1
$\begingroup$

Consider the classical U-tube, semi-permeable membrane experiment. Assume I define osmotic pressure as $\Pi=\frac{RT}{V}\gamma$, where $\gamma$ denotes the number of total solute particles (across all species) residing in a solvent of volume $V$.

I have seen people say something to the effect of, "At equilibrium conditions, the osmotic pressure $\Pi$ of the left compartment $C_L$ is equal to the osmotic pressure of the right compartment $C_R$"...equivalently: $\Pi_{C_L}=\Pi_{C_R}$ (at equilibrium).

Defining osmolarity $\Omega$ as $\Omega = \frac{\gamma}{V}$, simple arithmetic then tells us that:

$$\Pi_{C_L}=\Pi_{C_R} \implies \frac{\gamma_{C_L}}{V_{C_L}}=\frac{\gamma_{C_R}}{V_{C_R}} \implies \Omega_{C_L}=\Omega_{C_R}$$

i.e. if osmotic pressure of the two compartments is equal, then osmolarity of the two compartments is equal.

This result immediately caught my eye because then I thought of the following edge case:

Suppose I have a U-tube filled with pure solvent and I add some amount of solute (have your pick) to the left side. Suppose that the semi-permeable membrane allows solvent diffusion but prohibits solute diffusion. What happens?

Because $\gamma_{C_R}=0$, I would need an infinite amount of water to begin with in order for the non-zero $\frac{\gamma_{C_L}}{V_{C_L}}$ to approach $0$.

So what's wrong with my derivation? More specifically, when is my derivation appropriate? Do you always need non-zero solute in both compartment and a sufficiently large reservoir of solvent in order for equilibrium to be achieved? What exactly are the implicit assumptions of the procedure for the U-tube, semipermeable experiment?


Edit: Alternatively, is something wrong with my initial formulation of osmotic pressure?

$\endgroup$
0

1 Answer 1

1
$\begingroup$

$\gamma$ should not be number of particles, but molar amount of particles. Usually, $\gamma /V$ is replaced by amount concentration $c$. Additionally, for the case of electrolytes, there is the van't Hoff factor $i$ related to dissociation (>1) or association (<1) of particles.

So $\Pi = icRT$.

Osmotic pressures are at equilibrium equal only if hydrostatic pressures are equal. If a solution has osmotic pressure $\pu{1 atm}$, but has hydrostatic pressure greater by $\pu{1 atm}$ compared to pure water, osmosis does not occur.

If there is pure water and some solution on each side of semipermeable membrane, then ongoing osmosing causes growing difference of hydrostatic pressure that acts against the osmosis and finally balances it.


Response to feedback:

In biology, there is aside of osmolarity also tonicity. Osmolarity considers all solutes, tonicity just those not able to pass the membrane.

If osmolarity is equal, but tonicity is not, there is not equilibrium yet.

The same osmolarity but different tonicity leads to diffusion of solutes that takes with them solvent to rebalance osmolarity.


I read your updated response...I'm still not sure I understand why D is the correct answer to the U-tube problem presented in this post. Your responses so far would make me think that equality of osmolarity at equilibrium DOES NOT occur. I feel like there must be an assumption here about how the solution is sufficiently dilute and, therefore, no meaningful quantity of additional hydrostatic pressure is accumulated on the sucrose side.

If glucose concentrations are different, glucose would diffuse until they are the same. If summary glucose + sucrose concentrations are different, water would diffuse until they are the same.

Note that for $\pu{1 osmol/L}$, osmotic pressure is about $\pu{25 atm}$, or $\pu{250 m}$ of water column. The effect caused by hydrostatic pressure difference is therefore often negligible.

I think I must have a misunderstanding. Does your statement about the osmotic pressure mean that, at equilibrium conditions, for a U-tube with column radius $R$ meters, a net volume of $\pi R^2 \cdot 250 \pu{m3}$ of water flows from the right side (referring to this link ) of the U-tube to the left side??

No, it has nothing with volume nor flow. I see now it was rather a misleading, tanget info.

It means, that if you had "water||$\pu{1 osmol/L}$ of solute not passing membrane, and if the right part of U-tube had water level 250 m above the water level (or accordingly less based on density), then osmosis would not be ongoing, as counteracted by the hydrostatic pressure.

By other words, kinetic pressure of water molecules, decreased by solute presence, is counterbalanced by hydrostatic pressure. but there is low chance the membrane would survive such hydrostatic pressure.


If we take the meantioned U-tube, with passing glucose and not passing sucrose,

  • with 2 M sucrose + 1 M glucose on one side
  • with 1 M sucrose + 2 M glucose on the other side
  • with the equal initial volume

then

  • As the membrane is permeable for glucose, the final glucose concentration is 1.5 M, regardless of the final volumes on each side.
  • As the osmolarity on both sides must be the same (otherwise water would flow to balance them), sucrose concentration must be 1.5 M on both sides as well.
  • The same sucrose concentration is achieved by migration of water together with glucose, What increases glucose concentration and decreases sucrose concentration on the receiving side.
  • The equilibrium osmolarity and tonicity at equilibrium is the same.
  • The eventual effect of the optional external pressure is like $\Delta h \rho g = \Delta c RT$, but is mostly negligible.

The macroscopic conditions of osmotic equilibrium are:

  • concentrations (therefore chemical potential) of solutes, for which the membrane is permeable, are the same on both sides.
  • $\Delta h \rho g = \Delta c RT$, what means that
    • At zero difference of hydrostatic pressure, osmolarity on both sides is the same.
    • With the hydrostatic pressure difference, there is the difference in the equivalent osmolarity $\Delta c = \dfrac{\rho g}{RT} \Delta h$. This way is the solvent chemical potential of water(solvent) the same.
    • Tonicity is the same, or follows the same pattern as osmolarity wrt hydrostatic pressure difference.

$\Pi = h \rho g$ is then relation of the height of solution column of osmolarity $\Pi$, that is needed to stop osmosis of pure water on the other membrane side.


For components passing the membrane: $$\mu_{i,\text{L}} = \mu_{i,\text{R}}$$ $$c_{i,\text{L}} \approx c_{i,\text{R}} $$

For solvent water(w), solute(s) and sides left(L) and right(R):

$$\mu_\text{w,L} = \mu_\text{w,R}$$

If taken $a_i \approx x_i$:

$$\mu^\circ_\text{w} + V_\text{M,w}p_\text{L} + RT \ln{x_\text{w,L}}= \mu^\circ_\text{w} + V_\text{M,w}p_\text{R} + RT \ln{x_\text{w,R}}$$

$$V_\text{M,w}p_\text{L} + RT \ln{x_\text{w,L}}= V_\text{M,w}p_\text{R} + RT \ln{x_\text{w,R}}$$

$$V_\text{M,w}p_\text{L} + RT \ln{(1 - x_\text{s,L})}= V_\text{M,w}p_\text{R} + RT \ln{(1- x_\text{s,R})}$$

for $x_\text{s,R} \ll 1$ and $x_\text{s,R} \ll 1$:

$$\ln{(1- x_\text{s,R})} \approx = - x_\text{s,R}$$ $$\ln{(1- x_\text{s,L})} \approx = - x_\text{s,L}$$

$$V_\text{M,w}p_\text{L} - RT x_\text{s,L}= V_\text{M,w}p_\text{R} - RT x_\text{s,R}$$

if $x_i=ac_i$ then

$$V_\text{M,w}p_\text{L} - RTac_\text{s,L}= V_\text{M,w}p_\text{R} - RT ac_\text{s,R}$$

$$V_\text{M,w}(p_\text{L}-p_\text{R}) = RTa (c_\text{s,L} - c_\text{s,R})$$

$\endgroup$
5
  • $\begingroup$ I see that you just posted a very recent edit, but I will focus my attention on your prior edit ...the section titled: "Macroscopic conditions of osmotic equilibrium". Specifically, your comments here make me even FURTHER doubt the answer to the aforementioned problem about sucrose (biology.stackexchange.com/questions/113498/…). $\endgroup$
    – S.C.
    Commented Nov 18 at 18:02
  • $\begingroup$ Given that the left side has a larger concentration of sucrose (than the right side), net water movement will take place and hydrostatic pressure correspondingly accumulates in that left compartment (i.e. the water level is raised on the left side). Accordingly, this added hydrostatic pressure will offset the ability to balance the left and right osmolarities of sucrose by $\Delta c = \dfrac{\rho g}{RT} \Delta h$. $\endgroup$
    – S.C.
    Commented Nov 18 at 18:02
  • $\begingroup$ That is to say that, at equilibrium, the left side's sucrose osmolarity is $1.5+\Delta c$, and the right side's sucrose osmolarity is $1.5 - \Delta c$. Is this correct? If so, is the argument something like, "$\Delta h$ is negligibly small and, therefore, so, too, is $\Delta c$, in which case $1.5+\Delta c \approx 1.5$ and $1.5-\Delta c \approx 1.5$" $\endgroup$
    – S.C.
    Commented Nov 18 at 18:02
  • $\begingroup$ 1.5 +/- Δc/2 // And yes, your conclusion is otherwise right. Remember that in science, zero is not zero, but the value interval statistically not significantly different to zero. Δh can be managed to be minimal and c itself is subject to uncertainty as well. Δh 2.5 cm means dosing of respective sugar wit accuracy 0.01% and consider unknown content of absorbed moisture and weighting errors.. $\endgroup$
    – Poutnik
    Commented Nov 18 at 18:21
  • 1
    $\begingroup$ c1 = c0 + Δc1, c2 = c0 + Δc2, Δc = c2-c1 = Δc2 - Δc1. // [Δc2|/|Δc1| = V1/v2. // But those are rather details, mostly below the resolution level of the big picture of osmosis. $\endgroup$
    – Poutnik
    Commented Nov 18 at 18:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.