3.4.1. Relevant chemical properties

Authors: Joop Hermens, Kees van Gestel

Reviewer: Steven Droge, Monika Nendza

 

Learning Objectives

You should be able to:

 

Keywords: Hydrophobicity, octanol-water partition coefficients, volatility, Henry’s Law constant, ionized chemicals

 

Introduction

Different processes affect the fate of a chemical in the environment. In addition to the transfer and exchange between compartments (air-water-sediment/soil-biota), also degradation determines the concentration in each of these compartments (Figure 1).

 

Figure 1. Environmental fate: exchange between compartments and degradation affect the concentration in each compartment.

 

Some of these processes are discussed in other sections (see sections on Sorption and Environmental degradation of chemicals). Some chemicals will easily evaporate from water to air, while others remain mainly in the aqueous phase or sorb to sediment and accumulate into biota.

 

These differences are related to only a few basic properties:

 

Hydrophobicity

Hydrophobicity means fear (phobic) of water (hydro). A hydrophobic chemical prefers to “escape from the aqueous phase” or in other words “it does not like to dissolve in water”. Water molecules are tightly bound to each other via hydrogen bonds. For a chemical to dissolve in water, a cavity should be formed in the aqueous phase (Figure 2) and this will cost energy.

 

Figure 2. The formation of a cavity in water for chemical X.

 

Hydrophobicity mainly depends on two molecular properties:

It will take more energy for a chemical with a larger size to create the cavity making the chemical more hydrophobic, while interactions of the chemical with water will favour its dissolution making it less hydrophobic. Figure 3 shows chemicals with increasing hydrophobicity with increasing size and a decreasing hydrophobicity by the presence of polar groups (amino or hydroxy).

 

Figure 3. The effect of size and presence of polar groups on the hydrophobicity of chemicals. Increasing molecular size increases hydrophobicity; the introduction of polar groups leads to a decrease in hydrophobicity.

 

Most hydrophobic chemicals are non-polar organic micro pollutants. Well-known examples are the chlorinated hydrocarbons, such as polychlorinated biphenyls (PCBs) and polycyclic aromatic hydrocarbons (PAHs). Water solubility of these chemicals in general is rather low (in the order of a few ng/L up to a few mg/L).

The hydrophobic nature mainly determines the distribution of these chemicals in water and sediment or soil and their uptake across cell membranes. Additional Cl- or Br-atoms in a chemical, as well as additional (CH)x units, increase the molecular size, and thus a chemical’s hydrophobicity. The increased molecular volume requires a larger cavity to dissolve the chemical in water, while they only interact with water molecules via VanderWaals interactions.

Polar groups, such as the -OH and -NH units on the aromatic chemicals in Figure 3, can form hydrogen-bonds with water, and therefore substantially reduce the hydrophobicity of organic chemicals. The hydrogen bonding of hydroxy-substituents works in two ways: The oxygen of -OH bridges to the H-atoms of water molecules, while the hydrogen of –OH can form bridges to the O-atoms of water molecules. Nearly all molecular units consisting of some kind of (carbon-oxygen) combination reduce the hydrophobicity of organic contaminants, because even though they increase the molecular volume they interact via hydrogen bonds (H-bonds) with surrounding water molecules. Additional polar groups in a chemical typically decrease a chemical’s hydrophobicity.

 

Octanol-water partition coefficient:

A simple measure of the hydrophobicity of chemicals, originating from pharmacology, is the octanol-water partition coefficient, abbreviated as Kow (and sometimes also called Pow or Poct): this is the ratio of concentrations of a chemical in n-octanol and in water, after establishment of an equilibrium between the two phases (Figure 4). The -OH group in n-octanol does allow for some hydrogen bonding between octanol-molecules in solution, and between octanol and dissolved molecules. However, the relatively long alkyl chain only interacts through VanderWaals interactions, and therefore the interaction strength between octanol-molecules is much smaller than that between water-molecules, and it is energetically much less costly to create a cavity to dissolve any molecule.

 

Figure 4. Distribution of chemical X between octanol and water and an example of a chemical with log Kow of 5.0.

 

Experimentally determined Kow values were used in pharmacological research to predict the uptake and biological activity of pharmaceuticals. Octanol was selected because it appears to closely mimic the nonionic molecular properties of most tissue components, particularly phospholipids in membranes. Since the beginning of the 1970s, Kow values have also been used in environmental toxicology to predict the hazard and environmental fate of organic micro pollutants. Octanol may partially also mimic the nonionic molecular properties of most organic matter phases that sorb neutral organic chemicals in the biotic and abiotic environment.

Not unexpectedly, water solubility is negatively correlated with octanol-water partition coefficients.

In practice, three methods can be used to determine or estimate the Kow:

 

Equilibration methods

In the shake-flask method (Leo et al., 1971) and the 'slow-stirring' method (de Bruijn et al., 1989), the distribution of a chemical between octanol and water is determined experimentally. For highly lipophilic chemicals (log Kow > 5-6), the extremely low water solubility, however, hampers a reliable analytical determination of concentrations in the water phase. For such chemicals, these experimental methods are not suitable. During the last two decades, the use of generator columns has allowed for quantification of higher Kow values. Generator columns are columns packed with a sorbing material (e.g. Chromosorb®) onto which an appropriate hydrophobic solvent (e.g. octanol) is coated that contains the compound of interest. In this way, a large interface surface area between the lipophilic and water phases is created, which allows for a rapid establishment of equilibrium. When a large volume of (octanol-saturated) water (typically up to 10 litres) is passed slowly through the column, an equilibrium distribution of the compound is established between the octanol and the water. The water leaving the column is passed over a solid sorbent cartridge to concentrate the compound and allow for a quantification of the aqueous concentration. In this way, it is possible to more reliably determine log Kow values up to 6-7.

 

Chromatography

Kow values may also be derived from the retention time in a chromatographic system (Eadsforth, 1986). The use of reversed-phase High Performance Liquid Chromatography (HPLC), thin-layer chromatography or gas chromatography results in a capacity factor (relative retention time; retention of the compound relative to a non-retained chemical species), which may be used to predict the chemical distribution over octanol and water. HPLC systems have shown most successful, because they consist of stationary and mobile phases that are liquid. As a consequence, the nature of the phases can be most closely arranged to resemble the octanol-water system. Of course, this requires calibration of the capacity factors by applying the chromatographic method to a number of chemicals with well-known Kow values. Chromatographic methods may reliably be applied for estimations of log Kow values in the range of 2-8. For more lipophilic chemicals, also these methods will fail to reliably predict Kow values (Schwarzenbach et al., 2003).

 

Calculation

Kow values may also be calculated or predicted from parameters describing the chemical structure of a chemical. Several software programs are commercially available for this purpose, such as KOWWIN program of the US-EPA. These programs make use of the so-called fragment method (Leo, 1993; Rekker and Kort, 1979). This method takes into account the contribution to Kow of different chemical groups or atoms in a molecule, and in addition corrects for special features such as steric hindrance or other intramolecular interactions (equation 1):

log Kow = Ʃ fn + Ʃ Fp                                                                          (eq.1)

in which fn quantifies the contributions of each fragment n in a particular chemical (see e.g. Table 1) and Fp accounts for any special intramolecular interaction p between the fragments.

 

This fragment approach has been improved during the last decades and is available in the EPISUITE program from the US Environmental Protection Agency. Other programs for the calculation of Kow values are: ChemProp, and ChemAxon from Chemspider.

 

Table 1. Fragment constants (Kow) for a few fragments. (from the EPISUITE program)

Fragment

Fragment constant (f)a

-CH3 aliphatic carbon 

0.5473

Aromatic Carbon   

0.2940

-OH hydroxy, aromatic attach

-0.4802

-N aliphatic N, one aromatic attach

-0.9170

 

Note: the above calculations are given for non-ionized chemicals. The hydrophobicity of ionic chemicals is also highly affected by the degree of ionization (see below).

 

Kow values can also be retrieved from databases like echemportal or ECHA and others.

 

Volatility

Volatility of a chemical from the aqueous phase to air (see Figure 5) is expressed via the Henry’s law constant (KH).

 

Figure 5. Evaporation of chemical X from water to air.

 

Henry’s law constant (KH, in Pa⋅m3/mol) is the chemical distribution between the gas phase and water, as

                                                                         (eq.2)

where in an equilibrated water-gas system:

Caq is the aqueous concentration of the chemical (units in mol/m3), and Pi is the partial pressure of the chemical in air (units in Pascal, Pa), which is the pressure exerted by the chemical in the total gas phase volume (occupied by the mixture of gases the gas-phase above the water solution of the chemical). Note that Pi is a measure of the concentration in the gas phase, but not yet in the same units as the dissolved concentration (discussed below)!

 

For compounds that are slightly soluble in water, KH can be estimated from:

                                                                                      (eq.3)

 

 

where:

KH: Henry’s law constant (Pa⋅m3/mol), Vp is the (saturated) vapor pressure (Pa), which is the pressure of the chemical above the pure condensed (liquid) form of the chemical, and Sw is the maximum solubility in water (mol/m3).

 

The advantage of equation 3 is that both Vp and Sw can be experimentally derived or estimated. The rationale behind equation 3 is that two opposite forces will affect the evaporation of a chemical from water to air:

(i) the vapor pressure (Vp) of the pure chemical - high vapor pressure means more volatile, and

(ii) solubility in water (Sw) - high solubility means less volatile.

Benzene and ethanol (see Table 2) are good illustrations. Both chemicals have similar vapor pressure, but the Henry’s law constant for benzene is much higher because of its much lower solubility in water compared to ethanol; benzene is much more volatile from an aqueous phase.

 

Table 2. Air-water partition coefficients (Kair-water) calculated for five chemicals (ranked by aqueous solubility) by equation 3.

Chemical

Vapor pressure

(Pa)

Solubility (mol/m3)

KH

(Pa.m3/mol)

Kair-water

(L/L, or m3/m3)

Ethanol

7.50⋅103

1.20⋅104

6.25⋅10-1

2.53⋅10-4

Phenol

5.50⋅101

8.83⋅102

6.23⋅10-2

2.52⋅10-5

Benzene

1.27⋅104

2.28⋅101

5.57⋅102

2.25⋅10-1

Pyrene

6.00⋅10-4

6.53⋅10-4

9.18⋅101

3.71⋅10-4

DDT

2.00⋅10-5

2.82⋅10-6

7.08

2.86⋅10-3

 

Note: all chemicals at equilibrium have a higher concentration (in e.g. mol/L) in the aqueous phase than in the gas phase. Of these five, benzene is the chemical most prone to leave water, with an equilibrated air concentration about 4 times lower (22.5%) than the dissolved concentration.

 

Equations 2 and 3 are based on the pressure in the gas phase. Environmental fate is often based on partition coefficients, in this case the air-water partition coefficient (Kair-water). These partition coefficients are more or less ‘dimensionless’, because the concentrations are based on equal volumes (such as L/L), while KH has the unit Pa⋅m3/mol or something equivalent to the applied units (equation 4).

                                                                         (eq.4)

where:

Cair is the concentration in air (in e.g. mol/m3) and Caq is the aqueous concentration (in e.g. mol/m3).

Kair-water can be calculated from KH according to equation 5:

                                                                           (eq.5)

where R is the gas constant (8.314 m3⋅Pa⋅K−1⋅mol−1), and T is the temperature in Kelvin (Kelvin = oCelsius + 273).

This use of “RT” converts this gas phase concentration to a volume based metric, and applies the ideal gas law which relates pressure (P, in Pa) to temperature (T, in K), volume (V, in m3), and amount of gas molecules (n, in mol), according to the gas constant (R: 8.314 m3⋅Pa⋅K-1⋅mol-1):

P⋅V = n⋅R⋅T (note that the units of both terms will cancel out)                (eq.6)

At 25 oCelcius (298 K), the product RT equals 2477 m3⋅Pa⋅K-1⋅mol-1.

 

Examples of calculated values for Kair-water are presented in Table 2.

 

The influence of the chemical structure on volatility of a chemical from a solvent fully depends on the cost of creating a cavity in the solvent (interactions between solvent molecules) and the interactions between the chemical and the solvent molecules. For partitioning processes, the gas phase is mostly regarded as an inert compartment without chemical interactions (i.e. gas phase molecules hardly ever touch each other).

 

The molecules of a strongly dipolar solvent such as water that contain atoms that can interact as hydrogen acceptor (the O in an OH group) and hydrogen donor (the H in an OH group) strongly interact with each other, and it costs much energy to create a cavity. This cost increases strongly with molecular size, for nearly all molecules more than the energy regained by interactions with the surrounding solvent molecules. As a result, for most classes of organic chemicals, affinity with water decreases and volatility out of water into air slightly increases with molecular volume. For chemicals that are not able to re-interact via hydrogen bonding, e.g. alkanes, the overall volatility is much higher than for chemicals that do have specific interactions with water molecules besides van der Waals.

 

Degree of ionization

Acids and bases can be present in the neutral (HA and B) or ionized form (A- and BH+, respectively). For acids, the neutral form (HA) is in equilibrium with the anionic form (A-) and for bases the neutral form (B) is in equilibrium with the cationic form (BH+). The degree of ionization depends on the pH and the acid dissociation constant (pKa). Table 3 shows the equations to calculate the fraction ionized for acids and bases and examples of two acids (phenols) are presented in Table 4.

 

Table 3. Calculation of the fraction ionized for acids and bases.

Acids

Bases

 

 

 

 

  pKa =  - log Ka, where Ka is dissociation constant of the acidic form (HA or BH+).

 

The degree of ionization is thus determined by the pH and the pKa value and more examples for several organic chemicals are presented elsewhere (see Chapter Ionogenic organic compounds).

 

Table 4. The degree of ionization of two phenolic structures (acids).

Pentachlorophenol

Phenol

pKa = 4.60

pKa = 9.98

% ionized versus pH

% ionized versus pH

at pH of 7.0: 99.6 % ionized

at pH of 7.0: 0.1 % ionized

 

Examples for several organic chemicals are presented elsewhere (see section on Ionogenic organic compounds).

 

The fate of ionic chemicals is very different from that of non-ionic chemicals. The sediment-water sorption coefficient of the anionic species is substantially (>100x) lower than that of the neutral species. If the percentage of ionization is less than ~99 % (at a pH 2 units above the pKa), the sorption of the anion may be neglected (Kd is still dominated by the >1% neutral species) (Schwarzenbach et al., 2003). The reason for the low sorption affinity of the anionic acid form is twofold: anions are much better water soluble, but also most sediment particles (clay, organic matter, silicates) are negatively charged, and electrostatically repulse the similarly charged chemical. In that case the sorption coefficient Kd can be calculated from the sorption coefficient of the non-ionic form and the fraction of the non-ionized form (α):

                                                       (eq. 7)

In environments where the pH is such that the neutral acid fraction <1% (when pH >2 units above the pKa), the sorption of the anionic species to soil/sediment may significantly contribute to the overall “distribution coefficient” of both acid species.

For basic environmental chemicals of concern, among which many illicit drugs (e.g. amphetamine, cocaine) and non-illicit drugs (e.g. most anti-depressants, beta-blockers), the protonated forms are positively charged. These organic cations are also much more soluble in water than the neutral form, but at the same time they are electrostatically attracted to the negatively charged sediment surfaces. As a result, the sorption affinity of organic cations to sediment should not be considered negligible relative to the neutral species. The sorption processes, however, may strongly differ for the neutral base species and the cationic base species. Several studies have shown that the sorption affinity of cationic base species to DOM or sediment is even stronger than that of the neutral species.


References

De Bruijn, J., Busser, F., Seinen, W., Hermens, J. (1989). Determination of octanol/water partition coefficients for hydrophobic organic chemicals with the "slow-stirring" method. Environmental Toxicology and Chemistry 8, 499-512.

Eadsforth, C.V. (1986). Application of reverse-phase HPLC for the determination of partition coefficients. Pesticide Science 17, 311-325.

Leo, A., Hansch, C., Elkins, D. (1971). Partition coefficients and their uses. Chemical Reviews 71, 525-616.

Leo, A.J. (1993). Calculating log P(oct) from structures. Chemical Reviews 93, 1281-1306.

Rekker, R.F., de Kort H.M. (1979). The hydrophobic fragmental constant; an extension to a 1000 data point set. Eur.J. Med. Chem. - Chim. Ther. 14:479-488.

Schwarzenbach RP, Gschwend PM, Imboden DM (Eds.) 2003. Environmental organic chemistry. Wiley, New York, NY, USA.

 

Further reading:

Mackay, D., Boethling, R.S. (Eds.) 2000. Handbook of property estimation methods for chemicals: environmental health and sciences. CRC Press.

van Leeuwen, C.J., Vermeire, T.G. (Eds.) 2007. Risk assessment of chemicals: An introduction. Springer, Dordrecht, The Netherlands