In his article “The First Wetting Layer on a Solid,” Peter Feibelman (Physics Today, February 2010, page 34) points out that the first layer of water molecules on a solid surface embodies the boundary condition for water transport, pollution, corrosion, and other molecular transport phenomena. That observation and the revealing highresolution images presented bring to mind a fundamental problem of osmotic water transport.

In 1827 René Dutrochet pointed out that osmosis actually involves binary transport, 1 in which water moves one way and solute moves the other way. In 1855 Adolf Fick took the idea much further, 2 expanding on the work of other experimentalists. He considered a cylindrical pore in a hydrophilic membrane separating either water or a dilute salt solution on one side and a concentrated one on the other. 3 He reasoned that water will preferentially flow along the walls and salt will tend to migrate along the axis of the pore. As a consequence, he expected concentration gradients in the plane of the pore. Under certain conditions, he suggested, salt migration could be completely inhibited even though the pore might be large enough to allow migration of salt molecules. Subsequent contributions by Jacobus van’t Hoff 4 and Walther Nernst 5 established that molecular diffusion in aqueous solutions involves the migration of a solute in one direction driven by the gradient of osmotic pressure, and the flow of water in the opposite direction.

Binary transport in aqueous solutions is widely recognized, but the actual mechanisms are not clear. Solute diffusion involves the random migration of free molecules or ions. However, because water is a condensed phase, its migration cannot be visualized in terms of random motion of molecules. It is not clear if such a flow of water can be considered viscous, because viscosity typically involves wall effects and external forces. If it is viscous, what is the nature of that flow?

Work along the lines described in Feibelman’s article may throw more light on the nature of binary transport in osmosis and on molecular diffusion in which random motion of unattached molecules in one direction is accompanied by the migration of a condensed phase in the opposite direction.

1.
R.J. H.
Dutrochet
,
Ann. Chim. Phys.
35
,
393
(
1827
).
2.
A.
Fick
,
Philos. Mag. J. Sci.
10
,
30
(
1855
).
3.
T. N.
Narasimhan
,
Eos Trans. Am. Geophys. Union
85
(
47
),
499
(
2004
).
4.
J. H.
van’t Hoff
,
Z. Phys. Chem.
1
,
481
(
1887
).
5.
W. H.
Nernst
,
Z. Phys. Chem.
2
,
613
(
1888
);
2
,
634
(
1888
).