Family of MnIII4LnIII2(LnIII?=?SmIII, GdIII, DyIII) coordination clusters: Experimental and theoretical investigations

Article abstract of DOI:10.1016/j.poly.2016.08.050

The present work introduces a family of Mn^III₄Ln^III₂(Ln^III?=?Sm^III, Gd^III, Dy^III) coordination clusters having a multisite hydroxyl rich ligand, 3-[(2-hydroxy-3-methoxy-benzylidene)-amino]-propane-1,2-diol (H₃Vapd), namely [Mn^III₄Sm^III₂(Vapd)₄(OAc)₆]·4H₂O (1), [Mn^III₄Gd^III₂(Vapd)₄(OAc)₆]·4H₂O (2) and [Mn^III₄Dy^III₂(Vapd)₄(OAc)₆]·4H₂O (3). The Sm^IIIanalog is the first example of a Mn₄Ln₂species reported to date. The similarities and differences in terms of structure, topology and magnetic behaviors within the series are investigated extensively. DFT computations were carried out for Mn^III₄Gd^III₂to address the experimentally challenging questions regarding the nature of the magnetic interactions in this cluster.

Full text of DOI:10.1016/j.poly.2016.08.050

Accepted Manuscript
Family of MnIII 4LnIII 2 (LnIII= SmIII, GdIII, DyIII) coordination clusters: experi-
mental and theoretical investigations
Ritwik Modak, Yeasin Sikdar, Alina Bieńko, Maciej Witwicki, Maria
Jerzykiewicz, Sanchita Goswami
PII:
S0277-5387(16)30419-3
DOI:
http://dx.doi.org/10.1016/j.poly.2016.08.050
POLY 12187
Reference:
Polyhedron
To appear in:
Received Date:
Revised Date:
Accepted Date:
29 May 2016
12 August 2016
16 August 2016
Please cite this article as: R. Modak, Y. Sikdar, A. Bieńko, M. Witwicki, M. Jerzykiewicz, S. Goswami, Family of
MnIII 4LnIII (LnIII= SmIII, GdIII, DyIII) coordination clusters: experimental and theoretical investigations,
2
Polyhedron (2016), doi: http://dx.doi.org/10.1016/j.poly.2016.08.050
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1
2
3
4
5
6
7
Family of MnIII LnIII (LnIII= SmIII, GdIII, DyIII) coordination
clusters: experimental and theoretical investigations
4
2
8
9
Ritwik Modak,a Yeasin Sikdar,a Alina Bieńko,b Maciej Witwicki,b Maria Jerzykiewicz,bSanchita
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Goswami*a
a Department of Chemistry, University of Calcutta, 92, A. P. C. Road, Kolkata – 700009, India
E-mail: sgchem@caluniv.ac.in; Fax: +91 033 2350 9755; Tel: +91 033 2350 8386
bFaculty of Chemistry, University of Wroclaw, 14 F. Joliot - Curie, 50- 383 Wroclaw, Poland
ABSTRACT
The present work introduces a family of MnIII LnIII (LnIII = SmIII, GdIII, DyIII) coordination
4
2
clusters having a multisite hydroxyl rich ligand, 3-[(2-hydroxy-3-methoxy-benzylidene)-amino]-
propane-1,2-diol (H3Vapd), namely (1),
[MnIII SmIII (Vapd)4(OAc)6]·4H2O
[MnIII GdIII (Vapd)4(OAc)6]·4H2O (2) and [MnIII DyIII (Vapd)4(OAc)6]·4H2O (3). The SmIII
4
2
4
2
4
2
analogue is the first example of a Mn4Ln2 species reported to date. The similarities and
differences in terms of structure, topology and magnetic behaviours within the series are
investigated extensively. DFT computations were carried out for MnIII GdIII to address the
4
2
experimentally challenging questions regarding the nature of the magnetic interactions in this
cluster.
Keywords: Manganese, Lanthanides, Topology, Magnetic properties, Density functional
calculations
* Corresponding Author Address: E-mail: sgchem@caluniv.ac.in,sgoswamicu@gmail.com; Fax:
+91 033 2350 9755; Tel: +91 033 2350 8386
1. Introduction
The field of 3d-4f chemistry is broadening due to the remarkable physical and chemical
properties associated with this class of materials [1], like magnetism [2-4], luminescence [5,6],
gas adsorption [7] and catalysis [8], as well as their intriguing architectures. The 3d transition
metal ions exhibit a wide variety of physical and chemical properties by virtue of their diverse 3d
electrons. Moreover, the incorporation of LnIII ions possesses promising prospects for interesting
magnetic phenomena because of their inherent large amounts of anisotropy. Thus, the idea of
coupling LnIII centers with 3d metal ions opens up a promising avenue for the construction of
novel functional materials, boosted by the tremendous development of molecular magnetism [9].

1
2
3
4
5
6
7
8
9
The discovery of the first example of a compound showing single molecule magnet behavior
happens to be a (MnIII/MnIV) species, [Mn12O12(OAc)16(H2O)4] [10]. Thus, MnII/MnIII/MnIV
centers together with the power of LnIII ions, i.e. the Mn/Ln subclass of 3d−4f clusters, continues
to reside at the center of attraction as they are projected as components of magnetic spintronic
devices based on single molecule magnets [9]. A variety of structural types, compositions and
topologies have been identified so far for Mn−Ln clusters, of which high-nuclearity clusters
occupy an appreciable volume: MnIII LnIII [11a–11e], MnIV2LnIII [11f], MnIII LnIII [11g],
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
2
2
2
2
3
MnIII LnIII [11h], MnIII LnIII [9f,11i,j], MnII MnIII LnIII [11k,l], MnIII LnIII [11m-11o],
3
4
2
2
2
2
4
4
MnII MnIII LnIII [11p], MnIVMnIII LnIII [11q], MnIVMnIII LnIII [11r], MnIII LnIII [11s],
2
2
6
4
4
4
6
5
8
MnIII LnIII [11t,u], MnIV MnIII LnIII [11v], MnIII LnIV [11w], MnIII LnIII [11x], MnIII10LnIII
6
2
2
4
6
7
3
9
8
2
[11y], MnII MnIII LnIII2 [11z], MnIII11LnIII4 [11y,12a], MnIII11LnIV5 [11w], MnIIMnIII11LnIII [12b],
2
9
MnIII12LnIII6 [12c], MnIII12MnII LnIII [12d], MnIV MnIII18LnIII [12e]. Interestingly, low-nuclearity
6
3
aggregates are quite scarce and there are only a few reports of hexanuclear Mn4Ln2 complexes
[9f,11i-11l]. We envisage that further examples of Mn-Ln compounds would help to improve the
experimental and theoretical realization of the complicated magnetic exchange interactions
involving transition metals and lanthanide ions.
Here, we report an experimental demonstration of lanthanide contraction within members of
an alternately varied series of MnIII LnIII (LnIII = SmIII, GdIII, DyIII) coordination clusters:
4
2
[Mn4Ln2(Vapd)4(OAc)6]·4H2O [LnIII = SmIII (1), GdIII (2), DyIII (3) and H3Vapd = 3-[(2-hydroxy-
3-methoxy-benzylidene)-amino]-propane-1,2-diol]. The relative consequence of lanthanide
contraction has been assessed to clarify the fundamental chemistry of lanthanides. In addition,
their magnetic behaviors are investigated in detail. The concerted application of experimental
magnetism and theoretical consideration became necessary to define the magnetic behavior of
the MnIII LnIII series. We believe that the present investigations are significant because these
4
2
results can extend the knowledge base regarding prevailing interactions within MnIII-LnIII cores.
2. Result and Discussion
2.1. Syntheses and general characterization
The course of choosing ligands of the appropriate design has been demonstrated to be an
effective strategy for the generation of 3d-4f clusters. In our previous works on the construction
of homo and heterometallic 3d and 3d-4f clusters, NiII , MnIII NiII , CoIICoIII, CoII CoIII,
4
2
2
2
CuII DyIII, CuII DyIII3 [13a,b,4], we have used the multisite hydroxyl rich ligands 3-[(2-hydroxy-
4
6
benzylidene)-amino]-propane-1,2-diol
amino]-propane-1,2-diol (H3Vapd),
(H3Sapd),
3-[(2-hydroxy-3-methoxy-benzylidene)-
2-[(3-hydroxy2-[(3-hydroxy-propylimino)-methyl]-6-
methoxy-phenol (H2Vap) and 2-[(3-hydroxy-ethylimino)-methyl]-6-methoxy-phenol (H2Vae).
Indeed, a couple of groups have also employed these ligands to design promising homo and
heterometallic platforms [17c-17g,18,20b]. The ligand H3Vapd contains an O4N donor set with
five potential incorporable sites (one methoxo, one phenoxo, one imino and two alkoxo donor
atoms) admitting various coordination pockets with their different bindings modes to the 3d and

1
2
3
4
5
6
7
8
9
4f metal ions (shown in Scheme S1 and Schemes S2-S4 in the Supplementary Information). The
versatile ligand H3Vapd was prepared by a simple Schiff base condensation reaction using 3-
amino-1,2-propanediol and o-vanillin under refluxing conditions in ethanol. It is based on a basic
phenol-framework and contains two unsymmetrically disposed substituents. Adjacent to the
phenol unit, on one side is a 1,2 propanediol amine group and on the other side a methoxy group.
The syntheses of the complexes are similar, with the self-assembly of H3Vapd, Mn(OAc)2·4H2O,
and Ln(OAc)3·xH2O (in a 2:2:1 stoichiometric ratio) in the presence of triethylamine as a base, in
methanol, as in eq 1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
4Mn(OAc)2 + 2Ln(OAc)3 + 4H3Vapd + 8Et3N + O2 = [Mn4Ln2(Vapd)4(OAc)6] + 8Et3NH(OAc)
+ 2H2O
(1)
Dark brown block shaped single crystals were obtained by the slow diffusion of diethyl
ether into a methanolic solution of the complexes, with yields of 34-37 %, based on the ligand
(Scheme 1, see also, Experimental Section). It should be mentioned that NEt3 not only acts as a
proton acceptor, but also provides the basic condition to facilitate the aerial oxidation of the MnII
ion to the MnIII ion. It is interesting to note that among the previously reported five Mn4Ln2
complexes,
hydroxyethoxy)ethyl)imino)methyl]-4-methylphenol), salicylhydroxamic acid and 2-(((2-
hydroxy-3-methoxyphenyl)methylene)amino)-2-(hydroxymethyl)-1,3-propanediol, while the
other two complexes [11k,l] used N,N,N',N'-tetrakis(2-
MnII MnIII LnIII
three
MnIII LnIII
complexes
[9f,11i,j]
used
2,6-bis[((2-(2-
4
2
2
2
2
hydroxyethyl)ethylenediamine and {1,1,1-tris(hydroxymethyl)ethane, triethanolamine containing
pivalic acid} as coordination ligands (Scheme 2). In this work we report the first MnIII LnIII
4
2
aggregate in which the manganese centers Mn1 and Mn2 (along with their symmetry
equivalents) are six and five coordinated respectively.
The experimental powder XRD patterns of the bulk crystalline material of 1-3 are in good
agreement with the simulated XRD patterns from the single crystal X-ray diffraction results,
confirming the purity of the bulk samples (Fig. S1-S3 in the Supplementary Information).
Furthermore, in order to investigate the purity of the bulk crystalline material, energy-dispersive
X-ray (EDX) analysis of 1-3 confirmed the homogeneity of our crystals and also showed the
presence of both Mn and Ln ions in the complexes, with Mn:Ln ratios of 69.59:30.41 (for 1,
where Ln = Sm), 68.16:31.84 (for 2 where Ln = Gd) and 66.03:33.97 (for 3 where Ln = Dy),
which corroborated with the results obtained from single crystal X-ray diffraction (shown in the
Supplementary Information, Figs. S4-S6). The IR spectra of complexes 1-3 are very similar and
display the characteristic bands of the coordinated ligand and the acetato groups (Fig. S7). The
broad IR absorption at ~3430 cm−1 is attributed to the O–H stretching frequency of the H2O
ligands. A diagnostic sharp absortion at ~1643 cm-1 is due to the C=N stretching of the Schiff
base ligand, which is shifted to ~1619 cm−1 owing to complexation. In addition, the absorption at
~1225 cm−1 is assigned to the C−Ophen bond, which is shifted to ~1250 cm−1 in the metal
complexes, as the ligand coordinates in its deprotonated form. Due to the presence of both

1
2
3
4
5
6
7
8
9
asymmetric (η2: η1: µ2) and symmetric (syn-syn η1: η1: µ2) acetato bridging, two distinct (CO)
stretching frequency absorption bands at ~1560 and ~1438 cm−1 are observed, respectively[15].
ESI–MS spectra of complexes 1-3 were recorded in MeOH medium to explore the principal
species present in the solution (see Fig. S8-S10 in the Supplementary Information). All the
complexes show a similar kind of behavior by signalling the molecular ion peak at 525.027 and
555.037 a.m.u., which corresponds to C21H22N2O7Mn2, formulated as [Mn2(Sapd)(Vapd)+H+]+
(where the generation of Sapd3- originates from the loss of the –OMe group from Vapd3-), and
C22H24N2O8Mn2 for [Mn2(Vapd)2+H+]+, respectively. In the case of the Mn4Sm2 cluster (1), the
peaks at 825.718, 882.777 and 903.761 a.m.u. for [Mn4Sm2(Vapd)4(OAc)4]2+,
[Mn4Sm2(Vapd)4(OAc)6+ 2H+]2+ and [Mn4Sm2(Vapd)4(OAc)6+2Na+]2+ respectively, prove the
molecular integrity in the solution phase. Complex 2 was observed in the ESI–MS peak with the
loss of the –OAc group, with peaks appearing at 802.007 and 832.077 a.m.u., which correspond
to [Mn4Gd2(Vapd)4(OAc)3+H−]2+ and [Mn4Gd2(Vapd)4(OAc)4]2+ respectively. For complex 3 the
peaks at 806.032 and 735.974 a.m.u. well corroborate with the species
[Mn4Dy2(Vapd)4(OAc)3+H−]2+ and [Mn3Dy2(Vapd)3(OAc)5(H2O)+ H+]2+ respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
2.2. Crystal structure descriptions
Single-crystal X-ray diffraction measurements revealed that compounds 1-3 crystallize in the
triclinic space group with Z =1. The structures consist of heterometallic, hexanuclear Mn4Ln2
[LnIII = SmIII (1), GdIII (2) and DyIII (3)] clusters with the general formula
[MnIII LnIII (Vapd)4(OAc)6]·4H2O and with the asymmetric unit (ASU) containing half of the
4
2
complex, which lies upon an inversion center (see Figs. 1 and 2). The two different binding
modes of the tri-anionic ligand (Vapd3-) and the acetato group in the compounds have been
shown in Fig. 3. The oxidation states of the manganese ions and the protonation levels on the
coordinated oxygen atom of H3Vapd in the complexes were assigned by inspection of the metric
parameters, charge-balance considerations, Jahn–Teller elongation axes and by bond valence
sum analysis (see Tables S1 and S2 in the Supplementary Information) [16a].
Complexes 1 and 2 are isomorphs and therefore only the crystal structures of complex 1, i.e.
the Mn4Sm2 analogue, and complex 3 will be described here. Selected bond distances and bond
angles of all the complexes are summarized in the Supplementary Information, Table S3.
A close scrutiny reveals that the [MnIII SmIII ]18+ core of 1 can be described as being
4
2
III
composed of two parts; (i) a central Mn2 Sm2III core displaying a planar butterfly type topology
(comprising Mn1, Mn1#, Sm1 and Sm1#). The two SmIII ions are found at the “body” sites of the
butterfly, whereas the MnIII ions occupy the outer “wing” positions. Two µ3-alkoxo groups
bridge the two SmIII ions to each MnIII ion and (ii) two isolated MnIII ions (Mn2 and Mn2#)
capping the lanthanides from opposite sides. These are attached to the SmIII ions by means of two
µ2-alkoxo oxygen atoms and are located ~ 2.852 Å above and below the MnIII SmIII plane. The
2
2
overall core of 1 can be perceived as that of a spacecraft (Fig. 4). The central MnIII SmIII unit
2
2
forms the body of the spacecraft and the two capping MnIII centers constitute the wings.

1
2
3
4
5
6
7
8
9
The peripheral ligation is provided by two η1:η1:η2:η3:µ4 ligands (Vapd3-) and two
η1:η1:η2:η2:µ3 ligands (Vapd3-), the two rhombuses enjoy a second contact between the Mn1 and
Sm1# atoms (Mn1# and Sm1) from η2:η1:µ2acetato bridging. This bridging pattern is reflected in
the metal-metal distances of each rhombus.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
The Mn(1) and Mn(1#) ions are in six coordinated O5N1 distorted octahedral environments
(deviation of 0.808 from the ideal Oh symmetry, out of the range 0.1-3, CShM value) as pointed
out by CShM calculations [16b,c], with the Jahn–Teller axis µ2-OOAc….Mn(1/1#).…µ-OOAc
expected for d4 high spin MnIII ions (Fig. 5 and Table S4 in the Supplementary Information).
Applying the nomenclature developed by Stiefel and Brown, detailed geometrical calculations
reveal that the azimuthal expansion in the distorted octahedral geometry occurs with an average
azimuthal angle Ø of 65.2°, which is larger than the ideal value (Ø = 60°) [16d], further
supporting the J−T distortion characteristic of the d4 MnIII system (see Supplementary
Information, Fig. S12 and Table S5). The equatorial positions are occupied by O3N donor atoms,
contributed by the ligand phenoxo oxygen, imine nitrogen, µ3-alkoxo and µ2-alkoxo atoms. The
axial positions are occupied by one η2:η1:µ2 and one η1:η1:µ2 acetato group. In contrast, the
Mn(2) and Mn(2#) ions are five coordinated, adopting a distorted trigonal bipyramid geometry,
(deviation of 1.367 from the ideal D3h symmetry, out of the range 0.1-3, CShM value) with an
O4N environment, as depicted by CShM analysis, which is also supported by the Addison
parameter (τ) = 0.617 (Fig. 5) [16e]. The equatorial positions are occupied by O2N donor atoms
from one η1:η1:η2:η3:µ4 ligand (µ2-alkoxo), one η1:η1:η2:η2:µ3 ligand (imino nitrogen atom) and
one η1:η1:µ2acetato group. The axial positions are occupied by two oxygen atoms from the same
η1:η1:η2:η2:µ3 ligand (phenoxo and µ2–alkoxo). The average Mn–O and Mn–N bond distances in
the TBP and distorted octahedral coordination environments are 1.935(6), 2.030(6) and 1.987(7),
2.042(7) Å, respectively.
Each SmIII ion adopts a nine coordinated coordination polyhedron and can be best described
as a distorted spherical capped square antiprism with a minimum deviation of 1.938 (out of the
range 0.1-3, CShM value) from the ideal C4v symmetry (Fig. 5) [13b,c]. The analysis of its
coordination geometry with SHAPE is shown in the Supporting Information Table S4. The
coordination sphere of each SmIII ion is formed by two oxygen atoms from one η1:η1:η2:η2:µ3
ligand, three oxygen atoms from two η1:η1:η2:η3:µ4 ligands, two oxygen atoms from one η1:η2:µ2
acetato group and remaining two oxygen atoms from two η1:η1:µ2 acetato groups. The Sm–Oalk
and Sm–OOAc bond distances are in the ranges 2.363(6)-2.801(5) and 2.414(6)-2.555(6) Å
respectively.
The intracluster SmIII····SmIII and average SmIII····MnIII distances are 4.485(17) and
3.535(17) Å respectively in the hexanuclear core.
2.3. Differences between the solid state structures of complexes 1/2 with 3

1
2
3
4
5
6
7
8
9
The partially labelled structure of complex 3 is presented in Figure 4. The structure of 3 is to
some extent superimposable with that of 1 and 2. The coordination environment of the cores of
1/2 and 3 differ significantly: (i) In 3, the ligand Vapd3- binds in the η1:η1:η2:η2:µ4 mode instead
of the η1:η1:η2:η3:µ4 mode (ii) The coordination sites are filled with four alkoxo oxygen atoms
[Dy–Oalk = 2.310(3)-2.466(3) Å] and four acetato ions [Dy–OOAc = 2.352(4)-2.510(4) Å]. Here,
the Dy1 ···· O4 (Fig. 4) distance is 2.976 Å, which is borderline to be considered as a Dy–O
bond. Again, we have performed CShM calculations based on both eight (no Dy–O bond) and
nine (considering Dy–O bond) coordinated DyIII ions, but much higher values are observed for
the nine coordinated species (Table S4 in the Supplementary Information). Therefore, we have
considered the DyIII ion as possessing the typical eight-coordinate geometry without the above
mentioned Dy····O contact as judged by the reported Dy-O distances and CShM calculations.
The CShM results point towards a biaugmented trigonal prism arrangement with a minimum
deviation of 1.539 from the ideal C2v symmetry, out of the range 0.1-3, for the DyIII center (Fig. 5
and Table S4, in the Supplementary Information) [16b,c]. (iii) In 3, two symmetry related DyIII
and MnIII ions are arranged in a rhombus with two isolated MnIII ions capping from opposite
directions, but it is marked by the absence of two µ3–alkoxo bridges (Fig. 4), where the
intracluster DyIII….DyIII and average DyIII….MnIII distances are 4.616(4) and 3.522(8) Å
respectively in the hexanuclear core. There is no previous example of such a discrete unit in 3d-
4f chemistry.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
The molecules of 1, 2 and 3 are well separated within the crystal packing and no
intermolecular hydrogen bonds are observed in the crystal packing (Fig. S13 in the
Supplementary Information).
2.4. The effect of lanthanide contraction
The hypothesis underlying this work was that, by varying the LnIII ions along the most
crucial region of the lanthanide series, we might be able to visualize the direct
consequence of the lanthanide contraction on the structural outcome. Indeed, a minute
balance of lanthanide contraction affects the structural output of
1, 2 and 3. This
comparison is better quantitatively described by examination of the metrical data
provided in Table 1 which compares the Ln–O and Ln–Mn bond distances. The Ln–O(4)
bond distances generally decrease in the order
1
>
2
>
3
, in accord with the lanthanide
. The
contraction, and as a result the Ln1#–O(4) distances vary in the order
1
< 2 < 3
consequence of the lanthanide contraction is that the smaller DyIII ion has significantly
shorter Ln1–O4 distance, resulting in a stretched Ln1#–O4 separation of 2.976(4) Å. A
quick examination of Fig. 6 and Table 1 demonstrates that with decreasing Ln1–O4 bond
distance from SmIII to DyIII, the Ln1#–O4 distance difference increases and finally in the
dysprosium analogue (
O(4) connectivity is broken, resulting in a rhombus structural motif instead of a butterfly
motif. As expected, the Ln1–Ln1# distances follow the order
. The Ln1–Mn1,
Ln1–Mn1# and Ln1–Mn2 distances are also in accord with the lanthanide contraction.
3
), the O4 atom lies farthest from the Dy1# atom and the Dy1#–
1
< 2 < 3

1
2
3
4
5
6
2.5. Topological features of the hexanuclear core
7
8
9
To better appreciate the structural features of 1, 2 and 3, we have adopted the topological
approach established by Kostakis and co-workers [17a–17e]. Applying this method, the
centrosymmetric hexanuclear [MnIII SmIII ] and [MnIII GdIII ] cores can be represented as two
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
4
2
4
2
fused triangles comprising two LnIII and one MnIII nodes with two isolated MnIII nodes connected
to the LnIII centers (Fig. 7). Numerically, Mn1 is a 2–connected node, Ln1 is a 4–connected node
and Mn2 represents a 1–connected node; thus the cluster is described by the symbol 1,2,4M6–1
[17e]. Following a similar approach, the [Mn4Dy2] core can be described as a rectangle
comprising two DyIII and two MnIII nodes with two isolated MnIII nodes attached to the
lanthanides. Here, Mn1 is a 2–connected node, DyIII is a 3–connected node and Mn2 is a 1–
connected node and the cluster is depicted by symbol 1,2,3M6–1. We performed a comparative
study of the [Mn4Ln2] cores reported in the literature. [MnIII LnIII (μ3–O)2(Hbeemp)2(OAc)8(μ3–
4
2
OMe)2(H2O)2]C8H17OH (LnIII = GdIII, TbIII, YIII) [11i] has a similar topology, but the isolated
MnIII ion caps both the LnIII and MnIII ions of the [Mn2Ln2] rhombus thereby representing a
2,3,4M6–1
cluster.
The
Pecoraro
group
reported
[Mn4Ln2]
cores
for
.
[MnIII LnIII (OAc)(NO3)2(shi)4(μ–O)(μ–OH)](C5H5N)8 (C3H7NO)3.H2O (LnIII = YIII, TbIII, DyIII,
4
2
HoIII)
as
metallacrown
compounds
with
the
1,2,5M6–1
topology
[11j].
[MnIII LnIII (H2L)2(HL)2(OAc)4(CH3O)2(CH3OH)4](ClO4)2·6CH3OH (LnIII= LaIII, NdIII) clusters
4
2
exhibit close topological features [9f]. Here, the MnIII nodes are connected, therefore, the symbol
becomes 1,3,3M6–1. Scrutinizing the [Mn4Ln2] cores with the MnII MnIII LnIII2 distribution, two
2
2
different series are found: [MnII MnIII LnIII (O2CBut)6(edteH2)2(NO3)2] (LnIII = GdIII, TbIII) [11k]
2
2
2
and [MnII MnIII LnIII (Piv)8(thme)2(H2tea)2] (LnIII = PrIII, NdIII, SmIII, EuIII, GdIII, TbIII, DyIII)
2
2
2
[11l] with the topological motifs 4,5M6–1 and 2,3,4M6–1 respectively. It is interesting to note
here that other than complexes 1 and 2, the 1,2,4M6–1 topological motif can be found in the
dodecanuclear compounds reported by Yang et al. [17f]. The 1,2,3M6–1 topological motif is the
first example of Mn4Ln2 clusters (Table S6) along with another Co4Gd2 [17g] cluster.
2.6. Comparison with other related complexes
The new compounds 1, 2 and 3 join a small family of Mn4Ln2 complexes reported in the
literature. First of all, it should be noted that not all the three lanthanides chosen yielded the same
structural types. When SmIII and GdIII (larger and lighter among the series) were chosen as the
starting trivalent cation (1 and 2), a μ3-alkoxo bridged cluster was generated, whereas the DyIII
(3) analogue was devoid of any μ3–alkoxo bridging as a direct consequence originating from the
minute balancing of the lanthanide contraction effect, as discussed in the earlier section. It is
worth noting that the cores of 1 and 2 are very similar to two other Mn4Ln2 clusters reported
earlier [9f,11i]. In 2008, there was a report on compounds with a [MnIII LnIII ] core without a
4
2
systematic variation of the lanthanides (GdIII, TbIII and YIII) [11i], but the compounds were
isomorphous. In the hexanuclear core, the butterfly shaped MnIII LnIII fragment was supported
2
2

1
2
3
4
5
6
7
8
9
by μ2-alkoxo, μ3-methoxo and μ3-oxido groups. The remaining MnIII ions were linked to the
tetranuclear core by means of two acetato and a phenoxo oxygen atom, thereby capping the Mn–
Ln edges of the tetranuclear core (shown in Scheme 2). Here, all the MnIII ions were six
coordinated and the LnIII ions were nine coordinated with an O8N environment. Another
molecule that features a [MnIII LnIII ] core is reported by Ke et al. with close structural features
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
4
2
[9f]. The structures comprised a central MnIII LnIII planar rhombus with two MnIII ions located
2
2
above and below the rhombus (Scheme 2). The structure contains a μ3-methoxo bridge that links
the central MnIII ions. All the MnIII ions were six coordinated and the LnIII were nine
coordinated. An all oxygen ligating environment encompassed the LnIII centers. The metrical
parameters of these compounds are very similar to the cited compounds herein.
Interestingly, μ3-methoxo/μ3-oxo bridging is prevalent in connecting the MnIII centers with
the LnIII ions of the butterfly core in the previous reports; but in our case μ3-alkoxo bridges
contribute in linking the ions. The coordination environment around the MnIII ions differ from
the previously reported complexes in that two MnIII ions are hexacoordinated while two others
are penta coordinated. In this context it worth mentioning that Pecoraro and co-workers reported
a new metallacrown containing two LnIII and four MnIII ions and a partially systematic series was
studied [11j].
Recently two reports on new [Mn4Ln2] cores appeared with the [MnII MnIII LnIII ]
2
2
2
distribution [11k,l]. The molecule [MnII MnIII LnIII (O2CBut)6(edteH2)2(NO3)2], made by the
2
2
2
Christou group, has a face-fused, double cubane structure comprising four Mn and two Ln atoms
bridged by two μ4–oxo and four μ3–alkoxo groups [11k]. In
[MnII MnIII LnIII (Piv)8(thme)2(H2tea)2], four face sharing defective cubanes are held together by
2
2
2
alkoxo oxygen atoms with different bridging modes [11l].
Our series clearly demonstrates how the identity of the LnIII ion influences the structural
outcome of the [Mn4Ln2] core. The planar butterfly type orientation of the central MnIII LnIII
2
2
core of 1 and 2 contrasts markedly with that observed for 3. The Ln1–O4 bond length steadily
decreases from SmIII to DyIII, following the standard lanthanide contraction, indicating that the
LnIII size consistently affects the Mn2Ln2 central core. The presence of a smaller DyIII ion in 3
leads to stretching of the Ln1#–O4 bond to such an extent that it is broken.
To our knowledge, very few examples of hexanuclear 3d-4f complexes with a similar
topology have been reported in recent times. As far our knowledge goes, no example of
[MnIII LnIII ] aggregates including SmIII ions has been published to date.
4
2
In most of the series encountered in the literature, isostructural series are obtained, but we
have chosen alternate lanthanides to compare, thus providing a greater span, and this allowed for
the desired structural perturbation, highlighting the immediate effect of lanthanide contraction.
The assessment does not include variation of the MnIII centers and the ligand system. The
uniqueness of the reported series is twofold: (i) the μ3-alkoxo group holds the central core instead
of μ3-methoxo/ μ3-oxo groups. (ii) a clear demonstration of lanthanide contraction is given.
Finally, sufficient numbers of clusters having similar structures are required in order to frame a
complete structural correlation, and this work is underway in our laboratory.

1
2
3
4
5
6
2.7. Magnetic studies
7
8
9
The temperature dependence of the magnetic susceptibilities in the temperature range 2.0-300 K
under an external magnetic field of 0.5 T for the three complexes are shown in Fig. 8, as plots of
χMT vs. T and  vs. T. The field dependence of the magnetization up to 50000 Oe (5 T) at 2.0
K is shown in Fig. 9, as the plots of Mmol / (NAμB) vs. H.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
For complex 2, the χMT (where χM is the molar susceptibility per hexanuclear MnIII LnIII2 unit)
4
values decrease in the low temperature region. This data clearly indicates the presence of effects
favouring the formation of ground states possessing either low total spin S or low total atomic
angular momentum J (J = S + L) of the lanthanide ions [18a]. Such effects may result from spin-
orbit coupling or axial splitting of the ground states or/and antiferromagnetic exchange coupling
between two LnIII ions or MnIII and LnIII ions. This superexchange interaction should be weak
(compared with the crystal field splitting and spin-orbit coupling effects) because the magnetic f
orbitals of the lanthanide are better shielded than the d orbitals, leading to a low overlap between
the ligand and the f orbitals [18b].
For MnIII GdIII (2), the χMT value of 25.80 [cm3 K mol−1] at 300 K is close to the calculated
4
2
theoretical value of 27.74 [cm3 K mol−1] expected for four MnIII (S = 2) and two magnetically
8
isotropic GdIII ions possessing the S7/2 ground state (4f, J = 7/2, L = 0, S = 7/2), using the
expression [18a].
where
and
,
On lowering the temperature, χMT decreases very gradually (300-75 K) before a much bigger
decrease occurs below 75 K, reaching a value of 2.78 [cm3 K mol–1] at 0.5 T and 2 K. The molar
magnetic susceptibility of 2 increases on cooling down to T = 6.0 K, when it passes through a
maximum (Fig. 8).
The reciprocal magnetic susceptibilities in the temperature range of 50-300 K (Fig. S14)
follow the Curie-Weiss equation 1/χM = (T– Θ) / C with C = 26.4 [cm3 K mol−1] and Θ = −5.47
K. The small, negative value of the Weiss constant confirms a very weak antiferromagnetic
coupling, which can occur between the MnIII and GdIII ions or GdIII and GdIII ions.
Based on the structure, the χMT value of 2 can be viewed as contributions from a dimer of
GdIII ions (Gd1 and Gd1#) and four isolated MnIII ions (Mn1, Mn2, Mn1# and Mn2#). Therefore,

1
2
3
4
5
6
7
8
9
to analyze the magnetic susceptibility data of 2, we have employed an expression (see eq M1 in
the Supplementary Information; Table S7) which is the sum of a simple isotropic Heisenberg
dimer model of Gd [18c] with J describing the exchange interaction between the GdIII–GdIII ions
and the second term is a temperature dependent contribution from the MnIII ions (eqs M2 and M3
in the Supplementary Information) [18d]. The exchange interaction between the MnIII and GdIII
ions was introduced in terms of molecular field approximation (the MnIII–MnIII interaction was
omitted because of the much longer distance, ~ 6.694 Å, than the MnIII–GdIII distance of
3.522(9) Å or the GdIII–GdIII distance of 4.515(4) Å) (eq M4 in the Supplementary Information)
[18e]. The best agreement between the experimental and calculated results was obtained with J =
−0.16 cm−1, zJ’ = −0.09 cm–1, gMn = 2.02, gGd = 2.00 and the agreement factor, defined as
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
−5
is equal to 2.21×10 .
In comparison with literature data, similar weak GdIII····GdIII antiferromagnetic exchange
interactions and g values were found in a series of Schiff–base dinuclear GdIII complexes, such
GdIII (Salphen)3(H2O) (J = –0.082 cm−1) [18b], [GdIII (LA)(NO3)2]·4H2O (J = –0.106 cm−1)
2
2
[18f] or the well–known dinuclear GdIII carboxylato compounds with J values in the range from
0 to – 0.15 cm−1 [18b].
The reason for the small interaction between the paramagnetic centers results from the fact
that the 4f electrons are influenced very little by the surrounding environment [18h].
Theoretically, the GdIII ion, with a 4f7 electronic configuration, is very stable in energy and this
is responsible for the very difficult transfer of unpaired electrons to the d or s orbitals of the other
GdIII or MnIII ions.
The field dependence of the magnetization at low temperatures (Fig. 9) shows that the
magnetization increases smoothly with an increasing applied direct current (dc) field up to 2500
Oe, where a plateau exists, and then a rapid increase was observed without the expected
saturation value of 30 B.M. (M = 4gMnSMn + 2gGdSGd) for (gMn = 2.0). This behavior indicates the
presence of magnetic anisotropy and/or the lack of well–defined ground states, suggesting the
presence of low–lying excited states that might be populated when a field is applied [18i]. The
zero-field splitting effect causes a reduction of M to lower than 30, as shown in Fig. 9.
The magnetic behavior of the Mn4Sm2 complex 1 is completely different. As reported in Fig.
8, the value of χMT at room temperature is 11.94 [cm3 K mol−1], which is in reasonable agreement
with the expected value of 12.06 [cm3 K mol−1] for the four MnIII ions (S = 2) and two
6
independent SmIII ions possessing the H5/2 ground state (4f, g = 2/7, S = 5/2, L = 5, J = 5/2),
using the expression describe above. On lowering the temperature, the χMT value increased
gradually to reach a maximum of 12.471 cm3 Kmol−1 at 8.0 K and then decreased. The increase
in the higher temperature range indicates the presence of a ferromagnetic interaction, while the
decrease in the lower temperature region is ascribed to a weak intra and/or intermolecular
antiferromagnetic interaction.

1
2
3
4
5
6
7
8
9
The reciprocal magnetic susceptibilities in the temperature range 50-300 K follow the Curie-
Weiss equation 1/χM = (T – Θ) / C with C = 11.7 cm3 K mol−1 and Θ = + 0.18 K (Fig. S14). The
small, positive value of the Weiss constant confirms a very weak ferromagnetic coupling, which
can occur between the MnIII and SmIII ions or SmIII and SmIII ions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
The temperature dependence of the magnetic susceptibility of compound 1 can be calculated
by eq M5 in the Supplementary Information, as the sum of the independent contribution from
four MnIII ions (eq M3 in the Supplementary Information) and a dimer of SmIII ions. The average
exchange interaction was introduced in terms of a molecular field correction (eq M4 in the
Supplementary Information).
The magnetic susceptibility of the SmIII ion is much more complicated to describe because of
6
its H5/2 ground state, which splits into six states by spin-orbit coupling at room temperature,
which are not well separated from the excited ones (about 700 cm-1 from the ground state to the
first excited state), so both the crystal field effect and possible thermal population of the higher
energy states should be evaluated [18j]. As far as the existence of strong spin-orbit coupling for
the SmIII atoms is concerned, the SmIII dimer contribution for the magnetic data was analyzed by
the Van Vleck expression (eq M6 in the Supplementary Information) [18b].
Least-squares analysis of the magnetic data gives zJ’ = 0.52 cm−1, gMn = 2.01, gSm
=
2/7, λ = 198 cm−1 and R = 2.76×10−3. For the SmIII ions, λ–values are usually in the 160-
315 cm-1 range [18b], so our derived λ value is quite acceptable. Absolute zJ’ values for
LnIII complexes are commonly less than 1 cm-1 [18b,i] and our zJ’ estimate is
commensurate to this. It might also be noted that, as shown in the literature, the SmIII and
SmIII coupling usually has antiferromagnetic character [e.g. SmIII (Salphen)3(H2O) (zJ’ =
2
−0.97 cm−1) [18b] or {[SmIII (DPPA)2(μ2–C2O4)(H2O)2]·2H2O}n (J = −1.03 cm−1)] [18j],
2
however a ferromagnetic nature of the MnIII····LnIII coupling was found in MnIII GdIII ,
4
2
MnIII DyIII2 and MnIII TbIII2 complexes [11k].
4
4
The magnetization vs magnetic field plot for complex
1 at 2 K is shown in Fig. 9. The
expected value of saturation magnetization is 17.4 B.M. (M = 4gMnSMn + 2gSmSSm ) for
1
(gMn = 2.0). However, the observed saturation value at 2 K is only 12.5 B.M. This is due
6
to the crystal field effect on the SmIII ion that removes degeneration of the H5/2 (SmIII)
ground state and zero-field splitting at the MnIII or SmIII centers. The small differences
between the value of the magnetization with an increasing magnetic field and decreasing
magnetic field can be the result of torqueing effects of the sample.
The temperature dependence of the magnetic susceptibility of the Mn4Dy2 derivative 3 is
shown in Fig. 8. In the case of LnIII ions, except for GdIII, it is very difficult to analyze their
magnetic properties because of their orbital contribution. Usually, the 4fn configuration of a LnIII
ion is split into 2S+1LJ states by inter electronic repulsion and spin-orbit coupling. Further splitting
into Stark components is caused by crystal-field perturbation, which depends on the symmetry
site of the ion [18k]. Thus, the variable-temperature magnetic properties of a free ion of a rare
earth metal generally shows strong deviations from the Curie law and χMT decreases on cooling
because of the depopulation of the Stark levels.

1
2
3
4
5
6
7
8
9
However, as illustrated in Fig. 8, complex 3 obviously shows ferromagnetic coupling between
the DyIII ions and/or DyIII····MnIII ions. The χMT value at room temperature is 38.02 cm3 K mol-1,
which is a little bit lower than the value expected for two magnetically isolated DyIII ions
6
including the significant contribution of the 4f orbital (4f9, J = 15/2, S = 5/2, L = 5, H15/2) and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
four MnIII ions (S = 2). This can be a result of crystal field effects on the DyIII ion, which remove
the 16-fold degeneracy of the 6H15/2 ground state. As the temperature is lowered, the χMT product
increases very slowly to reach a maximum value of 40.39 cm3 K mol-1 at 5 K, and then decreases
abruptly and reaches 13.20 cm3 K mol-1 at 2 K, which is comparable with the result reported
previously [18j]. The increase in the higher temperature region and the positive value of Θ = 3.18
K indicate that there are ferromagnetic interactions between the DyIII····DyIII ions mediated by
an oxygen group and/or between DyIII····MnIII ions. Below 5 K, χMT decreases with cooling,
indicating the presence of possible antiferromagnetic interactions between the tetranuclear
molecules. The ligand field effects break the degeneration of the levels of the ground state
multiplet of the DyIII ion. So, the influence of the ligand field can be seen in the very low
temperature region where the theoretical model is unable to reproduce the experimental
susceptibility data [18j,l].
The field dependence of the magnetization (in Fig. 9) shows that, upon an increase of the
applied external magnetic applied field, the magnetization Mmol / (NAμB) decreases to zero, in
which the magnetization M/(NAμB) in the field decreasing mode is larger than the value in the
field increasing mode, due to a strong magnetic anisotropy.
The zero-field-cooling magnetization / field-cooling magnetization curves measured at 25 and
50 Oe coincide and thus confirm the absence of long–range ordering for all complexes. Their
course is identical with the temperature evolution of the dc magnetic susceptibility.
Generally, it is expected that Mn/Ln clusters exhibit slow relaxation behaviour because Ln
ions have a large magnetic anisotropy. According to the literature data a few examples of
hexanuclear 3d-4f complexes with a similar butterfly topology have been reported. Some of them
exhibit antiferromagnetic coupling, e.g. NiII LnIII [LnIII = GdIII, DyIII] [18m], or ferromagnetic
4
2
splitting with a slow magnetic relaxation of the magnetization, NiII HoIII [18n], whereas
4
2
NiII LnIII [LnIII = TbIII, DyIII] [13f,18n], FeIII DyIII [18o] or CoIII DyIII [13f] complexes with
4
2
4
2
2
4
ferromagnetic interactions are found to be SMMs.
However, in our systems, only the MnIII GdIII2 complex shows a very weak signal in the out of
4
phase component of the alternative current (ac) magnetic susceptibility under 1000 Oe dc, while
the MnIII SmIII and MnIII DyIII complexes do not, as the out–of–phase ac magnetic
4
2
4
2
susceptibility remains silent up to 2 K. One of the reasons for this may be the crystal structure
differences. On examining complexes 1-3, it can be seen that the lanthanide ions are situated at
the center of the core and manganese ions occupy the peripheral region (Fig. 4), opposite to the
examples presented in the literature. As a result, a possible exchange pathway can be realized
through a MnIII–(μ-O)2–LnIII–(μ-O)2–LnIII–(μ-O)2–MnIII or LnIII–(μ-O)2–MnIII–(μ-O)2–LnIII
bridge for complex 1 and 2 or MnIII–(μ-O)2–LnIII–O–MnIII–(μ-O)2–LnIII–(μ-O)2–MnIII for 3. In
fact there are no MIII–O–MIII [M = Mn] direct exchange interactions, as in NiII LnIII2 [13f,18n],
4

1
2
3
4
5
6
7
8
9
FeIII DyIII2 [18o] or CoIII DyIII4 [13f] complexes with the same topology. Different coordination
4
2
environments in compounds 1-3 and reported M4Ln2 complexes around the LnIII ion and/or the
different relative orientation of the local axes are probably responsible for the different relaxation
dynamics observed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
2.8. EPR Spectroscopic studies
The EPR spectra of both bands do not exhibit intensive wide lines for the measured complexes 1-
3 (Fig. S8 in the Supplementary Information). Only for 2, was it possible to determine the value
of the g-parameter as 2.00. This value is typical for complexes of gadolinium surrounded by
oxygen atoms [19]. The hyperfine structure was not observed. The spectra of 1 and 3 also exhibit
lines typical for a MnII sextet due, probably to very small amounts of MnII impurities.
2.9. Theoretical insight
The calculations were focused on the MnIII GdIII complex. The two other members of the
4
2
MnIII LnIII series are characterized by strong magnetic anisotropy on the LnIII sites induced by
4
2
crystal field splitting of their atomic J multiplets and to describe this effect multiconfigurational
ab initio methods should be used rather than DFT. However, the downside of these methods is
that for large systems they turn into a prohibitively expensive task.
To provide a deeper insight into magnetic properties of complex 2, the isotropic Heisenberg-
Dirac-van Vleck Hamiltonian was used:
(2)
Where the local spin operators for each paramagnetic centre and J is the Heisenberg
exchange coupling constant between two paramagnetic centres. On the assumption that there is
no coupling between MnIII ions, which is justified by the significant MnIII–MnIII distances in 2,
three different exchange pathways can be distinguished (Fig 10) and eq 2 can be rewritten as:
(3)
The JA, JB and JC values can be computed within the Broken Symmetry (BS) DFT
framework. For that purpose, single point calculations for four different spin configurations of
complex 2 have to be accomplished. The states used in this study are schematically shown in Fig.
10. They are the high spin configuration {+ + +} and three BS solutions {– + +}, {+ – +} and
{+ + –} with spins flipped on one GdIII ion, Mn1 and Mn1#, Mn2 and Mn2#, respectively.
Application of the Heisenberg-Dirac-van Vleck Hamiltonian from eq 3, to each of these four
single determinant wave functions leads to simple expressions of the related energies as shown in
eqs 4a-d.

1
2
3
4
5
6
7
8
9
(4a)
(4b)
(4c)
(4d)
–
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
From eqs 4a–d, analytical expressions for the JA, JB and JC values are obtained as eqs 5a–c.
–
(5a)
–
(5b)
(5c)
–
The results of the DFT calculations are summarized in Table S8. From this data, we can see
that all three predicted |J| values are not larger than 1.1 cm−1, which is an indication of a fairly
weak exchange coupling between the paramagnetic ions. It is important to emphasise that both
functionals brought about similar results, but the use of TPSSh tended to give slightly stronger
exchange couplings. A good illustration for this is that the JA values amounted to -0.60 and -0.82
cm−1 according to the B3LYP and TPSSh calculations, respectively.
Analysis of experimental magnetic data for complex 2 revealed that the coupling between the
two GdIII ions should be antiferromagnetic and the DFT results are in line with this conclusion
(the JA values computed for 2 at the B3LYP and TPSSh theory levels are slightly negative).
Moreover, the magnetic picture emerging from the theoretical investigation shows a weak
antiferromagnetic interaction between the GdIII ion and the neighbouring MnIII ions, which
agrees well with the experimentally determined negative value of zJ'. It is important to notice
that the magnetic interaction between the atoms Gd1 and Mn2 as well as between Gd1# and
Mn2# is slightly stronger than the coupling between Gd1 and Mn1 as well as between Gd1# and
Mn1# (|JC| > |JB|). The values of JB and JC predicted at the B3LYP theory level are -0.23 and -
0.93 cm−1, respectively.
In addition to the exchange couplings listed in the Supporting Information Table S8, in Fig.
11 the spin density isosurfaces in the high spin state {+ + +} are shown. This provides strong
evidence that for complex 2 in its high spin state the spin density is not only localised on the Gd
and Mn atoms, but also on the oxygen and nitrogen ligand donor atoms. To estimate the scale of
this effect, a Löwdin population analysis was carried out and this revealed that in 2 96.4%
(46.9% on two GdIII ions and 49.6% on four MnIII ions) of the spin population is located on the
metal sites. Hence, in the studied complexes the flow of spin density from the paramagnetic
cations toward the organic ligands can be considered as limited. As suggested by O. Kahn [18a],
the magnetic coupling can be qualitatively explained with the concept of spin delocalization and
spin polarization. Both phenomena transfer spin density onto ligands and thus are mirrored in its

1
2
3
4
5
6
7
8
9
isosurface. This was used in magnetochemistry to propose mechanisms governing magnetic
coupling in polynuclear clusters of transition metals [24a-f] and in complexes incorporating
lanthanides [23s,23t,24g-j]. From the fact that in complex 2 less than 4% of the spin population
is transferred onto the ligands it becomes apparent that the spin delocalisation and spin
polarization are minor and in consequence the antiferromagnetic interactions observed in
complex 2 are weak.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
3. Conclusions
In summary, we have investigated a MnIII LnIII2 (LnIII = SmIII, GdIII, DyIII) series to form a good
4
basis for structural and magnetic rationalization of several experimental observations.
Complexes 1-3 are heterometallic but homovalent, with two LnIII (Ln = SmIII, GdIII, DyIII) and
four MnIII ions. The coordination numbers of both the LnIII ions are same (nine for 1 and 2, eight
for 3). In contrast, two MnIII ions (Mn1, Mn1#) are six coordinate and the remaining two MnIII
centers (Mn2, Mn2#) are five coordinate. The central butterfly-like [MnIII LnIII ] core is capped
by two MnIII ions from the lanthanide side, resembling a spacecraft-like overall topology.
Synthetic access to the structurally related MnIII LnIII complexes has allowed a direct
comparison, showing that replacement of SmIII (1) and GdIII (2) with DyIII (3), leads to significant
structural perturbation in 3 as a result of lanthanide contraction. Allowing for the simplification
principle regarding topological analysis, the structures of 1 and 2 bear the 1,2,4M6–1 topology
whereas, the 1,2,3M6–1 topology is encountered in 3. Further, due to the variations of LnIII
and/or topologies, 1, 2 and 3 present quite different magnetic outcomes. DFT theoretical
calculations were done in order to provide additional insight into the magnetic interaction in 2.
2
2
4
2
4. Experimental Section
4.1 General Information
3-Amino-1,2-propanediol, o-vanillin and Ln(OAc)3·xH2O were procured from Aldrich Chemical
Co. Inc. (Germany) and were used as received. Mn(OAc)2·4H2O and triethylamine (TEA) were
purchased from SRL Pvt. Ltd., Mumbai, India. Commercially available solvents were purified
according to standard literature. 3-[(2-Hydroxy-3-methoxy-benzylidene)-amino]-propane-1,2-
diol (H3Vapd) was prepared according to literature procedures [20b]. Powder X-ray diffraction
(PXRD) patterns were acquired using a PANalytical, XPERT–PRO diffractometer (Netherlands)
operated at 40 kV, 30 mA, with graphite monochromatized Cu Kα radiation of wavelength
1.5406 Å and a nickel filter. Energy dispersive X-ray (EDX) analysis was done by field emission
scanning electron microscopy (FESEM−JEOL JSM 7600F). Elemental analysis for carbon,
hydrogen and nitrogen were carried out on a Perkin-Elmer 2400 II analyzer. IR spectra were
measured in the region 400-4000 cm–1 on a Perkin-Elmer RXI FTIR spectrophotometer with the
samples prepared as KBr disks. ESI-MS experiments for 1-3 were carried out using Waters Xevo
G2−SQTof instruments in LC-MS grade solvent.

1
2
3
4
5
4.2. Synthesis of the hexanuclear complexes 1-3
6
7
4.2.2. General procedure for the synthesis of [MnIII LnIII (Vapd)4(OAc)6]·4H2O (1 − 3)
4
2
8
9
The following general synthetic methodology was applied for the preparation of the
heterobimetallic hexanuclear metal complexes under aerobic conditions (shown in Scheme 1). A
methanolic solution of Et3N (2 equiv.) was added to a methanolic solution of H3Vapd (1 equiv.)
and the resulting mixture was stirred for 15 min. A methanolic solution of Mn(OAc)2·4H2O (1
equiv.) was added dropwise to the ligand solution under continuous stirring conditions and the
colour of the mixture changed from yellow to dark brown within a few minutes. After ca. 4 h.
stirring, a methanol-water solution (3:1, v/v) of Ln(OAc)3·xH2O (0.5 equiv.) was added and the
stirring was continued vigorously for further ca. 2.5 h. The final reaction mixture was filtered off
and the filtrate was evaporated to dryness under reduced pressure. The dark brown solid obtained
was left overnight in a vacuumed calcium chloride desiccator. The resulting solid was re-
dissolved in methanol. Crystals suitable for X-ray studies were obtained by slow diffusion of
diethyl ether into this methanolic solution over one week. The specific quantities and
stoichiometry of the reactants involved in each reaction, the yields of the products and their
characterization data are given below.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[MnIII4SmIII2(Vapd)4(OAc)6]·4H2O(1). Quantities: H3Vapd (0.112 g, 0.5 mmol),
.
.
Mn(OAc)2 4H2O (0.122 g, 0.5 mmol), Sm(OAc)3 xH2O (0.081 g, 0.25 mmol) and Et3N (0.138
mL, 1 mmol). Yield: 0.080 g (35% based on ligand). FT-IR data (KBr pellet) υ cm-1: 3435 (br),
1620 (vs), 1557 (vs), 1443 (s),1249 (s), 1049 (m), 742 (m). C56H74Mn4N4O32Sm2 (1835.67)
calcd: C 36.64, H 4.06, N 3.05; found: C 36.32, H 3.69, N 2.84. ESI–MS m/z, ion in CH3OH:
882.777, [Mn4Sm2(Vapd)4(OAc)6 + 2H+]2+.
[MnIII4GdIII (Vapd)4(OAc)6]·4H2O(2). Quantities: H3Vapd (0.112 g, 0.5 mmol),
2
.
.
Mn(OAc)2 4H2O (0.122 g, 0.5 mmol), Gd(OAc)3 xH2O (0.083 g, 0.25 mmol) and Et3N (0.138
mL, 1 mmol). Yield: 0.085 g (37% based on ligand). FT-IR data (KBr pellet) υ cm-1: 3435 (br),
1618 (vs), 1559 (vs), 1438 (s), 1249 (s), 1049 (m), 743 (m). C56H74Mn4N4O32Gd2 (1849.45)
calcd: C 36.37, H 4.03, N 3.03; found: C 36.20, H 3.54, N 2.81. ESI–MS m/z, ion in CH3OH:
832.077, [Mn4Gd2(Vapd)4(OAc)4]2+.
[MnIII4DyIII2(Vapd)4(OAc)6]·4H2O(3). Quantities: H3Vapd (0.112 g, 0.5 mmol),
.
.
Mn(OAc)2 4H2O (0.122 g, 0.5 mmol), Dy(OAc)3 xH2O (0.084 g, 0.25 mmol) and Et3N (0.138 ml,
1 mmol). Yield: 0.078 g (34% based on ligand). FT-IR data (KBr pellet) υ cm-1: 3567 (br), 1619
(vs), 1560 (vs), 1438 (s), 1249 (s), 1055 (m), 742 (m). C56H74Mn4N4O32Dy2(1862.03) calcd: C
36.16, H 4.01, N 3.01; found: C 35.53, H 3.79, N 2.82. ESI–MS m/z, ion in CH3OH: 806.032,
[Mn4Dy2(Vapd)4(OAc)3 + H−]2+.
4.3. Single crystal X-ray crystallography

1
2
3
4
5
6
7
8
9
Suitable single crystals of 1, 2 and 3 were chosen for X-ray diffraction studies.
Crystallographic data and details of the refinement are given in Table 2. The unit cell parameters
and the intensity data for the complexes (1, 2 and 3) were collected on a Nonius APEX–II
diffractometer with a CCD-area detector at 296 K using graphite-monochromated MoKα
radiation (λ = 0.71073 Å). The program SMART [21a] was used for collecting frames of data,
indexing reflections and determining lattice parameters, SAINT [21a] for integration of the
intensity of reflections and scaling, SADABS [21b] for absorption corrections and SHELXTL
[21c,d] for space group, structure determination and least-squares refinements on F2. All the
structures were solved by direct methods and refined by full-matrix least squares calculations
using SHELXTL software [21e]. The location of the Ln and Mn atoms were determined and O,
N and C atoms were subsequently determined from the difference Fourier maps. All the non-
hydrogen atoms were refined with anisotropic displacement parameters. All the hydrogen atoms
on carbon frameworks were included in the final stages of the refinement and were refined with a
typical riding model. Some hydrogen atoms on the solvent water molecules could not be located;
these were included in the molecular formula directly. The crystallographic figures have been
generated using Diamond 3.0 software [21f]. CCDC 1451600, 1451601 and 1451602 contain the
supplementary crystallographic data for this paper. These data can be obtained free of charge
from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/ cif.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
4.4. Magnetic measurement (experimental) details
Magnetic measurements in the temperature range 1.8-300 K were performed using a
Quantum Design SQUID-based MPMSXL-5-type magnetometer. The SQUID magnetometer
was calibrated with a palladium rod sample (Materials Research Corporation, measured
purity 99.9985%). The superconducting magnet was generally operated at a field strength
ranging from 0 to 5 T. Measurements were made at a magnetic field of 0.5 T. Corrections
were based on subtracting the sample-holder signal and contribution χD estimated from the
Pascal constants [22]. Magnetization measurements were conducted at 2 K in a magnetic
field from 0 to 5 T.
4.5. EPR spectroscopy (experimental) details
EPR spectra were obtained at room and liquid nitrogen temperatures, using a Bruker
Elexsys E500 spectrometer equipped with an NMR teslameter (ER 036TM). X-band
spectra were measured using a frequency counter (E 41 FC) and a double rectangular
cavity resonator (ER 4105DR) operating in the TD104 mode at a microwave power of 20
mW and a modulation amplitude of 5 G. In the case of low intensity signals, instead of
one scan, five scans were accumulated. Q-band spectra (34 GHz) were recorded by the
same spectrometer with an ER 5106QTW resonator and Bruker frequency counter using
the Li/LiH standard for g–factor calibration. The measurements parameters were as
follows: microwave power 11mW, modulation amplitude 10 G.

1
2
3
4
5
6
4.6. Computational details
7
8
9
All calculations were carried out with the ORCA 3.0.3 suite of programs [23a]. Due to the size of
the studied system, they were based on density functional theory (DFT). Since GdIII ions are
contained in the studied complex, scalar relativistic effects had to be taken into account in the
calculations. This was accomplished with the 0th order regular approximation (ZORA) [23b,c].
To facilitate the comparison between theory and experiment, the X-ray structure was used, but
only after the positions of all the hydrogen atoms, which were not reliably determined from the
X-ray diffraction, were optimised. These optimisations were performed with the GGA functional
BP86 [23d,e], taking advantage of the Split−RI−J approximation [23f] employing appropriate
auxiliary Coulomb fitting basis sets [23g]. Single point Broken-Symmetry (BS) DFT calculations
were performed with the hybrid B3LYP [23h,i,j] and hybrid meta-GGA TPSSh [23k]
functionals. In all the calculations, the segmented all-electron relativistically contracted (SARC)
basis sets on Gd, ZORA-def2-TZVP on Mn, O and N and ZORA-def2-SVP on C and H atoms
were used [23k,l], as well as increased integration grids (Grid5 in the ORCA convention) and
tight SCF convergence criteria. The TPSSh and B3LYP functionals were chosen as they were
shown to perform well in the studies of exchange coupled systems [23m-r], including complexes
with [NiIIGdIII], [NiIIGdIIINiII] and [CuIIGdIII] cores.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Acknowledgements
R.M and Y.S acknowledge UGC, India for a RFSMS fellowship (Sanction no.
UGC/740/RFSMS) and DST India (Sanction no. (SR/FT/CS-107/2011) for financial support.
DST–FIST and DST–PURSE are acknowledged for providing the X-ray diffraction and ESI–MS
facilities respectively at the Department of Chemistry, University of Calcutta. The magnetic
measurements were financed by a subsidy for statutory activity from the Ministry of Science and
Higher Education of the Republic of Poland for the Faculty of Chemistry, University of
Wroclaw. The DFT computations were performed using computers of the Wrocław Centre for
Networking and Supercomputing (Grant No. 47).
Appendix A. Supplementary data
The ligand structure, along with its different coordination pockets and binding modes, simulated
and experimental PXRD, EDX spectra, FT−IR, ESI−MS, EPR, crystal data, additional
crystallographic diagrams, Stiefel and Brown nomenclature, CShM results, metric parameters,
BVS calculations, topology, the magnetic equations used in fitting parameters and the theoretical
data of DFT calculations. CCDC 1451600, 1451601 and 1451602. These data can be obtained
free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-
033; or e-mail: deposit@ccdc.cam.ac.uk. Supplementary data associated with this article can be
found in the online version.

1
2
3
4
5
References
6
7
8
9
[1] (a) A. M ller, Nature, 447 (2007) 1034; (b) A. M ller, . K gerler, A. W. M. Dress, Coord.
Chem. Rev., 222 (2001) 193; (c) E. G. Mednikov, M. C. Jewell, L. F. Dahl, J. Am. Chem.
Soc., 129 (2007) 11619; (d) X. Yang, D. Schipper, R. A. Jones, L. A. Lytwak, B. J.
Holliday, S. Huang, J. Am. Chem. Soc., 135 (2013) 8468.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[2] (a) R. E. P. Winpenny, Chem. Soc. Rev., 27 (1998) 447; (b) M. Sakamoto, K. Manseki, H. O-
Kawa, Coord. Chem. Rev., 379 (2001) 219; (c) R. Sessoli, A. K. Powell, Coord. Chem. Rev.,
253 (2009) 2328; (d) M. Andruh, J. P. Costes, C. Diaz, S. Gao, Inorg. Chem., 48 (2009)
3342; (e) L. R. Piquer, E. C. Sañudo, Dalton Trans., 44 (2015) 8771; (f) K. Liu, W. Shi,
P.Cheng, Coord. Chem. Rev., 289-290 (2015) 74.
[3] (a) Y. N. Guo, G. F. Xu, P. Gamez, L. Zhao, S. Y. Lin, R. Deng, J. Tang, H. J. Zhang, J. Am.
Chem. Soc., 132 (2010) 8538; (b) P. H. Lin, T. J. Burchell, L. Ungur, L. F. Chibotaru, W.
Wernsdorfer, M. Murugesu, Angew. Chem. Int. Ed., 48 (2009) 9489; (c) X. Yi, K. Bernot,
F. Pointillart, G. Poneti, G. Calvez, C. Daiguebonne, O. Guillou, R. Sessoli, Chem. Eur. J.,
18 (2012) 11379; (d) K. Bernot, F. Pointillart, P. Rosa, M. Etienne, R. Sessoli, D. Gatteschi,
Chem. Commun., 46 (2010) 6458; (e) K. Bernot, J. Luzon, L. Bogani, M. Etienne, C.
Sangregorio, M. Shanmugam, A. Caneschi, R. Sessoli and D. Gatteschi, J. Am. Chem. Soc.,
131 (2009) 5573.
[4] R. Modak, Y. Sikdar, G. Cosquer, S. Chatterjee, M. Yamashita, S. Goswami, Inorg. Chem.,
55 (2016) 691.
[5] (a) M. D. Ward, Coord. Chem. Rev., 251 (2007) 1663; (b) B. Zhao, X. Y. Chen, Z. Chen,
W. Shi, P. Cheng, S. P. Yan, D. Z. Liao, Chem. Commun., (2009) 3113; (c) L.−A.Sorgho, C.
Besnard, P. Pattison, K. R. Kittilstved, A. Aebischer, J. C. G. Bu¨nzli, A. Hauser, C. Piguet,
Angew. Chem. Int. Ed., 50 (2011) 4108, (d) J. Jankolovits, C. M. Andolina, J. W. Kampf, K.
N. Raymond, V. L. Pecoraro Angew. Chem. Int. Ed., 50 (2011) 9660, (e) J. Long, J.
Rouquette, J-M. Thibaud, R. A. S. Ferreira, L. D. Carlos, B. Donnadieu, V. Vieru, L. F.
Chibotaru, L. Konczewicz, J. Haines, Y. Guari, J. Larionova Angew. Chem. Int. Ed., 54
(2015) 2236.
[6] M. Orfanoudaki, I. Tamiolakis, M. Siczek, T. Lis, G. S. Armatas, S. A. Pergantis, C. J.
Milios, Dalton Trans., 40 (2011) 4793.
[7] (a) C. J. Li, Z. J. Lin, M. X. Peng, J. D. Leng, M. M. Yang, M. L. Tong, Chem. Commun.,
(2008) 6348; (b) Y. Wang, P. Cheng, J. Chen, D. Z. Liao, S. P. Yan, Inorg. Chem., 46 (2007)
4530; (c) Z. G. Gu, H. C. Fang, P. Y. Yin, L. Tong, Y. Ying, S. J. Hu, W. S. Li, Y. P. Cai,
Cryst. Growth Des., 11 (2011) 2220.

1
2
3
4
5
6
7
8
9
[8] (a) M. Shibasaki, N. Yoshikawa, Chem. Rev., 102 (2002) 2187; (b) C. E. Plecñik, S. M. Liu,
S. G. Shore, Acc. Chem. Res., 36 (2003) 499; (c) K. Griffiths, P. Kumar, G. R. Akien, N. F.
Chilton, A. Abdul-Sada, G. J. Tizzard, S. J. Coles, G. E. Kostakis, Chem. Commun., 52
(2016) 7866; (d) K. Griffiths, C. W. D. Gallop, A. Abdul-Sada, A. Vargas, O. Navarro, G. E.
Kostakis, Chem. Eur. J., 21 (2015) 6358.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[9] (a) A. J. Blake, R. O. Gould, P. E. Y. Milne, R. E. P. Winpenny, J. Chem. Soc. Chem.
Commun., (1991) 1453; (b) X.−M. Chen, S. J. M. Aubin, Y.−L. Wu, Y.−S. Yang, T. C. W.
Mak, D. N. Hendrickson, J. Am. Chem. Soc., 116 (1995) 9600; (c) C. Benelli, D. Gatteschi,
Chem. Rev., 102 (2002) 2369; (d) X.−J. Kong, L.−S. Long, Z. Zheng, R.−B. Huang, L.−S.
Zheng, Acc. Chem. Res., 43 (2010) 201. (e) V. Mereacre, A. Baniodeh, C. E. Anson, A. K.
Powell, J. Am. Chem. Soc., 133 (2011) 15335; (f) H. Ke, L. Zhao, Y. Guo, J. Tang, Dalton
Trans., 41 (2012) 2314.
[10](a) T. Lis, Acta Crystallogr. Sect. B: Structural Science, B36 (1980) 2042; (b) A. Caneschi,
D. Gatteschi, R. Sessoli, A. L. Barra, L. C. Brunel, M. Guillot, J. Am. Chem. Soc., 113 (1991)
5873; (c) S. M. J. Aubin, M. W. Wemple, D. M. Adams, H.−L. Tsai, G. Christou, D. N.
Hendrickson, J. Am. Chem. Soc., 118 (1996) 7746.
[11](a) C. Benelli, M. Murrie, S. Parsons, R. E. P. Winpenny, J. Chem. Soc., Dalton Trans.,
(1999) 4125; (b) M. N. Akhtar, Y. Lan, V. Mereacre, R. Clérac, C. E. Anson, A. K. Powell,
Polyhedron, 28 (2009) 1698; (c) V. Mereacre, Y. Lan, R. Clérac, A. M. Ako, I. J. Hewitt, W.
Wernsdorfer, G. Buth, C. E. Anson, A. K. Powell, Inorg. Chem., 49 (2010) 5293; (d) A.
Mishra, W. Wernsdorfer, S. Parsons, G. Christou, E. K. Brechin, Chem. Commun., (2005)
2086; (e) C. Papatriantafyllopoulou, K. A. Abboud, G. Christou, Inorg. Chem., 50 (2011)
8959; (f) M. Murugesu, A. Mishra, W. Wernsdorfer, K. Abboud, G. Christou, Polyhedron,
26 (2006) 613; (g) M. N. Akhtar, Y. −Z. Zheng, Y. Lan, V. Mereacre, C. E. Anson, A. K.
Powell, Inorg. Chem., 48 (2009) 3502; (h) N. F. Chilton, S. K. Langley, B. Moubaraki, K. S.
Murray, Chem. Commun., 46 (2010) 7787; (i) T. Shiga, N. Hoshino, M. Nakano, H. Nojiri,
H. Oshio, Inorg. Chim. Acta, 361 (2008) 4113; (j) T. T. Boron III, J. F. Kampf, V. L.
Pecoraro, Inorg.Chem., 49 (2010) 9104; (k) A. Saha, M. Thompson, K. A. Abboud, W.
Wernsdorfer, G. Christou. Inorg. Chem., 50 (2011) 10476; (l) J. Liu, C. Ma, H. Chen, M. Hu,
H. Wen, H. Cui, X. Song, C. Chen, Dalton Trans.,42 (2013) 2423; (m) M. Li, A. M. Ako, Y.
Lan, W. Wernsdorfer, G. Buth, C. E. Anson, A. K. Powell, Z. Wang, S. Gao, Dalton Trans.,
39 (2010) 3375; (n) V. Mereacre, M. N.; Akhtar, Y. Lan, A. M. Ako, R. Clérac, C. E. Anson,
A. K. Powell, Dalton Trans., 39 (2010) 4918; (o) M. Li, Y. Lan, A. M. Ako, W.
Wernsdorfer, C. E. Anson, G. Buth, A. K. Powell, Z. Wang, S. Gao, Inorg. Chem. 49 (2010)
11587; (p) Z. Majeed, K. C. Mondal, G. E. Kostakis, Y. Lan, C. E. Anson, A. K. Powell,
Dalton Trans. 39 (2010) 4740; (q) V. Mereacre, A. M. Ako, R. Clérac, W. Wernsdorfer, I. J.
Hewitt, C. E. Anson, A. K. Powell, Chem. Eur. J., 14 (2008) 3577; (r) T. Shiga, T. Onuki, T.
Matsumoto, H. Nojiri, G. N. Newton, N. Hoshino, H. Oshio, Chem. Commun., (2009) 3568;
(s) A. M. Ako, V. Mereacre, R. Clérac, I. J. Hewitt, Y. Lan, G. Buth, C. E. Anson, A. K.
Powell, Inorg. Chem., 48 (2009) 6713; (t) G. Rigaux, R. Inglis, S. Morrison, A. Prescimone,
C. Cadiou, M. Evangelisti, E. K. Brechin, Dalton Trans., 40 (2011) 4797; (u) M. Holynska,
D. Premuzic, I.−R. Jeon, W. Wernsdorfer, R. Clérac, S. Dehnen, Chem. Eur. J.17 (2011)
9605; (v) C. M. Zaleski, E. C. Depperman, J. W. Kampf, M. L. Kirk, V. L. Pecoraro, Angew.

1
2
3
4
5
6
7
8
9
Chem. Int.Ed., 43 (2004) 3912; (w) A. E. Thuijs, A. Marton, T. C. Stamatatos, K. A.
Abboud, G. Christou, Polyhedron,103 (2016) 288; (x) S. Langley, B. Moubaraki, K. S.
Murray, Dalton Trans., 39 (2010) 5066; (y) V. Mereacre, D. Prodius, A. M. Ako, N. Kaur, J.
Lipkowski, C. Simmons, N. Dalal, I. Geru, C. E. Anson, A. K. Powell, C. Turta, Polyhedron,
27 (2008) 2459; (z)V. Mereacre, A. M. Ako, R. Clérac, W. Wernsdorfer, G. Filoti, J.
Bartolom´e, C. E. Anson, A. K. Powell, J. Am. Chem. Soc., 129 (2007) 9248.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[12](a) A. Mishra, W. Wernsdorfer, K. A. Abboud, G. Christou, J. Am. Chem. Soc., 126 (2004)
15648; (b) T. C. Stamatatos, S. J. Teat, W. Wernsdorfer, G. Christou, Angew. Chem. Int.
Ed., 48 (2009) 521; (c) J.−L. Liu, F.−S. Guo, Z.−S. Meng, Y.−Z. Zheng, J.−D. Leng, M.−L.
Tong, L. Ungur, L. F. Chibotaru, K. J. Heroux, D. N. Hendrickson, Chem. Sci., 2 (2011)
1268; (d) A. M. Ako, V. Mereacre, R. Clérac, W. Wernsdorfer, I. J. Hewitt, C. E. Anson, A.
K. Powell, Chem. Commun., (2009) 544; (e) C.Papatriantafyllopoulou, W. Wernsdorfer, K.
A. Abboud, G. Christou, Inorg. Chem.,50 (2011) 421.
[13](a) R. Modak, Y. Sikdar, S. Mandal, C. J. Gómez–García, S. Benmansour, S. Goswami,
Polyhedron,17 (2014) 155; (b) R. Modak, Y. Sikdar, S. Mandal, S. Goswami, Inorg. Chem.
Commun., 37 (2013) 193; (c) S. Nayak, H. P. Nayek, S. Dehnen, A. K. Powell, J. Reedijk,
Dalton Trans., 40 (2011) 2699; (d) A. K. Jami, V. Baskar, E. C. Sañudo, Inorg. Chem., 52
(2013) 2432; (e) L. Cong, X. Qin, W. Sun, Y.; Wang, S. Ding, Z. Liu, New J. Chem., 38
(2014) 545; (f) B. L. Liu, Q. X. Liu, H. P. Xiao, W. Zhang, R. J. Tao, Dalton Trans., 42
(2013) 5047; (g) Q. Gao, Y. Qin, Y. Chen, W. Liu, H. Li, B. Wu, Y. Li, W. Li, RSC Adv.,
5(2015)43195.
[14] M. Andruh, Dalton Trans., 44 (2015) 16633.
[15] A. K. Sharma, F. Lloret, R. Mukherjee, Inorg. Chem., 52 (2013) 4825.
[16](a) W. Liu, H. H. Thorp, Inorg. Chem., 32 (1993) 4102; (b) A. Ruiz-Martínez, D. Casanova,
S. Alvarez, Chem. Eur. J., 14 (2008) 1291; (c) A. Ruiz-Martinez, D. Casanova, S. Alvarez,
Dalton Trans., (2008) 2583; (d) E. I. Stiefel, G. F. Brown, Inorg. Chem., 11 (1972) 434; (e)
A. W. Addison, T. N. Rao, J. Reedijk, J. V. Rijn, G. C. Verschoor, J. Chem. Soc., Dalton
Trans., (1984) 1349.
[17](a) G. E. Kostakis, A. M. Ako, A. K. Powell, Chem. Soc. Rev., 39 (2010) 2238; (b) G. E.
Kostakis, A. K. Powell, Coord. Chem. Rev., 253 (2009) 2686; (c) G. E. Kostakis, A. K.
Powell, Chem. Eur. J., 16 (2010) 7983; (d) G. E. Kostakis, V. A. Blatov, D. M. Prserpio,
Dalton Trans., 41 (2012) 4634; (e) http://www.sussex.ac.uk/lifesci/kostakislab/picd/.
retrieved 25th March 2016. (f) X. Yang, Z. Li, S. Wang, S. Huang, D. Schipper, R. A. Jones,
Chem. Comm., 50 (2014) 15569, (g) Y-Z. Zheng, M. Evangelisti, F. Tuna, R. E. P.
Winpenny, J. Am. Chem. Soc., 134 (2012) 1057.
[18](a) O. Kahn, Molecular Magnetism, Wiley-VCH, New York, (1993); (b) E. A. Mikhalyova,
A. V. Yakovenko, M. Zeller, K. S. Gavrilenko, S. E.; Lofland, A. W. Addison, V. V.
Pavlishchuk, Inorg. Chim. Acta, 414 (2014) 97; (c) C. J. O’Connor, Prog. Inorg. Chem., 29
(1982) 203; (d) D. John, Eur. J. Inorg. Chem., (2006) 3503; (e) J. S. Smart, Effective Field

1
2
3
4
5
6
7
8
9
Theories of Magnetism, W. B. Saunders Comp. Philadelphia and London, (1966); (f) A. P.
Ginsberg, M. E. Lines, Inorg. Chem., 11 (1972) 2289; (g) P. Guerriero, S. Tamburii, P. A.
Vigato, C. Benelli, Inorg. Chim. Acta, 189 (1991) 19; (h) M. H. Molina, C. R. Perez, T.
Lopez, F. Lloret, M. Julve, Inorg. Chem., 42 (2003) 5456; (i) J. Feuersenger, D. Prodius, V.
Mereacre, R. Clérac, C. E. Anson, A. K. Powell, Inorg. Chem. Comm.,14 (2011) 1851; (j) X.
Feng, X.–L. Ling, H.–L. Song, L.–Y. Wang, S.–W. Ng, B. Y. Su, Dalton Trans., 42 (2013)
10292; (k) J.–C. G. Bunzli, G. R. Chopin, Lanthanide probes in Life, Chemical and Earth
Sciences, Elsevier, Amsterdam, (1989); (l) S. C. Manna, E. Zangrando, A. Bencini, C.
Benelii, N. R. Chaudhuri, Inorg. Chem., 45 (2006) 9114; (m) H. S. Ke, L. Zhao, Y. Guo, J.
Tang, Inorg. Chem., 51 (2012) 2699; (n) L. Zhao, J. Wu, H. Ke, J. Tang, Inorg. Chem., 53
(2014) 3519; (o) S. Chen, V. Mereacre, C. E. Anson, A. K. Powell, Dalton Trans., 45 (2016)
98.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[19] S. Rast, A. Borel, L. Helm, E. Belorizky, P. H. Fries, A.E. Merbach, J. Am. Chem. Soc., 123
(2001)2637.
[20](a) B. S. Furniss, A. J. Hannaford, P. W. G. Smith, A. R. Tatchell, Vogel’s Text book of
Practical Organic Chemistry, 5th ed. ELBS, Longman: London, (1989); (b) C.−M. Liu,
R.−G. Xiong, D.−Q. Zhang, D.−B. Zhu, J. Am. Chem. Soc., 132 (2010) 4044.
[21] (a) SMART & SAINT Software Reference manuals, Version 6.45; Bruker Analytical X-ray
Systems, Inc.: Madison, WI, (2003); (b) G. M. Sheldrick, SADABS, a software for
empirical absorption correction, Ver. 2.05; niversity of G ttingen G ttingen, Germany,
(2002); (c) SHELXTL Reference Manual, Ver. 6.1; Bruker Analytical Xray Systems, Inc.:
Madison, WI, (2000); (d) G. M. Sheldrick, SHELXTL, Ver. 6.12; Bruker AXS, Inc.:
Madison, WI, (2001); (e) G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr.,
64 (2008) 112; (f) K. Bradenburg, Diamond, ver. 3.0; Crystal Impact GbR: Bonn (Germany)
(2005).
[22] E. A. Boudreaux and L. N. Mulay, Theory and Applications of Molecular Paramagnetism,
Wiley−Interscience, New York (1976).
[23](a) F. Neese, Wiley Interdiscip. Rev. Comput. Mol. Sci., 2 (2012) 73; (b) C. van
Wuellen, J. Chem. Phys., 109 (1998) 392; (c) E. van Lenthe, A. E. Ehlers, E. J.
Baerends, J. Chem. Phys., 110 (1999) 8943; (d) A. D. Becke, Phys. Rev. A, 38 (1988)
3098; (e) J. P. Perdew, Phys. Rev. B, 33 (1986) 8822; (f) F. Neese, J. Comput. Chem.,
24 (2003) 1740; (g) F. Weigend, Phys. Chem. Chem. Phys., 8 (2006) 1057; (h) A. D.
Becke, J. Chem. Phys., 98 (1993) 5648; (i) C. Lee, W.; Yang, R. G. Parr, Phys. Rev.
B, 37 (1988) 785; (j) P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem., 98 (1994) 11623; (k) V. N. Staroverov, G. E.; Scuseria, J. M.; Tao, J. P.
Perdew, J. Chem. Phys., 119 (2003) 12129; (l) D. A. Pantazis, F. Neese, J. Chem.
Theory Comput., 5 (2009) 2229; (m) D. A. Pantazis, X. Y. Chen, C. R. Landis, F.
Neese, J. Chem. Theory Comput., 4 (2008) 908; (n) C. Baffert, M. Orio, D. A.
Pantazis, C. Duboc, A. G. Blackman, G. Blondin, F. Neese, A. Deronzier, M.–N.
Collomb, Inorg. Chem., 48 (2009) 10281; (o) D. A. Pantazis, M. Orio, T. Petrenko, S.
Zein, E. Bill, W. Lubitz, J. Messinger, F. Neese, Chem. Eur. J., 15 (2009) 5108; (p)

1
2
3
4
5
6
7
8
9
M. Orio, D. A. Pantazis, T. Petrenko, F. Neese, Inorg. Chem., 48 (2009) 7251; (q) Y.
Sikdar, R. Modak, D. Bose, S. Banerjee, D. Bieńko, W. Zierkiewicz, A. Bieńko, K.
D. Saha, S. Goswami, Dalton Trans., 44 (2015) 8876; (r) R. Biswas, C. Diaz, A.
Bauzá, M. Barceló–Oliver, A. Frontera, A. Ghosh, Dalton Trans., 43 (2015) 6455; (s)
S. K. Singh, N. K. Tibrewal, G. Rajaraman, Dalton Trans., 40 (2011) 10897; (t) G.
Rajaraman, F. Totti, A. Bencini, A. Caneschi, R. Sessoli, D. Gatteschi, Dalton
Trans., (2009) 3153.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
[24] (a) . . Romańczyk,K. Noga, A. J. Włodarczyk, W. Nitek, E. Broclawik, Inorg.
Chem., 49 (2010) 7676; (b) H. Bolvin, F. R. Wagner, Inorg. Chem., 51 (2012) 7112;
(c) R. Herchel, I. Nemec, M. Machata, Z. Travnícek, Inorg. Chem., 54 (2015) 8625;
(d) J.-D. Leng, S.-K. Xing, R. Herchel, J.-L. Liu, M.-L. Tong, Inorg. Chem., 53
(2014) 5458; (e) S. K. Singh, K. R. Vignesh, V. Archana, G. Rajaraman, Dalton
Trans., 45 (2016) 8201; (f) I. Oyarzabal, J. Ruiz, A. J. Mota, A. Rodríguez-Diéguez,
J. M. Seco, E. Colacio, Dalton Trans., 44 (2015) 6825; (g) E. Cremades, S. Gómez-
Coca, D. Aravena, S. Alvarez, E. Ruiz, J. Am. Chem. Soc., 134 (2012) 10532; (h) T.
Rajeshkumar, S. K. Singh, G. Rajaraman, Polyhedron, 52 (2013) 1299; (i) S.
Meskaldji, A. Zaiter, L. Belkhiri, A. Boucekkine, Theor. Chem. Acc., 131 (2012),
1151; (j) B. Teyar, L. Belkhiri, K. Costuas, A. Boucekkine, K. Meyer, Inorg. Chem.,
55 (2016) 2870.

Figures
Scheme 1. Synthesis of the heterobimetallic hexanuclear complexes 1-3.

Scheme 2. Examples of previously known MnIII LnIII2 and MnII MnIII LnIII2 oxido clusters.[9f, 11i–
4
2
2
11l]

Fig. 1. Structure of the clusters 1 (above) and 2 (below) with the selected numbering scheme
used. Hydrogen atoms and solvent molecules are omitted for clarity. Symmetry transformations
used to generate equivalent atoms: (#) 1–x, 1–y, 1–z.

Fig. 2. Structure of cluster 3 with selected numbering scheme used. Hydrogen atoms and solvent
molecules are omitted for clarity. Symmetry transformations used to generate equivalent atoms:
(#) 1-x, 1-y, 1-z.
−
Fig. 3. Binding modes of the tri–anionic ligands (Vapd3 ) and coordinated acetateo ions (OAc−).

Fig 4 Representative spacecraft-like [MnIII LnIII O10] core of 1 and 3. Symmetry transformations
4
2
used to generate equivalent atoms: (#) 1-x, 1-y, 1-z.
Fig. 5. Distorted octahedral geometry with azimuthal expansion around (a) Mn1/1# in 1-3.
Distorted trigonal bipyramid geometry around (b) Mn2/2# in 1-3. Distorted spherical capped
square antiprism geometry around (c) Sm1/1# in 1. Distorted biaugmented trigonal prism
geometry around (d) Dy1/1# in 3. Symmetry transformations used to generate equivalent atoms:
(#) 1–x, 1–y, 1–z.

Fig. 6. The decrease of the Ln1–O4 bond length (●) along with the increase of the Ln1#–O4
distance (●) against the number of f electrons of the trivalent lanthanide cation.
1,2,4M6-1 [MnIII SmIII ]
1,2,4M6-1[MnIII GdIII ]
1,2,3M6-1[MnIII DyIII ]
4
2
4
2
4
2
Fig. 7. Topological motif of the Mn4Ln2 clusters in 1-3.