Chemisorption of exchange-coupled [Ni2L(dppba)]+ complexes on gold by using ambidentate 4-(diphenylphosphino)benzoate Co-ligands

Article abstract of DOI:10.1002/chem.201300496

A new strategy for the fixation of redox-active dinickel(II) complexes with high-spin ground states to gold surfaces was developed. The dinickel(II) complex [Ni₂L(Cl)]ClO₄ (1ClO₄), in which L ^2- represents a 24-membered macrocyclic hexaaza-dithiophenolate ligand, reacts with ambidentate 4-(diphenylphosphino)benzoate (dppba) to form the carboxylato-bridged complex [Ni₂L(dppba)]⁺, which can be isolated as an air-stable perchlorate [Ni₂L(dppba)]ClO₄ (2ClO₄) or tetraphenylborate [Ni₂L(dppba)]BPh₄ (2BPh₄) salt. The auration of 2ClO₄ was probed on a molecular level, by reaction with AuCl, which leads to the monoaurated Ni ^II₂Au^I complex [Ni^II ₂L(dppba)Au^ICl]ClO₄ (3ClO₄). Metathesis of 3ClO₄ with NaBPh₄ produces [Ni ^II₂L(dppba)Au^IPh]BPh₄ (4BPh ₄), in which the Cl^- is replaced by a Ph^- group. The complexes were fully characterized by ESI mass spectrometry, IR and UV/Vis spectroscopy, X-ray crystallography (2BPh₄ and 4BPh ₄), cyclic voltammetry, SQUID magnetometry and HF-ESR spectroscopy. Temperature-dependent magnetic susceptibility measurements reveal a ferromagnetic coupling J=+15.9 and +17.9 cm^-1 between the two Ni ^II ions in 2ClO₄ and 4BPh₄ (H=-2 JS ₁S₂). HF-ESR measurements yield a negative axial magnetic anisotropy (D<0), which implies a bistable (easy axis) magnetic ground state. The binding of the [Ni₂L(dppba)]ClO₄ complex to gold was ascertained by four complementary surface analytical methods: contact angle measurements, atomic-force microscopy, X-ray photoelectron spectroscopy, and spectroscopic ellipsometry. The results indicate that the complexes are attached to the Au surface through coordinative Au-P bonds in a monolayer. Copyright

Full text of DOI:10.1002/chem.201300496

FULL PAPER
DOI: 10.1002/chem.201300496
Chemisorption of Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes on Gold
by Using Ambidentate 4-(Diphenylphosphino)benzoate Co-Ligands**
Matthias Golecki,[a] Jochen Lach,[a] Alexander Jeremies,[a] Frank Lungwitz,[b]
Michael Fronk,[b] Georgeta Salvan,[b] Dietrich R. T. Zahn,[b] Jaena Park,[c, d]
Yulia Krupskaya,[d] Vladislav Kataev,[d] Rꢀdiger Klingeler,[c] Bernd Bꢀchner,[d]
Benjamin Mahns,[d] Martin Knupfer,[d] Pablo F. Siles,[d, g] Daniel Grimm,[d, g]
Oliver G. Schmidt,[d, g] Andreas Reis,[e] Werner R. Thiel,[e] Daniel Breite,[f]
Bernd Abel,[f] and Berthold Kersting*[a]
Abstract: A new strategy for the fixa-
tion of redox-active dinickel(II) com-
plexes with high-spin ground states to
gold surfaces was developed. The di-
nickel(II) complex [Ni2L(Cl)]ClO4
(1ClO4), in which L2ꢀ represents a 24-
membered macrocyclic hexaaza-dithio-
phenolate ligand, reacts with ambiden-
N
tween the two NiII ions in 2ClO4 and
4BPh4 (H=ꢀ2 JS1S2). HF-ESR meas-
urements yield a negative axial mag-
netic anisotropy (D<0), which implies
a bistable (easy axis) magnetic ground
state. The binding of the [Ni2L-
AHCTUNGTRENNUNG
2
AHCTUNGTERNN(GUN dppba)]ClO4 complex to gold was as-
certained by four complementary sur-
face analytical methods: contact angle
measurements, atomic-force microsco-
py, X-ray photoelectron spectroscopy,
and spectroscopic ellipsometry. The re-
sults indicate that the complexes are at-
tached to the Au surface through coor-
tate
(dppba) to form the carboxylato-bridg-
ed complex [Ni2L
(dppba)]+, which can
be isolated as an air-stable perchlorate
[Ni2L(dppba)]ClO4 (2ClO4) or tetra-
phenylborate [Ni2L(dppba)]BPh4
4-(diphenylphosphino)benzoate
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ꢀ
(2BPh4) salt. The auration of 2ClO4
was probed on a molecular level, by re-
action with AuCl, which leads to the
monoaurated NiII AuI complex [NiII L-
dinative Au P bonds in a monolayer.
Keywords: chemisorption · gold ·
ligands
·
macrocyclic ligands
· nickel · phosphorus heterocycles
2
2
[a] M. Golecki, J. Lach, A. Jeremies, Prof. Dr. B. Kersting
Institut fꢀr Anorganische Chemie
[e] A. Reis, Prof. Dr. W. R. Thiel
Fachbereich Chemie, Technische Universitꢁt Kaiserslautern
Erwin Schrçdinger Strasse 54, 67663 Kaiserslautern (Germany)
Universitꢁt Leipzig, 04103 Leipzig (Germany)
Fax : (+49)341-973-6199
E-mail: b.kersting@uni-leipzig.de
[f] D. Breite, Prof. Dr. B. Abel
Chemische Abteilung, Leibniz-Institut
fꢀr Oberflꢁchenmodifizierung e. V.
04318 Leipzig (Germany)
[b] F. Lungwitz, Dr. M. Fronk, Prof. Dr. G. Salvan,
Prof. Dr. D. R. T. Zahn
Semiconductor Physics Institute
Chemnitz University of Technology
09107 Chemnitz (Germany)
[g] Dr. P. F. Siles, Dr. D. Grimm, Prof. Dr. O. G. Schmidt
Material Systems for Nanoelectronics
Chemnitz University of Technology
[c] J. Park, Prof. Dr. R. Klingeler
Reichenhainerstrasse 70, 09107 Chemnitz (Germany)
Kirchhoff Institute for Physics, University of Heidelberg
69120 Heidelberg (Germany)
[**] dppba=4-(diphenylphosphino)benzoate.
Supporting information for this article (including X-ray crystallograph-
ic data of 2ClO4 and 4BPh4 in the CIF format and XPS spectra) is
available on the WWW under http://dx.doi.org/10.1002/
chem.201300496.
[d] J. Park, Dr. Y. Krupskaya, Dr. V. Kataev, Prof. Dr. B. Bꢀchner,
B. Mahns, Prof. Dr. M. Knupfer, Dr. P. F. Siles, Dr. D. Grimm,
Prof. Dr. O. G. Schmidt
Leibniz Institute for Solid State and Materials Research
IFW Dresden, 01171 Dresden (Germany)
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7787

Introduction
The deposition of magnetically or electronically bistable
transition-metal complexes on planar surfaces in which the
spatially addressable functional units are assembled at the
molecular level is an attractive research field,[1,2] and has
perspectives for the storage of information at the molecular
level[3,4] and in molecular electronics.[5–10] Most research to
date has focused on the physisorption or chemisorption of
phthalocyanines,[11] porphyrines,[12] spin-crossover com-
pounds, and more recently also single molecule magnets,[13,23]
and several strategies have been developed to obtain these
materials as rows,[14] thin films,[15,16] or multilayers.[17] The
thermal evaporation of polynuclear transition-metal com-
plexes, however, has found limited applicability in surface
deposition due to their limited thermal and kinetic stabili-
ty,[18] and so new solution-based methods for surface func-
tionalization had to be devised. Due to the high affinity of
Figure 1. Left: Structure of dinuclear complexes [M2L(L’)]+ of the macro-
cyclic ligand H2L. Right: Sketch of chemisorbed
[Ni2L(O2CC6H4PPh2)]ClO4 complex on a gold surface.
a
Since the magnetic properties of the [Ni2L(L’)]+ com-
plexes can be switched by using chemical[33,34] or electro-
chemical methods,[35] we became interested in the immobili-
zation of such complexes on surfaces. We designed a strat-
egy that is based on the direct deposition of SMMs on surfa-
ces. We anticipated that the [LNi2]2+ fragment would react
with the hard donor function (H) of an ambidentate co-
thiolates for gold,[19,20] the ease of Au S bond formation
ꢀ
(i.e., deposition of thiols from solution) and the possibility
of surface patterning by soft-lithography,[21] the surface fixa-
tion of open-shell transition-metal complexes bearing thiol
functions have therefore become the focus of intense inves-
tigations.
ꢀ ꢀ
ꢀ
ligand H Y S to form a well-defined and isolable [LNi
(H
+
ꢀ
Y S)] complex as shown in Figure 1. The macrocyclic sup-
porting ligand L was expected to impose thermodynamic
and kinetic stability on the complex, thus preventing a back-
biting of the soft donor function (S) with the cluster core.
Two major strategies have evolved in the field of molecu-
lar magnetism that allow the deposition of polynuclear tran-
sition-metal carboxylato complexes with single molecule
magnetism (SMM) behavior on gold. In the first approach, a
+
ꢀ ꢀ
The soft phosphine donor in the precursor [LNi2ACTHNUTRGNE(NUG H Y S)]
self-assembled monolayer of a bifunctional thiol HS X L is
would then bind to the gold surface.[36,37] Notice the salt-like
composition of the adsorbate, containing monocations and
anions. Thus, the anion is deposited along with the cation,
but is only physisorbed.
ꢀ ꢀ
prepared (L is usually a carboxylate function, whereas X is
a spacer), which is then reacted with the carboxylato cluster
by using a ligand exchange reaction.[22] In the second ap-
proach, a SMM molecule with a properly functionalized
ligand shell is directly deposited.[23,24] Both strategies have
their advantages and disadvantages. In the latter method,
for example, the thiol has to be protected as it reacts with
the cluster core. The solution methods have the advantage
that they can be applied to more relevant substrates, like sil-
icon, silicon oxides, or ferromagnetic metal oxides (such as
La0.7Sr0.3MnO3).
The present study focuses on the ambidentate 4-(diphe-
nylphosphino)benzoate ligand (Hdppba)[38] with a soft di-
phenylphosphine and a hard carboxylate function. We
report the preparation and isolation of the complex salts
[Ni2LACHTNUTRGENNUG(dppba)]ClO4 and [Ni2LCAHTUNGTRNE(NUGN dppba)]BPh4, their successful
auration with AuCl, and their chemisorption on gold. The
crystal structures, electrochemistry, reactivity features, mag-
netic and electrochemical properties of the complexes are
described, as are the results of four complementary surface
analytical methods: contact angle measurements, atomic-
force microscopy, X-ray photoelectron spectroscopy, and el-
lipsometry. The coordination chemistry of transition-metal
complexes with phosphine complexes is well-established.
Phosphine-protected gold nanoparticles were reported.[39–44]
Little work has been performed in the field of phosphines
adsorbed on gold surfaces;[36,37,45] to the best of our knowl-
edge, this is the first approach for the chemisorption of ex-
change coupled metal clusters through ambidentate phos-
phine-carboxylate ligands.
Our team is interested in the synthesis and characteriza-
tion of exchange-coupled transition-metal complexes stabi-
lized by macrocyclic polyaza-polythiophenolate ligands. Par-
ticularly well-investigated are dinuclear complexes of the
N6S2 macrocycle H2L (Figure 1).[25,26] The N6S2 macrocycle
supports mixed-ligand complexes of the type [M2L(L’)]n+
with a doubly thiolato-bridged N3MACHTNUGTRNNEUG(m-SR)2ACHUTGNTREN(NUNG m-L’)MN3 core
structure. The complexes are distinguished by the fact that
the co-ligand L’ is readily exchanged. Thus, a large number
of closely related [NiII L(m-L’)]+ complexes (e.g., L’=Clꢀ,
ACTHNUTRGNEUNG
2
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
OHꢀ, SHꢀ, NO2 , NO3 , N3 , N2H4, CH3CO2 , HCO3 ,
H2PO4 , ClO4 , ReO4 , CrO42ꢀ, MoO42ꢀ, WO42ꢀ, alkyl car-
ꢀ
ꢀ
ꢀ
bonates (ROCO2 ), alkyl carbamates (RNCO2ꢀ), pyrazolate,
ꢀ
ꢀ
tetrazolate, and tetrahydridoborate (BH4 )) were prepared
Results and Discussion
and their stability,[27,28] reactivity,[29,30] and electronic struc-
tures[31,32] were studied as a function of the type of co-ligand
and oxidation state.
Synthesis and characterization of metal complexes: The syn-
thetic procedures are summarized in Scheme 1. Treatment
7788
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
Figure 2. ORTEP (left) and van der Waals plot (right) of the molecular
structure of the [Ni2L(O2CC6H4PPh2)]+ cation in crystals of 2BPh4. Ther-
mal ellipsoids are at 50% probability. Selected bond lengths [ꢃ] and
angles [8]: Ni1-O1 1.984(3), Ni1-N1 2.265(4), Ni1-N2 2.138(4), Ni1-N3
2.301(4), Ni1-S1 2.4742(13), Ni1-S2 2.4434(14), Ni2-O2 2.039(3), Ni2-N4
2.266(4), Ni2-N5 2.135(3), Ni2-N6 2.273(4), Ni2-S1 2.4713(16), Ni2-S2
2.4757(14); Ni2-S1-Ni1 89.47(5), Ni1-S2-Ni2 90.08(5).
those in PPh3 (103.3, 103.3, and 101.78)[48] indicating that the
structure of the dppba ligand is not markedly perturbed by
the [Ni2L]2+ fragment. To our knowledge, there are not
many paramagnetic 3d metal complexes of the dppba
ligand. However, the Walton group reported a number of
multiply bonded dimolybdenum and dirhenium complexes
with this ligand.[49] Overall, the [Ni2L]2+ complex binds the
dppba selectively through the carboxylate function in a che-
lating manner leaving an uncoordinated P atom for anchor-
ing to metal surfaces.
Scheme 1. Preparation of complexes 2–4.
We decided to examine the aurophilicity[50] of the [Ni2L-
of [Ni2L
A
ACHTUNGTREN(NGNU dppba)]ClO4 salt prior to surface deposition studies, by
benzoic acid in MeOH in the presence of NEt3 under Ar
provides a green solution from which upon addition of
treating it with suitable gold(I) compounds. An aurated spe-
cies was also sought for as a reference compound in the
LiClO4 the green, air-sensitive target compound, [Ni2L
A
XPS studies. [Ni2LACTHNGUTERNNUG
(dppba)]+ has three potential gold bind-
dppba)]ClO4 (2ClO4) precipitates in 86% yield. The electro-
spray ionization mass spectrum (ESI-MS) of a dilute metha-
nol solution of 2ClO4 exhibited a molecular ion peak for
ing sites, the two thiolate sulfur atoms from the macrocycle
L2ꢀ and the exposed P atom from the dppba ligand. No ap-
parent color changes took place when 2ClO4 was allowed to
[Ni2L
G
react with AuCl in a 1:1 ratio in CH2Cl2 for 24 h
sis agreed well with the formula 2ClO4. The presence of two
(Scheme 1), however, an ESI-MS spectrum of the green re-
action mixture showed the presence of a new species, formu-
ꢀ
IR absorption bands at 1550 cmꢀ1 (nas
(RCO2 )) and
ꢀ
1408 cmꢀ1 (ns
A
lated as the monogold(I) adduct [Ni2LACTHNGUTERNNUG
(dppba)AuCl]+ (3,
late function as anticipated. Electronic absorption bands at
650 and 1122 nm attributable to spin-allowed n2 and n1 tran-
sitions of a six-coordinated NiN3S2Ocarboxylate chromophore
suggest that this structure is maintained in the solution state
as well.[28]
m/z=1321.4), which can be isolated as a perchlorate salt
3ClO4 in 91% yield. The addition of further equivalents of
AuCl to 2ClO4 or 3ClO4 did not result in a polyaurated spe-
cies,[51] which shows that there is indeed only one gold bind-
ing site in 2.
Efforts to obtain single crystals of 2ClO4 suitable for X-
ray diffraction analysis were unsuccessful.[47] However, such
crystals could be obtained in the case of the tetraphenylbo-
rate salt 2BPh4 which was synthesized by metathesis of
2ClO4 with NaBPh4. Figure 2 shows that the dppba ligand
bridges the two NiII ions through its carboxylate function, in
a symmetrical fashion leading to a Ni···Ni distance of
3.481(2) ꢃ. The benzoate group of the dppba ligand is long
enough so that the P atom is exposed out of the binding
pocket of the [Ni2L]2+ fragment. In addition, the C-P-C
angles in 2ClO4 (101.1, 101.7, and 106.58) are very similar to
Unfortunately, we were not successful in growing single
crystals of 3ClO4. However, crystals of the tetraphenylbo-
rate salt of the related complex [Ni2LACHTNUGTRNEUNG(dppba)AuPh]BPh4
(4BPh4), which is accessible by metathesis of 3ClO4 with
NaBPh4 (Scheme 1), can be obtained by recrystallization
from a mixed CHCl3/CH2Cl2/hexanes solution. Notice that
the gold-bound chlorido ligand is replaced by a phenyl
group in this reaction. The ability of the [BPh4]ꢀ ion to
transfer a phenyl group is not unusual[52,53] and some of
these reactions occur even in aqueous solution.[54] The
phenyl compound is formed in high yield and is air stable.
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7789

B. Kersting et al.
The
crystal
structure
determination
of
were carried out to investigate the redox behavior of
4BPh4·3CH2Cl2·CHCl3·H2O revealed the presence of two
[Ni2L(dppba)]ClO4
(2ClO4),
[Ni2LACTHUGNTERNNU(G pdbba)AuCl]ClO4
crystallographically independent but chemically identical
(3ClO4), and [Ni2L
ACHTUNGTREN(UNNG pdbba)AuPh]BPh4 (4BPh4). Figure 4
molecules A and B of the trinuclear [Ni2L
A
shows the cyclic voltammograms of solutions of these com-
pounds in CH2Cl2. The CV of the acetato-bridged complex
complex 4. A perspective drawing of one of them is shown
in Figure 3, whereas selected bond lengths and angles are
[Ni2LACTHGNUTERNNUG
(OAc)]ClO4 was reported previously,[57] and is included
for comparison.
Figure 3. ORTEP (left) and van der Waals plot (right) of the molecular
structure of the [Ni2LACHTUNGTRENNUNG
(dppba)AuPh)]+ cation (molecule A) in crystals of
4BPh4·3CH2Cl2·CHCl3·H2O. Thermal ellipsoids are at 50% probability.
Selected bond lengths [ꢃ] and angles [8] for molecule a [molecule b]:
Au-C58 2.048(7) [2.041(5)], Au-P 2.3013(13) [2.294(1)], Ni1-O1 2.004(3)
[2.002(3)], Ni1-N1 2.259(4) [2.215(4)], Ni1-N2 2.144(4) [2.127(4)], Ni1-
N3 2.276(5) [2.335(4)], Ni1-S1 2.482(2) [2.465(1)], Ni1-S2 2.469(1)
[2.474(2)], Ni2-O2 2.019(3) [2.031(3)], Ni2-N4 2.232(4) [2.223(4)], Ni2-
N5 2.138(4) [2.165(4)], Ni2-N6 2.329(4) [2.298(4)], Ni2-S1 2.459(1)
[2.444(2)], Ni2-S2 2.478(2) [2.496(1)]; C58-Au-P 177.82(18) [175.73(14)],
Ni2-S1-Ni1 89.16(4) [89.54(5)], Ni1-S2-Ni2 88.99(5) [88.16(4)].
given in the Figure caption. The complexes are well separat-
I
ꢀ
ed from each other. The Au Ph fragment is attached to the
terminal phosphorus atom of the dppba co-ligand.[55] The co-
ordination geometry of the gold atom is almost perfectly
ꢀ
linear, with the C-Au-P angle being 177.8(2)8. The Au C
ꢀ
and Au P bond lengths are 2.048(7) and 2.3013(12) ꢃ, re-
spectively. The bond lengths and angles in 4 are very similar
ꢀ
ꢀ
to those in [Au(Ph)ACTHNUTRGNE(NUG PPh3)] (Au C 2.045(6), Au P
2.296(2) ꢃ, P-Au-C 175.5(2)8).[56] Notice that the binding of
the [AuPh]+ fragment does not significantly alter the struc-
ture of the [Ni2LACHTUNGTRENNUNG
(dppba)]+ complex. The main difference is
the smaller twist angle (148) between the planes through the
aromatic ring of the benzoate moiety and the carboxylato
group. The orientation of the PPh2 group with respect to the
tert-butyl substituents is also different in 4, indicating that
Figure 4. Cyclic voltammogram of 2ClO4–4ClO4 (in CH2Cl2) and [Ni2L-
AHCTUNGTRENNUNG
,
AHCTUNGTRENNUNG
ꢀ
the Ph2P group can freely rotate around the Ph2P C6H4CO2
bond. The cleft-like structure of the complex cations appears
to be of importance in the binding to the gold surface. In
fact, the cleft, which holds the ambidentate co-ligand, directs
the complexes to bind to the surface in an “upside-down”
fashion. In summary, the diphenylphosphino group in 2ClO4
is the only potential binding site for gold atoms, and the
binding does not alter the overall structure of the [Ni2L]2+
fragment.
The CV of 2ClO4 shows two redox waves, one at E11/2 = +
0.52 V (vs. a saturated calomel electrode (SCE)) with a
peak-to-peak separation DEp of 0.07 V and one at E21/2 = +
1.31 V with DEp =0.10 V. The first oxidation is a one-elec-
tron process as confirmed by controlled potential coulome-
try at 273 K. Thus, the oxidation of [Ni2LACHTNUTRGNEU(GN dppba)]ClO4 at an
applied potential of +1.0 V versus SCE in CH3CN solution
consumed n=0.95 eꢀ per complex, whereas re-reduction at
0.0 V versus SCE caused transfer of 96% of the charge
Cyclic voltammetry: Cyclic voltammetry (CV) experiments
7790
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
passed in oxidation. In addition, the CV of the electrogener-
ated [Ni2L
(dppba)]2+ dication is identical to that of the
parent [Ni2L
(dppba)]+ monocation.
The electrochemical behavior of [Ni2L
to that of the acetato-bridged complex [Ni2L
The redox chemistry of the two-coordinate AuI complexes
was investigated by other research groups. AuI complexes
with mixed Clꢀ and PPh3 donor ligands generally undergo
an irreversible reduction to Au0 at cathodic potentials of
less than ꢀ2 V.[60,61] The form and position of the reduction
peaks also depend on the Au coverage of the electrode. For
the present Ni2Au complexes there are no redox waves in a
potential window ranging from ꢀ2 to +2.5 V versus SCE
that can be attributed to reduction of the dppba-AuIPh
group, which is in good agreement with the reported
trend.[60] It can be concluded that this unit is redox inactive
in this potential range. The +III oxidation level is common
in nickel thiolate chemistry, and several other mixed-valent
nickel thiolate complexes have been reported in the past
two decades.[62–66] However, to our knowledge, trinuclear
NiIINiIIIAuI species are not described nor reported in the lit-
erature yet.
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
which shows two redox waves at E11/2 =0.56 V and E2
=
1/2
1.36 V.[56] The redox waves can be assigned to metal-cen-
tered (NiIINiII!NiIINiIII) and ligand-based oxidations
2+
(RSꢀ
(thiolate)!RS (thiyl radical)) within the [Ni2L] unit
C
as represented in Equations (1) and (2).[58]
Similarly, for 3ClO4 two redox waves are observed at
E11/2 =0.59 V (DEp =0.14 V) and E21/2 =1.38 V (DEp =
0.13 V) indicative of the same redox behavior. The first oxi-
dation yields a mixed-valent NiIINiIIIAuI complex 32+ as indi-
cated in Equation (3)[59] and the second oxidation is ligand
centered. The small shift to more anodic potentials is pre-
sumably a consequence of an electron-withdrawing effect of
the AuCl unit. This decreases the electron density of the car-
boxylate function, which increases the stability of the
NiIINiII form to some extent. For 4BPh4, the same is evident,
with the E1 and E2 values of 0.59 V (DEp =0.10 V) and
Magnetic properties of 2ClO4 and 4BPh4: Static magnetiza-
tion measurements: Static magnetization measurements for
the paramagnetic NiII and NiII AuI complexes 2ClO4 and
1/2
1/2
2
2
1.4 V (DEp =0.16 V). The irreversible redox wave at 0.91 V
is attributed to an oxidation of the tetraphenylborate anion.
The dication 32+ can also be prepared by chemical oxida-
tion. Green solutions of 3ClO4 turn dark red upon addition
4BPh4 were performed with a commercial SQUID (super-
conducting quantum interference device) magnetometer
from Quantum Design in the temperature range from 2 to
300 K in magnetic fields from 0 to 5 T. Figure 5 shows the
of one equivalent of CuACHTNUGTRNEUGN(ClO4)2. The UV/Vis spectra of such
solutions reveal several strong absorption bands (lmax =782
(4945), 517 (2735), 397 (2998), 330 (9864), and 304 nm
(13560mꢀ1 cmꢀ1) attributable to ligand-to-metal (RSꢀ!NiIII)
charge-transfer transitions. The same absorption features
are observed for electrochemically generated solutions of
32+, showing that both processes produce the same species.
Interestingly, there is also
a
feature at 1590 nm
(2538mꢀ1 cmꢀ1) attributable to an intervalence transition.
Electrospray ionization mass spectrometry provided final
conformation for the identity of the dication 32+, with a
peak at m/z=1420.29 for
a
mixed-valent [NiIINiIIIL-
ACHTUNGTRENNUNG(dppba)AuCl]ACHTUNGTRENNUNG
(ClO4)+ monocation and a peak at m/z=
661.61 for the mixed-valent [NiIINiIIIL
dication.
(dppba)AuCl]2+
ACHTUNGTRENNUNG
Figure 5. Magnetic field dependence of the static magnetization of 2ClO4
~
(
) and 4BPh4 (*) measured at T=1.8 K.
field dependence of the magnetization of 2ClO4 and 4BPh4
measured at T=1.8 K. The saturation magnetization is simi-
lar for the two complexes and is around 4.2 mB. This value
corresponds to a total spin Stot =2 in the ground state and in-
dicates a ferromagnetic coupling between Ni2+ ions (3d8,
SNi =1) in both complexes.
Figure 6 shows the temperature dependence of the mag-
netic susceptibility c(T)=M(T)/B and the inverse suscepti-
bility cꢀ1(T) of both complexes in an applied magnetic field
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7791

B. Kersting et al.
of the magnetic anisotropy parameter D were taken from
the ESR measurements (see below). We have found, howev-
er, that taking into account the D parameter does not
change the result of the fit of the susceptibility data, which
is most likely related to the smallness of the D value.[68] The
modeling reveals a ferromagnetic coupling of 23 (16 cmꢀ1)
and 26 K (18 cmꢀ1) between the two Ni ions for 2ClO4 and
for 4BPh4, respectively. These values agree well with those
of other carboxylato-bridged NiII complexes bridged by the
2
hexaaza-dithiophenolate macrocycle.[33]
High-field electron spin resonance (HF-ESR) measurements:
High-field ESR (HF-ESR) measurements were performed
on 2ClO4 and 4BPh4 with a homemade spectrometer on the
basis of a Millimeterwave Vector Network Analyzer from
AB Milimetre.[69] The samples were measured as oriented
powders in magnetic fields up to 15 T for excitation frequen-
cies n=80–350 GHz. The typical ESR spectra of 2ClO4 and
4BPh4 at different temperatures and
a frequency of
332 GHz are plotted in Figure 7. The spectra of both com-
Figure 6. Temperature dependence of the static magnetic susceptibility,
c=M/B and the inverse susceptibility cꢀ1 of a) 2ClO4 and b) 4BPh4 meas-
ured at B=1 T. Black lines correspond to the result of the fitting using
*
the Hamiltonian [Eq. (5)] and =data.
of 1 T. The data were analyzed using the Hamiltonian in
Equation (5):
Figure 7. The temperature dependence of the ESR spectra of a) 2ClO4
and b) 4BPh4 at n=332 GHz.
ꢀ
ꢁ
ð5Þ
2
2
X
X
1
3
Di S2z;i ꢀ SiðSi þ 1Þ
~
~
~
~
H ¼ ꢀ2JS1 ꢁ S2 þ mB
gSi ꢁ B þ
i¼1
i¼1
pounds look very similar. At T=4 K we observe a single
resonance line in the ESR spectrum, however, at higher T
the number of lines increases. Above T=20 K we observe
four resonance lines corresponding to the transitions within
the ground state multiplet S=2. Such a transfer of the spec-
tral weight to lower magnetic field with decreasing the tem-
perature is the first indication for an axial magnetic aniso-
tropy of the molecules (D<0).
Figure 8 shows the frequency versus resonance field de-
pendencies for 2ClO4 and 4BPh4 together with representa-
tive ESR spectra. The slopes of the four resonance branches
in both cases reveal a g-factor of 2.17. By the extrapolation
of the first resonance branch (line 1) to the zero field, the
magnetic anisotropy of both compounds was estimated. We
in which the first term describes the isotropic magnetic cou-
pling between the two Ni ions. Here, J is the Heisenberg ex-
change constant and S1 and S2 are the spin operators. The
second term describes the Zeeman interaction with the mag-
netic field B and the last term describes the single ion aniso-
tropy of the Ni ions. Here, g denotes the isotropic g-factor
of the ion, mB is the Bohr magneton, B is the external mag-
netic field and Di is the axial single-ion anisotropy constant.
Based on those parameters, the susceptibility c(T) and the
inverse susceptibility cꢀ1(T) were simulated by using the
julX simulation program.[67] The values of the g-factor and
7792
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
anisotropy axis to the applied magnetic field. The experi-
mental and modeled ESR spectra are plotted in Figure 9
with the calculated energy-level diagram. As can be seen,
Figure 8. The frequency dependence of the ESR spectrum of a) 2ClO4 at
T=20 K (D=40 GHz) and b) 4BPh4 at T=25 K (D=35 GHz; g=2.17
~
!
&
*
for both). =Line 1; =line 2; =line 3; and =line 4.
Figure 9. Experimental and modeled ESR spectra at n=332 GHz, T=
20 K and corresponding energy-level diagrams for a) 2ClO4 and
b) 4BPh4.
find anisotropy gaps of 40 and 35 GHz for 2ClO4 and
4BPh4, respectively.
For the analysis of the ESR spectra we introduced a
model that describes only the ground state of the molecule.
The effective spin Hamiltonian in this case can be written as
Equation (6):
the simulation reproduces the experimental ESR spectra
very well. The modeling confirms the negative sign of the
axial magnetic anisotropy parameter (D<0) and the D
values of ꢀ13.3 GHz (ꢀ0.063 K) and ꢀ11.6 GHz (ꢀ0.056 K)
for 2ClO4 and 4BPh4, respectively. Overall, the magnetiza-
tion and ESR measurements show no significant difference
between compounds 2ClO4 and 4BPh4. The magnetic prop-
erties of 2 are not significantly changed upon attachment of
the Au to the phosphorus atom. On the other hand, it is
also clear that the magnetic anisotropy barrier is too small
for getting sufficient retention of magnetization at finite
temperature.
ꢀ
ꢁ
1
3
2
~
~
H ¼ gmBS ꢁ B þ D Sz ꢀ SðS þ 1Þ
ð6Þ
in which the first term describes the Zeeman interaction of
the total spin of the molecule with the magnetic field B and
the second term describes the magnetic anisotropy of the
molecule. We assume that each molecule has a single total
spin S=2, a g-factor of 2.17, and a magnetic anisotropy pa-
Chemisorption of 2ClO4 on gold: To test whether 2ClO4 can
be immobilized on surfaces, it was brought in contact with a
planar gold substrate. The deposition of the [Ni2L-
rameter D=D/
(ꢀ0.063 K) for 2ClO4 and ꢀ11.6 GHz (ꢀ0.056 K) for 4BPh4.
G
E
The simulation of the ESR spectra was performed for n=
332 GHz, T=20 K and parallel orientation of the magnetic
ACHTUNGTRENNUNG
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7793

B. Kersting et al.
thiol monolayers on gold as described in the literature.[70]
A
thin film of the more lipophilic tetraphenylborate salt 2BPh4
on gold increases the contact angle by 5.28 (Table 1,
clean gold-coated Si wafer was immersed in a 1ꢄ10ꢀ3 m solu-
tion of 2ClO4 in EtOH for 24 h followed by washing with
EtOH and drying. To analyze the so modified Au surfaces
we used four complementary surface sensitive analytical
techniques (contact angle measurements, atomic-force mi-
croscopy (AFM), X-ray photoelectron spectroscopy (XPS),
and spectroscopic ellipsometry).[71]
ꢀ
entry 4), which could be expected since the BPh4 group is
ꢀ
more hydrophobic than the ClO4 ion. The contact angle
measurements also revealed that complexes such as [Ni2L-
ACHUTNGRENUN(G O2CPh)]BPh4 and [Ni2LCAHTUNGTRENN(UGN OAc)]ClO4, which lack end groups
for surface fixation, are not chemisorbed on the gold surfa-
ces (Table 1, entries 6 and 7). It is also interesting to note
that the physisorbed anions of the immobilized salts can be
exchanged for one another by surface metathesis reactions.
The chemisorbed cationic complexes remain attached on the
Contact angles and atomic-force microscopy: The immobili-
zation of 2ClO4 to gold was expected to change the wettabil-
ity and roughness of the surface. Therefore, the immobiliza-
tion of the complexes was probed by static water contact
angle measurements and AFM topography analysis. Table 1
lists the results.
surfaces. Thus [Ni2L
NaBPh4 undergoes an anion exchange reaction to produce
[Ni2L(dppba)]BPh4 (Table 1, entry 5). Such anion exchange
ACHTUNGTERNN(UNG dppba)]ClO4 on gold when treated with
AHCTUNGTRENNUNG
reactions were observed by others, for example, for polyca-
tions attached to solid surfaces.[78]
The trend observed with the contact angle measurements
is also reflected in the values of the root mean squared
(rms) roughness determined by using AFM. Figure 10 shows
Table 1. Water contact angles and rms roughness data obtained for gold
films modified with various dinickel(II) complexes.
Entry Compound
Contact
rms roughness
angle [8][a] [nm]
1
2
3
bare gold (EtOH)
PPh2(C6H4-4-CO2H)
[Ni2L(dppba)]ClO4
(2ClO4)
[Ni2L(dppba)]BPh4
(2BPh4)
[Ni2L(dppba)]ClO4
(2ClO4) (after washing with NaBPh4)
[Ni2L(O2CPh)]ClO4
[Ni2L(OAc)]BPh4
75.8 (1.5) 0.6 (1)
74.9 (1.4) n.d.[b]
71.5 (1.6) 1.7 (5)
G
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
4
5
R
A
76.7 (1.9) 1.6 (2)
ACHTUNGTRENNUNG
N
N
76.5 (2.1)
n.d.[b]
ACHTUNGTRENNUNG
6
7
T
G
75.9 (2.0) 0.6
75.8 (1.5) 0.6
A
ACHTUNGTRENNUNG
[a] The values represent the average of ten 4 mL drops of destilled, deion-
ized water. Standard deviations are given in parentheses. The “bare”
gold surfaces were identically treated to the modified surfaces except
with omission of any adsorbate in the solvent. [b] n.d.=not determined.
The contact angle measurements showed small but signifi-
cant variations, indicating that the complexes bind indeed to
the gold surface. For the unmodified gold-containing adven-
titious, nonpolar material a contact angle of 75.88 was ob-
tained, which agrees with the value reported in the litera-
ture.[72] Very little is known about water contact angle of ter-
tiary phosphines on the coinage metals.[35,36,73,74] The contact
angle of the free Hdppba ligand has not been reported pre-
viously.[75] The value of 74.98 is relatively high for a carboxy-
late terminated monolayer, suggesting that the CO2H
groups are involved in intermolecular H-bonding interac-
tions. Indeed, such an effect is known to increase the hydro-
phobicity of rather polar molecules.[76] Upon immobilization
Figure 10. AFM topography characteristics considering a 1ꢄ1 mm2 area,
a) Au substrate as grown with a roughness of 0.6(1) nm (rms), b) Au sub-
strate after deposition of [Ni2LACHTGNUTRNE(NUG dppba)]ClO4 with a roughness of
1.7(5) nm (rms), c) AFM profile following the white-dotted lines in (a)
and (b).
a comparison between the topography of [Ni2LACTHUNTRGNEU(GN dppba)]ClO4
of the [Ni2L
G
(2ClO4) and of the uncovered Au substrate along with the
duces slightly relative to the unmodified gold, indicating a
surface transition to a more hydrophilic state. This is not
surprising given the salt-like character of the adsorbates.
The contact angle for immobilized 2ClO4 on gold (Table 1,
entry 3) is comparable with those of alkane thiol monolayers
terminated with ester, alkoxy, or thioester groups.[77] The
angle seems to be quite high for an ionic compound, but
given that the charges are well-shielded by the apolar
groups of the macrocyclic ligands, this is not surprising. A
comparison of two line profiles. As can be seen, the rms
roughness of [Ni2L
than that of the pure gold substrate. In addition, the rms
roughnesses of the [Ni2L(O2CMe)]ClO4 and [Ni2L-
(O2CPh)]ClO4 compounds (Table 1, entries 6 and 7) had
similar rms values as the substrate.
ACHTUNGTERNN(UNG dppba)]ClO4 on gold is by 10 ꢃ larger
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
XPS studies: The attachment of [Ni2LACHTNUTRGNE(NUG dppba)]ClO4 (2ClO4)
to the gold surface was further analyzed by X-ray photoelec-
7794
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
tron spectroscopy (XPS).[69] XPS spectra were recorded for
bulk 2ClO4 and for a freshly prepared sample of 2ClO4
chemisorbed on gold. Photoelectron peaks of Ni, P, S, Cl, C,
and O were observed. However, the long exposure time led
to significant X-ray-induced damage, which particularly af-
ꢀ
fected the ClO4 groups, the bridging thiolate sulfur atoms
of the N6S2 macrocycle, and the P atoms as further detailed
below.[79] Table 2 lists the selected peak positions and their
assignments.
Table 2. XPS binding energies (eV) for 2ClO4 and for 2ClO4 chemisor-
bed on gold.[a]
Element
Bulk
2ClO4
2ClO4 on
Au
(111)[c]
Assignment
R3P
[b]
ACHTUNGTRENNUNG
P
P
N
)
)
132.2
133.2
33
131.1
132.1
39
Figure 11. XPS P
CAHTUTNGERNN(GU 2p3/2) and PACHTUNGTERN(NGUN 2p1/2) peaks for bulk 2ClO4 and for a mon-
Relative intensity [%]
olayer of 2ClO4 on gold.
P
P
U
)
)
134.1
135.1
49
139.9
140.9
18
856.5
861.7
865.6
873.4
878.2
882.7
163.7
164.7
100
133.4
134.4
61
R3PO, (or P(OR)3)
Relative intensity [%]
P
P
Relative intensity [%]
Ni
3ꢀ
)
)
n.o.[d]
n.o.[d]
PO4
PACHTUNGTRENNUG
(2p3/2
)
856.8
862.1
866.0
874.0
878.6
883.0
162.3
163.3
50
169.7
170.7
50
209
211
traces
199.5
201.5
100
399
533
284.9
286.1
288.2
Ni2+
“satellite”
“satellite”
Ni2+
“satellite”
“satellite”
AHCTUNGTRENNUNG
to the Au surface. The peak at 133.4 is attributed to X-ray
induced-damage. Notice that this value is different from that
of bulk 2ClO4. This suggests the formation of a different
oxygenated species, perhaps a phosphite group (P(OR)3).
Ni
ACHTUNGTRENNUNG(2p1/2) (main peak)
ꢀ
ꢀ
S
S
U
)
)
Ni mSR Ni
ꢀ
The incorporation of O atoms into the P C bonds (rather
ꢀ
than the Au P) would be consistent with the presence of
Total intensity [%]
S
S
(2p3/2
(2p1/2
)
)
169.8[b]
170.8[b]
traces
209
211
60
200
202
40
403
Ni mSO3R Ni
ꢀ
ꢀ
ꢀ
the Au P bonds. This finding is also in excellent agreement
ACHTUNGTRENNUNG
with the previous knowledge that the P donor in the [Ni2L-
AHCTUNGTRENNUNG
(dppba)AuPh]+ complex is much more resistant towards ox-
Total intensity [%]
Cl
Cl
Total intensity [%]
Cl
Cl
Total intensity [%]
N(1s)
O(1s)
ꢀ
G
)
)
ClO4
idation than in the parent [Ni2L
free -PPh2 group (see above). Notice that binding of the
phosphine to the gold surface shifts the P(2p1/2) peaks only
slightly. This specific behavior has been observed by others
for R3P ligands.[84]
(dppba)]+ complex with the
ACHTUNGTRENNUNG
E
N
)
)
Clꢀ
AHCTUNGTRENNUNG
G
NR3
The photoelectron peaks in the NiACHTUNGRTNE(UNG 2p3/2) region for bulk
533.5
287
RCO2, RSO3
CH (aliph.+arom.)
2ClO4 and for chemisorbed 2ClO4 on gold with the well-
known satellites at 862.5 and 863 eV are characteristic of di-
valent, six-coordinate nickel complexes (Figure 12). These
data imply that complex 2ClO4 undergoes no redox changes
upon surface fixation.[85–88] However, the Cl(2p) regions
C(1s)
ꢀ
ꢀ
C S, C P
RCO2
ꢀ
[a] The binding energies are referenced to Au 4f7/2 (84.0 eV). [b] Data ac-
quisition time: 58 h (165 cycles). [c] Data acquisition time: 44 h (106
cycles), [d] n.o.=not observed.
ꢀ
show, besides the presence of ClO4 ions (209 eV), addition-
al spectral features corresponding to chloride ions (see the
Supporting Information). The observed behavior is the
ꢀ
Of particular evidence for surface Au P single bonds are
the oxygenation reactions that 2ClO4 undergoes in the bulk
ꢀ
result of photoinduced decomposition of ClO4 ion into Clꢀ
and on the gold surface. These oxidations are induced by O
and was reported previously.[76] A similar photoinduced con-
ꢀ
ꢀ
atoms, which stem from the decomposition of the ClO4
version of the Ni-bound RSꢀ to RSO3 moieties can be ob-
ions by the incident X-ray beams.[80] The XPS spectrum of
served for chemisorbed 2ClO4. The O atoms most likely
ꢀ
bulk 2ClO4 shows two P(2p3/2) and P
U
stem again from the decomposition of the ClO4 ions. It is
binding energies typical for
a
interesting that these O atoms oxidize the P atoms in bulk
(Figure 11).[81,82] The feature at higher energies (134.1 eV) is
assumed to be due to a phosphine oxide (R3PO), which
arises from X-ray-induced damage. The oxidation of physi-
sorbed PR3 ligands to R3PO ligands was observed previous-
2ClO4 and the RSꢀ groups in chemisorbed 2ClO4. This can
ꢀ
be taken as additional evidence for the formation of an Au
P bond between molecules of 2 and the gold surface. As
mentioned before, the coordination of a PR3 ligand to a
gold atom increases its stability towards oxidation, and the
ly.[83] For chemisorbed 2ClO4 on gold, again two P
ACHTUNGRTNE(NUG 2p3/2) and
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7795

B. Kersting et al.
C angle is 1058. This deviation from the expected thickness
value of one monolayer most likely arises from the relative-
ly high film roughness of the gold substrates used rather
than from an ordering by intermolecular hydrogen-bonding
interactions between the adsorbate molecules. The strong
dependence of film thicknesses on the film roughness was
also observed by others.[94]
Nevertheless, the observed film thicknesses observed for
the perchlorate and tetraphenylborate salts are in good
agreement with the molecule size of ꢃ15 ꢃ for the
[Ni2L(O2CC6H4PPh2)]+ cation in 2ClO4 and 2BPh4 (on the
basis of the corresponding atomic distances in the crystal
structure of 2ClO4). Interestingly, the thicknesses of the
layers of the perchlorate and tetraphenylborate salts are
ꢀ
ꢀ
almost identical suggesting that the ClO4 and BPh4 ions
are located beneath the [Ni2L
AHCTUNGTRENNUNG
Figure 12. XPS NiACHTUNGTRENNUNG(2p3/2) peaks for bulk 2ClO4 and for a monolayer of
2ClO4 on gold.
than above them in a second layer. Therefore, the ellipso-
metric data further support the assumption that both 2ClO4
and 2BPh4 form monolayers rather than multilayer films,
same applies here as well. The absence of signals for free
phosphine groups also rules out the presence of overlayers.
Photoelectron peaks of C, O, and Cl are also observed
with the cationic [Ni2LACTHNGUTERNNUG
(dppba)]+ molecules oriented per-
pendicular to the Au surface.
(the Supporting Information). The C
G
Surface-binding model: If it can be assumed that the ob-
ꢀ
deconvoluted to three signals. The strong C(1s) signal at
284.9 eV can be attributed to the aliphatic and aromatic
(CH) carbon atoms of the dppba ligand and the N6S2 macro-
cycle. The shoulder at 286.1 eV is attributed to the five elec-
tron-deficient carbon atoms bonded to electronegative
sulfur and phosphorus atoms.[89] There is also a weak peak
at 288.2 eV which can be attributed to the carboxylate func-
tion.[90] Taken together, the XPS spectrum provides strong
evidence for a covalent Au-P linkage between 2 and the Au
surface in a monolayer.
served Au P bonding interactions in 4BPh4 are maintained
in chemisorbed [Ni2L
A
a hypothetical surface structure. Figure 13 shows a likely ori-
Optical thicknesses:To probe the thickness of the films spec-
troscopic ellipsometry measurements were carried out for a
monolayer of complex 2ClO4 on gold, the tetraphenylborate
salt 2BPh4, and the free phosphine Hdppba. The ellipsome-
try data were analyzed by using the Cauchy formula
[Eq. (7)] for thin monolayers,[91] in which three parameters
define the wavelength-dependent refractive index.
Figure 13. Proposed binding mode of 2ClO4 to gold (left: van der Waals
representation of the cation 2; right: representation with a Lewis formu-
la. The macrocycle is shown as an ellipse encircling the two Ni2+ ions for
clarity).
entation of 2ClO4 on the gold surface. The height of the [Ni2
LACTHNUTRGNEUNG
(dppba)]+ complex on the surface in this case is given by
the distance between two opposing NMe substituents, which
amounts to approximately 12.3 ꢃ. In an alternate orienta-
tion (not shown in Figure 12) the [Ni2L]2+ fragment is rotat-
B
C
ð7Þ
nðlÞ ¼ A þ
þ
l2 l4
The parameter A strongly correlates with the film thick-
ness d, which means that fitting both A and d at the same
time is not reasonable. Therefore A was kept constant at the
value of 1.45, which is commonly used for the refractive
index of monolayers described as transparent media.[92,93] By
using this model, reasonable fits were produced resulting in
average thicknesses of (4.8ꢂ4) ꢃ for Hdppba, (16.3ꢂ7) ꢃ
for 2ClO4, and (15.1ꢂ8) ꢃ for 2BPh4. The observed layer
thickness of the Hdppba ligand is somewhat smaller than
the calculated molecule size of approximately 7 ꢃ assuming
that the molecules are oriented with their pseudo-threefold
axis perpendicular to the surface and that the average C-P-
ꢀ
ed by 908 around the O2CC6H4 PPh2 bond. In this case, the
height is given by two opposing tert-butyl CH3 groups, which
amounts to 15.5 ꢃ (H32a···H37b).[95] This value is in reason-
able agreement with the thickness determined by ellipsome-
try. Irrespective of the actual orientation of the [LNi2]2+
fragment, the surface complex presumably is also stabilized
by van der Waals interactions between the Au surface and
the methylene or methyl groups of the supporting macrocy-
cle. The thiolate sulfur atoms, however, are out of range of
bonding interactions with the Au surface.
7796
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
plexes are hazardous and may explode. Only small quantities should be
prepared and great care should be taken.
Conclusion
A
ACHUTGTNRNE(NGU dppba)]ClO4 ACHTUNGTRENNUNG
The main findings of the present work can be summarized
a
ACHTUNGTRENNUNG
as
follows:
1) The
dinuclear
NiII
complex
0.108 mmol) in MeOH (25 mL). Triethylamine (16.2 mg, 0.16 mmol) was
then added and the ocher-colored suspension turned into a clear-green
solution, which was stirred for 48 h at RT. A solution of LiClO4·3H2O
(87.1 mg. 0.54 mmol) in EtOH (20 mL) was added, and stirring was con-
tinued for 30 min. The mixture was concentrated (rotary evaporator) to
ꢃ25 mL. The green, microcrystalline precipitate was filtered, washed
with EtOH and Et2O. Yield: 111 mg (86%). M.p. 320–3248C (decomp.);
IR (KBr pellet): n˜ =3443 (w), 2961 (m), 2867 (m), 1599 (s), 1550 (m, nas
(RCO2ꢀ)), 1461 (s), 1408 (s, ns (RCO2ꢀ)), 1363 (m), 1309 (w), 1265 (w),
1233 (w), 1201 (w), 1152 (w), 1092 (vs, ClO4ꢀ), 1040 (m), 931 (w), 913
(m), 881 (w), 825 (m), 817 (m), 768 (m), 746 (w), 719 (w), 699 (m), 623
(m, ClO4ꢀ), 495 cmꢀ1 (w); UV/Vis (CH2Cl2): lmax (e)=227 (48400), 272
(15000), 334 (11800), 385 (2100), 650 (34), 1122 nm (68 mꢀ1 cmꢀ1); MS
(ESI): m/z (CH3CN)=1089.4 [MꢀClO4]+; elemental analysis calcd (%)
for C57H78ClN6Ni2O6PS2 (1191.21): C 57.47, H 6.60, N 7.06, S 5.38; found:
C 57.25, H 6.39, N 6.97, S 5.27. The tetraphenylborate salt 2BPh4 was
prepared by adding a solution of NaBPh4 (14.4 mg, 42.0 mmol) in EtOH
(10 mL) to a solution of 2ClO4 (10.0 mg, 8.39 mmol) in CH2Cl2 (10 mL).
Yield: 11.5 mg (97%). IR (KBr pellet): n˜ =3424 (vw), 3052 (m), 3033
(m), 2961 (m), 2925 (m), 2839 (m), 1589 (s), 1548 (w), 1479 (m), 1458 (s),
1420 (vs), 1394 (m), 1363 (m), 1308 (w), 1264 (m), 1231 (w), 1199 (vw),
1181 (vw), 1150 (w), 1133 (w), 1075 (m), 1058 (m), 1039 (m), 1001 (w),
929 (w), 912 (m), 879 (m), 844 (m), 823 (m), 777 (vw), 748 (m), 733 (m),
703 (vs), 629 (w), 611 (m), 562 (w), 536 (w), 486 cmꢀ1 (m).
[Ni2L(O2CC6H4PPh2)]+ supported by a macrocyclic hexaza-
dithiophenolate ligand is readily prepared and can be isolat-
ed as the air-sensitive perchlorate [Ni2L
ACHTUNGTRENN(UNG dppba)]ClO4
(2ClO4) or tetraphenylborate [Ni2L(dppba)]BPh4 (2BPh4)
AHCTUNGTRENNUNG
salt. 2) The X-ray structure determination showed that the
ambidentate 4-(diphenylphosphino)benzoate co-ligand binds
specifically, through its carboxylate function, to the [Ni2L]2+
fragment such that the soft phosphorus atom is positioned
in the periphery of the complex. 3) The mode of auration of
2ClO4 with AuCl demonstrates that 2ClO4 exhibits only one
gold binding site. Only the phosphine group in 2ClO4 is aur-
ated. The two thiolate sulfur atoms in 2ClO4 exhibit no au-
rophilicity. 4) The [NiII L(dppba)]+ complexes exhibit an
ACHTUNGTERNUNNG
2
S=2 ground state that is attained by ferromagnetic coupling
of the two NiII (Si =1) ions. HF-ESR measurements yield a
negative axial magnetic anisotropy (D<0), which implies a
bistable (easy axis) magnetic ground state. 5) The [Ni2L-
ACHTUNGTRENNUNG
(dppba)]+ complex is redox-active and is also accessible in a
mixed-valent NiIINiIII form of a different S=3/2 ground spin
state. 6) Complexes of 2ClO4 can be chemisorbed on gold
A
ACHUTGTNRENG(NU dppba)AuCl]ClO4 ACHTUNGTRNE(NUGN 3ClO4): A green-colored solution of [Ni2L-
ꢀ
surfaces through the formation of Au P surface bonds as
AHCTUNGTRENNUNG
clearly established by contact angle measurements, AFM
spectroscopy, XPS spectroscopy, and spectroscopic ellipsom-
etry. Finally, 7) the complexes do not change their oxidation
state upon surface fixation. It would be interesting to see
whether the oxidation state of 2ClO4 (and hence the spin
state) can be switched by the tip of a scanning tunneling mi-
croscope. Such studies are underway.
a dropping funnel over a 30 min period to an ocher-colored suspension of
AuCl (19.5 mg, 80.0 mmol) in CH2Cl2 (15 mL). The resulting green solu-
tion was stirred for an additional 24 h at ambient temperature, filtered,
and the filtrate combined with a solution of LiClO4·3H2O (67.3 mg,
0.42 mmol) in EtOH. After stirring for a further 30 min, the CH2Cl2 was
removed by rotary evaporation to give the product as a green solid,
which was filtered and washed with cold EtOH and Et2O. Yield: 108 mg
(91% based on [Ni2LACTHNUTRGNEUNG(dppba)]ClO4). M.p. 268–2708C (decomposes with-
out melting.); IR (KBr pellet): n˜ = 3447 (w), 2953 (m), 2864 (m), 1602
(s), 1553 (m, nas (RCO2ꢀ)), 1461 (s), 1438 (m, ns (RCO2ꢀ)), 1398 (s), 1363
(m), 1308 (w), 1265 (w), 1233 (w), 1200 (vw), 1153 (w), 1097 (vs, ClO4ꢀ),
1039 (m), 999 (w), 930 (w), 912 (w), 881 (w), 825 (m), 768 (w), 752 (w),
726 (m), 694 (m), 623 (m, ClO4ꢀ), 564 (w), 534 (w), 490 (w), 415 cmꢀ1
(w); UV/Vis (CH2Cl2): lmax (e)=231 (47700), 304 (13400), 334 (9900),
381 (1900), 652 (35), 1119 nm (70 mꢀ1 cmꢀ1); MS (ESI): m/z (CH3CN)=
Experimental Section
Materials and methods: All reagents were purchased from commercial
sources unless otherwise specified. Compound [Ni2L(Cl)]ClO4 (1ClO4)
was prepared as described in the literature.[45] 4-(diphenylphosphane)-
benzoic acid was prepared by using a fluoride-catalyzed reaction from 4-
fluoroethylbenzoate and diphenyl(trimethylsilyl)phosphine.[96] Ethanol
was deoxygenated with nitrogen before use. Acetonitrile was distilled
from calcium hydride. The syntheses of the metal complexes were carried
out under a protective atmosphere of argon. Melting points were deter-
mined in open glass capillaries and are uncorrected. The infrared spectra
were recorded as KBr discs by using a Bruker Tensor 27 FT-IR spectro-
photometer. Electronic absorption spectra were recorded on a Jasco V-
670 UV/Vis/near-IR spectrophotometer. Elemental analysis was carried
out on a VARIO EL–elemental analyzer. ESI mass spectra were record-
ed on a Bruker Daltronics ESQUIRE3000 PLUS spectrometer. Cyclic
voltammetry measurements were carried out at 258C with an EG&G
Princeton Applied Research potentiostat/galvanostat model 263 A. Co-
baltocenium hexafluorophosphate (Cp2CoPF6) was used as an internal
standard. All potentials are given versus SCE but are standardized
against the ferrocenium/ferrocene couple.[97] Temperature-dependent
magnetic susceptibility measurements on powdered solid samples were
carried out by using a MPMS 7XL SQUID magnetometer (Quantum
Design) over a temperature range 2–330 K at an applied magnetic field
of 0.5 Tesla. The observed susceptibility data were corrected for underly-
ing diamagnetism. CAUTION! Perchlorate salts of transition-metal com-
1321.4
[MꢀClO4]+;
elemental
analysis
calcd
(%)
for
C57H78AuCl2N6Ni2O6PS2 (1423.63): C 48.09; H 5.52, N 5.90, S 4.50; found:
C 48.09, H 5.64, N 5.59, S 4.63.
A
N
(4BPh4):
A
solution of NaBPh4 (120 mg,
solution of [Ni2L-
a
AHCTUNGTRENNUNG
was stirred for 1 h, when CH2Cl2 was removed under reduced pressure.
The resulting green solid was isolated by filtration, washed with cold
EtOH and Et2O. Yield: 93% (based on [Ni2LACTHNUTRGNEUNG(dppba)AuCl]ClO4). The
crude product was purified by recrystallization from a mixed CH2Cl2/
CHCl3/n-hexanes solvent system. M.p. 263–2658C (decomposes without
melting.); IR (KBr pellet): n˜ =3444 (w), 3052 (m), 3049 (w), 2962 (m),
2857 (m), 1600 (s), 1552 (m, nas (RCO2ꢀ)), 1460 (vs), 1437 (m, ns
(RCO2ꢀ)), 1395 (s), 1263 (m), 1307 (w), 1265 (w), 1232 (w), 1151 (w),
1099 (w), 1076 (m), 1057 (m), 1041 (m), 999 (w), 930 (w), 912 (w), 881
(w), 825 (m), 768 (w), 728 (s), 703 (vs, BPh4ꢀ), 631 (m), 612 (m),
525 cmꢀ1 (m); UV/Vis (CH2Cl2): lmax (e)=229 (68100), 306 (11300), 334
(8000), 387 (1300), 651 (33), 1119 nm (68 mꢀ1 cmꢀ1); MS (ESI): m/z
(CH3CN)=1363.3 [MꢀBPh4]+; elemental analysis calcd (%) for
C87H103AuBN6Ni2O2PS2 (1685.06): C 62.01; H 6.16; N 4.99; S 3.81; found:
C 61.95; H 6.08; N 4.98; S 3.82.
Preparation of gold substrates: Substrates were prepared by magnetron
sputtering of 100 nm Au onto polished silicon wafers SiACHTUNTRGNEU(GN 111), pre-coated
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7797

B. Kersting et al.
with a 20 nm thick adhesion layer of titanium. The root mean square
roughness of the gold substrates was ꢄ0.6 nm (by using AFM). The sub-
strates were cut into pieces (1 cmꢄ1 cm), cleaned by ultrasonic treatment
in piranha solution (H2SO4/H2O2, 3:1; CAUTION! piranha solution
reacts violently with organic materials),[19,98] followed by washing with di-
chloromethane, ethanol, and deionized water.
using the 1.2-fold Uiso value of the parent atom except methyl protons,
which were assigned the 1.5-fold Uiso value of the parent C atoms. Unless
otherwise noted, all non-hydrogen atoms were refined anisotropically.
ORTEP-3 was used for the artwork of the structures.[105]
In the crystal structure of 4BPh4·1.5CH2Cl2·0.5CHCl3·0.5H2O, one tBu
group of one of the two crystallographically independent molecules was
found to be disordered over two positions. A split atom model with re-
Contact angles: Surface hydrophobicity was examined by performing
water contact angle measurements (Sessile Drop Method) with a DSA II
(Krꢀss, Hamburg, Germany) contact angle analyzer. The contact angle
measurements were collected by using a 4 mL drop size of deionized, dis-
tilled water. At least 10 contact angles per five different locations were
averaged. The spectra were analyzed according to Owens-Wendt-Rabel-
Kaelble.
ꢀ
strained C C and C···C distances was applied by using SADI instructions
implemented in SHELXTL. The site occupancies of the respective orien-
tations were refined as follows: C
(34b)=0.60:0.40. The C and Cl atoms of one CH2Cl2 solvate molecule
were refined isotropically.
CCDC-910079 ([Ni2L(dppba)]BPh4) and CCDC-910080 ([Ni2L-
AHCTUNGRTENNG(UN dppba)AuPh)]BPh4·1.5CH2Cl2·0.5CHCl3·0.5H2O) contain the supple-
ACHTUNGTRNENU(G 32a)CACUTHNTGRENNU(GN 33a)CACHTUNRTGENN(GUN 34a)/CAHCTUNGTREN(NNUG 32b)CACHTUNGTRENNUNG(33b)C-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
Atomic-force microscopy: Topography data were acquired with an Agi-
lent 5600 LS AFM system, under Argon (Ar)-controlled environmental
conditions to maintain the integrity of the organic system. Measurements
were performed in tapping mode to minimize the contact between the
AFM probe and the sample surface and avoid damage or modification of
the topographic characteristics. Special ultra-sharp (4–10 nm tip radius)
Olympus cantilevers were employed allowing high sensitivity measure-
ments. Data shown in the respective Figure 10 correspond to an area of
1ꢄ1 mm2, although a mapping of the topographic characteristics was per-
formed on different points of the samples to verify the uniformity of the
organic system over the Au substrate.
mentary crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Crystallographic data for [Ni2LACHTNUGRTENGNU(dppba)]BPh4: C81H98BN6Ni2O2PS2, Mw =
¯
1410.97, triclinic, space group P1, a=11.033(2), b=17.813(4), c=
19.616(4) ꢃ, a=100.48(3), b=104.44(3), g=98.58(3)8, V=3593.6(12) ꢃ3,
ACTHNUTRGNEUNG
Z=2, 1calcd =1.304 gcmꢀ3, m(MoKa)=0.656 mmꢀ1, crystal size 0.30ꢄ0.25ꢄ
0.20 mm, q limits: 1.42–27.398. T=183 K. Of 29738 measured reflections
(Rint =0.094), 15471 were unique and 8558 observed (I>2s(I)). Refine-
ment converged to R1 =0.0649, wR2 =0.1507 (for observed data), R1 =
0.1236, wR2 =0.1537 for all data. Max, min peaks: 0.990, ꢀ0.879 eꢃ3.
Ellipsometry: A M-2000 T-Solar ellipsometer (J. A. Woollam Co., Inc.)
operating with a Xe lamp was employed for ellipsometric measurements
in the spectral range from 0.7 to 5 eV. The cleaned gold substrates were
immersed in a 1ꢄ10ꢀ3 m solution of the complexes in dichloromethane
for at least 12 h, after which they were rinsed with absolute ethanol,
blown dry with ultrahigh purity nitrogen, and immediately measured
under ambient conditions. The reported thickness values represent the
average of at least nine measurements at different locations on the sur-
face of the sample at three different incident angles (65, 70, and 758).
Every sample series evaluated by spectroscopic ellipsometry consists of 3
to 5 samples of the same molecular complex produced under the same
experimental conditions. The error bars represent the standard deviation
from the mean thickness value within one sample series. The scatter ob-
served in the data was typically ꢂ0.6 nm, arising most likely from a film
roughness of the gold substrates of about 0.8 nm (measured by AFM).
Crystallographic
data
for
[Ni2LACHTUNGRTENNU(G dppba)AuPh)]ACHTUNGTRNEBNUNG Ph4·1.5CH2Cl2
·0.5CHCl3·0.5H2O: C89H107.50AuBCl4.50N6Ni2O2.50PS2, Mw =1881.12, mono-
clinic, space group P21/n, a=14.159(3), b=37.431(8), c=32.374(7) ꢃ, b=
93.17(3)8, V=17131(6) ꢃ3, Z=8, 1calcd =1.459 gcmꢀ3
,
mACHTUNGTRENNUNG(MoKa)=
2.403 mmꢀ1, crystal size 0.25ꢄ0.20ꢄ0.20 mm, q limits: 1.42–27.398. T=
181(2) K. Of 166354 measured reflections (Rint =0.11), 43106 were unique
and 24877 observed (I>2s(I)). Refinement converged to R1 =0.0522,
wR2 =0.1283 (for observed data), R1 =0.0993, wR2 =0.1428 for all data.
Max, min peaks: 1.937, ꢀ1.878 eꢃ3.
Acknowledgements
The data analysis was performed with the CompleteEASE software from
the J. A. Woollam Company.[99] The model used for the thickness deter-
mination consists of two layers, whereby the first one is the gold substrate
and the second one is a Cauchy layer. As the theory describing this layer
is applicable only for non-absorbing films, one usually has to restrict the
evaluated energy range to an appropriate spectral region. Because of the
small layer thickness the absorption was small enough to use the data of
the entire energy range from 0.7 to 5.0 eV.
This work was supported by the Deutsche Forschungsgemeinschaft DFG
FOR 1154 “Towards molecular spintronics” and the University of Leip-
zig. J. Lach thanks the Sꢁchsische Aufbaubank (SAB) for an ESF grant
(Europꢁischer Sozialfonds).
[1] For reviews on the organization of a magnetically bistable molecule
on surfaces, see: a) A. Cornia, M. Mannini, P. Sainctavit, R. Sessoli,
Chem. Soc. Rev. 2011, 40, 3076–3091; b) J. Gꢅmez-Segura, J. Veci-
ana, D. Ruiz-Molina, Chem. Commun. 2007, 3699–3707; c) M.
Cavallini, M. Facchini, C. Albonetti, F. Biscarini, Phys. Chem.
Chem. Phys. 2008, 10, 784–793.
[2] For reviews for integration of electronically bistable molecule or
molecular switches in thin film devices, see: a) K. Szaciłowski,
Chem. Rev. 2008, 108, 3481–3548; b) V. A. Dediu, L. E. Hueso, I.
Bergenti, C. Taliani, Nat. Mater. 2009, 8, 707–716; c) M. Wohlge-
nannt, Phys. Status Solidi RRL 2012, 6, 229–242.
X-ray photoelectron spectroscopy: XPS spectra were obtained on a com-
mercial PHI 5600 spectrometer equipped with two light sources. A mono-
chromatized AlKa source provided photons with an energy of 1486.6 eV,
which was used for the core level studies. The total energy resolution was
350 meV. The gold foils (Au, purity 99.59%) were cleaned in a ultrasonic
bath. The substrate was immersed in a 1ꢄ10ꢀ3 m solution of 2ClO4 in di-
chloromethane for at least 12 h, washed with dichloromethane, then etha-
nol, and immediately transferred into the spectrometer. Energy referenc-
ing was based on the Au 4f7/2 peak at 84.0 eV. All samples were studied
at room temperature at a pressure of approximately 2ꢄ10ꢀ10 mbar. Spec-
tra were analyzed using CasaXPS.[100]
[3] D. Gatteschi, R. Sessoli, J. Villain, Molecular Nanomagnets, Oxford
University Press, Oxford, 2006.
Crystallography: Single crystals of 2BPh4 suitable for X-ray crystallogra-
phy were obtained by recrystallization from MeCN/EtOH. Single crystals
of 4BPh4·1.5CH2Cl2·0.5CHCl3·0.5H2O were obtained by slow evaporation
of a mixed CHCl3, CH2Cl2, hexanes (1:1:1, v/v/v) solution. The diffraction
experiments were carried out at 203 K on a STOE IPDS-2T X-ray dif-
fractometer. The intensity data were processed with the program STOE
X-AREA.[101] Structures were solved by direct methods[102] and refined
by full-matrix least-squares on the basis of all data against F2 by using
[4] F. Troiania, M. Affronte, Chem. Soc. Rev. 2011, 40, 3119–3129.
[5] J. M. Seminario, Nat. Mater. 2005, 4, 111–113.
[6] A. R. Rocha, V. M. Garcꢆa-Suꢇrez, S. W. Bailey, C. J. Lambert, J.
Ferrer, S. Sanvito, Nat. Mater. 2005, 4, 335–339.
[7] L. Bogani, W. Wernsdorfer, Nat. Mater. 2008, 7, 179–186.
[8] V. Meded, A. Bagrets, K. Fink, R. Chandrasekar, M. Ruben, F.
Evers, A. Bernand-Mantel, J. S. Seldenthuis, A. Beukman, H. S. J.
van der Zant, Phys. Rev. B 2011, 83, 245415–1–13.
[9] F. Prins, M. Monrabal-Capilla, E. A. Osorio, E. Coronado, H. S. J.
van der Zant, Adv. Mater. 2011, 23, 1545–1549.
SHELXL-97.[103] PLATON was used to search for higher symmetry.[104]
atoms were placed in calculated positions and treated isotropically by
H
7798
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
[10] S. Shi, G. Schmerber, J. Arabski, J.-B. Beaufrand, D. J. Kim, S. Bou-
kari, M. Bowen, N. T. Kemp, N. Viart, G. Rogez, E. Beaurepaire,
H. Aubriet, J. Petersen, C. Becker, D. Ruch, Appl. Phys. Lett. 2009,
95, 043303–1–3.
[18] L. Margheriti, M. Mannini, L. Sorace, L. Gorini, D. Gatteschi, A.
Caneschi, D. Chiappe, R. Moroni, F. Buatier de Mongeot, A.
Comia, F. M. Piras, A. Magnani, R. Sessoli, Small 2009, 5, 1460–
1466.
[11] For thin films of phthalocyanines, see a) D. E. Barlow, L. Scudiero,
K. W. Hipps, Langmuir 2004, 20, 4413–4421; b) C. Iacovita, M. V.
Iacovita, B. W. Heinrich, T. Brumme, J. Kortus, L. Limot, J. P.
Bucher, Phys. Rev. Lett. 2008, 101, 116602–1–4; c) C. Wꢁckerlin,
D. Chylarecka, A. Kleibert, K. Mꢀller, C. Iacovita, F. Nolting, T. A.
Jung, N. Ballav, Nat. Commun. 2010, 1, 1–7; d) M. Schmid, A.
Kaftan, H.-P. Steinrꢀck, J. M. Gottfried, Surface Science 2012, 606,
945–949; e) M. Toader, M. Knupfer, D. R. T. Zahn, M. Hietschold,
Surface Science 2011, 605, 1507–1512; f) S. Lindner, M. Knupfer,
R. Friedrich, T. Hahn, J. Kortus, Phys. Rev. Lett. 2012, 109, 027601.
[12] For thin films of metallo-porphyrines, see: a) A. Scheybal, T.
Ramsvik, R. Bertschinger, M. Putero, F. Nolting, T. A. Jung, Chem.
Phys. Lett. 2005, 411, 214–220; b) H. Wende, M. Bernien, J. Luo,
C. Sorg, N. Ponpandian, J. Kurde, J. Miguel, M. Piantek, X. Xu, P.
Eckhold, W. Kuch, K. Baberschke, P. M. Panchmatia, B. Sanyal,
P. M. Oppeneer, O. Eriksson, Nat. Mater. 2007, 6, 516–520; c) M.
Bernien, J. Miguel, C. Weis, M. E. Ali, J. Kurde, B. Krumme, P. M.
Panchmatia, B. Sanyal, M. Piantek, P. Srivastava, K. Baberschke,
P. M. Oppeneer, O. Eriksson, W. Kuch, H. Wende, Phys. Rev. Lett.
2009, 102, 047202–1–4; d) P. M. Oppeneer, P. M. Panchmatia, N.
Sanyal, O. Eriksson, M. E. Ali, Prog. Surf. Sci. 2009, 84, 18–29.
[13] SMM arranged in thin films or multi-layers: a) M. Clemente-Leꢅn,
H. Soyer, E. Coronado, C. Mingotaud, C. J. Gꢅmez-Garcꢆa, P. Del-
haꢈs, Angew. Chem. 1998, 110, 3053–3056; Angew. Chem. Int. Ed.
1998, 37, 2842–2845; b) A. Cornia, A. C. Fabretti, M. Pacchioni, L.
Zobbi, D. Bonacchi, A. Caneschi, D. Gatteschi, R. Biagi, U. Del
Pennino, V. De Renzi, L. Gurevich, H. S. J. van Der Zant, Angew.
Chem. 2003, 115, 1683–1686; Angew. Chem. Int. Ed. 2003, 42,
1645–1648.
[14] Surface SCO-compounds arranged in rows: M. S. Alam, M. Stock-
er, M. Gieb, K. Mꢀller, P. Haryono, K. Student, A. Grohmann,
Angew. Chem. 2010, 122, 1178–1182; Angew. Chem. Int. Ed. 2010,
49, 1159–1163.
[15] a) W. Hieringer, K. Flechtner, A. Kretschmann, K. Seufert, W.
Auwꢁrter, J. V. Barth, A. Gçrling, H.-P. Steinrꢀck, J. M. Gottfried,
J. Am. Chem. Soc. 2011, 133, 6206–6222; b) T. Komeda, H. Isshiki,
J. Liu, Y.-F. Zhang, N. Lorente, K. Katoh, B. K. Breedlove, M. Ya-
mashita, Nat. Commun. 2011, 2, 217–223; c) T. Choi, S. Bedwani,
A. Rochefort, C.-Y. Chen, A. J. Epstein, J. A. Gupta, Nano Lett.
2010, 10, 4175–4180; d) T. G. Gopakumar, F. Matino, H. Naggert,
A. Bannwarth, F. Tuczek, R. Berndt, Angew. Chem. 2012, 124,
6367–6371; Angew. Chem. Int. Ed. 2012, 51, 6262–6266.
[16] SMMꢉs arranged in thin films: P. Gambardella, S. Stepanow, A.
Dmitriev, J. Honolka, F. M. F. de Groot, M. Lingenfelder; S. S.
Gupta, D. D. Sarma, P. Bencok, S. Stanescu, S. Clair, S. Pons, N.
Lin, A. P. Seitsonen, H. Brune, J. V. Barth, K. Kern, Nat. Mater.
2009, 8, 189–193; S. S. Gupta, D. D. Sarma, P. Bencok, S. Stanescu,
S. Clair, S. Pons, N. Lin, A. P. Seitsonen, H. Brune, J. V. Barth, K.
Kern, Nat. Mater. 2009, 8, 189–193.
[17] SCO compounds arranged in thin films or multi-layers a) H. Soyer,
C. Mingotaud, M.-L. Boillot, P. Delhaꢊs, Thin Solid Films 1998,
327–329, 435–438; b) A. B. Gaspar, M. Seredyuk, P. Gꢀtlich,
Coord. Chem. Rev. 2009, 253, 2399–2413; c) G. Agustꢆ, S. Cobo,
A. B. Gaspar, G. Molnꢇr, N. O. Moussa, P. ꢋ. Szilꢇgyi, V. Pꢇlfi, C.
Vieu, M. C. Muoz, J. A. Real, A. Bousseksou, Chem. Mater. 2008,
20, 6721–6732; d) I. Boldog, A. B. Gaspar, V. Martꢆnez, P. Pardo-
IbaÇez, V. Ksenofontov, A. Bhattacharjee, P. Gꢀtlich, J. A. Real,
Angew. Chem. 2008, 120, 6533–6537; Angew. Chem. Int. Ed. 2008,
47, 6433–6437; e) F. Volatron, L. Catala, E. Riviꢊre, A. Gloter, O.
Stꢊphan, T. Mallah, Inorg. Chem. 2008, 47, 6584–6586; f) C. Raja-
durai, F. Schramm, S. Brink, O. Fuhr, M. Ghafari, R. Kruk, M.
Ruben, Inorg. Chem. 2006, 45, 10019–10021; g) Y. Bodenthin, U.
Pietsch, H. Mçhwald, D. G. Kurth, J. Am. Chem. Soc. 2005, 127,
3110–3114.
[19] J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M. White-
sides, Chem. Rev. 2005, 105, 1103–1169.
[20] A. Ulman, Chem. Rev. 1996, 96, 1533–1554.
[21] Y. Xia, G. M. Whitesides, Angew. Chem. 1998, 110, 568–594;
Angew. Chem. Int. Ed. 1998, 37, 550–575.
[22] a) A. Naitabdi, J.-P. Bucher, P. Gerbier, P. Rabu, M. Drillon, Adv.
Mater. 2005, 17, 1612–1616; b) J. S. Steckel, N. S. Persky, C. R.
Martinez, C. L. Barnes, E. A. Fry, J. Kulkarni, J. D. Burgess, R. B.
Pacheco, S. L. Stoll, Nano Lett. 2004, 4, 399–402; c) S. Voss, M.
Burgert, M. Fonin, U. Groth, U. Rꢀdiger, Dalton Trans. 2008, 499–
505; d) E. Coronado, A. Forment-Aliaga, F. M. Romero, V. Corra-
dini, R. Biagi, V. De Renzi, A. Gambardella, U. Del Pennino,
Inorg. Chem. 2005, 44, 7693–7695.
[23] M. Mannini, F. Pineider, P. Sainctavit, C. Danieli, E. Otero,
A. M. T. Sciancalepore, M.-A. Arrio, A. Cornia, D. Gatteschi, R.
Sessoli, Nat. Mater. 2009, 8, 194–197.
[24] a) L. Zobbi, M. Mannini, M. Pacchioni, G. Chastanet, D. Bonacchi,
C. Zanardi, R. Biagi, U. Del Pennino, D. Gatteschi, A. Cornia, R.
Sessoli, Chem. Commun. 2005, 1640–1642; b) M. Mannini, F. Pine-
ider, C. Danieli, F. Totti, L. Sorace, P. Sainctavit, M.-A. Arrio, E.
Otero, L. Joly, J. C. Cezar, A. Cornia, R. Sessoli, Nature 2010, 468,
417–421; c) M. Pacchioni, A. Cornia, A. C. Fabretti, L. Zobbi, D.
Bonacchi, A. Caneschi, G. Chastanet, D. Gatteschi, R. Sessoli,
Chem. Commun. 2004, 2604–2605.
[25] V. Lozan, C. Loose, J. Kortus, B. Kersting, Coord. Chem. Rev. 2009,
253, 2244–2260.
[26] U. Lehmann, J. Lach, C. Loose, T. Hahn, B. Kersting, J. Kortus,
Dalton Trans. 2013, 42, 987–996.
[27] V. Lozan, B. Kersting, Inorg. Chem. 2008, 47, 5386–5393.
[28] U. Lehmann, J. Klingele, V. Lozan, G. Steinfeld, M. H. Klingele, S.
Kꢁss, A. Rodenstein, B. Kersting, Inorg. Chem. 2010, 49, 11018–
11029.
[29] B. Kersting in Activating Unreactive Molecules: The Role of Secon-
dary Interactions (Eds.: C. Bolm, F. E. Hahn), Wiley-VCH, Wein-
heim 2009, Chapter 1, pp. 1–17.
[30] B. Kersting, U. Lehmann, Adv. Inorg. Chem. 2009, 61, 407–470,
(Ed.: R. v. Eldik, C. D. Hubbard)
[31] P. Chaudhuri, V. Kataev, B. Bꢀchner, H.-H. Klauss, B. Kersting, F.
Meyer, Coord. Chem. Rev. 2009, 253, 2261–2285.
[32] C. Loose, E. Ruiz, B. Kersting, J. Kortus, Chem. Phys. Lett. 2008,
452, 38–43.
[33] G. Steinfeld, V. Lozan, H.-J. Krꢀger, B. Kersting, Angew. Chem.
2009, 121, 1988–1991; Angew. Chem. Int. Ed. 2009, 48, 1954–1957.
[34] J. Lach, T. Hahn, B. Kersting, J. Kortus, Eur. J. Inorg. Chem. 2012,
2381–2388.
[35] J. Lach, A. Jeremies, V. Lozan, C. Loose, T. Hahn, J. Kortus, B.
Kersting, Inorg. Chem. 2012, 51, 12380–12391.
[36] G. Westermark, I. Persson, Colloids Surf. A 1998, 144, 149–166.
[37] U. B. Steiner, P. Neuenschwander, R. W. Caseri, U. W. Suter, Lang-
muir 1992, 8, 90–94.
[38] B. M. Trost, M. R. Machacek, A. Aponick, Acc. Chem. Res. 2006,
39, 747–60.
[39] G. Schmid, Chem. Soc. Rev. 2008, 37, 1909–1930.
[40] B. K. Teo, X. Shi, H. Zhang, J. Am. Chem. Soc. 1992, 114, 2743–
2745.
[41] Z. Wang, L. Ma, Coord. Chem. Rev. 2009, 253, 1607–1618.
[42] J. M. Pettibone, J. W. Hudgens, J. Phys. Chem. Lett. 2010, 1, 2536–
2540.
[43] P. M. Shem, R. Sardar, J. S. Shumaker-Parry, Langmuir 2009, 25,
13279–13283.
[44] J. M. Pettibone, J. W. Hudgens, ACS Nano 2011, 5, 2989–3002.
[45] Y. S. Chung, K. Evans, W. Glaunsinger, Appl. Surf. Sci. 1998, 125,
65–72.
[46] B. Kersting, G. Steinfeld, Chem. Commun. 2001, 1376–1377.
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7799

B. Kersting et al.
ꢀ
ꢀ
[47] The fact that the BPh4 salts crystallize better than the ClO4 salts
has been observed in other dinuclear nickel complexes supported
[73] K. Hara, R. Akiyama, S. Takakusagi, K. Uosaki, T. Yoshino, H.
Kagi, M. Sawamura, Angew. Chem. 2008, 120, 5709–5712; Angew.
Chem. Int. Ed. 2008, 47, 5627–5630.
ꢀ
by H2L. It is assumed that the more polar ClO4 salts crystallize
with more solvate molecules. Indeed, crystals of the perchlorate
salts upon contact with air quickly loose the solvate molecules and
turn dull.
[74] K. Uvdal, I. Persson, B. Liedberg, Langmuir 1995, 11, 1252–1256.
[75] W. W. Weare, S. M. Reed, M. G. Warner, J. E. Hutchison, J. Am.
Chem. Soc. 2000, 122, 12890–12891.
[48] B. J. Dunne, A. G. Orpen, Acta. Crystallogr. Sect. C 1991, 47, 345–
347.
[49] S.-M. Kuang, P. E. Fanwick, R. A. Walton, Inorg. Chem. 2002, 41,
1036–1038.
[50] H. Schmidbaur, S. Cronje, D. Djordjevic, O. Schuster, Chem. Phys.
2005, 311, 151–161.
[51] J. D. E. T. Wilton-Ely, A. Schier, N. W. Mitzel, H. Schmidbaur, J.
Chem. Soc. Dalton Trans. 2001, 1058–1062.
[76] The different solubilities of a- and b-cyclodextrin are a representa-
tive example, see: J. W. Steed, J. L. Atwood, Supramolecular Chem-
istry, 2nd ed., Wiley, Chichester, 2009, pp. 327–336.
[77] C. D. Bain, E. B. Troughton, Y. T. Tao, J. Evall, G. M. Whitesides,
R. G. Nuzzo, J. Am. Chem. Soc. 1989, 111, 321–335.
[78] M. Ratel, M. Branca, J. Breault-Turcot, S. S. Zhao, P. Chaurand,
A. R. Schmitzer, J.-F. Masson, Chem. Commun. 2011, 47, 10644–
10646.
[52] a) N. W. Alcock, F. R. Hartley, D. E. Rogers, J. Chem. Soc. Dalton
Trans. 1973, 1070–1073; b) P.-K. F. Chin, F. R. Hartley, Inorg.
Chem. 1976, 15, 982–984.
[53] S. H. Strauss, Chem. Rev. 1993, 93, 927–942.
[54] J. M. Forward, J. P. Fackler, Jr., R. J. Staples, Organometallics 1995,
14, 4194–4198.
[55] J. P. Fackler, R. J. Staples, A. Elduque, T. Grant, Acta Crystallogr.
Sect. C 1994, 50, 520–523.
[56] X. Hong, K.-K. Cheung, C.-X. Guo, C.-M. Che, J. Chem. Soc.
Dalton Trans. 1994, 1867–1871.
[79] The sample integrity and the sensitivity to exposure to the X-ray
beam while taking XPS data has been checked by a comparison of
data sets that have been measured for different periods of time.
For instance, after an exposure to the X-rays of a few hours only
(2–3 h), no additional phosphorous signal can be observed as it is
the case for longer measurements. Unfortunately, the shorter meas-
urements do not allow a reliable analysis of the composition due to
bad statistics. The small amount of material in the monolayer re-
quired long acquisition times.
[80] R. G. Copperthwaite, J. Lloyd, J. Chem. Soc. Dalton Trans. 1977,
1117–1121.
[81] W. E. Swartz, Jr., J. K. Ruff, D. M. Hercules, J. Am. Chem. Soc.
1972, 94, 5227–5229.
[82] E. Fluck, D. Weber, Z. Anorg. Allg. Chem. 1975, 412, 47–58.
[83] L. Wang, D. Dehe, T. Philippi, A. Seifert, S. Ernst, Z. Zhou, M.
Hartmann, R. N. Klupp-Taylor, A. P. Singh, M. Jia, W. R. Thiel,
Catal. Sci. Technol. 2012, 2, 1188–1195.
[84] a) K. Velauthamurty, S. J. Higgins, R. M. G. Rajapakse, J. Bacsa, H.
van Zalinge, R. J. Nichols, W. Haiss, J. Mater. Chem. 2009, 19,
1850–1858; b) E. de Wolf, A. J. M. Mens, Gijzeman, J. H. van
Lenthe, L. W. Jenneskens, B.-J. Deelman, G. van Koten, Inorg.
Chem. 2003, 42, 2115–2124; c) T. Borkowski, W. Zawartka, P. Pos-
piech, U. Mizerska, A. M. Trzeciak, M. Cypryk, W. Tylus, J. Catal.
2011, 282, 270–277.
[85] L. J. Matienzo, L. I. Yin, S. O. Grim, W. E. Swartz, Jr., Inorg.
Chem. 1973, 12, 2762–2769.
[86] T. Gao, P. Norby, H. Okamoto, H. Fjellvꢌg, Inorg. Chem. 2009, 48,
9409–9418.
[87] I. Preda, A. Gutiꢊrrez, A. Abbate, F. Yubero, J. Mꢊndez, L. Alvar-
ez, L. Soriano, Phys. Rev. B 2008, 77, 075411–1–7.
[88] An analysis of the SAM stoichiometry shows a Ni/S atomic ratio of
0.24 rather than the nominal value (0.66), which is attributed to the
low intensity of the Ni and S signals.
[89] C. M. Whelan, C. J. Barnes, C. G. H. Walker, N. M. D. Brown, Surf.
Sci. 1999, 425, 195–211.
[90] M. Wells, D. L. Dermody, H. C. Yang, T. Kim, R. M. Crooks, Lang-
muir 1996, 12, 1989–1996.
[91] Optical Properties of Condensed Matter and Applications (Ed.: J.
Singh), Wiley, Chichester, 2006, p.7.
[57] Y. Journaux, T. Glaser, G. Steinfeld, V. Lozan, B. Kersting, Dalton
Trans. 2006, 1738–1748.
[58] Preliminary DFT calculations have been performed for the mixed-
valent [Ni2LACHTUNGTRENNUNG
(dppba)]2+ dication. These reveal an Stot =3/2 ground
state that is attained by a ferromagnetic exchange interaction be-
tween a NiII (S1 =1) and a low-spin NiIII (S2 =1/2) ion. The S=3/2
ground state for the [Ni2LACTHNUTRGNEUNG
(dppba)]2+ dication is also confirmed by
preliminary ESR spectroscopic measurements at X-band as well as
by temperature-dependent magnetic susceptibility measurements.
C. Loose, J. Kortus, R. Schnorr, B. Kersting, unpublished results.
[59] A solution of the one-electron oxidized 22+ dication has been pre-
pared by oxidation of [Ni2L
Cu
(ClO4)2. The electronic absorption spectrum of 22+ (lmax (e)=
782 (4945), 517 (2735), 397 sh (2998), 330 (9864), 304 nm
(13560 mꢀ1 cmꢀ1) is very similar to that of 32+
ACHTUNGTERNN(UNG dppba)]ClO4 with one equivalent of
ACHTUNGTRENNUNG
.
[60] Encyclopedia of Electrochemistry of the Elements, Vol. IV (Ed.:
A. J. Bard), Marcel Dekker, New York, 1975, p.101.
[61] R. Uson, A. Laguna, Organomet. React. Synth. 1986, 3, 322–342.
[62] H.-J. Krueger, R. H. Holm, Inorg. Chem. 1989, 28, 1148–1155.
[63] J. D. Franolic, W. Y. Wang, M. Millar, J. Am. Chem. Soc. 1992, 114,
6587–6588.
[64] G. Henkel, T. Krꢀger, B. Krebs, Angew. Chem. 1992, 104, 71–72;
Angew. Chem. Int. Ed. Engl. 1992, 31, 54–56.
[65] T. Beissel, T. Glaser, F. Kesting, K. Wieghardt, B. Nuber, Inorg.
Chem. 1996, 35, 3936–3947.
[66] B. Kersting, D. Siebert, Inorg. Chem. 1998, 37, 3820–3828.
[67] X. E. Bill, http://www.mpibac.mpg.de/bac/logins/bill/julXen.php.
[68] We have recorded reduced magnetization data for a similar set of
complexes, [Ni2(L)(O2CC6H
4ACHTUNTGERNNUGSH)]HCATUGNTREN[NNGU ClO4] and [Ni2(L)(O2CC6H4-
A
ACHTUNGTRENNUNG
[92] S. R. Wassermann, G. M. Whitesides, I. M. Tidswell, B. M. Ocko,
P. S. Pershan, J. D. Axe, J. Am. Chem. Soc. 1989, 111, 5852–5861.
[93] Others authors use A=1.50, which, in the case of the present mon-
olayers, would only make a thickness difference of about 0.1 nm,
see D. L. Allara, R. G. Nuzzo, Langmuir 1985, 1, 45–52.
[94] A. Ulman, Acc. Chem. Res. 2001, 34, 855–863.
the D parameter. A meaningful simulation of magnetization data
for these complexes necessitates knowledge of the relative orienta-
tions of zero-field splitting tensors, which is beyond the scope of
this study. J. Lach, Dissertation, Universitꢁt Leipzig, 2013.
[69] C. Golze, A. Alfonsov, R. Klingeler, B. Bꢀchner, V. Kataev, C.
Mennerich, H.-H. Klauss, M. Goiran, J.-B. Broto, H. Rakoto, S.
Demeshko, G. Leibeling, F. Meyer, Phys. Rev. B 2006, 73, 224403–
1–8.
[95] This value is calculated by adding the van der Waals radii of two H
atoms (2.4 ꢃ) to the atom distance of 13.123 ꢃ.
[96] A. Reis, D. Dehe, S. Farsadpour, I. Munstein, Y. Sun, W. R. Thiel,
New J. Chem. 2011, 35, 2488–2495.
[70] A. Ulman, An Introduction to Ultrathin Organic Films from Lang-
muir–Blodgett to Self-Assembly, Academic Press, New York, 1991,
pp. 237–300.
[71] The complexes cannot be deposited by using the gas phase. They
decompose without melting.
[97] Under our experimental conditions E(Cp2Co+/Cp2Co)=ꢀ1.345 V
versus E(Cp2Fe+/Cp2Fe); E(Cp2Co+/Cp2Co)=ꢀ0.937 V versus
SCE. N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.
[98] The cleaning of the substrate by the piranha solution proved neces-
sary to obtain reasonable surface thicknesses.
[72] B. S. Sumerlin, A. B. Lowe, P. A. Stroud, P. Zhang, M. W. Urban,
C. L. McCormick, Langmuir 2003, 19, 5559–5562.
7800
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 7787 – 7801

Exchange-Coupled [Ni2LACTHNUTRGNEUNG
(dppba)]+ Complexes
FULL PAPER
[99] CompleteEase, Software for Ellipsometry Data Aquisition and
Analysis, J. A. Wollam Co., Inc. Lincoln, NE, USA.
[100] Casa Software Ltd, Bay House 5 Grosvenor Terrace Teignmouth,
TQ14 8NE, United Kingdom.
[104] A. L. Spek, PLATON - A Multipurpose Crystallographic Tool;
Utrecht University, Utrecht, The Netherlands, 2000.
[105] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.
[101] Stoe, Cie, IPDS and X-RED32 2008.
[102] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467–473.
[103] G. M. Sheldrick, SHELXL-97, Computer program for crystal struc-
ture refinement, University of Gçttingen, Gçttingen, Germany,
1997.
Received: February 7, 2013
Published online: April 17, 2013
Chem. Eur. J. 2013, 19, 7787 – 7801
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7801