.00·10À6 cm3·gÀ1.[14] Magnetic moments were
calculated using the following equation: meff =2.828.
(T·cm)1/2, where meff is the effective magnetic moment, T is
the absolute temperature and cm is the molar susceptibility.
Effective magnetic moments values for all complexes in
both solution and solid state are listed in Table S2 of the
Supporting Information. Elemental analysis was performed
on a CHNS Thermo Scientific Flash EA 1112 equipped with
a thermal conductivity detector. ESI-MS were acquired on
a Quattro microTM API triple quadrupole mass spectrometer
from Waters equipped with electrospray ion source, using
anhydrous acetonitrile as solvent. FT-IR measurements
were carried out on a Bruker Vertex 70 spectrometer
equipped with DTGS detector and a Ge/KBr beam splitter.
The samples were analyzed in chloroform solutions or in the
solid state as KBr disks. UV-Vis spectra were collected on
a Perkin–Elmer Lambda EZ 201 spectrophotometer.
with a high degree of stereocontrol in the ring-closing
reaction, leading to the retention of configuration of
the starting epoxide (ꢀ97%) via a mechanism involv-
ing a double inversion of the initial epoxide configu-
ration. Finally the mechanistic studies show that in
spite of the dinuclear nature of the catalyst precursor
only one metal center is operative in the catalytic
cycle and that coordination of the substrate and the
subsequent reaction pathway can only take place by
dissociation of a hemilabile S atom from one of the
iron centers. These findings not only show that is pos-
sible to obtain highly active catalytic systems based
on iron(III) for the CO2/epoxides coupling but also
highlight the fundamental role of these sulfur-contain-
ing ligands in modulating the Lewis acidity of the iron
center and thus the reactivity with CO2 and oxiranes
for this family of catalysts.
Experimental Section
All manipulations involving air- and/or moisture-sensitive
compounds were performed under a nitrogen atmosphere
using standard Schlenk technique and a MBraun glovebox.
Toluene (99.5%; Sigma–Aldrich) and THF (99%; Sigma–Al-
drich) were used as received or refluxed for 48 h over
sodium or sodium ketyls and distilled before use for mois-
ture- and oxygen-sensitive reactions. All other reagents
were used as received (TCI or Sigma–Aldrich) or distilled
under reduced pressure over calcium hydride. The ligand
precursors: 2,6-[(3,5-di-tert-butyl-2-hydroxyphenyl)thio]-4-
tert-butylphenol (L1), 2,6-[(3,5-bis-a,a’-dimethylbenzyl-2-hy-
droxyphenyl)thio]-4-tert-butylphenol (L2), 2,6-[(2-hydroxy-
phenyl)thio]-4-tert-butylphenol (L3) and the corresponding
iron(III) complexes (1, 2, 3, Scheme 2) were synthesized ac-
cording to the reported procedures.[11] Deuterated solvents
were purchased from Euriso–Top or Sigma–Aldrich and
used as received. NMR spectra were collected on Bruker
Avance spectrometers (600, 400, 300 or 250 MHz for 1H):
the chemical shifts were referenced to tetramethylsilane
(TMS) as external reference using the residual protio signal
of the deuterated solvents. Measurements of effective mag-
Synthesis of the Pro-ligand 2,6-[(3,5-Dichloro-2-
hydroxyphenyl)thio]-4-tert-butylphenol (L4, Scheme
2)
The pro-ligand L4 was synthesized using a modification of
the previously reported procedure for L1.[11] A 100-mL two-
neck round-bottom flask equipped with condenser and mag-
netic stirring bar was charged with 0.76 g of 2-mercapto-4,6-
dichlorophenol (3.9 mmol) dissolved in 4 mL of ethanol,
1.4 g of Cs2CO3 (4.3 mmol) and the mixture was refluxed for
2 h until complete dissolution of Cs2CO3. 0.66 g of 2,6-dibro-
momethyl-4-tert-butylphenol (1.96 mmol) dissolved in 2 mL
of ethanol were slowly added at 08C and the mixture was
refluxed overnight. The solvent was distilled off, water was
added until dissolution of the residue and the aqueous
phase extracted twice with CH2Cl2. The combined organic
phases were dried with MgSO4 and, after evaporation of the
solvent, the product was purified by column chromatogra-
phy (eluent: CH2Cl2) and recovered as a white solid; yield:
0.94 g (85%). 1H NMR (600 MHz, CD2Cl2, 258C): d=1.12
(9H, s), 4.04 (4H, s), 6.80 (2H, s, Ar-H), 7.19 (2H, d, Ar-
H), 7.32 (2H, d, Ar-H); 13C NMR (600 MHz, CD2Cl2,
258C): d=31.36, 34.21, 36.55, 121.08, 121.98, 123.05, 124.99,
127.99, 130.83, 134.07, 143.75, 150.61, 152.14; elemental
analysis: calcd. for C24H22Cl4O3S2: C 51.08, H 3.93, S 11.36;
found: C 51.02, H 3.82. S 11.25; MS: m/z=586.7 (M+Na+).
netic moments were performed on
a Bruker Avance
600 MHz spectrometer at 258C in toluene-d8 using a 5 mm
Wilmad coaxial insert NMR tube.[12] Solutions of the com-
plex 4 (2.22 mM), 5 (2.29 mM) and 6 (2.16 mM) in toluene-
d8 with 1% (v/v) of TMS were prepared under a nitrogen at-
mosphere. The effective magnetic moment (meff) was calcu-
lated from meff =8cgMwT, where cg (cm3 gÀ1) is the corrected
molar susceptibility derived from cg =3Df/4pfoCMw +co. Df is
the shift in frequency (Hz) of the residual protio signal of
the solvent in the presence of the complex from the value of
the pure solvent, C and Mw are, respectively, the concentra-
tion (mol cmÀ3) and the molecular weight of the complex
(gmolÀ1), fo is the operating frequency of the spectrometer
(Hz), and co is the mass susceptibility of the pure solvent
(À0.6179ꢄ10À6 cm3·gÀ1 for toluene-d8). 4p/3 is the shape
factor for a cylindrical sample in a superconducting magnet.
The magnetic susceptibilities of all the complexes 1–6 in the
solid state were measured using a Sherwood Scientific MK
Synthesis of the Pro-ligand 2,6-[(3,5-Dimethyl-2-
hydroxyphenyl)thio]-4-chlorophenol (L5, Scheme 2)
The pro-ligand L5 was synthesized using a modification of
the previously reported procedure for L1.[11] A 100-mL two-
neck round-bottom flask equipped with condenser and mag-
netic stirring bar was charged with 0.80 g of 2-mercapto-4,6-
dimethylphenol (5.2 mmol) dissolved in 13 mL of DMF,
0.90 g of K2CO3 (6.5 mmol) and the resulting mixture was
refluxed for 2 h. 0.82 g of 2,6-dibromomethyl-4-chlorolphe-
nol (2.6 mmol) dissolved in 7 mL of DMF were slowly
Adv. Synth. Catal. 0000, 000, 0 – 0
9
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
added at 08C and was refluxed overnight. The solvent was
distilled off, water was added and the aqueous phase ex-
tracted twice with CH2Cl2. The combined organic phases
were dried with MgSO4 and, the product was recovered
after evaporation of the solvent as a white solid; yield:
0.97 g (81%). 1H NMR (400 MHz, CD2Cl2, 258C): d=2.17
(6H, s), 2.23 (6H, s), 3.86 (4H, s), 6.11 (1H, s broad, -OH),
6.54 (2H, s broad, -OH), 6.76 (2H, s, Ar-H), 6.92 (2H, s,
Ar-H), 6.95 (2H,s, Ar-H); 13C NMR (400 MHz, CD2Cl2,
258C): d=16.48, 20.36, 36.52, 116.53, 124.38, 125.27, 126.23,
129.69, 129.89, 133.83, 134.03, 151.25, 153.23; elemental
analysis calcd. for C24H25 Cl2O3S2: C 62.52, H 5.47, S 13.91;
found: C 62.49, H 5.42, S 13.87; MS: m/z=483.6 (M+Na+).
981.2 (M+Na+); 1354.8 [M(CH3CN)2K+]; UV-Vis: e585
=
5441 L·molÀ1·cmÀ1.
Synthesis of the Iron(III) Complex 5 (Scheme 2)
The complex 5 was synthesized using the same procedure
previously reported for complex 1.[11] Pro-ligand L5 (0.94 g;
2.04 mmol) was dissolved in THF (70 mL). The solution was
added to
a
suspension of sodium hydride (0.17 g;
7.08 mmol) in THF (30 mL) and the mixture stirred at room
temperature overnight. The resulting suspension was filtered
through celite and slowly added at room temperature to
0.325 g of anhydrous iron(III) chloride (2.0 mmol) dissolved
in 30 mL of THF. The rapid change of the color to the blue
was observed and the reaction kept overnight. The mixture
was then filtered through celite and the solvent removed
under reduced pressure affording a deep purple crystalline
solid; yield: 0.83 g (80%); elemental analysis calcd. for
C48H44Fe2Cl2O6S4: C 56.10, H 4.32, S 12.48; found: C 56.02,
Synthesis of the Pro-ligand 2,6-[(3,5-Dimethyl-2-
hydroxyphenyl)thio]-4-tert-butylphenol (L6, Scheme
2)
H 4.28, S 12.39; MS: m/z=981.2 (M+Na+); UV-Vis: e596
=
The pro-ligand L6 was synthesized using the previously re-
ported procedure for L1.[11] A 100-mL two-neck round-
bottom flask equipped with condenser and magnetic stirring
bar was charged with 3.14 g of 2-mercapto-4,6-dimethylphe-
nol (20.4 mmol) dissolved in 20 mL of ethanol, 0.90 g of
NaOH (22.4 mmol) and the mixture was refluxed about
1 hour until complete dissolution of the hydroxide. 3.42 g of
2,6-dibromomethyl-4-tert-butylphenol (10.2 mmol) dissolved
in 15 mL of ethanol were slowly added at 08C and the mix-
ture heated to the reflux of the solvent that was kept over-
night. The solvent was distilled off, water was added until
dissolution of NaBr by-product and the aqueous phase ex-
tracted twice with CH2Cl2. The combined organic phases
were dried with MgSO4, after evaporation of the solvent the
product was purified by column chromatography (petroleum
ether : ethyl acetate=95:5) and recovered as a white solid;
yield: 2.82 g (57.3%). 1H NMR (400 MHz, CD2Cl2, 258C):
d=1.05 (9H, s), 2.14 (6H, s), 2.17 (6H, s), 3.91 (4H, s), 5.79
(1H, s, -OH), 6.60 (2H, s, -OH), 6.69 (2H, s, Ar-H), 6.91
(4H, s, Ar-H); 13C NMR (400 MHz, CD2Cl2, 258C): d=
16.47, 20.34, 31.27, 34.08, 37.17, 117.18, 124.04, 124.20,
127.55, 129.60, 133.75, 134.12, 143.46, 150.09, 153.56; ele-
mental analysis calcd. for C28H34O3S2: C 69.67, H 7.10, S
13.29; found: C 69.55, H 7.09, S 13.18; MS: m/z=505.0
(M+Na+).
4757 L·molÀ1·cmÀ1.
Synthesis of the Iron(III) Complex 6 (Scheme 2)
The complex 6 was synthesized using the same procedure
previously reported for complex 1.[11] Pro-ligand L6 (1.62 g;
3.56 mmol) was dissolved in THF (100 mL). The solution
was added to a suspension of sodium hydride (0.30 g;
12.5 mmol) in THF (70 mL) and the mixture stirred at room
temperature overnight. The resulting suspension was filtered
through celite and slowly added at room temperature to
0.567 g of anhydrous iron(III) chloride (3.50 mmol) dis-
solved in 100 mL of THF. The rapid change of the color to
the deep purple was observed and the reaction kept over-
night. The mixture was then filtered through celite and the
solvent removed under reduced pressure affording a deep
purple crystalline solid; yield: 1.60 g (85%); elemental anal-
ysis calcd. for C56H62Fe2O6S4: C 62.30, H 5.83, S 11.98;
found: C 62.20, H 5.78, S 11.89; MS: m/z=1072.1 (M+H+);
UV-Vis: e586 =4962 L·molÀ1·cmÀ1.
Typical Procedure for CO2/Epoxide Coupling to
Cyclic Carbonates Catalyzed by 6/TBAB (Entry 15,
Table S1 in the Supporting Information)
A 60-mL stainless steel pressure reactor equipped with
a magnetic stirring bar was charged, under CO2 atmosphere,
with 7.7 mg of catalyst 6 (7.15ꢄ10À6 mol) and 23.0 mg of
TBAB (7.15ꢄ10À5 mol) dissolved in 5.0 mL of PO (7.15ꢄ
10À2·mol). The reaction mixture was pressurized with CO2 at
2.0 MPa and stirred at 1208C for 1 h. The reactor was
cooled with ice, the CO2 released, 1.0 mL of mesitylene
(7.15ꢄ10À3 mol) was added as an internal standard and the
mixture was analyzed by 1H NMR spectroscopy using
CDCl3 as solvent; yield: 52.0%.
Synthesis of the Iron(III) Complex 4 (Scheme 2)
The complex 4 was synthesized using the same procedure
previously reported for complex 1.[11] Pro-ligand L4 (0.54 g;
0.96 mmol) was dissolved in THF (30 mL). The solution was
added to
a suspension of sodium hydride (0.082 g;
3.4 mmol) in THF (20 mL) and the mixture stirred at room
temperature overnight. The resulting suspension was filtered
through celite and slowly added at room temperature to
0.148 g of anhydrous iron(III) chloride (0.91 mmol) dis-
solved in 15 mL of THF. The rapid change of the color to
blue was observed and the reaction kept overnight at room
temperature. The mixture was then filtered through a celite
path and the solvent removed under reduced pressure af-
fording a deep purple crystalline solid; yield: 0.44 g (78%);
elemental analysis calcd. for C48H38Fe2Cl8O6S4: C 46.70, H
3.10, S 10.39; found: C 46.61, H 3.03, S 10.35; MS: m/z=
Crystallographic Data Collection and Structure
Determination for Complex 6
The X-ray intensity data were measured on a Bruker
SMART Apex II CCD area detector diffractometer. Cell di-
mensions and the orientation matrix were initially deter-
mined from a least-squares refinement on reflections mea-
Adv. Synth. Catal. 0000, 000, 0 – 0
10
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
sured in three sets of 20 exposures, collected in three differ-
ent w regions, and eventually refined against all data. A full
sphere of reciprocal space was scanned by 0.38 w steps. The
software SMART[17] was used for collecting frames of data,
indexing reflections, and determination of lattice parame-
ters. The collected frames were then processed for integra-
tion by the SAINT program[17] and an empirical absorption
correction was applied using SADABS.[18] The structure was
solved by direct methods (SIR 2004)[19] and subsequent
Fourier syntheses and refined by full-matrix least-squares on
F2 (SHELXTL),[20] using anisotropic thermal parameters for
all non-hydrogen atoms. All hydrogen atoms were added in
calculated positions, included in the final stage of refinement
with isotropic thermal parameters, U(H)=1.2 Ueq(C)
[U(H)=1.5 Ueq(C–Me)], and allowed to ride on their carrier
carbons. Crystal data and details of the data collection for
complex 6 are reported in Table S3 and Table S4 in the Sup-
porting Information. CCDC 1477580 contains the supple-
mentary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/
cif.
grateful to Dr. Guglielmo Monaco from University of Sale-
rno for useful discussion, Dr. Patrizia Oliva and Dr. Patrizia
Iannece from University of Salerno for technical assistance.
References
[1] a) J. Qiao, Y. Liu, F. Hong, J. Zhang, Chem. Soc. Rev.
2014, 43, 631–675; b) C. Maeda, Y. Miyazaki, T. Ema,
Catal. Sci. Technol. 2014, 4, 1482–1497; c) M. Aresta,
A. Dibenedetto, A. Angelini, Chem. Rev. 2014, 114,
1709–1742; d) I. Omae, Coord. Chem. Rev. 2012, 256,
1384–1405; e) M. Cokoja, C. Bruckmeier, B. Rieger,
W. A. Herrmann, F. E. Kꢅhn, Angew. Chem. 2011, 123,
8662–8690; Angew. Chem. Int. Ed. 2011, 50, 8510–8537;
f) E. A. Quadrelli, G. Centi, J.-L. Duplan, S. Perathon-
er, ChemSusChem 2011, 4, 1194–1215; g) D. J. Dare-
nsbourg, Inorg. Chem. 2010, 49, 10765–10780; h) M.
Aresta, A. Dibenedetto, Dalton Trans. 2007, 2975–
2992; i) T. Sakakura, J.-C. Choi, H. Yasuda, Chem. Rev.
2007, 107, 2365–2387; j) Activation of Carbon Dioxide,
(Ed.: S. L. Suib), Elsevier, Amsterdam, 2013.
[2] a) C. J. Whiteoak, A. W. Kleij, Synlett 2013, 24, 1748–
1756; b) M. North, R. Pasquale, C. Young, Green
Chem. 2010, 12, 1514–1539; c) A. Decortes, A. M. Cas-
tilla, A. W. Kleij, Angew. Chem. 2010, 122, 10016–
10032; Angew. Chem. Int. Ed. 2010, 49, 9822–9837;
d) C. Martꢆn, G. Fiorani, A. W. Kleij, ACS Catal. 2015,
5, 1353–1370; e) J. W. Comerford, I. D. V. Ingram, M.
North, X. Wu, Green Chem. 2015, 17, 1966–1987.
[3] a) X.-B. Lu, D. J. Darensbourg, Chem. Soc. Rev. 2012,
41, 1462–1484; b) D. J. Darensbourg, S. J. Wilson,
Green Chem. 2012, 14, 2665–2671; c) X.-B. Lu, W.-M.
Ren, G.-P. Wu, Acc. Chem. Res. 2012, 45, 1721–1735;
d) S. Klaus, M. W. Lehenmeier, C. E. Anderson, B.
Rieger, Coord. Chem. Rev. 2011, 255, 1460–1479;
e) M. R. Kember, A. Buchard, C. K. Williams, Chem.
Commun. 2011, 47, 141–163; f) D. J. Darensbourg,
Chem. Rev. 2007, 107, 2388–2410; g) G. W. Coates,
D. R. Moore, Angew. Chem. 2004, 116, 6784–6806;
Angew. Chem. Int. Ed. 2004, 43, 6618–6639.
[4] a) Element Recovery and Sustainability, RSC Green
Chemistry series, Vol. 22, (Ed.: A. J. Hunt), RSC Pub-
lishing, Cambridge, 2013; b) R. M. Izatt, S. R. Izatt,
R. L. Bruening, N. E. Izatt, B. A. Moyer, Chem. Soc.
Rev. 2014, 43, 2451–2475.
[5] a) X.-F. Liu, Q.-W. Song, S. Zhang, L.-N. He, Catal.
Today 2016, 263, 69–74; b) M. Cokoja, M. E. Wilhelm,
M. H. Anthofer, W. A. Herrmann, F. E. Kꢅhn, Chem-
SusChem 2015, 8, 2436–2454; c) G. Fiorani, W. Guo,
A. W. Kleij, Green Chem. 2015, 17, 1375–1389; d) Q.
He, J. W. OꢂBrien, K. A. Kitselman, L. E. Tompkins,
G. C. T. Curtis, F. M. Kerton, Catal. Sci. Technol. 2014,
4, 1513–1528; e) H. Li, P. S. Bhadury, B. Song, S. Yang,
RSC Adv. 2012, 2, 12525–12551; f) Z.-Z. Yang, Y.-N.
Zhao, L.-N. He, RSC Adv. 2011, 1, 545–567.
Computational Protocol
All the DFT geometry optimizations were performed at the
GGA level with the Gaussian 09 set of programs,[21] using
the BP86 functional of Becke and Perdew.[22] The electronic
configuration of the molecular systems was described with
the split-valence plus one polarization function basis set of
Ahlrichs for H, C, N, O, S and Br (SVP keyword in Gaussi-
an).[23] For Fe, we used the small-core, quasi-relativistic
Stuttgart-Dresden effective core potential, with the associat-
ed triple-z valence basis set (SDD keywords in Gaussi-
an09).[24] Geometry optimizations were performed without
symmetry constraints, and the characterization of the locat-
ed stationary points was performed by analytical frequency
calculations. For better energetics, energies were re-evaluat-
ed via single point calculations on the BP86/SVP optimized
geometries with the triple-z plus one polarization function
basis set proposed by Ahlrichs (TZVP keyword in Gaussi-
an)[25] using the M06 functional.[26] Solvent effects (propyl-
ene oxide) were estimated with the polarizable continuum
solvation model PCM.[27] To this M06/TZVP electronic
energy in solvent, zero point energy and thermal corrections
were included from the gas-phase frequency calculations at
the BP86/SVP level.[28] All the calculations were performed
with an unrestricted DFT formalism, assuming the two Fe
centers in a high spin state.
Acknowledgements
Financial support is acknowledged from the Ministero dell’Is-
truzione dell’Universitꢀ e della Ricerca (MIUR, Roma, Italy
for FARB 2015) and the SPRING cluster (REBIOCHEM re-
search project CTN01 00063 49393). LC acknowledges the
King Abdullah University of Science and Technology for
supporting this research. The Centro di Tecnologie Integrate
per la Salute (Project PONa3_00138) for the 600 MHz NMR
instrumental time is acknowledged. The authors are also
[6] a) X.-B. Lu, Y.-J. Zhang, K. Jin, L.-M. Luo, H. Wang, J.
Catal. 2004, 227, 537–541; b) D. Tian, B. Liu, Q. Gan,
H. Li, D. J. Darensbourg, ACS Catal. 2012, 2, 2029–
2035; c) J. A. Castro-Osma, C. Alonso-Moreno, A.
Lara-Sanchez, J. Martinez, M. North, A. Otero, Catal.
Sci. Technol. 2014, 4, 1674–1684; d) J. Martꢆnez, J. A.
Adv. Synth. Catal. 0000, 000, 0 – 0
11
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
Castro-Osma, A. Earlam, C. Alonso-Moreno, A.
Otero, A. Lara-Sꢇnchez, M. North, A. Rodrꢆguez-Diꢈ-
guez, Chem. Eur. J. 2015, 21, 9850–9862; e) M. North,
S. C. Z. Quek, N. E. Pridmore, A. C. Whitwood, X. Wu,
ACS Catal. 2015, 5, 3398–3402; f) J. A. Castro-Osma,
M. North, X. Wu, Chem. Eur. J. 2016, 22, 2100–2107;
g) S. H. Kim, D. Ahn, M. J. Go, M. H. Park, M. Kim, J.
Lee, Y. Kim, Organometallics 2014, 33, 2770–2775;
h) M. Cozzolino, K. Press, M. Mazzeo, M. Lamberti,
ChemCatChem 2016, 8, 455–460; i) S. Supasitmongkol,
P. Styring, Catal. Sci. Technol. 2014, 4, 1622–1630; j) Y.
Ren, O. Jiang, H. Zeng, Q. Mao, H. Jiang, RSC Adv.
2016, 6, 3243–3249; k) M. A. Fuchs, C. Altesleben,
T. A. Zevaco, E. Dinjus, Eur. J. Inorg. Chem. 2013,
2013, 4541–4545.
Benet-Buchholz, A. W. Kleij, Adv. Synth. Catal. 2012,
354, 469–476; h) C. J. Whiteoak, B. Gjoka, E. Martin,
M. M. Belmonte, E. C. Escudero-Adꢇn, C. Zonta, G.
Licini, A. W. Kleij, Inorg. Chem. 2012, 51, 10639–
10649; i) J. E. Dengler, M. W. Lehenmeier, S. Klaus,
C. E. Anderson, E. Herdtweck, B. Rieger, Eur. J. Inorg.
Chem. 2011, 2011, 336–343; j) A. Buchard, M. R.
Kember, K. G. Sandeman, C. K. Williams, Chem.
Commun. 2011, 47, 212–214; k) D. Alhashmialameer, J.
Collins, K. Hattenhauera, F. M. Kerton, Catal. Sci.
Technol. 2016, DOI: 10.1039/C6CY00477F.
[11] a) A. Buonerba, F. Della Monica, A. De Nisi, E. Lucia-
no, S. Milione, A. Grassi, C. Capacchione, B. Rieger,
Faraday Discuss. 2015, 183, 83–95; b) A. Buonerba, A.
De Nisi, A. Grassi, S. Milione, C. Capacchione, S.
Vagin, B. Rieger, Catal. Sci. Technol. 2015, 5, 118–123.
[12] a) D. F. Evans, J. Chem. Soc. 1959, 2003–2005; b) J. L.
Deutsch, S. M. Poling, J. Chem. Educ. 1969, 46, 167;
c) Physical Methods in Bioinorganic Chemistry, (Ed.: L.
Que Jr), University Science Books, Sausalito, CA,
2000.
[13] Advanced Inorganic Chemistry, (Ed.: F. A. Cotton, G.
Wilkinson, C. A. Murillo, E. M. Bochmann), Wiley,
2009.
[14] a) B. L. Morris, A. Wold, Rev. Sci. Instrum. 1968, 39,
1937–1941; b) L. F. Lindoy, V. Katovic, D. H. Busch, J.
Chem. Educ. 1972, 49, 117.
[7] a) J. Melendez, M. North, P. Villuendas, Chem.
Commun. 2009, 2577–2579; b) W. Clegg, R. W. Harring-
ton, M. North, R. Pasquale, Chem. Eur. J. 2010, 16,
6828–6843; c) M. North, C. Young, Catal. Sci. Technol.
2011, 1, 93–99; d) C. Beattie, M. North, P. Villuendas,
C. Young, J. Org. Chem. 2013, 78, 419–426; e) J. A.
Castro-Osma, A. Lara-Sanchez, M. North, A. Otero, P.
Villuendas, Catal. Sci. Technol. 2012, 2, 1021–1026;
f) J. A. Castro-Osma, M. North, X. Wu, Chem. Eur. J.
2014, 20, 15005–15008; g) M. North, B. Wang, C.
Young, Energy Environ. Sci. 2011, 4, 4163–4170; h) J.
Melendez, M. North, P. Villuendas, C. Young, Dalton
Trans. 2011, 40, 3885–3902; i) W. Clegg, R. W. Harring-
ton, M. North, P. Villuendas, J. Org. Chem. 2010, 75,
6201–6207; j) M. North, P. Villuendas, C. Young, Chem.
Eur. J. 2009, 15, 11454–11457; k) M. North, R. Pas-
quale, Angew. Chem. 2009, 121, 2990–2992; Angew.
Chem. Int. Ed. 2009, 48, 2946–2948; l) J. Melꢈndez, M.
North, R. Pasquale, Eur. J. Inorg. Chem. 2007, 2007,
3323–3326.
[15] R. Metzinger, C. Z. Limber, Z. Anorg. Allg. Chem.
2012, 638, 2225–2234.
[16] a) V. D’Elia, A. A. Ghani, A. Monassier, J. Sofack-
Kreutzer, J. D. A. Pelletier, M. Drees, S. V. C. Vumma-
leti, A. Poater, L. Cavallo, M. Cokoja, J.-M. Basset,
F. E. Kꢅhn, Chem. Eur. J. 2014, 20, 11870–11882; b) V.
D’Elia, H. Dong, A. J. Rossini, C. M. Widdifield,
S. V. C. Vummaleti, Y. Minenkov, A. Poater, E. Abou-
Hamad, J. D. A. Pelletier, L. Cavallo, L. Emsley, J.-M.
Basset, J. Am. Chem. Soc. 2015, 137, 7728–7739.
[17] SMART & SAINT Software Reference Manuals, ver-
sion 5.051, Windows NT Version, Bruker Analytical X-
ray Instruments Inc.: Madison, WI, 1998.
[8] a) C. J. Whiteoak, N. Kielland, V. Laserna, E. C. Escu-
dero-Adꢇn, E. Martin, A. W. Kleij, J. Am. Chem. Soc.
2013, 135, 1228–1231; b) C. J. Whiteoak, N. Kielland, V.
Laserna, F. Castro-Gꢉmez, E. Martin, E. C. Escudero-
Adꢇn, C. Bo, A. W. Kleij, Chem. Eur. J. 2014, 20, 2264–
2275; c) J. Rintjema, W. Guo, E. Martin, E. C. Escu-
dero-Adꢇn, A. W. Kleij, Chem. Eur. J. 2015, 21, 10754–
10762.
[18] G. M. Sheldrick, SADABS, program for empirical ab-
sorption correction, University of Gçttingen, Germany,
1996.
[9] a) I. Bauer, H.-J. Knçlker, Chem. Rev. 2015, 115, 3170–
3387; b) C.-L. Sun, B.-J. Li, Z.-J. Shi, Chem. Rev. 2010,
111, 1293–1314; c) A. Correa, O. Garcia Mancheno, C.
Bolm, Chem. Soc. Rev. 2008, 37, 1108–1117; d) C.
Bolm, J. Legros, J. Le Paih, L. Zani, Chem. Rev. 2004,
104, 6217–6254.
[10] a) M. Taherimehr, J. P. C. C. Sert¼, A. W. Kleij, C. J.
Whiteoak, P. P. Pescarmona, ChemSusChem 2015, 8,
1034–1042; b) X. Sheng, L. Qiao, Y. Qin, X. Wang, F.
Wang, Polyhedron 2014, 74, 129–133; c) M. Adolph,
T. A. Zevaco, C. Altesleben, O. Walter, E. Dinjus,
Dalton Trans. 2014, 43, 3285–3296; d) M. A. Fuchs,
T. A. Zevaco, E. Ember, O. Walter, I. Held, E. Dinjus,
M. Doring, Dalton Trans. 2013, 42, 5322–5329; e) M.
Taherimehr, S. M. Al-Amsyar, C. J. Whiteoak, A. W.
Kleij, P. P. Pescarmona, Green Chem. 2013, 15, 3083–
3090; f) C. J. Whiteoak, E. Martin, E. Escudero-Adꢇn,
A. W. Kleij, Adv. Synth. Catal. 2013, 355, 2233–2239;
g) C. J. Whiteoak, E. Martin, M. M. Belmonte, J.
[19] M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini,
G. L. Cascarano, L. De Caro, C. Giacovazzo, G. Poli-
dori, R. Spagna, J. Appl. Crystallogr. 2005, 38, 381–388.
[20] G. M. Sheldrick, SHELXTLplus Windows NT Version,
Structure Determination Package, Version 5.1. Bruker
Analytical X-ray Instruments Inc.: Madison, WI, USA,
1998.
[21] M. J. Frisch, et al., Gaussian 09, Revision A.1, Gaussi-
an, Inc., Wallingford, CT, 2009.
[22] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100;
b) J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824;
c) J. P. Perdew, Phys. Rev. B 1986, 34, 7406–7406.
[23] A. Schꢊfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992,
97, 2571–2577.
[24] a) U. Haeusermann, M. Dolg, H. Stoll, H. Preuss, Mol.
Phys. 1993, 78, 1211–1224; b) W. Kuechle, M. Dolg, H.
Stoll, H. Preuss, J. Chem. Phys. 1994, 100, 7535–7542;
c) T. Leininger, A. Nicklass, H. Stoll, M. Dolg, P.
Schwerdtfeger, J. Chem. Phys. 1996, 105, 1052–1059.
Adv. Synth. Catal. 0000, 000, 0 – 0
12
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
asc.wiley-vch.de
[25] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120,
215–241.
[26] A. Schaefer, C. Huber, R. Ahlrichs, J. Chem. Phys.
1994, 100, 5829–5835.
[27] a) V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102,
1995–2001; b) J. Tomasi, M. Persico, Chem. Rev. 1994,
94, 2027–2094.
[28] a) A. Poater, E. Pump, S. V. C. Vummaleti, L. Cavallo,
J. Chem. Theory Comput. 2014, 10, 4442–4448; b) E.
Pump, C. Slugovc, L. Cavallo, A. Poater, Organometal-
lics 2015, 34, 3107–3111; c) S. Manzini, A. Poater, D. J.
Nelson, L. Cavallo, S. P. Nolan, Chem. Sci. 2014, 5,
180–188.
Adv. Synth. Catal. 0000, 000, 0 – 0
13
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

FULL PAPERS
14 Coupling of Carbon Dioxide with Epoxides Efficiently
Catalyzed by Thioether-Triphenolate Bimetallic Iron(III)
Complexes: Catalyst Structure–Reactivity Relationship and
Mechanistic DFT Study
Adv. Synth. Catal. 2016, 358, 1 – 14
Francesco Della Monica, Sai V. C. Vummaleti,
Antonio Buonerba, Assunta De Nisi, Magda Monari,
Stefano Milione, Alfonso Grassi, Luigi Cavallo,*
Carmine Capacchione*
Adv. Synth. Catal. 0000, 000, 0 – 0
14
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!