www.advmat.de
www.MaterialsViews.com
Yield: 2.30 g (52%); IR (KBr): ν = 3016 (vs), 1638 (vs), 815 cm− (vs)
1
(
for details, see Supporting Information, Figure S1); UV–vis (CH CN):
3
λmax (ε) = 206 (sh; ≈37 000), 248.5 nm (35 500); Anal. calcd for
C H N I : C 32.40, H 3.16, N 6.31, I 57.62; found: C 32.00, H 2.68, N
1
2
14 2 2
6
.26, I 58.35.
Single crystals of MV[Ni(dmit)2]2 were obtained from a diffusion
method of MV·I2 (30 mg) and [n-(C H ) N][Ni(dmit) ] (80 mg) in dry
4
9 4
2
acetonitrile (Wako; 60 mL) solution under a N atmosphere in an H-tube
2
with fine glass frits (G4) for separation of each compartment. The
solution stood in the dark for 10 d at room temperature (RT) to yield
diamond-shaped shiny black platelets of 0.5–1.0 mm in size.
−1
IR (KBr): ν = 1638 (w), 1344 (vs), 1056 (vs) cm (for details, see
Supporting Information, Figure S1); Anal. Calcd for C H N Ni S20:
24
14
2
2
Figure 4. Temperature (T) dependence of resistance (R) of MV[Ni] under
C 26.46, H 1.30, N 2.57, S 58.89; found: C 26.17, H 1.52, N 2.52, S 58.92.
In the following measurements on the single crystals, except for the
magnetic susceptibility, all of the single crystals were briefly checked
by X-ray oscillation photographs to identify the crystal quality and the
directions of the crystallographic axes.
2
UV irradiation showing the linear variation of R vs. logT at low T, which
is characteristic of carriers interacting with localized spins (the Kondo
effect).
interpretation. Additionally, a calculation of the overlap integrals
of the molecular orbitals (Supporting Information, Table S2,S3)
Irradiation Method: Various light sources with or without band-pass
filters (Δλ = 40–60 nm) and/or optical fibers were used to examine
the photoresponses in the physical properties and their wavelength
dependence: a high-pressure Hg/Xe lamp (250–450 nm) (SAN-EI
Electric, SUPERCURE-203S; 200 W), a Xe lamp (200–400 nm) (Asahi
2+
suggests that the localized spins on MV * and the carriers on
−
[
Ni] * interact. To obtain further convincing evidence for the
2+
interaction between the unpaired electrons on MV * (localized
spins) and on [Ni] * (carriers), the temperature-dependence of
Spectra, LAX-Cute; 100 W), a D /tungsten-halogen lamp (200–1100 nm)
2
−
(Hamamatsu Photonics K. K., L7893; 30 W (D ) & 5 W (halogen)), a
2
the resistivity under UV irradiation (Figure 4 ) was measured.
In contrast with the observed semiconducting behavior (∂T < 0;
high-power tungsten-halogen lamp (300–2600 nm) (Spectral Products,
∂
R
model ASBN-W100F-L; 100 W), and a laser (375 ± 5 nm) (NEOARK;
2
0 mW). The actual light power of a particular wavelength under
R = resistance, T = temperature) in the dark, MV[Ni] exhibits
2
ꢀ
ꢁ
each experimental condition was measured using a Si-diode power
meter (OPHIR, NOVA). Similarly, some of the lamps with filters and a
fiberscope above were used to examine the spectral changes in the ESR.
The details of the specifications of the light sources used in this work are
∂
∂
R
T
clear metallic behavior
> 0 above ≈200 K under UV irra-
diation. However, R increases linearly with logT when T is
reduced below ≈100 K. This behavior can be explained by the
Kondo effect.[ Electronic systems composed of carriers and
localized spins characteristically exhibit this effect at low tem-
perature due to the interaction between carriers and localized
spins. The Kondo effect is typically observed in metallic elec-
tronic systems containing dilute localized spin systems. Since
UV irradiation excites only the MV2 cations at the crystal sur-
10]
[23,24]
described in our previous papers.
Physical-Property Measurements: Single-crystal X-ray structural analysis
and magnetic susceptibility measurements were performed under dark
conditions. The details of the structural analysis are described in the cif
file deposited to the Cambridge Crystallographic Data Centre (CCDC).
Electrical resistivity and ESR measurements were performed both
under irradiation and in the dark. During the resistivity measurements
performed under irradiation, the sample temperature was monitored
using a Si-diode thermometer on which the sample was mounted;
both the sample and the thermometer were simultaneously irradiated.
Because of the high electrical resistivity under dark conditions, and for
the sake of ensuring an effective irradiation area, as wide as possible,
on the crystal surface, we applied a two-probe method (suitable for
samples with high resistance) instead of the standard four-probe
method (suitable for samples with low resistance). Because the accurate
penetration depth of the UV light was difficult to measure, and because
one could not discuss the absolute values of resistivity by the two-probe
method in the case of metallic samples, the resistivity under irradiation
is shown by an arbitrary unit in Figure 4. In the ESR measurements, the
background signal (originating from the cavity, the quartz sample tube,
and a polytetrafluoroethylene (Teflon) piece used to mount the sample)
was measured under both dark and irradiated conditions at RT and 77 K.
During UV irradiation in the ESR measurements, the temperature near
the sample was monitored using a Si-diode thermometer in the same
sample tube in an independent measurement (Supporting Information,
Figure S8). To estimate the effect of heating during UV irradiation,
we also measured the ESR spectra at 313 K, which was approximately
the temperature of the sample during irradiation at RT; no signal was
observed in this measurement. Further details of the experiments are
described in the Supporting Information. The Supporting Information
also describes the parameters (Supporting Information, Table S2) and
other details of the band dispersion and related calculations.
+
face to produce a dilute localized spin system in MV[Ni] , the
2
observed behavior of R is consistent with the Kondo effect, as
well as all of the data presented above.
In conclusion, it has been established that PE reversibly gen-
erates a metallic state with localized spins, and that the carriers
and localized spins interact in the photoexcited state of MV[Ni]2.
This is related to charge transfer, corresponding to a photochem-
ical partial redox reaction between MV2 and [Ni] ions, resulting
+
−
−
in doping (carrier injection) in the Ni(dmit) band; [Ni]
→
2
(1–δ)−
[Ni]
(δ ≈ 0.02). Organic materials are currently considered to
be the most-promising materials for realizing spintronics. Recent
advances in related studies have been remarkable, yet there still
remains much to be explored about the process of spin injection
into organic materials.[
11–21]
This novel type of photoconductor
provides an alternative pathway to organic spintronics.
Experimental Section
Sample Preparation: All of the chemicals were purchased as the
highest grade and used as received. [n-(C H ) N][Ni(dmit) ] was
4
9 4
2
synthesized according to the literature.[
22]
MV·I2 was prepared by
treating 4,4′-bipyridyl (Tokyo Chemical Industry, 1.56 g: 0.01 mol)
with a large excess of methyl iodide (Wako Pure Chemical Industries,
CCDC 881155 contains the supplementary 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.
2
8.0 g: 0.20 mol) in refluxing ethanol (Wako, 50 mL) for 3 h, followed
by recrystallization from hot methanol (Wako). Reddish orange needles
were produced.
4
wileyonlinelibrary.com
© 2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2012,
DOI: 10.1002/adma.201203153