ACD/pK_a DB

ACD/pK_a is very clear about which type of pK_a is being calculated. The pK_a of each particular ionization center in the drawn structure is called the microconstant (K_A, K_B, K_C, & K_D):

The pK_a values which are actually measured in aqueous solution, however, are called the macroscopic, or apparent, pK_a constants. The macroscopic constants are related to the above mentioned microconstants by the following equation:

We call such macroscopic constants the apparent exact pK_a values since they are obtained through the exact theoretical equations.

ACD/pK_a calculates macroconstants using the approximated theoretical equations which lead to the apparent approximate pK_a for very complex molecules where the calculation of exact pK_a is impossible.

It is very important to make sure that your pK_a software can handle calculations of both apparent and micro constants and clearly specifies which is which. For example, the molecule glycine, which is often expressed in chemical form as H₂N—CH₂—COOH in reality exists in solution at neutral pH in zwitterionic form, H₃N⁺--CH₂—COO^-. If you sketch in the neutral form, your pK_a software should be smart enough to handle this common problem of translating from the formula to the actual presence in solution.

In certain cases, however, you might want to look at a "what if?" case and thus you might want to calculate the pK_a for a "forced" equilibrium of the type:

For this case, it is helpful to have software which allows you to request the "microconstant" pK_a.

2. pK_a under what conditions?

Is this important?

Yes. The pK_a value depends critically on the experimental conditions under which the experiment was performed. ACD/pK_a depends on experimentally determined values in order to estimate their parameters. Therefore, for calculated pK_a values it is important to indicate the conditions (temperature, ionic strength, and especially solvent) presumed for the calculated values.

We have been careful to ensure that only data from the following experiments were used in creating our prediction algorithm:

• Completely aqueous solutions
• Experimental conditions of 25°C and zero ionic strength

If the second requirement could not be fulfilled, then we made an approximation of data to normal conditions (temperature = 25°C and ionic strength = 0.0) if and only if data at other conditions were available and gave a reasonable extrapolation.

3. Calculation Methods

Hammett and Taft Equations; Electronic Constants Used

In ACD/pK_a the inductive and resonance effects are considered separately. This is necessary because the ratio of inductive to resonance effect may vary substantially for the different equilibrium centers. The correlation equations for different types of atoms of the skeleton are calculated, and this is one of the contributions to the uncertainty which is listed with every calculation made by ACD/pK_a.

Hammett Equations

ACD/pK_a contains over 1,700 Hammett-type equations for about 1000 of the most popular ionizable functional groups. Each functional group has been characterized by several equations involving different types of substituent constants in order to achieve the most accurate calculation. All equations for a given functional group have been ranked according to their reliability (number of correlated structures, correlation coefficient and standard deviation) and reliability of available substituent constants. For example, the following ranking has been used for calculating pK_a values of para-substituted quinolines:

1. pK_a = 5.009 - 5.058*σ _I - 4.363*σ _R⁺, n = 10, r = 0.9989, sd = 0.13

2. pK_a = 4.874 - 4.561*σ _I - 5.63*σ _R, n = 10, r = 0.9878, sd = 0.46

3. pK_a = 5.179 - 5.318*σ _Para, n = 9, r = 0.9878, sd = 0.42

Electronic Constants

ACD/pK_a contains over 750 substituents with over 3,000 carefully derived experimental electronic constants. Included are the following constants:

σ _I - 592, σ * (Taft) - 265, σ _R - 453, σ _R^- - 157, σ _R⁺ - 143, σ _Para - 585, σ _Meta - 431, σ _Para^- - 142, σ _Para⁺ - 135, σ _Phosph (P-Acids) -68, σ _Ortho (Benzoic Acid) - 41, σ _Ortho (Phenol) - 37,

σ _Ortho (Aniline) - 30, σ _Ortho (Pyridine) - 48

When the required substituent constants are not available from the experimental data base they are calculated by special algorithms.

Transmission Effect

The calculation of the transmission effect is based on the following formula:

,

where all are substituent R electronic constants (inductive, resonance, etc.) and all are skeleton G transmission constants. The accuracy of the calculation is usually better than ± 0.05-0.1. Our algorithm contains 42 of the most frequently used skeletons G described by 126 such equations:

σ _I - 36, σ _R - 25, σ _R^- - 6, σ _R⁺ - 4, σ _Para - 24, σ _Meta - 24, σ _Phosph - 7

For example, the following constants are calculated for the carbamate functional group

σ_I = 0.45, σ_R = -0.34, σ_Meta = 0.32, σ_Para = 0.10, σ_R^- = -0.36, σ_R⁺ = -0.38, σ _Phosph = 0.0238.

The pK_a of 2-ammonio-4-thioxohexanedioate calculated by this method is 7.72 (experimental is 7.90):

Transmission Effect of Polyaromatics

The evaluation of the transmission effect of polyaromatics is based on the modified Dewar-Grisdale method. The original Dewar-Grisdale method (Dewar M.J.S., Grisdale P.J., (1962), J. Am. Chem. Soc., 84, 3539) can be used to calculate electronic transmission effects for only a very limited number of condensed polyaromatic systems. The improved ACD/pK_a method allows you to calculate these effects for virtually any polyaromatic system.

For example, the pK_a of 3-amino-5-hydroxy-2,7-naphthalenedisulfonate calculated by this method is 8.64 (exp. is 8.54):

Nonaromatic Ring Systems

The evaluation of the transmission effect of nonaromatic ring systems is based on the modified Exner-Fiedler method. The original Exner-Fiedler method can be used to calculate electronic transmission effects for only a very limited number of aliphatic cycles. The improved ACD/pK_a method allows calculation of these effects for any possible aliphatic (poly)cycles.

For example, the calculated transmission factor for

is 1.72 (exp. is 1.92).

Estimate of Error

In addition to calculating more accurate pK_a values, ACD/pK_a lets you know the type of error range there is for the values we estimate. Is the pK_a likely within ±0.01 pK_a units? Or within 0.5 units? It can make a big difference to how much you rely on the estimate, and as scientists we know this is an important aspect to any estimate. From what we have seen, pKalc simply does not provide this crucial information.