RSC Advances
Paper
Characterization methods. 1H NMR, 13C NMR and 19F NMR
were measured on a JEOL GSX-400 spectrometer (400 MHz for
1H, 100.5 MHz for 13C, and 376.2 MHz for 19F). All chemical
shis (d) are reported on parts per million downeld of aceto-
1
nitrile for H NMR (d ¼ 1.96) and 13C NMR (d ¼ 118.26) and
trichlorouoromethane for 19F NMR (d ¼ 0.0) were used as
internal standard for CD3CN solutions. ESI mass spectra were
measured on a Thermo Scientic Exactive mass spectrometer.
Optical rotations were measured on a JASCO P-2200 polarimeter
at room temperature, using the sodium D line.
Fig. 6 The molecular models: (a) enantiopure-CFs, (b) racemic-CFs.
General procedure. To a solution of 1,2-diaminocyclohexane
(0.31 mmol) in dry THF (3 ml) were added Et3N (0.93 mmol, 3.0
eq.) and peruoroalkanoyl chloride (0.93 mmol, 3.0 eq.) at 0 ꢀC,
and the mixture was stirred for 1 hour. The resulting mixture
was ltrated and the residue was washed with water. Chemical
data of CF7 and CF8 were previously reported.13,14
The above features suggested that the stacking interaction
among peruorinated chains has a main factor to bring about
chirality effects on gelation. According to the VCD analyses, the
stable aggregation mode for enantiopure CF9 was achieved by
the formation of two anti-parallel intermolecular hydrogen
bonds (Fig. 6(a)). The same connection was conrmed to exist in
the single crystal of CF4. Since a single array was estimated to
have a radius of >2 nm, the thin brils with a radius of >2 mm,
which were observed in the SEM images contained more than
ten such arrays. For enantiopure CF7 and CF8, the molecules
formed a single intermolecular hydrogen bond. Thus they
might form a weaker array. In both cases, however, the outer
surface of the array possessed the region of peruorinated
chains. It is speculated therefore that the odd-even effect
observed for the gelation ability of enantiopure CFn's (Fig. 1)
appeared when the array was bundle to a thin bril.
In contrast, RR- or SS-enantiomers formed a homochiral pair
and oriented in a head-and-tail way in crystal structure and each
array consisted of the opposite enantiomers arranged in an
alternative way through the two-antiparallel intermolecular
hydrogen bonds. It is proposed that the aggregation of the array
to form a thinner bril is achieved by the interdigitation of
peruorinated chains. Under such an aggregation mode, no
contact of the end groups of peruorinated chains would
appear, leading to the absence of odd–even effect.
trans-N,N0-Peruorobutanoyl-1,2-diaminocyclohexane (CF4).
(1R,2R)-CF4 [a]2D7 35.63 (c 0.19, THF), (1S,2S)-CF4 [a]2D8 ꢁ37.49 (c
1
0.19, THF); H NMR (400 MHz, CD3CN) d 7.64 (2H, m, NHCO),
3.87 (2H, m, CH2CHNH), 1.78 (4H, m, cyclohexyl), 1.47 (2H, m,
cyclohexyl), 1.34 (2H, m, cyclohexyl); 13C NMR (126 MHz,
CD3CN) d 52.3 (2C), 30.0 (2C), 23.3 (2C); 19F NMR (376 MHz,
CD3CN) d ꢁ80.3 (6F), ꢁ119.9 (4F), ꢁ126.5 (4F); HRMS (ESIꢁ)
C14H11O2N2F14 [M ꢁ H] (calcd 505.0597, found 505.0578).
trans-N,N0-Peruoropentanoyl-1,2-diaminocyclohexane (CF5).
(1R,2R)-CF5 [a]2D7 35.91 (c 0.20, THF), (1S,2S)-CF5 [a]2D7 ꢁ34.64 (c
1
0.19, THF); H NMR (400 MHz, CD3CN) d 7.63 (2H, m, NHCO),
3.86 (2H, m, CH2CHNH), 1.77 (4H, m, cyclohexyl), 1.45 (2H, m,
cyclohexyl), 1.32 (2H, m, cyclohexyl); 13C NMR (126 MHz, CD3CN)
d 52.8 (2C), 30.8 (2C), 23.7 (2C); 19F NMR (376 MHz, CD3CN) d
ꢁ81.8 (6F), ꢁ120.4 (4F), ꢁ124.1 (4F), ꢁ126.5 (4F); HRMS (ESIꢁ)
C16H11O2N2F18 [M ꢁ H] (calcd 605.0533, found 605.0524).
trans-N,N0-Peruorohexanoyl-1,2-diaminocyclohexane (CF6).
(1R,2R)-CF6 [a]2D8 20.17 (c 0.19, THF), (1S,2S)-CF6 [a]2D8 ꢁ22.40 (c
1
0.19, THF); H NMR (400 MHz, CD3CN) d 7.65 (2H, m, NHCO),
3.86 (2H, m, CH2CHNH), 1.75 (4H, m, cyclohexyl), 1.45 (2H, m,
cyclohexyl), 1.33 (2H, m, cyclohexyl); 13C NMR (126 MHz,
CD3CN) d 52.8 (2C), 30.8 (2C), 23.8 (2C); 19F NMR (376 MHz,
CD3CN) d ꢁ81.5 (6F), ꢁ120.2 (4F), ꢁ123.1 (4F), ꢁ123.3 (4F),
ꢁ126.7 (4F); HRMS (ESIꢁ) C18H11O2N2F22 [M ꢁ H] (calcd
705.0469, found 705.0461).
It should be emphasized that the aim of the above stacking
model was to rationalize the difference of gelation behavior
between racemic and enantiomeric gelators. It is intended to
present no speculative details of bril structures but simply
demonstrates the possible connectivity of the gelator mole-
cules. Presently an experimental support is now under progress
by nding any periodic structure that depends on molecular
chirality such as neutron diffraction or small angle X-ray
diffraction measurements.
trans-N,N0-Peruorononanoyl-1,2-diaminocyclohexane (CF9).
(1R,2R)-CF9 [a]2D7 19.492 (c 0.19, THF), (1S,2S)-CF9 [a]2D7 ꢁ23.841
(c 0.19, THF); 1H NMR (400 MHz, CD3CN) d 7.64 (2H, m, NHCO),
3.86 (2H, m, CH2CHNH), 1.76 (4H, m, cyclohexyl), 1.47 (2H, m,
cyclohexyl), 1.33 (2H, m, cyclohexyl); 13C NMR (126 MHz,
CD3CN) d 52.5 (2C), 30.9 (2C), 23.6 (2C); 19F NMR (376 MHz,
CD3CN) d ꢁ81.6 (6F), ꢁ119.7 (4F), ꢁ122.1 (4F), ꢁ122.5 (8F),
ꢁ123.1 (4F), ꢁ123.7 (4F), ꢁ126.6 (4F); HRMS (ESIꢁ)
C24H11O2N2F34 [M ꢁ H] (calcd 1005.0278, found 1105.0273).
trans-N,N0-Peruorodecanoyl-1,2-diaminocyclohexane (CF10).
(1R,2R)-CF10 [a]2D8 13.69 (c 0.19, THF), (1S,2S)-CF10
Experimental
Synthesis and characterization of gelators
Materials. All commercially available solvents and reagents
for synthesis and analysis were used as received without further
purication. Benzotriuoride (>98%) was obtained from Kanto
Chemical Co. Inc. Peruorobenzen (>99%) was obtained from
Aldrich. (1R,2R)-, (1S,2S)- and racemic-1,2-diaminocyclohexane
(>98%) and tetradecauorohexane (>96%) were obtained from
Tokyo Kasei, Ltd. Peruoroalkanoys chlorides were synthesized
according to previous report.35
1
[a]2D8 ꢁ16.53 (c 0.25, THF); H NMR (400 MHz, CD3CN) d 7.64
(2H, m, NHCO), 3.86 (2H, m, CH2CHNH), 1.76 (4H, m, cyclo-
hexyl), 1.47 (2H, m, cyclohexyl), 1.33 (2H, m, cyclohexyl); 13C
NMR (126 MHz, CD3CN) d 52.4 (2C), 30.7 (2C), 23.5 (2C); 19F
80546 | RSC Adv., 2015, 5, 80542–80547
This journal is © The Royal Society of Chemistry 2015