Do the chemists
know how
to align the spins
of electrons
in molecules,
parallel or antiparallel ?
to get magnetic compounds …
Understanding …
why the spins of two neighbouring electrons
(S = 1/2) become :
antiparallel ?
S=O
or parallel ?
S=1
Interaction Models between Localised Electrons
Sa
1
2
A
B
Sb
^
^ S
^
H = - J S
1 2
Scalar Coupling : describes, does not explain
Energy levels
J =
2 k
>0
if S = 0
Orthogonality
+
4ßS
<0
if S≠0;|ßS|>>k
Overlap
O2
H2
Hund
Aufbau
J. Miró
« Overlap » ?
Catalogue raisonné, N°1317
J. Miró, Pomme de terre, detail
An old game …
^
^
^
H = - J S1S
2
AF
F
Palace Museum, TaiPei, Neolithic period, Yang-Shao Culture
Exchange interactions can be very weak …
Energy
Exchange interactions
order of magnitude :
cm-1 or Kelvins …
≈
« Chemical » bonds
Robust !
order of magnitude :
>> 150 kJ mol-1 …
Michelangelo, Sixtin Chapel, Rome
Cu(II)
Cu(II)
≈ 5 Å
Negligible Interaction !
Problem :
How to create the interaction … ?
Cu(II)
Ligand
Cu(II)
≈ 5 Å
Orbital Interaction …
Solution :
The ligand !
A
Ligand
B
Examples with the ligand
• Cyanide
Non linear and linear bridges
Monet Claude, Charing Cross Bridge
Monet Claude, Waterloo Bridge
CNCyanide Ligand
Friendly ligand : small, dissymetric, forms stable complexes
Warning : dangerous, in acid medium gives HCN, lethal
Dinuclear µ-cyano
homometallic complexes
“Models” Compounds Cu(II)-CN-Cu(II)
J/cm-1
Compounds
exp
[Cu2(tren)2CN]3+
-160
[Cu2(tmpa)2CN]3+
-100
Overlap : antiferromatic coupling …
Rodríguez-Fortea et al. Inorg. Chem. 2001, 40, 5868
Cr(III)
Ni(II)
Dinuclear µ-cyano
heterometallic complexes
NB : A dissymetric ligand helps to get stable heterometallic complexes …
« Birds of the same feathers flock together » …
Polynuclear complex,
3-
synthetic strategy
2+
9+
+ 6
Hexacyanometalate “Heart” Mononuclear Complex
Lewis Base
Lewis Acid
Polynuclear Complex
Electrons in
Hexacyanochromate complex
eg
[CrIII(CN)6]3-
² oct
t2g
Cr(III)
z
t2g
x
Polynuclear complex,
ferromagnetic strategy
M-CN-M'


Orthogonality :
Ferromagnetism !
Example :
Cr(III) (t2g)
3
JF
Ni(II),(e g)2
Cr(III)Ni(II)6
S = 6x1+3/2 =
S = 15/2
Polynuclear complex,
ferrimagnetic strategy
M-CN-M’


Overlap = antiferromagnetism
Example
Cr(III) (t2g)3
Cr(III)Mn(II)6
JAF
Mn(II) (t2g)3
S = 6x5/2 + 3/2 =
S = 27/2
Ferromagnetic …
Paul KLEE
Eros
Ferrimagnetic …
Nocturnal separation, 1922
… High Spin Heptanuclear Complexes
C rC u6
S = 9/2
Hexagonal
R -3
a = b = 15,27 Å; c = 78,56 Å
a = b= 90°; g = 120°; V = 4831 Å3
C rNi6
S = 15/2
Hexagonal R -3
a = b = 15,27 Å; c = 41,54 Å
a = b= 90°; g = 120°; V = 8392 Å3
C rM n 6
S = 27/2
Hexagonal R -3
a = b = 23,32 Å; c = 40,51 Å
a = b= 90°; g = 120°; V = 19020 Å3
Marvaud et al., Chemistry, 2003, 9, 1677 and 1692
Chemists have transformed good old Prussian blue,
a blue pigment that Michael Faraday precipitated
in his time at Royal Institution,
into a room temperature magnet
also useful in display devices.
Alfred Bader, « End of Mistery », Chemistry in Britain, 37, 7, July 2001, 99
Greeting Card of RSC, Benevolent Fund, Thomas Graham House, Science Park, Milton Road,
Registered Charity N° 207890.
This painting does not depict neither W.T. Brande nor Michael Faraday making Prussian blue,
Thomas Philips RA, ca 1816 (from Alfred Bader, Hon FRSC)
1704 …
2004 : 300th anniversary !
Diesbach, draper in Berlin …
… prepares a blue pigment « Prussian blue »
… said to be the first coordination compound
Anna Atkins, Cyanotype
in Hart-Davis Adam
Chain Reactions,
pioneers of british science
and technology
National Portrait Gallery
London, 2000
Classical coordination chemistry …
Fe2+aq+ 6CN-aq
[Fe(CN)6]4-aq
OH2
+
Fe
[4-]
H2 O
H2 O
Fe
OH2
OH2
OH2
[3+]
3[Fe(CN)6]4-aq + 4Fe3+aq
{Fe4[Fe(CN)6]3}0•15H2O
Complexes as Ligands, or « bricks »
+ Lewis Acid-Base Interaction
since : 1936, modified 1972 …
an evergreen in inorganic chemistry …
Stoichiometry
[AII]4[BII(CN)6]3
or A4B3
[AII]4 [BII] 3 1
 = vacancy
a simple face-centered cubic structure …
J.F. Keggin, F.D. Miles, Nature 1936, 137, 577
A. Ludi, H.U. Güdel, Struct. Bonding (Berlin) 1973, 14, 1
Coming back to Prussian Blue
TC  z |J|
z : number of magnetic
neighbours
|J| : coupling constant
between nearest
neighbours
Néel, Annales de Physique, 1948
Coming back to Prussian Blue
TC  z |J|
z : number of magnetic
neighbours
|J| : coupling constant
between nearest
neighbours
TC = 5.6 K
Néel, Annales de Physique, 1948
Towards Prussian blue analogues …
TC  z |J|
JFerro > 0
Orthogonality
TC >> 5.6 K
Towards Prussian blue analogues …
TC  z |J|
JAntiferro < 0
Overlap …
TC >> 5.6 K
V4[Cr(CN)6]8/3.nH2O
Room Temperature TC
On a rational basis !
Ferlay et al. Nature, 1995
Mallah et al. Science 1993
Gadet et al., J.Am. Chem. Soc. 1992
V4[Cr(CN)6]8/3.nH2O
Room Temperature TC
On a rational basis !
2[CrIII (CN)6 ]3- +3Vaq 2+
[V 3[CrIII (CN6)]2]0
A blue, transparent, low density magnet at room temperature
Oscillating Magnet :
Experiment
Image
(2)
Magnet
Molecular
(3)
Holder
Screen
(1)
Len
Light, Sun …
(2)
Permanent
Magnet
A thermodynamical machine transforming
Light
in
Mechanical
Energy
QuickTime™ et un décompresseur
DV - PAL sont requis pour visualiser
cette image.
Other device
Magnetic Switch…
Or thermal probe …
Couple
Permanent
Magnet
Hot
Sample ( MM)
Up to 2004 …
magnetic analogues used as …
QuickTime™ et un décompresseur
DV - PAL sont requis pour visualiser
cette image.
… devices and demonstrators
Chemists have managed
to transform
isolated single molecules
into magnets
molecule-based magnets ?
Why ?
Specific properties
Low density
Transparent
Nanosized, identical molecules
Often biocompatible and biodegradable
Very flexible chemistry
Mild chemistry : Room T, Room P,
Solution Chemistry
To improve
Fragile
Aging
Diluted
To overcome
Top down
3D
Metals
Oxydes
•
•
•
New Physics
Quantum / Classical
Quantum tunneling
Fragments
Threads
Dots
•
Nanosystems •
Nice Chemistry
Single molecule magnets
Giant Molecular Clusters
0D, Molecules
Bottom up
• Applications (far …)
• Recording
• Quantum computing
… Single molecule magnets
Giant Molecular Clusters
High Spin + Anisotropy
∆E = DSz2
Mn12
Fe8
Idea
New
ConceptsTheory
Synthesis
New
Mn4
and many others
Materials
New
Properties
Functions
See Gatteschi
Hendrickson
Christou
Winpenny …
What is named
Single Molecule Magnet ?
=
High Spin
Anisotropic
High Spin Paramagnetic Molecule
H
Single Molecule Magnet
Below
T < TBlocking
H
Towards information storage at the molecular level ?
Single Molecule Magnet
Below
T < TBlocking
Towards information storage at the molecular level ?
Single molecule magnets
z
E
0
Thermal
Activation
y
DSz2
Anisotropy Barrier
x
Tunneling
- Sz
DSz = 400K ?
2
-4 -2 0 +2 +4
|D| = 1K
S = 20
Sz
+Sz
(D < 0)
N
C
2nd generation
NC
CN
M
NC
= Ni(II)(tetren)
Spin = 1
CN
C
N
= Ni(II)(Rad°)2
Spin = 2
Rad°
1rst generation
N
C
NC
CN
M
Complex
NC
CN
C
N
N
C
NC
CN
M
NC
CN
C
N
K. Vostrikova, P. Rey et al., JACS 2000, 122, 718
Some examples …
S = 14/2
S = 39/2 (AF)
S = 27/2
Rey, JACS 2000, 122, 718
Decurtins, Angewandte, 2000
Hashimoto, JACS, 2000
Marvaud, Chemistry, 2003, 9, 1677 y 1692
CoCu2
CrNi
CoCo2
CoNi2
CrNi2
7/2
5/2
CoCu3
CoCo3
CoNi3
Marvaud et al., Chemistry, 2003, 9, 1677 and 1692
Ariane Scuiller, Caroline Decroix, Martine Cantuel, Fabien Tuyèras …
CoNi5
Anisotropy
CoCo2
CoCu2
CrNi
CoNi2
7/2
CrNi2
5/2
CoCu3
High spin
CoCo3
CrCo3
CrMn6 27/2
CoMn6
15/2
CrNi6
CoNi3
CrNi3
9/2
CrCu6
CoCu6
CoCo6
CoNi5
CrNi3
V. Marvaud
[Mn12O12(CH3COO)16(H2O)4].2CH3COOH.4H20
or Mn12
Mn(III)
S=2
Mn(IV)
S=3/2
Ion Oxyde
Carbone
S
=8x2 -4x3/2 =
S=10
From D.Gatteschi and R. Sessoli
Mn12 is a hard magnet
Remnant Magnetisation
MAGNETIZATION ( B )
20
Bistability : in
zero field the
magnetisation can
be positive or
negative depending
of the story of
the sample
T=2.1K
10
0
-10
Coercive Field
-20
-3
-2
-1
0
1
2
3
M A G N ETIC FIELD (T)
From D.Gatteschi and R. Sessoli
Ground State Energy Levels
M=±8
M=±9
H = 0
M=±10
From D.Gatteschi and R. Sessoli
Energy Levels in a Magnetic Field
S
-S
H0
M=S
M=-S
At low temperature, a magnetic field populates
only the M = -S state
Going back to equilibrium :
Thermal activation :trivial
Axial symmetry
E(M) = DM2
H=0
M=S
E=DS2
M=-S
 = 0 exp(E/kBT)
From D.Gatteschi and R. Sessoli
Towards equilibrium :
Tunneling effect : new !
H=0
M=S
M=-S
From D.Gatteschi and R. Sessoli
Mn12 is a Hard Magnet
M
Mremnant
Msaturation
H
Hcoercive
+ Steps in the magnetisation curve
From D.Gatteschi and R. Sessoli
Resonnant Tunneling Effect
for H = nD/gB
M=S
H = nD/gB
M=-S
From D.Gatteschi and R. Sessoli
Conditions to observe
tunnelling effect
• Degenerated wave functions must superpose
• A transversal field must couple the two wave
functions
• Coupling splits the two levels : “tunnel splitting”
• Tunnelling Effect Probability increases with
tunnel splitting”
From D.Gatteschi and R. Sessoli
M
Tunneling Effect
H
From D.Gatteschi and R. Sessoli
No resonnant Tunneling Effect
with a magnetic field parallel to z
M=S
H  nD/gB
M = -S
From D.Gatteschi and R. Sessoli
M
H
No tunneling effect
From D.Gatteschi and R. Sessoli
Feasibility of « Molecular nanowires » (or SCM) ?
Anisotropic precursor
[Fe(III)(bipy)(CN)4]R. Lescouëzec, M. Julve, Valencia, Spain D. Gatteschi, W. Wernsdorfer
Angewandte Chem. 2003, 142, 1483-6
Trinuclear species
Double
zig-zag chains
Bis double
zig-zag chains
[{FeIII(L)(CN)4}2CoII(H2O)4] .4H2O
[{FeIII(L)(CN)4}2MII(H2O)4].4H2O
[L = 2,2’-bipy and 1, 10-phen),
M = Mn, Co, Cu, Zn]
[{FeIII(bpy)(CN)4}2MII(H2O)] .
CH3CN . 1/2H2O [M = Mn, Co, Cu]
Slow relaxation of the magnetisation …
Magnetization as a
function of time
1
1.5 K
M/M s
0.8
0.6
2.4 K
2.5 K
1.6 K
2.3 K
2.6 K
2.2 K
2.7 K
Thermally activated
relaxation of the
magnetization
ac: Ea= 142 K, 0 = 6.10-11 s
2.8 K
1.8 K
2.0 K
1.9 K
0.2
0
0.01
1.7 K
2.1 K
0.4
0.1
1
10
t (s)
100
1000
M vs. t plots along the b axis.
W. Wernsdorfer, Grenoble
The dream …
Magnetic Tip
H
High Spin
"down"
High Spin
"down"
≈ 10 nm
Surface
Magnetic Tip
H
High Spin
"down"
High Spin
"down"
≈ 10 nm
Surface
The dream …
Magnetic Tip
H
HSM «up»
High Spin
"down"
≈ 10 nm
Surface
… information storage at the molecular level !
Nanosciences …
H
HSM «up»
High Spin
"down"
≈ 10 nm
Surface
… a challenge for chemists and friends …