hooft@ruunsa.fys.ruu.nl (Rob Hooft) (11/12/90)
I've got enough osmosis-nonsense today. Please stop using all this empirical nonsense and start using plain thermodynamics. The effect is so simple to understand once you know the fundamental laws of thermodynamics. If more nonsense appears, I'll have to kill-file it! -- Rob Hooft, Chemistry department University of Utrecht. hooft@hutruu54.bitnet hooft@chem.ruu.nl chooft@fys.ruu.nl
shenkin@cunixf.cc.columbia.edu (Peter S. Shenkin) (11/13/90)
None of the diagrams is in error, and I disagree with your assessment of which
diagram best illustrates the phenomenon, as well.
In article <1990Nov11.132436.2836@newcastle.ac.uk: w.p.coyne@newcastle.ac.uk writes:
: 1.'Beaker'-with semi-permeable 2. 'U' shaped tube with membrane
: membrane down middle. at the bottom of the bend.
: | * | |^^^^^* | | | | | |^^| | |
: |^^^^^*^^^^^| | * | |^^| |^^| | | | |
: | high* lo | | *^^^^^| | | | | | | |^^|
: | * | | * | |hi'----*---'lo| | '----*---' |
: '-----------' '-----------' '-------*------' -------*------
: BEFORE AFTER BEFORE AFTER
:
: 3. Enclosed cylinder with semi-permeable membrane which can slide, so
: .---------------------. .---------------------.
: | high * lo | | * | as to alter the
: | * | | * | volumes on either side.
: '---------------------' '---------------------'
: BEFORE AFTER
:
:With both fig 1 and fig 2 mention is made of the levels ceasing to change
:when the difference in height becomes great enough.
:BUT
: I believe Fig 1 is false as no mention made of the changing membrane area
: available to each side.
The area of the membrane is totally irrelevant, and does not appear in the
equations describing the pheonomenon. This area will affect only the rate at
which osmotic equilibrium is achieved, not the position of the equilibrium.
The position of equilibrium is calculated by equating the chemical
potential of the solvent (actually, the chem. potential of all permeable
substances) on both sides of the membrane. This comes from the fact that if
there is a phase equilibrium at constant T, any ingredient common to both
phases must have the same chem. potential in both. Any part of the membrane
wet by one phase but not the other does not come into the equation, since at
such a part there is no phase equilibrium; only the part wet by both phases
enters into the calculation. Again, the area of this part does not
enter into the calculation.
:comments of the 3 diagrams:-
:Fig 1 is a very poor way of illustrating osmosis in that the effects of gravity
: and air pressure on both sides and the area of membrane exposed each side
: all change as one side rises and one falls
:Fig 2 much better as the area of membrane for each side is constant. But if
: the lo side falls enough it becomes the same as Fig 1.
I believe that two is clearer, but 1 is also correct.
:Fig 3 is the best in that it shows the result of osmosis without the added
: complications of the two others, unlike 1 and 2 the membrane should
: continue to move until both sides have same conc. or (in the case when
: the lo conc is actually pure water) all the water ends up on one side.
Fig 3 is also correct, but does not give a feeling for osmostic *pressure*.
Apparatus 3 is useless for calculating the molecular weight of a macromolecule,
for example. Apparatus 2 (or 1) could in principle be so used.
Note that in Figure 1 and Figure 2, after equilibrium has been achieved, the
high-pressure side still has a higher concentration than the low-pressure
side, but not by as much as at the beginning of the experiment. In Figure 3,
at equilibrium, there is neither a pressure nor a concentration difference
between the two sides.
-P.
************************f*u*cn*rd*ths*u*cn*gt*a*gd*jb**************************
Peter S. Shenkin, Department of Chemistry, Barnard College, New York, NY 10027
(212)854-1418 shenkin@cunixc.cc.columbia.edu(Internet) shenkin@cunixc(Bitnet)
***"In scenic New York... where the third world is only a subway ride away."***larry@kitty.UUCP (Larry Lippman) (11/14/90)
In article <1748@ruunsa.fys.ruu.nl>, hooft@ruunsa.fys.ruu.nl (Rob Hooft) writes: > I've got enough osmosis-nonsense today. > Please stop using all this empirical nonsense and start using plain > thermodynamics. The effect is so simple to understand once you know the > fundamental laws of thermodynamics. Amen! I was one of the first to respond to the question two weeks or so ago, and I gave a (somewhat) simple explanation based upon a thermodynamic approach. I briefly repeat: $$> My personal preference is to explain osmosis using a thermodynamic $$> approach. If one considers "chemical potential", as defined by the Gibbs $$> equilibrium theory, then it is simple to remember that in diffusional $$> transport (and chemical reactions in general, for that matter) chemical $$> substances move from higher to lower chemical potential. $$> $$> Osmosis represents a case where a solvent is common to both sides $$> of a semipermiable membrane. The chemical potential of such a pure solvent $$> would therefore be equal across such a membrane, and no transport would $$> occur. However, the chemical potential of a solvent containing a solute $$> is *less* than that of the solvent alone due to entropy (the solute being $$> dispersed in a random fashion within the solvent). $$> $$> Therefore, from a purely thermodynamic standpoint, the pure solvent $$> has a tendency to flow across the membrane into the side containing the $$> solvent and solute. Unless, of course, a pressure is exerted on the side $$> containing the solvent and solute which opposes the osmotic pressure $$> developed in that side. $$> $$> > Does it have something to do with the salt forming temporary weak bonds $$> > with the water molecules, so on the side of the membrane with the higher $$> > concentration fewer water molecules will be free to cross the membrane. $$> $$> No. It is important to understand that osmosis is a colligative $$> property of solutions in that the determining factors pertain solely to $$> the number of molecules of solute in solvent (and thermodyamic factors), $$> and are *independent* of actual chemical composition. It is also important to realize that as a concentration gradient across a membrane becomes "large", the relation of osmotic pressure becomes increasingly difficult to predict (i.e., "non-linear"). At least one reader already pointed this out. Larry Lippman @ Recognition Research Corp. "Have you hugged your cat today?" VOICE: 716/688-1231 {boulder, rutgers, watmath}!ub!kitty!larry FAX: 716/741-9635 {utzoo, uunet}!/ \aerion!larry
hooft@ruunsa.fys.ruu.nl (Rob Hooft) (11/14/90)
In <1990Nov13.111721.12306@newcastle.ac.uk> william@lorien.newcastle.ac.uk (William Coyne) writes: >In fig 1 in the AFTER part won't the water be cascading down >the area of membrane not exposed to the lo side, and would this not have an >effect on the equilibrium position? Wieeeee! There we have it again. Perpetual motion!! No, the water wouldn't be dropping to the low side. -- Rob Hooft, Chemistry department University of Utrecht. hooft@hutruu54.bitnet hooft@chem.ruu.nl hooft@fys.ruu.nl
mcdonald@aries.scs.uiuc.edu (Doug McDonald) (11/15/90)
In article <4171@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes: >In article <1748@ruunsa.fys.ruu.nl>, hooft@ruunsa.fys.ruu.nl (Rob Hooft) writes: >> I've got enough osmosis-nonsense today. >> Please stop using all this empirical nonsense and start using plain >> thermodynamics. The effect is so simple to understand once you know the >> fundamental laws of thermodynamics. > > Amen! I was one of the first to respond to the question two >weeks or so ago, and I gave a (somewhat) simple explanation based upon >a thermodynamic approach. I briefly repeat: > >$$> My personal preference is to explain osmosis using a thermodynamic >$$> approach. > >Larry Lippman @ Recognition Research Corp. "Have you hugged your cat today?" Sure. No problem. But that begs the question very seriously. The "usual" thermodynamic formulation ASSUMES ideal solutions. Are the big polymer solutions ideal? If so, this will work. Under what conditions ARE they ideal - does big size prevent ideality (yes, this is answered clearly in undergrad texts)? If they are not ideal WHY not - is it an entropic or energetic effect. What is the physical basis behind this? I don't know. But I will ask a genuine expert (Peter Wolynes). Stay tuned. Doug McDonald
mcdonald@aries.scs.uiuc.edu (Doug McDonald) (11/18/90)
In article <1990Nov14.171839.12177@ux1.cso.uiuc.edu> mcdonald@aries.scs.uiuc.edu (Doug McDonald) writes: >In article <4171@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes: >>In article <1748@ruunsa.fys.ruu.nl>, hooft@ruunsa.fys.ruu.nl (Rob Hooft) writes: >>> I've got enough osmosis-nonsense today. >>> Please stop using all this empirical nonsense and start using plain >>> thermodynamics. The effect is so simple to understand once you know the >>> fundamental laws of thermodynamics. >> > >Sure. No problem. > >But that begs the question very seriously. The "usual" thermodynamic >formulation ASSUMES ideal solutions. Are the big polymer solutions ideal? >If so, this will work. Under what conditions ARE they ideal - >does big size prevent ideality (yes, this is answered clearly in >undergrad texts)? If they are not ideal WHY not - is it an entropic >or energetic effect. What is the physical basis behind this? > >I don't know. But I will ask a genuine expert (Peter Wolynes). >Stay tuned. > OK, I did (ask Peter). We have that (in TeX) \int_0^{\Pi} V_A dP = -RT \log (P_A / P_A^*) where \Pi is the osmotic pressure, P_A^* is the vapor pressure of pure solvent, and P_A the vapor pressure of solution, and V_A is the partial molar volume of solvent. OF course for ideal solutions V_A = V_A*, and P_A / P_A^* = X_A, the mole fraction. This whole business assumes ideal solutions. In fact, it is hard to find membranes that will work well for many non-ideal ones. Biopolymers are reasonable ideal, at least at low concentrations. The sum of the volume of the unmixed things is the volume of the solution, and the heat of mixing is small. The free energy change of the solvent, which is what osmosis works on, of course, is then a purely entropic effect. The partial molar entropy of dilution of the solvent is - R \log (X_A). Being entropy only, the numbers just come out as a function of the numbers of different ways to distribute the molecules (of all kinds) in various orders. Note very importantly that since olny very dilute polymer solutions are ideal, a polymer never gets near another polymer - they are surronded by solvent molecules. Doug McDonald