Liquid crystals with multiple fluorinated bridges in the mesogenic core structure

Article abstract of DOI:10.1016/S0022-1139(01)00488-2

A rational, building block oriented design approach gives rise to a new class of "super fluorinated" liquid crystals suitable for the newest generation of active matrix LCDs. These materials contain combinations of different fluorinated linking groups in their mesogenic core structure, resulting in a highly advantageous property profile including a broad nematic phase range, low rotational viscosity and high dielectric anisotropy. The large-scale synthesis of such complex materials became feasible only by some recent advances in the synthesis of α,α-difluoroether linkages.

Full text of DOI:10.1016/S0022-1139(01)00488-2

Journal of Fluorine Chemistry 112 82001) 69±72
Liquid crystals with multiple ¯uorinated bridges in the
mesogenic core structure
Peer Kirsch*, Florian Huber, Marc Lenges, Andreas Taugerbeck
Merck KGaA, Liquid Crystals Division, D-64271 Darmstadt, Germany
Abstract
A rational, building blockoriented design approach gives rise to a new class of ``super ¯uorinated'' liquid crystals suitable for the newest
generation of active matrix LCDs. These materials contain combinations of different ¯uorinated linking groups in their mesogenic core
structure, resulting in a highly advantageous property pro®le including a broad nematic phase range, low rotational viscosity and high
dielectric anisotropy. The large-scale synthesis of such complex materials became feasible only by some recent advances in the synthesis of
a,a-di¯uoroether linkages. # 2001 Elsevier Science B.V. All rights reserved.
Keywords: Di¯uorooxymethylene bridge; Fluorination; Fluorodesulfuration; Liquid crystals; Mesogens; Tetra¯uoroethylene bridge
1. Introduction
effects on the application relevant physico-chemical proper-
ties of nematic liquid crystals [1±4] can be achieved by
the insertion of highly ¯uorinated linking groups into
typical mesogenic core structures, such as the phenylbicy-
clohexyl 1.
The most prominent objectives in the search for new,
superior liquid crystalline materials [1±4] for application in
active matrix liquid crystal displays 8AM-LCDs or thin-®lm
transistor, TFT-LCDs) [5] are the reduction of the rotational
viscosity 8g1), the extension of the nematic phase range and
the reduction of the driving voltage 8often by increasing the
dielectric anisotropy 8De) [6±8]).1 Another important devel-
opment target is the improvement of the ``reliability'' [1,2]
which among others includes the chemical and photoche-
mical stability of the liquid crystal, its speci®c resistivity and
its voltage holding ratio. Development objectives on the side
of the LCD manufacturers are video-compatible switching
times 8that is, less than 16.7 ms) for multimedia applica-
tions, low energy consumption for portable devices and the
reduction of production costs.
On the one hand, insertion of a tetra¯uoroethylene bridge
between two cyclohexane subunits 82) results in a dramatic
increase of the clearing temperature compared to the directly
linked analogues 81) by 50±70 K [10]. An undesirable side-
effect is a concomitant increase of the melting point and a
reduced solubility in liquid crystalline host mixtures. On the
other hand, the insertion of a di¯uorooxymethylene link
between a cyclohexane and a phenyl substructure 83) broad-
ens the nematic phase range substantially, increases the
dielectric anisotropy 8De) and reduces at the same time
the rotational viscosity 8g1) which is the parameter predo-
minantly determining the switching time of LCDs [9]
8Scheme 1).
A targeted, rational materials design is often impeded by
negative trade-off relations between the physico-chemical
parameters: in most cases an increase of the clearing tem-
perature 8e.g. by the elongation of the mesogenic core
structure) has to be ``paid'' for by a concomitant increase
of the rotational viscosity 8g1). A most notable exception of
this empirical ``rule of thumb'' are the materials containing a
di¯uorooxymethylene bridge, such as 3, where the opposite
tendency is observed. Based on this surprising ®nding, we
wanted to make use of the di¯uorooxymethylene bridge for
the compensation of the less desirable side-effects of the
clearing point increasing tetra¯uoroethylene link. A material
In our previous communications [9,10] we presented two
classes of nematic liquid crystals containing CF2O and
CF2CF2 bridges within their mesogenic core structures
which can be considered a signi®cant progress in this line
of research. The new materials demonstrate which bene®cial
* Corresponding author. Tel.: ‡49-6151-722218;
fax: ‡49-6151-722593.
E-mail address: peer.kirsch@merck.de 8P. Kirsch).
1 The anisotropy is defined as De ˆ e À e , the birefringence as
?
jj
Dn ˆ njj À n?, where || stands for parallel and ? perpendicular to the
nematic phase director, which can be approximated by the molecular
orientation axis or by the long molecular axis, respectively.
0022-1139/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 8 0 1 ) 0 0 4 8 8 - 2

70
P. Kirsch et al. / Journal of Fluorine Chemistry 112 +2001) 69±72
Scheme 1. The liquid crystalline basic structure 1. Its bridged analogues are extended by a tetrafluoroethylene 82) [10] and a difluorooxymethylene link8 3)
[9]. The combined analogue 4 contains both bridging groups.
to the dithianylium salt 17 ÁCF3SO3À [9]. The conversion to
the target compound 43 is accomplished by oxidative ¯uor-
odesulfuration [9,11±16] in the presence of 3,4,5-tri¯uor-
ophenol.
‡
containing both bridges 8such as 4) was expected to possess
a highly attractive combination of a high clearing point,
broad mesophase range, good solubility, high dielectric
anisotropy and reasonable rotational viscosity.
A comparative overview of the physical data of the target
compound 4 and its differently bridged analogues 1±3 is
given in Table 1.
2. Results and discussion
The data in Table 1 suggest, that at least for liquid crystals
based on the phenylbicyclohexyl ``backbone'' 81) the meso-
genic and viscoelastic properties can be modi®ed by the
quite independent choice of suitable bridge elements.
Roughly speaking, for compound 4 the high clearing point
8real as well as ``virtual'' TNI) can be attributed to the
tetra¯uoroethylene bridge [10], and especially the low rota-
tional viscosity 8g1) but also the broad nematic phase to the
di¯uorooxymethylene bridge [9].
In addition to the general improvement of the mesogenic
and viscoelastic properties, the substance class 4 shows a
signi®cant improvement of the so-called ``reliability'' para-
meters, such as speci®c resistivity and voltage holding ratio
[1,2]. This favorable combination of most of the applica-
tion relevant properties of 4, such as high clearing point,
broad nematic phase, low rotational viscosity, high dielec-
tric anisotropy and good reliability, shows that at least
some of the above-mentioned negative trade-off relations
can be overcome by a rational design of liquid crystalline
materials.
These properties possibly make 4 and its structural ana-
logues a potentially essential component for liquid crystal
mixtures for the next generation of fast switching ¯at-panel
displays for TV screens and portable devices capable of
displaying moving pictures, such as cellular phones, video
games, subnotebookcomputers and PDAs 8personal digital
assistants). On the synthetic side, the preparation of complex
``super ¯uorinated materials'' 8SFMs) such as 4 became
feasible only by a new synthetic route [9] allowing the easy
formation of a di¯uorooxymethylene linkage between a
The target compound 4 promises not only an interesting
combination of the physico-chemical in¯uences of the two
bridges. It is also very challenging from the synthetic point
of view containing two different ¯uorinated linking groups
either of which had proved dif®cult to synthesize in the past.
Retrosynthetic considerations 8Scheme 2) based on our
previous experience [9] suggested the following synthetic
route: the di¯uorooxymethylene linked aryl moiety is intro-
duced as the last step. This should be achievable via an
intermediate dithioorthoester from the dithianylium salt 5
which again is available from the ketone 6. As a central
intermediate for all kinds of liquid crystals based on a
di8cyclohexyl)tetra¯uoroethane substructure, the monoketal
7 was proposed. This again was expected to be accessible
from the commercially available 1,2-bis84-hydroxyphenyl)-
tetra¯uoroethane 8.2
The ®rst step towards the building block 7 is the catalytic
hydrogenation of 8, followed by the oxidation of the result-
ing diol 9 to the diketone 10. The selective monoketalization
to compound 7 was accomplished by re¯uxing a mixture of
10 and 2,2-dimethylpropane-1,3-diol in a two-phase system
of diluted sulfuric acid and cyclohexane 8Scheme 3).
The synthesis of the liquid crystal 4 from the building
block 7 is shown in Scheme 4. The side chain is introduced
by a Wittig reaction 8!11), followed by a catalytic hydro-
genation and the selective cystallization of trans-12 from
n-heptane. The ketal 12 is deprotected 8!13) and homolo-
gized via the enol ether 14 to the corresponding trans±trans
aldehyde 15. Oxidation of the aldehyde furnishes the car-
boxylic acid 16 which is condensed with 1,3-propanedithiol
3 The starting material 4 confirmed by GLC and HPLC to be
>99.8%.
2 The product 8 was supplied by Bayer AG, Leverkusen, Germany.

P. Kirsch et al. / Journal of Fluorine Chemistry 112 +2001) 69±72
71
Scheme 2. Retrosynthetic considerations for the synthesis of 4.
Scheme 3. Synthesis of the central building block 7: 8a) H2, 5% Rh-C, 2-propanol; 5 bar, 608C 876%). 8b) Pyridinium chlorochromate 8PCC), CH2Cl2; room
temperature, 18 h 879%). 8c) 2,2-Dimethylpropane-1,3-diol, cyclohexane, 0.01N H2SO4; reflux, 10 h 883%).
Scheme 4. Synthesis of 4: 8a) H7C3PPh3 BrÀ, KOtBu, THF; À108C: room temperature, 18 h 895%). 8b) 1: H2, 5% Pd±C, THF; 1 bar, room temperature; 2:
‡
‡
crystallization from n-heptane 864%). 8c) 98% HCOOH, toluene; room temperature, 18 h 876%). 8d) MeOCH2PPh3 Br± KOtBu, THF; À108C: room
temperature, 18 h 888%). 8e) 1:98% HCOOH, toluene; room temperature, 18 h; 2: MeOH, kat. NaOH; À208C: room temperature, 2 h 893%). 8f) CrO3/H2SO4
8Jones' reagent), acetone; 08C: room temperature, 18 h 868%). 8g) HS8CH2)3SH, CF3SO3H, toluene/isooctane 1:1; reflux, azeotropic removal of water 884%).
8h) 1: 3,4,5-Trifluorophenol, NEt3, CH2Cl2; À708C, 30 min; 2: NEt3Á3HF; À708C, 5 min; 3: Br2, CH2Cl2; À708C 81 h) À208C; 4: 0.5N NaOH 833%).
Table 1
Physical properties of some examples for the different liquid crystals 1±4a
Crystal
Mesophases
TNI
De
9.7
Dn
g1
1
2
3
4
C 66 N 94.1 I
C 70 SG 95 SB 102 N 168.6 I
C 44 N 105.3 I
74.7
122.0
91.5
0.075
0.060
0.067
0.074
171
269
145
250
9.0
10.5
9.4
C 49 SB 114 N 164.4 I
128.0
a ``Virtual'' clearing points 8TNI), dielectric anisotropies 8De), birefringences 8Dn) and rotational viscosities 8g1) are extrapolated from the Merckliquid
crystal mixture ZLI-4792. The temperatures are given in degree Celsius, the g1 values in milli Pascal seconds. C: crystalline, Sx: smectic X, N: nematic, I:
isotropic. The application oriented evaluation of liquid crystals for use in LCDs is centered around ``virtual'' clearing temperatures, electrooptic parameters
and viscosities. These data are obtained by extrapolation from a standardized nematic host mixture: TNI,extr, De, Dn and g1 were determined by linear
extrapolation from a 10% w/w solution in the commercially available Merckmixture ZLI-4792 8 TNI ˆ 92:88C, De ˆ 5:27, Dn ˆ 0.0964). The values thus
obtained are empirically corrected for changes in the order parameter. For the pure substances the mesophases were identified by optical microscopy, and the
phase transition temperatures by differential scanning calorimetry 8DSC).

72
P. Kirsch et al. / Journal of Fluorine Chemistry 112 +2001) 69±72
cyclohexane substructure and an aromatic ring with high
yield and under very mild conditions.
À117.1 8mc, 4F, CF2CF2), À79.3 8d, J ˆ 8:3 Hz, 2F,
‡
CF2O); MS 8EI): m=z ꢀ%† ˆ 504 [M ] 855), 356
‡
[M ÀF3PhOH] 8100), 148 872), 131 818), 81 821); HRMS
8EI) for C18H29F9O: calcd. 504.207470, exp. 504.208200.
3. Experimental
‡
À
3.1. Preparation of the dithianylium salt 17 ÁCF3SO3
Acknowledgements
A suspension of 17.4 g 843 mmol) of the carboxylic acid
16 in toluene/isooctane 84:6) was treated at room tempera-
ture with 6.0 g 855 mmol) of 1,3-propanedithiol. The mix-
ture was heated slowly to 608C, and after dropwise addition
of 8.0 g 853 mmol) of tri¯uoromethanesulfonic acid over a
time span of 5 min, re¯uxed for 3 h under azeotropic
removal of water. After cooling down to 858C, 100 ml of
dibutyl ether were added dropwise. Cooling down to 08C
under stirring precipitated the product which was ®ltrated-
off, washed twice with 20 ml of ice-cold methyl tert-butyl
ether and dried in vacuo. Yield: 20.8 g 884%) of
We thankDr. J. Krause, Dr. M. Heckmeier, Dr. G. LuÈssem
and their co-workers for the physical evaluation of the new
substances. A part of the workpresented in this article was
performed under the management of the Association of
Super-Advanced Electronics Industries 8ASET) in the
R&D program of the Japanese Ministry of International
Trade and Industry 8MITI) supported by the New Energy and
Industrial Development Organization 8NEDO).
‡
À
References
17 ÁCF3SO3 as yellowish crystals, which were used for
the following reaction step without further puri®cation.
[1] P. Kirsch, M. Bremer, Angew. Chem. 112 82000) 4384±4405.
[2] P. Kirsch, M. Bremer, Angew. Chem. Int. Ed. 39 82000) 4216±4235.
[3] S.M. Kelly, Flat Panel Displays: Advanced Organic Materials, RSC
Materials Monographs, The Royal Society of Chemistry, Cambridge,
2000.
3.2. Preparation of the liquid crystal 4
‡
À
A suspension of 7.0 g 812 mmol) of 17 ÁCF3SO3 in
80 ml of CH2Cl2 was cooled to À708C. Then, a mixture
of 3 ml 822 mmol) of triethylamine and 2.22 g 815 mmol) of
3,4,5-tri¯uorophenol was added dropwise, followed after 1 h
by 20 ml 8124 mmol) of NEt3Á3HF and, after another
10 min, by 10.0 g 863 mmol) of bromine in 30 ml of CH2Cl2
over a time span of 30 min. After stirring for 90 min at
À708C, the mixture was allowed to warm up to À208C and
poured into a stirred ice-cold mixture of 350 ml 1N NaOH
and 35 ml of saturated aqueous NaHSO3. The organic phase
was separated, and the aqueous phase was extracted twice
with 80 ml of n-pentane. The combined organic phases were
dried over Na2SO4 and evaporated to dryness. The semi-
solid residue was dissolved in n-hexane and ®ltrated over a
short silica gel column. A further puri®cation was achieved
by ¯ash-chromatography 8n-hexane; silica gel), followed by
two recrystallizations from n-pentane. Yield: 2.0 g 833%) of
4 as colorless crystals. For phase transitions see Table 1; 1H
NMR 8250 MHz, CDCl3, 303 K): d ˆ 6:82 8mc, 2H, ar-2,6-
H), 2.15À1.77 8m, 12H), 1.48À1.12 8m, 12H), 0.90 8t,
J ˆ 6:6 Hz, 3H); 19F NMR 8235 MHz, CDCl3, 303 K):
d ˆ À164:8 8mc, 1F, ar-4-F), À133.4 8mc, 2F, ar-3,5-F),
[4] D. Demus, J. Goodby, G.W. Gray, H.-W. Spiess, V. Vill 8Eds.),
Handbookof Liquid Crystals, Wiley-VCH, Weinheim, 1998.
[5] S. Kobayashi, H. Hori, Y. Tanaka, in: P.J. Collings, J.S. Patel 8Eds.),
Active Matrix Liquid Crystal Displays in Handbookof Liquid
Crystal Research, Oxford University Press, New York, 1997,
pp. 415±444.
[6] W. Maier, G. Meier, Z. Naturforschg. 16a 81961) 262±267.
[7] D. Demus, G. Pelzl, Z. Chem. 21 81981) 1±9.
[8] W.H. de Jeu, Physical Properties of Liquid Crystalline Materials,
Gordon & Breach, 1980.
[9] P. Kirsch, M. Bremer, A. Taugerbeck, T. Wallmichrath, Angew.
Chem. 113 82001) 1528±1532; Angew. Chem. Int. Ed. 40 82001)
1480±1484.
[10] P. Kirsch, M. Bremer, F. Huber, H. Lannert, A. Ruhl, M. Lieb, T.
Wallmichrath, J. Am. Chem. Soc. 123 82001) 5414±5417.
[11] J. Kollonitsch, S. Marburg, L.M. Perkins, J. Org. Chem. 41 81976)
3107±3111.
[12] S.C. Sondej, J.A. Katzenellenbogen, J. Org. Chem. 51 81986) 3508±
3513.
[13] G.K.S. Prakash, D. Hoole, V.P. Reddy, G.A. Olah, Synlett 81993)
691±693.
[14] M. Kuroboshi, T. Hiyama, J. Fluorine Chem. 69 81994) 127±128.
[15] R.D. Chambers, G. Sandford, M.E. Sparrowhawk, M.J. Atherton, J.
Chem. Soc., Perkin Trans. 1 81996) 1941±1944.
[16] C. York, G.K.S. Prakash, G.A. Olah, Tetrahedron 52 81996) 9±14.