SUPPRESSION IN ANALYSIS OF ARTIFICIAL SWEETENERS WITH

Transcription

1 APPROACHES TO DECREASE ION SUPPRESSION IN ANALYSIS OF ARTIFICIAL SWEETENERS WITH UPLC-TQ MS Noora Perkola Finnish Environment Institute 3rd Nordic MS Symposium September , Jurmala, Latvia

2 ARTIFICIAL SWEETENERS Acesulfame Potassium (ACS) Saccharin (SAC) Cyclamic acid (CYC) Sucralose (SCL)

3 ARTIFICIAL SWEETENERS Degradation in WWTPs 10 to 20% of SCL Up to 40% of ACS Over 90% of SAC and CYC Detected in environmental water samples Levels up to over 9 µg/l of ACS SAC and CYC up to 0.5 µg/l SCL up to 1 µg/l

4 ARTIFICIAL SWEETENERS ACS and SCL persistent in water Tracers of municipal wastewater Effects largely unknown SCL the most studied No acute toxicity or bioaccumulation observed Behavioral effects: Swimming speed and height (Daphnia magna) Locomotive behavior (gammarids) Respiration (bell shaped) (gammarids) Feeding rate (Calanus glacialis)

5 ION SUPPRESSION Decrease in MS ionization in surface and wastewaters CYC <20% ACS and SAC ~60% SCL ~40% Higher method detection and quantitation limits Can be caused by Interfering matrix compounds matrix dependant Ions originating from the sample, LC eluent

6 ION SUPPRESSION - LINEAR RANGE Compound name: Asesulfaami-K Correlation coefficient: r = , r^2 = Calibration curve: * x Response type: External Std, Height Curve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: None Response ng/ml Figure 1. ACS calibration curve linear range.

7 APPROACHES TO DECREASE MATRIX SUPPRESSION MS: Parameters and quantitation transitions LC: Separation of interfering matrix compounds from analytes Sample pretreatment Extraction method Final solution (to be injected) Dilution of the extract

8 MS PARAMETERS Cone gas flow less matrix into the MS Desolvation gas and energy adduct formation etc. Cone voltage and collision energy optimized for each transition highest obtainable response

9 QUANTITATION TRANSITIONS Transition SCL 395 > 359 Parent -> daughter SCL: m/z 395 > 35 (Berset et al.) Parent -> parent (pseudo MS/MS) SCL: m/z 395 > 395 (Minten et al.) SCL: Ammonium adducts 433 > higher peak area and S/N than for the [M- H] - Less matrix suppression Can t get rid of them => measure them Berset et al. Chemosphere 88 (2012) Minten et al. International journal of environmental and analytical chemistry 91 (2001)

10 LC SEPARATION Column materials: BEH C18 BEH C8 HSS C18 BEH Amide Retention of analytes vs. retention of interfering matrix compounds C18 and C8 similar

11 Figure 2. Chromatogram of standard solution; ACS (1), SAC (2), CYC (3) and SCL (4) with BEH C18 column (1a-4a) and HSS C18 column (1b-4b). Solvents: A = 2 mm NH4Ac (Aq) B = 2 mm NH4Ac (MeOH)

12 LC SEPARATION BEH Amide Good for polar analytes More organic solvent in the LC eluent => higher peak areas on MS

13 a) b) Figure 3. Chromatogram of standard solution; ACS (1), SAC (2), CYC (3) and SCL (4) with HSS (a) and Amide (b) columns.

14 LC SEPARATION BEH Amide Requires quite a strong buffer solution (10 mm) and higher ph for best retention of the sweeteners => ion suppression

15 a) b) Figure 4. Chromatogram of standard solution; ACS (1), SAC (2), CYC (3) and SCL (4) with Amide column: a) Solvents A: 10 mm NH4Ac (Aq) (ph 8) and B: MeCN. b) Solvents A: 10 mm NH4Ac in 50/50 MeCN/H2O (ph 8) and B: 10 mm NH4Ac in 90/10 MeCN/H2O (ph 8).

16 SAMPLE PRETREATMENT - SPE Cartridge ph Sample Recoveries J.T. Baker Bakerbond SDB Milli-Q, spiked SAC 63% ACS 100% CYC 106% SCL 86% Waters Oasis WAX 2.3 Milli-Q, spiked SAC 76% ACS 97% CYC 101% SCL 67% Waters Oasis WAX 3.3 Milli-Q, spiked SAC 75% ACS 91% CYC 86% SCL 74% Waters Oasis MAX 10.8 Milli-Q, spiked SAC 0% ACS 0% CYC 0% SCL 79% Waters Oasis MAX 11.5 Milli-Q, spiked SAC 0% ACS 0% CYC 0% SCL 87%

17 SAMPLE PRETREATMENT - SPE Cartridge ph Sample Recoveries J.T. Baker Bakerbond SDB WWTP effluent SAC 45% ACS 14% SCL 26% Waters Oasis WAX 2.0 WWTP effluent SAC 88% ACS 31% SCL 21% Waters Oasis WAX 2.2 WWTP effluent SAC 68% ACS 26% SCL 16% Waters Oasis WAX 3.0 WWTP effluent SAC 73% ACS 27% SCL 46%

18 SAMPLE PRETREATMENT - EXTRACT SPE with Oasis WAX ph 3 Extraction solvents MeOH for non-polar SCL 1% NH 3 in MeCN for polar analytes Less matrix suppression (final volume 0.5 ml) For SCL, when extracts are analysed separately For SAC and CYC, when extracts are mixed ACS no effect HSS C18 ACS SAC CYC SCL Separate extracts (n=4) Mixed extracts (n=4) Matrix suppression (%) Matrix suppression (%) SCL add

19 SAMPLE PRETREATMENT - DILUTION Dilution of the extract by 10-50%; Dilution of matrix and analytes Peak area decreases, but S/N stays ~same

20 DILUTION OF THE EXTRACT Area diluted sample / Area original extract (%) ACS SCL SAC CYC theoretical value 40 50% 60% 70% 80% 90% 100% 110% Portion of the original extract in diluted sample Figure 5. Dilution of a wastewater sample extract; effect on matrix suppression.

21 DILUTION OF THE EXTRACT SCL: matrix suppression decreased 30% with 1:1 dilution Method recovery for undiluted sample: SPE recovery (95%) matrix suppression (45%) = 50% Method recovery for 1:1 diluted sample: 95% * 45% = 64%

22 IDMS Isotope Dilution Mass Spectrometry Mass labelled standards for all analytes d 4 -ACS d 11 -CYC d 6 -SCL 13 C 6 -SAC Quantification using labelled standards as surrogates Reliable results despite low recoveries

23 Compound name: Asesulfaami-K Correlation coefficient: r = , r^2 = Calibration curve: * x Response type: Internal Std ( Ref 7 ), Height * ( IS Conc. / IS Height ) Curve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: None a Response ng/ml Compound name: Asesulfaami-K Correlation coefficient: r = , r^2 = Calibration curve: * x Response type: External Std, Height Curve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: None b Response ng/ml Figure 6. ACS calibration curve a) ISTD b) absolute response.

24 IDMS RECOVERY CORRECTED RESULTS c (ng/l) ACS: RSD 2.3% SCL; RSD 3.0% SAC; RSD 2.5% CYC; RSD 11% Portion of the extract in diluted sample (%) Figure 7. Dilution of a wastewater sample extract; effect on recovery corrected results obtained with IDMS.

25 FIT FOR PURPOSE Robust Accurate Method LOQs ACS and SAC 25 ng/l CYC 13 ng/l SCL 200 ng/l

26 Thank you! This study was financed by the association of Maa- ja vesitekniikan tuki Ry