Hello there! Welcome to our deep dive into the world of strong electrolytes. You've probably learned about "ideal solutions," where we pretend particles don't interact. Well, in the real world of ions—charged particles in a solvent—that assumption fails spectacularly! Ions attract and repel each other strongly. This means their "effective concentration," or activity ($a_i$), is very different from their actual concentration (molality $m_i$ or molarity $c_i$).
The whole point of the Debye-Hückel Theory (developed in 1923) is to provide a physical model to explain *why* and *how* this happens. It's our first great attempt to quantify the non-ideal behavior of ionic solutions.
Think of it this way: In an ideal gas, every molecule is "on its own." In an ionic solution, no ion is ever truly alone. It's like being in a crowded room where you're constantly being jostled and pulled by everyone around you. The Debye-Hückel theory is our way of describing that "crowd."
The central concept is the ionic atmosphere or ionic cloud.
Imagine we pick one specific ion, say a positive ion $j$, and call it our "central ion."
This atmosphere effectively "screens" the central ion. From a distance, its positive charge looks weaker than it really is because it's partially neutralized by its negative cloud. This screening is the physical origin of non-ideal behavior. Because of this screening, the ion's "activity" is lower than its concentration.
Before Debye and Hückel, the chemist G.N. Lewis empirically found that the non-ideal effects in solutions didn't just depend on concentration, but depended *more* on highly charged ions. He invented a quantity called ionic strength ($I$) to capture this.
$$ I = \frac{1}{2} \sum_i c_i z_i^2 $$Where $c_i$ is the molar concentration of ion $i$ and $z_i$ is its charge. Notice the $z_i^2$ term—this means a $Mg^{2+}$ ion (charge +2) contributes $2^2 = 4$ times as much to the ionic strength as an $Na^+$ ion (charge +1) at the same concentration. The Debye-Hückel theory would later provide the theoretical proof for why this $I$ is the correct quantity to use.
To build a mathematical model, we need to make some simplifying assumptions. These are the famous "Debye-Hückel assumptions":
This last assumption is what allows us to "linearize" the math, as we'll see next. It's also why the theory is a "limiting law"—it only becomes perfectly accurate at the limit of infinite dilution.
Our goal is to find the electric potential, $\psi(r)$, at any distance $r$ from our central ion $j$. We do this by combining two powerful pieces of physics.
Poisson's equation relates the electric potential $\psi$ at some point to the charge density $\rho(r)$ at that point. It's a fundamental law of electrostatics. For a spherically symmetric system (which our ion-in-a-cloud is), it is:
$$ \nabla^2 \psi = \frac{1}{r^2} \frac{d}{dr} \left( r^2 \frac{d\psi(r)}{dr} \right) = -\frac{\rho(r)}{\epsilon} $$Where $\epsilon = \epsilon_0 \epsilon_r$ is the permittivity of the solvent. This equation just says: "If you have a charge density $\rho$, you will have a potential $\psi$."
Now we need an expression for the charge density $\rho(r)$. The ions aren't fixed; they are distributed according to thermal energy. The Boltzmann distribution tells us the concentration of an ion $i$, $n_i(r)$, at a distance $r$ where the potential energy is $w_i(r)$.
The work (potential energy) to bring an ion $i$ with charge $z_i e$ to a point with potential $\psi(r)$ is $w_i(r) = z_i e \psi(r)$. So:
$$ n_i(r) = n_i^0 \exp\left(-\frac{w_i(r)}{k_B T}\right) = n_i^0 \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) $$Where $n_i^0$ is the bulk concentration (far away, where $\psi=0$).
The total charge density $\rho(r)$ is just the sum of all ion concentrations multiplied by their charges:
$$ \rho(r) = \sum_i n_i(r) z_i e = \sum_i n_i^0 z_i e \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) $$We combine Step 1 and Step 2 by plugging our expression for $\rho(r)$ into the Poisson equation:
$$ \nabla^2 \psi = -\frac{1}{\epsilon} \sum_i n_i^0 z_i e \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) $$This is a disaster. It's a complex non-linear differential equation that is extremely difficult to solve.
Here is where the "dilute solution" assumption saves us. We assume that $|z_i e \psi(r)| \ll k_B T$. This means the exponent is very small.
We can then use the Taylor expansion $e^{-x} \approx 1 - x$ for small $x$:
$$ \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) \approx 1 - \frac{z_i e \psi(r)}{k_B T} $$Now, let's plug this approximation back into our charge density $\rho(r)$ equation:
$$ \rho(r) \approx \sum_i n_i^0 z_i e \left(1 - \frac{z_i e \psi(r)}{k_B T}\right) $$ $$ \rho(r) \approx \sum_i n_i^0 z_i e - \frac{e^2 \psi(r)}{k_B T} \sum_i n_i^0 z_i^2 $$The first term, $\sum_i n_i^0 z_i e$, is the total charge in the bulk solution, which must be zero due to electroneutrality. So it vanishes! We are left with:
$$ \rho(r) \approx -\left( \frac{e^2 \sum_i n_i^0 z_i^2}{k_B T} \right) \psi(r) $$Now we plug *this* simple $\rho(r)$ back into Poisson's equation:
$$ \nabla^2 \psi = -\frac{1}{\epsilon} \left[ -\left( \frac{e^2 \sum_i n_i^0 z_i^2}{k_B T} \right) \psi(r) \right] $$ $$ \nabla^2 \psi = \left( \frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T} \right) \psi(r) $$This is a *much* friendlier equation. We define the entire cluster of constants in the parenthesis as $\kappa^2$ (kappa-squared).
$$ \kappa^2 = \frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T} $$So the Linearized Poisson-Boltzmann Equation is simply:
$$ \nabla^2 \psi = \kappa^2 \psi $$The solution to $\nabla^2 \psi = \kappa^2 \psi$ in spherical coordinates (with the boundary condition that $\psi \to 0$ as $r \to \infty$) is:
$$ \psi(r) = A' \frac{e^{-\kappa r}}{r} $$We find the constant $A'$ by knowing that as $r \to 0$, the potential must be dominated by the central ion $j$ itself, i.e., $\psi(r) \to \frac{z_j e}{4 \pi \epsilon r}$ (Coulomb's Law). This gives $A' = \frac{z_j e}{4 \pi \epsilon}$.
So, the final solution for the potential at a distance $r$ from the central ion is:
$$ \psi(r) = \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r} $$This is a beautiful result. It's just Coulomb's law, $\frac{z_j e}{4 \pi \epsilon r}$, multiplied by a "screening factor" $e^{-\kappa r}$.
The parameter $\kappa$ in that screening factor is crucial. Let's look at its units. Since $\kappa r$ must be unitless, $\kappa$ must have units of inverse distance (m⁻¹).
Therefore, $1/\kappa$ has units of distance. We call this the Debye length or the radius of the ionic atmosphere.
The Debye length $1/\kappa$ is the effective "thickness" of the ionic cloud. It's the distance over which the potential of the central ion is "screened" (drops by a factor of $1/e$).
Let's relate $\kappa$ back to the ionic strength $I = \frac{1}{2} \sum c_i z_i^2$. Remember $n_i^0$ is ions/m³ and $c_i$ is mol/L. $n_i^0 = c_i N_A \times 1000$.
$\sum_i n_i^0 z_i^2 = \sum_i (1000 c_i N_A) z_i^2 = 1000 N_A \sum_i c_i z_i^2 = 1000 N_A (2I)$.
$$ \kappa^2 = \frac{e^2 (2000 N_A I)}{\epsilon k_B T} = \left(\frac{2000 e^2 N_A}{\epsilon k_B T}\right) I $$This shows that $\kappa \propto \sqrt{I}$.
Example: For a 0.01 M solution of NaCl in water at 298 K, $I = 0.01$ M. The Debye length $1/\kappa$ is about 3.04 nm. For a 0.01 M solution of MgSO₄, $I = \frac{1}{2}(0.01 \times 2^2 + 0.01 \times (-2)^2) = 0.04$ M. Since $I$ is 4 times larger, $\kappa$ is 2 times larger, and $1/\kappa$ is 2 times smaller, about 1.52 nm.
We can prove the model is self-consistent. What is the total charge of the atmosphere $q_{cloud}$? We just integrate the charge density $\rho(r)$ from $r=0$ to $\infty$.
We know $\rho(r) = -\epsilon \kappa^2 \psi(r)$.
$$ q_{cloud} = \int_0^\infty \rho(r) 4\pi r^2 dr = \int_0^\infty \left( -\epsilon \kappa^2 \psi(r) \right) 4\pi r^2 dr $$ $$ q_{cloud} = \int_0^\infty -\epsilon \kappa^2 \left( \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r} \right) 4\pi r^2 dr $$ $$ q_{cloud} = -z_j e \kappa^2 \int_0^\infty r e^{-\kappa r} dr $$This standard integral $\int_0^\infty r e^{-\kappa r} dr$ evaluates to $1/\kappa^2$.
$$ q_{cloud} = -z_j e \kappa^2 \left( \frac{1}{\kappa^2} \right) = -z_j e $$This is a perfect result! It shows that the total charge in the atmosphere is exactly equal and opposite to the charge of the central ion.
Now we link this electrostatic model to thermodynamics. The goal is to find the activity coefficient $\gamma$.
The chemical potential $\mu_i$ of an ion $i$ is:
$$ \mu_i = \mu_i^{ideal} + \mu_i^{ex} = (\mu_i^\ominus + k_B T \ln c_i) + k_B T \ln \gamma_i $$The term $k_B T \ln \gamma_i$ is the excess chemical potential $\mu_i^{ex}$. It represents the *extra* energy of the ion due to all the non-ideal electrostatic interactions.
This $\mu_i^{ex}$ is exactly equal to the electrical work, $W_{el}$, done to "charge" a single ion $j$ from 0 to its full charge $z_j e$ *in the presence of its own atmosphere*.
The potential $\psi(r)$ at the central ion ($r \to 0$) can be split into two parts:
$$ \psi(r) \approx \underbrace{\frac{z_j e}{4 \pi \epsilon r}}_{\psi_{ion}} - \underbrace{\frac{z_j e \kappa}{4 \pi \epsilon}}_{\psi_{cloud}} \quad (\text{using } e^{-\kappa r} \approx 1 - \kappa r) $$The work of charging is done against the potential of the cloud, $\psi_{cloud}$.
$$ \mu_j^{ex} = W_{el} = \int_0^{z_j e} \psi_{cloud}(\lambda) d\lambda $$Where $\psi_{cloud}(\lambda) = - \frac{\lambda \kappa}{4 \pi \epsilon}$ is the potential of a cloud around a partially-charged ion $\lambda$.
$$ \mu_j^{ex} = \int_0^{z_j e} \left( - \frac{\kappa}{4 \pi \epsilon} \right) \lambda d\lambda = - \frac{\kappa}{4 \pi \epsilon} \left[ \frac{\lambda^2}{2} \right]_0^{z_j e} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon} $$Now we equate this to the thermodynamic definition:
$$ k_B T \ln \gamma_j = \mu_j^{ex} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon k_B T} $$ $$ \ln \gamma_j = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon k_B T} $$This is the activity coefficient for a single ion $j$, which we can't measure. We must find the mean ionic activity coefficient $\gamma_{\pm}$.
For an electrolyte $A_{\nu_+} B_{\nu_-}$, we have $\ln \gamma_{\pm} = \frac{1}{\nu_+ + \nu_-} (\nu_+ \ln \gamma_+ + \nu_- \ln \gamma_-)$.
Plugging in our $\ln \gamma_j$ expression and using the electroneutrality condition ($\nu_+ z_+^2 + \nu_- z_-^2 = \nu |z_+ z_-|$), we get:
$$ \ln \gamma_{\pm} = - \frac{e^2 |z_+ z_-| \kappa}{8 \pi \epsilon k_B T} $$We're at the final step. We take our expression for $\ln \gamma_{\pm}$ and substitute $\kappa \propto \sqrt{I}$ back into it.
$$ \ln \gamma_{\pm} = - \left[ \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right] \kappa $$ $$ \ln \gamma_{\pm} = - \left[ \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right] \sqrt{\frac{2000 e^2 N_A}{\epsilon k_B T}} \sqrt{I} $$This looks messy, but it's just a cluster of constants multiplied by $|z_+ z_-| \sqrt{I}$.
Converting from natural log ($\ln$) to base-10 log ($\log_{10}$), we get $\log_{10} \gamma_{\pm} = \ln \gamma_{\pm} / 2.303$.
$$ \log_{10} \gamma_{\pm} = - \left[ \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon k_B T)^{3/2}} \right] |z_+ z_-| \sqrt{I} $$We call that entire cluster of constants $A$.
$$ A = \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon_0 \epsilon_r k_B T)^{3/2}} $$This gives the famous Debye-Hückel Limiting Law:
$$ \log_{10} \gamma_{\pm} = -A |z_+ z_-| \sqrt{I} $$
For water at 298 K (25 °C), $A \approx 0.509 \text{ L}^{1/2}\text{mol}^{-1/2}$.
This law is a tremendous achievement. It predicts that a plot of $\log \gamma_{\pm}$ versus the square root of the ionic strength ($\sqrt{I}$) should be a straight line with a slope of $-A |z_+ z_-|$.
We can't measure $\gamma_{\pm}$ directly, but we can measure it indirectly from cell potentials (EMF), solubility measurements, or freezing point depression.
When we do this and plot the experimental data, we find the theory works perfectly... but only for extremely dilute solutions (e.g., $I < 0.01$ M for a 1:1 electrolyte). At higher concentrations, the experimental data deviates significantly from the straight line.
Why does the Limiting Law fail? Because our assumptions fail at higher concentrations.
We can improve the model by fixing the second assumption. We re-solve the Poisson-Boltzmann equation, but this time we say that the ionic atmosphere can only begin at a distance $a$ from the central ion's center. This $a$ is the ion-size parameter, representing the mean hydrated radius.
This modification (the math is a bit more involved) leads to a new potential for the cloud:
$$ \psi_{cloud}(a) = - \frac{z_j e \kappa}{4 \pi \epsilon (1 + \kappa a)} $$Notice that if $a \to 0$ (point charge), we get our old expression back. The $1 + \kappa a$ term in the denominator is the correction for finite ion size.
Following the exact same thermodynamic derivation as before, this $1 + \kappa a$ term just tags along. This leads to the Extended Debye-Hückel Equation:
$$ \log_{10} \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + \kappa a} $$Since $\kappa \propto \sqrt{I}$, we can write $\kappa a = (B \sqrt{I}) a$, where $B$ is another constant related to the solvent properties ($B \approx 0.329 \text{ L}^{1/2}\text{mol}^{-1/2}\text{Å}^{-1}$ for water at 298 K, if $a$ is in Ångströms).
$$ \log_{10} \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + B a \sqrt{I}} $$
This equation works much, much better, often giving good agreement up to $I \approx 0.1$ M. The ion-size parameter $a$ is generally treated as an adjustable "fitting" parameter, typically a few Ångströms.
Final Thought: The Debye-Hückel theory is a bit like the "ideal gas law" for ions. It's a "limiting law" that describes an idealized state (infinite dilution) perfectly. It provides the essential physical framework for understanding ion-ion interactions, even if we need more complex, empirical models (like the Brønsted-Guggenheim or Pitzer equations) to describe real-world, concentrated solutions.
Alright, let's go through these questions one by one. This is the core of physical electrochemistry, so we'll take our time to build the concepts step-by-step, just like a tutorial.
Question: Discuss the fundamental assumptions of the Debye-Hückel theory of strong electrolytes. Starting from the Poisson equation and the Boltzmann distribution function, derive the linearized Poisson-Boltzmann (PB) equation. Why is the linearization step necessary and under what condition (in terms of energy) is it valid?
The Debye-Hückel (DH) theory is a model to explain why and how the activity coefficients of ions in a solution deviate from the ideal value of 1. To make the problem solvable, Debye and Hückel made several key assumptions:
Our goal is to find the electric potential, $\psi(r)$, as a function of distance $r$ from a central reference ion $j$. We do this by combining two powerful equations.
1. The Poisson Equation: This equation from classical electrostatics relates the electric potential $\psi$ at a point to the local charge density $\rho(r)$ at that same point. For a spherically symmetric system (like an ion in its cloud), it is:
$$ \nabla^2 \psi = \frac{1}{r^2} \frac{d}{dr} \left( r^2 \frac{d\psi(r)}{dr} \right) = -\frac{\rho(r)}{\epsilon} $$Where $\epsilon = \epsilon_0 \epsilon_r$ is the permittivity of the solvent.
2. The Boltzmann Distribution: This equation from statistical mechanics tells us how the mobile ions are distributed in space due to a balance between electrostatic potential energy and thermal (kinetic) energy. The local concentration $n_i(r)$ of any ion $i$ at a distance $r$ is:
$$ n_i(r) = n_i^0 \exp\left(-\frac{w_i(r)}{k_B T}\right) $$Where $n_i^0$ is the bulk concentration (far from the ion) and $w_i(r)$ is the potential energy to bring that ion from infinity to the distance $r$. This energy is $w_i(r) = z_i e \psi(r)$, where $z_i e$ is the charge of the ion. So:
$$ n_i(r) = n_i^0 \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) $$3. Finding Charge Density $\rho(r)$: The local charge density is the sum of the concentrations of all ion types, multiplied by their respective charges:
$$ \rho(r) = \sum_i n_i(r) (z_i e) = \sum_i n_i^0 z_i e \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) $$4. The Full Poisson-Boltzmann (PB) Equation: We substitute this expression for $\rho(r)$ back into the Poisson equation. This gives the full PB equation:
$$ \nabla^2 \psi = -\frac{1}{\epsilon} \sum_i n_i^0 z_i e \exp\left(-\frac{z_i e \psi(r)}{k_B T}\right) $$This is a non-linear, second-order differential equation and is very difficult to solve.
Why is linearization necessary? Because the full PB equation is mathematically intractable for a simple, analytical solution. We need to simplify it to get a usable result.
We use the dilute solution assumption. In a very dilute solution, the ions are far apart, so the potential $\psi(r)$ is very small. This means the electrostatic potential energy $z_i e \psi(r)$ is much, much smaller than the average thermal energy $k_B T$ that keeps the ions moving.
Condition for validity: $|z_i e \psi(r)| \ll k_B T$.
Because the exponent $x = \frac{z_i e \psi(r)}{k_B T}$ is very small ($x \ll 1$), we can use the Taylor series expansion for $e^{-x}$ and keep only the first two terms:
$$ e^{-x} \approx 1 - x $$Applying this approximation to our $\rho(r)$ equation:
$$ \rho(r) \approx \sum_i n_i^0 z_i e \left(1 - \frac{z_i e \psi(r)}{k_B T}\right) $$Distributing the terms:
$$ \rho(r) \approx \sum_i n_i^0 z_i e - \left( \frac{e^2 \psi(r)}{k_B T} \right) \sum_i n_i^0 z_i^2 $$The first term, $\sum_i n_i^0 z_i e$, is the net charge density in the bulk solution. Since the solution as a whole is electrically neutral, this term is zero. This leaves us with a simple, linear relationship between $\rho(r)$ and $\psi(r)$:
$$ \rho(r) \approx - \left( \frac{e^2 \sum_i n_i^0 z_i^2}{k_B T} \right) \psi(r) $$5. The Linearized PB Equation: We plug this simplified $\rho(r)$ back into the Poisson equation:
$$ \nabla^2 \psi = -\frac{1}{\epsilon} \left[ - \left( \frac{e^2 \sum_i n_i^0 z_i^2}{k_B T} \right) \psi(r) \right] $$ $$ \nabla^2 \psi = \left( \frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T} \right) \psi(r) $$This is the Linearized Poisson-Boltzmann Equation. It's a standard differential equation and is much easier to solve.
Question: Solve the linearized Poisson-Boltzmann equation for a spherically symmetric ionic atmosphere around a central ion $j$. Derive the expressions for the electrical potential $\psi_j(r)$ as a function of distance $r$ and identify the term for the Debye length ($1/\kappa$). What is the physical significance of the Debye length?
From Question 1, we have our Linearized PB equation:
$$ \nabla^2 \psi = \left( \frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T} \right) \psi(r) $$Let's simplify this. We'll define the entire cluster of constants (which are all positive) as $\kappa^2$ (kappa-squared):
$$ \kappa^2 \equiv \frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T} $$So our equation becomes:
$$ \nabla^2 \psi = \kappa^2 \psi \quad \text{or} \quad \frac{1}{r^2} \frac{d}{dr} \left( r^2 \frac{d\psi}{dr} \right) = \kappa^2 \psi $$This is a standard linear, second-order differential equation. The general solution is:
$$ \psi(r) = C_1 \frac{e^{-\kappa r}}{r} + C_2 \frac{e^{\kappa r}}{r} $$Where $C_1$ and $C_2$ are integration constants. We find them by applying our physical boundary conditions:
By comparing these two, we can find $C_1$:
$$ \frac{C_1}{r} = \frac{z_j e}{4 \pi \epsilon r} \implies C_1 = \frac{z_j e}{4 \pi \epsilon} $$Substituting $C_1$ back into our solution, we get the final expression for the potential at any distance $r$ from the central ion $j$:
$$ \psi_j(r) = \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r} $$This is a beautiful result. It's just Coulomb's Law ($\frac{z_j e}{4 \pi \epsilon r}$) multiplied by an exponential "screening factor" ($e^{-\kappa r}$). This factor shows how the ionic atmosphere screens (dampens) the charge of the central ion.
The term $\kappa$ is defined as:
$$ \kappa = \sqrt{\frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T}} $$Let's check the units. The exponent $\kappa r$ must be dimensionless. Since $r$ has units of distance (meters), $\kappa$ must have units of inverse distance (m⁻¹).
This means that $1/\kappa$ has units of distance. This quantity is called the Debye length or the radius of the ionic atmosphere.
Physical Significance: The Debye length $1/\kappa$ represents the effective "thickness" of the ionic atmosphere. It is the distance over which the potential of the central ion is significantly "screened" by its cloud.
Mathematically, when $r = 1/\kappa$, the potential $\psi_j(r)$ has dropped to $1/e$ (about 37%) of the value it would have had without any screening.
Question: Define the "ionic cloud" or "ionic atmosphere." Derive an expression for the charge density ($\rho_j$) in the ionic atmosphere at a distance $r$ from the central ion. Show that the total charge of the ionic atmosphere is equal in magnitude and opposite in sign to the charge of the central ion.
The ionic cloud (or ionic atmosphere) is the time-averaged, spherically symmetric region of net charge that surrounds a central ion in an electrolyte solution.
Due to electrostatic forces, a central positive ion will, on average, have more negative ions in its immediate vicinity than positive ions. This creates a "cloud" of net negative charge. Conversely, a central negative ion will be surrounded by a "cloud" of net positive charge. This cloud is not static; ions are constantly moving due to thermal energy. The cloud is a statistical average of this distribution.
We already found an expression for the charge density $\rho(r)$ during the derivation of the Linearized PB equation (in Question 1, Part C).
We started with the Linearized PB equation: $\nabla^2 \psi = \kappa^2 \psi$
And we also know from the Poisson equation that: $\nabla^2 \psi = -\rho(r)/\epsilon$
By simply equating these two, we get a direct relationship between $\rho(r)$ and $\psi(r)$:
$$ \kappa^2 \psi(r) = -\frac{\rho(r)}{\epsilon} \implies \rho(r) = -\epsilon \kappa^2 \psi(r) $$Now, we just substitute the expression for $\psi_j(r)$ that we found in Question 2:
$$ \psi_j(r) = \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r} $$Plugging this in:
$$ \rho_j(r) = -\epsilon \kappa^2 \left( \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r} \right) $$ $$ \rho_j(r) = - \frac{z_j e \kappa^2}{4 \pi r} e^{-\kappa r} $$This is the expression for the charge density at a distance $r$ from the central ion $j$. Note that $\rho_j(r)$ always has the opposite sign of the central ion charge $z_j e$, which makes physical sense.
To find the total charge of the atmosphere, we must integrate the charge density $\rho_j(r)$ over all of space, from the ion's "surface" ($r=0$ for a point charge) to infinity. We integrate over a series of thin spherical shells of volume $dV = 4\pi r^2 dr$:
$$ q_{cloud} = \int_{0}^{\infty} \rho_j(r) dV = \int_{0}^{\infty} \rho_j(r) (4\pi r^2 dr) $$Substitute our expression for $\rho_j(r)$:
$$ q_{cloud} = \int_{0}^{\infty} \left( - \frac{z_j e \kappa^2}{4 \pi r} e^{-\kappa r} \right) (4\pi r^2 dr) $$Now, we simplify by canceling terms:
$$ q_{cloud} = - z_j e \kappa^2 \int_{0}^{\infty} r e^{-\kappa r} dr $$We just need to solve this standard integral. We can use integration by parts ($\int u dv = uv - \int v du$):
The first term, $\left[ -\frac{r}{\kappa}e^{-\kappa r} \right]_{0}^{\infty}$, is zero at both limits. (At $r \to \infty$, $e^{-\kappa r}$ goes to zero faster than $r$ goes to infinity. At $r=0$, the term is zero).
The second term becomes:
$$ = \frac{1}{\kappa} \int_{0}^{\infty} e^{-\kappa r} dr = \frac{1}{\kappa} \left[ -\frac{1}{\kappa}e^{-\kappa r} \right]_{0}^{\infty} $$ $$ = -\frac{1}{\kappa^2} \left[ e^{-\kappa r} \right]_{0}^{\infty} = -\frac{1}{\kappa^2} (e^{-\infty} - e^{0}) = -\frac{1}{\kappa^2} (0 - 1) = \frac{1}{\kappa^2} $$So, the definite integral $\int_{0}^{\infty} r e^{-\kappa r} dr$ is equal to $1/\kappa^2$.
Plugging this result back into our $q_{cloud}$ equation:
$$ q_{cloud} = - z_j e \kappa^2 \left( \frac{1}{\kappa^2} \right) $$ $$ q_{cloud} = - z_j e $$This is a crucial and elegant result. It shows that the model is self-consistent. The total charge of the ionic atmosphere ($q_{cloud}$) is exactly equal in magnitude and opposite in sign to the charge of the central ion ($z_j e$). The atmosphere perfectly neutralizes the central ion.
Question: Explain the process of calculating the activity coefficient from the Debye-Hückel theory. Show how the excess chemical potential ($\mu_i^{ex}$) of an ion $i$, representing the electrical work of charging, is related to the potential of the ionic cloud at the central ion.
This is where we connect the world of electrostatics (potentials, clouds) to the world of thermodynamics (chemical potential, activity).
The chemical potential $\mu_i$ of an ion $i$ in a real solution is:
$$ \mu_i = \mu_i^{\text{ideal}} + \mu_i^{\text{excess}} $$In thermodynamic terms, we define the activity $a_i$ and activity coefficient $\gamma_i$ to account for this excess term:
$$ \mu_i = \mu_i^\ominus + k_B T \ln(a_i) = \mu_i^\ominus + k_B T \ln(c_i \gamma_i) $$ $$ \mu_i = (\mu_i^\ominus + k_B T \ln c_i) + k_B T \ln \gamma_i $$Comparing our two equations, we see:
$$ \mu_i^{\text{ideal}} = \mu_i^\ominus + k_B T \ln c_i $$ $$ \mu_i^{\text{excess}} = k_B T \ln \gamma_i $$So, if we can find a way to calculate the excess chemical potential $\mu_i^{ex}$ from our physical model, we can directly calculate the activity coefficient $\gamma_i$.
What is this excess chemical potential? It's the extra Gibbs free energy the ion has due to being in a non-ideal solution. In the DH model, all non-ideality is due to electrostatic interactions.
Therefore, $\mu_i^{ex}$ is simply the net electrical work ($W_{el}$) done to "create" that ion in the presence of its ionic atmosphere.
We can calculate this work using a "charging process." Imagine we have an uncharged ion at position $r=0$. We slowly "charge" it up from 0 to its final charge $z_j e$. The work done $dW$ for each small bit of charge $d\lambda$ we add is $dW = \psi(\lambda) d\lambda$, where $\psi(\lambda)$ is the potential at the ion's surface *caused by its own atmosphere* while it has a temporary charge $\lambda$.
So, the total work is: $W_{el} = \mu_j^{ex} = \int_{0}^{z_j e} \psi_{cloud}(\lambda) d\lambda$
We need to find $\psi_{cloud}$ at the central ion (at $r=0$). Our total potential $\psi_j(r)$ from Question 2 is the sum of the potential from the central ion itself, $\psi_{ion}$, and the potential from its atmosphere, $\psi_{cloud}$:
$$ \psi_j(r) = \psi_{ion}(r) + \psi_{cloud}(r) $$Let's expand our solution for $\psi_j(r)$ using the Taylor series $e^{-x} \approx 1 - x$ (since $r$ is very small, $\kappa r$ is also very small):
$$ \psi_j(r) = \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r} \approx \frac{z_j e}{4 \pi \epsilon r} (1 - \kappa r) $$ $$ \psi_j(r) \approx \underbrace{\frac{z_j e}{4 \pi \epsilon r}}_{\psi_{ion}(r)} - \underbrace{\frac{z_j e \kappa}{4 \pi \epsilon}}_{\psi_{cloud}(r \to 0)} $$Look at that! The total potential $\psi_j(r)$ near the ion's center is the sum of:
This $\psi_{cloud}$ is what the ion "feels" from its surroundings. This is what we need for our charging process.
The potential of the cloud, $\psi_{cloud}$, is proportional to the charge of the central ion. If the central ion only has a partial charge $\lambda$, the cloud's potential will be:
$$ \psi_{cloud}(\lambda) = - \frac{\lambda \kappa}{4 \pi \epsilon} $$Now we integrate this from 0 to the final charge $z_j e$:
$$ \mu_j^{ex} = W_{el} = \int_{0}^{z_j e} \left( - \frac{\lambda \kappa}{4 \pi \epsilon} \right) d\lambda $$ $$ \mu_j^{ex} = - \frac{\kappa}{4 \pi \epsilon} \int_{0}^{z_j e} \lambda d\lambda = - \frac{\kappa}{4 \pi \epsilon} \left[ \frac{\lambda^2}{2} \right]_{0}^{z_j e} $$ $$ \mu_j^{ex} = - \frac{\kappa}{4 \pi \epsilon} \left( \frac{(z_j e)^2}{2} - 0 \right) = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon} $$This is the final expression for the excess chemical potential of a single ion $j$.
The process is: $\mu_j^{ex} = W_{el} = \int \psi_{cloud} d\lambda$. By finding $\psi_{cloud}$ from our $\psi(r)$ solution, we can calculate $W_{el}$, which gives us $\mu_j^{ex}$. And since $\mu_j^{ex} = k_B T \ln \gamma_j$, we have found the activity coefficient.
Question: Starting from the expression for the excess chemical potential ($\mu_i^{ex}$), derive the Debye-Hückel Limiting Law (DHLL) for the mean ionic activity coefficient ($\log \gamma_{\pm}$). Clearly show all the steps and define the terms in the final expression: $\log \gamma_{\pm} = -A |z_+ z_-| \sqrt{I}$.
From Question 4, we have the excess chemical potential for a single ion $j$:
$$ \mu_j^{ex} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon} $$We also know the thermodynamic definition:
$$ \mu_j^{ex} = k_B T \ln \gamma_j $$Equating these gives the (unmeasurable) single-ion activity coefficient:
$$ k_B T \ln \gamma_j = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon} $$ $$ \ln \gamma_j = - \frac{z_j^2 e^2 \kappa}{8 \pi \epsilon k_B T} $$We can't measure $\gamma_j$, but we *can* measure the mean ionic activity coefficient, $\gamma_{\pm}$. For a general electrolyte $A_{\nu_+} B_{\nu_-}$ (e.g., for $Al_2(SO_4)_3$, $\nu_+=2$, $\nu_-=3$), the mean is defined as:
$$ \gamma_{\pm}^\nu = (\gamma_+^{\nu_+}) (\gamma_-^{\nu_-}) \quad \text{where } \nu = \nu_+ + \nu_- $$Taking the natural log:
$$ \nu \ln \gamma_{\pm} = \nu_+ \ln \gamma_+ + \nu_- \ln \gamma_- $$Now, substitute our expression for $\ln \gamma_j$ from Step 1:
$$ \nu \ln \gamma_{\pm} = \nu_+ \left( - \frac{z_+^2 e^2 \kappa}{8 \pi \epsilon k_B T} \right) + \nu_- \left( - \frac{z_-^2 e^2 \kappa}{8 \pi \epsilon k_B T} \right) $$Factor out the common terms:
$$ \nu \ln \gamma_{\pm} = - \left( \frac{e^2 \kappa}{8 \pi \epsilon k_B T} \right) (\nu_+ z_+^2 + \nu_- z_-^2) $$For any salt, the electroneutrality condition is $\nu_+ z_+ = -\nu_- z_-$ (or $|\nu_+ z_+| = |\nu_- z_-|$). We can use this:
Wait, $z_+$ and $z_-$ have opposite signs, so $-z_+ z_-$ is a positive number. A more direct way is $|z_+ z_-| = |z_+| |z_-|$. Let's use $\nu_+ z_+ = -\nu_- z_-$.
$$ \nu_+ z_+^2 = (-\nu_- z_-) z_+ = -\nu_- z_- z_+ $$ $$ \nu_- z_-^2 = (-\nu_+ z_+) z_- = -\nu_+ z_+ z_- $$ $$ \nu_+ z_+^2 + \nu_- z_-^2 = -z_+ z_- (\nu_+ + \nu_-) = -z_+ z_- \nu $$Since $z_+$ and $z_-$ have opposite signs, $z_+ z_-$ is a negative number (e.g., (+2)(-1) = -2). So $-z_+ z_-$ is a positive number. We write this positive number as $|z_+ z_-|$.
$$ \nu_+ z_+^2 + \nu_- z_-^2 = |z_+ z_-| \nu $$Let's plug this back in:
$$ \nu \ln \gamma_{\pm} = - \left( \frac{e^2 \kappa}{8 \pi \epsilon k_B T} \right) (|z_+ z_-| \nu) $$Cancel $\nu$ from both sides:
$$ \ln \gamma_{\pm} = - \left( \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right) \kappa $$This is great, but we want the law in terms of $I$. Recall the definitions of $\kappa$ and $I$:
$$ \kappa = \sqrt{\frac{e^2 \sum_i n_i^0 z_i^2}{\epsilon k_B T}} $$ $$ I = \frac{1}{2} \sum_i c_i z_i^2 \quad \text{(where } c_i \text{ is molarity, mol/L)} $$We need to relate $n_i^0$ (ions/m³) to $c_i$ (mol/L).
$$ n_i^0 (\text{ions/m³}) = c_i (\text{mol/L}) \times N_A (\text{ions/mol}) \times 1000 (\text{L/m³}) = 1000 N_A c_i $$So, $\sum_i n_i^0 z_i^2 = \sum_i (1000 N_A c_i) z_i^2 = 1000 N_A \sum_i c_i z_i^2 = 1000 N_A (2I) = 2000 N_A I$.
Now substitute this into the $\kappa$ expression:
$$ \kappa = \sqrt{\frac{e^2 (2000 N_A I)}{\epsilon k_B T}} = \left( \sqrt{\frac{2000 e^2 N_A}{\epsilon k_B T}} \right) \sqrt{I} $$Plug this new expression for $\kappa$ back into our $\ln \gamma_{\pm}$ equation:
$$ \ln \gamma_{\pm} = - \left( \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right) \left( \sqrt{\frac{2000 e^2 N_A}{\epsilon k_B T}} \right) \sqrt{I} $$Let's group the constants:
$$ \ln \gamma_{\pm} = - \left[ \frac{e^3 \sqrt{2000 N_A}}{8 \pi (\epsilon k_B T)^{3/2}} \right] |z_+ z_-| \sqrt{I} $$We are almost there! We just need to convert from $\ln$ (base $e$) to $\log$ (base 10) using $\ln x = 2.303 \log_{10} x$:
$$ 2.303 \log_{10} \gamma_{\pm} = - \left[ \frac{e^3 \sqrt{2000 N_A}}{8 \pi (\epsilon k_B T)^{3/2}} \right] |z_+ z_-| \sqrt{I} $$ $$ \log_{10} \gamma_{\pm} = - \left[ \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon k_B T)^{3/2}} \right] |z_+ z_-| \sqrt{I} $$This entire cluster of constants depends *only* on the solvent (via $\epsilon$) and the temperature (via $T$). We call this constant $A$.
$$ A \equiv \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon_0 \epsilon_r k_B T)^{3/2}} $$This gives the final, famous Debye-Hückel Limiting Law:
$$ \log_{10} \gamma_{\pm} = -A |z_+ z_-| \sqrt{I} $$
Definitions:
Question: State the Debye-Hückel Limiting Law. Discuss in detail the methods for its experimental verification. Explain why the law is termed a "limiting law" and illustrate its success and failure by plotting $\log \gamma_{\pm}$ vs $\sqrt{I}$ for both theoretical and experimental data for a 1:1 and a 2:2 electrolyte.
The law states that for very dilute strong electrolyte solutions, the logarithm of the mean ionic activity coefficient ($\gamma_{\pm}$) is directly proportional to the product of the ion charges ($|z_+ z_-|$) and the square root of the ionic strength ($\sqrt{I}$).
The Law: $$ \log_{10} \gamma_{\pm} = -A |z_+ z_-| \sqrt{I} $$
This equation predicts that a plot of $\log \gamma_{\pm}$ vs $\sqrt{I}$ should be a straight line passing through the origin (since $\log(1) = 0$ at $I=0$) with a negative slope equal to $A |z_+ z_-|$.
We cannot measure $\gamma_{\pm}$ directly. Instead, we measure thermodynamic properties that *depend* on $\gamma_{\pm}$ and then calculate it.
It is called a "limiting law" because it is only strictly accurate at the limit of infinite dilution (i.e., as $I \to 0$).
The derivation relies on the linearization $e^{-x} \approx 1-x$, which is only valid when $x = |z_i e \psi| / k_B T$ is very small. This condition is only met in extremely dilute solutions. As concentration increases, this approximation breaks down, and the theory fails.
Here's how the plots of $\log \gamma_{\pm}$ vs $\sqrt{I}$ look:
Conclusion: The theory is a brilliant success in that it *perfectly* predicts the *initial slope* (the limiting behavior) for all electrolyte types. Its failure is that this range of validity is very small, especially for highly charged ions.
Question: The constant $A$ in the DHLL ($\log \gamma_{\pm} = -A |z_+ z_-| \sqrt{I}$) depends on the solvent and temperature. Derive the full expression for the constant $A$ in terms of fundamental constants ($e, N_A, k_B$) and solvent properties (dielectric constant $\epsilon_r$). Explain the physical origin of this dependence.
We already did the hard work in Question 5. We just need to isolate the constant $A$ from the final equation:
From Q5, Step 4, we had:
$$ \log_{10} \gamma_{\pm} = - \underbrace{\left[ \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon k_B T)^{3/2}} \right]}_{A} |z_+ z_-| \sqrt{I} $$So, the full expression for $A$ is:
$$ A = \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon k_B T)^{3/2}} $$We also know that the permittivity of the medium $\epsilon$ is $\epsilon = \epsilon_0 \epsilon_r$, where $\epsilon_0$ is the permittivity of free space (a fundamental constant) and $\epsilon_r$ is the relative permittivity (dielectric constant) of the solvent.
Substituting this in gives the most complete form:
$$ A = \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon_0 \epsilon_r k_B T)^{3/2}} $$This expression is for when $I$ is in Molarity (mol/L). A similar expression exists for molality ($m$), which also includes the solvent density $\rho_0$.
Let's look at the two key variables in $A$: $T$ and $\epsilon_r$.
1. Dependence on Temperature ($T$):
Temperature appears in the denominator as $(T)^{3/2}$. This means as $T$ increases, $A$ decreases.
Physical Reason: The entire DH theory is a battle between two forces:
When you increase the temperature $T$, you are giving the ions more kinetic energy. This thermal motion becomes more effective at breaking up the ordered ionic atmosphere. The solution behaves *more ideally*. A more ideal solution has $\gamma_{\pm}$ closer to 1, which means $\log \gamma_{\pm}$ is closer to 0. A smaller value of $A$ achieves exactly this.
2. Dependence on Dielectric Constant ($\epsilon_r$):
The dielectric constant $\epsilon_r$ also appears in the denominator as $(\epsilon_r)^{3/2}$. This means as $\epsilon_r$ increases, $A$ decreases.
Physical Reason: The dielectric constant is a measure of the solvent's ability to "shield" charges from each other.
This is why salts are *much* more non-ideal in solvents other than water.
Question: Critically evaluate the Debye-Hückel theory of activity coefficients. Discuss its major triumphs in explaining the behavior of dilute electrolyte solutions and its significant limitations that arise at higher concentrations.
The DH theory was a revolutionary achievement in physical chemistry. Its successes are fundamental to our understanding of electrolytes.
The theory's success is confined to very dilute solutions because its simplifying assumptions are only valid there. As concentration increases, these assumptions fail:
Question: Why does the Debye-Hückel Limiting Law fail at higher concentrations? Discuss the major modifications made to the theory to account for the behavior of more concentrated solutions, leading to the Extended Debye-Hückel Equation.
The DHLL ($\log \gamma_{\pm} = -A |z_+ z_-| \sqrt{I}$) fails as concentration increases primarily because its two most important assumptions are no longer physically realistic:
The first point (finite ion size) is the *easiest* to fix mathematically, and it leads directly to the Extended Debye-Hückel Equation.
The major modification is to stop treating ions as point charges.
Instead, we model the central ion as a "hard sphere" of radius $a$. This $a$ represents the mean distance of closest approach for other ions.
This changes our physical model in two ways:
In Question 2, we solved for the potential $\psi(r) = C_1 \frac{e^{-\kappa r}}{r}$ by assuming the ion was a point charge at $r=0$.
Now, we have a more complex boundary. The total potential for $r \ge a$ is still $\psi(r) = C_1 \frac{e^{-\kappa r}}{r}$. But to find $C_1$, we must match it to the potential *inside* the ion ($r < a$), which is more complex.
A simpler way (used by Debye and Hückel) is to modify the integration for the charging process (Question 4). Instead of integrating the cloud charge from $r=0$, we integrate from $r=a$ to infinity.
This new calculation for the electrical work $\mu_j^{ex}$ (which is shown in Q10) leads to a different result.
When this modification for a finite ion size $a$ is carried through the entire derivation, the final result for the excess chemical potential (Q4) gains a new denominator:
Original $\mu_j^{ex} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon}$
Modified $\mu_j^{ex} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon (1 + \kappa a)}$
This $1 + \kappa a$ term is the mathematical consequence of accounting for the ion's finite size.
When we carry *this* new $\mu_j^{ex}$ through the rest of the derivation (Q5), this $1 + \kappa a$ term simply follows along into the final equation for $\ln \gamma_{\pm}$:
$$ \ln \gamma_{\pm} = - \left( \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right) \frac{\kappa}{1 + \kappa a} $$Finally, we convert to $\log_{10}$ and substitute $A$ and $\sqrt{I}$ (from Q5) to get the Extended Debye-Hückel Equation:
$$ \log_{10} \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + \kappa a} $$Since $\kappa$ is also proportional to $\sqrt{I}$, we often group $\kappa a$ as $(B\sqrt{I})a$, where $B$ is another solvent/temp-dependent constant ($B = \kappa / \sqrt{I}$).
$$ \log_{10} \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + B a \sqrt{I}} $$This equation is a significant improvement and works reasonably well up to $I \approx 0.1$ M.
Question: Derive the expression for the mean ionic activity coefficient in the case of ionic clouds with finite-sized ions (the Extended Debye-Hückel equation). The final form is $\log \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + Ba\sqrt{I}}$. Clearly explain the physical significance of the ion-size parameter $a$ and the constant $B$.
The full derivation requires re-solving the PB equation with new boundary conditions, which is complex. A more direct "shortcut" is to re-calculate the electrical work (the charging process from Q4) with the new physical model.
Model: The central ion is a hard sphere of radius $a$. The ionic atmosphere (the cloud) can only exist for $r \ge a$.
1. Potential of the Cloud at the Ion: In Q4, we found the potential of the cloud at the central *point* ($r=0$) was $\psi_{cloud} = - \frac{z_j e \kappa}{4 \pi \epsilon}$. When we re-calculate this potential, but for a cloud that starts at $r=a$ instead of $r=0$, the new potential at the *surface* of the ion (at $r=a$) is found to be:
$$ \psi_{cloud}(a) = - \frac{z_j e \kappa}{4 \pi \epsilon (1 + \kappa a)} $$This is the key modification. The $1+\kappa a$ term in the denominator accounts for the fact that the ion "pushes" the cloud to start at $r=a$.
2. Electrical Work (Charging Process): Just as in Q4, the excess chemical potential $\mu_j^{ex}$ is the work to charge the ion (from 0 to $z_j e$) against the potential of its own cloud:
$$ \mu_j^{ex} = W_{el} = \int_{0}^{z_j e} \psi_{cloud}(\lambda, a) d\lambda $$Where $\psi_{cloud}(\lambda, a) = - \frac{\lambda \kappa}{4 \pi \epsilon (1 + \kappa a)}$ (assuming $a$ is constant and not dependent on $\lambda$).
$$ \mu_j^{ex} = \int_{0}^{z_j e} \left( - \frac{\lambda \kappa}{4 \pi \epsilon (1 + \kappa a)} \right) d\lambda $$ $$ \mu_j^{ex} = - \frac{\kappa}{4 \pi \epsilon (1 + \kappa a)} \int_{0}^{z_j e} \lambda d\lambda $$ $$ \mu_j^{ex} = - \frac{\kappa}{4 \pi \epsilon (1 + \kappa a)} \left[ \frac{\lambda^2}{2} \right]_{0}^{z_j e} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon (1 + \kappa a)} $$3. From $\mu_j^{ex}$ to $\gamma_{\pm}$: This is now identical to the steps in Q5, just with the $1 + \kappa a$ term tagging along.
$$ \mu_j^{ex} = k_B T \ln \gamma_j \implies \ln \gamma_j = - \frac{z_j^2 e^2 \kappa}{8 \pi \epsilon k_B T (1 + \kappa a)} $$Converting to the mean coefficient $\gamma_{\pm}$:
$$ \nu \ln \gamma_{\pm} = (\nu_+ \ln \gamma_+) + (\nu_- \ln \gamma_-) $$ $$ \nu \ln \gamma_{\pm} = - \left( \frac{e^2 \kappa}{8 \pi \epsilon k_B T (1 + \kappa a)} \right) (\nu_+ z_+^2 + \nu_- z_-^2) $$Using $\nu_+ z_+^2 + \nu_- z_-^2 = \nu |z_+ z_-|$ (from Q5):
$$ \ln \gamma_{\pm} = - \left( \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right) \frac{\kappa}{1 + \kappa a} $$4. Final Form: We now convert to $\log_{10}$ and substitute $\kappa$ in terms of $I$:
$$ \log_{10} \gamma_{\pm} = - \left[ \frac{1}{2.303} \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right] \frac{\kappa}{1 + \kappa a} $$We know from Q5 that $\log \gamma_{\pm} = -A |z_+ z_-| \sqrt{I}$ and $\ln \gamma_{\pm} = -(\text{const}) \kappa$.
The derivation from Q5 gives $\log \gamma_{\pm} = (\text{const}') \times \ln \gamma_{\pm}$
Let's use the constants $A$ and $B$ from the start.
$$ \log_{10} \gamma_{\pm} = \frac{\ln \gamma_{\pm}}{2.303} = \frac{1}{2.303} \left[ - \left( \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right) \frac{\kappa}{1 + \kappa a} \right] $$From Q5, $\kappa = \left( \sqrt{\frac{2000 e^2 N_A}{\epsilon k_B T}} \right) \sqrt{I}$. And $A = \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon k_B T)^{3/2}}$.
This is messy. Let's simplify. Let $A' = \left[ \frac{e^2}{8 \pi \epsilon k_B T} \right]$ and $B' = \sqrt{\frac{2000 e^2 N_A}{\epsilon k_B T}}$. So $\kappa = B' \sqrt{I}$.
$$ \ln \gamma_{\pm} = -A' |z_+ z_-| \frac{B' \sqrt{I}}{1 + a B' \sqrt{I}} $$ $$ \log_{10} \gamma_{\pm} = \frac{-A' B'}{2.303} |z_+ z_-| \frac{\sqrt{I}}{1 + a B' \sqrt{I}} $$If we define $A = \frac{A' B'}{2.303}$ and $B = B'/ \sqrt{1000 N_A}$ (to change units), we arrive at the standard form.
Let's just use the definitions:
$$ A = \left[ \frac{e^3 \sqrt{2000 N_A}}{2.303 \times 8 \pi (\epsilon_0 \epsilon_r k_B T)^{3/2}} \right] $$ $$ B = \kappa / \sqrt{I} = \sqrt{\frac{2000 e^2 N_A}{\epsilon_0 \epsilon_r k_B T}} \quad \text{(when I is in mol/L)} $$The expression $\ln \gamma_{\pm} = - \left( \frac{e^2 |z_+ z_-|}{8 \pi \epsilon k_B T} \right) \frac{\kappa}{1 + \kappa a}$ can be re-written as:
$$ \log_{10} \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + Ba\sqrt{I}} $$(Note: The definitions of $A$ and $B$ must be consistent. $B$ here is $\kappa/\sqrt{I}$ in units of L¹/² mol⁻¹/² m⁻¹ if $a$ is in m. Typically $B$ is given in Å⁻¹ L¹/² mol⁻¹/² so $a$ can be in Å).
Question: Compare and contrast the Debye-Hückel Limiting Law with the Extended Debye-Hückel equation. Discuss the role of the ion-size parameter $a$. How does the extended equation improve the agreement with experimental data, and what are its own limitations?
Let's do a side-by-side comparison.
| Feature | Debye-Hückel Limiting Law (DHLL) | Extended Debye-Hückel Equation | | :--- | :--- | :--- | | Equation | $\log \gamma_{\pm} = -A |z_+ z_-| \sqrt{I}$ | $\log \gamma_{\pm} = \frac{-A |z_+ z_-| \sqrt{I}}{1 + Ba\sqrt{I}}$ | | Ion Model | Ions are point charges with no volume ($a=0$). | Ions are hard spheres with a finite radius $a$. | | Plot vs $\sqrt{I}$ | Predicts a straight line with a negative slope. | Predicts a curve. The slope starts at $-A|z_+z_-|$ (same as DHLL) but becomes less negative as $I$ increases. | | Range of Validity | Extremely dilute solutions (e.g., $I < 0.005$ M for 1:1 salts). | Moderately dilute solutions (e.g., up to $I \approx 0.1$ M). | | Key Assumption | $a=0$ and $|z_i e \psi| \ll k_B T$ | $a > 0$, but still assumes $|z_i e \psi| \ll k_B T$ (linearization). |The parameter $a$ is the crucial modification.
Essentially, $a$ accounts for the fact that finite-sized ions "crowd" each other, which is a repulsive effect that *opposes* the electrostatic attraction of the cloud. This repulsive effect makes the solution behave slightly more ideally (moves $\gamma_{\pm}$ closer to 1) than the simple DHLL predicts.
The extended equation is an improvement, but it's not perfect. It still has its own limitations:
Question: Define and explain the physical significance of the following terms in the context of electrolyte solutions: (a) Ionic Strength ($I$), (b) Mean ionic activity ($a_{\pm}$), (c) Mean ionic activity coefficient ($\gamma_{\pm}$). How did the Debye-Hückel theory provide a theoretical justification for G.N. Lewis's empirical concept of ionic strength?
G.N. Lewis invented $I$ because he *observed* that it, not concentration, seemed to control activity coefficients. But he didn't know *why*.
The Debye-Hückel theory provided the "why."
The DH theory derivation showed that the key parameter controlling the ionic atmosphere was the screening parameter $\kappa$.
The derivation (Q2, Q5) showed that $\kappa^2$ was directly proportional to the sum $\sum_i n_i^0 z_i^2$:
$$ \kappa^2 = \left( \frac{e^2}{\epsilon k_B T} \right) \sum_i n_i^0 z_i^2 $$The theory also showed that the non-ideal energy ($\mu^{ex}$) and activity coefficient ($\ln \gamma_{\pm}$) were directly proportional to $\kappa$:
$$ \ln \gamma_{\pm} \propto \kappa $$Therefore, $\ln \gamma_{\pm} \propto \sqrt{\sum_i n_i^0 z_i^2}$.
Since $I = \frac{1}{2} \sum c_i z_i^2$ and $n_i^0 \propto c_i$, it means $I \propto \sum c_i z_i^2$.
By combining these, the DH theory proved that:
$$ \ln \gamma_{\pm} \propto \sqrt{I} $$In short, DH theory showed that the square root of the ionic strength was the fundamental parameter controlling activity coefficients because $I$ itself is the term that determines the "thickness" ($1/\kappa$) of the ionic atmosphere. Lewis's empirical $I$ was proven to be the correct physical quantity.
Question: Derive an expression for the radius of the ionic atmosphere, $r_D = 1/\kappa$. Calculate the value of $1/\kappa$ for a 0.01 M aqueous solution of NaCl and a 0.01 M aqueous solution of MgSO₄ at 298 K. (Given: $\epsilon_r$ for H₂O = 78.54, $\epsilon_0 = 8.854 \times 10^{-12} \text{ C}^2\text{N}^{-1}\text{m}^{-2}$, $k_B = 1.381 \times 10^{-23} \text{ J K}^{-1}$, $e = 1.602 \times 10^{-19} \text{ C}$, $N_A = 6.022 \times 10^{23} \text{ mol}^{-1}$).
From Question 2, the radius of the ionic atmosphere (Debye length) is $r_D = 1/\kappa$.
From Question 5, we found the relationship between $\kappa$ and ionic strength $I$:
$$ \kappa^2 = \frac{2000 e^2 N_A I}{\epsilon_0 \epsilon_r k_B T} $$Therefore, the expression for the Debye length is:
$$ r_D = \frac{1}{\kappa} = \sqrt{\frac{\epsilon_0 \epsilon_r k_B T}{2000 e^2 N_A I}} $$This equation has all terms in base SI units (m, J, K, C), *except* for $I$, which is in mol/L. The $2000$ factor (which is $2 \times 1000$) converts L to m³ and $\frac{1}{2} \sum$ to $\sum$.
Let's first calculate all the constant parts (at $T=298$ K):
Numerator = $\epsilon_0 \epsilon_r k_B T = (6.954 \times 10^{-10}) \times (4.116 \times 10^{-21}) = 2.862 \times 10^{-30}$
Denominator (without $I$) = $2000 e^2 N_A = 2000 \times (2.566 \times 10^{-38}) \times (6.022 \times 10^{23}) = 3.091 \times 10^{-11}$
So, $r_D = \sqrt{\frac{2.862 \times 10^{-30}}{3.091 \times 10^{-11} \times I}} = \sqrt{\frac{9.259 \times 10^{-20}}{I}} = \frac{3.043 \times 10^{-10}}{\sqrt{I}} \text{ (in meters)}$
Since $1 \text{ nm} = 10^{-9} \text{ m}$ and $1 \text{ Å} = 10^{-10} \text{ m}$, we can write:
$r_D \approx \frac{3.043 \text{ Å}}{\sqrt{I}}$ or $r_D \approx \frac{0.3043 \text{ nm}}{\sqrt{I}}$ (for water at 298 K, $I$ in mol/L)
Now we can solve the problems easily.
1. For 0.01 M NaCl:
2. For 0.01 M MgSO₄:
Tutor's Note: Look at the difference! Even though the concentration (0.01 M) is the same, the MgSO₄ solution has 4 times the ionic strength. This "pulls" the ionic cloud in, making its effective radius (the Debye length) half as large as in the NaCl solution.
Question: What is the "electrical potential" of the ionic atmosphere at the site of the central ion ($r=0$)? Derive the expression for this potential ($\psi_{cloud}$) and explain how it differs from the potential of the central ion itself. How is this "potential of the atmosphere" directly used to find the free energy of the ionic solution?
This is a very important concept. The total potential $\psi_j(r)$ at any point is the sum of the potential from the central ion itself, $\psi_{ion}$, and the potential from *everything else* (i.e., the atmosphere), $\psi_{cloud}$.
$$ \psi_j(r) = \psi_{ion}(r) + \psi_{cloud}(r) $$We derive this expression as we did in Question 4. We start with the DH solution for the *total* potential $\psi_j(r)$ and we expand it for small $r$ (i.e., at or near the central ion):
Total potential: $\psi_j(r) = \frac{z_j e}{4 \pi \epsilon r} e^{-\kappa r}$
Taylor expansion for $e^{-x} \approx 1 - x$ (for small $x = \kappa r$):
$$ \psi_j(r) \approx \frac{z_j e}{4 \pi \epsilon r} (1 - \kappa r) $$ $$ \psi_j(r) \approx \frac{z_j e}{4 \pi \epsilon r} - \frac{z_j e \kappa}{4 \pi \epsilon} $$Now we compare this to our conceptual equation:
$$ \psi_j(r) \approx \underbrace{\left( \frac{z_j e}{4 \pi \epsilon r} \right)}_{\psi_{ion}(r)} + \underbrace{\left( - \frac{z_j e \kappa}{4 \pi \epsilon} \right)}_{\psi_{cloud}} $$By inspection, we find the expression for the potential of the atmosphere at the central ion's location ($r=0$):
$$ \psi_{cloud} = - \frac{z_j e \kappa}{8 \pi \epsilon} $$Whoops, let's re-check that. $\frac{z_j e}{4 \pi \epsilon r}(\kappa r) = \frac{z_j e \kappa}{4 \pi \epsilon}$.
Correct: $$ \psi_{cloud} = - \frac{z_j e \kappa}{4 \pi \epsilon} $$
(The $8 \pi \epsilon$ term came in Q4 after integrating $W = \int \psi d\lambda$).
This is the potential at the central ion *due to its atmosphere*. It's finite, and it has the opposite sign of the central ion (as $z_j e \kappa / 4 \pi \epsilon$ is positive).
This $\psi_{cloud}$ is the key to finding the non-ideal contribution to the free energy.
The "free energy of the ionic solution" (specifically, the non-ideal *excess* free energy) is the excess chemical potential $\mu_j^{ex}$.
As shown in Question 4, this $\mu_j^{ex}$ is *defined* as the electrical work ($W_{el}$) of charging the ion from 0 to $z_j e$ against the potential of the cloud it creates.
$$ \mu_j^{ex} = G_{excess} = W_{el} = \int_{0}^{z_j e} \psi_{cloud}(\lambda) d\lambda $$Where $\psi_{cloud}(\lambda) = - \frac{\lambda \kappa}{4 \pi \epsilon}$ is the cloud's potential when the ion has a partial charge $\lambda$.
$$ \mu_j^{ex} = \int_{0}^{z_j e} \left( - \frac{\lambda \kappa}{4 \pi \epsilon} \right) d\lambda = - \frac{\kappa}{4 \pi \epsilon} \left[ \frac{\lambda^2}{2} \right]_0^{z_j e} $$ $$ \mu_j^{ex} = - \frac{(z_j e)^2 \kappa}{8 \pi \epsilon} $$Thus, the "potential of the atmosphere" $\psi_{cloud}$ is the integrand in the charging process that directly calculates the excess free energy ($\mu_j^{ex}$) of the ion, which in turn gives us the activity coefficient (since $\mu_j^{ex} = k_B T \ln \gamma_j$).
Question: Discuss the limitations of the Debye-Hückel model when applied to concentrated solutions. Explain the qualitative effects of (a) ion-solvent interactions (solvation) and (b) the breakdown of the "continuum" model for the solvent. How do empirical terms, like the one in the Brønsted-Guggenheim equation, attempt to correct for these deviations?
The DH model (even the Extended version) fails completely at high concentrations ($I > 0.1 - 0.5$ M) because its core assumptions are no longer just approximations, they are physically wrong.
The main limitations are:
(a) Ion-Solvent Interactions (Solvation):
(b) Breakdown of the "Continuum" (Dielectric Saturation):
At high concentrations, you have a complex battle: DH attraction (decreases $\gamma_{\pm}$), finite-size repulsion (increases $\gamma_{\pm}$), and solvation (increases $\gamma_{\pm}$). The solvation effect often "wins" at high $I$, causing the observed upturn.
Since the DH theory can't model these complex effects from first principles, we add empirical terms to "fix" the equation.
The Brønsted-Guggenheim equation is a famous example:
$$ \log_{10} \gamma_{\pm} = \underbrace{\frac{-A |z_+ z_-| \sqrt{I}}{1 + Ba\sqrt{I}}}_{\text{DH Extended Term}} + \underbrace{\beta I}_{\text{Empirical Term}} $$This equation simply says:
The parameter $\beta$ is an adjustable, empirical "fudge factor" that is specific to each salt. It has no deep theoretical meaning, but it is chosen to make the equation fit the experimental data at high concentrations. A positive $\beta$ is what allows this equation to correctly model the *upturn* in $\gamma_{\pm}$ when the $I$ term becomes large. More advanced models, like Pitzer equations, use even more complex empirical terms.
Time Remaining: 900s