Background Image
Table of Contents Table of Contents
Previous Page  214 / 302 Next Page
Information
Show Menu
Previous Page 214 / 302 Next Page
Page Background

4

Routine, Targeted and Non-Targeted Analysis of Environmental Contaminants of Emerging Concern – Development and Validation of a UHPLC Orbitrap MS Method

Table 2 lists screening results

samples along with their occu

312 CEC database. Of the 48

bisphenol A was the only one

TABLE 1. Results of targeted compound analysis.

FIGURE 3

.

Sample preparation and analysis using UHPLC-Orbitrap MS.

rbitrap MS Analysis

(CH

3

OH) were purchased from Fisher

er used for aqueous mobile phases

g reverse osmosis water through a

ater purification system (Mississauga,

SB) method E3454

1

was used to

s and non-targeted compound

anada) HLB solid phase extraction

xtraction. Method E3454 has been

ratory Accreditation (CALA) since

purchased from Sigma-Aldrich

labelled standards were purchased

and Cambridge isotope Laboratories

led intermediate standard solutions

k solutions in CH

3

OH. Five levels of

iluting intermediate solutions with

entific™ Dionex™ UltiMate™ 3000

p, WPS-3000 autosampler, and a

made by injecting 5

µ

L extracts into

mode) and an Agilent XDB C-18,

ctively, for positive and negative mode

lysis is available on request (Ref. 1).

ic™ Exactive™ Plus Orbitrap MS

interface. The Orbitrap MS system

e modes by infusion of standard

itrogen (>99%) was used in the ESI

00 and 3,200 V for positive and

cquired at a resolving power of

eak width at

m/z

200, R

FWHM

),

n using automatic gain control target

ted, quantitative analysis of 61 CECs.

n-targeted screening along with a

ically active compounds, steroids,

. TraceFinder software is used to

+Na)

+

in the positive mode and (M-

ounds listed in the database. The

gram (XIC) using a mass extraction

tically identified using an XIC area

(ppt) depending on compound), a 5

(M) and an isotopic (M+1) peak

of < 10%. Typical screening time was

se. Analytical results were interpreted

nd exported to Microsoft Excel® with

entation.

S4

S6

ound analysis in the collected

ction limits (MDL). A total of 21 of the

les analyzed.

Screening of Non-targeted Compounds

The identification of non-targeted compounds uses accurate mass of the monoisotopic

peak M and isotopic (M+1) peaks, relative intensities of the M/(M+1) peaks and

isotopic pattern of the halogenated compounds. Manual inspection of line shape of the

XIC chromatogram, major fragment ions from the mass spectrum will also improve the

confidence and credibility of analytical results. Figure 4 shows examples of true

positive identification of 3-phenoxybenzoic acid, a common primary metabolite of the

synthetic pyrethroid insecticides (4A); a halogen containing xenobiotic 3,5-dibromo-4-

hydroxybenzoic acid (4B); and an artificial sweetener sucralose (Splenda®) (4C).

Figure 5 shows an example of false-positive identification of desethylatrazine, an

environmental metobolite of the pesticide atrazine. The mono-isotopic peak M has a

mass error of 0.2454 ppm and the XIC has a perfect Gaussian shape and can be

considered as a positive identification. However, mismatch of chlorine isotopic pattern

concludes that the compound in question is not desethylatrazine.

the three sampling locations.

MDL #1, PS #1, PS #2, PS #1, TWAS #1, PE #2, PE #2, SE #1, FE Permate Permate

Acetamidophenol

100 <MDL <MDL 578.5 111.5 1952.5 4026.5 <MDL <MDL 105.0 <MDL

Atenolol

50 <MDL <MDL <MDL <MDL 143.5 579.5 288.5 <MDL <MDL <MDL

Atorvastatin

10 147.5 3419.5 3417.5 107.5 <MDL 165.0 <MDL 3419.5 110.0 72.5

Bezafibrate

20 <MDL <MDL <MDL 49.0 53.0 109.0 23.0 <MDL <MDL <MDL

Caffeine

20 147.5 75.5 525.0 34.5 4426.5 <MDL 35.0 <MDL 288.0 105.5

Carbadox

200 <MDL <MDL <MDL <MDL <MDL <MDL <MDL <MDL <MDL <MDL

Carbamazepine 2 332.5 <MDL <MDL 107.0 24.0 235.5 115.0 <MDL 262.0 183.5

Ciprofloxacin

100 918.0 289.0 289.5 316.0 304.5 319.5 315.0 298.5 606.5 536.0

DEET

150 <MDL <MDL <MDL <MDL <MDL 338.5 <MDL <MDL <MDL <MDL

Diclofenac sodium 100 <MDL <MDL <MDL 166.5 <MDL 148.5 <MDL <MDL <MDL <MDL

Hydrocortisone 5 1781.5 <MDL <MDL 7.5

8.0 11.0 9.5 <MDL 976.5 636.5

Lidocaine

10 38.0 <MDL <MDL 114.0 <MDL 118.5 54.5 <MDL 32.5 35.5

Oxolinic Acid

20 <MDL <MDL <MDL 81.0 <MDL 125.5 49.0 <MDL <MDL <MDL

Progesterone

20 29.0 <MDL <MDL <MDL <MDL 957.5 <MDL <MDL 108.5 <MDL

Bisphenol A

200 4458.0 617.5 1252.5 522.0 1211.0 1675.5 213.0 249.0 3621.5 3383.5

Equilin

50 1619.0 449.5 <MDL <MDL 678.0 1531.0 1337.0 <MDL 1441.0 345.5

Estriol

200 <MDL 472.5 <MDL 1006.0 216.5 <MDL <MDL <MDL <MDL <MDL

Gemfibrozil

10 <MDL 121.5 193.0 281.0 174.0 125.0 77.5 127.0 260.0 227.0

Oxybenzone

50 158.5 346.0 <MDL 170.0 237.0 166.5 170.0 196.0 159.5 159.5

Triclocarban

50 <MDL 734.0 429.0 366.5 1176.5 532.0 247.5 411.0 <MDL <MDL

Triclosan

120 3068.5 <MDL <MDL 2422.0 <MDL <MDL <MDL 343.5 777.5 912.0

ng/L

Compound

FIGURE 4

.

Examples of true

FIGURE 5

.

Example of a fal

Reconstituted in 250 ul vial

insert using 100 ul H20

Filtered samples were extracted

at pH7 using Waters™ HLB SPE

cartridge

Grab sample collected in

1-L brown glass bottles

Extracts evaporated

to near dryness

UHPLC Orbitrap MS Analysis

Quantitative and non-targeted

analysis by TraceFinder software