Two-dimensional films of discotic molecules at an air-water interface

HAL Id: jpa-00209517

https://hal.archives-ouvertes.fr/jpa-00209517

Submitted on 1 Jan 1982

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Two-dimensional films of discotic molecules at an air-water interface

F. Rondelez, D. Koppel, B.K. Sadashiva

To cite this version:

F. Rondelez, D. Koppel, B.K. Sadashiva. Two-dimensional films of discotic molecules at an air-water interface. Journal de Physique, 1982, 43 (9), pp.1371-1377. �10.1051/jphys:019820043090137100�.

�jpa-00209517�

Two-dimensional films of discotic molecules

at an

air-water interface

F. Rondelez, ^D.

Koppel

Laboratoire de Physique de la Matière Condensée, Collège ^de France, 11, place ^Marcelin Berthelot, 75231 Paris Cedex 05, ^France

and B. K. Sadashiva

Raman Research Institute - Bangalore, ^{560080 India}

(Reçu ^{le 11 mars} 1982, accepté le 30 avril 1982)

Résumé. ²⁰¹⁴ On montre pour la première fois que des molécules discotiques ^de benzène-hexa-alkanoates peuvent être étalées en monocouches à un interface eau-air. Les mesures de pression superficielles ^{donnent des résultats}

très analogues ^{à ceux des films d’acides} gras. Le noyau benzénique joue ^{le rôle} de la tête polaire grâce à six groupes

carboxyliques périphériques, ^{et se met à} plat ^sur l’interface. Les chaines alkanoates jouent ^{le rôle des chaines} aliphatiques. ^Les isothermes de pression permettent ^une estimation des dimensions moléculaires dans le plan ^de

l’interface. Ces dimensions sont pratiquement identiques ^{à celles déduites des mesures} de rayons X dans les phases

à colonnes formées par les benzène-hexa-alkanoates (cristaux liquides thermotropes). ^Ces expériences permettent de réfuter l’interprétation de la transition liquide expansé-liquide ^{condensé comme une} transition du second ordre

entre une phase isotrope ^{et une} phase cristal liquide nématique.

Abstract. ²⁰¹⁴ The possibility ^of forming Langmuir monolayers ^with disc-like molecules of benzene-hexa-alkanoates has been investigated for the first time. Surface _pressure measurements show that these monolayers ^{behave in}

many respects ^{as the} well-known films of fatty acids. The benzene ring plays ^{the role of the} polar head group while the alkanoates chains ^are the long aliphatic tails. Molecular dimensions can be derived from the surface _pressure- concentration isotherms. The results indicate that the benzene rings ^{lie flat at} the interface. The projection ^{of the}

molecular diameter onto the interface is practically ^{identical to} the lattice spacing measured in the liquid-crystalline

columnar mesophases ^formed by benzene-hexa-alkanoates in bulk. These experiments ^{allow to refute the inter-} pretation ^{of the} liquid expanded-liquid ^condensed transition in terms of a second order isotropic ~ ^nematic phase transition, ^as recently proposed by several theoretical groups.

Classification

Physics ^Abstracts

68.15 ^- 64.70J ^- 61.30E

1.

Introduction. - Soaps

^and

phospholipids

^are

classical

examples

^of

amphiphilic

molecules. Their molecular structure is

composed

^{of a}

hydrophobic hydrocarbon

chain tail and of a

hydrophilic polar

head. As a consequence of this dual character,

they generally

^{lie at} the interfaces of emulsions or foams,

forming

^a

large variety

^of structures : lamellar, ^hexa-

gonal,

^etc...

[1].

^{The numerous molecules}

forming thermotropic liquid crystalline phases

^{compose a}

much less known class of

amphiphilic

molecules. It is clear however that the formation of smectic

phases

for instance is due to the

partial segregation

^{of the}

aliphatic

^flexible chains from the

rigid

^{aromatic cores.}

To the best of our

knowledge,

there have been few attempts ^to check if such

compounds

could also be used ^as surfactants in aqueous solutions since

they

have no

prominent

water-soluble group. There is

only

one report

by

^{Dbrfler et al.}

[2],

ⁱⁿ which calamitic

(rod-

like)

^{molecules of 4-4’}

azoxy-a-methyl

cinnamate,

forming

^nematic

liquid crystalline phase

^{in bulk, have}

been

spread in.monolayers

^{at an air-water interface.}

Such observations are however _very

interesting

^since

they

could lead the way ^to the

discovery

^{of two-}

dimensional nematic films with

quasi long

^{range orien-}

tational order of the

long

^{molecular axes. This}

possi- bility

^{is not ruled out}

by theory

^as first shown

by

^de

Gennes as

early

^{as 1971}

[3].

In this paper, ^we show for the first time that the discotic class of

thermotropic liquid crystalline

^mate-

rials can also be

spread

^as

Langmuir monolayers.

Preliminary

^{results are}

presented

^on the benzene- hexa-alkanoates molecules which possess six

aliphatic

chains attached to a central benzene group. From surface _pressure measurements, the benzene

rings

^are

observed to lie flat on the water due to the pressure of six

carboxyl water-attracting

groups. The

aliphatic

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphys:019820043090137100

1372

chains are also, ^{at least}

partially, parallel

^{to the inter-}

face but

they

^{do not}

adopt

^a

fully

^extended

configura-

tion. The results will be

compared

^{with the} average molecular diameters deduced from X-rays ^lattice

spacing

measurements in the columnar bulk

liquid crystalline phases.

2.

Experimental.

^- Surface pressures were measur-

ed both

by

^the

Wilhelmy hanging plate technique

^and

by

the movable barrier method

[4].

^{The two methods}

are indeed

complementary.

^{In the former, the total sur-}

face area is fixed and the

amphiphilic

^{molecules are}

added as

drops

of solution in a volative solvent until the

spreading

^{is no}

longer

effective. In the latter, ^{a fixed}

number of

drops

^is

spread

^{at the}

beginning

^{of the}

experiment

^{and the surface area of the}

trough

^{is then}

continuously

^decreased up ^to the

collapse point

^{of the}

monolayer.

2.1 WILHELMY PLATE METHOD. ^- The force trans-

ducer was a Hewlett-Packard 8805-B pressure

ampli-

fier. The system ^was calibrated with

precisely

^known

weights.

^The output

voltage

^was

adjusted

^{so that 10 mV}

corresponded

^{to 1.0 x 10-4 N m -1. The} quartz

glass

cuvette

containing

^{the water}

subphase

^{had a surface of}

29.6 ^x 10-4 m2. It was

totally

enclosed in a massive,

temperature-controlled,

^{oven in which water from a}

Haake F3-thermostat was

constantly circulating.

^The temperature ^{was varied between 4} OC and 60°C. The set temperature ^was constant to within ± 0.1 ^{°C. This}

was a necessary step since the

Wilhelmy technique

cannot compensate ^{for the} temperature variation of the surface tension of _pure water

against

^{air. This}

variation can indeed be

large, typically

^{1.5 x} 10-4 N m - I OC - 1

[5].

The total drift

during

^{the course}

of an

experiment,

^after

allowing

for thermal

equili-

bration, ^was less than 0.4 ^x 10-4 N m-1.

2.2 MOVABLE BARRIER METHOD. ^- A commercial Lauda film balance was used. All

experiments

^were

performed

^{at room} temperature, ^{T = 21} ± ^{5°C. A}

rigorous

temperature ^{control was not}

employed

because the film balance

directly

^{measures the diffe-}

rence between the surface _pressure of _pure water and that of the

monolayer.

^{Therefore thermal} fluctuations in the water surface _pressure are

automatically

^can-

celled out The maximum drift ^was observed to be about 10- 5 N m - 1 _per minute. The

speed

^{at which the}

barrier was moved ^was 0.25 x 10- 2

m/min.

^This

corresponds

^{to an} average percentage ^{variation in the} total

monolayer

surface of 1-3

%, sufficiently

^{low to}

always

allow for mechanical

equilibration

^{of the}

monolayer,

while still

being compatible

^{with the}

experi-

mental drift

during

^the

measuring

^{time. The duration}

of one

single

^{run was about one hour. The surface}

pressure ^was monitored both upon

compression

^and

expansion

^{to check for}

possible hysteresis

^effect.

2. 3 CHEMICALS. ^- The

synthesis

of the benzene- hexa-alkonates (BH ^for

short)

^has

already

^{been des-}

cribed elsewhere

[6, 7].

Five successive

homologs

^{were studied}

starting

from n-pentanoate up ^to n-decanoate.

Small

quantities

(~ ^10-3

g)

^were

weighted

^{on a}

Cahn electrobalance to an accuracy of 10-6 _g and dissolved in 20 ml of

methylene

dichloride (Merck- spectro

grade).

This solvent had been checked _pre-

viously

^{to be a}

good spreading

solvent, ^{non miscible}

with water and

highly

volatile, ^{and also to} contain

only insignificant

fractions of surface-active

impuri-

ties

[8].

The surface pressure increase when pure solvent

was added and let evaporate ^was less than 5 ^x

10-6 N m-1 after 10

drops. (Up

^{to 60}

drops

^{could be}

added for ^a

complete experiment

^{with the}

Wilhelmy plate

^and up ^to 300

drops

with the Lauda

balance.)

3. Results. ^- For all

compounds,

^{the surface} pres-

sure measurements have been

performed

^{at several}

Fig. ^{1. 2013 Surface} pressure isotherms of benzene-hexa-n- pentanoate (BH-5) using ^either Wilhelmy hanging plate ^or

movable barrier (Lauda) ^methods. Substrate is pure ^water.

temperatures between 5 and 60 OC

using

^the

hanging plate,

^{and at room} temperature,

using

^{the movable}

barrier

techniques. Typical

^{results are shown on}

figures

^{1 and 2 for}

benzene-hexa-n-pentanoate (BH-5)

and on

figures

^{3 and 4} for benzene-hexa-n-nonanoate

(BH-9).

^{The curves} obtained in these two

examples

^are

somewhat different.

1)

For the lower

homolog,

the pressure 7c increases

regularly

when concentration C is increased

(Wilhelmy method)

^{or when} film surface A is decreased

(Lauda balance).

^At

relatively large

pressure values,

typically

between 6 ^x 10-3 N m -1 and 8 ^x 10-3 N m -1

depending

^on temperature, ^{there is a sudden}

change

^of

slope.

^The

corresponding

concentration will be called

Ckink in

^the

following.

Above

Ck;nk,

^{the 7T-C or n-A curves become}

nearly

parallel

^{to the abscissa} axis. This

region

^of

slight

Fig. ^{2. -} Continuously recorded surface pressure isotherms of benzene-hexa-n-pentanoate (BH-5). ^The monolayer ^is

first compressed (decreasing ^{areas per} molecule), ^{then re-} expanded (increasing areas). ^{The data} points ^{have also}

been reported ⁱⁿ figure ^{1 for} comparison ^{with the} Wilhelmy technique.

upward slope

^has

only

^a limited width.

Finally

^the

slope

is observed to re-increase

sharply

^{until a state of}

saturation is reached in the film. This saturation occurs

for different surface pressures values

according

^{to the}

technique

^of

investigation.

^{With the}

hanging plate technique,

^where

drops

^are

successively deposited

onto a fixed surface _area, the maximum surface _pres-

sure attainable

corresponds

^{to the}

equilibrium spread-

Fig. ^{3. -} Surface pressure isotherms of benzene-hexa-n- nonanoate (BH-9) usmg Wilhelmy hanging plate technique.

Substrate is pure ^water.

Fig. ^{4. -} Continuously recorded surface pressure isotherms of benzene-hexa-n-nonanoate (BH-9). ^The monolayer ^is

first compressed (decreasing ^{areas per} molecule), ^then re-expanded (increasing areas).

ing

pressure n,,. This is the surface _pressure obtained when ^the

monolayer

^{is in}

equilibrium

with the stable bulk

phase. Experimentally,

^as n,, is

approached,

^the

time

required

^for

equilibration

^after

deposition

^{of one}

further

drop

^increases

drastically.

^The

corresponding

surface concentration and surface area per molecule

are called

Ce and Ae respectively.

^{For BH-5 at room} temperature, ne ^{= 11.6 x 10- 3 N}

m-1, Ce

^{= 1.05 x} 10-3 _g m-2

and Ae

^{= 1.20 nm2.} With the movable barrier

technique,

where the film area is

continuously

reduced while the total number of molecules in the

monolayer

^is

kept

constant, the maximum surface pressure attainable

corresponds

^{to the}

collapse

pres-

sure nr. This is the

highest

^surface pressure ^to which the

monolayer

^{can be}

compressed

^without

expulsing

molecules into the third dimension. The

corresponding

surface area and surface concentration are called

Ac

and

Cc respectively.

^Above nc, three-dimensional islands start to form, ^{and if one tries to} compress the film even more, the surface _pressure is

actually

^observ-

ed to decrease

(Fig.

2). ^{For BH-5 at room} temperature, nc ⁼ 15.2 x 10-3 N

m-’, Ac

^{= 1.10 nm2 and}

Cc

1.15 ^x 10-3 _g m-2. _nc is

markedly higher

^than ne.

However this

gives only

little difference between

Ae and Ac

because of the large pressure

slope

^{in that}

domain of concentrations.

Experiments

^{at different}

compression speeds yielded

similar values of _nc to within 5 %. ^After compression ^{to an area} per molecule less than

A, large hysteresis

^{was observed} upon

subsequent

^film

re-expansion.

^{On the} contrary ^{there was no detectable}

hysteresis

^as

long

^{as the film surface} pressure ^was

kept

^below 7rc.

2) ^{For the}

higher homolog,

the overall _pressure increase at room temperature is much less

gradual

^than

in the lower

homolog

^{case. In}

figure

3, ^{the data}

points

at 22°C stay ^{on the zero} pressure line until a concen-

tration of 5 ^x 10-3 _g m-2 is reached, ^{at which}

point

the _pressure increases so

rapidly

that the maximum pressure ne ⁼ 2.9 ^x 10- 3 N m-’ is

already

^{reached at}

1374

Table I. ^- Characteristic

surface

concentrations and

surface

pressures observed in the

surface

pressure isotherms

of the benzene-hexa-n-alkanoates. Ck;nk corresponds ^{to the}

surface

concentration at the onset

of

^{the transition}

between the

liquid expanded

^{and the}

liquid

^{condensed states}

of

^the

monolayer. Ce corresponds

^{to the}

surface

^concen-

tration at the equilibrium

surface

pressure 7r,,.

Cc

corresponds ^{to the}

surface

concentration at the collapse

surface

pressure of ^the

film

7r,,.

a concentration of

Ce

^{= 0.68 x}

10-3 gM-2 .

^{The same}

features are observed on the Lauda curves. As film surface is decreased, the pressure increases first very

slowly

then much more

rapidly

^{until a maximum}

Pressure 7rc ⁼ 3.2 x 10- 3 N m -1 is reached. The

corresponding

^{surface area} per molecule is

A,, =

2.67 nm2. For still lower surface areas, ^a decrease in pressure is observed

indicating

^{that a}

collapse

^{of the}

film has taken

place.

In

figure

3, the results for a much

higher

temperature of 60°C have also been

plotted. Although

^{the measu-}

rements are less accurate at this elevated temperature,

we observe that some of the features discussed for the lower

homolog

^{case start. to} re-appear. A kink in the

curve can be detected for a concentration

Cki-k -

0.2 ^x 10-3 _g m - 2 . At

larger

concentrations it is followed

by

^a somewhat flat

region

^{until a} very

sharp

rise in surface _pressure occurs. At this temperature ^the

equilibrium spreading

pressure 7r,, is found to be 6.9 x 10-3 N m-1.

Unambiguous

^evidence for the reappearances of three distinct domains in the surface pressure-concen- tration

plots

^{when the} temperature ^{is increased is also}

yielded

^from measurements on the BH-8 and BH-10

homologs.

The

important

data for all the

homologs

^{which have}

been studied in the present

investigation

^{are summa-}

rized in table I.

4. Discussion. ^- All our

experimental

curves can be understood

by analogy

^{with the}

monolayer

^behaviour

of the

simpler

^chain

fatty

^acid substances which have been

extensively

studied in the past. ^{It is well reco-}

gnized

^{that such}

monolayers

^{can take at} least three characteristic film states at an air-water interface. In order of

decreasing

^surface concentration,

they

^are

referred as condensed

(solid

^or

liquid)

films,

liquid- expanded

^{films and}

finally

gaseous films

[4].

For a

particular

substance, all the different states may or may ^not be observed

depending

^{on the}

experi-

mental conditions

(e.g. temperature).

^For instance, ^at

room temperature,

myristic

acid shows the three

phases

while

palmitic

^{acid shows}

only

the condensed and the gaseous

phases.

^It should also been

pointed

^{out that}

the present discussion is

oversimplified.

Indeed, ^much

more intricate

phase diagrams, including

^mesomor-

phic

^states, have been described in the literature

[9].

However this crude level of

presentation

^is

quite

^suf-

ficient for our present purposes.

The existence of the three

phases

^is

clearly

^{evident in}

our

monolayers

of benzene-hexa-alkanoates. From table I, ^{it is seen that BH-5 at 5 and 21} °C, ^{BH-6 at}

5 and 21 oC, ^{BH-8 at 40} OC, ^{BH-9 at 40 and 60 °C and}

BH-10 at 40 OC, ^{all show}

expanded-liquid

^{films and}

condensed-liquid

^films

(no

attempts ^{to observe the}

always existing

gaseous film have been made since it

corresponds

^{to low surface} pressures and

requires

very accurate measurements

[10]).

The transition between the

liquid-expanded

^and

liquid-condensed phases

^is very obvious in the surface pressure isotherm

plots.

^In the transition

region

^the

compressibility

^{of the}

monolayer

becomes small and there is not much of a

surface _pressure

change

^{when surface} concentration is increased (see

Fig. 1).

^This transition, called the main

transition, ^is

generally

^{assumed to be} first-order

although

the isotherm shows a non-zero

slope

^{in the}

transition

region.

The fact that the main transition does not behave as an

ordinary

three-dimensional

solid-liquid phase

change has been discussed

by

many authors

[11]. Recently,

Albrecht et al.

[9]

^have

suggested

that this is due to the limited size of the

cooperatively transforming

^{units. For}

phospholipid monolayers, they

^{have estimated} that these domains

incorporates

about 150 molecules.

Our

experimental

data show a number of cases

where the transition between a

liquid-expanded

^{and a}

liquid-condensed

^{film is not observed.} The pressure isotherms stay very close to the horizontal axis until the concentration reaches a certain concentration at which the curves turn

upward

^almost

vertically.

^Here

the gaseous films ^are

directly

^{converted to condensed}

films. BH-9 at 21°C

gives

^a

good example

^{of such a}

behaviour

(Figs.

^{3 and} 4).

We thus observe here that the _range of

stability

^of

the different

phases

changes ^{in a}

regular

^{manner with} temperature. This behaviour is also encountered with

single

^chain

fatty

acids materials

[4].

^{For a}

given compound

^the occurrence of a

liquid expanded phase

becomes less and less

probable

^{as the} temperature ^is lowered.

Reciprocally,

^{at fixed} temperature, ^the

liquid- expanded

film becomes less and less

probable

^{as the}

chain

length

is increased from five to ten carbon atoms.

It is often found in other

compounds

^{that the addition}

of one carbon atom to the

length

^{of one}

aliphatic

chain shifts the curves in

nearly

^the very ^same fashion

as a decrease of 8-10 OC in temperature. ^The present

experiments

corroborate

roughly

that result.

Going

from BH-7 to BH-8, it is necessary ^to increase the temperature ^{from 21°C to 40 OC to} re-observe the

expanded-liquid

^film.

The _pressure isotherms

yield

information on the conformation of the benzene alkoanoates

spread

^{at an}

air-water interface. We will see that the present ^data

are

only

consistent with a model where the benzene

rings

^{lie flat on the water} while the chain

align

^{more or}

less

horizontally, although

^not

completely.

^{Such a}

conformation is

intuitively

evident since the presence of the six

water-attracting carboxyl

groups forces the benzene to be in direct contact with the aqueous

subphase.

^{On the other hand, the}

aliphatic

chains have

no reason to stay ^{close to the water.} However there is a

certain chain

rigidity

^which prevents ^{the first}

methy-

lene _groups to stand

directly upright.

This local

rigi- dity

^{has been}

amply

demonstrated in fluid

bilayers by

measurements of the order parameter ^{for each} seg- ment,

using

^EPR

[12]

^{or NMR}

[13]

spectroscopy. ^For

typical aliphatic

chains, ^{the first five or six}

methylene

groups ^are in a stretched

configuration.

^The average molecular dimensions can be estimated from the concentration at which the film reaches its

equilibrium spreading

^{pressure. Indeed at}

Ce

the molecules can be considered ^as closed

packed

and therefore the _average

area per molecule

Ae

^is

easily

^deduced

knowing

^the

molecular

weight

^{of the}

compound

^under

investigation.

The measured

Ce

for the various benzene-alkanoates have been

reported

^{in table I and the}

corresponding

calculated

Ae

^{values are}

given

in table II. The maximum molecular

length d,,

^can then be estimated

assuming

that there is no

interpenetration

^{of the}

aliphatic

^chains

between two

adjacent

molecules. Under these condi-

tions,

simple geometric

^{relations shows that}

d,, c!e

^1.24

A e ’I’

for molecules

having

^a

hexagonal shape.

The different values for

d,,

^{are also}

given

ⁱⁿ

table II. The

subscript

ⁿ indicates that the

d,,

^values

are derived from surface pressure

experiments.

^{It is}

true that the choice of

Ae

^to calculate the

d,,

^{values is}

somewhat

arbitrary.

The average molecular ^{areas at} the

collapse point, A, ,,

could also have been chosen.

At _any rate, ^the

Ae and Ac

^{values are very close and}

certainly

within the uncertainties of the

experiments.

Therefore the

dx

^{values are}

probably

correct to better than

10 %.

^A

comparison

^of these values with those derived from X-ray diffraction

photographs

^{in the}

three-dimensional columnar

mesomorphic

^{state is}

very

interesting.

^In that latter _case, Chandrasekhar

et al.

[7]

^have

proposed

^an

aligned, liquid crystalline

Table II. ^- Molecular dimensions

of

the benzene-hexa-n-alkanoates as derived

from

monolayer

surface

pressure

experiments

(dft)’ from

molecular models

assuming fully

extended chains

(dm) and from

X-ray ^lattice spacing ^bulk

measurements

(dx-ray)’ Ae

^and

Ac

^{are the areas} per molecule measured at the

equilibrium surface

pressure ne and at

the collapse pressure 7r,,,,

respectively.

^{All data}

are for

^{T = 21 °C.}

1376

structure of stacked discs

forming

columns with an

hexagonal close-packed

arrangement. ^{Therefore an} estimate of the disc molecular diameter ^can be derived

directly

^from measurements of the lattice

spacing.

Unfortunately,

^such

dx-,a,

^{data are}

^only

available for BH-7 and BH-8

[7-14],

since for the other benzene- hexa-alkanoates the columnar

mesomorphic

^{state does}

not exist or is

highly

metastable. A

comparison

^bet-

ween dn

^and

dx -,ay in

^{these two}

^existing

^{cases show an}

amazingly good

agreement between the two sets of data

(see

columns 3 and 5 of table II). ^{It therefore}

appears that the molecules

adopt nearly

^{the same}

conformation when

they

^are

spread

^{as a}

monolayer

at an air-water interface or when

they

^are

piled

^on top of each other in fluid columns.

In neither case are the

aliphatic

^chains

fully

^extended

as evidenced

by

^a

comparison

with the molecular diameter

dm

calculated for stretched chains from mole- cular

Dreiding

^models

(see

column 4 of table

II).

^The

dn

^{values are}

consistently

^{lower than}

dm

^{in all cases.}

Lastly,

several other conformations can be

certainly rejected.

^{Were the}

aliphatic

^chains

fully

^vertical

relative to the water surface, ^a much lower molecular surface area would be observed. From a

space-filling

model we can estimate that it should be in the order of 95

A2, independent

of the actual chain

length.

^{This is}

clearly

^not substantiated

by

^our observations which show a variation in

Ae

^{from 110 up to 3 30 A’. Another}

possibility

would be that the benzene nucleus is forced to take an

upright position

upon

compression.

Such a conformation has been indeed observed

by

Adam on benzene derivatives with various lateral substitutions

[15].

^The

projected

^{molecular area}

occupied by

^one benzene is then of the order

of 24 A2,

much too low

again

to account for our

experimental

results, ^{even when}

considering

^the additional area

occupied by

^each of the six

aliphatic

^{chains. It should}

be noted that the Adam

compounds

^{were not hexa-}

substituted with

water-attracting

groups. Therefore it was much easier for the benzene to tilt _away from the _aqueous

subphase.

5. Conclusions. ^- The present

experiments

^show

that benzene-hexa-alkanoates form stable

monolayers

when

spread

^{at an air-water} interface. The pressure isotherms exibit the well-known three states charac- teristic of

liquid-condensed, liquid-expanded

^and

gaseous films

according

^{to surface} concentrations and temperatures. Benzene-hexa-alkanoates constitute therefore ^{a new} class of surfactant materials with

properties

^as well defined as the classical

single long

chain

fatty

^{acids or}

phospholipids.

The benzene nucleus is observed to lie in direct contact with the water,

probably

because of the existence of the six

carboxyl

^«

pinning points

^{». The}

aliphatic

^{chains are} also, ^{at least}

partially,

parallel ^{to the water. A certain}

amount of randomness does exist however, ^{as indi-}

cated from a

comparison

between the measured mole-

cular

dimensions and that of

fully

extended chains.

In the two cases where the data are available in the literature, the measured molecular diameters are in close agreement, ^to better than 1

A,

with the diameters measured in the

mesomorphic

^bulk

liquid

^{state where}

these disc-like molecules are stacked in columns.

The

possibility

^of

forming monolayers

^{with mole-}

cules

having thermotropic mesomorphic

^{states in the}

bulk _open

interesting perspectives

^to try and demons- trate the

theoretically predicted

existence of

quasi-

nematic order in two dimensions

[3].

^{In that case it is}

true that calamitic molecules will have to be used rather than discotic molecules. However, ^the present work can

give

^{some hints on which} type of molecules will be best suitable.

Monolayers

of disc-like benzene-hexa-alkanoates may prove

important

^{as new} systems ^{on which to test} the current ideas on two-dimensional

melting.

^Their

six-fold symmetry ^{is well}

adapted

^to detect the ^occur-

rence of the defect-mediated hexatic

phase predicted by

Nelson and

Halperin [16].

This could

perhaps

^be

done

through

^surface

viscosity

measurements

although

such

experiments

^are

always

very delicate in mono-

layers.

Finally,

^the present

experiments

^{do not bear out}

the recent

interpretation

of the kink observed in the surface _pressure isotherms in terms of a second-order

phase

transition between the

liquid-expanded

^{and the}

liquid-condensed

^states

[11].

^Indeed the associated theoretical models all _suppose a

gradual

orientation of the

projection

^{of the}

aliphatic

chains in the inter- facial

plane.

^{Such a}

liquid crystalline,

uniaxial, ^nematic

ordering

^is

clearly

^not

possible

^{with molecules}

having

six-fold symmetry. Therefore the more classical inter-

pretation

^{of a} first-order

phase

transition ^{seems more} relevant here,

although

^{a true} flat transition

region

with infinite

compressibility

^{is not observed.}

Acknowledgments.

^{- We thank H.} Gruler for many

helpful

discussions

during

^his sabbatical stay ^{at the}

College

de France. This work has been

supported

under a

joint

program

by

the C.N.R.S.

(France)

^and

the C.S.I.R.

(India).

References

[1] ^{PERSHAN, P.} S., ^J. Physique Colloq. ⁴⁰ (1979) ^C3-423.

[2] DÖRFLER, ^H. D., KERSCHER, ^{W. and} SACKMANN, H., J. Phys. ^Chem. (Leipzig) ²⁵¹ (1972) ^314.

[3] ^DE GENNES, ^P. G., ^Farad. Symp. ⁵ (1971) ^16.

[4] GAINES, ^G. L., ^Insoluble monolayers ^at liquid-gas ^inter- faces (Wiley, ^New York) ^1966.

[5] ^Handbook of Physics (Chemical ^Rubber Company), Cleveland, ^Ohio.

[6] CHANDRASEKHAR, S., SADASHIVA, B. K. and SURESH, ^K.

A., ^{Pramana 9} (1977) ^471.

[7] CHANDRASEICHAR, S., SADASHIVA, B. K., SURESH, ^K. A., MADHUSUDANA, ^N. V., KUMAR, S., SHASHIDAR, ^R.

and VENKATESH, G., ^J. Physique Colloq. ⁴⁰ (1979)

C3-120.

[8] VILANOVE, ^{R. and} RONDELEZ, F., Phys. Rev. Lett. 45

(1980) 1502.

[9] ALBRECHT, O., GRULER, ^{H. and} SACKMANN, E., J.

Physique ³⁹ (1978) ^301.

[10] HAWKINS, G. A. and BENEDEK, ^G. B., Phys. ^{Rev. Lett.}

32 (1974) ^524.

KIM, ^{M. W. and} CANNELL, ^D. S., Phys. ^{Rev. A 13} (1976) ^411.

[11] ^{For a recent} review, ^see BARET, ^J. F., Progress ⁱⁿ Surf.

and Memb. Sci. ed. by ^{D. A.} Cadenhead and J. F. Danielli (Academic Press, ^New York) 1981, Vol. 14, p. 291.

[12] SEELIG, J., ^{J. Am.} Chem. Soc. 92 (1970) 3881; ⁹³ (1971)

5017.

[13] CHARVOLIN, J., MANNEVILLE, ^{P. and} DELOCHE, B.,

Chem. Phys. ^{Lett. 23} (1973) ^345.

[14] GOOZNER, ^{R. E. and LABES, M.} M., ^Mol. Cryst. Liq.

Cryst. ^{Lett. 56} (1979) ^75.

[15] ADAM, ^N. K., ^Proc. Roy. ^{Soc. 119} (1928) ^628.

[16] HALPERIN, ^{B. I. and} NELSON, ^D. R., Phys. ^{Rev. Lett.}

41 (1978) ^121. Phys. ^{Rev. B 19} (1979) ^2457.