Crown-linked dipyridylamino-triazine ligands and their spin-crossover iron(ii) derivatives: Magnetism, photomagnetism and cooperativity

Dalton

Transactions

View Article Online

View Journal | View Issue

PAPER

Crown-linked dipyridylamino-triazine ligands and their

spin-crossover iron(II) derivatives: magnetism,

photomagnetism and cooperativity†

Cite this: Dalton Trans., 2013, 42, 16494

a

a,b

a

Hayley S. Scott, Tamsyn M. Ross, Nicholas F. Chilton, Ian A. Gass,

a

c

Boujemaa Moubaraki, Guillaume Chastanet, Nicolas Paradis,

Jean-François Lètard, Kuduva R. Vignesh, Gopalan Rajaraman, Stuart R. Batten

and Keith S. Murray*

c

d

a

The syntheses, crystallography and magnetic properties of a series of compounds of formula trans-

II

1

II

2

II

3

[

Fe (L )

NCX)

2

(NCX)

2

] (X = S, Se, BH

3

(1–3)), cis-[Fe (L )(NCX)

2

]·CH

2

Cl

2

(X = S, Se, BH

3

(4–6)) and trans-[Fe (L )-

1

2

4

(

2

]

n

(X = S, Se (7–8)) are described (L = 6-chloro-N ,N -diethyl-N ,N -di(pyridin-2-yl)-1,3,5-triazine-

2

4

2

,4-diamine, L = 6,6’-(1,4,10,13-tetraoxa-7,16-diazacyclooctadecane-7,16-diyl)bis(N ,N -diethyl-N ,N -di-

3

(

pyridin-2-yl)-1,3,5-triazine-2,4-diamine,

L

=

6,6’-(1,4,10,13-tetraoxa-7,16-diazacyclooctadecane-7,16-

diyl)bis(N ,N ,N ,N -tetra(pyridin-2-yl)-1,3,5-triazine-2,4-diamine)). The magnetostructural properties of

–8 have been probed in detail by variable temperature magnetic measurements and crystallographic

2

4

1

methods. 1–6 display mononuclear structures while 7 and 8 form 1-D chain structures. Complexes 4–6

2

have the potential to form 1D-chains via L bridging, but instead form mononuclear complexes. Mag-

netic studies show that complexes 1, 2, and 4 remain in the high-spin (HS) state at all temperatures. An

aged, dry, powdered sample of 3 gives an abrupt HS to LS transition (T1/2 = 200 K), while a freshly pre-

2

pared, powdered sample of 3·1.5H O displays thermal hysteresis (Δ = 7 K). Complexes 5, 6 and 7 undergo

a gradual spin transition with T1/2 values of 100 K, 150 K and 130 K, respectively. Cooperativity para-

meters are compared, with 3 showing cooperativity (positive C) and 5 and 6 showing anticooperativity.

Photomagnetic LIESST (light induced excited spin state trapping) studies were performed on complexes

Received 8th July 2013,

Accepted 31st July 2013

5

and 6 and reveal T(LIESST) values lower than 60 K. An attempt has been made to understand the elec-

DOI: 10.1039/c3dt51839f

tronic structure of complex 3 and its cooperativity behaviour using density functional methods, the calcu-

www.rsc.org/dalton

lations reproducing the sign and, in part, the magnitude of the cooperativity.

II

materials, particularly of Fe , display thermal hysteresis, and

this history-dependence of their magnetic state renders them

Spin crossover (SCO) materials have the property of molecular important in the development of a new generation of advanced

Introduction

1

bistability, in that they can exist in two diﬀerent electronic materials for applications in data storage, display devices, and

2

states at a particular temperature. A growing number of SCO molecular electronics.

Many studies have been reported into mainly monomeric

II

SCO Fe materials in attempts to understand the fundamental

3

a

interactions responsible for the strong cooperativity between

SCO centres in hysteretic compounds, such interactions being

School of Chemistry, Building 23, Monash University, Clayton, Victoria 3800,

Australia. E-mail: keith.murray@monash.edu; Fax: +61-3-99054597;

Tel: +61-3-99054512

4

due to ligands and counterions involved in hydrogen bonds,

π–π stacking, NH/π and CH/π interactions and covalent

linkers. In recent times concerted eﬀorts have been made to

combine the spin state switching found in Fe -based SCO

compounds with secondary functions such as conductivity,

porosity, and chirality to produce so-called multifunctional

or hybrid SCO materials.

The bidentate chelating ability of 2,2′-dipyridylamine (dpa)

in Fe complexes has led to their use in simple mono-

b

School of Chemistry, University of Manchester, Oxford Road, Manchester, M13 9PL,

5

6

UK

7

c

CNRS, Université de Bordeaux, ICMCB, 87 avenue du Dr. A. Schweitzer, 33608

II

Pessac, France

d

8

Department of Chemistry, IITB, Powai, Mumbai, 400076, India

Electronic supplementary information (ESI) available: Figures of molecular

9

10

†

structures and intermolecular contacts. Figures of Slichter–Drickamer fits.

Figures of photomagnetism and reflectivity for 5 and thermal magnetism for 7.

Table of intermolecular interactions. CCDC numbers 883899–883904, 945424

and 945423. For ESI and crystallographic data in CIF or other electronic format

see DOI: 10.1039/c3dt51839f

II

1

12

II

meric systems, coordination polymers and oxo-based Fe

1

6494 | Dalton Trans., 2013, 42, 16494–16509

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

1

3

II

polynuclear complexes. However, dinuclear Fe complexes

incorporating dpa-based ligands can show: (a) ferromagnetic

II

coupling between the two HS Fe ions in, for example, [Fe

pyatriz) Cl ](CF SO (where dpyatriz = 2,4,6-tris(dipyridin-2-

ylamino)-1,3,5-triazine) when the triazine ring is also involved

in coordination; (b) gradual, incomplete SCO transitions

centred around 265 K in [Fe

2

(d-

2

3

3 2

)

1

4

II

2

(dpyatriz)

2 2 2 3 2

(H O) (CH CN) ]-

1

4

(

ClO ) , and; (c) a full two step SCO transition in the complex

4

2 2 4 2 2

[Fe (ddpp) (NCS) ]·4CH Cl , (where ddpp = 2,5-di(2′,2″-dipyri-

dylamino)pyridine)) the latter study revealing the first struc-

1

5

II

tural details at the HS-LS plateau temperature. 1-D Fe

chains of dpa-based ligands show features such as solvent-

1

6,17

dependent SCO behaviour,

alternating and crystallogra-

II

17

phically distinct HS and LS Fe centres along the chains and

SCO behaviour dependent on the degree of π–π inter-

1

7,18

actions.

From the perspective of using crown-appended ligands in

III

SCO studies, early work by Maeda et al. on the series [Fe M-

+

(

salten)(L)] (where M = Li , Na , K , Rb , salten = 4-aza-hepta-

methylene-1,7-bis(salicylideneiminate, L = N-(4-picolyl)-aza-15-

crown-ether) showed that encapsulation of alkali metal ions in

the aza-crown ether results in guest-dependent magnetic pro-

1

9

perties, crystal structures were unfortunately not presented

on any example. We have recently developed dipyridylamino-

substituted s-triazine polytopic ligands containing one or two

mono-aza-15-crown-5 units in an attempt to investigate any

host-dependent spin-crossover properties induced by cation

binding, and this resulted in formation of a polymeric

II

+

heterometallic (Fe –Na ) SCO compound which underwent a

2

0

gradual, one step transition with a T1/2 value of ∼240 K. Our

recent studies on a flexible, pyridyl-based linker ligand, N,N′-

2

1

bis(4-pyridyl-methyl)diaza-18-crown-6 (bpmdc) showed that

changing the guest in, or near, the crown-ether portion of the

bpmdc ligand resulted in diﬀerent supramolecular packing

1

2

3

Fig. 1 Structures of ligands L (top), L (middle) and L (bottom).

II

2+

arrangements of the well-known [Fe (3-bpp) ] SCO cation,

2

with abrupt spin transitions being observed at higher T1/2

II

22

II

values than those in conventional [Fe (3-bpp)

2

]X

2

salts. The intermolecular interactions. Mononuclear Fe SCO com-

II

2

4

initial aim of that work was to bridge two Fe (3-bpp)(NCX)

pounds containing a related ligand, DDE, viz. (N ,N ,N ,N -tetra-

2

6

moieties with the 2-connecting bpmdc ligand in order to ethyl-N ,N -di(pyridin-2-yl)-1,3,5-triazine-2,4,6-triamine) have

1

2

probe the hoped-for dinuclear product for any host–guest similar aliphatic (diethyl) groups to those in L and L and

II

2+

influences on the spin transition. The [Fe (3-bpp) ] SCO show a crystallographic phase transition accompanied by an

2

cation turned out to be the more stable product, and crystal- order–disorder transition of the diethyl groups, while under-

lised readily. With these previous studies in mind, we have going a complete HS ↔ LS spin transition, with T1/2 values

2

3

2

made further developments in crown-appended systems and above 205 K. The L ligand, containing a diaza-18-crown-

II

have investigated the host-dependent magnetic properties of a 6 moiety, was designed to produce heterometallic Fe SCO

II

new series of Fe SCO compounds formed with a dipyridyl- compounds where the chelating dipyridylamino groups would

II

amino-substituted s-triazine ligand that contains a bridging form an Fe SCO centre of the type [Fe (dpa)

2

(NCX)

) while having the potential

for guests such as the s-block cations to be chelated within the

2

] (where

−

diaza-18-crown-6 moiety.

NCX = NCS , NCSe and NCBH

3

1

We present here the new dipyridylamino-triazine ligand L

2

4

(

6-chloro-N ,N -diethyl-N ,N -di(pyridin-2-yl)-1,3,5-triazine-2,4- crown ring. As in the above-mentioned crown-dipyridyl (bpmc)

diamine) and the new diaza-18-crown-6-containing polytopic work, our initial aim was to make dinuclear SCO compounds

2

ligand

L

(6,6′-(1,4,10,13-tetraoxa-7,16-diazacyclooctadecane- in which L would provide the (functional) bridge. What

2

4

7

,16-diyl)bis(N ,N -diethyl-N ,N -di(pyridin-2-yl)-1,3,5-triazine- appeared possible ‘on paper’ led, however, to monomer for-

,4-diamine)), their structural formulae being shown in Fig. 1. mation, with L being suﬃciently flexible to allow the dpa-

Both ligands also contain non-coordinating aliphatic groups triazine arms to wrap around the Fe ions in an ‘ear-muﬀ’ con-

in an eﬀort to introduce crystal packing eﬀects via weak formation. Many attempts made to insert group 1a cations

2

II

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 16494–16509 | 16495

View Article Online

Paper

into the crown, either at pre- or post- (Fe ) coordination stages, CH

Dalton Transactions

II

2

2 2

Cl , filtered and then diﬀused with Et O to produce X-ray

led only to monomer formation and, disappointingly, no quality yellow crystals. The yields for complexes 1–3 were gen-

cation binding. Thus, we present here work using the ligands erally quite poor, although yields for complexes 4–6 were

1

2

L

and L , including synthesis, structures and magnetic somewhat better. Additional attempts at creating 1D chain

II

2

+

and photomagnetic (LIESST; light-induced excited state species of the formula [Fe (L )(NCX) ]·M were performed,

2

spin trapping) behaviour on the series of complexes trans- however no species of this type could be isolated.

II

1

II

2

[

(

Fe (L )

2

(NCX)

2

] (X = S(1), Se(2), BH

3

(3)) and cis-[Fe (L )

2

-

The 1-D complexes 7 and 8 were formed in sealed tubes by

reacting L , Fe (BF ) ·6H O and NaNCS–KNCSe in 1 : 1 : 2 ratio

4 2 2

3

II

NCX) ]·CH Cl (X = S(4), Se(5), BH (6)).

We also present a (known) double-dpa-triazine 18-crown- in dimethylformamide. The tubes were heated on a heat-block

-linked ligand, L , and the first 1-D linear chain Fe deriva- to 130 °C for 16 h, after which time slow cooling to ambient

tives thereof, trans-[Fe (L )(NCX) ] (X = S, 7; X = Se, 8), with the temperature yielded X-ray quality yellow crystals.

2

3

2

4

3

II

6

II

3

2

formation of the crown-bridged, polynuclear species, rather

than mononuclear structures of the kind found in 4–6, poss-

ibly being due, at least in part, to the use of a markedly

Complexes 1–3 crystallise in the monoclinic space group P2

diﬀerent synthetic methodology.

Single crystal structural analyses

1

/n,

with the asymmetric unit containing half the mononuclear

complexes and the unit cell containing two complete mono-

nuclear complexes. The complexes 1–3 are structurally similar,

Results and discussion

diﬀering only in a slight variation of bond lengths and angles

1

II

Complexes 1, 2 and 3 were isolated by the reaction of L with (Table 1). In 1–3 each Fe is coordinated with distorted trans

II

Fe (ClO

4 2 2 4 2 2

) ·6H O (1 and 2) or Fe (BF ) ·6H O (3) and NaNCS octahedral geometry (cis angles (N–Fe–N), 83.1(2)–96.9(2)° (1),

(

1), KNCSe (2) and NaNCBH (3) in a 2 : 1 : 2 ratio in MeOH. 83.2(1)–96.9(1)° (2), 86.6(3)–93.4(3)° (3); trans, 180° (1–3); octa-

3

2

5

For 1 and 2, the resulting yellow solution was filtered and then hedral distortion parameter Σ = 41.2° (1), 41.4° (2) and 32.6°

diﬀused with Et

1

2

O to produce X-ray quality yellow crystals. For (3)). Two dpa groups from two L ligands chelate via their

3

, the yellow solution was reduced in volume to produce a pyridyl nitrogens to form an equatorial plane. Completing the

−

yellow precipitate, which was then dissolved in CHCl , filtered FeN donor set are the two anionic cyano-based N

and then diﬀused with Et

̲

3

6

2

O to produce small, X-ray quality, (X = S (1), Se (2), BH

3

(3)) which bind apically, through the

yellow crystals. Complexes 4, 5 and 6 were isolated by the reac- nitrogen, resulting in the complexes adopting the trans confor-

2

II

tion of L with Fe (ClO ) ·6H O (4 and 5) or Fe (BF ) ·6H O (6) mation (Fig. 2). Fe–N_N_C_Xbond distances are 2.097(4), 2.11(4)

4

2

4 2

2

3

and NaNCS (4), KNCSe (5) and NaNCBH (6) in a 1 : 1 : 2 ratio and 1.946(5) Å while average Fe–Ndpa bond distances are 2.205,

in MeOH. The resultant yellow precipitate was dissolved in 2.22 and 1.985 Å for complexes 1, 2 and 3 respectively.

Table 1 Selected bond lengths and angles for complexes 1–8

a

c

b

c

b

a

Bond(s)

1

2

3

4

5

6

7

8

Lengths (Å)

N(1)–Fe(1)

N(2)–Fe(1)

N(3)–Fe(1)

N(1′)–Fe(1)

N(2′)–Fe(1)

N(3′)–Fe(1)

2.097(4)

2.231(4)

2.210(4)

2.097(4)

2.231(4)

2.210(4)

2.110(4)

2.227(3)

2.213(3)

2.111(4)

2.227(3)

2.213(3)

1.946(5)

1.993(6)

1.981(6)

1.946(5)

1.993(6)

1.981(6)

2.105(2)

2.182(2)

2.234(2)

2.105(2)

2.182(2)

2.234(2)

2.101(4)

2.182(4)

2.216(4)

2.101(4)

2.182(4)

2.216(4)

2.094(3)

2.134(3)

2.154(3)

2.094(3)

2.134(3)

2.154(3)

1.9535(15)

2.0107(15)

2.0149(16)

1.9535(15)

2.0107(15)

2.0149(16)

1.942(2)

1.993(2)

2.011(3)

1.942(2)

1.993(2)

2.011(3)

Angles (°)

N(1)–Fe–N(2)

N(1)–Fe–N(3)

N(1′)–Fe–N(1)

N(1)–Fe–N(2′)

N(1)–Fe–N(3′)

N(2)–Fe–N(3)

N(1′)–Fe–N(2)

N(2)–Fe–N(2′)

N(3′)–Fe–N(2)

N(1′)–Fe–N(3)

N(2′)–Fe–N(3)

N(3′)–Fe–N(3)

N(1′)–Fe–N(2′)

N(1′)–Fe–N(3′)

N(3′)–Fe–N(2′)

N(1)–C(1)–X(1)

Fe–N(1)–C(1)

91.91(16)

91.45(16)

180.00

88.09(16)

88.55(16)

83.07(16)

88.09(16)

180.00

96.93(16)

88.55(16)

96.93(16)

180.00

91.91(16)

91.45(16)

83.07(16)

179.30(16)

173.80(16)

91.99(13)

91.44(13)

180.0

88.01(13)

88.56(13)

83.08(12)

88.01(13)

180.0

96.92(12)

88.56(13)

96.92(12)

180.00

91.99(13)

91.44(13)

83.08(12)

178.3(4)

175.6(4)

92.3(2)

92.5(2)

180.00

87.7(2)

87.5(2)

86.6(3)

87.7(2)

180.00

93.4(3)

87.50(2)

93.4(3)

180.00

92.3(2)

92.5(2)

86.6(3)

176.90(7)

174.80(6)

87.83(9)

91.34(9)

92.47(13)

91.45(9)

172.71(8)

82.49(7)

91.45(9)

178.95(10)

98.29(8)

172.71(8)

98.29(8)

85.56(10)

87.83(9)

91.34(9)

82.49(7)

178.3(3)

161.9(2)

87.57(2)

91.13(16)

92.4(2)

86.68(11)

92.57(11)

89.21(16)

91.20(11)

174.71(11)

83.94(10)

91.20(11)

177.02(14)

98.25(10)

174.71(11)

98.25(10)

86.09(14)

86.68(11)

92.57(11)

83.94(10)

178.50(4)

166.97(4)

92.00(7)

92.38(7)

180.00

88.00(7)

87.62(7)

85.56(7)

88.00(7)

180.00

94.44(7)

87.62(7)

94.44(7)

180.00

92.00(7)

92.38(7)

85.56(7)

179.35(18)

171.93(15)

92.12(10)

92.44(10)

180.00

87.88(10)

87.56(10)

86.02(10)

87.88(10)

180.00

93.98(10)

87.56(10)

93.98(10)

180.00

92.12(10)

92.44(10)

86.02(10)

179.3(3)

171.6(3)

90.77(16)

172.50(16)

82.79(15)

90.77(16)

177.6(2)

98.98(14)

172.50(16)

98.98(14)

86.1(2)

87.57(16)

91.13(16)

82.79(15)

178.3(5)

160.2(4)

a

b

c

1

00 K. 123 K. 173 K.

1

6496 | Dalton Trans., 2013, 42, 16494–16509

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

1

Fig. 2 Molecular structure of the L family of complexes 1–3. H atoms omitted

3

for clarity. Nitrogen, dark blue; X (X = S (1), Se (2), BH (3)), purple; chlorine,

green; iron(II), turquoise.

2

Fig. 3 Structure of the L family of complexes, 4–6. H atoms omitted for clarity.

1

and 2 have typical Fe–N bond lengths for high-spin (HS)

3

Nitrogen, dark blue; X (X = S (4), Se (5), BH (6)), purple; oxygen, red; chlorine,

II

Fe while the shorter Fe–N bond lengths in 3 suggest the Fe

green; iron(II), turquoise.

is in the low-spin (LS) state at the temperatures specified in

1

Table 1. The diethyl ‘arms’ of L are oriented perpendicular to

the triazine ring in complexes 1–3 (Fig. 2) pointing “inwards” four oxygen atoms within the crown, making them available

in 1 and 2 and “outwards” in 3 (Fig. S1†). The closest inter- for potential hydrogen bonding to solvent molecules.

molecular contacts in 1 and 2 are C–H⋯Cl interactions

C18⋯Cl1, 3.415 Å (1), 3.401 Å (2)). The closest C⋯Cl distance moiety in 4 and 5 which could not be modelled unambigu-

in 3 is 3.689 Å between C17 and Cl1 (Fig. S2 and Table S1†). ously, however it was determined that 6 contains a single

The closest Fe–Fe distances are 10.151 Å in 1, 10.336 Å in 2 CH Cl molecule. The closest intermolecular contacts from

and 10.664 Å in 3.

this CH Cl molecule are with the ether oxygen atoms from

There is some residual electron density above the crown

(

2

Complexes 4–6 crystallise in the orthorhombic space group the crown (C25⋯O1, 3.425 Å). There are further intermolecular

Pbcn, with the asymmetric unit containing half the monomer contacts with two dipyridylamine pyridyl groups (C9⋯Cl1,

and the unit cell containing four complete mononuclear com- 3.690 Å) and two diethyl arms from the diethylamine portion

2

plexes (Fig. 3). Complexes 4–6 are structurally similar, diﬀering of L (C18⋯Cl1, 3.661 Å) from neighbouring molecules

only in a slight variation of bond lengths and angles (Table 1). (Fig. S3 and Table S1†). While the residual electron density

II

In 4–6 each Fe is coordinated with distorted octahedral geo- above the crown moiety in 4 and 5 suggests a number of poss-

metry (cis angles, 82.49(7)–98.29(8)° (4), 82.79(15)–98.98(14)° ible solvent combinations, the microanalytical data strongly

(

5), 83.94(10)–98.25(10)° (6); trans angles, 172.71(8)–178.95(10)° suggests it is one CH

2 2

Cl molecule, as in 6. Numerous un-

(

4), 172.50(16)–177.6(2)° (5), 174.71(11)–177.02(14)° (6); octa- successful attempts were made to bind alkali-metal cations to

2

5

II

hedral distortion parameter Σ = 48.43° (4), 47.37° (5) and the free diazacrown in these Fe complexes. However, the

2

4

7.48° (6)). Both bidentate dpa groups from one L ligand structure shown in Fig. 3 indicates that such binding is un-

chelate via their pyridyl nitrogens, and the N

completed by the two anionic N

6

donor set is likely to occur due to the distorted nature of the crown upon

−

II

̲

CX ligands (X = S (4), Se (5), Fe binding. To date, no cation binding has been achieved in

II

2

BH₃(6)), with complexes adopting the cis conformation crystals of the cis-[Fe (L )(NCX)

2

] system. The closest Fe–Fe dis-

Fig. 3). Fe–NNCX bond distances are 2.102(3), 2.106(5) and tances are 11.129 Å in 4, 11.267 Å in 5 and 11.128 Å in 6.

(

2

.094(3) Å, while average Fe–N_d_p_abond distances are 2.206,

Complexes 7 and 8 are isostructural and crystallise in the

.200 and 2.144 Å for complexes 4, 5 and 6, respectively. Com- triclinic space group P1

ˉ. The metal atom lies on an inversion

2

plexes 4, 5 and 6 have typical Fe–N bond lengths for high-spin centre and the asymmetric unit contains a section of the 1-D

II

3

Fe at the temperatures specified in Table 1, which for 6 lies chain encompassing one half of the L ligand bound to the

II

on the spin transition. The diethyl arms from the diethylamine Fe centre and one NCX (Fig. 5). Similarly to complexes 1 to 6,

portion of L are oriented in opposite directions, perpendicu- Fe in 7 and 8 is coordinated with distorted octahedral geome-

lar to the triazine ring, in each of 4–6 (Fig. 3).

try (7: trans angles equal to 180.00°; and cis angles ranging

The flexible diaza-crown ‘backbone’ of L ligand is substan- between 85.56(7)–94.44(7); 8: trans angles equal to 180.00°;

tially bent, adopting an ‘ear-muﬀ’ shape and resulting in an and cis angles ranging between 86.02(10)–93.98(10)), the

all-cis conformation around Fe . This geometry exposes the former yielding an octahedral distortion parameter of Σ =

2

II

2

II

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 16494–16509 | 16497

View Article Online

Paper

Dalton Transactions

3

Fig. 4 Molecular structure of the L pair of complexes 7 and 8. H atoms omitted for clarity. Nitrogen, dark blue; X (X = S (7), Se (8)), purple; oxygen, red; chlorine,

green; iron(II), turquoise.

II

of the complex [Co (NCS)

2

(bpmdc)(H

2

O)

2

] in which 1-D chains

are formed by the two-connecting ligand N,N′-bis(4-pyridyl-

methyl)diaza-18-crown-6 (bpmdc), as discussed by Batten

2

1

et al. In that case the crown ring had a trans-pyridyl confor-

mation with water molecules hydrogen bonded to each side of

the ring.

Magnetic susceptibility studies

DC magnetic susceptibility measurements were performed on

polycrystalline samples of 1, 2, 4–8 (Fig. 6–8), on an aged, dry,

powdered sample of 3 (Fig. 6) and a freshly prepared powder

sample of 3 (Fig. 7) in the 4–300 K temperature range under an

applied field of 0.5 T. Since complex 3 yielded only a few crys-

tals suitable for X-ray diﬀraction, a bulk powder sample was

used for the magnetic measurements.

3

−1

For complex 1 the χ

M

T value of 3.49 cm K mol at 270 K

remains essentially constant down to 130 K then decreases to

Fig. 5 Asymmetric unit of complex 7 (isostructural to complex 8). H atoms

omitted for clarity. Nitrogen, dark blue; sulphur purple; oxygen, red; chlorine,

green; iron(II), turquoise.

3

−1

3

.04 cm K mol at 70 K, remaining at this value until ∼10 K,

3

−1

before a more rapid decrease to 2.55 cm K mol at 2 K. Thus,

II

Fe remains high-spin in character throughout the tempera-

ture range measured, in line with the Fe–N distances, the

3

II

3

5.28° (7) and 34.16° (8). The L ligands bridge Fe centres to rapid decrease in χ T noted below 10 K being due to zero-field

M

give 1-D chains, with one dpa group from either end of each and Zeeman splitting eﬀects. Complex 2 behaves similarly.

II

ligand coordinating to adjacent Fe atoms. Two trans N

̲C X The origin of the broad inflection at ∼70 K is unclear, occur-

ligands complete the N coordination sphere (Fig. 4). The Fe– ring also in 4, and might indicate some small degree of spin

6

N

NCX bond distances are 1.9535(15) Å (7, X = S); 1.942(2) (8, X = crossover.

Se) Å, while average Fe–Ndpa bond distances are 2.013 Å (7);

In contrast to the magnetism of 1 and 2, the NCBH com-

.002 Å (8), respectively, which suggests that the spin states of pound 3 undergoes a very abrupt spin transition, beginning

3

2

3

−1

the complexes are between high-spin (HS) and low-spin (LS) at with a HS plateau and χ T value of 3.8 cm mol K at 280 K,

M

1

00 K. The two remaining and uncoordinated dpa groups on and followed by an abrupt reduction to a χ T of around

M

3

−1

each L ligand do not participate in any inter- or intra-mole- 0.9 cm K mol , characteristic of a mainly LS state, upon

cular interactions. The closest Fe⋯Fe intra-chain and inter- cooling to 170 K. This spin transition has a T₁_/₂= 210 K and

chain distances are 18.770(7) and 10.486(2) Å (7) and 18.790(7) shows little or no hysteresis (Fig. 6).

and 10.620(2) Å (8). The 1-D chains adopt a stepwise/zig-zag

To further investigate the magnetic behaviour of 3, suscepti-

shape around the ligand moiety and pack together as shown bility measurements were performed for this complex on a

in Fig. S4 and S5.† There are no solvent molecules present in freshly prepared and powdered sample, as seen in Fig. 7. This

the lattice in either complex.

sample was a hydrate, as determined by microanalysis and

The conformation of the 18-crown-6 ring in 7 and 8, TGA data (see Experimental section) and displayed an abrupt

3

together with the Fe⋯Fe intra-chain separation, is reminiscent SCO with hysteretic behaviour. The plateau value of 3.2 cm

1

6498 | Dalton Trans., 2013, 42, 16494–16509

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

T vs. T for 1 (○), 2 (■) and 3 (●) (aged, dry powder) between

–300 K in an applied ﬁeld of 0.5 T.

Fig. 8 Plot of χ

M

T vs. T for 4 (■), 5 (○) and 6 (●) between 5–300 K in an

Fig. 6 Plot of χ

M

applied ﬁeld of 0.5 T.

2

hysteresis. This is in good agreement to the thermal hysteresis

observed in the magnetic susceptibility measurements (Fig. 7)

given that the scanning rates in the MPMS settle mode (6.7 K

−1

min ) are a little faster than the DSC measurement scan rates

5 K min⁻). The enthalpy changes, ΔH, associated with the

1

(

spin transition have been calculated independently for the

heating and cooling cycles and the average values reported. This

results in an average ΔH value of 2.7 kJ mol⁻while the corres-

ponding average change in entropy value, ΔS, is calculated to be

1

3.1 J mol⁻

1

K

−1

(using the relation ΔH = TΔS (T = (T_e_x_o+ T_e_n_d_o)/

1

2

= 206.4 K)). The thermal behaviour was consistent through

repeated heating and cooling cycles and the ΔH and ΔS values

II

are within the range of that expected for Fe spin transitions.

3

−1

For complex 4 the χ T value of 3.18 cm K mol at 287 K

M

3

−1

increases slightly to reach a value of 3.32 cm K mol at 119 K

3

and then decreases down to a final χ T value of 2.57 cm K

M

−

1

II

mol at 10 K suggesting the Fe remains high-spin in charac-

ter throughout the temperature range measured. Compound 5

is shown to undergo a gradual spin transition, when measured

between the temperature range 261–10 K, with T₁_/₂occurring

at ∼100 K as evidenced by the gradual decrease of the room

M 2

Fig. 7 Plot of χ T vs. T for 3·1.5H

O (●) (freshly prepared powder) between

5

–300 K in an applied ﬁeld of 0.5 T.

mol⁻K observed for the HS state, at room temperature,

1

3

−1

decreases abruptly to 1.0 cm mol K at 170 K (T₁_/₂≈ 203 K).

While in the heating mode, T1/2 is shifted to a higher temp-

erature of ∼210 K; thus ΔT is ≈7 K. PXRD studies of these

powder samples of 3 revealed them to be largely amorphous

and, thus, definitive comparison to the calculated diﬀracto-

gram was not possible, although we are confident the structure

is similar to that of the crystallographic sample.

3

−1

temperature χ T value of 3.3 cm K mol to a final value of

M

3

−1

0

.7 cm K mol

at 10 K. Hysteretic behaviour was not

observed in the heating and cooling modes. For complex 6 the

3

−1

χ T value of 3.34 cm K mol at 300 K remains constant until

M

around 220 K, whereupon a gradual spin transition occurs

3

−1

leading to a final χ

T₁_/₂value is around 150 K with no evidence of hysteretic behav-

iour. Complex 7 showed a gradual spin transition, with T₁_/₂

30 K (Fig. S12; ESI†). Although we could obtain a suitable

M

T value of 0.26 cm K mol at 10 K. The

To further investigate the thermal hysteresis (width of 7 K)

found in the magnetic susceptibility measurements of

∼

1

3

·1.5H O (Fig. 7) a diﬀerential scanning calorimetry (DSC)

2

single crystal of complex 8 for crystallographic studies we were

unable to obtain enough pure crystalline product for magnetic

susceptibility measurements.

measurement was undertaken on the same sample between

1

23 K and 273 K with a heating and cooling rate of 5 K min⁻

and the experimental curve between 170 K and 240 K is shown

in Fig. S6 (ESI†). An exothermic peak was observed at 203.2 K

during the cooling cycle and upon heating an endothermic

Cooperativity and anticooperativity in 3, 5 and 6

peak at 209.6 K is observed. The endothermic peak is shifted As indicated above, the polycrystalline mononuclear spin

from the exothermic peak by 6.4 K indicating a small thermal crossover complexes 3, 5 and 6 generally display no hysteresis,

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 16494–16509 | 16499

View Article Online

Paper

Dalton Transactions

3

1

although the freshly prepared powder sample of 3·1.5H

2

O

trans-[Fe(L)

2

(NCS)

2

]

monomers.

We briefly explore this

shows an abrupt, single step hysteretic crossover with a diﬀer- method for complex 3, below.

ence of 7 K between T1/2↑ and T1/2↓. The aged powder of 3 Turning again to the Slichter and Drickamer approach, we

shows the same sharp transition, at the same T1/2, but with no have obtained good and reliable fits to eqn (1) for complexes 3,

hysteresis. Thus, prior to fitting the γ_H_Sdata (see below) we 5 and 6 when using the following parameter sets:

1

anticipated that complex 3 showed stronger cooperativity than

the others. At a qualitative level, the intermolecular inter- (T1/2 = 200 K, C = 0.83)

actions summarised in the Introduction that commonly con-

5: ΔH = 11.7 kJ; ΔS = 121 J K⁻mol⁻¹; Γ = −1.85 kJ mol⁻¹

tribute to cooperativity, viz., van der Waals, hydrogen (T1/2 = 100 K, C = −1.11)

3: ΔH = 18.8 kJ; ΔS = 89.4 J K⁻mol⁻¹; Γ = 2.75 kJ mol⁻¹

1

5

f

4

e

6: ΔH = 24.0 kJ; ΔS = 162 J K mol⁻¹; Γ = −5.05 kJ mol⁻¹

−1

bonding, π–π stacking, might be expected to reveal some

such eﬀects that are present and dominant in 3 but not in 5 (T₁_/₂= 150 K, C = −2.02)

and 6, even though 3 has diﬀerent ligand milieu compared

The plots for best fit are shown in Fig. 9, for 5 and Fig. S7

to 5 and 6. In complex 3 there is an intermolecular contact and S8 (ESI†) for 3 and 6. Use of positive Γ values for 5 and 6

1

C–H⋯Cl from a diethyl arm of L to the triazine bound gave very poor fits and thus we are confident that, within the

chlorine of a neighbouring complex (C17⋯Cl1, 3.689 Å) (Fig. 2 limitations of the mean field model, the crown complexes 5

and S2, Table S1†). It is possible the solvated species 3·1.5H

does display stronger interactions compared to 3. Complex 6 cooperative. While these fits were for the non-hysteretic

shows intermolecular contacts from the CH Cl solvate with sample of 3 and the experimental DSC values were for the hys-

the ether oxygens from the crown (C25⋯O1′, 3.425 Å) and teretic sample 3·1.5H O, the ΔH and ΔS values from DSC are

there are further intermolecular contacts from the CH Cl some six times smaller. Other groups have noted such discre-

2

O

and 6 display anticooperative interactions whilst complex 3 is

2

3

2

solvate with two dipyridylamine pyridyl groups (C9⋯Cl1, pancies but not to the same degree.

.690 Å) and with two diethyl arms from the diethylamine In general, the values of ΔH, ΔS and Γ (and C), above, are

3

2

II

portion of L (C18⋯Cl1, 3.661 Å) from neighbouring molecules typical for Fe L (NCX) spin-crossover materials, with Γ for 6

2

3

0,33,34

(Fig. S3 and Table S1†). Although we cannot unambiguously larger in size than reported values.

However, there are

model the residual electron density above the crown in 5 as a few, if any, systems with negative Γ values reported to make

dichloromethane molecule, the microanalytical data on the comparisons to 5 and 6 for both of which the shapes and

bulk sample suggests a single dichloromethane solvate as in 6 slopes of the χ

M

T (or γHS) curves are ‘gradual’, albeit with

and as such we would expect similar intermolecular inter- diﬀerent T1/2 temperatures. When Γ is positive, larger Γ (or C)

3

actions to complex 6, described above.

values result in more abrupt curves, but we don’t have

However, it is diﬃcult to make definitive conclusions in enough data to make such correlations when Γ is negative. In

regard to diﬀerences in cooperativity from the above discus- comparing outcomes for 5 and 6, it may be noted that the ΔH

sion of weak crystallographic interactions. More quantitative and ΔS values are bigger in the X = NCBH₃case 6; such

2

6–28

analyses of cooperativity

have involved calculations of the increases in best-fit ΔH values when changing NCSe for

3

3,34

Slichter and Drickamer thermodynamic (mean-field) type NCBH

eqn (1)), to simulate spin transition γ_H_S(fraction of high-

3

have also been noted in other crossover siblings.

2

9

(

spin) data

ln½ð1 ꢀ γHSÞ=γHSꢁ ¼ ðΔHHS$LS þ Γð1 ꢀ 2γHSÞÞ=RT ꢀ ΔSHS$LS=R

ð1Þ

and deduce interaction parameters, Γ (or cooperativity factors

3

0

C = Γ/2RT1/2). Tchougréeﬀ et al. have recently discussed the

sign of Γ, with a positive sign being indicative of cooperative

interactions and a negative sign being indicative of antico-

operative interactions. This group used atom–atom potential

calculations to underpin the Γ values. It contrasts with the

3

1

work of Robert et al. who have calculated electrostatic Made-

lung fields acting on the Fe and N centres, by use of ab initio

CASSCF methods to deduce the cooperativity parameter Γ_p_o_l

(eqn (2)) where:

Γpol ¼ ΔQðδVHS ꢀ δVLSÞ

ð2Þ

Fig. 9 Plot of high-spin fraction, vs. temperature, for complex 5 with the red

Here δVHS and δVLS stand for the diﬀerences in potential upon

the Fe and N atoms, and ΔQ is the diﬀerence in atomic

line being the best ﬁt to the Slichter–Drickamer mean-ﬁeld model (eqn (1) in

nd

M

script) and the ΔH and ΔS and Γ values given in the script. The LS χ (TIP; 2

charges. This approach was applied to families of cis- and order Zeeman) contribution (see Fig. 8) has not been included.

1

6500 | Dalton Trans., 2013, 42, 16494–16509

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

Theoretical studies on 3

the weak interactions between the molecules, and Γpol,

accounting for the charge redistribution due to ligands, geo-

metry reorganization for diﬀerent spins and lattice expansion

within the system. Here the Γvdw is estimated as the energy

diﬀerence between the like spin and unlike spin in the crystal

lattice, i.e. Γvdw = (2ELS,HS − ELS,LS − EHS,HS). For complex 3 this

is estimated to be 70 K and this value is much smaller than

the value reported for other complexes. As this value is

related to intermolecular interactions, and as the crystal struc-

ture suggests that the complex 3 does not have any significant

intermolecular interactions, a small computed Γ_v_d_wis consist-

ent with this expectation. When taking into consideration only

To further understand the cooperativity, we have performed

DFT calculations on complex 3 to compute the Γ and the C

parameters and also to gain insight into the electronic struc-

ture of the HS molecule. For calculations we have employed

3

5

B3LYP-D2 including dispersion eﬀects of Grimme, as weak

interactions need to be accounted for. All calculations employ

3

1

3

6

the Ahlrich TZV basis set for all the atoms and have been

3

7

performed using the Gaussian 09 suite of programs. We have

analyzed the HS wavefunction of the complex 3 and the com-

puted spin density plot is shown in Fig. 10. The spin density

on the Fe atom is estimated to be 3.81 revealing a moderate

delocalization of spin densities to other atoms, particularly to

the coordinating N atoms. Within the two sets of nitrogen

Γ

vdw, the C parameter is estimated to be 0.21. Although the

sign is reproduced compared to the experiment, the magni-

tude is underestimated. Assuming Γvdw = 70 K, one can esti-

mate Γpol ∼ 260 K from the experimental measurement. This

is in line with the fact that the Γ_p_o_lhas been termed as the pro-

atoms in the coordination sphere (Npy and N of NCBH ), the

3

distinction is visible with nitrogen of the NCBH group having

3

a spin density of 0.03 while the Npy atoms have close to 0.05. A

3

1

minent factor controlling the C value. Unfortunately, since

the crystal structure of the HS form is unavailable, Γpol cannot

be directly computed from eqn (2), above. Using the approxi-

mation that the LS lattice structure also applies to the HS

form, we get an overestimation of the C to be 5.2, which is

perhaps not surprising.

smaller spin density on the N atom of NCBH group is due to

3

the presence of the BH group (weak σ donating ability) and

3

these electronic diﬀerences observed suggest that the axial lig-

ation can significantly influence the observed magnetic pro-

perties as evident in the complexes 1–3 studied. While only a

negligible amount of spin density is detected at the BH

3

group, other pyridine-type nitrogen atoms propagate spin den-

sities via spin polarisation (see green shades in Fig. 10).

However, as the ligands are large the propagation does not

reach toward the end of the molecular chain, revealing that

electronic factors such as spin polarization are unlikely to be

important in determining the SCO properties. Thus, the

expected cooperativity is purely electrostatic in nature.

The computed Mulliken charges reiterate this statement as the

charge computed on Fe is 0.98 for HS and 0.70 for the LS state

Reflectivity and photomagnetic studies on the crown-ligand

2

(

L ) – iron(II) SCO complexes 5 and 6

Optical reflectivity and photomagnetic LIESST (light induced

excited spin state trapping) studies were carried out on com-

pounds 5 and 6. The spin crossover can be monitored by fol-

lowing the thermal behaviour of the visible spectrum

measured by diﬀuse reflectance. The results for 5 and 6 were

similar, so the data for 6 are focussed upon here.

while the B atom of the BH group bears a significant negative

3

For both complexes, the absorption spectra at 280 K are

mainly made up of a large band around 850 nm that can be

charge both in the HS and LS state (∼−0.34).

To further probe the cooperativity eﬀects, we have employed

II

3

1

attributed to the d–d transition of Fe in its HS state. Upon

the protocol of Robert et al. and have attempted to compute

3

1

cooling, an intense band grows around 600 nm that reflects

the MLCT occurring in the LS state and indicates that the HS

to LS spin crossover occurs between 280 K to 80 K (Fig. 11 and

S9†). For 5, the small variation of the 850 nm band indicates

that some HS residue could be present at low temperature. In

the case of 6, an almost isosbestic point is observed at 740 nm,

indicating the occurrence of an equilibrium. In addition, the

band at 850 nm almost disappears at 80 K, indicating that the

spin-crossover is complete. From the reflectivity data recorded

at 600 nm (Fig. 11 and S9†), the spin crossover temperatures

can be estimated at 130 K for 5 (Fig. S9†) and 170 K for 6

the C parameter. This group computed the cooperativity

using CASSCF methodology. In their report, they found that

the Γ parameter has two contributions, Γ_v_d_w, accounting for all

(

Fig. 11). Moreover, in 6, the evolution of the reflectivity

signals at 550 nm and 850 nm are inversed in agreement with

the equilibrium observed in the absorption spectra.

Below 80 K, the 600 nm band decreases while the 850 nm

band increases (in 6 especially) as we decrease temperature.

This indicates the occurrence of photo-induced colour change

and therefore the LIESST eﬀect at the surface of the sample.

This is particularly visible on the reflectivity measurements at

600 nm (5) and 550 nm (6), where the signal increases,

Fig. 10 B3LYP-D2 computed spin density plot for complex 3. Here red and

green shade represent positive and negative spin densities.

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 16494–16509 | 16501

View Article Online

Paper

Dalton Transactions

Fig. 12 Thermal behaviour of the χ

M

T product of 6 before irradiation (○),

during irradiation (△) and in the dark after irradiation (■). The insert presents

the derivative of the thermal χ T as a function of T (in the dark after irradiation);

M

the minimum allows the determination of the T(LIESST) value.

baseline is recovered. The minimum value of the derivative

∂

M

χ T/∂T gives the value of the T(LIESST) temperature estimated

3

8

at 55 K for 5 and 48 K for 6.

Several kinetic runs were recorded to characterise the relax-

ation process (Fig. 13 and S11†). For both 5 and 6, the relax-

ation curves follow an exponential shape. To extract the

relaxation rate constant at each temperature, we employed a

stretched exponential law, with a distribution σ of the acti-

a

vation energy E . Fig. 13 and S10† show the fit of the kinetics.

The thermodynamic parameters E , k and σ were extracted

a

∞

from the Arrhenius plot (ln kHL vs. 1/T) and the best-fit para-

−

1

−1

meters for 5 are: E

a

= 540 cm , k

∞

= 1500 s , σ = 33 cm ;

and for 6, are: E = 196.4 cm , k_∞= 0.659 s , σ = 49.7 cm⁻¹.

−

1

−1

a

These low activation energies are probably indicative of little

communication occurring in the lattices of 5 and 6, in broad

agreement with the anticooperativity discussion above.

Fig. 11 Thermal dependency of the absorption spectra under light irradiation

(

top) and intensity of the reﬂectivity signal at 550 (green) and 850 nm (red)

II

2

3 2

bottom) for cis-[Fe (L )(NCBH ) ], 6.

indicating the population of the HS state. In 6, the decrease of

the 850 nm signal also indicates the occurrence of HS popu-

lation below 80 K.

For the photomagnetic studies on 5 and 6, irradiation was

performed at diﬀerent wavelengths and the best one was

found to be at in the green band (514 nm) with a power of

mW cm⁻. Only the LIESST eﬀect was observed and no

2

5

reverse-LIESST. When the photostationary limit was reached,

the T(LIESST) curve was recorded (Fig. 12 and S10†). The

M

increase of χ T from 10 to 40 K observed for 5 usually follows

II

from the Fe zero-field splitting. This increase is not observed

in 6, probably due to eﬃcient relaxation of the photo-induced

HS state. The maximum χ

the photoconversion eﬃciency, this being around 90% for 5

and 66% for 6. Above 40 K, the χ

M

T value gives some information on

Fig. 13 Plot of the diﬀerent relaxation kinetics recorded as a function of the

temperature (35, 40, 45, 47.5 and 50 K) for 6. The red lines are calculated using

M

T product decreases and the the parameters given in the text. The inset is the ﬁt to the Arrhenius equation.

1

6502 | Dalton Trans., 2013, 42, 16494–16509

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

3

forms discrete monomers (4–6) while L forms extended 1-D

Discussion

2

3

chains (7 and 8). The structural diﬀerences of L and L can be

II

1

2

Complexes 1–3 form a series of type trans-[Fe (L )

where they diﬀer only in N

2

(NCX)

2

], seen in Fig. 1, where the diethylamine arms found in L are

−

3

̲

and Se derivatives remain mainly high-spin, the stronger field (Fig. 3) there seems little reason on steric grounds for com-

3

X = BH complex shows spin-crossover. The T1/2 values in the plexes involving L not to form discrete monomers, as in 4–6,

2

II

2

L

complexes cis-[Fe (L )(NCX)

the expected changes for the ligand-field strengths NCS < The synthesis of 7 and 8 to form the 1-D chains, however, is

NCSe < NCBH . Only the powder sample of 3·1.5H O showed markedly diﬀerent to that of 4–6 (see Experimental section).

2 2 2

]·CH Cl , 4–6, likewise follow where the additional dpa group could remain uncoordinated.

3

2

thermal hysteresis. In attempting to rationalise the hysteresis The synthesis of 4–6 was undertaken at ambient temperature

and associated cooperativity in 3, we note that the most signifi- in methanol, whereas 7 and 8 are formed by heating in DMF at

cant feature of 3, at least when considering the crystal struc- 130 °C for 16 hours in a sealed vial and cooling to get crystal-

ture, is the presence of a single C–H⋯Cl contact (vide infra). line product. It is reasonable to suggest that the higher reac-

For a weakly hysteretic sample we would expect a degree of tion temperatures in the synthesis of 7 and 8 may play a large

cooperativity mediated by such interactions. We can postulate role in the formation of 1-D chains instead of discrete mono-

that the water of hydration molecule suggested by microana- mers as is the case in 4–6. There are other examples of temp-

lysis and TGA in the freshly prepared powder sample of 3 may erature aﬀecting the crystalline product, for example in

4

0

introduce some additional H-bonding intermolecular contacts, polynuclear MOF products made by hydrothermal methods.

structural information not available to us from our single Regarding the photomagnetic studies, compounds 5 and 6

crystal structure determination. Using the Slichter and Dricka- are sensitive to the LIESST eﬀect and T(LIESST) values were

2

9

mer model, fits to the χ

M

T vs. T data for 3 confirm the co- measured giving values of 55 K for 5 and 48 K for 6 which

= 100 K line in the graph of T(LIESST) vs.

where T(LIESST) = T₀− 0.3T₁_/₂. From databases on

The anticipated chain structure of complexes containing L , many mononuclear FeN coordination environments, this line

(NCX) coordination spheres, as

CX ligands being trans to each other, did not form. expected. Moreover, with stronger NCX ligand fields we

operative nature resulting in a positive cooperativity factor C as belong to the T

0

3

0,33,34

38

seen for similarly abrupt spin transitions.

T₁_/₂,

2

6

II

with the dpa chelating groups binding to separate Fe centres, is indicative of FeN

4

2

−

38

and the N

̲

2

The ‘ear-muﬀ’ conformation of L appears, from the data thus would expect higher T1/2 and lower T(LIESST) values. Com-

far obtained, to be a very stable form that favours formation of pounds 5 and 6 clearly follow this trend. Despite the large

mononuclear SCO products. Incorporation of an alkali-metal crown moiety in these complexes, no particular constraints

or alkaline earth metal cation into the crown prior to forming seem to be imposed to the coordination sphere. This is linked

II

the Fe compound would be expected to ‘flatten’ the crown to the small octahedral distortion parameter Σ = 47.37° (5) and

ring and assist polymer formation, but we have not been able 47.48° (6) in the HS state. This parameter has been shown to

2

to achieve cation binding to the L ligand to date. Finally the be important in the improvement of the photo-induced HS

dichloromethane solvate molecule present in 6 (and assumed state lifetime. Such weak Σ agree with the relatively small T

to be present in 4 and 5 based on microanalytical data) is (LIESST) values.

involved in several C–H⋯O and C–H⋯Cl intermolecular inter-

actions (Fig. S3 and Table S1†) which may play a part in the

gradual spin transitions in 5 and 6. In contrast to 3, use of the

2

9

Slichter and Drickamer model to fit the χ

M

T vs. T data for 5

Conclusion

and 6 resulted in negative cooperativity factors (C) suggesting

II

the observed gradual spin transitions are anticooperative in Six new Fe mononuclear complexes and two 1-D chain com-

3

9

nature. Guionneau et al. reported an inverse correlation plexes of dpa-triazine ligands have been synthesised and fully

between T1/2 and the change in the geometric distortion para- characterised, three of them showing spin crossover. The

meter ΔΘ for low-spin and high-spin states in a range of com- mononuclear compounds may be categorised into two

], where L is a bidentate diﬀerent series of discrete [Fe L (N ̲C X) ] complexes, with the

2 n 2

N-donor chelate. Unfortunately we have no structural details common variant within each series being the terminal N

II

plexes of type cis-[Fe (NCS)

2

(L)

5

f

−

̲C X

II

of 5 and 6, which are of the [Fe (NCS) (L) ] type, in the low- ligands. Both series contained diethylamino substituents in

2

spin state and so cannot make a direct comparison to this the triazine ring of the ligands to see if they engendered weak

correlation. intermolecular interactions that would influence cooperativity

-D chain structures were formed by ligand L , wherein in the spin transition. No such influence could be detected.

only one of the dpa moieties on each end of the crown linker The series with a linking crown ligand situated between two

3

1

II

2

binds to Fe , and the remaining dpa moieties remain un- dpa moieties, L , did not form the anticipated chain trans

coordinated between neighbouring 1-D chains (Fig. S4†). structures, but rather formed mononuclear cis species. 1-D

These uncoordinated dpa moieties are not involved in π–π chain complexes were formed, however, when using the crown-

3

stacking with equivalent groups on neighbouring chains, linked ligand L that has two dpa units at each end, only one

2

II

despite appearances in Fig. S4.† It remains unclear why L

of which coordinates to Fe . Compounds 5 and 6, containing

This journal is © The Royal Society of Chemistry 2013

Dalton Trans., 2013, 42, 16494–16509 | 16503

View Article Online

Paper

Dalton Transactions

2

4

the crown-linked ligand L , gave well resolved photomagnetic/ 6-Chloro-N ,N -diethyl-N ,N -di(pyridin-2-yl)-1,3,5-triazine-2,4-

1

LIESST behaviour and relaxation kinetics.

diamine (L )

Further, we have attempted to shed light on cooperativity in

the SCO examples by structural and more quantitative calcu-

lational approaches of the Slichter–Drickamer mean-field

4

5

,6-Dichloro-N,N-diethyl-1,3,5-triazine-2-amine

.84 mmol) and 2,2′-dipyridyl-amine (1.0 g, 5.84 mmol) were

(1.29

g,

dissolved in 30 ml of MeCN. DIPEA (2.9 ml, 16.6 mmol) was

added, and the solution was stirred under reflux for three

days. Following this, the solution was reduced in volume

in vacuo to give a yellow solid. 20 ml of deionised water was

2

9

approach (for 3, 5, 6) and by DFT and ab initio methods (for

), with broad agreement found between the methods but

3

without being able to pin down precise intermolecular struc-

tural reasons for the weak cooperativity noted.

3

added, and the product extracted with CHCl (2 × 30 ml). The

Further studies in this area will examine the confluence of

such cooperativity eﬀects and the weak intermolecular con-

tacts such as those reported here. We will also seek to investi-

gate the role, if any, of the synthetic methods used to form

organic extracts were combined and reduced, giving a brown

solid that was subsequently recrystallised from a 1 : 1 (v/v)

acetone–hexane solution to produce yellow needles, isolated by

filtration. Yield: 0.988

g

(44.5%). MW: 355.82. MP:

II

monomeric/1-D chain species in related Fe dpa based SCO

−1

1

2

61.1–161.5 °C. IR (ATR cm ): 3337 (m), 3039 (w), 2946 (w),

807 (w), 2358 (w), 2101 (w), 1995 (w), 1658 (m), 1593 (s), 1554

materials. Finally, in regard to cation binding to the crown

2

3

moieties in the L and L compounds, no binding has been

noted to date but this will be pursued further.

(

s), 1502 (s), 1449 (s), 1303 (m), 1244 (m), 1148 (m), 1095 (m),

027 (m), 984 (m), 934 (m), 882 (m), 839 (m), 771 (sh), 671 (m).

H-NMR (400 MHz, DMSO-d ): δ 8.35 (2H, m), 7.87 (2H, m),

1

6

7.58 (2H, m), 7.26 (2H, m); 3.51 (2H, q, J = 7.0 Hz), 3.23 (2H, q,

J = 7.0 Hz), 1.09 (3H, t, J = 7.0 Hz), 0.92 (3H, t, J = 7.80 Hz).

Experimental

Syntheses

1

3

C-NMR (500 MHz, CDCl ): δ 166.07, 155.39, 149.13, 145.13,

3

1

38.33, 123.74, 122.00, 110.38, 58.07, 13.66. MS (ESI+) m/z: 356

+

All manipulations were performed under aerobic conditions [M + H] , 378 [M + Na] . Microanalysis: Found C 56.88, H 5.29,

using chemicals as received, unless otherwise stated. Caution! N 26.31%. Calculated for C17H18ClN , C 57.38, H 5.10, N

7

Care should be taken when using the potentially explosive per- 27.55%.

II

chlorate Fe salts. 4,6-Dichloro-N,N-diethyl-1,3,5-triazine-2-

amine (below) was prepared using a modification to the con- 6,6′-(1,4,10,13-Tetraoxa-7,16-diazacyclooctadecane-7,16-diyl)bis-

4

1

3

2

4

ditions reported by Hermon et al. Ligand L was prepared (N ,N -diethyl-N ,N -di(pyridin-2-yl)-1,3,5-triazine-2,4-diamine)

using the method of Gamez et al.

2

4

2

(L )

1

L

(1.66 g, 4.67 mmol), diaza-18-crown-6 (0.6 g, 2.29 mmol)

4

,6-Dichloro-N,N-diethyl-1,3,5-triazine-2-amine

and diisopropylethylamine, DIPEA (1.2 ml, 6.87 mmol) were

dissolved in 40 ml of 1 : 1 (v/v) MeCN–toluene, and refluxed

over the course of four days. After this time, the solution was

reduced in vacuo to give a brown solid, which was subsequently

re-dissolved in 30 ml of deionised water and extracted with

CH Cl (2 × 30 ml). The organic extracts were combined and

reduced in vacuo to give an oﬀ-white coloured powder. This

powder was then washed with warm ethanol to yield a clean

white powder. Yield: 1.81 g (87.7%). MW: 901.07. MP:

Cyanuric chloride (3.5 g, 18.9 mmol) and N,N-diisopropylethyl-

amine (DIPEA) (2.9 ml, 16.6 mmol) were dissolved in 100 ml

THF and the solution was stirred and cooled to −30 °C. Separ-

ately, diethylamine (0.7 g, 9.57 mmol) was dissolved in 100 ml

of THF and added, dropwise, to the stirred solution of cyanuric

chloride–DIPEA kept at −30 °C, over the course of four hours.

After complete addition, the solution was left to reach room

temperature and stirred for a further 24 h. The solution was

then reduced in volume, in vacuo, to form a yellow solid which

was dissolved in 200 ml of deionised water. This was extracted

2

1

7

1

(

1

98.4–200.1 °C. H NMR (200 MHz, d

6

-DMSO): δ 8.30 (m, 4H),

.62 (m, 4H), 7. 55 (m, 4H), 6.99 (m, 4H), 3.65–3.07 (m, 32H),

1

3

.08 (t, 6H, J = 6.8 Hz), 0.91 (t, 6H, J = 6.8 Hz). C NMR

): δ 211.90, 165.98, 156.58, 148.50, 136.86,

23.29, 120.30, 70.59, 69.81, 63.48, 58.47, 48.22, 41.46, 18.61,

2

with Et O (2 × 50 ml) and the combined organic extracts were

500 MHz, DMSO-d

6

filtered, dried, and reduced in vacuo to give a yellow solid

product. The yellow solid was dissolved in boiling hexane, the

solution filtered, and the clear filtrate was slowly reduced

under a flow of N to give white crystals. Yield: 1.74 g (41.6%).

MW: 221.09. MP: 78.1–78.6 °C. IR (ATR cm ): 2996 (w),

2

1

7

+

3.53, 8.71. MS (ESI+) m/z: 901 [M + H] , 923 [M + Na] . IR

cm ): 2969 (m), 2894 (m), 2088 (w), 1588 (m), 1547 (s),

−

1

(

2

−

1

1519 (s), 1496 (s), 1474 (s), 1417 (s), 1388 (s), 1367 (m),

1

292 (s), 1259 (m), 1234 (m), 1190 (w), 1165 (w), 1142 (m),

075 (m), 995 (w), 932 (w), 879 (w), 808 (m), 774 (m). Micro-

952 (w), 2874 (w), 2114 (w), 1982 (w), 1582 (w), 1559 (s),

471 (s), 1432 (s), 1376 (m), 1347 (s), 1325 (s), 1226 (s),

154 (s), 1091 (m), 1038 (m), 980 (m), 904 (w), 809 (s), 791 (m),

analysis: Found C 60.92, H 6.85, N 24.54%. Calculated for

C H N O , C 61.31, H 6.71, N 24.87.

46 60 16 4

+

1

82 (m), 594 (w). MS (ESI+) m/z: 221 [M + H] . H-NMR

(

7

1

200 MHz, CDCl ): δ 3.62 (q, 4H, J = 7.1 Hz), 1.22 (t, 6H, J =

3

II

1

3

trans-[Fe (L ) (NCS) ] (1)

1

2

.1 Hz). C-NMR (400 MHz, CDCl

2.71. Microanalysis: Found C 38.25, H 4.68, N 25.28%. Calcu- L (20 mg, 0.056 mmol) was dissolved in 3 ml of MeOH. Separ-

3

): δ 170.14, 164.08, 42.63,

II

lated for C H Cl N C 38.03, H 4.56, N 25.34%.

ately, Fe (ClO ) ·6H O (9.8 mg, 0.027 mmol), NaNCS (4.5 mg,

7

10

2

4

4 2

2

1

6504 | Dalton Trans., 2013, 42, 16494–16509

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

0

.056 mmol) and 10 mg ascorbic acid were added to 3 ml Although the exact number of solvate waters is not known the

MeOH, and the mixture was stirred until all solids were dis- physical analyses (vide supra) are consistent with each other

solved. The Fe(NCS) solution was then added dropwise to the and with that of a hydrated sample.

2

1

solution containing L , and the resultant yellow solution was

filtered, and the filtrate placed in an open vessel within a

vapour-diﬀusion chamber containing diethyl ether. X-ray diﬀr- The L ligand (50 mg, 0.056 mmol) was dissolved in 3 ml of

II

2

2 2 2

cis-[Fe (L )(NCS) ]·CH Cl (4)

2

II

action quality crystals formed over the course of two days, and MeOH. Separately, NaNCS (9.2 mg, 0.112 mmol), Fe -

were isolated via filtration. Yield: 6 mg (9.7%). MW: 883.66. IR (ClO ) ·6H O (20.3 mg, 0.056 mmol) and 10 mg ascorbic acid

4

2

1

(

ATR cm⁻room temperature): 3336 (w), 3066 (w), 2976 (m), were dissolved in 3 ml of MeOH. The Fe(NCS)

2

solution was

2

1

6

934 (m), 2871 (w), 2052 (s), 1571 (s), 1501 (s), 1461 (s), added dropwise to the solution of L , giving a yellow solution,

401 (s), 1376 (s), 1361 (s), 1332 (m), 1282 (s), 1236 (sh), which was filtered. Over time, a precipitate formed, and it was

181 (s), 1153 (s), 1115 (m), 1087 (s), 1054 (m), 1024 (s), re-dissolved in 3 ml of CH Cl and placed in an open vessel

011 (s), 983 (s), 891 (s), 817 (m), 777 (s), 756 (s), 673 (s), within a vapour diﬀusion chamber containing diethyl ether.

32 (m). Microanalysis: % found C 49.06, H 4.11, N 25.04; % X-ray diﬀraction quality crystals formed after 1 week, and were

2

calculated for C36

H

36Cl

2

FeN16

S

2

, C 48.93, H 4.11, N 25.

isolated via filtration. Yield: 20 mg (31%). MW: 1073.09. IR

(

ATR cm⁻room temperature): 3941 (w), 3077 (w), 2971 (w),

1

II

1

trans-[Fe (L ) (NCSe) ] (2)

2 2

2930 (m), 2863 (m), 2059 (s), 1599 (s), 1558 (s), 1507 (s),

This complex was synthesised in low yield using a similar pro- 1479 (s), 1461 (s), 1433 (s), 1417 (s), 1362 (s), 1330 (s), 1300 (s),

cedure to that used for compound 1, except that KNCSe 1258 (s), 1233 (s), 1194 (s), 1169 (s), 1154 (s), 1110 (s), 1076 (s),

(

8.1 mg, 0.056 mmol) was used in place of NaNCS. Yield: 5 mg 1060 (s), 1029 (s), 1015 (s), 974 (s), 928 (s), 908 (s), 878 (s),

18.9%). MW: 977.45. IR (ATR cm room temperature): 2976 849 (s), 834 (s), 803 (s), 774 (s), 755 (s), 726 (s), 689 (s), 673 (s),

w), 2065 (s), 1673 (w), 1566 (s), 1499 (s), 1479 (s), 1468 (s), 628 (s). Microanalysis: % found C 50.78, H 5.46, N 21.35; %

−

1

407 (s), 1359 (s), 1334 (s), 1285 (s), 1237 (s), 1181 (s), 1158 (s), calculated for C H Cl FeN O S (4·CH Cl ), C 50.82, H 5.40,

49 62 2 18 4 2 2 2

1

090 (s), 1055 (s), 1011 (s), 984 (s), 822 (s), 778 (s), 757 (s), N 21.77.

6

74 (s), 632 (s). Microanalysis: % found C 43.18, H 2.90,

II

2

cis-[Fe (L )(NCSe)

2

]·CH

2

Cl

2

(5)

N 23.24; % calculated for C H B Cl FeN Se , C 44.24,

3

6

36

2

16

2

H 3.71, N 22.93.

Compound 5 was synthesised using a similar method to that

given for complex 4 but with KNCSe (8.1 mg, 0.056 mmol)

II

1

2 3 2 2

trans-[Fe (L ) (NCBH ) ] (3) and 3·1.5H O

used instead of NaNCS. Yield: 28 mg (40%). MW: 1166.87. IR

A similar method was employed to that used for 1 and 2, but (ATR cm⁻room temperature): 3503 (w), 3077 (w), 2970 (m),

1

II

using Fe (BF

4 2 2

) ·6H O (9.1 mg, 0.027 mmol) in place of Fe - 2930 (s), 2861 (m), 2360 (w), 2063 (s), 1600 (s), 1558 (s),

(

ClO ) ·6H O and with NaNCBH (3.5 mg, 0.056 mmol) used 1506 (s), 1478 (s), 1462 (s), 1433 (s), 1363 (s), 1330 (s), 1301 (s),

4

2

3

instead of NaNCS–KNCSe. However, once the precursor solu- 1260 (s), 1235 (s), 1155 (s), 1110 (s), 1060 (s), 1016 (s), 974 (s),

tions had been combined, the solvent was reduced suﬃciently 928 (s), 908 (s), 878 (s), 803 (s), 774 (s), 754 (s), 689 (s), 674 (s).

to a point where a precipitate formed in solution. The solution Microanalysis: % found C 46.73, H 4.82, N 19.63; % calculated

was then filtered and the yellow precipitate that was isolated for C49

was re-dissolved in a small amount of CHCl . This CHCl solu-

tion was placed in an open vessel within a vapour-diﬀusion

chamber containing diethyl ether, giving a small number of A similar method was employed to that used for 4, however

H

62Cl

2

FeN18

O

4

Se

2

(5·CH

2

Cl

2

), C 47.01, H 4.99, N 20.14.

3

II

2

cis-[Fe (L )(NCBH

3

)

2

] (6)

II

yellow crystals of 3 that were suitable for X-ray crystallographic Fe (BF

4

)

2

·6H

2

O (18.8 mg 0.056 mmol) was used instead of

II

analysis. Yield: 1 mg (4%). The bulk powder (3·1.5H O) was Fe (ClO ) ·6H O, and NaNCBH (7 mg, 0.112 mmol) used

2

4 2

2

3

synthesised as above where the MeOH solvent was reduced in instead of NaNCS. Yield: 15 mg (23%). MW: 1121.56. IR (ATR

volume forming a yellow precipitate which was isolated via fil- cm⁻room temperature): 2931 (w), 2862 (w), 2347 (m),

tration. This was then analysed using magnetic susceptibility, 2184 (s), 1601 (s), 1570 (s), 1559 (s), 1507 (s), 1481 (s), 1463 (s),

infrared, microanalytical, TGA and DSC measurements. MW: 1435 (s), 1417 (s), 1363 (s), 1331 (s), 1300 (s), 1260 (s), 1235 (s),

1

−

1

8

47.20 (anhydrous). IR (ATR cm

room temperature): 1156 (s), 1121 (s), 1085 (s), 1062 (s), 1016 (s), 975 (s), 928 (s),

3

1

390(br), 2933 (w), 2326 (m), 2158 (m), 1565 (s), 1500 (s), 909 (s), 879 (s), 804 (s), 775 (s), 756 (s), 690 (s), 675 (s), 631 (s).

480 (s), 1469 (s), 1407 (s), 1362 (s), 1286 (s), 1241 (s), 1181 (s), Microanalysis: % found C 53.78, H 6.31, N 22.54; % calculated

1

6

088 (s), 1023 (s), 985 (s), 823 (s), 797 (s), 779 (s), 759 (s), for C49

75 (s). The microanalysis suggested four solvate waters (%

2 4 2 2 2

H68Cl FeN18O B (6·CH Cl ), C 52.47, H 6.11, N 22.48.

II

3

trans-[Fe (L )(NCS) ] (7)

2

n

found C 47.18, H 5.00, N 24.24; Cl 7.38% calculated for

3

II

C

36

H

50

B

2

Cl

.71) whereas the TGA analysis (Fig. S13†) showed a gradual and NaNCS (2.9 mg, 0.036 mmol) were dissolved in 3 ml of di-

weight loss of ca. 3% between 25 °C and 230 °C, equivalent to methylformamide and placed in a sealed vial. The tube was

the loss of 1.5H O. The bulk powder sample has been labelled heated on a hot-plate and held at 130 °C for 16 h, after which

throughout the text as (3·1.5H O) based on the TGA analysis. time the vial was left to slowly reach ambient temperature.

2 4 2 4 2 2

FeN16O (3·4H O), C 47.04, H 5.48, N 24.38, Cl L (20 mg, 0.018 mmol), Fe (BF ) ·6H O (6.1 mg, 0.018 mmol)

7

2

Dalton Trans., 2013, 42, 16494–16509 | 16505

View Article Online

Paper

Dalton Transactions

Upon cooling, small yellow crystals appeared on the side of the measurements on 4 and 5 were performed at 173(2) K on a

vial. Enraf-Nonius Kappa CCD with Mo Kα radiation, λ = 0.71073 Å.

Yield: 7 mg (30.6%). MW: 1269. IR (ATR cm⁻room temp- X-ray data were processed using the DENZO program and

1

43

4

erature) 3432 (br), 2866 (m), 2052 (s), 1590 (s), 1552 (s), data was corrected for absorption using the SORTAV package.

1

%

524 (s), 1463 (s), 1407 (s), 1371 (s), 1295 (s), 1265 (s), 1177 (s), X-Ray measurements were performed at 100(2) K for 1, 3, 7

097 (s), 996 (s), 804 (s), 775 (s), 738 (s), 666 (s). Microanalysis: and 8 at the Australian Synchrotron MX2 beam-line. The data

4

5

found C 56.96, H 4.65, N 24.00; % calculated for collection and integration were performed within Blu-Ice

4

6

C H FeN O S (7), C 56.78, H 4.45, N 24.28.

and XDS

software programs. Crystallographic data and

6

0

56

22 4 2

refinement parameters for 1–8, given in Tables 2 and 3, were

II

3

trans-[Fe (L )(NCSe)

3

2

]

n

(8)

solved by direct methods (SHELXS-97), and refined

2

47

II

(SHELXL-97) by full least-squares on all F data. All other

non-hydrogen atoms in 1–8 were refined anisotropically and

all hydrogen atoms were placed in calculated positions.

Residual electron density located above the crown moiety in

complexes 4 and 5 was unable to be assigned unambiguously

L (20 mg, 0.018 mmol), Fe (BF ) ·6H O (6.1 mg, 0.018 mmol)

4

2

and KNCSe (5.2 mg, 0.036 mmol) were dissolved in 3 ml of di-

methylformamide and placed in a sealed vial. The tube was

heated on a hot-plate and held at 130 °C for 16 h, after which

time the vial was allowed to slowly reach ambient temperature.

Upon cooling, small yellow crystals appeared on the side of the

vial. Yield: 6 mg (24.5%). MW 1363. IR (ATR, cm , room temp-

erature) 3443 (br), 2864 (m), 2054 (s), 1591 (s), 1553 (s),

524 (s), 1464 (s), 1409 (s), 1372 (s), 1295 (s), 1266 (s), 1177 (s),

097 (s), 997 (s), 805 (s), 775 (s), 744 (s), 666 (s). Microanalysis:

4

8

and so was treated with SQUEEZE which suggested a rela-

tively small amount of electron density (4.5 e per complex for

−

1

4

, 6.5 e per complex for 5). This is in contrast to the micro-

analysis which suggests the presence of a single dichloro-

methane molecule. CCDC numbers 883899–883904 (1 to 6),

1

%

945424 (7) and 945423 (8).

found C 52.04, H 4.05, N 22.36; % calculated for

Se , C 52.87, H 4.14, N 22.61.

C

60

H

56FeN22

O

4

2

Physical measurements

X-ray crystallography

Elemental analyses (C, H, N) were performed by the Campbell

X-ray crystallographic measurements on 2 and 6 were per- Microanalytical Laboratory, University of Otago, Dunedin, New

formed at 173(2) K (2) and 123(2) K (6) using a Bruker Smart Zealand. Infrared spectra were recorded on a Bruker Opus/IR

Apex X8 diﬀractometer with Mo Kα radiation, λ = 0.71073 Å. IFS 55 spectrophotometer in a range 4000–400 cm⁻. Diﬀeren-

Single crystals were mounted on a glass fibre using oil. The tial scanning calorimetry (DSC) was performed on a DSC Q100

data collection and integration were performed within SMART series from TA instruments with 4–5 mg of sample in closed

and SAINT+ software programs, and corrected for absorption aluminium pans, at a ramp rate of 5 K per minute. Two

1

4

2

using the Bruker SADABS program. X-ray crystallographic cooling–heating cycles were performed where complex 3 was

Table 2 Crystallographic details for complexes 1–6

1

2

3

4

5

6

Formula

FW (g mol

T (K)

C

36

H

36Cl

2

FeN16

S

2

C

36

H

36Cl

2

FeN16Se

2

C

36

H

42Cl

2

FeN16

B

2

C

48

H

60FeN18

O

4

S

2

C

48

H

60FeN18

O

4

Se

2

C

49 68 2 2 4

H B Cl FeN18O

−1

)

883.68

100

Monoclinic

977.48

173

Monoclinic

847.23

123

Monoclinic

1073.11

173

Orthorhombic

1166.91

173

Orthorhombic

1121.58

123

Orthorhombic

Crystal system

Space group

Z

P2

2

1

/n

P2

2

1

/n

P2

2

1

/n

Pbcn

4

Pbcn

4

Pbcn

4

a (Å)

b (Å)

c (Å)

10.151(2)

16.530(3)

12.110(2)

90

10.336(1)

16.467(2)

12.172(1)

90

10.664(2)

16.148(3)

11.932(2)

90

16.646(3)

19.181(4)

17.078(3)

90

16.781(3)

19.268(4)

17.107(3)

90

16.657(1)

19.263(1)

16.898(1)

90

α (°)

β (°)

105.50(3)

90

1958.1(7)

1.499

104.658(3)

90

2004.1(3)

1.620

101.77(3)

90

2011.5(7)

1.399

90

5422.1(4)

1.374

0.439

γ (°)

V (Å )

3

5452.9(19)

1.307

0.413

5531.3(19)

1.401

1.648

−

3

ρcalc (g cm )

−1

μ (mm

)

0.680

2.377

0.558

Measured/

independent (Rint

24 043/3254

(0.0386)

12 951/4591

(0.0732)

14 356/4022

(0.0667)

37 799/6906

(0.1000)

20 679/6673

(0.1338)

53 578/6221

(0.0870)

)

reflections

Observed

3062

2608

3081

4418

3035

4372

a

b

0.0532, 0.0844

0.1216, 0.1034

0.971

0.1213, 0.2990

0.1448, 0.3176

1.052

0.0648, 0.1567

0.1108, 0.1751

1.075

0.0855, 0.1813

0.1933, 0.2171

0.946

0.0676, 0.1839

0.0984, 0.2049

1.045

0.0668, 0.1936

1.161

a

b

2

o

2 2

2

o

1/2

R

1

o

= ∑||F | − |F

c

||/∑|F

o

|. wR

2

= {∑[w(F

∑F

c

) ]/∑[w(F

)]}

.

1

6506 | Dalton Trans., 2013, 42, 16494–16509

View Article Online

Dalton Transactions

Paper

green light (λ = 532 nm at 5 mW cm⁻was found to be the

2

Table 3 Crystallographic details for complexes 7 and 8

most eﬃcient) until photosaturation was reached. Then, in the

7

8

absence of irradiation, the temperature was increased at a rate

of 0.3 K min⁻. The extreme of the δχ

1

T/δT vs. T plot gave the

Formula

FW (g mol

T (K)

C

60

1 22

H56Fe N O

S

4 2

C

60 1 22 4 2

H56Fe N O Se

M

−1

)

1269.24

100 (2)

Triclinic

P1

1

10.486(2)

12.423(3)

13.133(3)

71.52(3)

72.10(3)

66.56(3)

1456.0(5)

1.448

1363.04

100 (2)

Triclinic

Pˉ1

T(LIESST) values for compounds 5 and 6, defined as the temp-

erature for which the light-induced HS information is

Crystal system

Space group

Z

a (Å)

b (Å)

c (Å)

α (°)

β (°)

3

8

ˉ

erased. At 10 K, compounds 5 and 6 were again irradiated to

photosaturation, and in the absence of irradiation the relax-

ation kinetics at 35, 40, 45, 47.5 and 50 K were measured for

compound 6;. The relaxation kinetics were measured at 10, 20,

45, 50, 54, 56, 60, 65, 70 and 75 K for 5. See ESI† for further

details.

1

10.620(2)

12.570(3)

13.150(3)

70.65(3)

71.26(3)

65.99(3)

1477.2(5)

1.532

γ (°)

V (Å )

1

13

3

H NMR and C NMR spectra were recorded on a Bruker

DPX 200 MHz, Bruker Avance 400 MHz and a Varian Inova

00 MHz. Electrospray ionization mass spectra (ESI-MS) were

recorded with Micromass (now Waters) ZMD with

−

3

ρ

calc (g cm )

−1

μ (mm

)

0.401

1.557

5

Measured/independent

int) reflections

Observed reflections

I > 2σ(I)]

56 483, 8096

(0.0509)

7449

46 822, 6541

(0.0588)

5450

a

(

R

Waters alliance e2695 HPLC system for automatic sample

[

a

b

injections. MeOH was the mobile phase and had a flow rate of

R

1

, wR

2

[I > 2σ(I)]

(all data)

0.0508, 0.1340

0.0543, 0.1382

1.050

0.0476, 0.1286

0.0581, 0.1356

1.040

00 μL min⁻

1

.

1

2

Goodness-of-fit on F

a

b

2

o

2 2

2

o

1/2 1/2

R

1

o

= ∑||F | − |F

c

||/∑|F

o

|. wR

2

= {∑[w(F

∑F

c

) ]/∑[w(F

)]} ]}

.

Acknowledgements

cooled to 123 K, held for 3 minutes and heated to 273 K. This work was supported by an ARC Discovery Grant to KSM.

Thermal scans below room temperature were calibrated via the KSM, KRV and GR thank the Australia-India DEST-AISRF

−

1

cyclohexane solid–solid transition (186.1 K, 6.70 kJ mol ) and program and the Monash-IITB Academy for financial and

−1

melting point (279.7 K, 2.63 kJ mol ). Transition tempera- scholarship support. Part of this research was undertaken

tures are reported using the peak maximum of the thermal on the MX1 beamline at the Australian Synchrotron, Victoria,

transition. Temperature values were determined with a ±4.8 K Australia. The authors would like to thank Dr Rachel Williams

accuracy and enthalpy scans were estimated with an experi- from the MX1 beamline at the Australian Synchrotron for her

mental uncertainty of ±12% for a scan rate of 10 K min⁻

1

.

excellent help. Dr Suzanne Neville is thanked for valuable

Variable temperature solid state direct current (dc) mag- discussions.

netic susceptibility data down to 5 K were collected with an

applied field of 0.5 T on a Quantum Design MPMS 5 T

magnetometer calibrated by use of a standard palladium

Notes and references

sample (Quantum Design) of accurately known magnetisation

or by use of magnetochemical calibrants such as CuSO ·5H O.

Microcrystalline samples were dispersed in Vaseline to avoid

torquing of the crystallites. The sample mulls were contained

in a calibrated capsule held at the centre of a drinking straw

that was fixed at the end of the sample rod.

4

2

1 See spin crossover articles by various authors in: (a) Spin-

crossover in Transition metal compounds, ed. P. Gütlich and

H. A. Goodwin, Top. Curr. Chem., Springer, Berlin, 2004,

pp. 233–235; (b) Spin Crossover Materials. Properties and

Applications. ed. M. A. Halcrow, Wiley, London, 2013;

(c) Cluster issue “Spin Crossover Complexes”. Eur. J. Inorg.

Chem., 2013, 573–1067.

2 (a) P. Gütlich, Y. Garcia and H. A. Goodwin, Chem. Soc.

Rev., 2004, 29, 419–427; (b) O. Kahn, Nature, 1999, 21, 319.

3 K. S. Murray and C. J. Kepert, Top. Curr. Chem., 2004, 233,

195.

4 (a) M. Yamada, M. Ooidemizu, Y. Ikuta, S. Osa,

N. Matsumoto, S. Iijima, M. Kojima, F. Dahan and

J.-P. Tuchagues, Inorg. Chem., 2003, 42, 8406;

(b) G. S. Matouzenko, G. Molnar, N. Brefuel, M. Perrin,

A. Bousseksou and S. A. Borshch, Chem. Mater., 2003, 15,

550; (c) Y. Ikuta, M. Ooidemizu, Y. Yamahata, M. Yamada,

S. Osa, N. Matsumoto, S. Iijima, Y. Sunatsuki, M. Kojima,

F. Dahan and J.-P. Tuchagues, Inorg. Chem., 2003, 42, 7001;

(d) Y. Sunatsuki, Y. Ikuta, N. Matsumoto, H. Ohta,

Photomagnetic characterisation for compounds 5 and 6

were carried out at CNRS, Université de Bordeaux, ICMCB,

with a Spectrum Physics Series 2025 Kr+ laser coupled through

an optical fibre into the cavity of the MPMS-55 Quantum

design SQUID magnetometer operating at 2 T. Samples were

prepared as a thin layer (ca. 0.1 mg) to promote full pene-

tration of the irradiated light. The sample weight was obtained

by comparing its thermal spin transition behaviour with an

3

8

accurately weighed sample. The sample was first slowly

cooled to 10 K by ensuring that potential trapping of HS

species at low temperatures did not occur. Irradiation to

photosaturation was carried out a number of times at diﬀerent

wavelengths to determine which source was most eﬃcient at a

−

2

power intensity of 5 mW cm (to prevent warming of the

sample). Samples were then cooled to 10 K and irradiated with

Dalton Trans., 2013, 42, 16494–16509 | 16507

View Article Online

Paper

Dalton Transactions

M. Kojima, S. Iijima, S. Hayami, Y. Maeda, S. Kaizaki, 16 M. Quesada, V. A. de la Pena-O’Shea, G. Aromí, S. Geremia,

F. Dahan and J.-P. Tuchagues, Angew. Chem., Int. Ed., 2003,

2, 1614; (e) B. Weber, W. Bauer, T. Pfaﬀeneder,

C. Massera, O. Roubeau, P. Gamez and J. Reedijk, Adv.

Mater., 2007, 19, 1397.

4

M. M. Dîrtu, A. D. Naik, A. Rotaru and Y. Garcia, 17 S. M. Neville, B. A. Leita, D. A. Oﬀermann, M. B. Duriska,

Eur. J. Inorg. Chem., 2011, 3193.

B. Moubaraki, K. W. Chapman, G. J. Halder and

5

(a) J. F. Lètard, P. Guionneau, E. Codjovi, O. Lavastre,

K. S. Murray, Eur. J. Inorg. Chem., 2007, 1073.

G. Bravic, D. Chasseau and O. Kahn, J. Am. Chem. Soc., 18 S. M. Neville, B. A. Leita, G. J. Halder, C. J. Kepert,

1

997, 119, 10861; (b) S. Hayami, Z.-Z. Gu, M. Shiro,

B. Moubaraki, J.-F. Létard and K. S. Murray, Chem.–Eur. J.,

2008, 14, 10123.

Y. Einaga, A. Fujishima and O. Sato, J. Am. Chem. Soc.,

2

000, 122, 7126; (c) R. Boča, M. Boča, L. Dlháň, M. Vrbova 19 Y. Maeda, M. Suzuki, S. Hirose, S. Hayami, T. Oniki and

and R. Werner, Inorg. Chem., 2001, 40, 3025; (d) J. A. Real, S. Sugihara, Bull. Chem. Soc. Jpn., 1998, 71, 2837.

B. Gallois, T. Granier, F. Suez-Panama and 20 T. M. Ross, B. Moubaraki, S. M. Neville, S. R. Batten and

J. Zarembowitch, Inorg. Chem., 1992, 31, 4972; K. S. Murray, Dalton Trans., 2012, 41, 1512.

e) Z.-J. Zhong, J.-Q. Tao, Z. Yu, C.-Y. Duan, Y.-J. Liu and 21 M. B. Duriska, S. M. Neville and S. R. Batten, Chem.

(

X.-Z. You, J. Chem. Soc., Dalton Trans., 1998, 327;

Commun., 2009, 5579.

(

f) M. A. Halcrow, Chem. Soc. Rev., 2011, 40, 4119.

22 I. A. Gass, S. R. Batten, C. M. Forsyth, B. Moubaraki,

C. J. Schneider and K. S. Murray, Coord. Chem. Rev., 2011,

255, 2058.

6

(a) S. Arata, H. Torigoe, T. Iihoshi, N. Matsumoto, F. Dahan

and J.-P. Tuchagues, Inorg. Chem., 2005, 44, 9288;

(

b) D. L. Reger, J. R. Gardinier, W. R. Gemmil, M. D. Smith, 23 T. M. Ross, B. Moubaraki, K. S. Wallwork, S. R. Batten and

A. M. Shahin, G. J. Long, L. Rebbouh and F. Grandjean,

J. Am. Chem. Soc., 2005, 127, 2303.

(a) Y. Garcia, V. Niel, M. C. Muňov and J. A. Real, Top. Curr.

K. S. Murray, Dalton Trans., 2011, 40, 10147.

24 T. J. Mooibroek, S. J. Teat, C. Massera, P. Gamez and

J. Reedijk, Cryst. Growth Des., 2006, 6, 1569.

7

8

Chem., 2004, 233, 229, Springer, Berlin; (b) J. A. Real, 25 P. Guionneau, M. Marchivie, G. Bravic, J.-F. Létard and

A. B. Gaspar, V. Niel and M. C. Muňov, Coord. Chem. Rev.,

003, 236, 121.

D. Chasseau, Top. Curr. Chem., 2004, 234, 97.

26 K. S. Murray, in Spin Crossover Materials. Properties

and Applications, ed. M. A. Halcrow, Wiley, London, 2013,

ch. 1.

2

(a) C. Faulmann, K. Jacob, S. Dorbe, S. Lampert, I. Malfant,

M.-L. Doublet, L. Valade and J. A. Real, Inorg. Chem., 2007,

4

6, 8548; (b) K. Takahashi, H.-B. Cui, Y. Okano, 27 C. Carbonera, C. A. Kilner, J. F. Lètard and M. A. Halcrow,

H. Kobayashi, H. Mori, H. Tajima, Y. Einaga and O. Sato,

J. Am. Chem. Soc., 2008, 130, 6688.

(a) G. J. Halder, C. J. Kepert, B. Moubaraki, K. S. Murray

Dalton Trans., 2007, 1284.

28 A. Bousseksou, G. Molnár, L. Salmon and W. Nicolazzi,

Chem. Soc. Rev., 2011, 40, 3313.

9

and J. D. Cashion, Science, 2002, 298, 1762; 29 C. P. Slichter and H. P. Drickamer, J. Chem. Phys., 1972, 56,

b) S. M. Neville, G. J. Halder, K. W. Chapman, 2142.

M. B. Duriska, P. D. Southon, J. D. Cashion, J.-F. Létard, 30 A. V. Sinitskiy, A. L. Tchougréeﬀ and R. Dronskowski, Phys.

B. Moubaraki, K. S. Murray and C. J. Kepert, J. Am. Chem. Chem. Chem. Phys., 2011, 13, 13238.

Soc., 2008, 130, 2869; (c) S. M. Neville, G. J. Halder, 31 (a) M. Kepenekian, B. Le Guennic and V. Robert, Phys. Rev.

(

K. W. Chapman, M. B. Duriska, B. Moubaraki, K. S. Murray

and C. J. Kepert, J. Am. Chem. Soc., 2009, 131, 12106.

0 (a) Y. Sunatsuki, Y. Ikuta, N. Matsumoto, H. Ohta,

M. Kojima, S. Iijima, S. Hayami, Y. Maeda, S. Kaizaki,

F. Dahan and J.-P. Tuchagues, Angew. Chem., Int. Ed., 2003,

B: Condens. Matter, 2009, 79, 094428; (b) M. Kepenekian,

B. Le Guennic and V. Robert, J. Am. Chem. Soc., 2009, 131,

11498; (c) M. Kepenekian, J. S. Costa, B. Le Guennic,

P. Maldivi, S. Bonnet, J. Reedijk, P. Gamez and V. Robert,

Inorg. Chem., 2010, 49, 11057.

1

4

2, 1614; (b) M. Clemente-León, E. Coronado, M. López- 32 J. A. Real, H. Bolvin, A. Bousseksou, A. Dworkin, O. Kahn,

Jordà and J. C. Waerenborgh, Inorg. Chem., 2011, 50, 9122.

F. Varret and J. Zarembowitch, J. Am. Chem. Soc., 1992, 114,

1

1 R. Carballo, B. Covelo, E. M. Vázquez-López, E. Garciá-Mar-

4650.

tinez, A. Castineiras and C. Janiak, Z. Anorg. Allg. Chem., 33 (a) K. Nakano, N. Suemura, S. Kawata, A. Fuyuhiro, T. Yagi,

2

005, 631, 2006.

S. Nasu, S. Morimoto and S. Kaizaki, Dalton Trans., 2004,

982; (b) K. Nakano, N. Suemura, K. Yoneda, S. Kawata and

S. Kaizaki, Dalton Trans., 2005, 740.

1

2 J. Y. Lu, T. J. Schroeder, A. M. Babb and M. Olmstead, Poly-

hedron, 2001, 20, 2445.

3 P. Cortés, A. M. Atria, M. T. Garland and R. Baggio, Acta 34 L. Capes, J. F. Lètard and O. Kahn, Chem.–Eur. J., 2000, 6,

Crystallogr., Sect. C: Cryst. Struct. Commun., 2006, 62, m297. 2246.

4 M. Quesada, P. de Hogg, P. Gamez, O. Roubeau, G. Aromi, 35 (a) S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.

B. Donnadieu, C. Massera, M. Lutz, A. L. Spek and

J. Reedijk, Eur. J. Inorg. Chem., 2006, 1353.

Phys., 2010, 132, 154104; (b) S. Grimme, J. Comput. Chem.,

2006, 27, 1787.

1

5 J. J. M. Amoore, C. J. Kepert, J. D. Cashion, B. Moubaraki, 36 (a) A. Schäfer, H. Horn and R. Ahlrichs, J. Chem. Phys.,

S. M. Neville and K. S. Murray, Chem.–Eur. J., 2006, 12,

220.

1992, 97, 2571; (b) A. Schäfer, C. Huber and R. Ahlrichs,

J. Chem. Phys., 1994, 100, 5829.

8

1

6508 | Dalton Trans., 2013, 42, 16494–16509

View Article Online

Dalton Transactions

Paper

3

7 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, 40 Y. B. Go, X. Wang, E. V. Anokhina and A. J. Jacobson, Inorg.

M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, Chem., 2005, 44, 8265.

B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, 41 T. Hermon and E. Y. Tshuva, J. Org. Chem., 2008, 73, 5953.

X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, 42 G. M. Sheldrick, SADABS, Program for area detector adsorp-

J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,

J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,

tion correction, Institute for Inorganic Chemistry, University

of Göttingen, Germany, 1996.

H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, 43 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,

F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, 307.

K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, 44 R. H. Blessing, J. Appl. Crystallogr., 1997, 30, 421.

K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, 45 T. M. McPhillips, S. E. McPhillips, H. J. Chiu, A. E. Cohen,

J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,

J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,

R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,

R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,

K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,

J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,

A. M. Deacon, P. J. Ellis, E. Garman, A. Gonzalez,

N. K. Sauter, R. P. Phizackerley, S. M. Soltis and P. Kuhn,

Blu-Ice and the Distributed Control System: software for

data acquisition and instrument control at macromolecular

crystallography beamlines, J. Synchrotron Radiat., 2002, 9,

401.

J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, 46 W. Kabsch, J. Appl. Crystallogr., 1993, 26, 795.

GAUSSIAN 09 (Rev. A.02), Gaussian, Inc., Wallingford, CT, 47 (a) G. M. Sheldrick, SHELXL-97, Program for refinement of

2

009.

crystal structures, University of Göttingen, Germany, 1997;

(b) A. L. Spek, Acta Crystallogr., Sect. A: Fundam. Crystallogr.,

1990, 46, C34.

3

8 J.-F. Létard, J. Mater. Chem., 2006, 16, 2550.

9 M. Marchivie, P. Guionneau, J.-F. Létard and D. Chasseau,

Acta Crystallogr., Sect. B: Struct. Sci., 2005, 61, 25.

48 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.

Dalton Trans., 2013, 42, 16494–16509 | 16509

Article Doi

Crown-linked dipyridylamino-triazine ligands and their spin-crossover iron(ii) derivatives: Magnetism, photomagnetism and cooperativity

DOI: 10.1039/c3dt51839f

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c3dt51839f

Full text of DOI:10.1039/c3dt51839f

Products guided by the article

R&D Labs maybe for 127795-40-8

Relevant to this article

Hot Product