Tasting "Weird and Wild" Jelly Bean Flavors UsingHS SPME Gas Chromatography Mass SpectrometryLECO USACorporation; Saint Joseph, Michigan

Key Words: GC-TOFMS, GC-MS, Pegasus BT, Flavor®

1. Introduction

Bring Your Child to Work Day is a national event in the United States, which is often difficult for children of scientists tocelebrate due to lab safety protocols. This application note details a safe experiment designed to stimulate interest in thesciences for all ages while demonstrating the ease-of-use and capabilities of a new benchtop Time-of-Flight GC-MS toeffectively differentiate exotic flavor profiles in store-purchased jelly beans.

As part of the experiment, children participated in a game, taking turns selecting jelly beans from one of several visually-identical pairs. Each jelly bean pair had either an appealing or appalling taste. After tasting their selections, each childquickly realized there was no way to successfully predict which jelly bean was the appealing taste. In the second part of theexperiment, the children were allowed to prepare and analyze a portion of their jelly beans using the BT GC-MS,Pegasusand use the data to assist them in making their selection. When using the GC-MS data, the children were consistentlysuccessful at selecting their preferred flavor.

Life

Sci

ence

and C

hem

ical A

naly

sis

Solu

tions

Figure 1. The figure above shows GC-MS total ion chromatograms (TIC) for a good-flavored jelly bean (Top, lime) vs. a bad-flavored one(Bottom, green grass) along with Peak True (deconvoluted) and NIST library spectra for a key flavor-contributing compound from eachjelly bean. The table contains additional flavor-contributing compounds and their flavor descriptors, which each child used to successfullychoose good-flavored jelly beans.

2.5 5 7.5 10 12.5 15 17.5 20

0

0.3e9

0.6e9

0.9e9

1.2e9

1.5e9

1.8e9

2.1e9

2.4e9

2.7e9

Time (min)

Lime Green GrassCompound Name & Descriptors Compound Name & Descriptors

D-Limonene (citrus, orange, fresh, sweet) (Z)-3-hexen-1-ol (fresh green cut grass)

ç-Terpinene (lemon/lime, tropical) estragole (sweet, phenolic, weedy)

terpinolene (fresh, citrus, lemon and lime-like, terpy) benzyl alcohol (floral, rose)

octanal (green with peely citrus orange)

40 60 80 100 120 140 160

0

0.2e7

0.4e7

0.6e7

0.8e7

1.0e7

1.2e7

1.4e7

1.6e7

0

100

200

300

400

500

600

700

800

900

1000

M/Z

93.06121.08

121

51

Peak True - sample "HS SPME Lime", Terpinolene, at 9.52912 min, Area (Counts)

Library Hit - Similarity: 907 - Library: replib - Terpinolene, Abundance

40 60 80 100 120

0

1e7

2e7

3e7

4e7

5e7

6e7

0

100

200

300

400

500

600

700

800

900

1000

M/Z

67.0741.07

Peak True - sample "HS SPME Grass Clippings", 3-Hexen-1-ol, (Z)-, at 10.2783 min, Area (Counts)

Library Hit - Similarity: 923 - Library: replib - 3-Hexen-1-ol, (Z)-, Abundance

Lime

Green Grass

2. Experimental

As part of the experiment, children participated in a game in which they took turns selecting a jelly bean from one ofthe matching pairs shown in Table I below.

Table I. Description of Jelly Bean Pairs

They were then asked to taste their selection to see if they had "won" or "lost" by selecting an appealing vs. appallingflavor. It was evident that selections based on visual appearance of the jelly beans alone led to random resultsduring the taste tests. Example responses from our willing participants are shown in the photos below.

A second round of the game was started, and each child was given an option to have LECO's BT GC-MSPegasussystem "taste" the jelly beans first, prior to the children making a selection. The components identified by the GC-MS,and their respective taste and odor descriptors, were used to aid in selecting jelly beans in this round.

The children participated in safely preparing the jelly beans for a HS-SPME GC-MS experiment that would providethe data used to influence their selections. A portion of a single jelly bean was placed into a 20 mL HS vial along with5 mL of HPLC grade H O, and 1g of NaCl.2

Delive

ring t

he R

ight

Resu

lts

Good Bad

Fruity Smelly Socks

Popcorn Spoiled Eggs

Peach Vomit

Lime Green Grass

Blueberry Minty Toothpaste

Chocolate Dog Food

The children were then able to load the samples into the autosampler of the GC-MS system and start the sample queueon their own.

A 1 cm PDMS/Carboxen/DVB SPME fiber was used to extract aroma compounds from the headspace of each jelly beanand deliver them to the GC–MS for analysis. An Rtx-200 MS column was used for the chromatographic separation. TheTOFMS data were acquired from 30 to 500 m/z at 20 spectra/second. The jelly beans were analyzed on the BTPegasusunder the conditions shown below in Table II.

Table II. GC-TOFMS BT) Conditions(Pegasus

3. Results and Discussion

The benchtop time-of-flight GC-MS data was used to differentiate the visually identical jelly beans based on thecomponents detected and identified in their headspace. The data was compiled into tables with key analytes and theirtaste/odor descriptors so that the kids could use the information to influence their decision when making their second-round jelly bean taste selections.

The Total Ion Chromatograms (TICs) with tables containing the key differentiating ingredients for each Jelly Bean typeare shown in Figures 2 through 6. Each chromatogram also includes a Peak True (deconvoluted) mass spectrum fromone of the key taste/odor analytes and the NIST library match with similarity score for tentative identification.

Life

Sci

ence

and C

hem

ical A

naly

sis

Solu

tions

Gas Chromatograph Agilent 7890 with LECO L-PAL 3 Autosampler

Injection 2 min SPME desorption, Split 10:1 at 250 ° C

Carrier Gas He @ 1.0 mL/min, Constant Flow

Column Rtx-200 MS, 30 m x 0.25 mm i.d. x 0.25 µm coating (Restek)

Oven Program 5 min at 35 °C, ramp 20 °C/min to 240 °C, hold 5 min

Transfer Line 250 °C

Mass Spectrometer LECO Pegasus BT

Ion Source Temperature 250 °CMass Range 30-500 m/z

Acquisition Rate 20 spectra/s

Delive

ring t

he R

ight

Resu

lts

Figure 2. The figure above shows GC-MS total ion chromatograms (TIC) for a good-flavored jelly bean (Top, Fruity) vs. a bad-flavoredone (Bottom, Smelly Socks) along with Peak True (deconvoluted) and NIST library spectra for a key flavor contributing compound fromeach jelly bean. The table contains additional flavor contributing compounds and their flavor descriptors.

Figure 3. The figure above shows GC-MS total ion chromatograms (TIC) for a good-flavored jelly bean (Top, Popcorn) vs. a bad-flavored one (Bottom, Spoiled Eggs) along with Peak True (deconvoluted) and NIST library spectra for a key flavor contributingcompound from each jelly bean. The table contains additional flavor contributing compounds and their flavor descriptors.

2.5 5 7.5 10 12.5 15 17.5 20

0

0.3e9

0.6e9

0.9e9

1.2e9

1.5e9

1.8e9

2.1e9

2.4e9

2.7e9

Time (min)

Fruity Smelly SocksCompound Name & Descriptors Compound Name & Descriptors

isobutyl acetate (sweet, fruity, tropical) hexanoic acid (sour, fatty, sweat, cheesy)

ethyl butyrate (fruity, pineapple, sweet) octanoic acid (rancid, cheesy, oily)

3-methyl-1-butanol-acetate (sweet, fruity, banana) isovaleric acid (sour, stinky feet, sweaty, cheese)

d-limonene (sweet, orange, citrus) 2-methyl-pentanoic acid (sour, cheese)

linalool (citrus, floral, sweet, orange, lemon)

(E)-citral (citrus, lemon)

40 60 80 100 120 140

0

0.2e8

0.4e8

0.6e8

0.8e8

1.0e8

0

100

200

300

400

500

600

700

800

900

1000

M/Z

43.07

Peak True - sample "HS SPME Tutti Fruitti", Ethyl butyrate, at 6.28611 min, Area (Counts)

Library Hit - Similarity: 956 - Library: mainlib - Ethyl butyrate, Abundance

40 60 80 100 120 140

0

1e7

2e7

3e7

4e7

5e7

0

100

200

300

400

500

600

700

800

900

1000

M/Z

60.01

Peak True - sample "HS SPME Stinky Socks", Isovaleric acid, at 12.3157 min, Area (Counts)

Library Hit - Similarity: 885 - Library: replib - Isovaleric acid, Abundance

Fruity

Smelly Socks

2.5 5 7.5 10 12.5 15 17.5 20

0

0.3e9

0.6e9

0.9e9

1.2e9

1.5e9

1.8e9

2.1e9

Time (min)

Popcorn Spoiled EggsCompound Name & Descriptors Compound Name & Descriptors

2,3-pentanedione (toasted, buttery, creamy, caramellic) 2-methyl-benzenethiol (meaty, sulferous, egg)

Butyl butyrolactate (creamy, buttery, buttermilk) dimethyl sulfide (sulfery, onion, cabbage)

2-acetyl pyridine (popcorn, heavy corn) dimethyl disulfide (sulferous cabbage, malt, cream)

gamma decalactone (coconut, buttery, sweet) s-(methylthio) butyrate (sulfury, cheese, putrid, musty, limberger-type cheese)

40 60 80 100 120 140

0

0.2e7

0.4e7

0.6e7

0.8e7

1.0e7

1.2e7

1.4e7

1.6e7

0

100

200

300

400

500

600

700

800

900

1000

M/Z

79.02

Peak True - sample "HS SPME Butter Popcorn", 2-Acetylpyridine, at 11.9238 min, Area (Counts)

Library Hit - Similarity: 918 - Library: replib - 2-Acetylpyridine, Abundance

40 60 80 100 120 140

0

1e7

2e7

3e7

4e7

0

100

200

300

400

500

600

700

800

900

1000

M/Z

46.99

Peak True - sample "HS SPME Rotten Eggs", Dimethyl sulfide, at 1.98721 min, Area (Counts)

Library Hit - Similarity: 977 - Library: replib - Dimethyl sulfide, Abundance

Popcorn

Spoiled Eggs

Life

Sci

ence

and C

hem

ical A

naly

sis

Solu

tionsFigure 4. The figure above shows GC-MS total ion chromatograms (TIC) for a good-flavored jelly bean (Top, Peach) vs. a bad-flavored

one (Bottom, Vomit) along with Peak True (deconvoluted) and NIST library spectra for a key flavor contributing compound from eachjelly bean. The table contains additional flavor contributing compounds and their flavor descriptors.

Figure 5. The figure above shows GC-MS total ion chromatograms (TIC) for a good-flavored jelly bean (Top, Blueberry) vs. a bad-flavored one (Bottom, Minty Toothpaste) along with Peak True (deconvoluted) and NIST library spectra for a key flavor contributingcompound from each jelly bean. The table contains additional flavor contributing compounds and their flavor descriptors.

2.5 5 7.5 10 12.5 15 17.5 20

0

0.3e9

0.6e9

0.9e9

1.2e9

1.5e9

1.8e9

Time (min)

Peach VomitCompound Name & Descriptors Compound Name & Descriptors

isoamyl butyrate (fruity, green, apricot, pear, green apple) 2-nonanone (cheesy, green, fruity, dairy, dirty)

hexyl acetate (fruity, fresh, sweet banana) diallyl sulfide (sulfurous, onion-garlic, horseradish)

geranyl isobutyrate (peach, sweet, floral) 2-undecanone (creamy cheese like notes)

gamma-Dodecalactone (peach, sweet fruity) hexanoic acid (sour, fatty, sweat, cheesy)

gamma-decalactone (fruity, creamy, peach) octanoic acid (rancid, soapy, cheesy, fatty)

carvacrol (spicy, herbal, medicinal)

40 60 80 100 120 140 160 180 200

0

0.2e8

0.4e8

0.6e8

0.8e8

1.0e8

1.2e8

1.4e8

1.6e8

1.8e8

0

100

200

300

400

500

600

700

800

900

1000

M/Z

85.01

Peak True - sample "HS SPME Peach", ç-Decalactone, at 14.9241 min, Area (Counts)

Library Hit - Similarity: 977 - Library: replib - ç-Decalactone, Abundance

40 60 80 100 120 140

0

1e6

2e6

0

100

200

300

400

500

600

700

800

900

1000

M/Z

44.98

Peak True - sample "HS SPME Barf", Diallyl sulfide, at 7.99876 min, Area (Counts)

Library Hit - Similarity: 916 - Library: replib - Diallyl sulfide, Abundance

Peach

Vomit

2.5 5 7.5 10 12.5 15 17.5 20

0

0.3e9

0.6e9

0.9e9

1.2e9

1.5e9

1.8e9

2.1e9

2.4e9

2.7e9

Time (min)

Blueberry Minty ToothpasteCompound Name & Descriptors Compound Name & Descriptors

Linalool (citrus, orange, sweet) menthol (cooling, minty)

ethyl nonanoate (fruity and tropical) carvone (minty, spearmint)

dimethylbenzylcarbinyl acetate (sweet, fruity, berry, jasmin) methyl-salicylate (minty, wintergreen)

ethyl butyrate (fruity, pineapple) cinnamaldehyde (cinnamon, red hots, spicy)

ethyl isovalerate (sweet, fruity, apple, pineapple)

ethyl hexanoate (sweet, pineapple, banana)

50 100 150 200 250 300 350 400 450 500

0

0.1e8

0.2e8

0.3e8

0.4e8

0.5e8

0.6e8

0.7e8

0.8e8

0.9e8

1.0e8

0

100

200

300

400

500

600

700

800

900

1000

M/Z

43.02

Peak True - sample "HS SPME Berry Blue", Dimethylbenzylcarbinyl acetate, at 12.8965 min, Area (Counts)

Library Hit - Similarity: 923 - Library: mainlib - Dimethylbenzylcarbinyl acetate, Abundance

60 80 100 120 140 160 180

0

1e7

2e7

3e7

4e7

5e7

6e7

0

100

200

300

400

500

600

700

800

900

1000

M/Z

81.99

Peak True - sample "HS SPME Tooth Paste", (-)-Carvone, at 12.7556 min, Area (Counts)

Library Hit - Similarity: 952 - Library: replib - (-)-Carvone, Abundance

Blueberry

Minty Toothpaste

Delive

ring t

he R

ight

Resu

lts

Figure 6. The figure above shows GC-MS total ion chromatograms (TIC) for a good-flavored jelly bean (Top, Chocolate) vs. a bad-flavored one (Bottom, Dog Food) along with Peak True (deconvoluted) and NIST library spectra for a key flavor contributingcompound from each jelly bean. The table contains additional flavor contributing compounds and their flavor descriptors.

LECO Corporation | 3000 Lakeview Avenue | St. Joseph, 49085 | Phone: 800-292-6141 | 269-985-5496MIinfo@leco.com • www.leco.com | -9001:2008 | -Q-994 |ISO HQ LECO LECOis a registered trademark of Corporation.

LECO LECO, Pegasus, and ChromaTOF are trademarks of Corporation.

Form No. 203-821-529 10/16- 0 © 2016 CorporationREV LECO

4. Conclusion

With the use of the BT data, the children were able to predict the appealing and appalling flavored jelly beansPegasusbased on the taste profile for the analytes which were identified by GC-MS in each of the jelly bean varieties.

The kids found the experiment exciting and were exposed to analytical chemistry in a fun, safe, and educational way.

2.5 5 7.5 10 12.5 15 17.5 20

0

1.2e8

2.4e8

3.6e8

4.8e8

6.0e8

7.2e8

8.4e8

Time (min)

Chocolate Dog FoodCompound Name & Descriptors Compound Name & Descriptors

3-methlbutanal (chocolate, peach, fatty) 1-octen-3-ol (mushroom, earthy, raw chicken, oily, fungal)

vanillin (creamy chocolate, sweet vanilla) 2-acetylthiazole (cornchip with musty background, popcorn, roasted peanuts)

tetramethyl pyrazine (musty, nutty, chocolate coffee, cocoa) 2,4-decadienal (fatty, oily, chicken, fried, slight rancid tallow)

triacetin (creamy, with an oily mouth feel)

bis(2-methyl-3-furyl)disulfide (strong meaty, brothy, roasted savory, cooked onion, garlic and black pepper)

40 60 80 100 120 140 160 180

0

1e6

2e6

3e6

4e6

5e6

0

100

200

300

400

500

600

700

800

900

1000

M/Z

54.04

Peak True - sample "HS SPME Chocolate Pudding", Pyrazine, tetramethyl-, at 10.9723 min, Area (Counts)

Library Hit - Similarity: 945 - Library: replib - Pyrazine, tetramethyl-, Abundance

50 75 100 125 150 175 200 225

0

1e6

2e6

3e6

4e6

5e6

6e6

7e6

8e6

0

100

200

300

400

500

600

700

800

900

1000

M/Z

112.99

113

Peak True - sample "HS SPME Dog Food", bis(2-methyl-3-furyl) disulfide, at 14.9571 min, Area (Counts)

Library Hit - Similarity: 893 - Library: mainlib - Furan, 3,3'-dithiobis[2-methyl-, Abundance

Chocolate

Dog Food