If Tina Turner was an inexperienced chromatographer, perhaps she would sing “What’s salt got to do with it?”

One of the important aspects of forensic testing is the validity of the result.  The validity of the result is comprised of assuring that a test result is:

• Precise-a measure of how closely the results can be to one another.  It is characterized by a low Standard Deviation, but may or may not have a high average deviation from the true (actual) value; Accurate-a measure of how closely the results are to the true value.  It is characterized by a high Standard Deviation, but may or may not have a low average deviation from the true (actual) value

• Specific-a test that reacts only with the analyte of interest and nothing else so that when there is a measure it is unique to that which is examined.  It is not simply selective, which is an exhibition of a degree of preference for the substance tested for, but not necessarily unique

• Sensitive-the quantitative measure is within the demonstrated linear dynamic range in that the unknown amount exceeds both the Limit of Detection and the Limit of Quantification but does not exceed the Limit of Linearity

• Repeatable-day-in and day-out the test using the same method on the same instrumentation on the same unknown arrives at the same result

• Reproducible-refers to the ability of a test or experiment to be accurately reproduced, or replicated, by someone else working independently

• Reliable-inversely related to random error, it is the consistency of a set of measurements or of a measuring instrument; Traceable-the completeness of the information about every step in a process chain

• Verifiable-the ability to take all information provided and arrive at the same conclusions as the original analyst without the need to independently test

• Valid-documented proof that the process undertaken is suitable for its intended use and achieves the intended reported result correctly and uniquely

• Robust-the method and instrument combine to have a low percent false positive and low percent false negative throughout all designed analytical conditions

• True- the value that characterizes a quantity perfectly in the conditions that exist when that quantity is considered. It is an ideal value, which could be arrived at only if all causes of uncertainty (error) are eliminated.

So we are looking for a verified valid and true result, correct?

An important but overlooked issue has to do with the addition of salt in the chromatographic method.  Most methods call for it.  The problem is that standards, controls and verifiers infrequently have the salt added in them.  They are instead made up of aqueous solutions without the inorganic salts or without the same amount of inorganic salts as the standards, controls and verifiers as they are not placed in the sample tube, the vacutainer tube, and therefore are devoid of some measurable amount of inorganic salt.

There are two reasons that inorganic salts are in an unknown sample.

1. Most blood samples are taken in a form of gray tube top that typically has some anti-fermentation agent (i.e., an anti-glycotic agent).  This is usually Sodium Fluoride (NaF).  It is usually in a 1% or 2% ratio.
2. Some sort of inorganic salt can be added in the sample preparation phase.  The reason that inorganic salt is added is to change the (phase ratio or B) partition ratio of the analyte in the liquid phase to lower the K or partition ratio and therefore allow it to elute into the gaseous phase more readily.

WHAT DOES “SALTING OUT” MEAN?

“Salting out” can occur when “too much” inorganic salt, such as NaF, is present.  There is no way  to explain it away as anything other than that which it is.  It is a large issue in chromatography.  If present, it can massively effect the proper quantitation of the analyte of interest examined.

“Too much” inorganic salt means too high a ratio of blood to the inorganic salt (usually in excess of 1%-2%) exists.  It is not as a measure of an absolute, but the proportion.  In the case of a vacutainer tube that is certified to fill to 10 mL +/-10% which means that it would be considered properly filled at 9-11 mL will contains 200 mg of NaF.  If the fill is only to 7 or 8 mL of blood, the ratio of blood to NaF will exceed 2%.  This is an example of a potential “salting out” condition

WHY ADD SALT TO BEGIN WITH IN THE PREPARATION PHASE?

In Headspace Gas Chromatography, we use an indirect measure to devine the amount of analyte of interest in the liquid phase.  Briefly described, once the headspace contents in a closed system arrive to equilibrium at a given temperature, pressure and flow, then we can equate the amount of analyte in the headspace to the amount in the liquid phase.  The translation between the headspace (i.e., the gaseous phase) and the liquid phase is called the partition ratio (K).  The higher the partition ratio the harder it is for the analyte of interest to elute or go into the gas phase.  Think of it as a resistance amount.  A higher number, a higher K, means more resistance into the gas phase.  The higher the K, the more the analyte likes to remain in the liquid phase.  Conversely, compounds that have low K values will tend to partition more readily into the gas phase.  So, in HS-GC we want low K.

Why would we want lower K values?

Compounds that have low K values have relatively high responses and low limits of detection.  Sensitivity is increased as the partition coefficient is decreased and volatiles can more readily enter the gas phase.  This is a desirable feature.  The problem is that at ETOH naturally has a very high K value comparatively speaking.

Do we want to add salt?

We don’t have to do so.  Some chemists and chromatographers believe it is desirable to do so as ETOH naturallyhas a very high K value.

	50 degree	60 degree	80 degree
Ethanol	1220	630	240
n-Propanol	520	350	150

Table 1:  Partition ratios of ETOH and n-propanol at verious temperatures

How do we obtain lower K values?

We can add salt.  But most importantly we also have to mindful that not all salts that are added to a sample have the same magnitude of response.

Most Effect

• Potassium Carbonate
• Lithium Sulfate
• Ammonium Phosphate
• Ammonium Sulfate
• No Salt

Least Effect

There is a different layer of context when using inorgnaic salts to lower K values that we must be aware of.  Generally, compounds with K values that are already relatively low will experience very little change in the partition coefficient after adding a salt to an aqueous sample matrix.  So, the impact of adding salt will not be uniform across analytes.  As a result, the impact of adding inorganic salt to n-propanol will be less than when inorganic salt is added and ETOH is present because ETOH has a higher K value to begin with.  Generally, volatile polar compounds in polar matrices (aqueous samples) will experience the largest shifts in K and have higher responses after the addition of salt to the sample matrix.

Again, to re-state it as the concept is very important:  High salt concentrations in aqueous samples decrease the solubility of polar organic volatiles in the sample matrix and promote their transfer into the headspace, resulting in lower K values. The magnitude of the salting-out effect on K, however, is not the same for all compounds.

Some authors who have great names in the field of toxicology do not understand this concept well.  They can design the general concept of the experiment, but they do not conduct them.  They get someone else to design the method.  As a result, some of them clearly get it wrong when it comes to “salting out.”  A well-intentioned toxicologist is not an analytical chemist and vice versa.  Most toxicologists are experts in pharmacokinetics and/or pharmacodynamics, not in chemistry or chromatography.

What is the effect of the “Salting Out” Effect?

Great question.  However, the most comprehensive answer is that when inorganic salts are not within the expected ratio, there is a “salting out effect”.  By adding these salts to an aqueous matrix, some analytes such as ETOH can be forced to enter the gaseous phase above that which would happen if no salts were present.  Remember it is a closed system.

If you don’t have the same amount of inorganic salt added consistently throughout your verifiers, standards, the calibrators, the n-propanol blanks and the unknowns, then you have changed the conditions.  You have changed the method.

How much?  What is the result in terms of quantification?  All a scientist would be able to do is to say honestly is that there is a change and how much that difference is unknown.  There needs to be study.

When you change your method, you have to validate it.  It is just that simple.  A method that has not been validated does not necessarily mean it is invalid just not proven to be valid.

In the most recent American Chemical Society Forensic Chromatography class with Dr. Harold McNair (who is widely recognized as the father of modern adsorption based gas chromatography), Dr. Lee Polite and Lew Fox, we designed an experiment, albeit a simple one, to see if there was an effect in the fact that the blood tube has NaF when the verifiers, standards, the calibrators and the n-propanol blanks do not.  The short experiment did report a difference.

Like most things in life, changing one variable to “solve” a perceived problem often results in another.  Think of the snake who tries to eat its own tail.

When it comes to salting out one must remember that it causes a shift in partition (K) values.

As the standards, the calibrators, the n-propanol blanks and the verifiers are made as to one volume and hence have one K value, then in this situation, in the short draw situation or a large dump of inorganic salt, the K is wrong will lead to an aberrant result.