Structural Diversity in a New Series of Halogenated Quinolyl Salicylaldimides-Based FeIII Complexes Showing Solid-State Halogen-Bonding/Halogen···Halogen Interactions

Article abstract of DOI:10.1021/acs.cgd.8b00753

A new series of tridentate N-8-quinolyl-salicylaldimine ligands, Hqsal-5,7-X₂ [X = Cl(1), Br(2), I(3)], halo-substituted at the 5,7 position of the aminoquinoline moiety and their corresponding complexes with Fe^III were synthesized and formulated as [Fe(qsal-5,7-X₂)(NCS)(MeO)]₂·solv. (X = Cl (1a), Br (2a), I (3a, solv = ¹/₂MeOH), [Fe₄(qsal-5,7-X₂)₄(NCS)₂(MeO)₂]·solv. (X = Br (2b), I (3b; solv = 4CH₂Cl₂)) by single-crystal X-ray diffraction analysis. 1a and 2a are isostructural dimers where each Fe^III metal ion, showing a distorted octahedral environment, is bound by a N,N,O tridentate (qsal-5,7-X₂)^- (X = Cl and Br) ligand, a N-coordinated SCN^- anion, and a bridging methanolate anion. 2b and 3b are centrosymmetric tetramers where each Fe^III is bound by three nitrogen atoms and three oxygen atoms derived from a tridentate (qsal-5,7-X₂)^- (X = Br and I), a SCN^-, a bridging methanolate anion, and a bridging μ²-oxy moiety. In 3b, the iodine atoms dominate the packing interactions through the establishment of a halogen-bonding network. The magnetic behavior of 1a-3a dimers and 2b-3b tetramers indicate the presence of strong antiferromagnetic interactions between Fe^III centers (S = 5/2), mediated by the alkoxy bridges. Experimental data can be modeled with an isotropic Hamiltonian, H = -2J(S₁ · S₂) for dimers (J = -15 cm^-1) and H = -2J(S₂ · S₃) - 2J′(S₁ · S₂ + S₃ · S₄), for tetramers (J = -24 cm^-1, J′ = -11 cm^-1). The magnetic behavior of 1a-3a dimers indicates the presence of strong antiferromagnetic interactions between Fe^III centers (S = 5/2), mediated by the alkoxy bridges. 2b and 3b show the same magnetic behavior since they contains analogous bridges between paramagnetic centers, but for a linear tetramer. Density functional theory (DFT) calculations, based on hybrid functional mPW1PW paralleled by the Def2SVP all-electron split-valence basis sets, support the experimental results, showing that monomers could possibly show a spin crossover (SCO) behavior, even though the formation of complexes with an even number of metal ions results in the strong antiferromagnetic interactions. Accordingly, broken symmetry DFT calculations carried out on 1a clearly show that the antiferromagnetic coupling of the two HS Fe^III centers results in the lowest energy electron configuration of the complex.

Full text of DOI:10.1021/acs.cgd.8b00753

Subscriber access provided by University of Massachusetts Amherst Libraries
Article
Structural Diversity on a New Series of Halogenated Quinolyl
Salicylaldimides-Based FeIII Complexes Showing Solid-
State Halogen-Bonding/Halogen···Halogen Interactions
Suchithra Ashoka Sahadevan, Enzo Cadoni, Noemi Monni, Cristina Saenz de Pipaon, Jose Ramon Galan-
Mascaros, Alexandre Abhervé, Narcis Avarvari, Luciano Marchio', Massimiliano Arca, and Maria Laura Mercuri
Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.8b00753 • Publication Date (Web): 31 May 2018
Downloaded from http://pubs.acs.org on May 31, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Structural Diversity on a New Series of Halogenated
Quinolyl SalicylaldimidesꢀBased FeIII Complexes
Showing SolidꢀState Halogenꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bonding/HalogenꢁꢁꢁHalogen Interactions
Suchithra Ashoka Sahadevan,†,‡ Enzo Cadoni,† Noemi Monni,† Cristina Sáenz de Pipaón,∏ Joséꢀ
Ramon Galan Mascaros,∏,ǁ Alexandre Abhervé,‡ Narcis Avarvari,‡ Luciano Marchiò,#
Massimiliano Arca,† Maria Laura Mercuri*†
†
Dipartimento di Scienze Chimiche e Geologiche, Università degli Studi di Cagliari, S.S.
554 − Bivio per Sestu − I09042 Monserrato (Cagliari), Italy
‡
Laboratoire MOLTECHꢀAnjou UMR 6200, UFR Sciences, CNRS, Université d’Angers,
Bât. K, 2
Bd. Lavoisier, 49045 Angers, France
∏
Institute of Chemical Research of Catalonia, Barcelona Institute of Science and
Technology, Avinguda Paı¨sos Catalans 16, Tarragona 43007, Spain
ǁ
Institució Catalana de Recerca i Estudis Avançats , Passeig Lluis Companys 23,
Barcelona 08010, Spain.
ACS Paragon Plus Environment
1

Crystal Growth & Design
Page 2 of 46
1
2
3
4
5
6
#
Dipartimento di Chimica, Università di Parma, Parco Area delle Scienze 17A, I43124
Parma, Italy.
7
8
9
KEYWORDS (Word Style “BG_Keywords”). Halogen bonding, quinolyl salicylaldimides
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hqsal, Schiff bases, antiferromagnetic dimers, tetramer, DFT, Iron
ABSTRACT
A new series of tridentate Nꢀ8ꢀquinolylꢀsalicylaldimine ligands, Hqsalꢀ5,7ꢀX2 [X = Cl(1), Br(2),
I(3)], haloꢀsubstituted at the 5,7 position of the aminoquinoline moiety and their corresponding
complexes with FeIII were synthesized and formulated as [Fe(qsalꢀ5,7ꢀX2)(NCS)(MeO)]2ꢁsolv. (X
= Cl (1a), Br (2a), I (3a, solv = ½ MeOH), [Fe4(qsalꢀ5,7ꢀX2)4(NCS)2(MeO)2]ꢁsolv. (X = Br (2b),
I (3b; solv = 4CH2Cl2) by singleꢀcrystal Xꢀray diffraction analysis. 1a and 2a are isostructural
dimers where each FeIII metal ion, showing a distorted octahedral environment, is bound by a
N,N,O tridentate (qsalꢀ5,7ꢀX2)– (X = Cl and Br) ligand, a Nꢀcoordinated SCN– anion, and a
bridging methanolate anion. 2b and 3b are centrosymmetric tetramers where each FeIII is bound
by three nitrogen atoms and three oxygen atoms deriving from a tridentate (qsalꢀ5,7ꢀX2)– (X = Br
and I), a SCN–, a bridging methanolate anions and a bridging ꢀ2ꢀoxo moiety. In 3b, the iodine
atoms dominate the packing interactions through the establishment of a halogenꢀbonding
network. . The magnetic behavior of 1a–3a dimers and 2b-3b tetramers indicate the presence of
strong antiferromagnetic interactions between FeIII centers (S = 5/2), mediated by the alkoxy
bridges. Experimental data can be modeled with an isotropic Hamiltonian, ꢀ = −2ꢁ(ꢂꢃ ∙ ꢂꢄ)
–1
ꢆ
(
)
(
)
for dimers (J = –15 cm ) and ꢀ = −2ꢁ ꢂꢄ ∙ ꢂꢅ − 2ꢁ ꢂꢃ ∙ ꢂꢄ + ꢂꢅ ∙ ꢂꢇ , for tetramers (J = –24
cm–1, J' = –11 cm–1). The magnetic behavior of 1a–3a dimers, indicate the presence of strong
antiferromagnetic interactions between FeIII centers (S = 5/2), mediated by the alkoxy bridges. 2b
ACS Paragon Plus Environment
2

Page 3 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
and 3b show the same magnetic behavior since they contains analogous bridges between
paramagnetic centers, but for a linear tetramer. DFT Calculations, based on hybrid functional
mPW1PW paralleled by the Def2SVP allꢀelectron splitꢀvalence basis sets, support the
experimental results, showing that monomers could possibly show a SCO behavior, even though
the formation of complexes with an even number of metal ions results in the strong anti
ferromagnetic interactions. Accordingly, Broken Symmetry DFT calculations carried out on 1a
clearly show that the antiferromagnetic coupling of the two HS FeIII centers results in the lowest
energy electron configuration of the complex.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
One of the most powerful strategies to design halogenꢀbonded systems is based on the use of
tectons, in particular metallotectons, namely metal complexes bearing peripheral halogen atoms,
able to be involved in well identified nonꢀcovalent intermolecular interactions, such as halogen
bonds. Over the last decade halogen bonding (XB), a strong, specific and highly directional nonꢀ
covalent interaction, has gained great interest among the known nonꢀcovalent bonds, for the
construction of one dimensional (1D), 2D, 3D selfꢀassembled supramolecular architectures1–4.
This interaction plays a role as crucial as hydrogen bonding (HB)5 in driving highly specific
crystal packing patterns and several applications of XB are reported[1] in various research areas
such as materials science and synthetic chemistry. Halogenꢁꢁꢁhalogen (XX) interactions are also a
peculiar class of halogens bonds interactions, that although weak, play a key role in the
formation of supramolecular networks ranging from 1ꢀ3D materials to selfꢀassembled
monolayers, and can be considered as a reliable supramolecular tool in molecular selfꢀ
assembly.6–9 In this context the combination of haloꢀsubstituted anilates, the 3,6ꢀdihalo
ACS Paragon Plus Environment
3

Crystal Growth & Design
Page 4 of 46
1
2
3
4
5
6
7
8
9
derivatives of 2,5ꢀdihydroxybenzoquinone (H4C6O4), X2An2ꢀ (X = Cl, Br, I), with paramagnetic
metals has been shown to be particularly successful.10–15 In the search of novel haloꢀsubstituted
tectons where the halogen atoms can play a role on the crystalline packing pattern and/or
physical properties, we focused our attention on the tridentate Nꢀ8ꢀquinolylꢀ5ꢀXꢀ
salicylaldiminate (X = H, Hqsal) ligand and its haloꢀsubstituted derivatives on the 5ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
salicylaldiminate moiety (qsalꢀX) (X = F, Cl, Br, I), because of the interesting spin crossover
16–21
(SCO) properties shown by their FeIII/FeII complexes.
Temperatureꢀinduced spin transition
22–25
in fact was observed in [Fe(qsal)2]NCS and [Fe(qsal)2]NCSe
with wide thermal hysteresis
loop of 140 K and 180 K respectively, while switching spin state upon exposure to light
(LIESST effect) was also found.24 Halogen effects on the SCO behavior was reported in
[FeIII(qsalꢀX)2]NCS·solvent (X = F, Cl, Br; solvent= 1/4CH2Cl2·1/2MeOH) complexes, as the
spin transition temperature was found to increase on going from F to Br.17,26–31 Very recently, an
abrupt twoꢀstep SCO above room temperature32,33 was observed in the [FeII(qsalꢀCl)2] complex
due to the halogen substitution on the salicylaldiminate moiety of the ligand that has the potential
to improve the SCO properties within FeIIꢀSCO systems. Therefore, the introduction of halogen
atoms at certain key positions of the metal complexes influences significantly their physical
properties, although keeping their structure unaltered. In this work we investigate the influence
of the halogen substitution on the quinolyl moiety of the Hqsal ligand, which has never been
studied so far, with the aim to highlight how a simple change in the positions of the halogen
atoms in the same ligand can dramatically change the crystal structure and physical properties of
the related metal complexes. We report herein the synthesis and structural characterization of
the novel class of haloꢀsubstituted Hqsalꢀ5,7ꢀX2 ligands (X = Cl, 1, Br, 2, I, 3), shown in Chart 1,
and their FeIII complexes: [Fe(qsalꢀCl2)(NCS)(MeO)]2 (1a), [Fe(qsalꢀ5,7ꢀBr2)(NCS)(MeO)]2
ACS Paragon Plus Environment
4

Page 5 of 46
Crystal Growth & Design
1
2
3
4
(2a), [Fe4(qsalꢀ5,7ꢀBr)4(NCS)2(MeO)2] (2b), [Fe(qsalꢀ5,7ꢀI2)(NCS)(MeO)]2, (3a,·½MeOH) and
5
6
[
Fe4(qsalꢀ5,7ꢀI2)4(NCS)2(MeO)2] (3b,·4CH2Cl2). The magnetic properties of the 1a–3a, 2b–3b
7
8
9
complexes are also reported.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 1: Structure of Hqsalꢀ5,7ꢀX2 ligand. X = Cl (1), Br (2), I (3).
RESULTS AND DISCUSSIONS
Synthesis of Hqsal-5,7-X2 ligands and their FeIII complexes
An overview of the synthetic procedure of the Hqsalꢀ5,7ꢀX2 ligands is shown in Scheme 1.
Scheme 1. Overview of the synthetic procedure of Hqsalꢀ5,7ꢀX2 ligands.
ACS Paragon Plus Environment
5

Crystal Growth & Design
Page 6 of 46
1
2
3
4
5
6
7
8
9
The haloꢀsubstituted Hqsalꢀ5,7ꢀX2 ligands were prepared in good yields (48% to 89%) by
treating 5,7ꢀdihaloꢀ8ꢀaminoquinoline in EtOH with salicylaldehyde and glacial acetic acid. The
synthesis of the Hqsalꢀ5,7ꢀX2 ligands was performed according to the synthetic pathway shown
in Scheme S1 in Supporting Information (SI). The first stage of the process is a simple formation
of an imine, starting from amine and aldehyde, at room temperature and with good yields despite
of the substituents used. The reaction is carried out in ethanol and in the presence of acetic acid.
Acetic acid was used both for increasing the electrophilicity of the carbonyl aldehyde group,
given the low nucleophilicity of the aromatic amine, and for favoring the water elimination in the
second stage. The reagents and the products are poorly soluble in ethyl alcohol, therefore
reaction times of almost 24 hours are required for the complete disappearance of the amines as
reported in Scheme S2 in SI. More details on the synthesis 34–37 are given in SI.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The FeIII complexes, 1a–3a, 2b–3b, were synthesized according to the general synthetic
strategy recently reported for analogous complexes20 (Scheme 2).
ACS Paragon Plus Environment
6

Page 7 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. General synthetic scheme for 1aꢀ3a, 2bꢀ3b complexes; X = Cl (1), Br (2), I (3);
a=dimer; b=tetramer.
These complexes were obtained by carefully layering the ligands 1–3 with Et3N in CH2Cl2 and
on top FeCl3ꢁ6H2O and NaSCN in 1:1 ratio, dissolved in MeOH. ESIꢀMS spectra recorded on
freshly prepared CHCl3/MeOH solutions, in the 100ꢀ2000 m/z mass range, showed only the
molecular ions, which can be assigned to [Fe(qsalꢀ5,7ꢀX2)2]+ monomeric species [X = Cl (1c+),
Br (2c+), and I (3c+)]. Mass spectra recorded on solutions with concentration values at the upper
solubility limit (10 mg/mL) and lower values of needle voltage of 4500 V did not show dimeric
forms (vide infra). Therefore, to improve peak resolution, spectra were recorded in 100ꢀ1000 m/z
mass range. The isotopic patterns of the ligands 1–3 and of the monomeric species 1c+ , 2c+ and
3c+ are reported in SI (Figure S1ꢀS2).
ACS Paragon Plus Environment
7

Crystal Growth & Design
Page 8 of 46
1
2
3
4
Structural Studies
5
6
7
8
9
Among the ligands, only single crystals of 3 could be obtained by recrystallization of the crude
product in dichloromethane, confirming the expected composition (Figure S3).
Single crystals of the iron complexes were obtained by slow diffusion of ligand in
dichloromethane and FeIII salt with NaSCN in methanol. The Xꢀray diffraction analysis
unexpectedly established the complexes to be dimers or tetramers according to Scheme 3 and
Table 1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1: Summary of the obtained FeIII complexes.
FeCl3ꢁ6H2O + NaSCN
X = Cl
Br
2a
2b
I
[Fe(qsalꢀ5,7ꢀX2)(NCS)(MeO)]2 dimer
1a
–
3aa
3bb
[Fe4(qsalꢀ5,7ꢀX2)4(NCS)2(MeO)2] tetramer
a Structurally characterized in 3a·½MeOH. b Structurally characterized in 3b·4CH2Cl2.
The presence of halogen atoms on the ligands periphery expands the types of interactions that
can be exchanged by the iron complexes.38,39 Here, we will analyze the influence of the halogen
substitution (Cl, Br, I) on the second coordination sphere and crystal packing.
The complexes 1a and 2a crystallize in the triclinic space group Pꢀ1; 2b crystallizes in
monoclinic space group P21/c. The molecular structure of 1a is reported in Figure S4. The
molecular structure of 2a, 3a, 2b, 3b are reported in Figures 1ꢀ3. A summary of Xꢀray
crystallographic data for 3, 1a-3a, 2b-3b are reported in Table S1, and selected bond distances
(Å) and angles (°) for 1a-3a, 2b-3b are reported in Table S2ꢀS7.
ACS Paragon Plus Environment
8

Page 9 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2a
3a
Figure 1. Molecular structure of 2a (symmetry code ’ = ꢀx; 1ꢀy; 1ꢀz). and 3a (symmetry code ’ =
2ꢀx; 1ꢀy; ꢀz, ’’ = 1ꢀx; ꢀy; 1ꢀz). Thermal ellipsoids are drawn at the 30% probability level. Color
code: C = grey, Cl = green, Br = brown, O = red, S = yellow, Fe = orange, N = blue, H = white.
Compounds 1a and 2a are isostructural and their molecular features can therefore be described
together. Both complexes lie on an inversion center and half dimer represents the asymmetric
unit. The FeIII metal is in a distorted octahedral environment and it is bound by a N,N,O
tridentate ligand, a SCNꢀ anion, and a bridging methanolate anion. For the tridentate ligand, the
FeꢀN bond distances are significantly longer (0.13ꢀ0.25 Å) than the remaining bond distances,
and the shortest distance is that between the metal and the phenolate moiety, Table S2 in SI. The
SCNꢀ anion is bound to the metal center through the nitrogen atom, and consistently, it exhibits a
linear coordination geometry. The asymmetric unit of compound 3a comprises two half dimeric
units and one methanol molecule of crystallization (Figure 1). The overall structural arrangement
is nevertheless similar to that of 1a and 2a, since the presence of the iodine on the periphery of
the aromatic rings has a negligible influence on the iron coordination environment. The methyl
alcohol molecule forms a hydrogen bond with the oxygen atom of the phenolate moiety, O(11).
ACS Paragon Plus Environment
9

Crystal Growth & Design
Page 10 of 46
1
2
3
4
5
6
7
8
9
In 1aꢀ3a, the metalꢀmetal distance is approximately 3.15 Å. The crystal packing of 1a and 2a is
reported in Figure S5. Adjacent molecules exchange different types of interactions, namely,
Cl···Cl (3.39 Å) and Br···Br (3.51 Å) with symmetry related halogen atoms, CH···S with the
aromatic ring and the SCNꢀ anion and πꢀπ stacks with adjacent bromoꢀquinoline rings. The
crystal packing of 3a is more complex due to the presence of two different molecular entities and
the solvent of crystallization. In this system, the presence of the iodine atom with respect to the
bromine and chlorine in 1a and 2a, expands the possible types of interactions that can be formed
between adjacent molecules. In fact, I(22) and I(21) both interacts with S(1T) of the thiocyanate
group. The geometry of the I(22)···S(1T) interaction (3.49 Å) is consistent with the presence of a
halogen bond, whereas the geometry of the I(21)···S(1T) interaction (3.56 Å) presumably occurs
through the negatively charged corona on iodine towards the terminal sulphur consistent with the
presence of a chalcogenide bond.1,4,40
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
10

Page 11 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Portion of the crystal packing of 2a (a and b) and 3a (c), with the methanol
molecule of crystallization highlighted. Dashed lines represent short interactions between
symmetry related molecules. Selected weak interaction distances are indicated in Å.
2b and 3b are centrosymmetric tetramers where the two outermost metal have a different
coordination environment than the two innermost metals (Figure 3).
ACS Paragon Plus Environment
11

Crystal Growth & Design
Page 12 of 46
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2b
3b
Figure 3. Molecular structure of 2b (left) and 3b (right), with thermal ellipsoids drawn at the
30% probability level. Solvent molecules and hydrogen atoms were removed for clarity.
Symmetry code ’ = 1ꢀx; 1ꢀy; 1ꢀz. Color code C = grey, Br = brown, I= violet, O = red, S=
yellow, Fe = orange, N = blue, H = white.
In 3b, four dichloromethane molecules of crystallization per tetrameric complex are present. In
both systems Fe(1) is bound by three nitrogen atoms and three oxygen atoms deriving from a
N,N,O tridentate ligand, a SCNꢀ anion a bridging methanolate and a bridging oxo moiety.
Differently, Fe(2) is bound by two nitrogen atoms and four oxygen atoms deriving from a N,N,O
tridentate ligand, two bridging oxo moieties and a bridging methanolate anion. In both
compounds, the Fe(1)ꢀFe(2) and Fe(2)ꢀFe(2)’ bond distances are 3.07 and 2.96 Å, respectively.
The crystal packing of 2b is reported in Figure 4. Ideally, the molecules form layers by means of
πꢀπ stacks, CH···S and CH···C/N interactions between the aromatic ring and the SCNꢀ anion.
ACS Paragon Plus Environment
12

Page 13 of 46
Crystal Growth & Design
1
2
3
4
5
6
These layers interacts by forming halogen bonds (Br(3)···Br(4), 3.62 Å) and CH···Br (3.77 Å)
interactions between symmetry related molecules.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Portion of the crystal packing of 2b. Dashed lines represent short interactions between
symmetry related molecules. Selected weak interaction distances are indicated in Å.
In 3b, the iodine atoms dominate the packing interactions, as shown in Figure 5. According to
the geometries of the interactions, the negative corona of I(1) interacts with the σꢀhole of I(4) of
a symmetry related molecule. Likewise, the lone pair distribution of S(1) interacts with the σꢀ
hole of I(3).1,4,40 Hence, I(3) acts as XB donor with respect to the carbon atom of an adjacent
SCNꢀ anion. Hence I(2), I(3) and I(4) acts as XB donor whereas acts as XB acceptor.
Dichloromethane molecules of crystallization are also involved in weak interactions among
adjacent complex molecules.
ACS Paragon Plus Environment
13

Crystal Growth & Design
Page 14 of 46
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Highlights on the interaction exchanged by the iodine atoms in 3b. Dashed lines
represent short interactions between symmetry related molecules, distances are indicated in Å.
Different views of the crystal packing of 3b are reported in Figure S6. In all ligand systems,
the aromatic moieties linked by the imino group are not coplanar since they form a dihedral
angle in the 31ꢀ41° range. In 2b and 3b, the tridentate ligands positioned on the same side with
ACS Paragon Plus Environment
14

Page 15 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
respect to the Fe4O4 core, form πꢀπ stacks with the quinoline and phenolate residues. The shortest
contact is between the aromatic rings of the phenolate groups (3.4ꢀ3.5 Å range), whereas the
distances between the quinoline groups is approximately 3.6 Å for 2b and greater than 3.7 Å for
3b.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Magnetic Measurements
In these series, compounds type a (dimers) and b (tetramers) are magnetically identical, if we
consider the intermolecular magnetic interactions as negligible. Indeed, their magnetic behavior
is indistinguishable within experimental error. Therefore, we only describe the magnetic
properties of 1a and 2b, as representatives of these two magnetic structures.
The χT product for 1a at room temperature is 5.01 emu K mol–1, lower than the expected value
for the spinꢀonly contribution of two high spin (HS) FeIII S = 5/2 centers (χTHS ≈ 8.750 emu K
mol–1), but significantly higher than the spinꢀonly contribution of two low spin (LS) FeIII S = 1/2
centers (χTLS. ≈ 0.750 emu K mol–1). This confirms the HS state of the FeIII centers and suggests
the presence of strong antiferromagnetic interactions between them, mediated by the alkoxo
bridges. In the 2–300 K range, χT decreases with temperature (Figure 6, left), reaching values
close to zero at very low temperatures. The experimental data can be fitted to the Hamiltonian:
ꢀ = −2ꢁ(ꢂꢃ ∙ ꢂꢄ)
where S1 = S2 = 5/2. The best fit (Figure 6, left) was found for g = 2.09, J = –21 K (–14.6
cm–1), with the addition of a small paramagnetic impurity, equivalent to ≈2% of the total spin
carriers.
The magnetic behavior for 2b is similar, since it contains analogous bridges between
paramagnetic centers, but now in a linear tetramer. The room temperature χT product is 8.50 emu
K mol–1. This value can only be assigned to four HS FeIII centers, antiferromagnetically coupled
ACS Paragon Plus Environment
15

Crystal Growth & Design
Page 16 of 46
1
2
3
4
5
(χTHS ≈ 17.50 emu K mol–1, χTLS ≈ 1.50 emu K mol–1). In this case, the appropriate
phenomenological Hamiltonian is:
6
7
8
9
ꢀ = −2ꢁ(ꢂꢄ ∙ ꢂꢅ) − 2ꢁ′(ꢂꢃ ∙ ꢂꢄ + ꢂꢅ ∙ ꢂꢇ)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
where S1 = S2 = S3 = S4 = 5/2. The temperature dependence of the χT product was satisfactorily
reproduced with parameters: g = 2.03, J = –35 K (24.3 cm–1), J' = –16 K (11.1 cm–1), and a
paramagnetic impurity equivalent to ≈ 2.5 % of the total spin carriers. (Figure 6, right). It is
particularly significant the fact that the shape of the curve can be reproduced only if J >J'. We
found a ratio of 2 between both exchange parameters as optimum to reproduce the data. This can
be rationalize if we look into the Fe–O–Fe angles. Whereas the Fe1–O–Fe2 bridging angles are
asymetric (≈ 98º and 105º), the two Fe2–O–Fe2 angles are almost identical, forming a perfect
square. This permits a better overlap between the magnetic orbitals, and a stronger
antiferromagnetic superexchange in the central bridge.
Figure 6. Thermal variation of χMT versus T plots for 1a (left) and 2b (right). In red, the fitting
obtained with the following parameters: g = 2.09, J = –21 K (–14.6 cm–1) (1a) and g = 2.03, J =
–35 K (24.3 cm–1), J' = –16 K (11.1 cm–1), (2b).
ACS Paragon Plus Environment
16

Page 17 of 46
Crystal Growth & Design
1
2
3
4
DFT Calculations
5
6
7
8
9
In order to investigate the electronic structure and reactivity of the ligands 1–3 and the relevant
iron(III) complexes, QuantumꢀMechanical (QM) calculations were carried out at the Density
Functional Theory level.41 Previous theoretical investigations showed that the correct modeling
of high/low spin equilibria at DFT level depends dramatically on the choice of the functional,42
which can unevenly favor one of the two spin states. In particular, it was observed that the pure
functionals PW9143,44 and BLYP45 favor the LS state, while the paradigmatic hybrid functional
B3LYP,46–48 featuring a 20% HartreeꢀFock (HF) exchange, favors the HS state.[37] Kepp
comparatively examined twelve different functionals and demonstrated that the B3LYP* hybrid
functional (HF exchange 15%)49,50 provide an accurate evaluation of the HSꢀLS energy gap for
30 different SCO iron complexes.51 This notwithstanding, in the peculiar case of qsalꢀX– (qsalꢀ
X– = monosubstituted quinolylsalicylaldiminate; X = H, OMe, F, Cl, Br, I) FeIII trisꢀcomplexes,
the B3LYP* functional was recently proved to overestimate the HS–LS enthalpy difference
ꢁHSCO.52 In this work, based on the very good results achieved in the past on a variety of both
organic53–55 and inorganic systems,56–58 we turned to test the hybrid functional mPW1PW by
Adamo and Barone,59 paralleled by the Def2SVP allꢀelectron splitꢀvalence BS, including
polarization functions.60
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Whatever the combination of BSs and functional, HS and LS configurations show different
structural features, in particular as far as the metal ion is involved, so that a comparison between
theoretically optimized and structural metric parameters allows for the attribution of the correct
spin state. A validation of the computational setup chosen for this work was therefore carried out
by comparing the structural bond distances and angles determined for the complex [Fe(qsalꢀF)2]+
(4+), isolated in 4(NCS)17, featuring a LS at 100 K and a HS FeIII center at 270 K, with the
ACS Paragon Plus Environment
17

Crystal Growth & Design
Page 18 of 46
1
2
3
4
corresponding optimized metric parameters. An examination of the bond distances involving the
5
6
central metal ion shows that on passing from LS to HS the Fe–N distances are sensibly modified
7
8
[14]
[average values 1.960(2) and 1.983(2) Å at 100 and 270 K,
respectively], while the Fe–O
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bond lengths remain almost unaltered [average values 1.875(2) and 1.866(2) Å at 100 and 270
K,[14] respectively]. DFT optimized distances are in very good agreement with the structural ones
and display the trend found experimentally on passing from LS to HS configurations [LS: Fe–N,
1.973; Fe–O 1.857; HS: Fe–N, 2.173; Fe–O, 1.898 Å]. Notably, the uncorrected total electronic
energies of the LS and HS configurations calculated for 4+ in the gas phase differ by only 2.80
kcal mol–1, while the enthalpy contribution ꢁHSCO to the spinꢀcrossover (SCO) amounts to 3.84
kcal mol–1, in agreement with the temperatureꢀcontrolled SCO behavior observed
experimentally.[14] The structures of the neutral protonated compounds Hqsalꢀ5,7ꢀX2 (1–3) and
the corresponding Oꢀdeprotonated (qsalꢀ5,7ꢀX2)– anions were optimized at the same level of
theory. A comparison of the metric parameters calculated for 3 with the corresponding ones
determined by single crystal Xꢀray diffraction (see above) clearly shows an excellent agreement.
In particular, it is worth noting that the C–I distances are optimized at values extremely close to
the experimental ones (optimized C–I distance = 2.109 Å). Also the formation of the
intramolecular O12–H12···N2 bond is reproduced correctly with an optimized O···N distance
only slightly overestimated [calculated 2.598 Å; structural value 2.652(4) Å]. In agreement with
the crystal structure analysis, the aminoquinoline moiety is twisted with respect to the plane of
–
phenyl ring (τ = 44.63°). On passing from 3 to the corresponding deprotonated form qsalꢀ5,7ꢀI2 ,
the metric parameters remain almost unvaried, with a small decrease in the τ dihedral angle [τ =
39.98° for (qsalꢀ5,7ꢀI2)–] and the loss of planarity of the 2ꢀhydroxyphenylmethanimine moiety. In
the deprotonated form, KohnꢀSham (KS) frontier molecular orbitals can be envisaged with large
ACS Paragon Plus Environment
18

Page 19 of 46
Crystal Growth & Design
1
2
3
4
5
6
contributions from the lone pairs of electrons located on the two nitrogen and the oxygen atoms
(KSꢀHOMOꢀ6, HOMOꢀ3, and HOMOꢀ1 for (qsalꢀ5,7ꢀI2)–; Figure 7].
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MO 83
MO 86
MO 88
Figure 7. Isosurface drawings of KSꢀHOMOꢀ6 (MO 83), KSꢀHOMOꢀ3 (MO 86), and KSꢀ
HOMOꢀ1 (MO 88) with large contributions from the LPs on the N and O donor atoms in (qsalꢀ
5,7ꢀI2)–. Cutoff value = 0.05 |e|.
These atoms feature remarkable negative natural61,62 charges Q (QN1 = ꢀ0.463, QN2 = ꢀ0.504,
and QO12 = ꢀ0.645 |e|) and are therefore all available to interact with the iron(III) ions. On passing
from qsalꢀ5,7ꢀI2)– to the corresponding brominated and chlorinated anions, the natural charges
are basically unvaried [(qsalꢀ5,7ꢀCl2)–: QN1 = ꢀ0.465, QN2 = ꢀ0.499, and QO12 = ꢀ0.650 |e|; (qsalꢀ
5,7ꢀBr2)–: QN1 = ꢀ0.464, QN2 = ꢀ0.500, and QO12 = ꢀ0.647 |e|), thus indicating similar donor
abilities of the three donor atoms. On varying the nature of the halogen substituents, KSꢀHOMOꢀ
1, KSꢀHOMOꢀ3, and KSꢀHOMOꢀ6 keep the same composition in terms of their constituent
atomic orbitals, their eigenvalues becoming progressively slightly more stable on passing from
ACS Paragon Plus Environment
19

Crystal Growth & Design
Page 20 of 46
1
2
3
4
5
6
7
8
9
(qsalꢀ5,7ꢀI2)– to (qsalꢀ5,7ꢀCl2)– [eigenvalues ε: ΚΣ−HOMOꢀ1: ꢀ2.082, ꢀ2.156, ꢀ2.230; HOMOꢀ3:
ꢀ3.504, ꢀ3.570, ꢀ3.605; HOMOꢀ6: ꢀ4.760, ꢀ4.734, and ꢀ4.601 eV for (qsalꢀ5,7ꢀCl2)–, (qsalꢀ5,7ꢀ
Br2)–, and (qsalꢀ5,7ꢀI2)–, respectively].
DFT calculations have been extended to the monomer cationic complexes [Fe(qsalꢀ5,7ꢀX2)2]+
[X = Cl (1c+), Br (2c+), I (3c+)], identified in solution by mass spectrometry (see above). These
complexes, although not isolated in the solid state, represent interesting model complexes of the
potential spin crossover (SCO) behavior. In order to model the possible electron configurations,
both the HS (S = 5/2, sixtuplet) and LS (S = 1/2, doublet) configurations63 were modeled for the
three FeIII complexes. The optimized structures of complexes 1c+–3c+ show the central ion in a
distorted octahedral coordination achieved by the two N1 and N2 atoms and by the O atoms
(Figure S8 for 1c+ in the HS state). Notably, upon coordination the phenyl and the quinolyl rings
assume tend to a more planar structure as compared to the free ligand in order to maximize the
interaction to the metal ion (for example τ = 15.73° for 1c+ in the HS state).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In Table 2, the FeꢀN and FeꢀO bond lengths optimized for 1c+, 2c+, and 3c+ both in the LS and
HS states are listed, showing that the nature of the halogen affects only marginally the bond
distances, which depend as expected on the spin state.
Table 2: FeꢀN and FeꢀO bond distances optimized for the model
complexes 1c+–3c+ at DFT level in the gas phase in the LS and HS
configurations.a
1c+
2c+
3c+
HS
FeꢀN2
FeꢀN1
2.131
2.192
2.129
2.192
2.126
2.192
ACS Paragon Plus Environment
20

Page 21 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
FeꢀO1
FeꢀN2
FeꢀN1
FeꢀO1
1.906
1.949
1.994
1.856
1.908
1.911
1.944
1.994
1.860
LS
1.947
1.994
1.858
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Labels of the (qsalꢀ5,7ꢀX2)– ligands as in Figure 2 (X = Cl, Br,
and I, for 1c+, 2c+, and 3c+, respectively).
An examination of the crystal structures of compounds 1a, 2a and 3a clearly shows that FeꢀO
distances fall in the range 1.92ꢀ1.94 Å, FeꢀN1 bond lengths are all around 2.18 Å, while FeꢀN2
distances are slightly shorter (2.13 Å). Therefore, the structural parameters clearly point to a HS
configuration of the complexes. Accordingly, the total electronic energies calculated for HS and
LS complexes 1c+–3c+ in the gas phase testify for a larger stability of HS as compared to LS,
with energy differences ꢁESCO of 3.0, 3.2, and 3.5 kcal mol–1 for 1c+, 2c+, and 3c+, respectively,
which are only slightly larger than those calculated for 4(NCS)+ at the same level of theory. A
thermochemical analysis shows that by taking into account the Zero Point Energy (ZPE)
corrections and the thermal corrections to enthalpy, the same trend is confirmed (ꢁHSCO = 4.07,
4.28, and 4.52 kcal mol–1 for 1c+, 2c+, and 3c+, respectively). The sum of electronic and thermal
free energy calculated for HS and LS model complexes in the gas phase (ꢁGSCO = 7.59, 7.91, and
7.90 kcal mol–1 for 1c+, 2c+, and 3c+, respectively) shows that the entropy term is essentially
independent on the nature of the halogen (ꢁSSCO = 1.18·10–2, 1.22·10–2, and 1.13·10–2 kcal mol–1
for 1c+, 2c+, and 3c+, respectively), so that T1/2 depends exclusively on the enthalpy term (|ꢁHLSꢀ
HS/ꢁSLSꢀHS| = 344, 351, and 359 for 1c+, 2c+, and 3c+, respectively). The trend in the ꢁHSCO
values was confirmed when solvation was implicitly considered at IEFꢀPCM SCRF level64
ACS Paragon Plus Environment
21

Crystal Growth & Design
Page 22 of 46
1
2
3
4
5
(ꢁESCO = 2.49, 2.78 and 3.11 kcal mol–1; ꢁHSCO = 3.50, 3.73, and 4.62 kcal mol–1 for 1c+, 2c+,
and 3c+, in CHCl3, respectively).
6
7
8
9
These results strongly testify for a SCO nature of the complexes 1c+–3c+, both in the solid state
and in chloroform solution. Therefore, theoretical results suggest that while monomers could
possibly show a SCO behavior, the formation of complexes with an even number of metal ions
results in the strong anti ferromagnetic interactions, evidenced experimentally (see above), that
prevent any SCO behavior in the solid state. In order to verify this hypothesis, the electronic
structure of 1a was investigated at the same level of theory, in the open– and closed–shell singlet
configurations (2S+1 = 1, 1asing), as independent HS FeIII–FeIII complex without coupling (2S+1
= 11, 1aHS-HS), and eventually as antiferromagnetic singlet (1aafc). The latter configuration was
obtained by using a broken symmetry approach65 obtained from a guess of the ground state
defined by molecular fragments, including one [Fe(qsalꢀ5,7ꢀX2)]2+ ion with five unpaired αꢀspin
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
electrons (sextet), and another one with five βꢀspin electrons (sextet). The resulting wave
function was optimized to avoid restricted/unrestricted spin instabilities and the geometry of the
complex was finally reꢀoptimized. In Table 3 some selected bond lengths are listed for the
examined electronic configurations, along with the total electronic energies and spin densities
(Figure S10) calculated for 1asing, 1aHS-HS, and 1aafc.
Table 3. Selected bond lengths calculated for 1a in the singlet configuration (1asing), with two
independent HS FeIII ions without coupling (1aHS-HS), and as antiferromagnetic singlet (1aafc).
For each configuration the total electronic energy ETOT (Hartree) and the spin densities (|e|/au3)
on the metal ions are also reported.
ACS Paragon Plus Environment
22

Page 23 of 46
Crystal Growth & Design
1
2
3
4
5
1asing
1
1aHS-HS
11
1aafc
Struct.
2S+1
1
–
6
7
8
9
FeꢀN2
1.946
1.962
1.900
1.919/1.933
1.881
2.196
2.196
2.139
2.183
1.920
1.981/2.033
2.106
–
FeꢀN1
2.206
2.206
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
FeꢀO1
1.948
1.948
FeꢀO2/O2'
FeꢀN1T
ETOT
1.981/2.042
1.972
1.981/2.042
1.972
ꢀ7177.6703036
0, 0
ꢀ7177.7815725
+4.30, +4.30
ꢀ7177.7827629
+4.29, –4.29 b
Spin density on FeIII
–
a Labels of the ligand donor atoms as in Fig. 2. b Figure S10.
A comparison of the optimized bond lengths and angles clearly shows that only the
configurations of 1aHS-HS and 1aafc reproduce reliably the coordination pattern at the iron(III)
centers. In addition, the closedꢀshell singlet displays, as expected, a RHF/UHF instability. The
total electronic energies calculated for 1aHS-HS and 1aafc are consistent with the antiferromagnetic
state being the ground state, in perfect agreement with the magnetic measurements discussed
above. Notably, the spin densities of ±4.29 |e|/au3 (Figure S10) are in excellent agreement with 5
unpaired electrons on each FeIII center, since part of the spin density delocalizes onto the ligands.
Conclusions
A new series of the tridentate Nꢀ8ꢀquinolylꢀsalicylaldimine ligands Hqsalꢀ5,7ꢀX2 (X = Cl(1),
Br(2), I(3)) and their corresponding complexes with FeIII, formulated as [Fe(qsalꢀ5,7ꢀ
X2)(NCS)(MeO)]2ꢁsolv. (X = Cl (1a), Br (2a), I (3a)), [FeIII (qsalꢀ5,7ꢀX2)4(NCS)2(MeO)2]ꢁsolv.
4
(X = Br (2b), I (3b; solv = 4 CH2Cl2) have been synthesized and structurally characterized. 1a
and 2a are isostructural dimers where each FeIII metal is bound by the N,N,O tridentate qsalꢀ5,7ꢀ
ACS Paragon Plus Environment
23

Crystal Growth & Design
Page 24 of 46
1
2
3
4
5
6
7
8
9
X2 (X = Cl and Br) ligand, a SCNꢀ anion and two bridging methanolate anions in a distorted
octahedral environment. The SCNꢀ anion exhibit a linear geometry since it is bound to the metal
center through the nitrogen atom. Several Cl···Cl (3.39 Å) and Br···Br (3.51 Å) halogen
intermolecular interactions with symmetry related halogen atoms, CH···S with the aromatic ring
and the SCNꢀ anion and πꢀπ stacks with adjacent bromoꢀquinoline rings occur between adjacent
molecules. In 3a iodine atoms interact through a linear I22···S1T geometry with S1T of the SCNꢀ
group via halogen bonding. 2b and 3b are centrosymmetric tetramers where each FeIII is bound
by three nitrogen atoms and three oxygen atoms deriving from the tridentate qsalꢀ5,7ꢀX2 (X = Br
and I), a SCNꢀ anion, a bridging methanolate and a bridging oxo moiety. Unexpectedly, the haloꢀ
substitution at the 5,7 position of the aminoquinoline moiety yield dimers and tetramers instead
of the mononuclear complexes with a dramatic change of the magnetic properties of the reported
complexes, on going from SCO magnetic behavior observed in the mononuclear complexes to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
strong antiferromagnetic interactions between FeIII centers (S = 5/2), mediated by the alkoxo
66,67
bridges in dimers and tetramers herein reported.
DFT Calculations support the structural
finding, showing that monomers could possibly show a SCO behavior, even though the
formation of complexes with an even number of metal ions results in the strong anti
ferromagnetic interactions. DFT clearly show as well that the antiferromagnetic coupling of the
two HS FeIII centers in 1a results in the lowest energy electron configuration of the complex.
In conclusion the search for halogenꢀbonded systems where halogenꢀbonds dominates the
structure and influence the physical properties of a given material still represents a challenge in
material science, even though the results reported in this work show that a proper choice of the
halogen substitution on the ligand coordinated to the metal can provide chemists with a versatile
strategy for the crystalꢀengineering design of new compounds with novel physical properties.
ACS Paragon Plus Environment
24

Page 25 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
Further studies will involve the replacement of halogen atoms with different substituents in order
to investigate their role, both at the electronic and supramolecular level, in determining the
molecular packing pattern and the related physical properties.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
Materials.
All manipulations were performed in air with reagent grade solvents. All organic chemicals
were purchased from TCI (Zentek) and were used without further purification, and inorganic
chemicals were purchased from SigmaꢀAldrich and used as received. FTꢀIR spectra were
performed on KBr pellets and collected with a Bruker Equinox 55 spectrophotometer. Melting
points were obtained on a Kofler hot stage microscope and are uncorrected. Elementary analyses
were performed by –CHNS/O Perkin Elmer 2400 series II. The aminoquinoline derivative was
separated by a chromatographic column (25mm x 180 mm) filled with DAVISIL silica gel (40ꢀ
63 m), packed with Buchi Cꢀ670 cartridge.
Synthesis of 5,7-dichloro-8-aminoquinoline. A solution of sulfuryl chloride (60.0 mmol) in
12.5 mL of glacial acetic acid was added slowly with stirring to a solution of 8ꢀaminoquinoline
(4.38 g, 30.4 mmol) in 21 mL of glacial acetic acid cooled to 10°C by an ice bath. The mixture
was then warmed gradually to room temperature and finally was heated for 25 min on the steam
bath at 70ꢀ75°C. Then, the reaction mixture was quenched by a solution of 20 g of sodium
acetate in 150 mL of water. The precipitated crude product was removed by filtration and dry in
air. Yield: 4.98g (77 %) Melting point: 183ꢀ184°C. ESIMS(+): 213.00 m/z (M+H+) with the
expected isotope distribution.
ACS Paragon Plus Environment
25

Crystal Growth & Design
Page 26 of 46
1
2
3
4
5
6
7
8
9
Synthesis of 2-[(5,7-Dichloroquinolin-8-ylimino)-methyl]-phenol (Hqsal-5,7-Cl2), 1. 5,7ꢀ
dichloroꢀ8ꢀaminoquinoline (3.34g, 15.7 mmol), salicylaldehyde (2.02g, 24.3 mmol) and glacial
acetic acid (0.34 mL) were dissolved in 40 mL of dry ethanol . The reaction mixture was stirred
at room temperature for 14 hours. The end of the reaction was verified by TLC (2:8 of
diethylether: petroleum ether. When the reaction was completed, the mixture reaction was
filtered, washed, dried in air. Yield: 3.74 (74.9 %). Melting point: 201°C with decomposition.
Elem anal. Calcd for C16H10Cl2N2O (found), mass %: C, 60.59 (59.73); H, 3.18 (2.94); N, 8.33
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(8.55). FTIR: νmax(KBr)/cmꢀ1 3435 (νOH), 3057, 2921 (νCH), 1621 (νC=N). H NMR
(DMSOꢀd6/DCl): δ = 13.45 (s), 9.41 (s), 9.25 (dd), 8.89 (dd), 8.06 (s), 7.86 (m), 7.78 (dd), 7.74
(td), 7.61 (s), 7.37 (d), 7.28 (td). ESIMS(+): 317.1 m/z (M+H+), with expected isotope
distribution.
Synthesis of 5,7-dibromo-8-aminoquinoline. NꢀBromosuccinimide (2.44 g, 13.0 mmol ) was
added slowly to a stirred solution of 8ꢀaminoquinoline (1.0 g, 6.9 mmol) and (1.4g,) of
NaClO4/SiO2 1:4 ratio in CH2Cl2 (20 mL). The reaction mixture was stirred at room temperature
for 5ꢀ10 min and the progress controlled by TLC (50:50 of diethylꢀether: petroleum ether). When
the reaction was completed the mixture was filtered and the catalyst washed with CH2Cl2 (2x10
mL). Then, the combined organic layer was washed with water, dried over anhydrous sodium
sulfate and concentrated under reduced pressure using a rotary evaporator. Further purification
was carried out by flash chromatography on silica gel (diethylether/ petroleum ether: 4/6). Yield:
1.39 (66.9 %). Melting point: 143 °C. ESIMS(+): 303.1 m/z (M+H+) with expected isotope
distribution.
Synthesis of 2-[(5,7-Dibromoquinolin-8-ylimino)-methyl]-phenol (Hqsal-5,7-Br2), 2. 5,7ꢀ
dibromoꢀ8ꢀaminoquinoline (1.60g, 5.3 mmol), salicylaldehyde (1.00g, 8.1 mmol) and glacial
ACS Paragon Plus Environment
26

Page 27 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
acetic acid (0.1 mL) were dissolved in dry ethanol (40 mL). The reaction mixture was stirred at
room temperature for 22 hours. The end of the reaction was verified by TLC (diethylether/
petroleum ether2:8). When the reaction was completed, the mixture reaction was filtered,
washed, dried in air. Yield: 1.9g (89.0 %). Melting point: 180°C with decomposition . Elem anal.
Calcd for C16H10Br2N2O (found), mass %: C, 47.33 (46.68); H, 2.48 (2.23); N, 6.90 (6.64).
FTIR: νmax(KBr)/cmꢀ1 3435 (νOH), 3049, 2962 (νCH), 1621 (νC=N). 1H NMR (DMSOꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
d6/DCl): δ = 13.32 (s), 9.30 (s), 9.19 (dd), 8.84 (dd), 8.41 (s), 7.84 (m), 7.76 (dd), 7.72 (td), 7.54
(s), 7.36 (d), 7.27 (td). ESIMS(+): 404.9 m/z (M+H+), with expected isotope distribution.
Synthesis of 5,7-diiodo-8-aminoquinoline. Potassium iodide (1.0 g, 6.0 mmol) was added
slowly to a stirred solution of 8ꢀaminoquinoline (0.43 g, 3.0 mmol), NaIO4 (1.28 g, 6.0 mmol),
NaCl (0.69 g, 12.0 mmol) in 40 mL of acetic acid/water (9/1). The reaction mixture changes
from orange to black, then it was stirred at room temperature for 3 hours. When the reaction was
completed to the mixture was added 30 mL of water, extracted with CH2Cl2 (3x20 mL). Then,
the combined organic layer was dried over anhydrous sodium sulfate and concentrated under
reduced pressure using a rotary evaporator. Further purification was carried out by flash
chromatography on silica gel (diethylether/petroleum ether: 1/1 ). Yield: 0.66 g (44.0 %) Melting
point: 144ꢀ145°C with decomposition. ESIMS(+): 397.1 m/z (M+H+).
Synthesis of 2-[(5,7-Diiodoquinolin-8-ylimino)-methyl]-phenol (Hqsal-5,7-I2), 3. 5,7ꢀ
diiodoꢀ8ꢀaminoquinoline (0.58 g, 1.5 mmol), salicylaldehyde (0.39 g, 1.6 mmol) and glacial
acetic acid (0.1 mL) were dissolved in dry ethanol (20 mL). The reaction mixture was stirred at
room temperature for 14 hours. The end of the reaction was verified by TLC (2/8 of
diethylether/petroleum ether. When the reaction was completed, the mixture reaction was
filtered, washed and dried in air. Yield: 0.35g (48.0 %). Melting point: 173ꢀ174°C with
ACS Paragon Plus Environment
27

Crystal Growth & Design
Page 28 of 46
1
2
3
4
decomposition. Elem anal. Calcd for C16H10I2N2O (found), mass %: C, 38.43 (39.12); H, 2.02
5
6
(2.31); N, 5.60 (5.32). FTIR: νmax(KBr)/cmꢀ1 3441 (νOH), 3057, 2921 (νCH), 1612 (νC=N). 1H
7
8
9
NMR (DMSOꢀd6/DCl): δ = 13.19 (s), 9.23 (s), 9.17 (dd), 8.95 (s), 8.75 (dd), 7.86 (m), 7.81 (m),
7.62 (s), 7.44 (d), 7.34 (td). Mass spectrometry (MS): 501.1 m/z (M+H+).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of [Fe(qsal-5,7-Cl2)(NCS)(MeO)]2 (1a) Compound 1 (20.0 mg, 0.06 mmol) was
dissolved in CH2Cl2 (5ml) with Et3N (8.36 ꢂL, 0.06 mmol) to give a yellow solution and was
layered at the bottom of test tube. FeCl3ꢁ6H2O (8.10 mg, 0.03 mmol) in 1.5ml MeOH and
NaSCN (2.68 mg, 0.03 mmol) in 1.5 mL MeOH were sonicated for 5 min to give dark purple
solution and were carefully layered on top of the above layer. A blank solution of MeOH (1mL)
was layered in between the above mentioned solutions. Dark red crystals were formed at the
interface after 5 days. Yield = 54.5 %. Elem Anal. Calcd for C36H24Cl4Fe2N6O4S2 (found), mass
%: C, 46.88 (47.12); H, 2.62 (2.54); N, 9.11(8.87), Cl 15.38(15.22), O 6.94(6.95), S 6.95(6.56),
Fe 12.11(12.01). FTIR: νmax(KBr)/cmꢀ1 2920 (νArꢀH), 2061 (νC≡N), 1605 (νC=N). MS
(ESI+): m/z 688 [Fe(qsalꢀCl2)]+.
Synthesis of [Fe(qsal-5,7-Br2)(NCS)(MeO)]2 (2a), [Fe4(qsal-Br)4(NCS)2(MeO)2] (2b)
The synthetic procedure of 2a was the same as 1a but using 2 (25.0 mg, 0.06 mmol) instead of
1. The use of a very slight excess of Et3N (10.03 ꢂL, 0.07mmol) yielded 2b. Yield=56.7 % (2a)
and 48% (2b). Elem Anal. Calcd C36H24Br4Fe2N6O4S2 2a (found), mass %: C, 39.31 (39.12); H,
2.2 (2.31); N, 7.64 (7.32) Br, 29.05(28.62); O 5.82(5.78), S 5.83(5.56), Fe 10.15(10.91) Elem
Anal. Calcd C68H42Br8Fe4N10O8S2 2b (found), mass %: C, 39.77 (39.12); H, 2.06 (1.99); N, 6.82
(6.32) Br, 31.12(30.52); O 6.23 (5.83), S 3.12 (3.42), Fe 10.88 (10.71). FTIR: νmax(KBr)/cmꢀ1
2922 (νArꢀH), 2049 (νC≡N), 1603 (νC=N). MS (ESI+): m/z 866 [Fe(qsalꢀBr2)]+.
ACS Paragon Plus Environment
28

Page 29 of 46
Crystal Growth & Design
1
2
3
4
5
6
7
8
9
Synthesis of [Fe(qsal-5,7-I2)(NCS)(MeO)]2 (3a), [Fe4(qsal-5,7-I2)4(NCS)2(MeO)2] (3b)
The synthetic procedure of 3a was the same as 1a but using 3 (30 mg, 0.06 mmol) instead of 1.
The use of a very slight excess of Et3N (10.03 ꢂL, 0.07 mmol) yielded 3b. Yield = 49.34% (3a)
and 41% (3b). Elem Anal. Calcd C37H28Fe2I4N6O5S2 3a (found), mass %: C, 33.66 (33.75); H,
2.14 (2.39); N, 6.37 (6.22) I, 38.45(37.54); O 6.06 (7.1), S 4.86 (4.74), Fe 8.46(8.26) Elem Anal.
Calcd C72H50Cl8Fe4I8N10O8S2 3b (found), mass %: C, 31.22 (30.24); H, 1.82 (1.85); N, 5.06
(4.77), Cl 10.24 (8.47) I, 36.66(34.54); O 4.62 (4.35), S 2.32 (2.18), Fe 8.07 (7.6). FTIR:
νmax(KBr)/cmꢀ1 2919 (νArꢀH), 2054 (νC≡N, 3a), 2046 (νC≡N, 3b), 1601 (νC=N). MS (ESI+):
m/z 1054 [Fe(qsalꢀI2)]+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
X-ray Crystallography. Single crystal data were collected with a Bruker APEX DUO
diffractometer with a Quazar MX Multilayer Optics diffractometer (Mo Kα radiation; λ =
0.71073 Å) at 100 K (1a, 2a, and 2b) and with a Bruker Smart APEXII at 200 K (3a and 3b). A
summary of data collection and structure refinement for 1aꢀ3a, 2b, and 3b are reported in Table
1. The structure was solved by direct methods (SHELXSꢀ97) and refined on F2 with fullꢀmatrix
least squares (SHELXLꢀ97), using the Wingx software package. The nonꢀH atoms were refined
with anisotropic displacement parameters. The unit cell parameters were obtained using 60 ωꢀ
frames of 0.5° width and scanned from three different zone of reciprocal lattice. The intensity
data were integrated from several series of exposures frames (0.3° width) covering the sphere of
reciprocal space68. Absorption correction were applied using the program SADABS 69. The
structures were solved by the dual space algorithm implemented in the SHELXT code.70 Fourier
analysis and refinement were performed by the fullꢀmatrix leastꢀsquares methods based on F2
implemented in SHELXLꢀ2014.71 Graphical material was prepared with the Mercury 3.972
ACS Paragon Plus Environment
29