LC-MS Applications for Environmental Analysis

Although the observed resolution of alkylated PAHs may

not be enough to replace capillary GC-MS for PAH

fingerprinting applications, the resolution obtained by

SPE-LC-MS/MS could be enough to be used as a screening

tool to decide if a given sample should be analyzed using

those time-consuming techniques, taking advantage of the

low sample consumption and the speed of this

methodology. Additionally, the absence of sample

preparation could provide the ability to track in almost

real time the extent of a contamination by monitoring for

the alkylated PAH-specific concentration patterns

observed at the pollution source. With the gradient

separation used, baseline resolution of the 16 priority

PAHs from their isobaric interferences present in Standard

Reference Material 2260a was obtained except for

benzo[

]fluoranthene, which coeluted with perylene.

Attempts to separate these compounds without a

significant increase in run time were unsuccessful, and

since method speed was a priority, these compounds were

quantified as a group.

Optimization of the Online SPE Procedure

SPE column loading, washing, and reconditioning

parameters were optimized for extraction recovery,

seawater salt elimination, and prevention of carryover

using isotopically labeled PAHs as testing compounds.

Same-day 10 mL injections of 100 ng/L (online SPE) and

100 μL direct injections of 10,000 ng/L solutions in 70%

methanol/water were made, accounting for 1.0 ng on

column for each compound (the 5 mL injection mode was

tested against 50 μL direct-injection, 0.5 ng on column).

Percent recoveries were obtained using averaged peak

areas, using at least three direct-injection runs and two

online SPE runs. The direct-injection method had the

same analytical gradient as the online SPE method. The

observed retention times were in agreement with an 8 min

offset due to the online SPE time, ensuring similar APPI

source conditions at elution in both injection modes thus

enabling the direct comparison of peak areas. Passing at

least 2 mL of aqueous mobile phase through the loading

column after the SPE step was enough to prevent the

transfer of salt residues to the APPI source.

Method Validation

Calibration and quality control

Calibration curves were obtained by plotting the peak

area ratio of each PAH to an isotopically labeled PAH

internal standard against concentration in nanograms per

liter. Linearity was observed for all analytes in the range

used (R

>0.99; 5 to 500 ng/L). Calibration stability was

evaluated every 10 runs by injecting seawater fortified at

100 ng/L. Calibration and method accuracy was verified

by injecting artificial seawater fortified with serially

diluted standard reference materials 1491a and 2260a.

With every analysis batch, a negative (reagent and

sampling) and a positive (fortified at 100 ng/L) blank were

also used. Additionally, one sample duplicate and one

fortified matrix experiment were always analyzed per

every five samples. The system was continuously tested for

carryover by injecting a reagent blank after the highest

calibration standard and after every calibration

verification standard. Compound identification was

considered positive when signals with a S/N ratio above

3 were present in both the quantification and confirmation

SRM transitions, with a maximum retention time

difference of 0.2 min relative to calibration standards or

standard reference materials. Calculated concentrations

below method detection limits (MDLs) were considered

non-detections. A reporting limit (RL) of three times the

MDLs was set in order to reduce the risk of false positives

and ensure data quality.

Determination of method detection limits

MDLs were calculated by multiplying the standard

deviation from seven measurements by the Student

value

(

(7–1, 99)

=3.143), according to procedures outlined by the

US EPA,

using natural seawater (from FIU Campus

Beach, see Table 2), fortified at 50 ng/L. For sensitivity

comparison, MDLs for the traditional LLE+GC-MS

methodology were determined using 1,000 mL of the

same seawater sample also fortified to 50 ng/L and

extracted three times with 50 mL portions of methylene

chloride. The extract was obtained, evaporated, and

cleaned according to established methods (EPA 3510C

and 3630C)

10,11

and analyzed by a GC-MS method

available elsewhere.

The average MDLs corrected for

sample size obtained by LLE-GC-MS analysis are an order

of magnitude higher than those obtained by

SPE-LC-MS/MS.

Although in practice lower MDL

values can be obtained with LLE due to the possibility of

using larger sample volumes, the higher per volume

sensitivity of the online SPE approach is more useful when

limited amounts of sample are available. Also, the low

sample volume required and high sample throughput of

this method facilitate the analysis of multiple quality

controls such as duplicates and fortified matrix

experiments.