Polarity: The Invisible Sorting Mechanism

Why polarity governs selectivity

The rule chemists abbreviate as “like dissolves like” has a rigorous theoretical basis. A 2021 PNAS study derived it from first principles: the thermodynamic driving force for two substances to mix — or not — is determined by the similarity of their dielectric properties, which reflect the distribution of electrical charge in each molecule.¹ Polar substances, with asymmetric charge distribution, are stabilized by interaction with other polar molecules. Non-polar substances, with symmetric electron distribution, are stabilized by interaction with other non-polar molecules. The energy cost of forcing dissimilar molecules to associate drives separation.

In liquid-phase adsorption, this is the primary selectivity mechanism. A dissolved contaminant molecule will leave the fluid phase and bind to an adsorbent surface if that surface offers a more energetically favorable environment than staying in solution. For polar contaminants, polar surfaces are more favorable. For non-polar contaminants, non-polar surfaces are more favorable. Everything else in media selection — pore size, surface area, dose — operates within the constraints this polarity match sets.

What polar and non-polar mean at the surface level

Polar adsorbents carry surface functional groups with uneven electron distribution: silanol groups (Si-OH) on silica, hydroxyl groups on alumina (Al-OH), and mixed silicate/hydroxyl chemistry on magnesium silicate. These groups create local dipoles on the surface — partial negative charge at the oxygen, partial positive at the hydrogen. A polar contaminant molecule approaching this surface experiences electrostatic attraction. The interaction can involve hydrogen bonding, dipole–dipole forces, and ion–dipole interactions depending on the specific molecule and surface.

Non-polar adsorbents present a chemically inert, electron-rich surface. The graphitic surface of activated carbon is composed largely of aromatic carbon rings — extensively conjugated, with delocalized pi electrons but no permanent dipole. Moreno-Castilla (2004) describes the dominant mechanism for organic molecule adsorption on carbon as dispersive (van der Waals) and hydrophobic interactions, not electrostatic ones.² A non-polar molecule approaching a graphitic carbon surface experiences dispersive attraction without the energetic penalty of disrupting an ordered water structure. The molecule leaves the aqueous phase because water’s hydrogen-bonded network actively excludes it.

The Polanyi adsorption potential theory, applied to carbon systems by Toth and formalized in reviews of carbon nanomaterial adsorption, quantifies this driving force as a function of the molar volume and polarizability of the adsorbate — higher polarizability and lower aqueous solubility correspond to stronger adsorption on carbon surfaces.³

Media polarity in practice

Each major adsorbent class has a characteristic polarity that determines what it captures effectively:

Media	Surface chemistry	Effective for
Silica gel	Silanol groups (Si-OH) — high polar	Polar organics, pigments, waxes, water from non-polar solvents
Magnesium silicate	Mixed Mg-OH / Si-OH — moderate polar	Polar impurities from non-polar streams, soaps, oxidation products
Acid-activated montmorillonite (T-5®)	Modified aluminosilicate clay — high polar	Color bodies, chlorophyll, carotenoids, phospholipids in oil processing
Activated carbon	Graphitic aromatic carbon — non-polar / weak polar	Non-polar organics, PFAS (hydrophobic tail), volatile organics, taste/odor
Alumina	Al-OH — high polar, amphoteric	Acidic compounds, fluoride, arsenic in some configurations

Research on bleaching earth (acid-activated montmorillonite) confirms the selectivity pattern: polar pigment compounds — chlorophyll, carotenoids, oxidized lipids — are retained by the clay surface through dipole and ion-exchange interactions, while the non-polar oil phase passes through with minimal retention.⁴ This is the mechanism that makes bleaching earth effective for edible oil decolorization: the target impurities are polar, the product is non-polar, and the clay surface is selective exactly for that polarity class.

Activated carbon’s selectivity is the inverse. The hydrophobic graphitic surface captures non-polar organics and the non-polar (fluorinated alkyl) tails of PFAS molecules, while polar inorganic species — nitrate, most metal ions under typical conditions — pass through with low affinity.

The constraint that dose cannot overcome

Polarity mismatch is not a problem you solve with more media. This is the key point for practical system design.

Adding more activated carbon to a system trying to remove polar compounds does not improve removal of those compounds. The graphitic surface simply does not offer favorable energetics for polar molecule adsorption — adding more graphitic surface area provides more sites with the same unfavorable character. Similarly, bleaching earth dosed into a system targeting non-polar organic contaminants does nothing useful: the non-polar molecules have no thermodynamic reason to leave the oil phase and bind to a polar clay surface.

The correct response to inadequate removal in an adsorption system is to diagnose the cause before adjusting dose. If polarity is mismatched, the solution is changing the adsorbent. If polarity is matched but removal is insufficient, the problem is capacity (surface area, dose, or contact time) or pore geometry (target molecules can’t access available surface). These require different fixes.

Multi-stage systems: polarity gradients, not strength gradients

Process streams with multiple target contaminants — a color remediation application removing both polar pigments and non-polar oxidation byproducts, for example — cannot be addressed with a single polarity class of adsorbent. This is why multi-stage media systems exist.

The stages in a well-designed separation stack are arranged by polarity class: each layer is selected for its affinity to a specific class of target compound. The sequence matters. Placing the highest-loading adsorbent stage first protects downstream media from premature saturation with compounds that could have been captured earlier and more cheaply.

This is a polarity gradient, not a strength gradient. The purpose is not to subject the process stream to progressively more aggressive treatment. It’s to present each contaminant class with the surface chemistry it will preferentially bind to, in the order that optimizes media economics and prevents fouling. Understanding that each stage addresses a specific polarity class — rather than contributing a fraction of some aggregate “adsorptive power” — is what distinguishes effective multi-stage system design from trial-and-error media stacking.

The solvent matrix is part of the system

Polarity matching operates in the context of the surrounding fluid. A polar contaminant dissolved in a non-polar solvent behaves differently than the same contaminant in water. In a non-polar solvent, the polar contaminant is already energetically unfavorable in the fluid phase — it wants to leave. The driving force onto a polar adsorbent surface is strong. In water, the same contaminant is stabilized by hydrogen bonding with water molecules — the driving force onto a polar surface is weaker, because the alternative (staying in water) is not as unfavorable.

This is why media that perform reliably in oil-phase applications may require reconsideration in aqueous systems, and vice versa. The adsorbent’s polarity is fixed; the thermodynamic driving force for adsorption depends on the polarity contrast between the adsorbent surface and the fluid phase the contaminant is leaving.

References

Footnotes

1. Alavi, S.H., & Rizvi, S.S.H. (2021). Like dissolves like: A first-principles theory for predicting liquid miscibility and mixture dielectric constant. Proceedings of the National Academy of Sciences, 118(3), e2020389118. https://pmc.ncbi.nlm.nih.gov/articles/PMC7880597/ ↩

2. Moreno-Castilla, C. (2004). Adsorption of organic molecules from aqueous solutions on carbon materials. Carbon, 42(1), 83–94. https://doi.org/10.1016/j.carbon.2003.09.022 ↩

3. Chemical Reviews. (2010). Adsorption of organic compounds by carbon nanomaterials in aqueous phase: Polanyi theory and its application. Chemical Reviews, 110(10), 5989–6008. https://doi.org/10.1021/cr100059s ↩

4. Hassanien, R., & Abdel-Hameed, S.A.M. (2022). Montmorillonite for adsorption and catalytic elimination of pollutants from wastewater: A state-of-the-arts review. Sustainability, 14(24), 16441. https://doi.org/10.3390/su142416441 ↩