That's why one always uses standard states. To define a standard state for dG is equivalent to stating the units of Kd.

Thus, if you use Kd [L/g], this means that the dG is defined with respect to a standard state - "theoretical" ideal solution with concentration 1 mol/L and "theoretical" ideal adsorption layer state with adsorption 1 mol/g. Or 1 mg/L and 1mg/g, or any other couple of units leading to Kd [L/g].

Changing the standard state of dG is the result of changing the units of Kd. dH is usually independent of the conversion, as long as the conversion factor is independent of T. However, dS always contains a unit term that has little physical meaning.

If the surface is well defined, then it is a good idea to use Kd [m] (i.e. standard state mol/m^3 in the bulk and mol/m^2 at the surface). This allows one to compare Kd's & dG's. If the adsorption length is known, then use the dimensionless Kd/delta. dG = RTln(Kd/delta) is the only option which makes your dG equal to the micromechanical adsorption free energy. 

Hallo,

Other alternatives I have also come across is employing dimensionless,Equilibrium constant, Kc, instead of Kd,  given as the ratio of solid phase concentration (Ca) in mg/L (or any units of concentration) to the equilibrium concentration (Ce).

Check this paper.

Dear Radomir Slavchov,

When I use Kd in the unit of L/g the Kd values becomes lower than 1 such as 0,4672 etc and this causes dG to become positive. However when I use Kd values as mL/g, then it dG becomes negative, meaning that the adsorption is occured spontaneously. I prefered to use Kd as mL/g, because I used 0.2 g adsorbent / 25 mL adsorbate solution; meaning the units are given as g and mL. Thus it caused dG to become negative.

The term "occur spontaneously" refers to a rather strange process, namely: there's a standard theoretical ideal solution of concentration 1 mol/L, and a standard theoretical ideal monolayer of adsorption 1 mol/g. One puts them in contact. If dG < 0, matter will spontaneously adsorb from the 1 mol/L solution to the 1 mol/g surface. If dG > 0, matter will desorb. That is, the sign of dG is not very useful and it makes little sense to comment on it.

"Spontaneous adsorption" means: if there is a solution of C = X and adsorption layer of Gamma = Y, on its way to equilibrium, will the system see desorption or adsorption? The answer depends on both X and Y, obviously. By choosing your units, you choose X and Y.

The best approach is to use Kd = qe/ce. The unit of your Kd now is L/g( i.e mg/g divide by mg/L). Convert the Kd in L/g to a dimensionless quantity by multiplying by molar mass of the element in g/mol, 15.56 mol/L. That is L/g * g/mol * 15.56 mol/L = dimensionless Kd. Contact me on [email protected] for better explanations.

Using Cad/Ce might not be a dimensionless quantity in all sense because Cad is concentration of adsorbate in adsorbent in mg/g and not mg/L hence you might not have a dimensionless quantity

The Kd is in L/g but should be converted to be dimensionless so that the units of dG would be as it should.

We just developed a small program in python to convert DelG to Kd values more efficiently:

http://shityakovlab.uphero.com/hardware-and-software.html

I recommend you this paper.

Kd : Distribution coefficient is defined as the ratio of equilibrium concentrations C of a dissolved test substance in a two phase system consisting of a sorbent (soil or sewage sludge) and an aqueous phase; it is a dimensionless value when concentrations in both phases are expressed on a weight/weight base. In case the concentration in the water phase is given on a weight/volume base the dimension then is ml⋅/g. Kd can vary with sorbent properties and can be concentration dependent.

Hope it will help

This paper will be of help

Is Kd constant for Isothermal process?