Synthesis, structure, and properties of 4,8,12-trioxa-12c-phospha-4,8,12,12c-tetrahydrodibenzo[cd,mn]pyrene, a molecular pyroelectric

Article abstract of DOI:10.1021/ja962023c

The title compound (4) was synthesized, and its crystalline structure was determined. The molecule has C(3v) point symmetry and crystallizes in the trigonal space group R3m. Crystal data for 4: a = 16.6710(13) ?, b = 16.6710(13) ?, c = 4.2590(3) ?, α = β

Full text of DOI:10.1021/ja962023c

1208
J. Am. Chem. Soc. 1997, 119, 1208-1216
Synthesis, Structure, and Properties of
4,8,12-Trioxa-12c-phospha-4,8,12,12c-tetrahydrodibenzo-
[cd,mn]pyrene, a Molecular Pyroelectric
Frederik C. Krebs,* Peter S. Larsen, Jan Larsen, Claus S. Jacobsen,†
Carlo Boutton,‡ and Niels Thorup§
Contribution from the Macromolecular Chemistry Group, Department of Solid State Physics,
Risø National Laboratory, DK-4000 Roskilde, Denmark
ReceiVed June 17, 1996. ReVised Manuscript ReceiVed NoVember 5, 1996X
Abstract: The title compound (4) was synthesized, and its crystalline structure was determined. The molecule has
C3V point symmetry and crystallizes in the trigonal space group R3m. Crystal data for 4: a ) 16.6710(13) Å, b )
16.6710(13) Å, c ) 4.2590(3) Å, R ) â ) 90°, γ ) 120°, Z ) 3, R(F) ) 0.0234. The material has a permanent
polarization and consequent pyroelectric properties. The room temperature pyroelectric coefficient was found to be
-3 ( 2 µC m-2 K-1, which is in accordance with a calculated value of -3.2 µC m-2 K-1. The molecular dipole
moment was determined to be 3.3 ( 0.2 D, the direction of which was unambiguously assigned with respect to the
molecular coordinates. The thermal expansivity was determined at temperatures in the range -93 to 200 °C. The
relative dielectric permittivity tensor was obtained at optical frequencies (ꢀ11 and ꢀ22 ) 3.16 and ꢀ33 ) 2.48) and in
the microwave region at 35 GHz (ꢀ11 and ꢀ22 ) 5.2 ( 0.6 and ꢀ33 ) 2.9 ( 0.2), and at low frequencies (120 Hz and
1 kHz), the isotropic permittivity was determined (ꢀ120 Hz ) 4.7 ( 0.8 and ꢀ1 kHz ) 4.7 ( 1.1). Finally, an estimate
of the molecular heat capacity was calculated (Cp ) 900 J Kg-1 K-1) and the material was considered for potential
use in infrared detection as its detectivity merit factor, Mr, was determined (Mr ) 8.8 × 10-2 m2 C-1).
Introduction
mers,5,6 and liquid crystals.36 Various approaches have been
employed in attempting to control the nature of the pyroelectric
effect depending on the area of chemistry from which the
pyroelectric material stems. The majority of known pyroelectric
materials fall within the category of inorganic/ceramic materials,
and the approach to control the pyroelectric effect in this respect
has been by doping and varying composition without affecting
the structure3,31,32 and by development of solid solution sys-
tems,35 and also, the effect of the crystal growth method has
been investigated.34 The inorganic/organic salts are typically
(14) Lang, S. B.; Cohen, M. D.; Steckel, F. J. Appl. Phys. 1965, 36,
3171-3173.
(15) Gavrilova, N. D.; Drozhdin, S. N.; Novik, V. K.; Maksimov, E. G.
Solid State Commun. 1983, 48, 129-133.
(16) Asaji, T.; Weiss, A. Z. Naturforsch. 1985, 40a, 567-574.
(17) Fleck, S.; Weiss, A. Z. Naturforsch. 1987, 42a, 645.
(18) Asaji, T.; Taya, M.; Nakamura, D. Phys. Status Solidi A 1987, 102,
815-818.
(19) Zheludev, I. S.; Gladkii, V. V. SoV. Phys. Crystallogr. 1966, 11,
366-368.
(20) Simhony, M.; Shaulov, A. J. Appl. Phys. 1971, 42, 3741.
(21) Unruh, H. -G. Solid State Commun. 1970, 8, 1951-1954.
(22) Nomura, S. J. Phys. Soc. Jpn. 1961, 16, 2440-2452.
(23) Lang, S. B. Phys. ReV. B 1971, 4, 3603-3609.
(24) Ackermann, W. Ann. Phys. 1915, 46, 197-230.
(25) Gladkii, V. V.; Zhedulev, I. S.; Sidnenko, E. V. Bull. Acad. Sci.
USSR, Phys. Ser. 1969, 33, 276-281.
(26) Novik, V. K. SoV. Phys. Crystallogr. 1965, 10, 89-90.
(27) Sawada, S.; Nomura, S.; Asao, Y. J. Phys. Soc. Jpn. 1961, 16, 2207-
2212.
(28) Ro¨ntgen, W. C. Ann. Phys. 1914, 45, 737-800.
(29) Savage, A. J. Appl. Phys. 1966, 37, 3071-3072.
(30) Heiland, G. Ibach, H. Solid State Commun. 1966, 4, 353-356.
(31) Liu, S. T.; Heaps, J. D.; Tufte, O. N. Ferroelectrics 1972, 3, 281-
285.
(32) Liu, S. T.; Kyonka, J. Ferroelctrics 1974, 7, 167-169.
(33) Thatcher, P. D. J. Appl. Phys. 1968, 39, 1996-2002.
(34) Wemple, S. H.; Didomenico Jr., M.; Camlibel, I. J. Phys. Chem.
Solids 1968, 29, 1797-1803.
(35) Iwasaki, H.; Miyazawa, S.; Koizumi, H.; Sugii, K.; Niizeki, N. J.
Appl. Phys. 1972, 43, 4907-4915.
(36) Tournilhac, F.; Blinov, L. M.; Simon, J.; Subachius, D. B.;
Yablonsky, S. V. Synth. Met. 1993, 54, 253-261.
Materials exhibiting pyroelectric properties are used as the
active components in infrared detectors,1-4 in infrared video
cameras,5,6 in pyroelectric image tubes,7 and in the formation
of pyro-electrooptic transient gratings,8 to mention only a few
areas of application. From a materials science point of view,
there is thus an interest in developing new materials where better
performance can be achieved with respect to detectivity, ease
of incorporation in devices, compatibility with known technol-
ogy, etc. There is however also a fundamental interest in
obtaining materials in which the nature of the pyroelectric effect
can be controlled. This essentially implies the development of
a deeper understanding of the phenomenon. The pyroelectric
effect itself is observed in molecular materials,9-18 inorganic/
organic salts,19-27 inorganic/ceramic materials,20,25-35 poly-
† Department of Physics, Technical University of Denmark, DK-2800
Lyngby, Denmark.
‡ Laboratorium voor Chemische en Biologische Dynamica, Celestijnen-
laan 200D, B-3001 Heverlee, Belgium.
§ Department of Chemistry, Technical University of Denmark, DK-2800
Lyngby, Denmark.
X Abstract published in AdVance ACS Abstracts, January 15, 1997.
(1) Glass, A. M. Appl. Phys. Lett. 1968, 13, 147-149.
(2) Byer, R. L.; Roundy, C. B. Ferroelectrics 1972, 3, 333-338.
(3) Glass, A. M. J. Appl. Phys. 1969, 40, 4699-4713.
(4) Liu, S. T.; Maciolek, R. B. J. Electron Mater. 1975, 4, 91-100.
(5) Dubois, J. C. Mol. Cryst. Liq. Cryst. 1994, 255, 231-242.
(6) Robin, P.; Facoetti, J. C.; Dubois, C.; Dautriche, P.; Pourquier, E.
Onde Electr. 1994, 74, 34-36.
(7) Taylor, R. G.; Boot, A. H. Contemp. Phys. 1973, 14, 55-87.
(8) Ducharme, S. Optics Lett. 1991, 16, 1791-1793.
(9) Koptsik, V. A. SoV. Phys. Crystallogr. 1960, 4, 197-200.
(10) Gavrilova, N. D. SoV. Phys. Crystallogr. 1965, 10, 91-92.
(11) Sworakowski, J. Molecular Electronics; Ashwell, G. J., Ed.;
Research Studies Press Ltd.: John Wiley & Sons, England, 1992; pp 266-
331.
(12) Terauchi, H.; Kojima, T.; Sakaue, K.; Tajiri, F. J. Chem. Phys. 1982,
76, 612-615.
(13) Giermanska, J.; Nowak, R.; Sworakowski, J. Ferroelectrics 1985,
65, 165.
S0002-7863(96)02023-9 CCC: $14.00 © 1997 American Chemical Society

InVestigation of a Molecular Pyroelectric
J. Am. Chem. Soc., Vol. 119, No. 6, 1997 1209
salts of organic or inorganic acids, where the approach has been
to observe the effect of exchanging the cations,24 but also, the
effect of experimental conditions like crystal growth temperature
has been investigated.37 Within the polymeric pyroelectrics,
the approach has been to substitute halogen atoms and copoly-
merize similar monomers.5,6,38 The liquid crystalline approach
has been to control chirality and composition of binary mixtures
of polyphillic molecules.36 Up until 1983 few purely molecular
organic pyroelectric materials were known and character-
ized;9,10 from this date, however, systematic studies on the
pyroelectricity of molecular materials began11 and quite a few
molecular pyroelectrics are known and characterized today.12-18
Many of these materials stem from meta-substituted benzene
derivatives13-18 as these derivatives have a pronounced tendency
to crystallize in a polar space group.39 In order for a molecular
material to exhibit pyroelectric properties, the molecule itself
must have a permanent dipole moment and it must crystallize
in one of the 10 pyroelectric point groups.40 While most organic
molecules possess a permanent dipole moment, few crystallize
in one of the pyroelectric point groups.41,42 We have synthesized
and crystallized the molecule 4 which differs from most other
molecular pyroelectrics by having a high molecular symmetry
and not showing strong intermolecular interactions as evidenced
by sublimation at temperatures just above room temperature.
The molecular material 4 investigated in this paper is built from
this simple symmetrical molecule with a particular shape that
implies the presence of a permanent dipole moment and, at the
same time, allows the molecules to stack, the consequence of
which may be reflected in the exceptionally clear and simple
crystal structure that enables one to see directly that the material
is polarized. The molecular dipole moment vector coincides
with the molecular C3 axis, which again coincides with the polar
axis of the crystal. The nature of the pyroelectric effect as
observed from 28 to 110 °C can be developed by the combina-
tion of a measurable molecular property, the dipole moment,
and the crystalline structural properties, the molecular arrange-
ment and the thermal expansivity.
When considering a material for use as the active component
in a detector system, it is important that its ability to detect a
given signal can be expressed in a manner such that a
comparison with other materials is possible. The way to achieve
this is by deriving a merit factor1-7,31 which takes into account
certain physical properties pertaining to the material and some
physical factors pertaining to the geometry and construction of
the detector in question. The physical factors which depend
solely upon the material are the dielectric constant, the density,
the heat capacity, the themal conductivity,7 the loss tangent,4,31
and of course the pyroelectric coefficient. The physical factors
which depend upon the geometry and construction are the
thickness and the area of the detector material in the actual
detector system, the assocciated electrical capacitance2,3 of the
detector crystal with electrodes, and the frequency at which
detection is done. The physical factors having an influence on
the magnitude of the merit factor therefore depend upon the
disposition of the chosen detector system, for instance, whether
it is time dependent or time independent, whether potential or
charge is read,5,6 or whether it is a low-noise detector system.7
A large detector response to a given energy input is not solely
granted by a large pyroelectric coefficient. Other factors which
interplay are the density, dielectric constant along the polar axis,
and the heat capacity, which all have to be small in value to
get a large response. This is because the magnitude of the
detector response is directly proportional to the pyroelectric
coefficient but inversely proportional to the latter three quanti-
ties. The main advantage of molecular materials in this respect
is that even though their pyroelectric coefficient often is only
moderate in magnitude the value of their relative dielectric
constant is rarely above 10.40 Many inorganic/ceramic materials
have large pyroelectric coefficients but often relative dielectric
constants with values in excess of several hundred.40 The
pyroelectric coefficient and the dielectric constant therefore often
balance each other with respect to the magnitude of the
detectivity merit factor. Another advantage of molecular
materials is of course their low density compared to that of most
inorganic/ceramic materials. The heat capacity of molecular
materials however is usually comparable to slightly higher in
value than the corresponding inorganic/ceramic materials. In
this paper a time and physical disposition independent detectivity
merit factor, Mr, for compound 4 was evaluated so that the
material could be compared with other known materials without
considering how an actual detector system should be constructed.
Experimental Section
Synthetic Methods and Materials. All reagents used were standard
grade unless otherwise mentioned. THF was dried by distillation from
Na/benzophenone under argon. NMP was dried by distillation from
CaH2 under vacuum. NMR spectra were recorded on a 250 MHz
BRUKER NMR spectrometer. 31P-NMR spectra were recorded with
a capillary containing 85% H3PO4(aq) as a reference. Melting points
were determined using a BU¨ CHI melting point apparatus or if stated
determined by DSC on a Perkin Elmer DSC-4. Mass spectra were
determined using a JEOL JMS HX110-110T mass spectrometer.
Elemental analysis was done at the University of Copenhagen,
Department of Chemistry, Elemental Analysis Laboratory, Univer-
sitetsparken 5, 2100 Copenhagen, Denmark.
1-[2-Tetrahydropyranyl)oxy]-3-fluorobenzene (1). 3-Fluorophe-
nol (75 g, 0.669 mol) and dihydropyran (79 g, 0.928 mol) were mixed
in a 500 mL conical flask while being cooled to a temperature of -35
°C in a CO2(s)/acetone bath. The cooling was removed, and the solution
was stirred while 37% HCl(aq) (1 mL) was added. The temperature
was allowed to rise while the solution was stirred. After 15 min, the
reaction took place, and the mixture acquired room temperature and
was stirred for one-half hour, during which crystallization occurred.
Ether (500 mL) was added, and the organic layer was washed with 1
M NaHCO3(aq) (200 mL) and dried (Na2SO4). The ether was
evaporated on a rotary evaporator, and the crude oil was crystallized
by dissolution in boiling hexane (200 mL) followed by cooling. This
1
gave 1 (110 g, 83%) as colorless crystals: mp 47-48 °C; H-NMR
(CDCl3) δ 7.26-7.14 (m, 1H), 6.82 (td, 1H, J1 ) 6.0 Hz, J2 ) 2.2 Hz,
J3 ) 0.6 Hz), 6.79 (dt, 1H, J1 ) 11.3 Hz, J2 ) 2.2 Hz), 6.67 (tdd, 1H,
J1 ) 8.5 Hz, J2 ) 2.5 Hz, J3 ) 1 Hz,), 5.38 (t, 1H, J ) 3.1 Hz),
3.93-3.82 (m, 1H), 3.65-3.55 (m, 1H), 2.1-1.5 (m, 6H); 13C-NMR
(CDCl3) δ 163.91 (d, J ) 245.2 Hz), 158.81 (d, J ) 10.9 Hz), 130.43
(d, J ) 9.7 Hz), 112.54 (d, J ) 3 Hz), 108.69 (d, J ) 21.2 Hz), 104.56
(d, J ) 24.8 Hz), 96.95 (s), 62.42 (s), 30.63 (s), 25.52 (s), 19.06 (s)
Anal. Calcd for C11H13FO2: C, 67.33; H, 6.67. Found: C, 67.20; H,
6.67.
Tris[[(2-tetrahydropyranyl)oxy]-6-fluorophenyl]phosphine (2). 1
(12 g, 61.1 mmol) was dissolved in dry THF (250 mL). The solution
was cooled to -78 °C on a CO2(s)/acetone bath under argon. n-BuLi
(1.6 M, 38.2 mL, 61.1 mmol) was added, and the reaction mixture
was stirred for one-half hour. PBr3 (1.78 mL, 19 mmol) was added.
After the mixture stirred for 1 h, the reaction was quenched with MeOH
(5 mL) before the temperature was allowed to rise (NOTE: a rise in
temperature above -50 °C before quenching leads to elimination and
formation of colored products). The mixture was poured into brine
(200 mL) and extracted with ether (500 mL). The organic phase was
(37) Eisner, J. Ferroelectrics 1972, 4, 213-219.
(38) Wada, Y.; Hayakawa, R. Jpn. J. Appl. Phys. 1976, 15, 2041-2057.
(39) Curtin, D. Y.; Paul, I. C. Chem. ReV. 1981, 81, 525-541.
(40) Liu, S. T. Landolt-Bo¨rnstein, Numerical Data and Functional
Relationships in Science and Technology; Hellwege, K.-H., Ed.; Springer-
Verlag: New York, 1979; Vol. 11/III, pp 471-494.
(41) Mighell, A. D.; Rodgers, J. R. Acta. Crystallogr. 1980, A36, 321-
326.
(42) Wilson, J. C. Acta. Crystallogr. 1988, A44, 715-724.

1210 J. Am. Chem. Soc., Vol. 119, No. 6, 1997
Krebs et al.
Table 1. Crystallographic Data for Phosphangulene (4)
dried (Na2SO4), and the solvents were evaporated on a rotary evaporator.
The semicrystalline material was crystallized from boiling MeOH (500
mL) which gave 2 (8.23 g, 65%) as a colorless compound that was a
mixture of isomers: mp 145-147 °C; 1H-NMR (δ ppm, CDCl3) 7.24-
7.1 (m, 1H), 6.98-6.84 (m, 1H), 6.65-6.50 (m, 1H), 5.44-5.27 (m,
1H), 3.88-3.66 (m, 1H), 3.58-3.45 (m, 1H), 1.9-1.2 (m, 6H); 31P-
NMR (δ ppm, CDCl3) -70 to -72 (m) Anal. Calcd for C33H36PF3O6:
C, 64.27; H, 5.88. Found: C, 64.36; H, 5.73.
formula
formula wt
crystal system
space group
Z
a, Å
b, Å
C18H9PO3
304.22
trigonal
R3m
3
16.6710(13)
16.6710(13)
4.2590(3)
90
120
1025.09(13)
1.478
c, Å
Tris(2-hydroxy-6-fluorophenyl)phosphine (3). (2 (8 g, 13 mmol)
was stirred in MeOH (800 mL) containing 37% HCl(aq) (16 mL) under
argon. After 4 h the solution was evaporated to a volume of 20 mL.
The mixture was then stirred with CHCl3 (300 mL) and 1 M
NaHCO3(aq) (300 mL) under argon (NOTE: this is very important as
the hydrochloride is the initial product). The clear organic phase was
separated from the clear aqueous phase and dried (Na2SO4), followed
by evaporation of the solvent. Crystallization from boiling toluene (100
mL) followed by careful addition of heptane (200 mL) under argon.
The product was left to cryrstallize in the freezer. This gave 3 (4.32
g, 91%) as colorless crystals which were dried in a vacuum oven at 80
R, â
γ
V, Å3
F
calc, g cm-3
crystal dimens, mm
T, K
0.4 × 0.05 × 0.05
293(2)
type of radiation
Cu KR
µ, mm-1
1.877
R(F), Rw(F2) all data
GOF (on F2)
0.0234, 0.0570
1.057
1
°C: mp 176.5-177 °C; H-NMR (δ ppm, CDCl3) 7.4-7.2 (m, 1H),
6.8-6.6 (m, 1H), 5.1 (s, 1H); 13C-NMR (δ ppm, CDCl3) 165.4 (dd,
JC-F ) 245 Hz, JC-P ) 5.8 Hz), 161.0-160.6 (m), 130.6-130.4 (m),
110.46 (s), 109.31 (s), 106.39 (d, J ) 25.8 Hz); 31P-NMR (δ ppm,
CD3OD) -80 (q, JP-F ) 26 Hz). Anal. Calcd for C18H12PF3O3: C,
59.34; H, 3.32. Found: C, 59.94; H, 3.55.
function having a Gauss/Lorenz ratio of 0.8. The zero point was fixed
by assigning the 110 peak a 2θ value obtained from the calculated
powder diffractogram. The following peaks were used in unit cell size
determination: 110, 300, 220, 101, 021, 211, 330, 600, 520.
Measurement of the Dipole Moment. The dipole moment was
determined by measuring the capacitance at 21 °C of chloroform
4,8,12-Trioxa-12c-phospha-4,8,12,12c-tetrahydrodibenzo[cd,mn]-
pyrene (4) (Phosphangulene). 3 (0.75 g, 2.06 mmol) was placed in
a dry 50 mL round-bottomed flask under argon. t-BuOK (95%) (0.8
g, 6.18 mmol) was added, followed by dry NMP (20 mL). The solution
acquired a slight tan color. The flask was then placed on an oil bath
(200 °C) and stirred for 1 h, during which a precipitate formed and the
color became dark brown. After cooling to room temperature, the
mixture was poured into ether (200 mL), and the organic phase was
then washed with water (2 × 150 mL). The organic layer was colorless,
and all the colored products remained in the aqueous phase. The
organic phase was dried (Na2SO4) and concentrated in vacuum to yield
the pure phosphangulene with some NMP. The mixture was dissolved
in boiling EtOAc (20 mL) and left to crystallize. Well-formed needles
of 4 (0.54 g, 86%) were obtained (the product can be sublimed at 160
°C and 0.1 mmHg or at 200 °C and atmospheric pressure; in both cases,
it takes 1 h): DSC showed melting in the interval 283-298 °C peak
solutions of 4 with concentrations in the range 0.01-0.001 mol L-1
.
The measurement of the capacitance was accomplished with an
impedance analyzer from Hewlett-Packard (type 4192A LF). The
dipole moment was obtained using the Onsager equation43 and was
found to be 3.3 ( 0.2 D.
Heat Capacity. The heat capacity of the crystal was estimated,
from the calculated heat capacity for the gas phase molecule, to be Cp
) 900 J kg-1 K-1. The value for the solid is expected to be a little
smaller than this value. The free molecule (gas phase) heat capacity
was calculated with MOPAC and PM3.44 Calculations were performed
on a Power Iris Indigo and a Power Challenger from SGI.
Measurement of Dielectric Constants at Optical Frequencies.
The relationship between the refractive indices and the relative
permittivity was used to determine the values for the components of
the dielectric permittivity tensor at optical frequencies. The material
is uniaxial as is shown by the crystal structure. The refractive index
n3, for white light, was determined by means of a microrefractometer
spindle-stage, using calcite as the refractometer crystal,45 and found to
be 1.575 ( 0.002. The refractive indices n1 ) n2, are much higher
than n3; the crystal is thus optically negative like calcite. These
components could not be determined directly because the crystals
dissolved too rapidly in the refractometer oils available. In R-bromo-
naphthalene, however, the crystals dissolved at a rate sufficiently slow
to allow interpolation in a sequence of microrefractometric measure-
ments.46 The path difference between the liquid and the crystal seen
along the c axis gave the grain thickness, and the path difference
between the liquid and the crystal seen perpendicular to the c axis then
gave n1,n2 - 1.657, which gives n1,n2 ) 1.78 ( 0.01 for white light.
The birefringence is thus 0.20. Using
at 290.44 °C, 29.95 KJ mol-1; mass spectra (EI+) 304 m/z; H-NMR
1
(δ ppm, CDCl3) 7.31-7.23 (m, 1H), 7.1-7.04 (m, 2H); 13C-NMR (δ
ppm, CDCl3) 156.66 (d, J ) 7.8 Hz), 130.37 (s), 115.29 (d, J ) 1.2
Hz), 114.97 (d, 12.7 Hz); 31P-NMR (δ ppm, CDCl3) -133 (s). Anal.
Calcd for C18H9PO3: C, 71.05; H, 2.98. Found: C, 70.95; H, 3.00.
4,8,12-Trioxa-12c-oxyphospha-4,8,12,12c-tetrahydrodibenzo-
[cd,mn]pyrene (5). 4 (0.15 g, 0.5 mmol) was dissolved in CHCl3 (50
mL) and stirred vigorously overnight with 35% H2O2(aq) (4 mL) in a
conical flask (100 mL) using a magnetical stirrer. The organic phase
was separated and dried (Na2SO4). Evaporation and recrystallization
from EtOAc gave 5 (0.1 g, 63%) as colorless crystals: mp 310-311
°C; mass spectra (EI+) 320 m/z; 1H-NMR (δ ppm, CDCl3) 7.47 (t, 1H,
J ) 7.8 Hz), 7.18 (dd, 2H); 13C-NMR (δ ppm, CDCl3) 160.31 (s),
133.89 (s), 115.38 (d, J ) 4.8 Hz), 110.32 (d, J ) 108.4 Hz); 31P-
NMR (δ ppm, CDCl3) -48(s) Anal. Calcd for C18H9PO4: C, 67.50;
H, 2.83. Found: C, 67.04; H, 2.46.
n )
ꢀ
(1)
x
i
r,i
Statistical Analysis. All numerical results given with an indication
of experimental error represent
x ( 2σ, where x represents the
The relative dielectric permittivity tensor at optical frequencies can thus
be written as
population mean and 2σ represents two standard deViations.
Crystallographic Methods. The crystal structure of 4 was deter-
mined by X-ray crystallography using direct methods and least-squares
refinement. Hydrogen atoms were found close to the observed positions
and included in the final refinements. Data were collected on an Enraf
Nonius CAD4 four-circle diffractometer. The absolute structure was
confirmed by a Flack parameter of -0.01 ( 0.1. Crystallographic data
are given in Table 1.
3.16 0
0
ꢀoi,pj tical ) 0
3.16 0
0 2.48
(2)
(
)
0
(43) Bo¨ttcher, C. J. F. Theory of Electric Polarization, 2nd ed.;
Elsevier: Amsterdam, 1973; Vol. 1, pp 159-204.
(44) (a) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 221. (b)
Computational results obtained using the MOPAC program from MSI of
San Diego, CA.
Powder diffraction was done using a STOE powder diffractometer
scanning the interval 8° e 2θ e 40° at a rate of 0.8° 2θ/min at 10 °C
intervals in the range 30-200 °C (it was not possible to elevate the
temperature above 200 °C as the material sublimed at a rate too fast to
complete the scan). The peaks were fitted using a pseudo Voigt
(45) Medenbach, O. Fortschr. Mineral. 1985, 61, 111-113.
(46) Beyer, H. Theorie und praxis der interferenzmikroskopie; Aka-
demische Verlagsgesellschaft; Geest & Portig K.-G.: Leipzig, 1974.

InVestigation of a Molecular Pyroelectric
J. Am. Chem. Soc., Vol. 119, No. 6, 1997 1211
At 35 GHz. The components of the relative dielectric permittivity
tensor were measured by the microwave cavity perturbation method.47,48
The Q-factor of a resonant cavity was measured when a crystal was
placed in the resonant cavity with the crystallographic axis of interest
aligned with the electrical field vector in the cavity. The frequency
shift observed upon introducing the sample into the cavity is related to
the dielectric constant in the direction aligned with the electric field as
well as the appropriate filling and depolarization factors, which are
estimated from the sample dimensions. The dielectric constants could
Scheme 1
thus be determined. The components were found to be ꢀ11 and ꢀ22
5.2 ( 0.6 and ꢀ33 ) 2.9 ( 0.2.
)
At 120 Hz and 1 kHz. The low-frequency isotropic relative
permittivity was obtained by measuring the capacitance of a series of
discs of the material prepared by compressing finely ground 4 under
vacuum with a mechanical pressure of 400 kg m-2 to a density of 1425
( 78 kg m-3
. The diameter of the discs were 13 mm, and their
thicknesses varied from 0.1 to 0.4 mm. The values obtained were ꢀ120 Hz
) 4.7 ( 0.8 and ꢀ1 kHz ) 4.7 ( 1.1.
Pyroelectric Measurements. Several methods have been described
for the measurement of the pyroelectric coefficients.24,40,49-55 Since
we were dealing with relatively small single crystals (1 × 0.4 × 0.4
mm3), the only method that worked well was the temperature step
method.24 When subjecting the crystal to a step change in temperature,
while measuring the potential appearing across a known capacitance,
the pyroelectric coefficient can be obtained at any given temperature
as the slope of the curve (or as the change in potential upon stepping)
according to
Ctotal ∆V
∆P
∆T
T
) p3
)
(3)
A
∆T
T
P being the polarization, p3 the stress free pyroelectric coefficient, A
the electrode area, Ctotal the total capacitance, V the voltage, and T the
temperature. The charge generated was obtained by measuring the
potential appearing across a capacitor (220 pF) placed in parallel with
the sample. The potential measured across the sample and capacitor
was buffered using a null configured electrometer operational amplifier
AD549JH from Analog Devices. The temperature was controlled by
means of a peltier element (MI1020T (0.9 W) from Marlow Industries
Inc.) thermally connected to a heat sink (3.5 K W-1) with heat
conducting glue stuck to one side. The other side had a copper coating
(0.01 mm) and was covered with a thin slice of mica (0.05 mm). Silver
paste electrodes (Leitsilber 200 from DEMETRON) were applied to
the crystal perpendicular to the c axis. Electrical connection to the
null amplifier circuit was made with silver wire (0.05 mm Ø). The
sample was placed in transistor heat transfer silicon grease on the mica-
covered side of the peltier element. The sample/amplifier assembly
was enclosed in a gold-plated brass box with gas inlet/outlet. The
temperature was measured with an Al-Ni/Ni-Cr thermocouple. Data
acquisition was achieved using a 12 bit A/D converter with digital
output lines AX5210 from Axiom Inc. The digital output lines were
used to control the temperature through a D/A converter (DAC0808LCN
from National Semiconductor), driving the peltier element by means
of a power operational amplifier (L165V from S G S Thompson).
Calibration of Pyroelectric Measurement Setup. In order to
ensure that the measurements were correct the pyroelectric coefficients
of reference samples were determined. Five LiNbO3 samples with a
known direction of polarization were used as reference materials. The
samples were cut from plates of the material having a thickness of 0.5
mm with the c axis along the 0.5 mm dimension. Dimensions of
individual samples typically measured 2 × 2 × 0.5 mm3 with the c
axis perpendicular to the 2 × 2 mm2 surface. The positive direction
of polarization was known and connected to signal ground. When the
temperature was raised the voltage which developed across the material
had a positive sign with respect to signal ground. The capacitance of
the samples had values of around 8.5 pF and the reference capacitor
placed in parallel with the reference samples had a value of 15 nF.
When the pyroelectric coefficient was measured as described in the
section above, a value of -39 ( 3 µC m-2 K-1 was obtained at a
temperature of 28 °C. This value is in good agreement with the literaure
value29 of -40 µC m-2 K-1 at a temperature of 25 °C.
Thermogravimetric Measurements. The weight of a 6.603 mg
sample was measured in the temperature range 20-200 °C. The rate
of temperature change was set to 1 °C/min. At this rate of change of
temperature, the weight loss became detectable at 127 °C. In order to
show that sublimation took place at lower temperatures where the
pyroelectric measurements were made, the weight of a 21.296 mg
sample placed at a constant temperature of 100 °C was monitored over
a 15 h period of time. During this period, 0.021 mg evaporated,
corresponding to a rate of sublimation of 1.1 × 10-4 % (weight) min-1
at 100 °C. The measurements were made on a TGA51 thermogravi-
metric analyzer in conjunction with a Thermal Analyst 2000 from
Dupont Instruments.
(47) Buranov, L. I.; Shchegolev, I. F. Probl. Tekh. Eksp. 1971, 2, 171.
(48) Ong, N. P. J. Appl. Phys. 1977, 48, 2935-2940.
(49) Fabel, G. W.;Henisch, H. K. Solid State Electron. 1971, 14, 1281-
1283.
(50) Hartley, N. P.; Squire, P. T.; Putley, E. H. J. Phys. 1972, E5, 787-
789.
(51) Gladkii, V. V.; Zheludev, I. S. SoV. Phys. Crystallogr. 1965, 10,
50-53.
Results and Discussion
(52) Chynoweth, A. G. J. Appl. Phys. 1956, 27, 78-84.
(53) Lang, S. B.; Steckel, F. ReV. Sci. Instrum. 1965, 36, 929-932.
(54) Shaulov, A.; Simhony, M. Appl. Phys. Lett. 1972, 20, 6-7.
(55) Gavrilova, N. D. SoV. Phys. Crystallogr. 1965, 10, 278-281.
Compound 4 is a phosphine derived from triphenylphosphine.
Its synthesis is fairly straightforward (Scheme 1), starting from
commercially available 3-fluorophenol by reaction with 2,3-

1212 J. Am. Chem. Soc., Vol. 119, No. 6, 1997
Krebs et al.
as a measure of the amount of charge generated as the material
underwent a step change of temperature (see the Experimental
Section). This type of measurement results in an average value
T
of the pyroelectric coefficient, p3 , over the temperature step
region. The magnitude of the pyroelectric effect depends upon
some of the physical properties of the molecule and of the
material. The molecular dipole is responsible for the polariza-
tion. Since the volume of the crystal is unconfined during the
course of the experiments, the pyroelectric coefficient obtained
here is the pyroelectric coefficient at constant stress. It is thus
composed of two different pyroelectric contributions40
Figure 1. Molecular structure of phosphangulene (4), which is
nonplanar, and the corresponding carbon compound, which in its
cationic form is planar.
E
pnT ) pnS + enµT Rµ
(4)
where p, n, T, S, E, e, µ, and R represent the pyroelectric
coefficient, the polar axis, constant stress, constant strain,
constant electric field, piezoelectric stress coefficient, tensor
components, and thermal expansion constant, respectively. The
second term represents the change in polarization per unit change
of volume with temperature, which is due to the thermal
expansion. This is known as the secondary pyroelectric
coefficient. The first term is the primary pyroelectric coefficient,
and when observed, this term is due to molecular or atomic
movements within the crystal, leading to partial or full cancel-
lation of the dipole moments. This does not necessarily lead
to nor depend upon any change of volume of the unit cell. In
both cases, the change in polarization leads to the generation
of charges at the faces of the crystal perpendicular to the polar
axis. When, in turn, the charge generation is considered across
a resistance and a capacitance, be it the resistance and
capacitance of the crystal itself or of the measuring device, the
charge generation can be read as an electrical potential across
the crystal. The magnitude depends upon the dielectric constant
of the material across which the potential is read.
In order to predict the nature of the secondary pyroelectric
properties of this material, or of any molecular material for that
matter, a knowledge of the thermal expansivity, of the structure
and of the molecular dipole is required. In the case of the
prediction of primary pyroelectric coefficients, which depend
upon the acoustical vibration modes of the material, it is
necessary to consider the structural components as anharmonic
oscillators and summing over the acoustical frequency band.23
Their contribution to the heat capacity function of the material
is proportional to the primary pyroelectric coefficient. Here
we expect the primary effect to be negligible. In order,
therefore, to predict the secondary pyroelectric properties, the
thermal expansion was determined in the temperature range -95
to 200 °C, in part (-95 to 45 °C) using the four-circle
diffractometer and in part (30-200 °C) using powder diffraction
(Figure 5). The change in unit cell parameters (volume, axis
lengths) upon thermal expansion clearly shows a linear relation-
ship, keeping the results obtained by the two different methods
separate from one another. The unit cell volume and unit cell
parameters as a function of temperature are shown (Figure 5).
Since no discontinuities in unit cell parameters are observed as
a function of temperature, no abnormal dielectric behavior can
be anticipated on this account. Furthermore DSC does not
indicate that a phase transition takes place before the melting
point is reached.
Figure 2. ORTEP drawing of phosphangulene (4). Only the asym-
metric unit has been labeled.
dihydropyrane to give 1. Selective lithiation in the 2-position
and subsequent reaction with phosphoroustribromide gave 2,
which upon deprotection with hydrochloric acid gave 3.
Reaction of 3 with base and subsequent aromatic nucleophilic
substitution by the phenolate ion gave the ring-closed product
4. Compound 4 could be oxidized by hydrogen peroxide in
chloroform. This gave the phosphine oxide 5. It is possible to
prepare tangible quantities of 4 (0.25-1.0 g scale possible). The
corresponding carbon compound is planar in its cationic form.56
Compound 4 is not planar, and therefore, its point symmetry is
lower (Figure 1). It is this nonplanarity which gives rise to the
appearance of a permanent dipole moment.
The crystal structure (Table 1) reveals that apart from being
triangular (Figure 2) the shape of the molecule is similar to
that of a Chinese hat (Figure 3) and also that these hats stack
on top of one another forming a directed stack (Figure 3). An
interesting point is that they stack in an eclipsed manner which
allows the stacks to pack next to each other (Figure 4). The
space group is rather unusual for an organic compound.41,42 The
crystal symmetry may be a consequence of the molecular
symmetry of the molecule combined with the molecule’s large
aptitude toward stacking and packing. The particular arrange-
ment of these molecular dipoles in the crystalline state thus gives
the material the permanent polarization and the consequent
pyroelectric properties.
The measurement of the pyroelectric properties in the
temperature range 28-110 °C was accomplished by measuring
the charge generation of a single crystal of the compound along
the polar axis (the c axis). A known capacitance was connected
in parallel (much larger than the capacitance of the electrical
circuit (crystal, amplifier, and electrical connections)), and the
electrical potential measured across this capacitance was taken
The expression for the polarization, P, is, when it is known
that the axis of the molecular dipole, µ, and the c axis are
collinear (their directions may be opposite),
0
Z
V(T)
(56) Martin, J. C.; Smith, R. G. J. Am. Chem. Soc. 1964, 86, 2252-
2256.
P(T) )
0
( )
µ
(5)

InVestigation of a Molecular Pyroelectric
J. Am. Chem. Soc., Vol. 119, No. 6, 1997 1213
Figure 3. Drawing that emphasizes the manner in which the molecules stack on top of each other and how all the stacks point in the same
direction.
The presence of ferroelectric properties would imply that the
polarization of the material could be reversed. Possible mech-
anisms to achieve this are (a) a mechanism that would require
the molecules to somehow undergo inversion either by inverting
around the phosphorous atom or by rotating in the crystal or
(b) a mechanism where the overall dipole moment is a result
of intermolecular dipoles rather than intramolecular dipoles as
expected. Inversion around phosphorous is highly unlikely first
of all because phosphines generally do not invert as evidenced
by the fact that optically active and optically stable phosphines
can be made, indicating that the inversion barrier is too high
for inversion in the temperature span where organic molecules
are stable.57 Further, inversion around phosphorous in the
crystalline phase would imply that all the molecules invert at
the same time in one concerted step since it would be impossible
to invert just one molecule and place it inverted on its former
site. This last statement also explains why rotation of the
individual molecules in the crystal can be ruled out as a possible
mechanism of ferroelectric behavior. In addition, such a
mechanism would severely affect the magnitude of the structure
factors, and even though the overall periodicity of the lattice is
not disrupted, such a mechanism would affect the appearance
of the X-ray powder diffractogram, an observation that has not
been made. The last mechanism would require that the
molecules were very polarizable, that two different distributions
of electrons in the molecule or in between molecules were
possible, and that each distrubution would result in a dipole
moment with a direction opposite to the other such that the
effective direction of polarization could be reversed. A way to
establish whether the direction of polarization can be reversed
is by establishing whether the field dependence of the polariza-
tion shows any hysteresis, a characteristic feature of ferroelectric
materials.58 Attempts were made to measure the field depend-
ence of polarization by the Sawyer method.59 The results were
not conclusive because the sample capacitances that could be
obtained ranged between 0.5 and 2 fF. These capacitance values
were found to be much too small for practical purposes. Even
though no numerical information on the field dependence of
the polarization could be obtained, it is concluded that compound
4 does not exhibit ferroelectric behavior, mainly because the
molecule itself does not have any molecular structural features
which suggest that the molecule can alter its conformation nor
Figure 4. Drawing that shows the packing of the molecules in the ab
plane.
where P, V, T, Z, and µ equal the polarization, volume (given
as V(T) ) aT + b), temperature, molecular entities per unit
cell, and molecular dipole moment, respectively. The polariza-
tion at a given temperature can thus be calculated; at room
temperature P ) 0.032 C m-2. Using the linear equation for
thermal expansion and upon differentiating with respect to
temperature, one obtains an expression for the secondary
pyroelectric coefficient at a given temperature
0
dP(T)
dT
Z
) p3(T) ) -
0
( )
µ
(6)
b2
a
aT2 + 2bT +
(
)
A plot of the experimentally determined pyroelectric coefficient
and of the pyroelectric coefficient obtained using the equation
above shows a good similarity between the two at temperatures
below 60 °C (Figure 6). At temperatures above 60 °C, however,
the observed pyroelectric coefficient becomes nonlinear and
increases rather rapidly in magnitude. This is due to sublimation
of the material, the effect of which cannot be separated from
the true secondary pyroelectric effect, since they both lead to a
decrease in spontaneous polarization (or in effect, a decrease
of molecular dipoles per unit volume). The fact that sublimation
occurs and that this is solely responsible for the observed
deviation from the calculated pyroelectric coefficient was
quantified by thermogravimetric measurements (Figure 7). An
obvious practical argument is that the material can be sublimed
at 90 °C in a vacuum-sealed tube over 24 h. Any reaction with
oxygen can be excluded as the material sublimes in open air
without formation of any oxide as shown by phosphorous NMR
of a sample sublimed in air. The rate of sublimation at 100
°C, determined by thermogravimetry, compares quite well to
the rate of disappearance of molecular dipole moments obtained
as the increase in the pyroelectric signal at 100 °C compared to
the value of the pyroelectric signal at 28 °C.
(57) Horner, L.; Winkler, H.; Rapp, A.; Mentrup, A.; Hoffmann, H.;
Beck, P. Tetrahedron. Lett. 1961, 161.
(58) Jona, F.; Shirane, G. Ferroelectric Crystals; Pergamon Press Inc.:
Oxford, U.K., 1962; pp 1-23.
(59) Sawyer, C. B.; Tower, C. H. Phys. ReV. 1930, 35, 269-273.
(60) Brandrup, J.; Immergut, E. H. Polymer Handbook, 3rd ed.; John
Wiley & Sons: New York, 1989; pp V51-V54.

1214 J. Am. Chem. Soc., Vol. 119, No. 6, 1997
Krebs et al.
Figure 6. Calculated secondary pyroelectric coefficient as a function
of temperature obtained by combination of eq 6 and the linear expansion
from Figure 5. This is compared to the experimentally observed
pyroelectric response that shows an increase in magnitude at increased
temperature.
Figure 7. Result of thermogravimetric measurements in the temper-
ature range 20-200 °C. At a scan rate of 1 °C/min, sublimation
becomes detectable at 127 °C. The upper curve represents the weight
of the sample in percent of the initial sample weight. The lower curve
represents rate of loss of material per minute. Weight loss given in
percent of the initial sample weight per minute.
Figure 5. Plots as a function of temperature of the unit cell volume
(above), the a,b-axis length (middle), and the c-axis length (below).
The values in the lower part of the temperature range were obtained
using a four-circle diffractometer (CAD4) on a single crystal, and the
higher temperature range values were obtained using powder diffraction
(STOE).
temperature, the pointed end of the crystal acquires an electrical
potential higher than that of the flat end. We know from the
structure/morphology relationship (Figure 8) that at the pointed
end all the concave sides of the molecules are exposed. The
appearance of positive charges at the electrode at the pointed
end when subjecting the crystal to a temperature rise (Figure
10) implies that the pointed end of the crystal must have the
negative charge. These results thus show that the molecular
dipole moment is directed from the concave to the convex side
of the molecule (following the convention for the direction of
the molecular dipole moment from - to +). In other words,
the negative end of the molecule is the benzene rings and the
positive end of the molecule is at the phosphorous atom (Figure
11). To further substantiate these findings, an identical experi-
ment with LiNbO3 as the pyroelectric sample with a known
direction of polarization was carried out. This showed that the
direction of polarization could be found and further that the
literature value for the pyroelectric coefficient of LiNbO3 could
be reproduced with the experimental setup.
exhibit any polarization properties much different from those
of its constituent phenyl rings taken individually.
The assignment of the direction of the molecular dipole
moment, with respect to the molecule itself, can be made when
the relationship between the physical morphology of the crystal
and the crystal structure is known (Figure 8) and when, at the
same time, the orientation of the crystal during measurement
of the pyroelectric coefficient is known. The physical morphol-
ogy of the crystals is similar to the shape of a pencil sharpened
at one end. This makes it very easy to orient the crystals with
the positive c axis in the direction of interest when performing
measurements on the crystals. The sign of the secondary
pyroelectric coefficient is negative because linear expansion
leads to a decrease in polarization.
The experimentally observed sign of the pyroelectric response
(Figure 9) depends upon the orientation of the crystal during
measurement, and it is observed that, upon increasing the
It is noticeable that the dielectric permittivity tensor has the
largest components perpendicular to the c axis. This is in good
agreement with the notion that the benzene rings are very

InVestigation of a Molecular Pyroelectric
J. Am. Chem. Soc., Vol. 119, No. 6, 1997 1215
Figure 8. Drawing showing the orientation of the molecules with
respect to the physical morphology of a typical crystal.
polarizable in the plane of the ring. The planes of the benzene
rings are (due to the hat shape of the molecule and the
orientation of the molecules in the unit cell) forming an angle
with the c axis. The polarizable benzene rings thus have a
component in the c direction and a much larger component in
the a, b direction. The only significant contribution to the
permittivity at room temperature is thus believed to be electronic
in nature. The fact that the magnitude of the relative permittivity
is smallest in the c direction is a great advantage when
considering the performance of a pyroelectric detector system.
When evaluating a material for potential use in a detector
system, the detectivity merit factor is derived. The detectivity
merit factor, Mr, for a pyroelectric material, can be expressed
as a function of its dielectric permittivity, heat capacity, density,
and of course its pyroelectric coefficient at the given temper-
ature:2
Figure 9. Plots of the step temperature change (top graph), the
corresponding pyroelectric response when orientating the crystal with
the flat end electrically connected to ground (orientation 1, middle
graph), and the pointed end electrically connected to ground (orientation
2, bottom graph). Schematic drawings of the electrical setups and the
orientations are shown below the graphs.
pn(T)
was found to be comparable in performance to many of the
known pyroelectric matarials but also lower in performance
when compared to the most yielding materials such as PVDF,
TGS, and Li2SO4 hydrate. A few examples of merit factors
for various materials are given at 25 °C (Table 2). In infrared
detection, the interest is to have a material which responds to
small variations in temperature (small differences in energy),
the smaller the better. The properties outlined above are
important when considering the magnitude of the pyroelectric
voltage obtainable with a given material. The use of phos-
Mr )
(7)
ꢀoꢀrcpF
The units are m2 C-1 and express the signal voltage obtainable
with a given material when multiplied with the energy absorbed
at the surface:
signal voltage (J C-1) ) Mr (m2 C-1) × energy (J m-2) (8)
When compared to other known materials, phosphangulene (4)

1216 J. Am. Chem. Soc., Vol. 119, No. 6, 1997
Krebs et al.
Table 2. Comparison of Phosphangulene (4) with Some Known
Materials (at 298 K)
Pna
cp
Mr
(C m-2 K-1
(J kg-1
F
(m2 C-1
material
× 10-6
)
ꢀr
2.9
K-1
)
(kg m-3
)
× 10-2
8.4
)
4b
3.2
1000 1478
PVDFc
27-40
8.4-13.5 806 1750-1800 23-25
LiNbO3
40d
28e
43
43
8
400
1630
635e 4640e
430 7450
1500 1700
400 2050
440 5326
490 5328
5.4
14.4
28.8
172
7.2
3.1
e
LiTaO3
176
280
TGSf
Li2SO4H2Oe 100
SBNg
600
1170
La:SBNg
a Numerical value only. b This work. c Data from ref 60. d Data from
ref 29. e Data from ref 2. f Data from ref 7. g Data from ref 4.
because of the significant vapor pressure of the material at
temperatures just above room temperature.
Conclusions
Phosphangulene (4) was synthesized and its crystalline
structure determined. It was found to belong to a rare space
group and to have pyroelectric properties, the nature of which
were characterized in the temperature region 20-110 °C. The
thermal expansivity in the temperature region -95 to 200 °C
was also determined and used to predict the nature of the
secondary pyroelectric properties using the value of the mo-
lecular dipole moment measured in solution. The success of
this approach is due to the fact that the molecule has a locked
conformation. Only when the conformation is the same in
solution as in the solid state can the assumption be made that
the dipole moment is similar in magnitude both in the solid
state and in solution. The direction of the molecular dipole
moment with respect to the molecule itself was determined. The
molecular heat capacity was calculated. The refractive indices
and the dielectric constants at optical frequencies, at 35 GHz,
and the isotropic dielectric constant at low frequencies (120 Hz
and 1 kHz) were determined. The material was considered for
potential use in infrared detection and compared with that of
known materials. It was found to be comparable in performance
to known materials but to have a significant vapor pressure just
above room temperature, making an application difficult.
Figure 10. Illustration of the observed pyroelectric response when a
crystal is subject to a thermally induced volume expansion. Above,
the crystal is in thermal and electrical equilibrium at one temperature.
Below, thermal equilibrium has been reached at a new and higher
temperature. The material has expanded, and an increase in volume is
seen. Electrical equilibrium is, however, not achieved, and charges are
consequently observed at the electrodes (thick vertical lines). The
orientation of the dipoles in the dielectric between the capacitor plates
is thus deduced from the sign of the potential difference.
Acknowledgment. We thank D. Micheelsen, Ole V. Peter-
sen, Mikkel Jørgensen, Per Michael B. Johansen, and Finn W.
Poulsen. Special thanks to Klaus Bechgaard for reading this
manuscript as well as making this work possible.
Figure 11. Drawing of the molecule together with the unambiguously
assigned direction of the molecular dipole moment with respect to the
molecular geometry (following the convention that the direction is
toward the positive charge).
Supporting Information Available: Tables of fractional
coordinates, equivalent isotropic and anisotropic thermal pa-
rameters, bond lengths, bond angles, and unit cell parameters
in the temperature range -93 to 200 °C (7 pages). See any
masthead page for ordering and Internet access instructions.
phangulene in a detector system, although a possibility based
on the value of the merit factor, is essentially problematic
JA962023C