Although most analysts already know which approach they need to use for Analysis, this is occasionally a topic of discussion.  While this blog post is not meant to give an absolute answer for each specific application, I hope to provide some tips to steer you in the general direction towards a solution, if this is your dilemma. I will focus here on characteristics of the analyte compound, which should be the primary concern. Other factors, such as cost and detection methods will not be discussed in this blog post.

To use GC for analysis:

First of all, the analyte must be volatile.  This is because it will need to exist in a vapor state in order to partition between the carrier gas stream (mobile phase) and the stationary Phase inside the column.  While it depends somewhat on the choice in column, generally the compound should have a boiling below about 400°-500°C (at atmospheric pressure of 760 mm Hg).  For this reason, most GC analytes are smaller compounds with a molecular weight of less than 1000.  An example of a compound that works well for GC analysis is naphthalene, which has a boiling point of 217.9°C.  Although we have many examples of analyses that include naphthalene, here is a chromatogram that represents one of the most common applications (EPA method 8270):

http://www.restek.com/images/cgram/gc_ev1418.pdf

In order for partition to occur in the vapor state, the molecule must also remain intact. Ideally, it should not decompose upon heating.  In other words, it must be thermally stable (not thermally labile).  For example, riboflavin decomposes between 278-282°C and is generally not analyzed using GC.  Often in cases like this, GC analysis can be done if the compounds are derivatized.

Molecules that can be analyzed by GC or LC:

There are some compounds that could be analyzed equally well by GC or LC.  Bisphenol A is a good example of this.  Here is a link to example chromatograms for both LC and GC analyses of this:

http://www.restek.com/chromatogram/search?s=bisphenol%20A

Some other good examples include compounds like nitrobenzene:

http://www.restek.com/chromatogram/search?s=nitrobenzene

Sometimes analytes that need to be derivatized for GC analysis do not need derivatization for LC.  Chlorophenoxy acid herbicides are a good example of this. Here is an example of GC analysis for these herbicides (derivatized to methyl ester form):

http://www.restek.com/images/cgram/gc_ev00091.pdf

And an example of LC analysis (underivatized):

http://www.restek.com/images/cgram/lc_ev0355.pdf

Although there are many examples of compounds that can be done either way, LC is considered more universal and generally does not require derivatization as often.

To use LC for analysis:

Analytes for LC need to be soluble in a suitable mobile phase, but they do not need to be volatile at all. As a result, compounds range from small to very large.  As is the case for GC, most applications for LC are for organic molecules. Although it may be possible under certain circumstances for some inorganic compounds to be analyzed by LC, this would be beyond the scope of what Restek products can accomplish and will not be discussed here.

LC analysis is usually more difficult for the smallest of molecules, particularly if they coincide with the solvents in the mobile phase, for example methanol, acetonitrile or water.  Also, compounds that exist as gases at room temperature cannot be analyzed by LC, or at least it would not be practical.

In LC, to allow partitioning between the liquid mobile phase and the stationary phase inside the column, a compound must be reasonably soluble in the mobile phase and it must have some affinity toward the stationary phase. The phrase “like dissolves like” is very applicable when considering solubility. A compound’s chemical interaction toward a stationary phase could occur in several different ways.  If interested in reading more on this topic, I suggest reading USLC Column Selection and Mobile Phase Adjustment Guide. As discussed in the guide, four of the primary mechanisms are dispersion, polarizability, hydrogen-bonding and cation exchange.

As in GC, an analyte for LC also must remain intact and not decompose. Fortunately, thermal stability is not a concern for LC, since analysis can usually be performed at or near room temperature.  I mentioned earlier that riboflavin does decompose upon heating.  We find that riboflavin is analyzed fairly easily by LC, though, as shown in the following chromatogram:

http://www.restek.com/images/cgram/lc_ff0558.pdf

While decomposition is not common with LC analyses, ionization in aqueous solution occurs quite often. Consequently, such analytes are affected dramatically by the pH of the mobile phase.  To control these affects, most analysts will use buffer in the mobile phase. If interested in reading more on this topic, please refer to the following:

When should you use a buffer for HPLC, how does it work and which one to use?

New Advice on an Old Topic: Buffers in Reversed-Phase HPLC

Resources available:

A good tool to use that is at your disposal is Restek Searchable Chromatogram Library.  To look for example analyses for compound(s) of interest, simply type their name or CAS number in the search box. You may see examples for the analysis done by LC or GC, or perhaps both.  If the compounds of interest tend to decompose, become reactive or are difficult to detect, you may see examples of the analysis done by derivatization.

If interested in reading more on this topic, here are some articles that may be helpful:

http://bitesizebio.com/29109/run-fly-comparison-hplc-gc/

http://www.news-medical.net/life-sciences/Liquid-Chromatography-versus-Gas-Chromatography.aspx

http://lab-training.com/2014/04/02/what-are-the-differences-between-gc-and-hplc/

I hope you find this helpful. Thank you for reading.