Chemistry Letters Vol.37, No.7 (2008)
761
of 2 favors molecular packing suitable for the meta-stable phase.
In summary, the thermal cycle of 2 showed a very wide
irreversible loop. It behaves as a SCO compound with
(a)
1
TC ¼ 300 K, but once heated above 375 K it irreversibly
0
became a SCO compound with TC ¼ 210 K. Pressurization
elevates the melting point and prohibits it to enter a stable phase.
This work suggests a potential utility of 2 as a magnetic-detec-
tion pressure sensor as well as a write-once information storage
material at ambient pressure.13
(b)
This work was supported by Grants-in-Aid for Scientific
Research (Nos. 17550166 and 19550135) from the Ministry of
Education, Culture, Sports, Science and Technology, Japan.
References and Notes
1
Compounds I, II, and III, Springer-Verlag, Berlin, 2004; P. Gutlich,
P. Gutlich, H. A. Goodwin, Spin Crossover in Transition Metal
¨
¨
2
3
Figure 3. (a) DSC measurements on 2 at a warming rate of
3 K minꢁ1. A dotted line in the inset shows the data when the
temperature scan was suspended at 393 K for 180 min. (b) Free
energy diagram for 2.
´
J.-F. Letard, P. Guionneau, E. Codjovi, O. Lavastre, G. Bravic,
Hayami, Z. Gu, H. Yoshiki, A. Fujishima, O. Sato, J. Am. Chem.
´
across 390 K, but it gradually melted as indicated by the disap-
pearance of the XRD peaks. An overheated portion might remain
above the melting point.
´
´
´
J. R. Galan-Mascaros, M. Monrabal-Capilla, J. Garcıa-Marthinez,
´
A. B. Gaspar, V. Ksenofontov, M. Seredyuk, P. Gutlich, Coord.
A relaxation from a meta-stable phase to a ground phase is
suggested by the exothermic peak at 375 K (ca. 25 kJ/mol). Each
phase has low-spin (ls) and high-spin (hs) states. A G vs. T dia-
gram is usually convenient (Figure 3b), where we named the
phases as S1 (meta-stable solid), S0 (stable solid), and L (liquid).
Lls is energetically unstable and undetectable. The thermal cycle
of 2 is interpreted as: S1 ! (TC1) ! S1 ! (mp1) !
4
5
¨
H. Inokuchi, G. Saito, P. Wu, K. Seki, T. B. Tang, T. Mori, K.
Imaeda, T. Enoki, Y. Higuchi, K. Inaka, N. Yasuoka, Chem. Lett.
W. Zhang, F. Zhao, T. Liu, M. Yuan, Z.-M. Wang, S. Gao, Inorg.
M. Seredyuk, A. B. Gaspar, V. Ksenofontov, S. Reiman, Y.
6
ls
hs
S0 ! (mp0) ! Lhs on heating and Lhs ! (mp0) ! S0
!
hs
hs
(TC0) ! S0 on cooling. Since mp1 is located lower than
Galyametdinov, W. Haase, E. Entschler, P. Gutlich, Chem. Mater.
¨
ls
mp0, a partial melt induces an exothermic solid–solid phase
transition or resolidifying occurs immediately after the melt
(an arrow in Figure 3b).
S. Hayami, R. Moriyama, Y. Shigeyoshi, R. Kawajiri, T. Mitani,
1: mp >250 ꢂC. Anal. Calcd for C36H26FeN8S2: C, 62.61; H, 3.79;
N, 16.23; S, 9.29%. Found: C, 62.47; H, 3.73; N, 15.63; S, 8.84%.
2: mp 117–134 ꢂC. Anal. Calcd for C68H90FeN8S2: C, 71.68; H,
7.96; N, 9.83; S, 5.63%. Found: C, 71.84; H, 8.16; N, 9.99; S,
5.34%.
Two phenomena were separated by applying pressure, dem-
onstrating that the magnetic properties shown in Figure 2a are
caused by the combination of the SCO and solid–solid phase
transitions. We measured the SCO of 2 using a Cu–Be clamp-
type cylinder cell (ElectroLAB, Japan).12 As Figure 2b shows,
the melt phase transition was removed, but the SCO transition
remained at 330 K, that is, shifted higher by 30 K, under
0.11 GPa. The thermal loop completely disappeared in a temper-
ature range available. A pressure effect on the SCO transition
temperature was confirmed on a similar experiment on 1; the
TC was elevated only by ca. 40 K under 0.23 GPa.
Pressure effects to stabilize low-temperature phases are
common, and the transition temperatures are elevated according
to the Clapayron–Clausius equation. In the present case, the melt
largely depends on the volume change due to the bulky substitu-
ents, and consequently the mp is more sensitive to pressure than
the SCO TC. On applying pressure on 2, TC1 is left behind below
400 K while the mp1 is shifted over 400 K.
7
8
9
Selected crystallographic data of 1. At 100 K: monoclinic,
˚
C2=c, a ¼ 13:681ð3Þ, b ¼ 11:360ð3Þ, c ¼ 21:345ð5ÞA, ꢁ ¼
ꢂ
3
˚
93:5819ð17Þ , V ¼ 3310:9ð14Þ A . At 400 K: monoclinic, C2=c,
ꢂ
˚
a ¼ 13:62ð4Þ, b ¼ 11:644ð18Þ, c ¼ 22:25ð3Þ A, ꢁ ¼ 92:617ð10Þ ,
3
˚
V ¼ 3525ð13Þ A . For details, see CCDC 684391 and 684392.
10 B. Gallois, J.-A. Real, C. Hauw, J. Zarembowitch, Inorg. Chem.
11 The TC is defined by the temperature where dðꢀmolTÞ=dT is
maximum, because a conventional T1=2 value cannot be defined
owing to a ꢀmolT jump. An error within ꢃ5 K was estimated.
12 Y. Uwatoko, Koatsuryoku no Kagaku to Gijutsu 2001, 11, 181;
13 For experimental details, the ꢀmolT vs. T plots on 1 and the C4 and
C
14 derivatives, the results on 1 under pressure, the XRD profiles on
A problem still remains; we could not obtain S0 of 2 by
recrystallization from any alcoholic solvents investigated here.
It may be because the solvation sphere around each molecule
2 at 360 and 380 K, and the DSC profile on the C14 derivative, see
Supporting Information, being available electronically on the CSJ-