Coordination cluster analogues of the high-Spin [Mn19] system with functionalized 2,6-Bis(hydroxymethyl)phenol ligands

Article abstract of DOI:10.1002/ejic.201402327

A series of 2,6-bis(hydroxymethyl)-4-R-phenol ligands (H₃L^R; R = H, F, Cl, Br, I, Ph, NH_2, NO₂, SMe) have either been newly synthesized or the existing syntheses have been significantly improved to investigate ligand-functionalized analogues of the previously published coordination cluster [Mn^III₁₂Mn^II₇(μ₄-O)₈(μ₃-N₃)₈(HL^Me)₁₂(MeCN)₆]Cl₂·10MeOH·MeCN (1) with S = 83/2. The crystal structures and magnetic properties of three such Mn₁₉ clusters, namely, [Mn^III₁₂Mn^II₇(μ₄-O)₈(HL^H)₁₂(μ₃-Cl)₇(μ₃-OMe)(MeOH)₆]Cl₂·16H₂O·10MeOH·MeCN (3), [Mn^III₁₂Mn^II₇(μ₄-O)₈(HL^I)₁₂(μ₃-N₃)₈(MeOH)₆](O₂CH)₂·16MeOH·10MeCN (4) and [Mn^III₁₂Mn^II₇(μ₄-O)₈(μ₃-Cl)_7.7(μ₃-OMe)_0.3(HL^SMe)₁₂(MeOH)₆]Cl₂·27MeOH (5) are reported and compared to those of the parent cluster. When these ligands are functionalized with substituents of moderate electronegativity, it is possible to synthesize Mn₁₉ analogues; however, when such ligands bear highly electron-donating (amino) or -withdrawing (nitro) substituents, the Mn₁₉ analogues are no longer accessible. The Mn₁₉ cluster framework is both magnetically and structurally robust with respect to the electron-donor/acceptor characteristics of the ligand substituent; therefore, the Mn₁₉ system is an excellent platform for peripheral chemical engineering. The robustness of the inorganic {Mn^III₁₂Mn^II₇(μ₄-O)₈} core of Mn₁₉ systems with variously functionalized encapsulating ligands is demonstrated by the invariance of the record S = 83/2 spin state. Chemical modification aimed towards attaching the molecule to various substrates does not interfere with the electronic structure.

Full text of DOI:10.1002/ejic.201402327

FULL PAPER
DOI:10.1002/ejic.201402327
Coordination Cluster Analogues of the High-Spin [Mn19]
System with Functionalized 2,6-Bis(hydroxymethyl)phenol
Ligands
Samir Mameri,*[a][‡] Ayuk M. Ako,*[b,c] Fatma Yesil,[a]
Marcel Hibert,[a] Yanhua Lan,[b] Christopher E. Anson,[b] and
Annie K. Powell*[b,c]
Keywords: Cluster compounds / Ligand design / Tridentate ligands / Molecular magnetism / Manganese
A series of 2,6-bis(hydroxymethyl)-4-R-phenol ligands (H3LR;
R = H, F, Cl, Br, I, Ph, NH2, NO2, SMe) have either been
newly synthesized or the existing syntheses have been sig-
nificantly improved to investigate ligand-functionalized ana-
logues of the previously published coordination cluster
(μ4-O)8(μ3-Cl)7.7(μ3-OMe)0.3(HLSMe
)12(MeOH)6]Cl2·27MeOH
(5) are reported and compared to those of the parent cluster.
When these ligands are functionalized with substituents of
moderate electronegativity, it is possible to synthesize Mn19
analogues; however, when such ligands bear highly elec-
tron-donating (amino) or -withdrawing (nitro) substituents,
the Mn19 analogues are no longer accessible. The Mn19 clus-
ter framework is both magnetically and structurally robust
with respect to the electron-donor/acceptor characteristics of
the ligand substituent; therefore, the Mn19 system is an ex-
cellent platform for peripheral chemical engineering.
[MnIII12MnII7(μ4-O)8(μ3-N3)8(HLMe
MeCN (1) with S = 83/2. The crystal structures and magnetic
properties of three such Mn19 clusters, namely, [MnIII12MnII
)12(MeCN)6]Cl2·10MeOH·
-
7
(μ4-O)8(HLH
)12(μ3-Cl)7(μ3-OMe)(MeOH)6]Cl2·16H2O·
10MeOH·MeCN (3), [MnIII12MnII7(μ4-O)8(HLI)12(μ3-N3)8-
(MeOH)6](O2CH)2·16MeOH·10MeCN (4) and [MnIII12MnII
-
7
Introduction
a common MnII vertex. The core is then encapsulated
within a coordination shell made up of 12 organic ligands
and eight face-bridging μ3,η1-X (e.g., X = azide) ligands. In
this coordination cluster, the ferromagnetic coupling of all
of the MnII/III centres leads to the record ground spin state
of S = 83/2,[1] which was further corroborated by EPR spec-
troscopy[2] and DFT calculations.[3]
The design and synthesis of molecular materials with
predictable magnetic properties continues to remain a chal-
lenging task, because the structural factors that govern ex-
change coupling between paramagnetic centres are complex
and elusive. Previously, we reported the [Mn19] coordina-
tion cluster [MnIII12MnII7(μ4-O)8(μ3-η1-N3)8(HLMe)12-
(MeCN)6]Cl2·10MeOH·MeCN (1), which was synthesized
from the commercially available 2,6-bis(hydroxymethyl)-4-
methylphenol (H3LMe). The coordination cluster 1 can be
The 4-substituted H3LR ligand system based on 2,6-bis-
(hydroxymethyl)phenol (Scheme 1) employed in the synthe-
sis of 1 is quite appealing for a number of reasons,[1,4]
though its use in coordination chemistry has been quite lim-
ited.[1,5,6] For instance, when partially or completely depro-
tonated, the phenolic and two hydroxy groups of the H3LR
ligand system have the capacity to bridge three metal ions
in a linear fashion, and the relatively short distances be-
tween those metal ions could favour ferromagnetic ex-
change pathways. Given the very large ground-state spin of
1, a natural extension to our previous work on the [Mn19]
described
in
terms
of
its
central
inorganic
{MnIII12MnII7(μ4-O)8} core, which is formed from the fu-
sion of two supertetrahedral {MnIII6MnII4} units that share
[a] Laboratoire d’Innovation Thérapeutique (Medicinal Chemistry
Laboratory), University of Strasbourg, UMR 7200-CNRS,
Faculté de Pharmacie,
74, route du Rhin, 67401 Illkirch, France
http://medchem.u-strasbg.fr/pages/welcome.php
[b] Institut für Anorganische Chemie, Karlsruhe Institute of
Technology (KIT),
Engesserstr. 15, 76131 Karlsruhe, Germany
E-mail: annie.powell@kit.edu
manase.ayuk@kit.edu
http://www.aoc.kit.edu/
[c] Institute of Nanotechnology, Karlsruhe Institute of Technology
(KIT),
P. O. Box 3640, 76021 Karlsruhe, Germany
http://www.int.kit.edu/home.php
[‡] Current address: Institut de Chimie, UMR 7177/CNRS-
Unistra, University of Strasbourg, 4, rue Blaise Pascal, 67000
Strasbourg, France
Scheme 1. Chemical structure of ligands based on 2,6-bis(hydroxy-
methyl)-4-R-phenol, H3LR; R = Me, H, Cl, Br, F, I, NH2, NO2,
SMe and OMe.
E-mail: mameri@unistra.fr
Eur. J. Inorg. Chem. 2014, 4326–4334
4326
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
system is the use of analogous ligands in which the 4-methyl H3LBr, H3LH, H3LI and H3LNO2 are more problematic and
substituent of the H3LMe ligand system is replaced by other often require large-scale column chromatography.[10] In this
R groups. Such functionalization could be aimed at further paper, we expand on our earlier work by reporting im-
chemical elaboration of the system. For example, the –NH2 proved and optimized syntheses of the latter five ligands
group was targeted as a gateway to other functionalities and describe the syntheses of the new compounds H3LNH
2
such as peptides or for C–N amination cross-coupling reac- and H3LSMe. We then describe the syntheses, structures and
tions, and the ligand bearing a pendant iodo end group is magnetic properties of the [Mn19] aggregates derived from
a good candidate for cross-coupling reactions in solution or three of these ligands (H3LH, 3; H3LI, 4; H3LSMe, 5) and
solid state. Alternatively, a substituent may aid the interac- compare these with the aggregates involving the ligands
tion of the cluster with a substrate. To this effect, we re- with R = Me[1] and OMe.[4]
cently demonstrated the use of the ligand H3LOMe in the
synthesis of the Mn19 aggregate [Et3NH]2[MnIII12MnII7(μ4-
O)8(μ3-N3)7.4(μ3-Cl)0.6(HLOMe)12(MeOH)6]Cl4·14MeOH
(2), molecules of which were successfully deposited onto
Results and Discussion
highly oriented pyrolytic graphite (HOPG) substrates and
Synthetic Studies
investigated by scanning tunneling microscopy (STM) and
current imaging tunneling spectroscopy (CITS).[4] A sulfur-
Previously, we reported the Mn19 aggregate [MnIII
-
12
based substituent such as –SCH3 would allow attachment MnII7(μ4-O)8(μ3-N3)8(HLMe
)12(MeCN)6]Cl2·10MeOH·
of the cluster to the surfaces of metals such as gold.[7] As MeCN (1), which displays the highest spin ground state
changes to the electron-donor/acceptor properties of the known for a 3d molecular species to date: ST = 83/2.[1] The
para substituent will alter the electron density of the bridg- organic ligand (HLMe 2–
in this complex is obtained by de-
)
ing phenoxo group within the ligand, it is essential to know protonation of the commercially available 2,6-bis(hydroxy-
whether the core structure of the [Mn19] aggregates as well methyl)-4-methylphenol (H3LMe). It was considered useful
as their magnetic properties are retained upon ligand func- to prepare analogous aggregates with ligands bearing alter-
tionalization, before any further manipulation and investi- native functional groups at the 4-position. Although the
gation. High-spin isotropic molecules are also of interest published syntheses of the ligands H3LCl and H3LOMe are
because of their potential for applications that utilize their relatively convenient, those for H3LH, H3LF, H3LBr, H3LI
magnetocaloric effect, for example, in refrigeration or in and H3LNO are not. The fluoro and bromo ligands, in par-
2
theranostics, for which such complexes could be used as ticular, require time-consuming large-scale column
magnetic resonance imaging contrast agents.[8,9] Convenient chromatography.[10] Our modified syntheses of the H3LBr
syntheses of the ligands H3LCl, H3LPh and H3LOMe are al- and H3LF ligands were realized by treating the correspond-
ready readily available, but those published for H3LF, ing para-substituted phenols with formaldehyde in basic
Scheme 2. Synthetic scheme for preparation of 2,6-bis(hydroxymethyl)-4-R-phenol ligands H3LR. Reagents and conditions: (i) EtOH,
H2SO4, reflux; (ii) BBr3, CH2Cl2, –78 °C; (iii) LiAlH4, tetrahydrofuran (THF), 0 °C to room temp.; (iv) BTMA·ICl2, NaHCO3, MeOH,
room temp.; (v) formaldehyde, NaOH, H2O/MeOH, Δ; (vi) paraformaldehyde, AcOH, H2SO4, 50 °C; (vii) conc. HCl, water, reflux; (viii)
H2, Pd/C, 60 psi, room temp.
Eur. J. Inorg. Chem. 2014, 4326–4334
4327
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
media, and the workup for these two ligands was also opti-
Although the syntheses of the complexes reported here
mized. The new methylmercapto ligand H3LSMe could be were generally similar to that reported for 1,[1] the establish-
prepared in a similar way. However, similar reactions in- ment of the appropriate conditions in each case was not
volving 4-iodo-, 4-nitro- and 4-aminophenols failed to pro- straightforward and represented quite a substantial syn-
duce isolable materials (with at best only traces observable thetic and crystallization challenge, as had been the case for
by HPLC for the iodo compound). Therefore, we explored 2.[4] In general, the complexes described in this work were
new synthetic routes (Scheme 2).
obtained upon reaction of the appropriate H3LR ligand
Access to both the parent ligand H3LH and the iodo- with a base (NaOAc·3H2O or Et3N) and MnCl2·4H2O or
substituted ligand H3LI was made possible by a multistep MnBr2·4H2O in the presence or absence of NaN3 in MeCN/
synthesis from commercially available 2-methoxyiso- MeOH (typically 5:1 v/v) mixtures and crystallized by slow
phthalic acid. The product of the first step of this synthesis, evaporation of the mother liquor at room temperature.
diethyl-2-methoxyisophthalate (I), was prepared in high Various reaction conditions such as reaction components,
yield (ca. 90%) from 2-methoxyisophthalic acid by using stoichiometry, concentration, reaction time, pH and tem-
H2SO4 for activation. The removal of the methyl group of perature were investigated, and we report here the optimum
the anisole methoxy group with BBr3 to give the corre- conditions that we have established. For instance, with
sponding phenol (compound II) was quantitative in dry H3LH,
the
Mn19
[MnIII12MnII7(μ4-O)8(μ3-Cl)7(μ3-
CH2Cl2 at –78 °C. A hydride reduction of the ester groups OMe)(HLH)12(MeOH)6]Cl2·16MeOH·MeCN (3) complex
of II was then successfully performed to afford the unsub- could only be obtained under ambient conditions when
stituted parent ligand H3LH. This could then be readily Et3N was employed as a base and in the absence of NaN3;
iodinated with benzyltrimethylammonium dichloroiodate the synthesis is sensitive to concentration and requires very
(BTMA·ICl2) in the presence of NaHCO3 to yield H3LI.[10] dilute solutions. By contrast, crystalline products involving
For the synthesis of the amino-substituted ligand H3LI could only be obtained from reactions involving
H3LNH , we first synthesized the benzodioxene derivative 8- NaN3 under reflux conditions to yield [MnIII12MnII7(μ4-O)8-
2
acetoxymethyl-6-nitro-1,3-benzodioxene (III) from p-nitro- (μ3-N3)8(HLI)12(MeOH)6](O2CH)2·16MeOH·10MeCN (4).
phenol by using paraformaldehyde under acidic condi- On the other hand, the ligand H3LSMe required similar re-
tions.[10d,10e] Notably, the temperature must be carefully action conditions to those used for 3 and yielded [MnIII
-
12
controlled to obtain good yields of dioxene III; even a small MnII7(μ4-O)8(μ3-Cl)7.7(μ3-OMe)0.3(HLSMe)12(MeOH)6]Cl2·
deviation of a few degrees from the optimal +50 °C led to 27MeOH (5).
side-products. Subsequently, the dioxene was heated to re-
Attempted syntheses using the nitro-substituted ligand
flux in acidic media to free the alcoholic chelating moiety H3LNO only gave a crystalline product after extended reac-
2
and, thus, afford the nitro ligand H3LNO . Finally, a palla- tion times, and the small crystal size meant that synchro-
2
dium-catalyzed reduction of the nitro group in H3LNO with tron radiation was necessary to obtain a good-quality data-
2
H2 under high pressure was then performed to give the new set. Initial crystallographic analysis showed that the tem-
diformylated amino species H3LNH . The chemical struc- perature factor for the central metal ion of the nonadecanu-
2
tures of all of the ligands from our syntheses were con- clear complex obtained was significantly higher than those
1
firmed by H, 13C and 19F NMR (for H3LF) spectroscopy, for the other Mn centres (unlike the other Mn19 systems we
high-resolution mass spectroscopy (HRMS) and elemental have structurally characterized to date), which indicates that
analysis.
this position is not purely occupied by MnII ions. Refine-
To exploit the broad application of the H3LR ligand sys- ment of this metal atom position as a disordered overlay of
tem and to test the robustness of the [Mn19] system to Mn and Na ions supported this and gave relative occupanc-
changes in ligand substituent, we have investigated the li- ies of ca. 80:20 for Mn and Na; this complex was not inves-
gands in the synthesis of the corresponding [Mn19R] aggre- tigated further. It is unclear whether this Mn/Na mixture at
gates. Of the ligands for which the syntheses are reported the central position is directly related to the highly electron-
here, we describe the structures of the Mn19 aggregates ob- withdrawing nitro group on the ligand (Hammett σpara
=
tained with H3LH (the unsubstituted “parent” ligand), H3LI +0.78[14]) or whether the longer reaction times that were
(as the iodo substituent should allow further ligand func- necessary with this ligand are also responsible. No crystal-
tionalization) and H3LSMe (as the sulfur-containing substit- line product was obtained with the strongly electron-donat-
uent will allow addressing onto gold surfaces). The synthe- ing (Hammett σpara = –0.66[11]) ligand H3LNH2. Therefore,
sis and structure of the complex [Et3NH]2[MnIII12MnII7(μ4- we conclude that isostructural Mn19 aggregates can be syn-
O)8(μ3-N3)7.4(μ3-Cl)0.6(HLOMe)12(MeOH)6]Cl4·14MeOH (2) thesized with a wide range of R substituents on the organic
has been reported elsewhere.[4] This bears a moderately elec- ligand, providing the most strongly electron-donating or
tron-donating methoxy substituent at the ligand 4-position -withdrawing functional groups (amino or nitro) are avo-
(Hammett σpara = –0.27, compared to –0.17 for the methyl ided.
group in the original aggregate 1).[11] Similarly, we have pre-
viously reported the heterometallic aggregate [MnIII
-
12
Structures of the Mn19 Aggregates
MnII6LuIII(μ4-O)8(μ3-η1-N3)6(μ3-η1-Cl)2(HLF)12(MeCN)6]-
Cl3·3H2O·7MeOH·MeCN,[12] in which the fluoro substitu-
ent on the ligand has Hammett σpara = +0.06.[11]
The structures for 3–5 were determined by X-ray diffrac-
tion studies (Figure 1), and the crystallographic data are
Eur. J. Inorg. Chem. 2014, 4326–4334
4328
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
an octahedron of MnIII centres inscribed within them
linked by the μ4-O2– bridges. This inorganic core is encapsu-
lated within a coordination shell made up of 12 organic
–
ligands and eight μ3,η1-X (X = N3 , Cl– or –OMe–) ligands,
which cap the faces of the MnIII octahedra. The three
6
metal centres (MnII···MnIII···MnII) along each supertetra-
hedral edge are bridged by two oxygen atoms of the phen-
oxo group and one alkoxo group, whereas the second alk-
oxy oxygen remains protonated and ligates the outer MnII
centres (Scheme 3). The differences between these three
structures, and also those of 1 and 2, involve the R substitu-
ent of the organic ligand, the (μ3-X)– face-bridging ligands
and, in one case (4), the counterions.
Scheme 3. Coordination of (HLR)2– ligands (see Table 1). In li-
gands of type A, Ob coordinates to the central MnII ion; for ligands
of type B, Ob coordinates to an apical MnII ion.
¯
Aggregate 3 crystallizes in space group R3 with Z = 3;
the cell parameter a is similar to that for 1, but c is some-
what longer, which is perhaps unexpected as the methyl
groups in 1 have been replaced by smaller hydrogen atoms
in 3. Although the face-bridging ligands are different (seven
chlorido and one methoxo in 3 instead of the eight azides
in 1) the terminal ligands on the outer MnII centres are
now MeOH instead of MeCN, and it appears that the extra
hydrogen-bonding capabilities that these provide have re-
sulted in the longer unit cell for 3. A similar effect was
found for the structure of 2, which also has terminal MeOH
ligands.[4] Complex 5 crystallizes in the cubic space group
¯
Pa3 with Z = 4, and the crystallographic site symmetry is
¯
therefore once again 3 but with a more open crystal struc-
ture than for 1–3 and a significantly higher lattice solvent
content.
By contrast, 4 crystallizes in space group P21/n with Z =
2 and is, thus, the only compound from 1–5 not to crys-
¯
tallize with 3 site symmetry. Although the 4-iodo substitu-
ent is substantially larger than most of the others, the dif-
ferent packing is in this case driven by the change in coun-
terion. Instead of chloride ions, as for the other compounds,
the charge of the aggregate in 4 is balanced by two formate
ions, which presumably result from oxidation of methanol
solvent molecules. Each formate ion acts as a supramo-
lecular bridge between two cluster molecules and accepts
Figure 1. Molecular structures of the Mn19 clusters in 3 (top), 4
(middle) and 5 (bottom). H atoms, counterions and lattice solvent
molecules omitted for clarity; MnIII dark pink, MnII pale pink, O
red, N blue, Cl green, F pale green, I dark brown.
summarized in Table 3. The structures of these compounds two hydrogen bonds from each of them.
are all very similar to those observed previously for the
The principal focus of the work presented here is to study
[Mn19] coordination clusters 1 and 2.[1,4] As noted in the the effect of variation of the substituents on the organic
Introduction, such coordination clusters can be described ligands on the properties of the corresponding aggregates.
in terms of a central inorganic {MnIII12MnII7(μ4-O)8} core For these purposes, it is convenient to quantify the electron-
formed from the fusion of two supertetrahedral donating or -withdrawing properties of the various R
{MnIII6MnII4} units, in which the MnII centres describe the groups in terms of their Hammett σpara parameters.[11] The
tetrahedra and share a common MnII vertex and each has bond lengths and angles involving the oxygen atoms of
Eur. J. Inorg. Chem. 2014, 4326–4334
4329
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 1. Bond lengths [Å] and angles [°] involving oxygen atoms of the (HLR)2– ligands in 1–5.[a]
1[1]
2[4]
3
4[a]
5
R
Me
–0.17
(N3)8
OMe
–0.27
(N3)7.4(Cl)0.6
H
–
I
SMe
0.00
(Cl)7.7(OMe)0.3
[11]
Hammett σpara
μ3 ligands
+0.18
(N3)8
(Cl)7(OMe)
Type A Ligands[b]
Oa–MnII
2.2862(17)
103.76(7)
1.8944(17)
1.8545(18)
109.29(8)
2.3446(17)
2.221(2)
2.353(3)
101.41(12)
1.891(3)
1.854(3)
110.64(13)
2.315(3)
2.187(3)
2.306(3)
102.63(12)
1.891(3)
1.842(3)
110.04(12)
2.304(3)
2.191(3)
2.302(4)
102.51(18)
1.897(4)
1.850(4)
109.6(2)
2.332(4)
2.214(5)
2.316(3)
101.70(11)
1.905(2)
1.848(2)
111.22(10)
2.309(2)
2.216(4)
MnII–Oa–MnIII
Oa–MnIII
Ob–MnIII
MnIII–Ob–MnII
Ob–MnII
Oc–MnII
Type B Ligands[b]
Oa–MnII
2.3602(17)
101.27(7)
1.8947(17)
1.8782(17)
104.75(8)
2.1856(17)
2.211(2)
2.325(3)
100.78(12)
1.905(3)
1.876(3)
104.17(13)
2.204(3)
2.232(3)
2.364(3)
104.75(14)
1.900(3)
1.872(3)
101.57(12)
2.188(3)
2.218(4)
2.366(5)
101.8(2)
1.896(4)
1.874(5)
104.8(2)
2.216(5)
2.199(5)
2.334(3)
102.08(11)
1.902(3)
1.872(3)
104.66(12)
2.216(3)
2.215(3)
MnII–Oa–MnIII
Oa–MnIII
Ob–MnIII
MnIII–Ob–MnII
Ob–MnII
Oc–MnII
[a] For 4, the parameters given are each the mean value of the three corresponding chemically equivalent but crystallographically indepen-
dent bond lengths or angles. [b] : Type A ligands bridge a supertetrahedral edge involving the central MnII ion [Mn(1)] and an “outer”
MnII ion. Type B ligands bridge supertetrahedral edges between two “outer” MnII centres.
these ligands in 1–5 are compared in Table 1. As the atomic the decisive factor governing the magnetic coupling within
numbering schemes are not the same for all of the com- the aggregate, but rather it is the bonding parameters in-
plexes, the ligand oxygen atoms are designated according to volving the μ4-O bridges that govern the overall magnetic
Scheme 3. Within each structure, there are two distinct structure of the aggregate.[3]
types of ligand. Although both types have the same connec-
tivity as shown in Scheme 3, six ligands per aggregate
Table 2. Summary of magnetic data for 3–5.
1[1]
2[4]
3
4
5
bridge edges of a supertetrahedron that link the central
MnII ions [Mn(1) in each structure] to an outer MnII ion,
whereas the other six bridge edges between two outer MnII
centres; these are designated Types A and B, respectively, in
Table 1.
χЈT [cm3 Kmol–1] at 300 K 93
83
85
89
86
χЈT [cm3 Kmol–1] max.
894 831
770 (at 4.0 K) 904 864
688
χЈT [cm3 Kmol–1] at 1.8 K 894 831
904 864
83.9 82.1
Msat [μB] at 1.8 K, 70 kOe 84.5 84.1 83.6
In 1–5, it is not only the organic R group that changes;
in 1, 2 and 4, the face-bridging ligands are all or mostly
all azide ligands, whereas they are predominantly chloride
ligands in 3 and 5. However, consideration of the param-
eters presented in Table 2 should allow us to attribute the
Magnetic Studies
The magnetic properties of freshly prepared polycrystal-
effects of the various structural changes. Any direct effect line samples of 3–5 were measured. For all compounds, the
on the ligand–cluster bonding resulting from changes in the alternating current (ac) susceptibility was checked but
R group are likely to result primarily in changes to the geo- showed no out-of-phase signal above 1.8 K and no fre-
metries around the phenoxo oxygen atoms Oa, that is, quency dependence of the in-phase component. The tem-
MnIII–Oa and MnII–Oa, as the methanolic oxygen atoms perature dependence of the χT products and magnetizations
Ob and Oc are not in conjugation with R. Changes resulting for 3–5 are summarized in Table 2 and are very similar to
from the face-bridging ligands, on the other hand, would the published data for 1 and 2.[1,4]
result in changes to the MnIII–Oa and MnIII–Ob geometries.
For all of the compounds, the χT value at 300 K is signif-
However, although some sets of corresponding parameters icantly higher than the theoretical values for the 19 inde-
in Table 1 do show differences at the 3σ significance level, pendent metal ions (66.6 cm3 Kmol–1 for gav = 2.00) and
it is not really possible to discern any overall trends that then continuously increases as the temperature decreases to
can be assigned to changes in R or the face-bridging li- reach values consistent with an S = 83/2 ground state at
gands. What appears to be the case here is that neither 1.8 K (the theoretical value is 881.9 cm3 Kmol–1 for g =
changing the ligand substituent nor the nature of the face- 2.00), provided that the susceptibility is measured with a
bridging ligand has any significant effect on the Mn–O small ac field (3 Oe at 200 Hz) rather than the more usual
bonding in terms of the observed bond lengths or angles. 1000 Oe direct current (dc) field to avoid saturation effects.
We note here that the DFT calculations performed on 1 The slight exception to this is 3, for which χT reaches a
suggested that the nature of the face-bridging ligands is not maximum value of 770 cm3 Kmol–1 at 4.0 K before drop-
Eur. J. Inorg. Chem. 2014, 4326–4334
4330
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Figure 2. Temperature dependence of the χT products for 3, 4 and 5. Insets: reduced magnetizations at the indicated temperatures.
ping slightly to 688 cm3 Kmol–1 at 1.8 K; this is presumably sistent with the S = 83/2 state observed for the original
the effect of weak antiferromagnetic intermolecular interac- [Mn19] aggregate, 1. Thus, the present work demonstrates
tions. Ferromagnetic interactions and the S = 83/2 ground the structural robustness of the core of the [Mn19] system,
state for all the compounds are further confirmed by the which remains essentially unchanged in spite of various
field dependences of their magnetization below 10 K with changes to the encapsulating ligands. This is in line with the
very quick saturation of magnetization with even a small suggestion from the DFT calculations on the [Mn19] sys-
external field to reach values close to 83 μB (Figure 2). The tems, which have led to the conclusion that it is the bonding
almost perfect superimposition of the reduced magnetiza- parameters within the central inorganic {MnIII12MnII7(μ4-
tion curves at different temperatures onto a master curve O)8} core that determine the nature of the magnetic struc-
for each compound indicates a lack of anisotropy (insets in ture. Indeed, the robustness of the magnetic behaviour and
Figure 2), and they are in very good agreement with Bril- electronic structure of the core with respect to encapsulat-
louin functions calculated for S = 83/2 and g = 2.0, as was ing ligand functionalization is particularly appealing in
also the case for 1.[1] This demonstrates the robust nature view of possible applications. Thus, as synthetic routes to
of the central inorganic core of these [Mn19] systems, as the functionalized encapsulating ligands have been developed,
nature of the encapsulating ligands has very little effect on it should now be possible to provide a library of further
the overall electronic and magnetic structure.
ligands to tailor the [Mn19] system for attachment to a wide
variety of possible substrates ranging from metal, semicon-
ductor or insulator surfaces through to biological macro-
molecules with the maximum spin ground state maintained
at 83/2.
Summary and Conclusions
We have demonstrated the tunability of the H3LR ligand
system based on 4-substituted 2,6-bis(hydroxymethyl)phen-
ols by developing convenient synthetic methodologies to the
ligands with R = H, Br, Ph, F, Cl, I, NO2 and NH2 to sit
alongside the existing commercial available H3LMe and the
high-yield syntheses of H3LOMe and H3LCl.[10] By tuning
the reaction conditions, new [Mn19] complexes with R =
H, I and SMe have been synthesized and structurally and
magnetically characterized. A comparison of the data for
these complexes with the data obtained for the previously
published aggregates with R = Me[1] and OMe[4] indicates
that changes to the ligand substituent do not result in any
significant changes to the core structures or to the magnetic
properties of the aggregates. All of the compounds de-
Experimental Section
Ligand Synthesis
General Procedures: Reactions were performed under an argon at-
mosphere unless specified otherwise. Solvents were purified by
standard methods before use [CH2Cl2 (CaH2), MeOH (Mg), THF
(Na/benzophenone)]. EtOH (HPLC grade) was used as purchased.
2-Methoxyisophthalic acid was purchased from Aldrich (85% pu-
rity; the impurity was identified as 2-methoxy-3-methylbenzoic
acid, vide infra). TLC was performed with silica gel plates (Merck,
60F-254). Flash chromatography was performed with 40–63 μm
SiO2. 1H, 13C and 19F NMR spectra were recorded at ambient tem-
scribed here exhibit ferromagnetic ground spin states con- perature at 400 MHz with tetramethylsilane (TMS) as an internal
Eur. J. Inorg. Chem. 2014, 4326–4334
4331
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
standard (Bruker UltraShield Plus). Chemical shifts (δ) of samples
in CDCl3, CH3OD or [D6]acetone are in parts per million (ppm).
The splitting patterns are designated as: s singlet, br s broad singlet,
d doublet, t triplet, q quartet, and m multiplet. Coupling constants
(J) between two nuclei separated by n chemical bonds are denoted
in Hertz (Hz). FTIR spectra were recorded with a Nicolet 380 ap-
paratus. HRMS spectra were recorded with an Agilent Technol-
ogies 6520 Q-TOF apparatus. Melting points were recorded with
EtOAc (100 mL) was added, and the organic phase was further
washed with water (30 mL) and dried with cotton. Evaporation of
the solvents yielded H3LH as a white crystalline solid (404 mg,
98%). Rf = 0.49 (CH2Cl2/AcOEt, 1:1), m.p. 95.4 °C. 1H NMR
([D6]acetone): δ = 4.57 (br s, 2 H), 4.76 (s, 4 H), 6.78 (t, J = 7.6 Hz,
1 H), 7.12 (d, J = 7.6 Hz, 2 H), 8.56 (br s, 1 H) ppm. 13C NMR
([D6]acetone): δ = 62.3, 119.8, 127.1, 127.8, 154.7 ppm. Selected IR
(neat): ν = 784 (m), 989 (s), 1008 (s), 1061 (s), 1204 (s), 1456 (s),
˜
a Büchi B-540 melting point apparatus. Elemental analyses were 1594 (m), 2884 (w), 2921 (w), 3292 (m), 3401 (m) cm–1. C8H10O3
obtained at the analytical facility of the IUT Robert Schuman,
University of Strasbourg. The synthesis of 2,6-bis(hydroxymethyl)-
4-chlorophenol (H3LCl) was performed by following the literature
procedure.[9] The 2,6-bis(hydroxymethyl)-4-R-phenol species (R =
Br, F, H, I, NO2) were synthesized by using improved versions of
the literature procedures,[10a–10f] as described below.
(154.17): calcd. C 62.33, H 6.54; found C 62.19, H 6.82.
4-Fluoro-2,6-bis(hydroxymethyl)phenol (H3LF):[10c] Formaldehyde
(50 mL, 37% w/w in water) was added to a solution of NaOH (5 g,
125 mmol) and 4-fluorophenol (5.6 g, 50 mmol) in a hydroalcoholic
mixture (37 mL, 23% methanol). The resulting solution was stirred
for 3 d at 35 °C. Then, glacial acetic acid (7.5 mL) in water (25 mL)
was added dropwise at 0 °C. The resulting solution was concen-
trated in vacuo. Extraction with hot EtOAc (3ϫ 50 mL) and fur-
ther drying over cotton produced an orange oil that was purified
by flash chromatography (eluent EtOAc/CH2Cl2, 70:30 to 100:0) to
yield the title fluorinated ligand H3LF (6.8 g, 79%) as a yellow so-
lid. Rf = 0.77 (AcOEt), m.p. 140 °C (ref. 137–139 °C).[9a] 1H NMR
([D6]acetone): δ = 4.69–4.71 (m, 2 H), 4.75 (d, J = 5.2 Hz, 4 H),
6.91 (d, J = 9 Hz, 2 H), 8.38 (s, 1 H) ppm. 13C NMR ([D6]acetone):
Diethyl 2-Methoxyisophthalate (I): Conc. H2SO4 (600 μL,
11 mmol) was added to a solution of 2-methoxyisophthalic acid
(940 mg, 4.78 mmol) in EtOH (45 mL). The resulting mixture was
heated in the open air under reflux for 24 h. Then, the solvents
were removed, and the resulting yellowish oil was dissolved in
CH2Cl2 and dried with cotton. Purification by column chromatog-
raphy (eluent: CH2Cl2) allowed separation of the ester I (1.05 g,
88%) as a colourless oil from the monoester byproduct (vide infra).
1
Rf = 0.5 (CH2Cl2). H NMR (CDCl3): δ = 1.39 (t, J = 7.2 Hz, 6 δ = 61.6, 112.5, 112.8, 129.71, 129.78, 150.0, 156.0, 158.3 ppm. 19
F
H), 3.92 (s, 3 H), 4.38 (q, J = 7.2 Hz, 4 H), 7.18 (t, J = 7.8 Hz, 1 NMR ([D6]acetone): δ = 50.79 (t, J = 9.4 Hz) ppm. Selected IR
H), 7.90 (d, J = 7.6 Hz, 2 H) ppm. 13C NMR (CDCl3): δ = 14.4, (neat): ν = 699 (s), 868 (s), 966 (s), 993 (s), 1014 (s), 1070 (s), 1124
˜
61.5, 63.8, 123.6, 127.2, 134.9, 159.5, 165.9 ppm. Selected IR
(s), 1201 (s), 1245 (s), 1313 (m), 1333 (m), 1458 (s), 1482 (s), 3274
(CDCl ): ν = 1004 (m), 1024 (m), 1148 (s), 1199 (s), 1244 (s), 1303
(m), 3406 (m) cm–1. HRMS: calcd for. C8H9FO3 [M]+ 172.05375;
˜
3
(s), 1465 (m), 1590 (m), 1725 (s), 2360 (w), 2982 (w) cm–1. C13H16O5 found. 172.05375. C8H9FO3 (172.16): calcd. C 55.81, H 5.27; found
(252.27): calcd. C 61.90, H 6.39; found C 62.06, H 6.42.
C 55.60, H 5.38.
Ethyl 2-Methoxy-3-methylbenzoate (byproduct): The impurity in the
commercial 85% 2-methoxyisophthalic acid is assumed to be 2-
methoxy-3-methylbenzoic acid, as its ethyl ester could be isolated
4-Bromo-2,6-bis(hydroxymethyl)phenol (H3LBr): Formaldehyde
(55 mL, 37% w/w in water) was added to a solution of NaOH
(5 g, 125 mmol) and 4-bromophenol (8.1 g, 47 mmol) in a water/
methanol mixture (39 mL, 21% methanol). The resulting solution
as a byproduct from the synthesis of I: white solid. Rf = 0.97
1
(CH2Cl2/AcOEt, 1:1). H NMR (CDCl3): δ = 1.40 (t, J = 7.2 Hz, was stirred overnight at 65 °C. Then, glacial acetic acid (9 mL) in
3 H), 2.32 (s, 3 H), 3.83 (s, 3 H), 4.38 (q, J = 7.1 Hz, 2 H), 7.05 (t,
water (30 mL) was added dropwise at 0 °C. The solid was collected
J = 7.6 Hz, 1 H), 7.33 (d, J = 7.6 Hz, 1 H), 7.63 (d, J = 7.8 Hz, 1 from the resulting slurry by filtration and washed with cold water
H) ppm. 13C NMR (CDCl3): δ = 14.4, 16.2, 61.1, 61.6, 123.6, 125.2,
129.2, 132.8, 135.1, 158.4, 166.7 ppm.
(50 mL), CH2Cl2 (60 mL) and cold acetone (10 mL) to yield the
desired brominated phenol (10.3 g, 94%) as a white powder. All
analysis data correspond to those previously published.[10]
Diethyl 2-Hydroxyisophthalate (II): Compound
I
(750 mg,
2.97 mmol) was dissolved in CH2Cl2 (15 mL). The solution was
cooled to –78 °C [acetone/CO2(s) bath]. BBr3 (1 m in CH2Cl2,
3.3 mL, 3.3 mmol) was then added dropwise. The bath was re-
moved, and the resulting mixture was further stirred for 15 min.
The bath was replaced, and water (10 mL) and HCl (1N, 5 mL)
were added slowly. The organic phase was separated and further
6-Bis(hydroxymethyl)-4-iodophenol (H3LI):[10d] Ligand H3LH
(100 mg, 649 μmol) was dissolved in MeOH (4 mL), and NaHCO3
(275 mg, 3.27 mmol) and benzyltrimethylammonium dichloro-
iodate (290 mg, 833 μmol) were added. The resulting orange sus-
pension was stirred overnight at room temperature, and then the
solvent was evaporated. The resulting solid was dissolved in EtOAc
washed with H2O (3ϫ 15 mL), saturated NaHCO3 (3ϫ 15 mL) (75 mL) and H2O (25 mL). The organic phase was further washed
and H2O (30 mL). Drying over cotton and evaporation of the sol-
vent gave a yellow oil, which was dissolved in AcOEt and filtered
over silica to yield the phenol II (705 mg, 99%) as a colourless oil.
with water (6ϫ 10 mL) and dried with cotton. Evaporation of the
solvent gave the iodo ligand H3LI (168 mg, 93%) as a yellow solid.
1
Rf = 0.6 (CH2Cl2/AcOEt, 1:1), m.p. 153.1 °C. H NMR ([D6]acet-
1
Rf = 0.93 (CH2Cl2/AcOEt, 1:1). H NMR (CDCl3): δ = 1.41 (t, J
one): δ = 4.69–4.74 (m, 6 H), 7.47 (s, 1 H), 8.67 (s, 1 H) ppm. 13C
NMR ([D6]acetone): δ = 61.3, 81.5, 131, 135.3, 154.4 ppm. Selected
= 7.0 Hz, 6 H), 4.41 (q, J = 7.0 Hz, 4 H), 6.91 (t, J = 7.8 Hz, 1 H),
8.04 (d, J = 8.0 Hz, 2 H) ppm. 13C NMR (CDCl3): δ = 14.4, 61.6, IR (neat): ν = 697 (m), 867 (s), 929 (s), 1003 (s), 1064 (s), 1193 (s),
˜
117.0, 118.4, 136.2, 161.8, 167.8 ppm. Selected IR (CDCl ): ν =
1206 (s), 1261 (s), 1332 (s), 1418 (s), 1455 (s), 1475 (s), 2357 (w),
˜
3
759 (s), 1024 (s), 1150 (s), 1187 (s), 1250 (s), 1439 (s), 1613 (s), 1671 2884 (w), 3307 (s), 3396 (s) cm–1. C8H9IO3 (280.06): calcd. C 34.31,
(s), 1704 (s), 1732 (s), 2360 (w), 2983 (w) cm–1. C12H14O5 (238.24): H 3.24; found C 34.53, H 3.26.
calcd. C 60.50, H 5.92; found C 60.83, H 6.01.
2,6-Bis(hydroxymethyl)-4-nitrophenol (H3LNO ):
Nitrophenol
[10e]
2
2,6-Bis(hydroxymethyl)phenol (H3LH): Phenol II (640 mg,
2.69 mmol) was dissolved in THF (12 mL). Slowly, LiAlH4
(408 mg, 10.75 mmol) was added at 0 °C. The resulting grey sus-
pension was stirred overnight at room temperature. Et2O (10 mL)
and HCl (1 n, 15 mL) were then added slowly to the cold solution.
(10.5 g, 75 mmol) was added to a mixture of paraformaldehyde (11,
37.8 mmol), acetic acid (45 mL) and concentrated H2SO4 (23 mL).
The resulting suspension was heated at 50 °C for 20 h. Then, water
(60 mL) was added, and the mixture was neutralized by the slow
addition of ground K2CO3 (53 g, 0.38 mol), and stirring was main-
Eur. J. Inorg. Chem. 2014, 4326–4334
4332
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
tained for 2 h. The solid was collected by filtration under vacuum
and washed with cold water (50 mL), Et2O (70 mL) and CH2Cl2
(30 mL) and further recrystallized from EtOH (100 mL). Purifica-
then filtered. Dark brown blocks of 3 were obtained overnight and
dried in vacuo, yield 64% (based on Mn). [C103H123Cl7Mn19O51]Cl2·
13H2O·MeCN (3815.2): calcd. C 33.06, H 4.02, N 0.37; found C
tion by column chromatography (eluent CH2Cl2) gave 8-acetoxy- 33.09, H 4.19, N 0.32. Selected IR data (KBr): ν = 474 (w), 548
˜
methyl-6-nitro-1,3-benzodioxene (III; 15.8 g, 83%) as a white solid.
(m), 644 (s), 810 (m), 862 (w), 985 (m), 1024 (m), 1160 (m), 1226
(m), 1251 (m), 1384 (s), 1470 (s), 1633 (w), 2930 (m), 3370 (br, s)
cm–1.
Compound III was heated to reflux in a conc. HCl./water mixture
(60/90 mL) for 4 h to give the title nitro ligand H3LNO (12.40 g,
2
The same reaction in the presence of NaN3 afforded a brown solu-
tion, which failed to yield crystalline material for further analysis.
[MnIII12MnII7(μ4-O)8(μ3-N3)8(HLI)12(MeOH)4(H2O)2](O2CH)2·
99%) as a white solid. All analyses are in agreement with those
previously reported.[10e] 1H NMR ([D6]acetone): δ = 4.87 (s, 4 H),
5.03 (t, J = 5.0 Hz, 2 H), 8.12 (s, 2 H), 9.71 (s, 1 H) ppm. 13C
NMR ([D6]acetone): δ = 61.3, 122.4, 129.1, 141.4, 160.0 ppm.
HRMS: calcd. for C8H10NO5 [M + H]+ 200.05076; found.
200.05071.
16MeOH·10MeCN (4):
A slurry of MnCl2·4H2O (0.04 g,
0.2 mmol), NaN3 (0.039 g, 0.6 mmol), NaO2CMe·3H2O (0.021 g,
0.15 mmol) and H3LI (0.14 g, 0.5 mmol) in MeCN (15 mL) and
MeOH (3 mL) was stirred at room temperature for 1.5 h and then
heated at reflux for 2.5 h, after which the dark brown reaction mix-
ture was slowly cooled and filtered. Dark brown crystals of the title
cluster were obtained after 2 d, washed with a small amount of
MeCN and dried in air over several days, yield 25% (based on
Mn). [C102H204I12Mn19N24O50](O2CH)2·5MeCN·5MeOH (5589.0):
calcd. C 26.02; H 2.66; N 7.39; found C 26.05; H 2.51; N 7.23.
2,6-Bis(hydroxymethyl)-4-aminophenol (H3LNH ): In a sealed Paar
2
flask, Pd/C catalyst (10% w/w; 50 mg) was added to a solution of
H3LNO (405 mg, 2.03 mmol) in MeOH (20 mL). The resulting
2
black suspension was shaken overnight at room temp. under a H2
atmosphere (60 psi) and then filtered through paper. The resulting
solution was evaporated to dryness to yield the amino ligand
H3LNH (343 mg, 99%) as a brown crystalline solid. M.p. 108.5–
2
1
Selected IR (KBr pellet): ν = 527 (m), 612 (s), 636 (s), 761 (m), 881
˜
109.5 °C. H NMR (MeOD): δ = 4.63 (s, 4 H), 6.64 (s, 2 H) ppm.
13C NMR (MeOD): δ = 62.0, 116.3, 129.4, 140.1, 147.5 ppm. Se-
(m), 964 (w), 996 (m), 1012 (m), 1080 (w), 1199 (s) 1221 (m), 1264
(s), 1312 (m), 1384 (w), 1417 (w), 1464 (vs), 1501 (w), 1607 (m),
2060 (vs, N3), 2820 (m), 2926 (m), 3024 (m), 3060 (m), 3361 (s, br)
cm–1.
lected IR (neat): ν = 879 (m), 932 (m), 988 (m), 1066 (s), 1186 (s),
˜
1216 (s), 1273 (m), 1330 (m), 1464 (s), 1614 (m), 3280 (m), 3360
(m) cm–1. HRMS: calcd. for C8H11NO3 [M]+ 169.07387; found.
169.07389.
[MnIII12MnII7(μ4-O)8(μ3-Cl)7.7(μ3-OMe)0.3(HLSMe
)12(MeOH)6]Cl2·
2,6-Bis(hydroxymethyl)-4-(methylmercapto)phenol (H3LSMe): Form-
aldehyde (25 mL, 37% w/w in water) was added to a solution of
NaOH (5 g, 125 mmol) and 4-(methylmercapto)phenol (2.8 g,
20 mmol) in a hydroalcoholic mixture (16 mL, 38% methanol). The
resulting solution was stirred overnight at 50 °C. Glacial acetic acid
(4.4 mL) in water (8 mL) was then added dropwise at 0 °C to pH
5–6, and the resulting solution was concentrated in vacuo. The re-
sulting orange solid was dissolved in EtOAc (50 mL) and H2O
(30 mL). The organic phase was then dried with cotton wool to
afford a white solid, which was purified by flash chromatography
(eluent: CH2Cl2/MeOH, 100:0 to 95:5) to yield the methylmercapto
ligand H3LSMe (2.9 g, 73%) as a white solid. Rf = 0.36 (CH2Cl2/
27MeCN (5): A slurry of MnCl2·4H2O (0.06 g, 0.3 mmol), Et3N
(0.041 g, 0.4 mmol) and H3LSMe (0.06 g, 0.3 mmol) in MeCN
(15 mL) and MeOH (3 mL) was stirred for 2 h under ambient con-
ditions to afford a dark brown solution, which was then filtered.
Dark brown blocks of 5 were obtained overnight and dried in
vacuo, yield 64% (based on Mn). [C114.3H144.9Cl7.7Mn19O50.3S12]-
Cl2·7H2O·2.5MeCN (4324.8): calcd. C 32.99, H 3.85, N 0.75, S
8.91; found C 32.88, H 3.64, N 0.89, S 8.96. Selected IR data (KBr
pellet): ν = 550 (s), 622 (s), 674 (s), 773 (s), 875 (w), 1020 (s), 1109
˜
(w), 1244 (s), 1277 (m), 1446 (s), 1462 (s), 1587 (m), 2918 (m), 3207
(br, s) cm–1. The same reaction in the presence of NaN3 afforded a
brown solution, which failed to yield crystalline material for further
analysis.
1
MeOH 95:5), m.p. 90–91 °C. H NMR ([D6]acetone): δ = 2.40 (s,
3 H), 4.60 (t, J = 5.4 Hz, 2 H), 4.75 (d, J = 5.6 Hz, 4 H), 7.15 (s,
X-ray Data Collection and Structure Refinement: The data for 3 and
4 were collected at 180(2) K with a Stoe IPDS II diffractometer by
using graphite-monochromated Mo-Kα radiation. Crystals of 5
were invariably small and weakly-diffracting, and the dataset was
recorded at 150(2) K with a Bruker SMART Apex diffractometer
on the SCD beamline of the ANKA synchrotron source, Karlsruhe,
Germany with Si-monochromated radiation with λ = 0.8000 Å.
The data were corrected for absorption, the structures were solved
by direct methods (SHELXTL), and full-matrix least-squares re-
finement against F2 (all data) was performed with SHELXL-
2013.[13] All ordered non-H atoms were refined anisotropically. The
face-bridging ligands on the threefold axes in 3 and 5 were a disor-
dered superposition of chloride and methoxo ligands, and these
were assigned relative occupancies of 50:50 and 85:15, respectively.
Disordered lattice solvent molecules and counterions that could
not be refined satisfactorily with partial occupancies were handled
by using the SQUEEZE option in PLATON.[14] The crystallo-
graphic data and structure refinement details for 3–5 are listed in
Table 3.
2 H), 8.53 (s, 1 H) ppm. 13C NMR ([D6]acetone): δ = 17.9, 61.84,
61.97, 127.6, 128.9, 153.1 ppm. Selected IR (neat): ν = 875 (s), 1005
˜
(s), 1063 (s), 1205 (s), 1262 (m), 1337 (m), 1423 (s), 1435 (s), 1455
(s), 1476 (s), 3280 (m), 3368 (m) cm–1. HRMS: calcd. for C9H12O3S
[M]+ 200.05071; found. 200.05076.
Synthesis of Complexes 3–5
General Details: All reactions were performed under aerobic condi-
tions by using the ligands described above and commercially avail-
able reagents, which were used as received without further purifica-
tion. Elemental analyses (CHN) were performed with a Vario EL
elemental analyzer. FTIR spectra were recorded with a Perkin–El-
mer Spectrum One spectrometer with samples prepared as KBr
pellets.
Caution! Although no such tendency was observed during the present
work, azides (N3) are potentially explosive and should be handled
with care and in small quantities.
[MnIII12MnII7(μ4-O)8(μ3-Cl)7(μ3-OMe)(HLH)12(MeOH)6]Cl2·
16H2O·10MeOH·MeCN (3): A slurry of MnCl2·4H2O (0.1 g, CCDC-953780 (for 3), -953782 (for 4) and -922474 (for 5) contain
0.5 mmol), Et3N (0.08 g, 0.8 mmol) and H3LH (0.077 g, 0.5 mmol)
in MeCN (15 mL) and MeOH (3 mL) was stirred for 2 h under
ambient conditions to afford a dark brown solution, which was
the supplementary 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.
Eur. J. Inorg. Chem. 2014, 4326–4334
4333
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 3. Crystallographic data for 3–5.
Mn19H (3)
115H198Cl9Mn19NO77
Mn19I (4)
Mn19SMe (5)
Formula
C
C140H204I12Mn19N34O70
C168.3H225.9Cl9.7Mn19N27O50.3S12
Mr
4189.65
trigonal
R3
180(2)
20.3770(8)
20.3770(8)
37.263(2)
90
6050.02
monoclinic
P21/n
5204.51
cubic
Crystal system
Space group
T [K]
¯
¯
Pa3
180(2)
150(2)
27.5286(11)
27.5286(11)
27.5286(11)
90
a [Å]
b [Å]
23.3872(12)
16.8363(5)
26.1412(13)
90
c [Å]
α [°]
β [°]
90
120
13399.6(10)
3
1.558
94.774(4)
90
10257.5(8)
2
1.959
3.019
90
90
γ [°]
V [Å3]
20861.8(14)
4
1.659
Z
ρcalcd. [gcm–3]
μ [mm–1]
1.510
1.927
F(000)
λ [Å]
6417
5906
0.71073
58795
18779
0.0633
11301
1127/32
0.918
0.1324
0.0540
+1.60/–1.02
10660
0.8000
91393
7396
0.0344
6087
292/21
1.108
0.2059
0.0603
+1.03/–0.52
0.71073
34489
6320
0.0656
5044
340/14
0.989
0.1015
0.0484
+0.40/–0.45
Reflections collected
Unique data
Rint
Data with IϾ2σ(I)
Parameters/restraints
S on F2 (all data)
wR2 (all data)
R1 [IϾ2σ(I)]
Largest residuals [eÅ–3]
[5] T. Glaser, I. Liratzis, A. M. Ako, A. K. Powell, Coord. Chem.
Rev. 2009, 253, 2296–2305, and references cited therein.
[6] C. Lampropoulos, K. A. Abboud, T. C. Stamatatos, G.
Christou, Inorg. Chem. 2008, 47, 813–815.
Magnetic Measurements: The magnetic susceptibility measure-
ments were obtained with a Quantum Design MPMS-XL super-
conducting quantum interference device (SQUID) magnetometer
over the temperature range 1.8–300 K first with a dc field of
1000 Oe and then in zero dc field with an oscillating ac field of
3 Oe and an ac frequency of 200 Hz. The magnetization measure-
ments were made over a temperature range of 1.8–300 K with dc
applied fields from 0 to 70 kOe. The measurements of M vs. H at
100 K were additionally used to check for the presence of ferromag-
netic impurities, which were found to be absent. All measurements
were performed on fresh finely ground crystalline samples re-
strained in grease, and the magnetic data were corrected for the
sample holder.
[7] e.g.M. Mannini, P. Sainctavit, R. Sessoli, C. Cartier dit Moulin,
F. Pineider, M.-A. Arrio, A. Cornia, D. Gatteschi, Chem. Eur.
J. 2008, 14, 7530–7535, and references cited therein.
[8] a) M. Evangelisti, E. K. Brechin, Dalton Trans. 2010, 39, 4672–
4676, and references cited therein; b) S. Nayak, M. Evangelisti,
A. K. Powell, J. Reedijk, Chem. Eur. J. 2010, 16, 12865–12872.
[9] a) Y.-Z. Zheng, M. Evangelisti, R. E. P. Winpenny, Angew.
Chem. Int. Ed. 2011, 50, 3692–3695; Angew. Chem. 2011, 123,
3776; b) Y. Wang, W. Li, S. Zhou, D. Kong, H. Yang, L. Wu,
Chem. Commun. 2011, 47, 3541–3543; V. Ozdemir, B. Williams-
Jones, S. J. Glatt, M. T. Tsuang, J. B. Lohr, C. Reist, Nature
Biotechnol. 2006, 24, 942–946.
[10] a) A. A. Moshfegh, B. Mazandarini, A. Nahid, G. H. Hakimel-
ahi, Helv. Chim. Acta 1982, 65, 1229–1232; b) R. S. Drago,
M. J. Desmond, B. B. Corden, K. A. Miller, J. Am. Chem. Soc.
1983, 105, 2287–2296; c) P. D. Hampton, Z. Bencze, W. Tong,
C. E. Daitch, J. Org. Chem. 1994, 59, 4838–4843; d) G. T.
Crisp, P. D. Turner, Tetrahedron 2000, 56, 407–415; e) J.
De Mendoza, P. M. Nieto, P. Prados, C. Sánchez, Tetrahedron
1990, 46, 671–682; f) T. Routasalo, J. Helaja, J. Kavakka,
A. M. P. Koskinen, Eur. J. Org. Chem. 2008, 3190–3199.
[11] C. Hansch, A. Leo, Substituent Constants for Correlation
Analysis in Chemistry and Biology, Wiley-Interscience, New
York, 1979.
Acknowledgments
This work was supported by the Deutsche Forschungsgemeinschaft
(DFG) (Center for Functional Nanostructures) and Centre
National de la Recherche Scientifique (CNRS). The authors
acknowledge the Synchrotron Light Source ANKA for provision
of instrumentation on the SCD beamline. Dr. Gernot Buth is
thanked for experimental assistance. S. M. is thankful to Alain
Hazemann of the IUT Robert Schuman (Unistra) for elemental
analyses.
[12] A. M. Ako, B. Burger, Y. Lan, V. Mereacre, R. Clérac, G. Buth,
S. Gómez-Coca, E. Ruiz, C. E. Anson, A. K. Powell, Inorg.
Chem. 2013, 52, 5764–5774.
[13] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[14] P. van der Sluis, A. L. Spek, Acta Crystallogr., Sect. A 1990,
46, 194–201.
[1] A. M. Ako, I. J. Hewitt, V. Mereacre, R. Clérac, W.
Wernsdorfer, C. E. Anson, A. K. Powell, Angew. Chem. Int. Ed.
2006, 45, 4926–4929; Angew. Chem. 2006, 118, 5048–5051.
[2] O. Waldmann, A. M. Ako, H. U. Güdel, A. K. Powell, Inorg.
Chem. 2008, 47, 3486–3488.
[3] E. Ruiz, T. Cauchy, J. Cano, R. Costa, J. Tercero, S. Alvarez,
J. Am. Chem. Soc. 2008, 130, 7420–7426.
[4] a) A. M. Ako, M. S. Alam, S. Mameri, Y. Lan, M. Hibert, M.
Stocker, P. Müller, C. E. Anson, A. K. Powell, Eur. J. Inorg.
Chem. 2012, 4131–4140.
Received: May 15, 2014
Published Online: July 24, 2014
Eur. J. Inorg. Chem. 2014, 4326–4334
4334
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim