pKa of amino acids: the full table, titration curves and the Henderson-Hasselbalch equation

A pKa is a pH address

A pKa is a pH address. Below that address a group keeps its proton. Above it the group has let go. Stand exactly on the address and half the molecules hold the proton while half have released it. Every amino acid has at least two of these addresses, one for the carboxyl group and one for the amino group, and seven of the twenty have a third on the side chain.

This guide lists all of them in one table, shows how to read the titration curve those numbers draw, works through the Henderson-Hasselbalch equation with real values, and explains why the table in one textbook never quite matches the table in the next. The charged five residues get their own treatment in the acidic and basic amino acids guide; this page is the full set.

What is the pKa of an amino acid?

The pKa of an amino acid is the pH at which one of its ionizable groups is exactly half protonated. A free amino acid in water does not sit as a neutral molecule with a loose carboxylic acid and a loose amine. The carboxyl gives its proton to the amine inside the same molecule, which leaves a negative carboxylate and a positive ammonium on one backbone. OpenStax describes the result as a dipolar ion, or zwitterion, and notes that these zwitterions are internal salts with many of the physical properties of salts.

So the two backbone pKa values describe two separate events. The first, written pKa1, is the carboxyl group losing its proton as the pH climbs out of strong acid. The second, pKa2, is the ammonium group losing its proton as the pH climbs toward base. Side chains that can ionize add a third event, usually written pKa3 or pKaR, and its position depends entirely on the chemistry of that side chain: a second carboxyl on aspartate sits low, a guanidino group on arginine sits very high.

Each pKa is a property of one group, measured in one set of conditions. That last clause is the reason the numbers wander between sources, and it gets its own section below.

pKa table for all 20 amino acids

The table below carries the values printed in OpenStax Organic Chemistry, Table 26.1, which is a common classroom reference. The side-chain column is blank for the thirteen residues whose side chain does not ionize. The pI column is the isoelectric point, the pH at which the free amino acid carries no net charge, and it is worked out from the pKa values in the way the later section explains.

A few patterns are worth reading off the table. The alpha-carboxyl values all sit in a narrow band from 1.82 (histidine) to 2.83 (tryptophan). The alpha-amino values spread wider, from 8.80 (asparagine) to 10.60 (proline), whose secondary amine is part of a ring. The seven side-chain values span the whole scale: aspartate at 3.65, glutamate at 4.25, histidine at 6.00, cysteine at 8.18, tyrosine at 10.07, lysine at 10.53 and arginine at 12.48. Those seven are the ones that decide a peptide's charge, and the polar amino acids guide sorts them alongside the uncharged polar residues.

Why do pKa values differ between textbooks?

Because each number was measured by somebody, somewhere, at some temperature, and different books copy from different measurements. PubChem makes this visible by listing every source it has for each amino acid side by side. For tyrosine it carries 2.20, 9.11 and 10.07 from Lehninger's 1975 textbook, and separately 2.06 and 9.145 from another source. For threonine it lists 2.63 and 10.43 from the Merck Index against the 2.09 and 9.10 printed in OpenStax. Arginine's guanidino group is 12.48 in OpenStax and 13.2 in the Merck Index entry PubChem carries. None of these is a typo. They are different experiments.

Temperature is the biggest single reason. PubChem's lysine record gives pK1 2.18, pK2 8.95 and pK3 10.53 and notes the set was measured at 38 °C. Its glycine record carries a carboxyl pKa of 2.37 measured at 20 °C beside a Merck Index pair of 2.34 and 9.60 with no temperature stated. For valine, leucine and isoleucine the Kortum 1961 compilation reports carboxyl values measured at 13 °C or 0 °C. A pKa moves with temperature, so a table assembled from mixed sources is a table of mixed conditions.

Assignment is the second reason, and cysteine is the clearest case. OpenStax lists its amino group at 10.28 and its thiol at 8.18. The Merck Index entry in PubChem lists the amine at 8.33 and the sulfide at 10.78. The two groups ionize so close together that the measured constants are hard to pin to one group or the other, and different authors have pinned them differently. The same tangle shows up in the aspartate side chain, 3.65 in OpenStax against 3.87 in the IUPAC compilation PubChem cites.

When we compare two published tables in-house, the first thing we look for is the temperature line, because that one detail accounts for more of the gap than anything else. The practical rule is to take all values from one table and say which table it was.

How do you read a titration curve?

A titration curve plots pH against the amount of base added, and for an amino acid it has one flat step per ionizable group. OpenStax walks through alanine. Below pH 1 the molecule is fully protonated and carries a positive charge. At pH 2.34, the alpha-carboxyl pKa, the solution is a 50:50 mix of the protonated form and the zwitterion. At pH 6.01 the amino acid is entirely in the neutral zwitterion form; that is the isoelectric point. At pH 9.69, the alpha-amino pKa, the mix is 50:50 zwitterion and the deprotonated, negatively charged form. Above pH 11.5 it is entirely deprotonated.

In base-equivalent terms, 0.5 equivalent of sodium hydroxide strips the proton from half the carboxyl groups and 1.0 equivalent finishes that step, leaving the zwitterion dominant. Another 0.5 equivalent takes half the ammonium groups and 2.0 equivalents complete the titration. So the first leg of the curve, roughly pH 1 to 6, is the carboxyl dissociating, and the second leg, roughly pH 6 to 11, is the ammonium dissociating.

Each flat region is a buffer zone, centered on a pKa, where added base changes the pH slowly. Each steep region is where one species has been used up and the next has not yet started to react. An amino acid with a charged side chain adds a third flat step, and the position of that step on the pH axis is the side-chain pKa read directly off the plot.

How does the Henderson-Hasselbalch equation use pKa?

The Henderson-Hasselbalch equation turns a pKa into a ratio: pH = pKa + log([A-]/[HA]), where HA is the protonated form of a group and A- is the form that has lost its proton. When pH equals pKa the log term is zero and the two forms are present in equal amounts. One pH unit above the pKa the ratio is ten to one in favor of the deprotonated form. One unit below, it is ten to one the other way.

OpenStax works the example of 1.00 M alanine at pH 9.00. Because 9.00 is far closer to pKa2 (9.69) than to pKa1 (2.34), the calculation uses pKa2. The log of the ratio is 9.00 minus 9.69, which is minus 0.69, so the ratio of deprotonated to zwitterion is 0.20. With the two forms summing to 1.00 M, that gives 0.83 M zwitterion and 0.17 M deprotonated alanine: 83 percent neutral, 17 percent negatively charged.

The equation is an approximation and it has edges. A Chemistry LibreTexts treatment of the Henderson-Hasselbalch approximation gives the working conditions: the log term should stay between minus 1 and plus 1, the buffer concentration should be large compared with the acid constant, and the approximation is at its best for pKa values between about 5 and 9. Near the very start or end of a titration, where one form is almost gone, the simple ratio breaks down. For the backbone groups of an amino acid, which sit near 2 and near 9.5, that caveat matters most at the carboxyl end.

How do pKa values give the isoelectric point?

The isoelectric point is the average of the two pKa values that flank the neutral zwitterion. Which two depends on the side chain.

1. For a residue with a neutral side chain, average the alpha-carboxyl and alpha-amino values. Alanine: (2.34 + 9.69) / 2 = 6.01, which is the value in the table.
2. For an acidic side chain, average the two lowest values, because the zwitterion sits between the two carboxyl dissociations. Aspartic acid: (1.88 + 3.65) / 2 = 2.77.
3. For a basic side chain, average the two highest values, because the zwitterion sits between the two amine dissociations. Lysine: (8.95 + 10.53) / 2 = 9.74.

Run the same arithmetic on the rest of the table and every pI column entry falls out, including histidine at (6.00 + 9.17) / 2 = 7.59 and arginine at (9.04 + 12.48) / 2 = 10.76. OpenStax notes that neutral amino acids land near pI 5.0 to 6.5, acidic ones lower and basic ones higher. Solubility in water is usually at its lowest at the pI, where the molecule carries no net charge, which is why pI is the number a lab checks before choosing a buffer.

For a whole peptide the count is longer, because the chain has one N-terminal amine, one C-terminal carboxyl and whatever ionizable side chains the sequence contains. The peptide isoelectric point guide walks that calculation step by step with a worked example.

Do the values change inside a peptide or protein?

Yes, and sometimes by a lot. The free amino acid values above describe a molecule floating alone in water with a free carboxyl and a free amine on the same carbon. Once a residue sits inside a chain, those two backbone groups have become amide bonds and no longer ionize at all. Only the side chains and the two chain ends keep a pKa.

Grimsley, Scholtz and Pace compiled 541 measured pK values from 78 folded proteins and published the summary in Protein Science in 2009. As a baseline they used values from alanine pentapeptides, where the group of interest is free of charge interactions and burial: Asp 3.9, Glu 4.3, His 6.5, Cys 8.6, Tyr 9.8, Lys 10.4, C-terminus 3.7 and N-terminus 8.0. Note that the N-terminal amine of a peptide sits near 8.0 rather than the 9 to 10 of a free amino acid.

Inside folded proteins the averages were close to those baselines, Asp 3.5, Glu 4.2, His 6.6, Lys 10.5, Tyr 10.3, but the spread was wide. Aspartate ranged from 0.5 to 9.2, cysteine from 2.5 to 11.1, lysine from 5.7 to 12.1. The authors tie the shifts to three things: burial in a low-dielectric interior raises acid pK values and lowers amine pK values, a positively charged neighborhood lowers pK values while a negative one raises them, and a hydrogen bond pushes the value up or down depending on which form it stabilizes. Lysine stays near its baseline because its amine is usually exposed to solvent.

For a short research peptide with no stable fold, the pentapeptide baselines are the better set to use. For a folded protein, the table on this page is a starting point and nothing more.

Where pKa shows up on a certificate

A certificate of analysis never prints a pKa, yet the charge arithmetic on this page sits behind several lines of it. The purity method on our certificates is reversed-phase HPLC on a C18 column, which separates by hydrophobicity, and the HPLC purity versus identity guide explains what that number can and cannot prove. The identity line comes from LC-MS, where a peptide picks up protons during electrospray and the basic residues in the table above are where many of those protons land. The net content line is reported separately, because a basic peptide carries counterions paired to its positive charges and those add mass to the powder.

Each certificate we publish reports those three lines by lot number, and the certificate index lists them. If you want to estimate a peptide's charge at the pH of your own buffer, pull the side-chain values from one table, add the two chain ends, and run the Henderson-Hasselbalch arithmetic group by group.

FOR LABORATORY AND IN-VITRO RESEARCH USE ONLY. NOT FOR HUMAN OR ANIMAL CONSUMPTION.

Bottom line

Every amino acid has an alpha-carboxyl pKa near 2 and an alpha-amino pKa near 9 to 10, and seven have a side-chain pKa between 3.65 and 12.48. The isoelectric point is the average of the two values that flank the zwitterion. Textbooks disagree by a few tenths because the measurements come from different temperatures and different assignments, so take every value from one table and name it. Inside a chain, only the side chains and the two ends still ionize, and a fold can move a side-chain value by several pH units.

Common questions

More science guides

• What Is an Amino Acid Residue in a Peptide Chain?
• Peptide Antigenicity in the Lab: Epitopes, Haptens and Carrier Proteins
• Certified peptides: what the word can honestly mean on a research label
• Is compounded semaglutide FDA approved? No, and here is what that means under 503A and 503B
• Compounded Tirzepatide Explained: 503A, 503B, the Shortage and Where FDA Stands Now

Sources

• OpenStax, Organic Chemistry, 26.1 Structures of Amino Acids (Table 26.1). The full table of alpha-carboxyl, alpha-amino and side-chain pKa values and pI for the 20 common amino acids, and the description of the zwitterion as an internal salt.
• OpenStax, Organic Chemistry, 26.2 Amino Acids and the Henderson-Hasselbalch Equation: Isoelectric Points. The alanine titration curve, the worked example of 1.00 M alanine at pH 9.00, and the rules for averaging pKa values to get pI.
• Chemistry LibreTexts, Henderson-Hasselbalch Approximation. Derivation from Ka and the working limits of the approximation.
• Grimsley GR, Scholtz JM, Pace CN. A summary of the measured pK values of the ionizable groups in folded proteins. Protein Science 2009;18(1):247-251 2009 doi:10.1002/pro.19. Alanine pentapeptide baseline pK values, averages and ranges across 541 measurements in 78 folded proteins, and the causes of the shifts.
• PubChem, L-lysine (CID 5962), Dissociation Constants. pK1 2.18, pK2 8.95, pK3 10.53 at 38 degrees Celsius from the Merck Index.
• PubChem, L-tyrosine (CID 6057), Dissociation Constants. Lehninger values 2.20, 9.11 and 10.07 beside the 2014 Pharmaceutical Research values 2.06 and 9.145.
• PubChem, L-cysteine (CID 5862), Dissociation Constants. Merck Index assignment of pK2 8.33 to the amine and pK3 10.78 to the sulfide.
• PubChem, L-threonine (CID 6288), Dissociation Constants. Merck Index values pKa1 2.63 and pKa2 10.43.
• PubChem, L-arginine (CID 6322), Dissociation Constants. Merck Index values pK1 2.18, pK2 9.09, pK3 13.2.
• PubChem, L-aspartic acid (CID 5960), Dissociation Constants. IUPAC compilation values pK1 1.92, pK2 3.87, pK3 9.87.
• PubChem, Glycine (CID 750), Dissociation Constants. Carboxyl pKa 2.37 at 20 degrees Celsius beside the Merck Index pair 2.34 and 9.60.
• PubChem, L-valine (CID 6287), L-leucine (CID 6106) and L-isoleucine (CID 6306), Dissociation Constants. Kortum 1961 carboxyl values measured at 13 and 0 degrees Celsius.