DaltonTransactions

PAPER

Cite this: Dalton Trans., 2013, 42, 5711

Received 18th November 2012,Accepted 16th January 2013

DOI: 10.1039/c3dt32752c

www.rsc.org/dalton

New 3d–4f heterometallic clusters built from mixedglycine and iminodiacetate acid: dioctahedron {La2Ni9}and onion-like {Gd5}⊂{Ni12} with interestingmagnetocaloric effect†

Zhong-Yi Li,a,b Jiang Zhu,c Xiao-Qun Wang,a Jun Ni,a Jian-Jun Zhang,*a Shu-Qin Liua

and Chun-Ying Duan*a,b

The preparation, structures and properties of 3d–4f compounds, undecanuclear [La2Ni9(Gly)12(IDA)3-

(μ3-OH)3][La(H2O)9][Na3(H2O)7(ClO4)3](ClO4)6·5H2O (1) and isostructural heptadecanuclear [Ln5Ni12-

(Gly)12(IDA)6(μ3-OH)9(H2O)3](ClO4)6·11H2O (Ln = Gd (2); Nd (3); Sm (4); Tb (5); Dy (6); Y (7)) based on

mixed glycine (HGly) and iminodiacetate acid (H2IDA) ligands were described. The structure of the

[La2Ni9(μ3-OH)3(IDA)3(Gly)12]3+ cationic cluster in 1 can be described as a face-shared and La-centered

dioctahedron. However, the [Ln5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]6+ cationic clusters in 2–7 bear an

onion-like {Ln5}⊂{Ni12} structure, where the trigonal bipyramid {Ln5} core is encapsulated by the outer

triangular orthobicupola {Ni12} shell. Magnetic studies have been performed for these compounds, and 2

displays dominant ferromagnetic coupling and has a large magnetocaloric effect (21.8 J kg−1 K−1,

ΔH = 7 T).

Introduction

Recently, the research of magnetic refrigeration based on theprinciple of the magnetocaloric effect (MCE), which relies onthe change of magnetic entropy upon application of a mag-netic field, has attracted great attention due to its energy-efficient and environmentally friendly features and possiblereplacement of the expensive helium-3 method in ultra low-temperature refrigeration.1,2 For a good molecular magneticrefrigerant, large spin ground state, negligible magnetic aniso-tropy and dominant ferromagnetic exchange are all necessary.3

High-nuclearity 3d–4f heterometallic clusters, especially 3d-Gdcompounds are good candidates to meet these requirementsdue to their high isotropic spins of lanthanide ions (for Gd3+

ion, S = 7/2), as well as the preference for strong ferromagneticinteraction between 3d and heavy 4f ions.4,5 Although morethan ten reports about the magnetocaloric effect of 3d–4f

heterometallic clusters have been published, the relationshipsbetween structure and property are not very clear yet. Thus, adetailed understanding of these relationships is still requiredfor rational molecular magnetic refrigerant design through thesynthesis of new 3d–4f compounds bearing different nuclearityand configuration.

Initially, 3d–4f heterometallic compounds were generallyconstructed by a single multidentate ligand containing bothN- and O-donor atoms. Later on, many small-size or monoden-tate ligands containing either N- or O-donor atoms, such asN3

−,6 HCOO− 7 and (CH3)3CCO2− 8 anions, were employed into

the reaction system as auxiliary ligands. Cooperating with themain ligands, the auxiliary ligands can be used to bridge themetal ions or to meet the different coordination spheres ofmetal ions, which could increase the nuclearity, affect themagnetic interactions or even stabilize the resulting com-pounds in the crystallization process.9 On the other hand, thesynthesis of such compounds is not easy, due to the hugedifficulty arising from the different coordinating affinities of3d and 4f metal ions to the two different ligands during theself-assembly process, especially when the two ligands havesimilar sizes or donor sets.

α-Amino acids (AAs) and H2IDA, both containing theN- and O-donors and displaying various bridging modes,have been demonstrated to be excellent ligands for the con-struction of high-nuclearity 3d–4f cluster compounds, suchas {Gd6Cu24},

10 {Ln6Cu27},11 {Ln20Ni30}

12 and {Gd54Ni54}.13

†Electronic supplementary information (ESI) available: Crystal data, additionalcrystallographic diagrams and magnetic diagrams, IR spectra and PXRD pat-terns. CCDC 910376 (1) and 910375 (2). For ESI and crystallographic data in CIFor other electronic format see DOI: 10.1039/c3dt32752c

aChemistry College, Dalian University of Technology, Dalian 116024, China.

E-mail: zhangjj@dlut.edu.cnbState Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian,

116024, China. E-mail: cyduan@dlut.edu.cncSchool of Chemistry and Chemical Engineering, Liaoning Normal University,

Dalian 116029, China

This journal is © The Royal Society of Chemistry 2013 Dalton Trans., 2013, 42, 5711–5717 | 5711

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article OnlineView Journal | View Issue

Recently, our interest has been focused on the design andsynthesis of high-nuclearity 3d–4f heterometallic clustersbased on AA and H2IDA as coligands, and some interestingresults have been obtained. Here, we report the syntheses,structures and magnetic properties of a series of high-nuclearity 3d–4f clusters: [La2Ni9(Gly)12(IDA)3(μ3-OH)3]-[La(H2O)9][Na3(H2O)7(ClO4)3]·(ClO4)6·5H2O (1) and [Ln5Ni12-(Gly)12(IDA)6(μ3-OH)9(H2O)3]·(ClO4)6·11H2O (Ln = Gd (2);Nd (3); Sm (4); Tb (5); Dy (6); Y (7)), based on mixed ligands ofHGly and H2IDA.

Experimental sectionMaterials and physical measurements

All chemicals were obtained from commercial sources andused without further purification. Aqueous solutions of lantha-num perchlorate were prepared by digesting lanthanide oxidesin concentrated perchloric acid. Caution! Perchlorate salts ofmetal complexes are potentially explosive. Only a small amountshould be used and handled with great care. X-ray powder dif-fraction (XPD) measurements were carried out on a RigukuD/Max-2400 X-ray Diffractometer using Cu Kα (λ = 1.5418 Å)radiation at room temperature. Elemental analyses were deter-mined using a Vario EL III elemental analyzer. FT-IR spectrawere recorded in the range of 4000–400 cm−1 on a JASCO FT/IR-430 spectrometer with KBr pellets. Magnetic measurementswere performed on a Quantum Design MPMS XL-7. The datawas corrected for the sample holder and the diamagneticcontributions.

Synthesis of 1–7. Compounds 1–7 were prepared under thesame conditions. 0.2 ml 1 M Ln(ClO4)3 aqueous solution(0.2 mmol; Ln = La, Gd, Nd, Sm, Tb, Dy and Y), 0.439 gNi(ClO4)2·6H2O (1.2 mmol), 0.045 g HGly (0.6 mmol), 0.027 gH2IDA (0.2 mmol) and 2 ml deionized water were placed in a20 ml scintillation vial. 1 M NaOH aqueous solution was usedto adjust the pH value of the solution to ∼6.2. Then the solu-tion was heated at 90 °C for 24 h and cooled to room temp-erature. The blue crystals were collected, washed three timeswith deionized water and dried in air.

[La2Ni9(Gly)12(IDA)3(μ3-OH)3][La(H2O)9][Na3(H2O)7(ClO4)3]·(ClO4)6·5H2O (1): yield: 65%. Anal. Calc. for C36H104Cl9La3N15-Na3Ni9O94: C, 12.05; H, 3.03; N, 5.85%. Found: C, 11.98; H,3.27; N, 5.61%. IR (KBr pellet, cm−1): 3353 s, 1593 s, 1452 m,1415 s, 1309 w, 1089 s, 987 w, 806 w, 727 w, 627 m.

[Gd5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]·(ClO4)6·11H2O (2):yield: 59%. Anal. Calc. for C48H115Cl6Gd5N18Ni12O95: C, 13.83;H, 2.78; N, 6.05%. Found: C, 13.45; H, 3.00; N, 5.59%. IR (KBrpellet, cm−1): 3441 s, 2967 w, 1587 s, 1450 m, 1409 s, 1306 w,1098 s, 927 w, 728 w, 626 m.

[Nd5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]·(ClO4)6·11H2O (3):yield: 50%. Anal. Calc. for C48H115Cl6Nd5N18Ni12O95: C, 14.05;H, 2.83; N, 6.15%. Found: C, 13.50; H, 2.88; N, 5.85%. IR (KBrpellet, cm−1): 3442 s, 2963 w, 1587 s, 1450 m, 1410 s, 1305 w,1096 s, 943 w, 725 w, 625 m.

[Sm5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]·(ClO4)6·11H2O (4):yield: 55%. Anal. Calc. for C48H115Cl6Sm5N18Ni12O95: C, 13.95;H, 2.80; N, 6.10%. Found: C, 13.56; H, 2.87; N, 5.77%. IR (KBrpellet, cm−1): 3439 s, 2948 w, 1589 s, 1451 m, 1410 s, 1307 w,1090 s, 942 w, 727 w, 626 m.

[Tb5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]·(ClO4)6·11H2O (5):yield: 49%. Anal. Calc. for C48H115Cl6Tb5N18Ni12O95: C, 13.80;H, 2.78; N, 6.04%. Found: C, 13.54; H, 2.92; N, 5.67%. IR (KBrpellet, cm−1): 3423 s, 2948 w, 1597 s, 1452 m, 1411 s, 1307 w,1090 s, 942 w, 728 w, 626 m.

[Dy5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]·(ClO4)6·11H2O (6):yield: 69%. Anal. Calc. for C48H115Cl6Dy5N18Ni12O95: C, 13.75;H, 2.76; N, 6.01%. Found: C, 13.56; H, 2.86; N, 5.82%. IR (KBrpellet, cm−1): 3451 s, 2960 w, 1589 s, 1451 m, 1410 s, 1306 w,1098 s, 928 w, 727 w, 625 m.

[Y5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)4]·(ClO4)6·11H2O (7): yield:56%. Anal. Calc. for C48H115Cl6Y5N18Ni12O95: C, 15.07; H, 3.03;N, 6.59%. Found: C, 15.23; H, 3.17; N, 6.22%. IR (KBr pellet,cm−1): 3442 s, 2948 w, 1601 s, 1453 m, 1411 s, 1308 w, 1088 s,942 w, 731 w, 627 m.

X-ray crystallography

The intensity data were measured at 293(2) K on a BrukerSMART APEX II CCD area detector system with graphite-monochromated Mo Kα (λ = 0.71073 Å) radiation. Datareduction and unit cell refinement were performed withSmart-CCD software.14 The structures were solved by directmethods using SHELXS-97 and were refined by full-matrixleast-squares methods using SHELXL-97.15

For 1, all non-hydrogen atoms were refined anisotropically.Hydrogen atoms on the OH− group and the coordinated waterwere initially found on Fourier difference maps and thenrestrained by using the DFIX instruction. The hydrogen atomswere included in the structural model as fixed atoms (usingidealized sp2-hybridized geometry and C–H bond lengths of0.95 Å) “riding” on their respective carbon atoms. Since thedisordered H2O solvent molecules and Na+ ions could not beunambiguously modeled, the Platon Squeeze16 option wasutilized based on the model that included the cationiccluster. Squeeze indicates a total solvent accessible areavolume of 1110 Å3, corresponding to about 302 electrons percell or approximately 3Na+ and 12 water molecules performula, which was further confirmed by the elemental ana-lysis data.

For 2, all non-hydrogen atoms were refined anisotropicallyexcept N10, C15, C19 and C21 (atoms from HGlys). C19 andC21 are disordered and the occupancy factors are 0.60 forC19A and C21A and 0.40 for C19B and C21B, respectively.Since the disordered H2O solvent molecules and perchlorateanions could not be unambiguously modeled, the PlatonSqueeze16 option was utilized based on the model thatincluded the cationic cluster. Squeeze indicates a total solventaccessible area volume of 4119 Å3, corresponding to about1418 electrons per cell or approximately 6 ClO4

− and 11 watermolecules per formula, which was further verified by theelemental analysis data. A summary of the most important

Paper Dalton Transactions

5712 | Dalton Trans., 2013, 42, 5711–5717 This journal is © The Royal Society of Chemistry 2013

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article Online

crystal and structure refinement data of 1 and 2 is given inTable 1. Selected bond lengths and angles are given in TablesS1 and S2.† Crystal data of compounds 3–7 were also collected.Due to the weak intensities, only preliminary structures canbe observed. All of them are isomorphous with 2, as alsoconfirmed by powder XRD patterns and elemental analysisdata. Their cell parameters are listed in Table S3.†

Results and discussionCrystal structure of 1

Compound 1 has two types of cations: [La(H2O)9]3+ and

[La2Ni9(μ3-OH)3(IDA)3(Gly)12]3+, which are balanced by ClO4

−

anions. Besides, [Na3(H2O)7(ClO4)3] and water solvent mole-cules are also observed in the lattice. In the [La(H2O)9]

3+

cation, the La3+ ion is terminated by nine water molecules andits coordination polyhedron can be described as a distortedmonocapped square antiprism. The average La–O bond dis-tance is 2.55(3) Å.

[La2Ni9(μ3-OH)3(IDA)3(Gly)12]3+ cationic cluster features a

face-shared and La-centered dioctahedron structural motif(Fig. 1). Each La3+ ion has an O12 donor set which could bebest described as an icosahedron. Nine carboxylate oxygenatoms from three IDA and six Gly ligands coordinate to thecentral La3+ ion with bond distances in the range of 2.764(6)–2.815(7) Å. The 12-coordinated icosahedron is completedby three μ3-OH− groups with an average bond distance of2.502(1) Å. There are three independent Ni2+ ions in thecluster. Ni1 and Ni2 have N2O4 donor sets and both arechelated by two Gly ligands in cis-mode from the equatorial

plane with average Ni–N and Ni–O bond distances of 2.05(2)and 2.038(7) Å respectively. A slightly distorted octahedral con-figuration around each nickel atom is completed by theadditional binding of two carboxylate atoms (one from a Glyand another from an IDA ligand with the correspondingaverage Ni–O distances of 2.06(2) and 2.078(1) Å, respectively)in axial positions. Ni3 adopts an octahedral six-coordinatedNO5 donor set which consists of one oxygen atom fromμ3-OH− group, two oxygen atoms from two Gly ligandsand NO2 donor atoms from a chelating IDA ligand. The Ni–Ocarboxylate bond distances are in the range of 2.042(7)–2.053(7) Å, close to that of the Ni–N bond (2.084(7) Å), butlonger than that of the Ni–OOH bond (1.973(5) Å).

In 1, each Gly ligand is employed to chelate one Ni2+ ionand also uses two oxygen atoms to coordinate to one Ni2+ andone La3+ ions, as shown in Scheme 1. The coordination modemay be described as μ3-ηO1:ηO1:ηO1N1. IDA ligand adopts aμ5-ηO1:ηO1:ηO2N1:ηO1:ηO1 coordination mode. Each IDA is usedto chelate one Ni2+ ion, but each of its carboxylate oxygenatoms is also employed to coordinate to one other metal ion.There are three OH− anions in the center of the cluster, andeach of them is used to bridge two La3+ and one Ni2+ ions. Thedeprotonated degree of the μ3-coordinated groups was con-firmed by BVS calculations (Table S4†). The average ∠La–O–Laand ∠La–O–Ni are 103.5(2)° and 110.3(3)°, respectively.

The 11 metal atoms are arranged in an interesting topology:the 9 Ni2+ ions form two octahedrons sharing one face and thetwo La3+ ions are located in the centers of the two octahedrons

Table 1 Crystal data and structure refinement of 1 and 2

1 2

Formula C36H108Cl9La3-N15Na3Ni9O94

C48H115Cl6Gd5-N18Ni12O95

Mr 3620.51 4168.05Cryst. system Hexagonal OrthorhombicSpace group P6̄2c P212121a/Å 15.000(2) 17.069(3)b/Å 15.000(2) 26.593(4)c/Å 29.211(9) 26.750(5)α/° 90 90β/° 90 90γ/° 120 90V (Å3), Z 5691(2), 2 12142(3), 4dcalcd, g cm−3 2.113 2.280F(000) 3612 8188θ range (°) 2.61–24.99 1.93–25.00Reflections collected/unique 25 475/3403 58 285/21 385R(int) 0.0453 0.0719Goodness-of-fit on F2 1.073 0.997R1

a (I > 2σ(I)) 0.0298 0.0449wR2

b (all data) 0.0796 0.1029Max/mean shift in final cycle 0.001/0.000 0.002/0.000Absolute structure parameter — 0.032(10)

a R1 = ∑(||Fo| − |Fc||)/∑|Fo|.bwR2 = {∑w[(F2o − F2c)]/∑w[(F2o)

2]}0.5,w = [σ2(F2o) + (aP)2 + bP]−1, where P = (F2o + 2 F2c)/3]. 1, a = 0.0511,b = 2.7365; 2, a = 0.0406, b = 0.0000.

Fig. 1 (a) Structure of the [La2Ni9(μ3-OH)3(IDA)3(Gly)12]3+ cationic cluster;

(b) metal skeleton of the cluster. Symmetry code: A: −y + 1, x − y, z; B: −x + y +1, −x + 1, z; (c) structure of the {Ni9} dioctahedron.

Scheme 1 The coordination modes of the ligands. (a) μ3-ηO1:ηO1:ηO1N1 for Gly;(b) μ2-ηO1:ηO1N1 for Gly; (c) μ5-ηO1:ηO1:ηO2N1:ηO1:ηO1 for IDA.

Dalton Transactions Paper

This journal is © The Royal Society of Chemistry 2013 Dalton Trans., 2013, 42, 5711–5717 | 5713

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article Online

respectively (Fig. 1b). The edge lengths (Ni2+⋯Ni2+) of theoctahedrons are in the range of 5.214(1)–5.398(1) Å, while theLa3+⋯Ni2+ (same octahedron) and La3+⋯La3+ separations are3.684(1)–3.770(1) and 3.929(1) Å, respectively. As we know, todate there is no polynuclear compound reported that containsthe same face-shared and Ln-centered dioctahedral structureas in 1. However, several [LaNi6(Gly)12]

3+ clusters whose struc-tures can be described as a La3+ ion located in the center of anoctahedral {Ni6} cage were reported.17,18 In these compounds,the average La⋯Ni and Ni⋯Ni separations are in the range3.63–3.71 Å and 5.13–5.25 Å, respectively, which are both con-sistent with those in 1.

Crystal structural of 2

Compound 2 crystallizes in the orthorhombic space groupP212121. The [Ln5Ni12(Gly)12(IDA)6(μ3-OH)9(H2O)3]

6+ cationiccluster in 2 bears a onion-like {Gd5}⊂{Ni12} structure. Thecentral {Gd5} cluster exhibits a trigonal bipyramidal structure,while the outer {Ni12} shell possesses a triangular orthobicu-pola framework (Fig. 2). The {Gd5} core and {Ni12} shell arebridged by μ3-OH− groups and carboxylate oxygen atoms. Withthe help of 9 μ3-OH− groups, the five Gd3+ ions are connectedinto a compressed trigonal bipyramid. The equatorial edges(∼6.2 Å) determined by Gd1, Gd2 and Gd3 ions are muchlonger than the side edges (∼4.0 Å). Besides, the axial sepa-ration (Gd4⋯Gd5) is only 3.4 Å. This configuration is very rarein {Ln5} clusters, most of which prefer to adopt a pyramidalskeleton19 and up to now only one {Dy5} compound bearing a

regular trigonal bipyramidal structure has been reported.20 Allthe Gd3+ ions adopt a nine-coordinated O9 donor set whichconsists of oxygen atoms either from a μ3-OH− group orcarboxylate groups and can be described as a distorted mono-capped square antiprism. The Gd–O(OH) and Gd–O(carboxy-late) bond lengths are in the range of 2.416(6)–2.593(8) and2.329(6)–2.492(6) Å, respectively.

The structure of the outer {Ni12} shell may be described as atriangular orthobicupola with Ni⋯Ni separations of 5.2 or7.7 Å. The 12 Ni2+ ions can be divided into three groups. Thefirst group (Ni1, Ni3 and Ni5) has an octahedral N2O4 donorset. Each metal ion is chelated by two N, O-donor Gly ligandsin cis-mode from the equatorial plane and further coordinatedby one terminal water molecule and one carboxylate atomfrom one IDA in axial positions. The coordination polyhedronof the second group (Ni2, Ni4 and Ni6) is almost the same asthat of the first group except that the axial position is occupiedby two carboxylate atoms from two IDA respectively. The metalion in the third group has an octahedral NO5 donor set whichconsists of one oxygen atom from a μ3-OH− group, one carboxy-late oxygen atom from a Gly ligand, one carboxylate oxygenatom from an IDA ligand and NO2 donor atoms from anotherchelating IDA ligand. The average Ni–N and Ni–O bondlengths are 2.07 and 2.06 Å, respectively, which are in theexpected range for related Ni–Ln complexes with Gly or IDA asligand.12,13,21

The Gly ligands in 2 have two coordination modes. Oneis μ3-ηO1:ηO1:ηO1N1, the same as that in 1. The other isμ2-ηO1:ηO1N1, in which the ligand uses its ON donor set tochelate a Ni2+ ion and the carboxylate oxygen atom to coordi-nate to a Gd3+ ion, as shown in Scheme 1b. IDA ligand adoptsa μ5-ηO1:ηO1:ηO2N1:ηO1:ηO1 coordination mode. For the nineμ3-OH− groups, three are used to bridge three Gd3+ ionsrespectively with ∠Gd–O–Gd in the range of 91.264(8)–115.251(8)°. Each of the remaining six are used to connect oneNi2+ and two Gd3+ ions with ∠Gd–O–Gd and ∠Gd–O–Ni in therange of 109.504(8)–110.663(8) and 100.251(9)–104.218(9)°,respectively.

Syntheses

Simultaneously using HGly and H2IDA as ligands for the syn-thesis of Ni–Ln clusters led to the formation of either undeca-nuclear {La2Ni9} or heptadecanuclear {Ln5Ni12} (Ln = Gd, Nd,Sm, Tb, Dy, Y) compounds whose structures are quite differentfrom those solely based on HGly or H2IDA ligand.10–13,21 Theformation of the products is not significantly affected by theratio of reactants. In experiments a wide range of ratios of Ln :Ni : HGly : H2IDA, such as 1 : 2 : 2 : 1, 1 : 4 : 2 : 1, 1 : 4 : 3 : 1,1 : 6 : 3 : 1 and 1 : 6 : 3 : 3, can produce the same products.However, the ratio of 1 : 6 : 3 : 1 provides the best yield of theproduct. Generally high nuclearity 3d–4f clusters wereobtained by long time for evaporation, sometimes even severalmonths.12b Herein, we tried a time-saving moderate temp-erature heating method (90 °C) and reactions were complete in24 h with satisfying yields.

Fig. 2 Structure of 2. (a) Overall structure of the [Gd5Ni12(Gly)12(IDA)6-(μ3-OH)9(H2O)3]

6+ cationic cluster; (b) metal skeleton of the cluster; (c) thecompressed trigonal bipyramid structure of the [Gd5(μ3-OH)9]6+ core; (d) thetriangular orthobicupola structure of the {Ni12} shell.

Paper Dalton Transactions

5714 | Dalton Trans., 2013, 42, 5711–5717 This journal is © The Royal Society of Chemistry 2013

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article Online

The radii of the 4f ions play a key role in determining theconfigurations of compounds 1–7. As the lanthanide ion’sradius decreases with increasing atomic number, light lantha-nide ions prefer high coordination numbers (12 for La3+ ion in1), while heavy lanthanide ions like low coordination numbers(9 for other lanthanide ions in 2–7). The difference inducescompounds 1 and 2–7 to adopt two completely different struc-tures. Similar phenomenon was also observed in other 3d–4fheterometallic clusters.11,12b,22 Compounds 1–7 are stable andcan remain crystalline after washing with water and exposureto air, as confirmed by the powder X-ray diffraction results.

Magnetic properties

The magnetic properties of 1–7 were studied on polycrystallinesamples. At room temperature, the χMT products (in which χMis molar magnetic susceptibility) of 1 and 7 are 13.22 and13.84 cm3 mol−1 K (Fig. 3), larger than the expected values of9.00 cm3 mol−1 K (calculated for nine spin-only Ni2+ ions (S =1, g = 2) and two diamagnetic La3+ ions) and 12.00 cm3 mol−1 K(calculated for twelve non-interacting Ni2+ ions (S = 1, g = 2)and five diamagnetic Y3+ ions) for 1 and 7 respectively. The dif-ference may be attributed to the ferromagnetic exchange.1e,2b

For 1, upon lowering the temperature, the χMT valuedecreases slightly to a minimum of 12.05 cm3 mol−1 K at 70 Kbefore it increases rapidly to a maximum of 16.63 cm3 mol−1 Kat around 4 K, indicating a very weak magnetic interaction.After that temperature, χMT decreases abruptly to 12.08 cm3

mol−1 K at 2 K which may be caused by the intercluster anti-ferromagnetic interactions and/or zero-field splitting of theNi2+ ions. For 7, upon cooling, χMT decreases slightly to aminimum of 13.19 cm3 mol−1 K at 50 K, and then increasesgradually to a maximum of 13.42 cm3 mol−1 K at 10 K, alsoindicating a very weak magnetic coupling. With furthercooling, the χMT value decreases abruptly to 10.26 cm3 mol−1 Kat 2 K, which can be ascribed to the same reasons as in 1.The overall magnetic behavior of the two compounds may

be described as ferrimagnetic, similar to that of other high-nuclearity Ni compounds.23,24

For 3, 4 and 5, the χMT values at 300 K are 23.20, 17.76 and73.08 cm3 mol−1 K, respectively, compared with the expectedvalues, 20.18 cm3 mol−1 K for 3 (five uncoupled Nd3+ (S = 3/2,L = 6, g = 8/11) and twelve Ni2+ ions (S = 1, g = 2)), 12.45 cm3

mol−1 K for 4 (five uncoupled Sm3+ (S = 5/2, L = 5, g = 2/7) andtwelve Ni2+ ions (S = 1, g = 2)) and 71.06 cm3 mol−1 K for 5(five Tb3+ (S = 3, L = 3, g = 3/2) and twelve Ni2+ ions (S = 1, g =2)). On lowering the temperature, the χMT products of 3 and 5decrease gradually to 17.36 cm3 mol−1 K and 69.08 cm3 mol−1 Karound 8 K and then quickly decrease to 11.31 cm3 mol−1 Kand 67.25 cm3 mol−1 K at 2 K, respectively. The χMT product of4 firstly decreases gradually to a plateau (15.86 cm3 mol−1 K)between 30 and 14 K, then decreases abruptly to 11.99 cm3

mol−1 K at 2 K. The decrease of χMT values may be ascribed toa combination of the antiferromagnetic interaction betweenmetal centers, the thermal depopulation of excited Stark sub-levels and zero-field splitting of the Ni2+ ion.12b,25

At 300 K, the observed χMT value of 82.92 cm3 mol−1 K for 6is consistent with the expected value of 82.83 cm3 mol−1 K forfive isolated Dy3+ (S = 5/2, L = 5, g = 4/3) and twelve Ni2+ ions(S = 1, g = 2). Upon cooling, the χMT value decreases sizablyand reaches a minimum of 75.52 cm3 mol−1 K at 52 K before itincreases gradually to a value of 76.45 cm3 mol−1 K at 15 K,then decreases rapidly to 57.34 cm3 mol−1 K at 2 K. Thedecrease of χMT at higher temperature region may be associ-ated with crystal field effects of the Dy3+ ion.2b,26 While a com-bination of the intermolecular antiferromagnetic interaction,the thermal depopulation of excited Stark sublevels of theDy3+ 6I15/2 state and zero-field splitting of the Ni2+ ion maybe responsible for the decrease of χMT below 15 K.27

As compounds 3–7 are isomorphous, we can deduce theNi⋯Ni interaction of 3–6 by subtracting the χMT (7) from theχMT of 3–6, similarly to those reported in the literature.18a,28

The ΔχMT values are the sum of Ln⋯Ln, Ln⋯Ni interactionsand the crystal field effect of the Ln3+ ions, as shown inFig. S8–S11.† The ΔχMT values for Nd–Y and Sm–Y decreaseslowly with decreasing temperature till 15 K, then decreaseabruptly and tend to zero as the temperature approaches absol-ute zero, suggesting that the Nd⋯Nd and Nd⋯Ni (or Sm⋯Smand Sm⋯Ni) interactions may be small or negligible. ΔχMTvalues for Dy–Y and Tb–Y both decrease slowly with decreasingtemperature. But the slight increase of ΔχMT value for Dy–Yin the 50–14 K region indicates possible Dy⋯Dy or Dy⋯Niferromagnetic interactions.

As shown in Fig. 4, the χMT value of 2 at 300 K is 53.80 cm3

mol−1 K, which is close to the expected value of 51.38 cm3

mol−1 K for five uncorrelated Gd3+ (S = 7/2, g = 2) and twelveNi2+ ions (S = 1, g = 2). Upon cooling, the χMT value increasesgradually before 50 K and then increases abruptly to 81.03 cm3

mol−1 K at 2 K, revealing the dominant Gd⋯Gd or Gd⋯Niferromagnetic coupling. The field-dependence of the mag-netization at 2 K appears to be saturated at 7 T to 59.2Nβ,which is close to the value expected if all the metal ions wereferromagnetically coupled (59Nβ).Fig. 3 The χMT vs. T plots of 1 and 3–7 under a 1000 Oe dc field.

Dalton Transactions Paper

This journal is © The Royal Society of Chemistry 2013 Dalton Trans., 2013, 42, 5711–5717 | 5715

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article Online

For 2, both the large value of the magnetization and domi-nate ferromagnetic interaction due to the presence of Gd3+

ions favor a large MCE. Thus, the field-dependent magnetiza-tions were measured in the ranges of 0–7 T and 2–10 K (insetof Fig. 4) and the equation ΔSm(T) = ∫ [∂M − (T,H)/∂T]HdH fromthe M(H,T) data was used to calculate the magnetic entropychange ΔSm.2 The results are shown in Fig. 5. It can be seenthat −ΔSm increases gradually with increasing ΔH, reaching amaximum value of 21.8 J kg−1 K−1 at 4 K for a field change of0–7 T. This value is larger than the reported value of17.6 J kg−1 K−1 (3 K, ΔH = 5 T) for a {Gd2Ni6} compound basedon valine ligand5 but lower than the value of 38.2 J Kg−1 K−1

(2 K, ΔH = 7 T) for a {Gd42Ni10} compound based on acetateligand,4b indicating that the MCE may be proportional tothe density of the gadolinium which is in agreement with thepreviously reported results.1c,3b,29 Moreover, the value isalso lower than the calculated maximum entropy value of46.1 J kg−1 K−1 based on −ΔSm = ∑Rln(2S + 1), correspondingto the sum of individual contributions from the Ni2+ and Gd3+

spins. The reason may be ascribed to the weak antiferromag-netic interactions between some Ni2+ ions, as observed in 7,and to the anisotropy of the Ni2+ ions.2b

Conclusions

In summary, an undecanuclear {La2Ni9} and six heptadecanuc-lear {Ln5}⊂{Ni12} compounds with a new metal skeleton weresuccessfully synthesized through a time-saving moderate temp-erature heating method. On the one hand, the result providesa good example of the construction of high-nuclearity 3d–4fheterometallic clusters through a mixed ligands method. Onthe other hand, compounds 2 exhibits ferromagnetic couplingand relative large magnetocaloric effect (21.8 J kg−1 K−1, ΔH =7 T) and may be a promising candidate for a magnetic refriger-ant. Work is underway to synthesize 3d–4f compounds basedon other mixed ligand systems and study their magneticproperties.

Acknowledgements

This work was supported by the National Science Foundationof China (21071025, 21101020 and 91122031), the Fundamen-tal Research Funds for the Central Universities (DUT12YQ04and DUT11LK22), and fund of Ministry of Education of China(20100041120021 and ROCS).

Notes and references

1 (a) J. B. Peng, Q. C. Zhang, X. J. Kong, Y. Z. Zheng,Y. P. Ren, L. S. Long, R. B. Huang, L. S. Zheng andZ. P. Zheng, Angew. Chem., Int. Ed., 2011, 50, 10649;(b) J. D. Leng, J. L. Liu and M. L. Tong, Chem. Commun.,2012, 48, 5286; (c) Y. Z. Zheng, E. M. Pineda, M. Helliwelland R. E. P. Winpenny, Chem.–Eur. J., 2012, 18, 4161;(d) M. Evangelisti and E. K. Brechin, Dalton Trans., 2010,39, 4672; (e) S. Nayak, M. Evangelisti, A. K. Powell andJ. Reedijk, Chem.–Eur. J., 2010, 16, 12865.

2 (a) V. K. Pecharsky and K. A. Gschneidner, Jr., J. Magn.Magn. Mater., 1999, 200, 44; (b) Y. Z. Zheng, M. Evangelistiand R. E. P. Winpenny, Angew. Chem., Int. Ed., 2011, 50,3692; (c) M. Evangelisti, F. Luis, L. J. D. Jongh andM. Affronte, J. Mater. Chem., 2006, 16, 2534.

3 (a) C. M. Liu, D. Q. Zhang and D. B. Zhu, Dalton Trans.,2010, 39, 11325; (b) Y. Z. Zheng, M. Evangelisti, F. Tunaand R. E. P. Winpenny, J. Am. Chem. Soc., 2012, 134, 1057;(c) G. Karotsis, M. Evangelisti, S. J. Dalgarno andE. K. Brechin, Angew. Chem., Int. Ed., 2009, 48, 9928.

4 (a) P. O. Adelani, A. G. Oliver and T. E. Albrecht-Schmitt,Chem. Sci., 2011, 2, 99; (b) J. B. Peng, Q. C. Zhang,X. J. Kong, Y. Z. Zheng, Y. P. Ren, L. S. Long, R. B. Huang,L. S. Zheng and Z. P. Zheng, J. Am. Chem. Soc., 2012, 134,3314; (c) A. S. Dinca, A. Ghirri, A. M. Madalan, M. Affronteand M. Andruh, Inorg. Chem., 2012, 51, 3935;(d) G. Karotsis, S. Kennedy, S. J. Teat, C. M. Beavers,D. A. Fowler, J. J. Morales, M. Evangelisti, S. J. Dalgarnoand E. K. Brechin, J. Am. Chem. Soc., 2010, 132, 12983.

Fig. 5 Calculated −ΔSm(T) from the magnetization data at different fields(0–7 T) and temperatures (2–10 K).

Fig. 4 The χMT vs. T plot of 2 under a 1000 Oe dc field. Inset: the field-dependent experimental magnetization plots at indicated temperatures.

Paper Dalton Transactions

5716 | Dalton Trans., 2013, 42, 5711–5717 This journal is © The Royal Society of Chemistry 2013

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article Online

5 A. Hosoi, Y. Yukawa, S. Igarashi, S. J. Teat, O. Roubeau,M. Evangelisti, W. Cremades, E. Ruiz, L. A. Barrios andG. Arom, Chem.–Eur. J., 2011, 17, 8264.

6 (a) J. Rinck, G. Novitchi, W. V. D. Heuvel, L. Ungur,Y. H. Lan, W. Wernsdorfer, C. E. Anson, L. F. Chibotaruand A. K. Powell, Angew. Chem., Int. Ed., 2010, 49, 7583;(b) P. P. Yang, X. L. Wang, L. C. Li and D. Z. Liao, DaltonTrans., 2011, 40, 4155.

7 (a) T. Liu, Y. J. Zhang, Z. M. Wang and S. Gao, J. Am.Chem. Soc., 2008, 130, 10500; (b) C. Papatriantafyllopoulou,W. Wernsdorfer, K. A. Abboud and G. Christou,Inorg. Chem., 2011, 50, 421; (c) M. Y. Li, Y. H. Lan,A. M. Ako, W. Wernsdorfer, C. E. Anson, G. Buth,A. K. Powell, Z. M. Wang and S. Gao, Inorg. Chem., 2010, 49,11587.

8 (a) V. M. Mereacre, A. M. Ako, R. Clérac, W. Wernsdorfer,G. Filoti, J. Bartolomé, C. E. Anson and A. K. Powell, J. Am.Chem. Soc., 2007, 129, 9248; (b) S. Nayak, O. Roubeau,S. J. Teat, C. M. Beavers, P. G. Gamez and J. Reedijk, Inorg.Chem., 2010, 49, 216.

9 (a) C. Lampropoulos, T. C. Stamatatos, K. A. Abboud andG. Christou, Inorg. Chem., 2009, 48, 429;(b) N. F. Chilton, S. K. Langley, B. Moubaraki andK. S. Murray, Chem. Commun., 2010, 46, 7787; (c) T. Birk,K. S. Pedersen, C. A. Thuesen, T. Weyhermüller,M. Schau-Magnussen, S. Piligkos, H. Weihe, S. Mossin,M. Evangelisti and J. Bendix, Inorg. Chem., 2012, 51,5435.

10 J. J. Zhang, S. M. Hu, S. C. Xiang, T. L. Sheng, X. T. Wu andY. M. Li, Inorg. Chem., 2006, 45, 7173.

11 J. J. Zhang, S. Q. Xia, T. L. Sheng, S. M. Hu, G. D. Leibeling,F. Meyer, X. T. Wu, S. C. Xiang and R. B. Fu, Chem.Commun., 2004, 1186.

12 (a) X. J. Kong, Y. P. Ren, L. S. Long, Z. P. Zheng,R. B. Huang and L. S. Zheng, J. Am. Chem. Soc., 2007, 129,7016; (b) X. J. Kong, Y. P. Ren, L. S. Long, Z. P. Zheng,G. Nichol, R. B. Huang and L. S. Zheng, Inorg. Chem., 2008,47, 2728.

13 X. J. Kong, Y. P. Ren, W. X. Chen, L. S. Long, Z. P. Zheng,R. B. Huang and L. S. Zheng, Angew. Chem., Int. Ed., 2008,47, 2398.

14 SMART, SAINT and SADABS, Bruker AXS Inc., Madison, WI,1998.

15 (a) G. M. Sheldrick, SHELXS-97, Program for X-ray CrystalStructure Determination, University of Gottingen, Germany,1997; (b) G. M. Sheldrick, SHELXL-97, Program for X-rayCrystal Structure Refinement, University of Gottingen,Germany, 1997.

16 P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A:Found. Crystallogr., 1990, 46, 194.

17 D. M. J. Doble, C. H. Benison, A. J. Blake, D. Fenske,M. S. Jackson, R. D. Kay, W. S. Li and M. Schröder, Angew.Chem., Int. Ed., 1999, 38, 1915.

18 (a) J. J. Zhang, S. C. Xiang, S. M. Hu, S. Q. Xia, R. B. Fu,X. T. Wu, Y. M. Li and H. S. Zhang, Polyhedron, 2004, 23,2265; (b) Y. Yukawa, G. Aromí, S. Igarashi, J. Ribas,S. A. Zvyagin and J. Krzystek, Angew. Chem., Int. Ed., 2005,44, 1997; (c) T. Peristeraki, M. Samios, M. Siczek, T. Lis andC. J. Milios, Inorg. Chem., 2011, 50, 5175.

19 (a) R. J. Blagg, C. A. Muryn, E. J. L. McInnes, F. Tuna andR. E. P. Winpenny, Angew. Chem., Int. Ed., 2011, 50, 6530;(b) M. T. Gamer, Y. H. Lan, P. W. Roesky, A. K. Powell andR. Clérac, Inorg. Chem., 2008, 47, 6581.

20 J. B. Peng, X. J. Kong, Y. P. Ren, L. S. Long, R. B. Huangand L. S. Zheng, Inorg. Chem., 2012, 51, 2186.

21 (a) J. J. Zhang, S. M. Hu, L. M. Zheng, X. T. Wu, Z. Y. Fu,W. X. Du, H. H. Zhang and R. Q. Sun, Chem.–Eur. J., 2002,8, 5742; (b) J. J. Zhang, S. M. Hu, S. C. Xiang, X. T. Wu,L. S. Wang and Y. M. Li, Polyhedron, 2006, 25, 1.

22 (a) A. J. Blake, R. O. Gould, C. M. Grant, P. E. Y. Milne,S. Parsons and R. E. P. Winpenny, J. Chem. Soc., DaltonTrans., 1997, 485; (b) J. J. Zhang, T. L. Sheng, S. M. Hu,S. Q. Xia, G. Leibeling, F. Meyer, Z. Y. Fu, L. Chen, R. B. Fuand X. T. Wu, Chem.–Eur. J., 2004, 10, 3963.

23 (a) C. Benelli, A. J. Blake, E. K. Brechin, S. J. Coles,A. Graham, A. G. Harris, S. Meier, A. Parkin, S. Parsons,A. M. Seddon and R. E. P. Winpenny, Chem.–Eur. J., 2000,6, 883; (b) T. C. Stamatatos, K. A. Abboud, S. P. Perlepesand G. Christou, Dalton Trans., 2007, 3861; (c) S. S. Tandon,S. D. Bunge, J. Sanchiz and L. K. Thompson, Inorg. Chem.,2012, 51, 3270.

24 X. J. Kong, L. S. Long, R. B. Huang, L. S. Zheng,T. D. Harris and Z. P. Zheng, Chem. Commun., 2009, 4354.

25 V. Chandrasekhar, B. M. Pandian, R. Boomishankar,A. Steiner, J. J. Vittal, A. Houri and R. Clérac, Inorg. Chem.,2008, 47, 4918.

26 C. Benelli and D. Gatteschi, Chem. Rev., 2002, 102, 2369.27 (a) V. Mereacre, A. M. Ako, R. Clérac, W. Wernsdorfer,

I. J. Hewitt, C. E. Anson and A. K. Powell, Chem.–Eur. J.,2008, 14, 3577; (b) C. Papatriantafyllopoulou,T. C. Stamatatos, C. G. Efthymiou, L. Cunha-Silva,F. A. A. Paz, S. P. Perlepes and G. Christou, Inorg. Chem.,2010, 49, 9743.

28 (a) M. L. Kahn, P. Lecante, M. Verelst, C. Mathonière andO. Kahn, Chem. Mater., 2000, 12, 3073; (b) J. P. Costes,F. Dahan, A. Dupuis and J. P. Laurent, Chem.–Eur. J., 1998,4, 1616; (c) E. Colacio, J. Ruiz-Sanchez, F. J. White andE. K. Brechin, Inorg. Chem., 2011, 50, 7268.

29 F. S. Guo, J. D. Leng, J. L. Liu, Z. S. Meng and M. L. Tong,Inorg. Chem., 2012, 51, 405.

Dalton Transactions Paper

This journal is © The Royal Society of Chemistry 2013 Dalton Trans., 2013, 42, 5711–5717 | 5717

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by D

alia

n U

nive

rsity

of

Tec

hnol

ogy

on 2

7/12

/201

3 00

:50:

56.

View Article Online