The Importance of Using GC-TOFMS for Flavor and Off-Flavor Analysis Related to Packaging Issues For Leco Corporation September 19, 2013

Transcription

1 The Importance of Using GC-TOFMS for Flavor and Off-Flavor Analysis Related to Packaging Issues For Leco Corporation September 19, 2013 Ray Marsili Marsili Consulting Group

2 Overview Raspberry syrup shelf life study with two types of bottle. Chemicals extracted from paperboard packaging. 2-Nonenal in wine bottle closure (resin based). Fungal metabolites in casein powder. Cheese powder: Finding a small peak embedded in a large peak. General remarks regarding spectral skewing and fast acquisition scan rates. 2

3 Packaging Related Off-Flavors 3 Water Light CO Microbes 2 Off-Flavors Flavors Oxygen Water External Contaminants Scalping: Loss of flavor chemicals from product into/through packaging. Leaching: OF chemical contamination from packaging into product (plasticizers, antioxidants, slip agents, photoinitiators, ink solvents, etc. ).

4 The GERSTEL Twister Automated Extraction from Liquid Samples 4 Magnetic core, sealed in glass and coated with PDMS l = 10 mm, df = 0.5 mm df = 1.0 mm l = 20 mm, df = 0.5 mm df = 1.0 mm 24 µl 63 µl 47 µl 126 µl SPME : max. 0.5 µl

5 The GERSTEL Twister Stir Bar Sorptive Extraction (SBSE) [a.k.a. Gum Phase Extraction (GPE)] 5 Same principle as liquid/liquid extraction Extraction is based on sorption Predictable recovery due to log Ko/w PDMS sorbent Highly inert Retains no water Selectivity eliminates polar matrix interferences Desorption with TDS or TDU Extremely low detection limits (ppt to ppq) Marsili Consulting Group 11/2007

6 Raspberry Syrup Shelf Life Study: Two Types of Bottles Advantages of Peak Deconvolution Capabilities for Flavor Research Needs to be Emphasized 6 Monitored flavor chemicals Which are constant Which increase (e.g., leaching) Which decrease (e.g., scalping) Determine if PCR PETE (polyethylene terephthalate) bottles perform as well as regular PETE bottles. Shelf life conditions: 90 F for 0, 13, 25 and 49 Days. Two SPME methods: Method A: 2gms syrup extracted with CARB/DVB/PDMS fiber with a 40 min C. Method B: Same as above, only using a 10 min 40 C. Marsili Consulting Group 11/2007

7 Critical Flavor Chemicals That Remained Constant Raspberry ketone, methyl ether 7 α-trans-ionone Note: Not detected with quadruple GC-MS Natural Raspberry Flavored Syrup Shelf Life Study: PCR PETE Bottles vs. Regular PETE Bottles Marsili Consulting Group 11/2007

8 SCALPING RESULTS 16 Flavor-Important Chemicals Showed Steady Declines in Concentrations During Shelf Life in Both Types of PETE Bottles: Ethyl acetate Ethyl-2-methylbutanoate* 2-Hexenal 3-Heptanol Ethyl hexanoate 2-Hexen-1-ol, acetate Hexyl acetate Benzyl alcohol 3-Methylbutyl-2-methylpropanoate Phenyl methyl acetate 6-Camphenol 2-Butoxy-1-methyl-2-oxoethylbutanoate Methyl-2-aminobenzoate Decanoic acid delta-decalactone Ethyl decanoate 1,2,3-Trimethyl-cyclopent-2-enecarboxaldehyde *Seven chemicals in red were not detected by quadruple GC-MS 8

9 LEACHING RESULTS Five Chemicals that Showed Steady Increases in Concentration During Shelf Life for Both Types of Bottles: 9 CHEMICALS GENERATED BY MAILLARD REACTIONS, ETC.: Furfural (Maillard/carmelization reaction) Ethyl benzoate (from reaction between benzoic acid and ethyl alcohol) Benzothiazole (Maillard reaction product)* CHEMICALS LEACHED FROM PETE RESIN: 2,4-di-tert-Butylphenol (Chemical leached from packaging) 2,2,4-Trimethyl-1,3-pentanediol diisobutyrate (KodaflexTXIB, a plasticizer) *not detected by quadruple GC-MS

10 Conclusion of Raspberry Flavor Shelf Life Study 10 Same two major flavor chemicals remained constant for two types of bottles (α-trans-ionone and raspberry ketone, methyl ether). 16 important flavor chemicals (especially esters) were scalped at same rate in the two types of bottles. Five chemicals showed steady increases in concentration in both types of bottles (leaching and Maillard reactions). Even though scalping and leaching occurred, they occurred at same rate, so two bottle types were equivalent.

11 Chemicals in Packaging Inks Direct Analysis by Dilute and Shoot (EtAc) or by Direct Thermal Desorption (DTD) e+007 3e e+007 2e+007 EtAc Dilution (60-70 mg ink/5ml EtAc) Benzene Photoinitiators (acrylate derivatives) derivatives e+007 1e+007 5e DTD (Microvial), 20 mg, t1 DTD (Microvial), 20 mg, t2 0 Time (s) T IC 20 mg N.Y :C ink (no I.S.) t1 microvial:1 T IC 20 mg N.Y :C ink (no I.S.) t2 microvial:1 TIC N.Y:C EtAc solvent 1uL SW69.5mg:1 (1) 2,3-epoxypropyl acrylate ( ); (2) 1,2-benzenedicarboxlic acid, mono(2- ethylhexyl) ester (MEHP) ( ); (3) PETIA ( ); (4) 11 photocure sb907 ( ); (5) Glycidyl acrylate ( ); (6) diisooctyl adipate (Monoplex DIOA) ( ); (7) acrylic acid; (8) butyl acrylate; (9) 2-methyl benzoquinone; (10) 1,1,2- trimethylpropyl benzene; (11) BHT

12 Extract Ink Chemicals From Paperboard Using Matte Cell Extraction 12 Simulant Fill Top Plate A B C D Teflon Spacer Paperboard (A) Assembled cell; (B) Paperboard (50 cm 2 ) mounted on bottom plate; (C) Teflon spacer to allow fill of 30 ml of simulant (10% EtOH solution); (D) Top plate screwed down to make A. (Design from Prof. T.G. Hartman, Rutgers)

13 13 Matte Cell Extraction of Paperboard: MeCl 2 vs SBSE 3.5e+007 3e e+007 2e e+007 1e+007 5e Time (s) Analytes extracted from 30 ml simulant (10% EtOH) exposed to 50 cm 2 of paperboard at 40 C for 10 days. Twister (SBSE) Plasticizers and Photoinitiators Methylene Chloride Extraction TIC Matte Cell #1, 30uL 1.1mg 2-undecanone/mL EtOH in 1 ml MeCl2, 1ul inj:1 TIC Matte Cell #1C 10%EtOH SBSE 22hrs, 10uL 1.1mg 2-undecanone/mL IS: (1) decanal (2) 2-undecanone (I.S.) (3) 1,2,2-trimethylpropyl benzene (4) 2-dodecanone (5) Kodaflex TXIB ( ) (6) benzophenone (7) Irqacure 184 ( ) (8) 2,3-epoxypropyl acrylate ( ) (9) methyl o-benzylbenzoate (10) Parsol MOX ( ) (11) Padimate ( ) (12) 4-phenylbenzophenone (Runtecure 1059) ( ) (13) MEHP ( )

14 Matte Cell Extraction of Paperboard by SBSE (30 mls 10% EtOH simulant exposed to 50 cm 2 of paperboard for 10 o C) Peak Deconvolution of Leco Pegasus Needed to Find Dimethoxyphenyl acetophenone (DMPA), a Photoinitiator 14 8e+006 7e+006 6e+006 5e Ethanone, 2,2-dimethoxy-1,2-diphenyl- (DMPA) DMPA 8,9-Dihydrocyclopenta[def]phenanthrene Undecanoic acid, 2-methyl- 4e+006 3e s 996s 2e+006 1e Time (s) TICx Time (s) TIC

15 Matte Cell Extraction of Paperboard by SBSE (30 mls 10% EtOH simulant exposed to 50 cm 2 of paperboard for 10 o C) Peak Deconvolution of Leco Pegasus Needed to Find Dimethoxyphenyl acetophenone (DMPA), a Photoinitiator 15

16 Quantitation of Benzophenone in Paperboard Packaging (SBSE) 13.3 ng benzophenone/cm 2 16 Time (s) TIC 2-undecanone (I.S.) Time (s) x10 Matte Cell Extraction: 30 mls 10% EtOH/water simulant, 40 C for 10 days; 22 ppb 2-undecanone I.S. added.

17 Quantitation of Benzophenone in Paperboard Packaging (SBSE) ppb Benzophenone Peak Area Benzophenone/Peak Area IS vs. ppb Benzophenone Benzophenone standards: ppb Internal Standard: 2-undecanone (22 ppb) Extracted from 30 mls 10% EtOH for 3 hrs. y = x R 2 = Peak Area Benzophenone/Peak Area IS Note: Different Twister used for each standard.

18 Critical Peak Deconvolution Advantage of the Leco Pegasus GC-TOF 1000 Peak Deconvolution for (Z)-2-Nonenal From Wine Bottle Closure (Resin-based) Caliper - sample "12 wk heat abused beer SBSE 5uL IS 10 ml SW 1hr t1:1", s to s - 0 s to 0 s Peak True - sample "12 wk heat abused beer SBSE 5uL IS 10 ml SW 1hr t1:1", peak 307, at s Library Hit - similarity 815, "2-Nonenal, (Z)-" Caliper True (Deconvoluted) Library Ethyl octanoate 73, 88, 101 from Ethyl octanoate Z-2-Nonenal Time (s)

19 Fungal Metabolites by SBSE 19 Major off-flavor problem with casein powder. Environmental plus fungal contamination. Problem: Two instances of TCA contamination of casein powder and one instance of TBA contamination of casein powder occurred in summer of Tainted (musty) caseins were used to make nutritional beverages. Several law suits resulted; strong musty taint in nutritional beverages made with contaminated caseins.

20 MUSTY COMPOUNDS 2-MIB (2-methyl isoborneol) Geosmin Musty Taint in Casein Powder Odor threshold 5-10 ng/l 1-10 ng/l Odor type earthy musty Molecular mass 2,4,6-TCA (2,4,6-trichloroanisole) 0,1-2 ng/l musty 212 2,3,6-TCA (2,3,6-trichloroanisole) 0,1-2 ng/l musty 212 2,3,4-TCA (2,3,4-trichloroanisole) 0,2-2 ng/l musty 212 2,4,6-TBA (2,4,6-tribromoanisole) 0,15-2 ng/l musty 346

21 Musty Taint in Casein Powder: Trihaloanisole Contamination Cl From pesticides and wood preservatives Odor thresholds for TCAs and TBAs are ng/l 21 Cl OH Cl Cl chlorophenol O- methyltransferase Microbial (i.e., fungal enzymatic) o-methylation to form the strong musty trichloroanisole from the trichlorophenol. Cl Cl O Problem: Two instances of TCA contamination of casein powder and one instance of TBA contamination of casein powder occurred in summer of Tainted (musty) caseins were used to make nutritional beverages. Several law suits resulted. Marsili Consulting Group 4/19/08

22 Musty Taint in Casein Powder 1gm musty casein + 25 ml water, stir 30 RT; SBSE for 3 hr with 2 cm x 0.5 mm Twister 22 Time (s) TIC Marsili Consulting Group 4/19/08

23 Musty Taint in Casein Powder Finding chloroanisoles in the chromatogram 23 2,4-dichloroanisole 2,4,6-trichloroanisole 7ppb 2,3,6- trichloroanisole + more! Time (s) Marsili Consulting Group 4/19/08

24 Musty Taint in Casein Powder The value of peak deconvolution 24 2,4-dichloroanisole 2,4,6-trichloroanisole 2,4,6-trichlorophenol 2,3,6-trichloroanisole butyl butanoate Time (s) Marsili Consulting Group 4/19/08 Time (s)

25 Musty Taint in Casein Powder TBA and TBAd Spike (133 ppb) in Musty Casein Time (s) TBAd spike (m/z 351) 2,4,6-tribromoanisole (m/z 346) Technique is called SIDA (Stable Isotope Dilution Analysis). SIDA is arguably the most accurate technique for quantitative determination of organic compounds. (Alan D. Harmon in Flavor, Fragrance and Odor Analysis.) O 50 Br Br Br (mainlib) Benzene, 1,3,5-tribromo-2-methoxy-

26 Musty Taint in Casein Powder TBAd Standard Curve in Control Casein TBAd spikes in 3g Casein Powder 26 ul Spike (0.02ug/uL TBAd) ppb y = x 2 R = Peak amu 667 ppb Marsili Consulting Group 4/19/08

27 Automatic Peak Find and Spectral Deconvolution: Dimethyl Sulfone in Cheese Powder 27 Analysis by TOF Raw spectrum Butyric Acid Deconvoluted spectrum Dimethyl sulfone Library spectrum Marsili Consulting Group 7/2006

28 Rel. Abund. Rel. Abund Spectral Bias (Skewing) TOFMS m/z m/z Intensity Intensity GC Peak Time (sec) Time (sec) Rel. Abund. Rel. Abund. Scanning MS m/z m/z 28 Rel. Abund Intensity Rel. Abund m/z Time (sec) m/z

29 Acquistion Rate and Data Density (18-20 spectra across peak needed for automated peak find) spectra/second needed to obtain the ideal spectra across this peak. Ten spectra/sec (MSD) can distort peak shape, peak height and apex location.

30 Spectral Acquisition Rate and Automated Peak Find for Nine (9) Pesticides 30 Peak definition is very poor. 1 Spectra/sec

31 Spectral Acquisition Rate 31 2 Spectra/sec Typical scanning instrument rate Automated peak find is not possible.

32 Spectral Acquistion Rate 32 5 Spectra/sec 5 peaks located

33 Spectral Acquisition Rate Spectra/sec 6 peaks located

34 Spectral Acquisition Rate Spectra/sec 8 peaks located

35 Spectral Acquisition Rate Spectra/sec Metolachlor 238 Malathion 158 Fenthion 278 Chlorpyrifos 199 Parathion 291 DCPA 301 Cyanazine 68 Isodrin 193 Trichloronat 269 All 9 peaks located

36 Why Leco GC-TOFMS for Peak Deconvolution? New sample extraction techniques result in high extraction recoveries and more complex chromatograms with overlapping peaks. 36 Best peak deconvolution with: o An array detector like TOF with no MS skewing. o High data density per chromatographic peak (e.g., 40 spectra/sec). Leco TOFMS provides high sensitivity and peak deconvolution offers advantages of improved mass spectral matching and more accurate peak area determinations. ChromaTOF auto peak find saves time.

37 Thank you For your Time and Attention For More Information Contact: LECO Corporation Lakeview Ave. St. Joseph, MI Phone: