Analysis of acid herbicides in drinking and surface water using on-line SPE

Figure 1. Molecular structures of selected analytes.

Introduction

Acidic herbicides cover a broad range of compounds which are widely used in crop protection and general weed control. They tend to be extremely water soluble, and can easily enter surface and ground waters through natural drainage. Acid herbicides are toxic to humans and aquatic organisms, and are therefore monitored to low ng/L levels in potable and ground waters.

The on-line approach to sample preparation has grown in popularity because of its advantages in improved workflow and reduced sample handling:

• Little or no sample pre-treatment
• Totally automated procedures
• High precision and accuracy
• Elimination of blow down and reconstitution steps
• Reduced solvent use and disposal costs

Trace analysis of organics in water traditionally involves large sample volumes, labor intensive procedures and relatively high use of solvents. The on-line SPE approach uses a simple, well established hardware setup, and fully integrates sample preparation into the analytical workflow. Typical sample volumes of 1-10 mL, and lower solvent usage mean sample collection, transport and handling, along with solvent disposal costs, are much reduced.

This application note describes the use of ISOLUTE® ENV+ on-line cartridges in a fully automated on-line SPE-LC-MS/MS method for extraction and quantification of 16 acid herbicides in drinking and surface water.

Analytes

2,4,5-T, 2,4-D, 2,4-DB, Asulam, Benazolin, Bentazone, Bromoxynil, Dicamba, Dichlorprop, Ioxynil, MCPA, MCPB, Mecoprop, Pentachlorophenol, Quinmerac, Triclopyr

Sample preparation procedure

Format:

ISOLUTE® ENV+ On-line SPE cartridge 30 mm x 2.1 mm, part number OSPE-916-32150

Overview

100 µL of pre-treated sample (hard, medium or soft drinking water or surface water) was injected and loaded (trapped) onto the ISOLUTE ENV+ on-line SPE cartridge using a mobile phase consisting of 2% acetonitrile/98% 0.01% formic acid (aq). After 1 min, valve positions were switched to enable transfer of the analytes (in backflush mode) onto the analytical cartridge. Analytes were separated using the gradient conditions shown in Table 1 and analyzed using the MS/MS conditions described in Table 3. The total cycle time was 18.5 mins.

Sample pre-treatment

A 5 mL aliquot of each sample was taken to which internal standard solution and 5% (aq) formic acid (75 µL was added. The samples were vortexed to ensure thorough mixing before 2 mL was transferred to a 2 mL auto sampler vial for analysis

Instrument set up

SPE-HPLC system: Agilent 1200 series HPLC with 2 binary pumps, switching value, column oven and temperature controlled autosampler (set to 10 ⁰C).

HPLC column: Zorbax Eclipse plus C18 2.1 x 100mm, 1.8 micron analytical column.

Binary Pump 1

Mobile Phase Channel A: 0.01% Formic acid in high purity water
Mobile Phase Channel B: Acetonitrile
Figure 2. Switching valve configuration a) (Top) Valve position 1→2; b) (Bottom) Valve position 1→6

HPLC conditions

Table 1. Pump 1.

Stop time: 16.50 minutes  Post run: 2.00

Binary pump 2

2% Acetonitrile: 98% 0.01% Formic acid in high purity water (isocratic). Flow 0.3 mL/min.

Auto sampler injection parameters

Injection Volume: 100 µL
Injection with needle wash: Needle wash flush port 5.0 Sec. Draw Speed: 150 µL/min
Eject Speed: 1000 µL/min
Draw Position: 0.0 mm
Equilibration Time: 3 secs
Sample Flush-Out Factor: 15 times injection volume Vial/Well Bottom Sensing: Enabled
Stop time: As pump

Column compartment

Temperature: Thermostatically Controlled at 50 °C
Enable analysis: when temp. is within ± 0.8 °C (Left & Right)

Table 2. Switching valve timing.

Stop time: As pump

Mass spectrometer conditions

Instrument set up

Agilent 6490 QQQ with ifunnel

Table 3. MS/MS conditions.

Table 4. Internal standard allocation.

Calibration curves

Representative samples were spiked at a range of concentrations (shown in table 5) and calibration curves were constructed. Selected examples are shown below. Calibration curves showed a quadratic fit with a r² >0.999 achieved for all compounds. See figures 3a–3c for calibration curves for selected analytes.

Table 5. Calibration range.

Example calibration curves

Figure 3a. Asulam.

Figure 3b. MCPA.

Figure 3c. Pentachlorophenol.

Figure 4. Typical chromatogram.

Average analyte recovery (n=10), along with measured trueness and precision data for 4 different matrices are shown in tables 6a and 6b.

Precision was calculated by using ((2 x standard deviation)/ Mean) * 100. Trueness was calculated by using ((Mean – Expected value) / Expected value).

Limits of detection, calculated using 3 times the relative within batch standard deviation of repeat analysis of a low level standard, were in the low ng/l range for all analytes.

Table 6a. Analytical data (soft and medium water).

Table 6b. Analytical data (hard and surface water).

Conclusions

• The ISOLUTE® ENV+ On-line SPE cartridge provided reproducible pre-concentration of a wide range of acid herbicides in a variety of water matrices
• Limits of detection in the low ng/mL range are achievable for all analytes
• Reproducibility between batches was seen on the data produced over three separate days. The data was shown to be comparable by using a paired T-test
• No carryover was seen in a randomized batch containing blank and spiked samples.

Acknowledgements:

The author would like to thank the analysts in the Chemistry Department of DCWW’s Glaslyn Laboratory.

Ordering information

Literature Number: AN863