Cynthia Elmore and J. Garrett SlatonOI Analytical, P.O. Box 9010, College Station, TX 77845-9010 • (800) 653-1711 • www.oico.com

Volatile Organic Compounds in Energy Drinks asDetermined by GC-MS with Purge and Trap Sample Concentration

Introduction

Experimental

The use of purge and trap for foaming liquid samples has been demonstrated for a range of somewhat difficult matrices. The samples were purged in the vials, rather than in a fritted sparge vessel and usable results for samples typically analyzed by headspace were obtained.

Trace levels of chloroform, attributed to drinking water disinfectant byproducts, were found in several samples. Benzene was found in a few of the drinks containing benzoate preservatives.

The presence of other compounds, such as p-cymene, limonene, and butanoate esters is also to be expected based on their use as flavorings. Ethanol is detected in nearly all of the energy drinks to an appreciable level, but its source cannot be determined from the data.

Summary and Conclusions

References1. “The American Energy Drink Craze in Two Caffeinated Charts”, Roberto A.

Ferdman, Quartz, March 26, 2014.2. “Energy Drinks: Busting Your Health for the Buzz”, Pennsylvania Medical

Society, September 19, 2008.3. USEPA Method 8260B Volatile Organic Compounds by Gas

Chromatography/Mass Spectrometry (GC/MS) Revision 2 December 1996.4. ICBA Guidance Document to Mitigate the Potential for Benzene Formation

in Beverages, International Council of Beverages Associations, 2006.5. Data on Benzene in Soft Drinks and Other Beverages, U.S. FDA, 2007.

Chloroform and benzene are compounds on the NPDWRs list. Chloroform was found in several of the samples at reportable levels (>2 ppb), still well below the United States Environmental Protection Agency (USEPA) maximum contaminant level (MCL). Chloroform is known to develop in drinking water as a disinfectant byproduct. Flavoring agent p-cymene, used as part of the calibration but not on the list, was also detected above the reporting limit.

Benzene is carcinogenic at high concentrations and is monitored in drinking water analysis. It may form when benzoate salts and ascorbic acid are present. Various factors foster the formation, such as exposure to heat and light, while sugar and EDTA can inhibit it4. In 2005, the United States Food and Drug Administration (FDA) received private laboratory results showing low levels of benzene in some soft drinks that contained benzoate salts and ascorbic acid. Because the FDA has no standard for benzene in beverages other than bottled water, the NPDWR limit of 5 ppb was adopted as a standard5. Benzene was detected in two evaluated drinks at trace levels (2.17 and 1.85 ppb). The drinks contained benzoate salts, but no ascorbic acid.

All samples contained varying amounts of other compounds (TICs) identified as natural and artificial flavorings frequently found in fruit-flavored beverages. Most drinks also contained appreciable (>200 ppb) ethanol, perhaps left over as an aid to dissolve essential oils or from decomposition of ethyl esters.

Results and DiscussionEnergy drinks and shots are approximately a $30 billion dollar industry with major beverage companies, such as PepsiCo, and focused energy drink manufacturers, like Red Bull GmbH, participating1. A 2014 World Health Organization report estimated 30% of adults, 68% of adolescents, and 18% of children under the age of 10 consume energy drinks2.

Prior to public outcry in the wake of the discovery of benzene in soft drinks in the early 1990s, prepared beverages were not regulated. They are still not broadly held to the National Primary Drinking Water Regulations, though the benzene content is now regulated to 5 ppb. Beverages can be foamy when analyzed by purge and trap methodology, and thus, are traditionally analyzed by headspace coupled to a GC-MS. We have analyzed a selection of these challenging samples using GC-MS with a new purge and trap sample concentrator and present both quantitative and some qualitative data on identified VOCs both on and off the National Primary Drinking Water Regulations (NPDWRs) list with good results.

Figure 1. 4760 Purge & Trap with 4100 Sample Processor

Figure 2. Energy drinks evaluated in the study

Figure 3. 50 ppb Calibration Standard

Figure 4. Energy Drink 1

Eth

ano

l 50

0 p

pb

SS1

& 1

S1SS

2 IS2 SS

3B

uta

no

ate

este

r 1

60

pp

bB

uta

no

ate

este

r 1

80

pp

bB

uta

no

ate

este

r 9

00

pp

b

IS4

IS3 O

rgan

ic a

cid

65

pp

bIS

4

Bu

tan

oat

e es

ter

23

0 p

pb

Bu

tan

oat

e es

ter

62

0 p

pb

Figure 5. Energy Drink 2

Eth

ano

l 12

60

pp

b

SS1

& 1

S1SS

2IS

2

SS3

Bu

tan

oat

e es

ter

12

50

0 p

pb

3-m

eth

yl-1

1-B

uta

no

l 36

90

pp

b

SS4

Hex

ano

ate

eth

yl e

ster

39

0 p

pb

Bu

tan

oat

e es

ter

14

20

0 p

pb

Bu

tan

oat

e es

ter

26

40

pp

b

Hex

ano

ate

este

r 2

96

0 p

pb

Pen

tan

oat

e es

ter

10

30

pp

b

Figure 6. Energy Drink 3

Eth

ano

l 24

0 p

pb

SS1

& 1

S1SS

2IS

2

SS3 .b

eta.

– P

hel

lan

dre

ne

40

0 p

pb

SS4

IS3

Ben

zald

ehyd

e 1

40

pp

b

Unknown alcohol 170 ppb

Lin

no

nen

e 2

67

00

pp

bg

amm

a Te

rpin

ene

21

40

pp

bTe

rpin

ole

ne

80

0 p

pb

Table 1. Method Parameters

Purge-and-Trap Eclipse 4760 P&T Sample Concentrator

Trap #10 trap; Tenax® / Silica gel / CMS

Purge Gas Zero grade Helium at 40 mL/min

Purge Time 11 min

Sparge Mount Temp. 45 °C soil

Sample Temp. 45 °C soil

Desorb Time 0.5 min

Bake Time 5 min

OI #10 Trap Temp. Ambient during purge180 °C during desorb pre-heat190 °C during desorb210 °C during bake

Water Management 120 °C during purgeAmbient during desorb240 °C during bake

Transfer Line Temp. 140 °C

Six-port Valve Temp. 140 °C

Gas Chromatograph Agilent 7890A

Column Restek Rxi-624 Sil MS 30 meter, 0.25 mm, 1.4 µm film

Carrier Gas Zero grade helium

Inlet Temp. 250 °C

Inlet Liner Agilent Ultra Inert 1 mm straight, taper

Column Flow Rate 0.8 mL/min

Split Ratio 150:1

Oven Program Hold at 40 °C for 1.5 min16 °C/minute to 180 °C 40 °C/minute to 220 °CHold at 220 °C for 2.0 minTotal GC Run is 13.25 min

Mass Spectrometer Agilent 5975C

Mode Scan 35 - 300 amu

Scans/Second 5.19

Solvent Delay 1.40 min

Transfer Line Temp. 250 °C

Source Temp. 300 °C

Quadrupole Temp. 200 °C

Draw-out Plate 6 mm

Table 2. Calibration DataAnalyte Compound AVG % RSD1 pentafluorobenzene (IS)

2 dichlorodifluoromethane 0.154 7.19

3 chloromethane 0.197 6.43

4 vinyl chloride* 0.251 5.40

5 bromomethane 0.248 9.68

6 chloroethane 0.145 8.16

7 trichlorofluoromethane 0.515 5.21

8 ethyl ether 0.158 4.22

9 1,1-dichloroethene* 0.398 6.06

10 carbon disulfide 1.121 5.24

11 1,1,2-trichloro-1,2,2-trifluoroethane 0.409 5.61

12 methyl iodide 0.826 5.22

13 allyl chloride 0.187 2.10

14 methylene chloride* 0.420 7.19

15 acetone 0.026 9.93

16 trans-1,2-dichloroethene* 0.460 9.31

17 methyl tert-butyl ether 0.784 3.34

18 chloroprene 0.666 1.93

19 1,1-dichloroethane 0.743 3.41

20 acrylonitrile 0.115 7.14

21 cis-1,2-dichloroethene* 0.510 3.00

22 2,2-dichloropropane 0.329 6.24

23 bromochloromethane 0.250 2.78

24 chloroform* 0.847 2.78

25 methyl acrylate 0.300 4.67

26 carbon tetrachloride* 0.576 2.38

27 dibromofluoromethane (SS) 0.528 1.82

28 1,1,1-trichloroethane* 0.623 3.13

29 2-butanone 0.029 5.01

30 1,1-dichloropropene 0.570 3.99

31 1,4-difluorobenzene (IS)

32 benzene* 1.045 8.69

33 methacrylonitrile 0.113 4.17

34 1,2-dichloroethane-d4 (SS) 0.043 1.57

35 1,2-dichloroethane* 0.374 6.83

36 trichloroethene* 0.310 3.74

37 dibromomethane 0.195 2.68

38 bromodichloromethane* 0.358 3.82

39 1,2-dichloropropane* 0.263 2.33

40 methyl methacrylate 0.118 6.88

41 2-chloroethyl vinyl ether 0.073 13.29

42 cis-1,3-dichloropropene 0.350 8.09

43 chlorobenzene-d5 (IS)

Analyte Compound AVG % RSD44 toluene-d8(ss) 1.301 1.72

45 toluene* 0.790 3.81

46 4-methyl-2-pentanone 0.021 4.12

47 tetrachloroethene* 0.329 5.37

48 trans-1,3-dichloropropene 0.346 9.58

49 ethyl methacrylate 0.262 6.78

50 1,1,2-trichloroethane* 0.237 3.12

51 chlorodibromomethane* 0.313 9.24

52 1,3-dichloropropane 0.335 2.28

53 1,2-dibromoethane* 0.305 3.12

54 2-hexanone 0.154 5.73

55 chlorobenzene* 0.945 3.95

56 ethylbenzene* 1.496 4.11

57 1,1,1,2-tetrachloroethane 0.276 1.99

58 m,p-xylenes* 0.588 4.86

59 o-xylene* 0.575 4.39

60 styrene* 0.956 3.97

61 bromoform* 0.181 9.29

62 isopropylbenzene 1.446 4.59

63 1,4-dichlorobenzene-d4 (IS)

64 4-bromofluorobenzene (SS) 1.088 1.90

65 bromobenzene 0.880 4.57

66 n-propylbenzene 3.822 2.51

67 1,1,2,2-tertrachloroethane 0.687 4.20

68 2-chlorotoluene 2.173 3.12

69 1,3,5-trimethylbenzene 2.669 1.84

70 1,2,3-trichloropropane 0.656 6.60

71 4-chlorotoluene 2.545 3.50

72 tert-butylbenzene 2.520 1.84

73 pentachloroethane 0.382 6.77

74 1,2,4-trimethylbenzene 2.682 2.11

75 sec-butylbenzene 3.346 1.81

76 p-isopropytoluene 2.635 2.95

77 1,3-dichlorobenzene 1.478 4.77

78 1,4-dichlorobenzene* 1.449 4.65

79 n-butylbenzene 2.536 2.84

80 1,2-dichlorobenzene* 1.339 4.00

81 1,2-dibromo-3-chloropropane* 0.090 12.86

82 hexachlorobutadiene 0.574 2.68

83 1,2,4-trichlorobenzene* 0.943 4.35

84 naphthalene 1.868 9.34

85 1,2,3-trichlorobenzene 0.858 5.72

* National Primary Drinking Water Regulated Compounds (NPDWR)

Sample concentration and introduction was done using an OI Analytical 4760 Purge and Trap and a 4100 Sample Processor, while an Agilent 5975C Mass Spectrometer and 7890A Gas Chromatograph were used for separation and detection. See Figure 1 for a photograph of the equipment and Table 1 for the method parameters. Twenty samples were analyzed from top selling brands in the U.S. (Figure 2) Due to the potential for excessive foaming, samples were run in the “soil mode” of the 4100 where the sample is purged in a 40-mL vial rather than in a frit sparger.

A multi-point calibration was run, which included the regulated compounds from the NPDWRs list with concentrations ranging from 2 ppb to 200 ppb. Calibrations were generated using USEPA Method 82603. (Table 2) All method criteria were met. Samples were quantitated using this calibration and then a library search was performed on peaks not identified by the calibration method, i.e., tentatively identified compounds (TICs).

VOCs-in-Energy-Drinks-by-GC-MS-purge-and-trap_r3.indd 1 2/25/16 5:22 PM