Solid phase extraction methods usually require a large amount of elution solvent to pull everything off of the sorbent bed. However, evaporating and reconstituting those samples can take a lot of time. So is it possible to do an elution that doesn’t require evaporation and reconstitution? We were able to develop a method that resulted in recovery of all compounds, all without evaporating our samples.
We started by using the method that we developed when determining our optimal sample size for a 10 mg sorbent bed. We analyzed our 100 compound drugs of abuse panel in urine and used a 50 uL sample volume. We decided to try different volumes of 2 separate elution solvents: 98:2 methanol/ammonium hydroxide and 2.5% formic acid in water. We wanted to make sure to elute all of our compounds of interest without putting a too basic or too aqueous solution onto our mass spec. We tried the following combinations:
98:2 MeOH/NH4OH | 2.5% formic acid (aq) | |
Combination 1 | 1 x 150 µL | 1 x 100 µL |
Combination 2 | 2 x 75 µL | 2 x 75 µL |
Combination 3 | 2 x 50 µL | 2 x 50 µL |
Combination 4 | 1 x 100 µL | 1 x 100 µL |
Of all of the elution combinations, Combination 1 resulted in the highest recoveries for most of our analytes of interest. The below chart compares our microelution protocol with a “normal SPE protocol” that included evaporation and reconstitution.

There were a few compound classes that had decreased recoveries, including tricyclic antidepressants (nortriptyline), selective serotonin reuptake inhibitors (duloxetine), and some opioids (EDDP). While the recoveries were decreased for these compounds, we did still achieve recovery of all analytes in our panel. We did not see an increase in matrix effects when using our microelution protocol, which means our samples were fairly clean!

While a 10 mg sorbent bed mass isn’t usually used with a microelution with no evaporation, it is possible to do it! Depending on the compounds being analyzed, recoveries can be as good as recoveries seen when using a “normal” SPE protocol.