Langmuir films of amphiphilic crown ethers

Article abstract of DOI:10.1021/jp990787j

We have used X-ray reflectivity and grazing-incidence X-ray diffraction to study amphiphilic crown ethers 1 and 2 at the air-water interface. A new synthetic route was employed, which substantially reduced the level of impurities. Unlike previous studies of 1, we saw no evidence of a first-order lying-down to standing-up transition of the alkyl tails. We did observe in both compounds a gradual reorientation of the tails away from the interface as the film was compressed, followed by an irreversible collapse to a multilayer structure. The monolayer films have a fluid-like inplane structure, while the multilayer films are crystalline. ? 1999 American Chemical Society.

Full text of DOI:10.1021/jp990787j

6206
J. Phys. Chem. B 1999, 103, 6206-6214
Langmuir Films of Amphiphilic Crown Ethers
Paul A. Heiney,* MacKenzie R. Stetzer, and Oksana Y. Mindyuk
Department of Physics and Astronomy and Laboratory for Research on the Structure of Matter,
UniVersity of PennsylVania, Philadelphia, PennsylVania 19104
Elaine DiMasi
Physics Department, BrookhaVen National Laboratory, Upton, New York 11973
Andrew R. McGhie
Laboratory for Research on the Structure of Matter, UniVersity of PennsylVania,
Philadelphia, PennsylVania 19104
Hu Liu and Amos B. Smith III*
Department of Chemistry and Laboratory for Research on the Structure of Matter,
UniVersity of PennsylVania, Philadelphia, PennsylVania 19104
ReceiVed: March 5, 1999; In Final Form: June 1, 1999
We have used X-ray reflectivity and grazing-incidence X-ray diffraction to study amphiphilic crown ethers
1 and 2 at the air-water interface. A new synthetic route was employed, which substantially reduced the
level of impurities. Unlike previous studies of 1, we saw no evidence of a first-order lying-down to standing-
up transition of the alkyl tails. We did observe in both compounds a gradual reorientation of the tails away
from the interface as the film was compressed, followed by an irreversible collapse to a multilayer structure.
The monolayer films have a fluid-like inplane structure, while the multilayer films are crystalline.
1. Introduction
) 0 to 27 mN/m at approximately 240 Å2/molecule, a horizontal
plateau, and then a second sharp rise below 76 Å2, leading to
Langmuir films continue to attract attention both as model
two-dimensional thermodynamic systems and for their potential
applications. Grazing incidence diffraction (GID) and X-ray
reflectivity (XR) have proven to be particularly useful structural
probes and have shown that in model rodlike amphiphiles many
different structures are possible, involving combinations of
conformational rearrangements (chain melting, etc.), reorienta-
tion of the entire molecule (tilting transitions), and long-range
cooperative effects. (See, e.g., refs 1 and 2 for reviews or ref 3
for a recent example.) Macrocyclic amphiphiles4,5 present a
different set of issues. Such molecules are often capable of
complexation with metal and molecular ions and are particularly
interesting candidates as components of supramolecular ionic
devices. Structural transitions in alkyl-substituted azacrowns and
similar molecules have been associated either with rearrange-
ments from flattened arrangements (with the macrocycle core
cofacial with the water surface) to edge-on arrangements or with
layering transitions.6-12
Disubstituted crown ethers such as 1 (Figure 1) are in some
ways intermediate between rodlike and macrocyclic amphiphiles,
in that they possess both a macrocyclic core capable of
complexation with metal ions and also two long chains that
might be expected to behave individually like traditional rodlike
amphiphiles.
Wei et al.13-18 studied 1 (as prepared by Xie et al.19) at the
air-water water interface and reported pressure-area (Π-A)
isotherms with a stepped structure, with an initial rise from Π
collapse at 65 mN/m. These features were explained as arising
from a first-order transition from a lying-down configuration,
with the central core and carbonyl groups in contact with the
water subphase but the chains extending away from the subphase
(as in Figure 1b), to a condensed phase with the alkyl chains
and carbonyl groups directed away from the interface (Figure
1c). They also reported that monolayer Langmuir-Blodgett
films of 1 could be used as alignment layers for liquid
crystals.14-18 If the films were deposited with low spreading
pressure, they obtained homogeneous alignment of the liquid
crystal, while with high spreading pressure (presumably the
conformation in Figure 1c) homeotropic alignment was obtained.
Wei et al. inferred that the differing tail conformations in the
Langmuir film were transferred to the solid substrate and
provided a basis for an altered interaction with the liquid crystal
molecules.
We were interested in further investigating the lying-down
to standing-up transition in 1 using the greatly increased
structural sensitivity made possible by X-ray scattering tech-
niques. Accordingly, we prepared a sample using the synthetic
sequence of Xie et al.,19 as outlined in Scheme 1. In our hands,
however, the reaction between 4-(heptyloxy)benzoic chloride
(3) and 4-hydroxybenzoic acid (4) gave a 70:16:14 mixture
consisting of the desired 4-[(4-(heptyloxy)benzyl)oxy]benzoic
acid (5), trimeric acid 6, and 4-(heptyloxy)benzoic acid (7).
Although acid 7 could be readily removed from 5 and 6 by
column chromatography, the mixture of 5 and 6 proved
inseparable either by fractional precipitation or column chro-
* To whom correspondence should be addressed.
10.1021/jp990787j CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/09/1999

Langmuir Films of Amphiphilic Crown Ethers
J. Phys. Chem. B, Vol. 103, No. 30, 1999 6207
Figure 1. Structures of compounds 1 and 2 and some possible conformations of 1 at the air-water interface, drawn schematically: (a) crown ether
core cofacial with water, phenyl rings, and alkyl chains in contact with water; (b) core and phenyl rings in contact with water, alkyl chains projecting
away from the interface; (c) core in contact with water, phenyl rings and alkyl chains projecting directly away from water; (d) core in contact with
water, phenyl rings and alkyl chains making an angle φ with the interface.
matography. All attempts to suppress the formation of 6 and 7
by modifying the reaction conditions failed. Not withstanding
this difficulty, the mixture of 5 and 6 was converted to a mixture
of acid chlorides with thionyl chloride and then treated with
diaza-18-crown-6-ether 8 to furnish crown ether liquid crystal
1, which was contaminated as expected with corresponding
unsymmetrical crown ether 9. We were again unable to purify
1. It is important to note that the ratio of 1 and 9 was not
significantly altered by repeated fractional precipitation from
ethanol, which is the purification method described by Xie et
al.19 Furthermore, the Π-A isotherms we obtained using our
(impure) sample of 1 were inconsistent with those previously
reported.13-18
trifluoroacetic acid in the presence of anisole to capture the
transient tertiary butyl carbocation afforded acid 5 in 74% yield
without contamination. Formation of the corresponding acid
chloride, followed by reaction with diaza-18-crown-6-ether 8
according to the reported procedure19 gave pure 1 in 68% yield.
Crown ether liquid crystal 2 was prepared in a similar fashion
(Scheme 3).
We then used Π-A isotherms, XR, and GID to study the
pure samples of 1 and 2 thus obtained. We did not see the same
behavior at the air-water interface as the Wei group, but we
did see evidence of multilayer formation and collapse, as well
as intralayer self-organization at higher pressures. In the
remaining sections of this paper we will describe in more detail
the synthesis of pure 1 and 2, and the results of our physical
measurements on these samples.
We thus turned to an alternative synthetic route for the
preparation of pure acid 5, and in turn pure crown ether liquid
crystal 1 (Scheme 2).
Toward this end, treatment of 4-hydroxybenzoic acid (4) with
N,N-dimethylformamide di-tert-butyl acetal furnished tert-butyl
4-hydroxybenzoate (10) in 52% yield,20 which was then
converted to tert-butyl ester 11 in 89% yield upon treatment
with 4-(heptyloxy)benzoic chloride (3). Acidic hydrolysis with
2. Experimental Section
Synthesis.22 tert-Butyl Ester 10. A solution of 4-hydroxy-
benzoic acid (4, 1.58 g, 11.44 mmol) in anhydrous toluene (35
mL) was heated to 80 °C and dimethylformamide di-tert-butyl

6208 J. Phys. Chem. B, Vol. 103, No. 30, 1999
Heiney et al.
SCHEME 1
SCHEME 2
acetal (3.48 g, 22.88 mmol) added slowly. After 4 h, the reaction
mixture was cooled to room temperature, diluted with brine (10
mL), and extracted with ethyl acetate (3 × 20 mL). The
combined organic extracts were washed with brine (2 × 10 mL),
dried over MgSO4, filtered, and concentrated. Flash chroma-
tography (ethyl acetate/hexane, 20:80) provided 10 (1.15 g, 52%
yield) as a white solid: mp 129-131 °C [lit.20 mp 129-131
°C]. The spectral properties of 10 were identical with literature
values.20
tert-Butyl Ester 11. A solution of ester 10 (0.39 g, 1.99 mmol)
in anhydrous benzene (11 mL) was cooled to 0 °C. Triethyl-
amine (0.81 g, 8.0 mmol) was added followed by 4-(heptyloxy)-
benzoic chloride (3, 0.71 g, 2.79 mmol). The reaction was stirred
at room temperature for 12 h and then concentrated. Flash

Langmuir Films of Amphiphilic Crown Ethers
J. Phys. Chem. B, Vol. 103, No. 30, 1999 6209
SCHEME 3
chromatography (ethyl acetate/hexane, 3:97) provided 11 (0.73
g, 89% yield) as a white solid: mp 94-95 °C; IR (KBr) 2960
1.76 (m, 4H), 1.49-1.42 (m, 4H), 1.39-1.24 (m, 12H), 0.88
(t, J ) 6.9 Hz, 6H); 13C NMR (500 MHz, CDCl3) δ 171.5,
164.6, 163.7, 151.8, 134.1, 132.3, 128.1, 121.8, 121.3, 114.4,
70.7, 69.8, 68.4, 50.1, 46.4, 31.7, 29.1, 29.0, 25.9, 22.6, 14.0;
high-resolution mass spectrum (FAB, NBA) m/z 961.4821 [(M
+ Na)+; calcd for C54H70N2O12Na 961.4826]. Anal. Calcd for
C54H7oN2O12: C, 69.06; H, 7.51; N, 2.98. Found: C, 68.97; H,
7.55; N, 2.86.
1
(m), 1740 (s), 1710 (s), 1610 (s) cm-1; H NMR (500 MHz,
CDCl3) δ 8.12 (d, J ) 6.8 Hz, 2H), 8.04 (d, J ) 6.7 Hz, 2H),
7.24 (d, J ) 6.8 Hz, 2H), 6.95 (d, J ) 6.9 Hz, 2H), 4.03 (t, J
) 6.6 Hz, 2H), 1.84-1.77 (m, 2H), 1.59 (s, 9H), 1.49-1.42
(m, 2H), 1.39-1.28 (m, 6H), 0.89 (t, J ) 6.9 Hz, 3H); 13C
NMR (500 MHz, CDCl3) δ 165.1, 164.4, 163.7, 154.5, 132.4,
131.0, 129.5, 121.6, 121.2, 114.4, 81.1, 68.4, 31.8, 29.1, 29.0,
28.2, 25.9, 22.6, 14.1; high-resolution mass spectrum (FAB,
NBA) m/z 413.2326 [(M + H)+, calcd for C25H33O5 413.2328].
Anal. Calcd for C25H32O5: C, 72.79; H, 7.82. Found: C, 73.02;
H, 8.08.
tert-Butyl Ester 12. A solution of acid 5 (0.130 g, 0.365
mmol) in thionyl chloride (neat, 4 mL) was heated to reflux
under argon for 5 h. The reaction was then cooled to room
temperature, concentrated, diluted with anhydrous benzene (5
mL), and cooled to 0 °C. The resultant mixture was then treated
with a solution of tert-butyl ester 10 (0.071 g, 0.365 mmol)
and triethylamine (0.146 g, 1.46 mmol) in anhydrous benzene
(3 mL), warmed to room temperature, stirred for 16 h, and
concentrated. Flash chromatography (ethyl acetate/hexane, 5:95)
afforded 12 (0.159 g, 82% yield) as a white solid: mp 127-
130 °C (dec); IR (KBr) 2960 (m), 1750 (s), 1720 (m), 1610 (s)
cm-1; 1H NMR (500 MHz, CDCl3) δ 8.28 (d, J ) 8.8 Hz, 2H),
8.15 (d, J ) 8.9 Hz, 2H), 8.08 (d, J ) 8.7 Hz, 2H), 7.38 (d, J
) 8.7 Hz, 2H), 7.28 (d, J ) 8.8 Hz, 2H), 6.99 (d, J ) 9.0 Hz,
2H), 4.06 (t, J ) 6.5 Hz, 2H), 1.84-1.81 (m, 2H), 1.61 (s,
9H), 1.52-1.46 (m, 2H), 1.41-1.29 (m, 6H), 0.91 (t, J ) 7.0
Hz, 3H); 13C NMR (500 MHz, CDCl3) δ 165.0, 164.3, 164.0,
163.9, 155.6, 154.2, 132.4, 131.9, 131.0, 129.8, 126.6, 122.2,
121.5, 121.0, 114.5, 81.2, 68.4, 31.8, 29.1, 29.0, 28.2, 26.0, 22.6,
14.1; high-resolution mass spectrum (FAB, NBA) m/z 555.2362
[(M + Na)+; calcd for C32H36O7Na 555.2359]. Anal. Calcd for
C32H36O7: C, 72.16; H, 6.81. Found: C, 72.52; H, 6.74.
Acid 6. A solution of tert-butyl ester 12 (0.140 g, 0.263 mmol)
in methylene chloride (10 mL) at 0 °C was treated with anisole
(1 mL) followed by trifluoroacetic acid (2.5 mL). The reaction
mixture was stirred for 3 h and then concentrated. The resultant
solid was washed with ethyl ether (3 × 6 mL) and filtered to
give 6 (0.096 g, 77% yield) as a white solid: mp 220-225 °C;
IR (KBr) 2960 (m), 1750 (s), 1700 (s), 1610 (s) cm-1; 1H NMR
(500 MHz, DMF-d7) δ 8.32 (d, J ) 8.8 Hz, 2H), 8.17 (d, J )
8.9 Hz, 2H), 8.16 (d, J ) 8.6 Hz, 2H), 7.62 (d, J ) 8.8 Hz,
2H), 7.55 (d, J ) 8.7 Hz, 2H), 7.20 (d, J ) 9.0 Hz, 2H), 4.18
(t, J ) 6.5 Hz, 2H), 1.84-1.80 (m, 2H), 1.50-1.27 (m, 8H),
0.89 (t, J ) 7.1 Hz, 3H); 13C NMR (500 MHz, DMF-d7) δ
167.4, 164.7, 164.6, 164.5, 156.4, 155.2, 132.9, 132.3, 131.7,
129.7, 127.3, 123.4, 122.9, 121.5, 115.5, 69.0, 32.3, 29.6, 29.5,
26.4, 23.0, 14.2; high-resolution mass spectrum (FAB, NBA)
Acid 5. A solution of tert-butyl ester 11 (0.20 g, 0.49 mmol)
in methylene chloride (10 mL) at 0 °C was treated with anisole
(1 mL) followed by trifluoroacetic acid (2.5 mL). The reaction
mixture was stirred for 3 h and then concentrated. Flash
chromatography (methanol/chloroform, 5:95) provided 5 (0.13
g, 74% yield) as a white solid: mp 156-160 °C; IR (KBr) 1740
1
(s), 1690 (s), 1610 (s) cm-1; H NMR (500 MHz, CDCl3) δ
8.18 (d, J ) 8.8 Hz, 2H), 8.13 (d, J ) 8.9 Hz, 2H), 7.33 (d, J
) 8.7 Hz, 2H), 6.97 (d, J ) 8.8 Hz, 2H), 4.04 (t, J ) 6.6 Hz,
2H), 1.86-1.78 (m, 2H), 1.49-1.44 (m, 2H), 1.41-1.28 (m,
6H), 0.89 (t, J ) 6.9 Hz, 3H); 13C NMR (500 MHz, CDCl3) δ
169.6, 164.3, 163.9, 155.5, 132.4, 131.9, 126.5; 122.0, 121.0,
114.4, 68.4, 31.8, 29.1, 29.0, 26.0, 22.6, 14.1; high-resolution
mass spectrum (FAB, NBA) m/z 357.1697 [(M + H)+; calcd
for C21H25O5 357.1702]. Anal. Calcd for C21H24O5: C, 70.77;
H, 6.79. Found: C, 70.82; H, 6.88.
Crown Ether Liquid Crystal 1. A solution of acid 5 (0.127
g, 0.357 mmol) in thionyl chloride (neat, 4 mL) was heated to
reflux under argon for 16 h. The reaction was then cooled to
room temperature, concentrated, and diluted with anhydrous
benzene (3 mL). The resultant mixture was treated with a
solution of crown ether 8 (0.043 g, 0.162 mmol) and triethyl-
amine (0.073 g, 0.714 mmol) in anhydrous benzene (5 mL) and
then stirred at room temperature for 50 min, heated to reflux
for 16 h, cooled to room temperature, and concentrated. Flash
chromatography (methanol/chloroform, 3:97) afforded 1 (0.133
g, 88% yield) as a white solid, which was recrystallized twice
from ethanol to give 1 (0.103 g, 68% yield) as a white solid:
mp 125-128 °C; IR (KBr) 1740 (s), 1640 (s), 1610 (s) cm-1
;
1H NMR (500 MHz, CDCl3) δ 8.10 (d, J ) 8.8 Hz, 4H), 7.44
(d, J ) 8.4 Hz, 4H), 7.22 (d, J ) 8.4 Hz, 4H), 6.95 (d, J ) 8.9
Hz, 4H), 4.02 (t, J ) 6.6 Hz, 4H), 3.84-3.45 (m, 24H), 1.83-

6210 J. Phys. Chem. B, Vol. 103, No. 30, 1999
Heiney et al.
m/z 477.1904 [(M + H)+; calcd for C28H29O7 477.1913]. Anal.
Calcd for C28H28O7: C, 70.57; H, 5.92. Found: C, 70.44; H,
5.93.
Crown Ether Liquid Crystal 2. A solution of acid 6 (0.056
g, 0.118 mmol) in thionyl chloride (neat, 3 mL) was heated to
reflux under argon for 5 h. The reaction was then cooled to
room temperature, concentrated, diluted with anhydrous benzene
(2 mL), and cooled to 0 °C. The resultant mixture was then
treated with a solution of crown ether 8 (0.014 g, 0.05 mmol)
and triethylamine (0.024 g, 0.24 mmol) in anhydrous benzene
(2 mL), warmed to room temperature, stirred for 16 h, and
concentrated. Flash chromatography (methanol/chloroform,
5:95) afforded 2 (0.049 g, 78% yield) as a white solid, which
was recrystallized twice from ethanol to give 2 (0.034 g, 54%
yield) as a white solid: mp 192-193 °C; IR (KBr) 2750 (m),
1
1750 (s), 1620 (s) cm-1; H NMR (500 MHz, CDCl3) δ 8.25
(d, J ) 8.6 Hz, 4H), 8.13 (d, J ) 8.9 Hz, 4H), 7.47 (d, J ) 8.0
Hz, 4H), 7.35 (d, J ) 8.6 Hz, 4H), 7.25 (d, J ) 8.3 Hz, 4H),
6.96 (d, J ) 8.9 Hz, 4H), 4.04 (t, J ) 6.6 Hz, 4H), 3.84-3.48
(m, 24H), 1.84-1.78 (m, 4H), 1.49-1.42 (m, 4H), 1.39-1.29
(m, 12H), 0.89 (t, J ) 7.1 Hz, 6H); 13C NMR (500 MHz, CDCl3)
δ 171.4, 164.3, 164.1, 163.9, 155.6, 151.6, 134.4, 132.4, 131.8,
128.2, 126.6, 122.2, 121.8, 121.0, 114.5, 70.7, 69.8, 68.4, 50.1,
46.4, 31.7, 29.1, 28.9, 25.9, 22.6, 14.0; high-resolution mass
spectrum (FAB, NBA) m/z 1201.5291 [(M + Na)+; calcd for
C68H78N2O16Na 1201.5249]. Anal. Calcd for C68H78N2O16: C,
69.25; H, 6.67; N, 2.38. Found: C, 68.85; H, 6.33; N, 2.05.
Calorimetry. Differential scanning calorimetry (DSC) mea-
surements employed a TA Instruments DSC 2920 calorimeter.
The samples were loaded in air into crimped Al DSC pans, and
the measurements were made under flowing He.
Figure 2. Differential scanning calorimetry data for 1 (top) and 2
(bottom). Data were collected at a scan rate of 2 K/min in He.
X-ray Scattering. X-ray measurements were conducted at
beam line X22B (λ ) 1.55Å) of the National Synchrotron Light
Source, Brookhaven National Laboratory, using a liquid surface
spectrometer as described previously.10,21,23,24 X-ray measure-
ments on 1 were conducted at 291 K, and those on 2 at 293 K,
unless otherwise indicated.
results were not affected by incomplete mixing of the sample
with the solvent, selected solutions were sonicated before
deposition, without producing any qualitative change from the
isotherms shown in Figure 3. X-ray measurements employed a
movable barrier/Langmuir float trough, as previously de-
scribed.10,21,23 Pressure-area isotherms were remeasured for
each compound immediately prior to the scattering measure-
ments. Subsequently, a new film was deposited and compressed
to the desired pressure. Except as otherwise indicated, the
spreading pressure decreased by less than 10% during the course
of an X-ray measurement. The combined uncertainty in mo-
lecular area (due to measurement and deposition error and
pressure variation during the measurement) was on the order
of 10%. It is worth noting that the isotherms measured on the
two troughs were consistent to within 10% in pressure and area
even though the source of water for the subphases was different.
A monochromatic X-ray beam was deflected toward the water
surface via Bragg reflection from a Ge(111) crystal. For XR
measurements, the momentum transfer is normal to the interface
(in the Qz direction). In this case the instrumental resolution
was determined primarily by the slits between the sample and
the detector. These slits were always set to optimize the
resolution in the scattering plane while increasing the signal by
lowering the resolution out of the scattering plane. The resolution
in the direction perpendicular to the water surface (Qz) during
XR measurements was 0.008 Å-1. For GID measurements, the
incident beam was almost parallel to the water surface and the
scattered photons were collected with a Braun position sensitive
direction oriented with the wire in the vertical direction. The
resolution in the Qxy scattering plane was set by the incoming
3. Results
Figure 2 shows characteristic DSC data for 1 and 2. Xie et
al. previously reported19 a nematic liquid crystal phase in 1
between 393 and 408 K, as observed by DSC and optical
microscopy. Similarly, we observed two endotherms upon
heating 1, with onsets at 399 and 410 K, and two related
exotherms at low temperatures upon cooling. Optical microscopy
under crossed polarizers showed that the region between the
two features observed in calorimetry was indeed associated with
a nematic liquid crystalline mesophase. However, the situation
was complicated by the growth of at least two and possibly
three different crystal modifications at low temperature, with
the relative quantities of the different phases depending on the
cooling rate. Upon heating, these crystal phases melted at
slightly different temperatures. Consistent with this observation,
slits to be approximately 0.007 Å-1
.
Pressure-Area Isotherms. All subphases employed water
purified using a Millipore filtration system with a resulting
resistivity F > 18 MΩ/cm. Small quantities of solute (≈10-3
g) were weighed on a Denver Instruments A-200DS balance to
an accuracy of 10-4 g, giving a typical mass uncertainty of
3-5%. All monolayers were spread from 10-3 to 10-4
M
chloroform solutions. The Langmuir trough was cleaned be-
tween experiments using a hexane rinse followed by an ethanol
rinse and three rinses with clean H2O. Initial pressure-area
isotherms, including those shown in Figure 3, were collected
on a commercially available Lauda film balance using the
movable barrier/Langmuir float technique. To verify that our

Langmuir Films of Amphiphilic Crown Ethers
J. Phys. Chem. B, Vol. 103, No. 30, 1999 6211
Fi, thickness Li, and roughness σi describing the interface
between slabs i and i + 1. The reflectivity is then given, as a
function of photon momentum transfer Qz, by
N+1
R(Qz)
Fi - Fi-1
F0
2
-Q′z2σi2/2
e-iQ′ Di e
(1)
(2)
z
)
∑
(
)
|
|
RF(Qz)
i)1
Q ′ ) Q 2 - q 2
x
z
z
c
where Q ′z is the refraction-corrected wavevector, qc is the
critical wavevector below which the reflectivity is unity, R(Qz)/
RF(Qz) is the measured reflectivity divided by the Fresnel
reflectivity calculated for a perfectly sharp subphase interface,
F0 is the electron density of the subphase, FN+1 ∼ 0 is the density
of the air or He above the sample, and Di ) ∑i Lj is the
j)l
distance from the water surface to the interface between layer
i and layer (i + 1) (i.e., D0 ) 0). Our general procedure for
analyzing the data was to use as few adjustable parameters as
possible, consistent with obtaining reasonable agreement be-
tween the model and data. Although the error bars in fitted
parameters obtained from a particular model were sometimes
quite tight, our general experience from fitting similar data
sets10,21,23 has been that model-dependent uncertainties (ob-
served, for example, in changing from a 1-box to a 2-box model)
lead to systematic uncertainties in all fitted parameters on the
order of 10%. For an N-box fit, the fitting parameters were the
overall intensity (constrained, however, to lie within 10% of
the measured incident photon flux), as well as a density and
thickness for each layer. Except as noted below, the roughness
terms σi were all constrained to be identical to the roughness
of the water surface σ0 induced by capillary waves. This was
taken to be an adjustable parameter in the fits but always
converged to a value in the physical reasonable range.
Figure 3. Pressure-area isotherms of 1 (top) and 2 (bottom) at the
air-water interface. Arrows show points along the isotherms at which
X-ray reflectivity profiles were measured.
we found that the detailed shape of the DSC traces, and the
integrated enthalpy change under each peak, depended on the
thermal history of the sample.
In 2 we observed a single broad endotherm (possibly with
some internal structure) on heating, exhibiting an onset at 465
K and an enthalpy change of 64 J/g. Two exotherms were
observed upon cooling. Optical microscopy confirmed that the
region between the two exotherms corresponded to a monotropic
nematic liquid crystalline phase.
In our XR measurements we observe changes in the surface-
normal structure of 1 with pressure (Figure 4). Data are shown
normalized to the calculated Fresnel reflectivity RF, at various
surface pressures. The maximum in R/RF is due to constructive
interference between the subphase-monolayer and monolayer-
air interfaces. The rapid decrease in R/RF at larger Qz shows
that the density profile, averaged over the rough water surface,
is smeared out over a length scale σ ∼ 3-4 Å. Solid lines show
the results of least-squares fits to the data, and the parameters
obtained with these fits are presented in Table 1. At Π ) 2.7
mN/m good agreement was obtained with a one-box model,
while for higher pressures two-box models were required. We
interpret the total thickness, L1 + L2, as the thickness of a
Langmuir monolayer, and the need for two boxes as representing
density variations within the layer. As the pressure is increased,
the layer thickness gradually increases.
Typical pressure-area isotherms for 1 and 2 at the air-water
interface, measured on compression, are presented in Figure 3.
Molecular areas at the beginning of the rapid rise were on the
order of 100 Å2 in both cases. If the barrier was stopped before
reaching the plateau, the pressure remained constant to better
than 10%. If the barrier was stopped on the horizontal plateau,
the pressure for both samples was observed to drop to around
25 mN/m and then stabilize. Thus, we believe that the vertical
rise is an equilibrium feature, while the behavior for areas less
than 60 Å2/molecule is indicative of nonequilibrium processes.
Indeed, the pressure maximum observed in both samples at the
right-hand side of the plateau is a clear indication of nonequi-
librium behavior. Furthermore, we visually observed stripes on
the trough for areas A < 60 Å2/molecule, indicating the
formation of bi- or multilayers. We never observed the pressure
rise previously reported13-18 in the vicinity of 240 Å2.
In an XR measurement, the photon momentum transfer is
perpendicular to the air-water interface, and the reflected X-ray
intensity is determined by the Fourier transform of the gradient
of the electron density perpendicular to the water surface. We
followed the common practice1,25 of analyzing the XR data by
modeling the electron density at the interface as a stack of N
uniform slabs (boxes), each with a different electron density
Good agreement was obtained between the calculated charge
density of 1 (504 e-/molecule) and the electron densities
deduced from experiment. From the measured molecular area
A and fitted parameters we can calculate
N
N ) A
F L
i i
(3)
∑
e
i)1
At Π ) 2.7 mN/m the experimental value of 600 differs by
19% from the calculated value, but at higher pressures measured
values lie in the range 527-537 e-/molecule, within 7% of the
value predicted. The relative thicknesses of the two layers and
the calculated charge densities are internally consistent if we

6212 J. Phys. Chem. B, Vol. 103, No. 30, 1999
Heiney et al.
At the highest compression (point iv on the isotherm in Figure
3) we were unable to obtain a satisfactory fit to the reflectivity
profile with any reasonable model. The solid line in Figure 4-iv
shows the best fit to a two-box model, with F1 ) 0.427, L1 )
12.7, F2 ) 0.292, L2 ) 4.5, and each layer allowed to have a
different roughness σi. This model is inconsistent with the
measured molecular charge and does not provide a good
description of the maximum at Qz ∼ 0.5 Å-1. A four-box model
corresponding to a 60 Å film provides a better fit to the data,
but with unphysical internal thicknesses. The most likely
interpretation of the data is that we have coexistence of different
layer thicknesses and that this coexistence cannot be represented
by a simple slab model.
We now consider the molecular conformations of 1 that could
give rise to the measured areas and reflectivity profiles. We
used simple molecular modeling to obtain approximate dimen-
sions of space-filling models starting from the configurations
shown in Figure 1; calculated dimensions from these models
are given in Table 2. If the entire molecule is parallel to the
water, we obtain a molecular area of 310 ( 60 Å2 and a
thickness of 5-7 Å. No signature of such a structure was seen
in either our Π-A isotherms or in XR. For the core and phenyl
rings in contact with water, but alkyl chains projecting away
from the interface, we obtain a molecular area of 205 ( 35 Å2
(depending in part on how the molecules are packed in two
dimensions) and a thickness of 10 Å, while if both chains
(including both biphenyl and alkyl portions) project away from
the interface the area and thickness are 72 Å2 and 21 Å,
respectively. We note that in this case the tails do not touch
each other, so that the molecular area is determined by the core
region in contact with the water. Finally, we can consider a
structure intermediate between (a) and (c), such as that in Figure
1d, in which the tails are tilted an angle φ from the horizontal.
For 0 < φ < 90 the molecular area could be as large as 310
Å2, if the entire molecule is projected onto the surface, or as
small as 70 Å2, if we assume that the tails can arrange
themselves as necessary to allow the cores to close-pack at the
surface, and the layer thickness could range between 5 and 16
Å.
Figure 4. Reflectivity profiles (total reflectivity divided by Fresnel
reflectivity) of 1 measured at points on the isotherm shown in Figure
3: (i) Π ) 3 mN/m; (ii) Π ) 14.8 mN/m; (iii) Π ) 27.0 mN/m; (iv)
full compression. Solid curves show fits to the data, as discussed in
the text. The inset shows calculated electron density F (e-/Å3) versus
distance from the interface z.
TABLE 1: Results of Fits to X-ray Reflectivity Dataa
Π
A
F1
L1
F2
L2
σ
compd (mN/m) (Å2) (e-/Å3) (Å) (e-/Å3) (Å) (Å) Ne
1
1
1
2
2
2.7(5) 125 0.425 11.3
3.5 600
14.8(2)
27.0(5)
1.7(5)
80 0.416 14.3 0.159 4.8 2.7 537
70 0.415 16.3 0.143 5.4 3.1 527
95 0.439 15.0
80 0.435 18.0
If we compare these models with the XR and isotherm data,
we see that the Π ) 2.7 mN/m data are not consistent with any
of the first three models. The most likely model is something
similar to that in Figure 1d, with the tails neither touching the
water surface nor completely vertical. A measured thickness of
11.3 Å corresponds to φ ) 23°. As the film is compressed,
both the Π ) 14.8 and Π ) 27.0 data sets are consistent with
either model 1c or 1d; i.e., the tails project away from the surface
but are not necessarily normal to the surface. Thus the increase
in film thickness with applied pressure suggests that the only
structural change upon compression in this range is a continuous
tilting of the tails toward a more vertical position.
3.3 626
3.8 626
19.2(2)
a Π and A are the pressure and molecular area at which the
measurement was made, obtained by correlating the measured pressure
with the isotherms in Figure 3. Li and Fi are the thickness and electron
density of the first and second layers, σ is the Gaussian surface
roughness, assumed for these fits to be the same for all interfaces, all
obtained from least squares fits to the reflectivity data. The uncertainty
in each of these parameters is on the order of 10%. Ne is the number
of electrons per molecule calculated from the molecular area, layer
thicknesses, and electron densities, to be compared with Ne(calcd) )
504 (for 1) and 628 (for 2).
Figure 5 shows the X-ray reflectivity from 2, normalized to
the calculated Fresnel reflectivity RF. The results are generally
similar to those obtained from measurements of 1. For Π )
1.7 and 19 mN/m, good fits were obtained with a one-box model
(with parameters given in Table 1). The data taken at full
compression cannot be described with a one- or two-box model.
The high frequency of the oscillations indicates that the film
must be much thicker. Indeed, a good fit was obtained using a
four-box model with a total thickness of 63 Å, but with
unphysical internal layer thicknesses. In any case, it seems clear
that the nonequilibrium maximum observed in the isotherm, the
pressure instability observed at small areas, and the visual
associate the second layer with the last 3-4 carbons in the
chains: 1.5 Å per carbon yields L2 ) 4.5 - 6.0 Å, and the
calculated charge ratio is F2L2/F1L1 ) 0.11 f 0.15, consistent
with the ratios of 0.13 and 0.11 obtained from Table 1. In reality,
of course, the charge density distribution must be more
complicated, but our spatial resolution is limited to ê ∼ πÅ/
0.7 ∼ 5 Å by the range in Qz provided by the XR measurements.
Fine details of the structure may also be obscured by the
roughness of the water subphase. The fact that the data are better
modeled by a two-layer model simply reflects the general
decrease in electron density from the bottom to the top of the
film.

Langmuir Films of Amphiphilic Crown Ethers
J. Phys. Chem. B, Vol. 103, No. 30, 1999 6213
TABLE 2: Molecular Dimensions for Various Conformations of 1 and 2 Calculated from Simple Molecular Modeling (Chem
3D)a
Compd
conformation
l (Å)
w (Å)
t (Å)
A (Å2)
1
1
1
1
2
2
2
2
(a) parallel to water (Figure 1a)
52 ( 3
6 ( 1
6 ( 1
6 ( 1
6 ( 1
6 ( 1
6 ( 1
6 ( 1
6 ( 1
6 ( 1
310 ( 60
(b) alkyl chains normal to water (Figure 1b)
(c) chains and phenyl rings normal to water (Figure 1c)
(d) tails tilted (Figure 1d)
34 ( 2
10 ( 1
205 ( 35
12 ( 1
21 ( 2
70 ( 15
12 + 40 cos φ
64 ( 3
6 + 15 sin φ
6 ( 1
70 + 240 cos φ
385 ( 70
(a) parallel to water (Figure 1a)
(b) alkyl chains normal to water (Figure 1b)
(c) chains and phenyl rings normal to water (Figure 1c)
(d) tails tilted (Figure 1d)
47 ( 2
10 ( 1
280 ( 50
12 ( 1
28 ( 2
75 ( 15
12 + 52 cos φ
6 + 22 sin φ
70 + 315 cos φ
a The corresponding schematic representations for 1 are given in Figure 1. The thickness t should be compared with the total film thickness ∑iLi
measured by XR measurements. The length l and width w are dimensions parallel to the surface, and their product is the area A, which should be
compared with that obtained from Π-A isotherms.
Figure 5. Reflectivity profiles (total reflectivity divided by Fresnel
reflectivity) of 2 measured at points on the isotherm shown in Figure
5: (v) Π ) 2 mN/m; (vi) Π ) 19 mN/m; (vii) full compression. Solid
curves show fits to the data. The inset shows calculated electron density
F (e-/Å3) versus distance from the interface z.
observation of stripes on the surface are all indicative of
multilayer formation.
We can apply the same conformational analysis to the
reflectivity data for 2 at Π ) 1.7 and Π ) 19 mN/m as that
used above for 1. The calculated molecular charges (Table 1)
are fortuitously in extremely good agreement with the data. The
areas and thicknesses calculated from molecular models similar
to those in Figure 1 are given in Table 2. No evidence of a
structure with the entire molecule in contact with water was
seen in Π-A or XR measurements: the measured areas are
too small and the measured thicknesses too large. Thicknesses
of 15 and 18 Å measured from XR correspond to tilt angles
(Figure 1d) of 26 and 34°, respectively, but these observations
could also be described by some statistical combination of
molecules exhibiting a range of tail tilts.
We now discuss our GID measurements, which probe the
in-plane momentum transfer Qxy. No features were seen in GID
profiles of 1 at Π ) 2.7, 14.8, or 27.0 mN/m. The most likely
explanation is that the two-dimensional structure was fluid, or
at least sufficiently disordered that no Bragg peaks were
produced, although it is remotely possible that highly oriented
crystallites were not detected in the single sample orientation
at which measurements were made. Figure 6, top, shows GID
Figure 6. Top: GID profile from 1 at full compression (position iv in
Figure 3). Points indicate the integrated X-ray counts (at 30 s/point)
between Qz ) 0 and Qz ) 0.5 Å-1 as a function of Qxy. The inset
shows the Qz dependence of the intensity at Qxy ) 1.51 Å-1, obtained
by counting for 300 s at the peak position and subtracting the
background intensity at (0.035 Å-1 from the peak. Bottom: GID
profile from 2 at full compression (position vii in Figure 3). Points
show Qz-integrated intensity; the intensity shows the Qz dependence at
Qxy ) 1.39 Å-1 with the background intensity at (0.35 Å-1 subtracted.
of 1 at full compression, position (iv) in Figure 3. There is a
strong feature at Qxy ) 1.51 Å-1, corresponding to a spacing
of 4.1 Å, together with several weaker features at smaller angles.
A “Bragg rod” scan at Qxy ) 1.51 (inset) shows some structure
up to Qz, ) 0.7 Å-1, consistent with thin-film diffraction. (The
length scale of the film thickness is determined by the inverse
of the Bragg rod length, 2π Å/0.7 ∼ 9 Å.) Keeping in mind
that the XR data indicate the probable existence of multilayers,
the precise structure cannot be determined. However, the most
likely interpretation of the data is a crystalline mono- or
multilayer structure. The unit cell might include one or several
molecules, but the structure factor is dominated by chain-chain

6214 J. Phys. Chem. B, Vol. 103, No. 30, 1999
Heiney et al.
interactions with a characteristic spacing of 4 Å. GID patterns
of 1 at full compression were collected at several temperatures
between 282 and 303 K, and no changes were observed,
indicating that the film crystal structure is stable up to at least
303 K.
Similarly, the bottom panel of Figure 6 shows the GID profile
of 2 at full compression. Again, a strong feature at 1.39 Å-1
(d ) 4.5 Å) indicates that the structure factor is dominated by
interchain correlations. The Bragg rod has less structure but
again extends to ∼0.7 Å-1, consistent with thin film diffraction.
References and Notes
(1) Als-Nielsen, J.; Jacquemain, D.; Kjaer, K.; Leveiller, F.; Lahav,
M.; Leiserowitz, L. Phys. Rep. 1994, 246, 251 and references therein.
(2) Jacquemain, D.; Grayer Wolf, S.; Leveiller, F.; Deutsch, M.; Kjaer,
K.; Als-Nielsen, J.; Lahav, M.; Leiserowitz, L. Angew. Chem., Int. Ed. Engl.
1992, 31, 130.
(3) Durbin, M. K.; Malik, A.; Richter, A. G.; Yu, C. J.; Eisenhower,
R.; Dutta, P. Langmuir 1998, 14, 899.
(4) Lednev, I. K.; Petty, M. C. AdV. Mater. 1996, 8, 615 and references
therein.
(5) Lehn, J.-M. Supramolecular Chemistry: Concepts and PerspectiVes;
VCH: Weinheim, 1995.
(6) Malthete, J. J.; Poupinet, D.; Vilanove, R.; Lehn, J.-M. J. Chem.
Soc. Chem. Commun. 1989, 1016.
(7) Mertesdorf, C.; Ringsdorf, H. Liq. Cryst. 1989, 5, 1757.
(8) Maliszewskyj, N. C.; Heiney, P. A.; Blasie, J. K.; McCauley, J.
P., Jr.; Smith, A. B., III. J. Phys. II 1992, 2, 75.
(9) Heiney, P. A.; Gidalevitz, D.; Maliszewskyj, N. C.; Satija, S.;
Vaknin, D.; Pan, D.; Ford, W. T. Chem. Commun. 1998, 1483.
(10) Gidalevitz, D.; Mindyuk, O. Y.; Stetzer, M. R.; Heiney, P. A.;
Kurnaz, M. L.; Schwartz, D. K.; Ocko, B. M.; McCauley, J. P., Jr.; Smith,
A. B., III. J. Phys. Chem. B 1998, 102, 6688.
4. Conclusions
Our best model for the interfacial behavior of both 1 and 2
is the following: At high dilution the structure is unknown but
most likely resembles the conformation in Figure 1d, with the
tails projecting slightly away from the air-water interface (a
small but nonzero value of φ). As the film is compressed, the
tails are directed on the average farther and farther away from
the interface, corresponding to larger values of φ or to a higher
concentration of conformation (c) in statistical mixtures of (b),
(c), and (d). This results in smaller molecular areas and thicker
films. The molecular area changes by almost a factor of 2 in
this process, corresponding to a high compressibility of the film.
Within the film, the molecules are most likely fluid. When the
molecular area is on the order of 70 Å2, the tails are all
approximately vertical and packed as closely as possible within
a monolayer, and at higher compression the film undergoes an
irreversible collapse to a multilayer structure, which has,
however, sufficient internal order to produce several Bragg
peaks.
(11) El Abed, A. M.; Peretti, P.; Gallet, F.; Billard, J. J. Phys. II Fr.
1993, 3, 851.
(12) El Abed, A. M.; Daillant, J.; Peretti, P. Langmuir 1993, 9, 3111.
(13) Fang, J. Y.; Lu, Z.-H.; Pan, S. P.; Chen, Z. L.; Wei, Y.; Qin, J.;
Xie, M. C. Solid State Commun. 1992, 82, 1023.
(14) Zhu, Y. M.; Lu, Z.-H.; Jia, X. B.; Wei, Q. H.; Xiao, D.; Wei, Y.;
Wu, Z. H.; Hu, Z. L.; Xie, M. G. Phys. ReV. Lett. 1994, 72, 2573.
(15) Zhu, Y.-M.; Wei, Y. J. Chem. Phys. 1994, 101, 10023.
(16) Zhu, Y. M.; Lu, Z. H.; Wei, Y. Phys. ReV. E 1995, 51, 418.
(17) Dong, X.; Yang, X. M.; Zhu, Y. M.; Lu, Z. H.; Wei, Y. Thin Solid
Films 1994, 251, 40.
(18) Zhu, Y. M.; Jia, X. B.; Xiao, D.; Lu, Z. H.; Wei, Y.; Wu, Z. H.;
Hu, Z. L.; Xie, M. G. Phys. Lett. A 1994, 188, 287.
(19) Xie, M. G.; Qin, J.; Hong, F.; Wang, L. Mol. Cryst. Liq. Cryst.
1991, 209, 309.
(20) Kricheldorf, H. R.; Kaschig, J. Liebigs Ann. Chem. 1976, 882.
(21) Gidalevitz, D.; Mindyuk, O. Y.; Heiney, P. A.; Ocko, B. M.;
Henderson, P.; Ringsdorf, H.; Boden, N.; Bushby, R. J.; Martin, P. S.;
Strzalka, J.; McCauley, J. P., Jr.; Smith, A. B., III. J. Phys. Chem. B 1997,
101, 10870.
(22) Materials and Methods: All reactions were carried out in oven-
dried glassware under an argon atmosphere, unless otherwise noted. All
solvents were reagent grade. Tetrahydrofuran (THF) was freshly distilled
from sodium/benzophenone under argon before use. Dichloromethane and
benzene were freshly distilled from calcium hydride before use. Triethy-
lamine was distilled from calcium hydride and stored over potassium
hydroxide. Unless stated otherwise, all reactions were magnetically stirred
and monitored by thin layer chromatography using 0.25 mm Whatman
precoated silica gel plates. Flash column chromatography was performed
with the indicated solvents using silica gel-60 (particle size 0.023-0.040
mm) supplied by E. Merck.
We observed no signs of a first-order lying-down to standing-
up transition as reported by Wei et al.13-18 Although a definitive
explanation for the difference in behaviors of our sample and
theirs is at present unknown, it is highly likely that the crown
ether liquid crystal samples prepared and studied by Wei et al.
were contaminated with the same crown ether liquid crystal
impurity 9 that we encountered in our initial syntheses.
Acknowledgment. We thank D. Vaknin, R. Composto, and
J. K. Blasie for the use of their equipment. We thank D.
Gidalevitz and B. M. Ocko for their assistance and for useful
discussions. OYM, MRS, PAH, and ABS were supported by
the MRSEC Program of the National Science Foundation (NSF)
under Award No. DMR96-32598. Research was carried out in
part at the National Synchrotron Light Source, Brookhaven
National Laboratory, supported under U.S. DOE Contract No.
DE-AC02-98CH10886.
(23) Gidalevitz, D.; Mindyuk, O. Y.; Heiney, P. A.; Ocko, B. M.; Kurnaz
M. L.; Schwartz, D. K. Langmuir 1998, 14, 2910.
(24) Ocko, B. M.; Wu, X. Z.; Sirota, E. B.; Sinha, S. K.; Gang, O.;
Deutsch, M. Phys. ReV. E 1997, 55, 3164 and references therein.
(25) Tidswell, I. M.; Ocko, B. M.; Pershan, P. S.; Wasserman, S. R.;
Whitesides, G. M.; Axe, J. D. Phys. ReV. B 1990, 41, 1111.