Beer is a popular beverage produced by the fermentation of hopped malt extracted from barley and other grains. Some compounds (flavors) have a positive effect on aroma (attributes) and some have a negative effect (defects). This presentation will focus on a new method that enables the investigation and characterization of flavors and defects of beer in one analysis using HS trap/GC/MS. Classically, this analysis is performed on four separate detectors. This new method employs one detector (MS) to provide these solutions required for the production and the testing of beer. The outcome is a more cost effective, accurate means to ensure the validity and the quality control of their product. Other benefits include enhanced productivity, attaining more information from a single analysis, and requiring less bench space. The following experiments and results will be discussed. • Quantitation of dimethyl sulfide (DMS), 2,3-butanedione (diacetyl), 2,3-pentandione and t,2-nonenal • Characterization of several types of beers • Fermentation profiling • Analysis of raw materials • Aging studies Originally presented at Pittcon 2012.

1. Determining Flavors and
“Defects” in Beer by
Headspace Trap/Gas
Chromatography/Mass
Spectrometry
Andrew Tipler, Chromatography R&D Manager
Lee Marotta, Field Application Scientist
1 © 2009 PerkinElmer
2012

2. Content
This presentation will describe a system that provides an almost comprehensive
analysis of flavor compounds and defects in beer
 Theory
 Design and operation
 Beer Application
The system comprises the following components:
 A headspace trap sampling system
 A gas chromatograph
 A mass spectrometer
2

3. Replacing Multiple Systems with One …
Ethanol - FID
Defects
 Vicinal Diketones (‘butterscotch’) - ECD
 Dimethyl Sulfide (sulfury character) – FPD or SCD
 Aldehydes (oxidation ‘cardboard’ products) - FID
 Thiols (skunkiness) - SCD
Flavor Compounds – MS (characterization)
 Alcohols
 Ketones
 Esters
 Acids
 Turpenes
Fermentation Markers
 Diacetyl
3

4. Acknowledgement
The authors would like to thank Bill Yawney of the LongTrail Brewery, Vermont, for
his advice, inspiration and some of the analytical data used in this presentation
4

5. Headspace Instrumentation
– Theory, Design and
Operating Principles
5 © 2009 PerkinElmer

6. Principles Behind Headspace Sampling
6

7. Tasting Beer
Besides actually making beer, much of the fun
is associated with drinking it!
It’s nutritious, it makes the world easier to live
in and it tastes good.
Taste is obviously subjective but we beer
connoisseurs generally consider the following
when drinking a fine beer:
 Don’t drink out of the bottle
 Don’t cool the beer to Arctic temperatures
 Use an appropriately shaped glass
 Don’t fill the glass completely
7

8. Tasting Beer
Besides actually making beer (of course),
much of the fun is associated with drinking it!
It’s nutritious, it makes the world easier to live
in and it tastes good.
Taste is obviously subjective, but beer
connoisseurs will generally consider the
following when drinking a fine beer:
 Don’t drink out of the bottle
 Don’t cool the beer to Arctic temperatures
 Use an appropriately shaped glass
 Don’t fill the glass completely
 These are all done to ensure that the beer
aroma is involved in the tasting process
(beer aroma, or „nose‟ as it‟s called, is an
important part of the formal beer-judging
process)
8

9. Headspace Sampling
Headspace sampling is a bit like smelling the
aroma
Step 1 – put beer sample into a vial and seal it
Step 2 – heat the vial for a period of time at a
constant temperature
Step 3 – extract some of the vapor and
analyze it by gas chromatography
9

10. Theory
During the equilibration step, molecules
distribute themselves according to their
partition coefficients
K=Cl/Cv
Molecules with low partition coefficients
favor the vapor (headspace) phase
whereas molecules with high partition
coefficients favor the liquid (sample) Compound
phase
Partition coefficients are reduced as the
temperature is increased Liquid Sample
At equilibrium, the concentration in the
headspace phase is proportional to the
original concentration in the sample
Determining the composition of the
headspace phase enables the
composition of the sample to be
established.
10

11. Key Components in the PerkinElmer Headspace System
11

12. Enhanced Sensitivity with the Headspace Trap
Since polar compounds in water (or beer)
have very high partition coefficients – often
less than 0.5% of the compound in the
sample may pass into the headspace.
With headspace without the trap, only a
small fraction of the total headspace vapor
will enter the column
The headspace trap technique can
enhance detection limits by 100 times by
withdrawing the entire HS volume and
enabling several injections from same vial
to be focused on trap
12

13. Sample Vial Thermal Equilibration
valve detector
seal trap column
vial
oven
Headspace Sampler Gas Chromatograph
13

14. Vial Pressurization
column isolation
valve detector
seal trap column
vial
oven
Headspace Sampler Gas Chromatograph
14

15. Trap Load
valve detector
seal trap column
vial
oven
Headspace Sampler Gas Chromatograph
15

16. Vial Re-Pressurization
valve detector
seal trap column
vial
oven
Headspace Sampler Gas Chromatograph
16

17. Trap Re-Load
valve detector
seal trap column
vial
oven
Headspace Sampler Gas Chromatograph
17

18. Dry Purge
valve detector
seal trap column
vial
oven
Headspace Sampler Gas Chromatograph
18

19. Trap Desorption
valve detector
seal trap column
vial
Trap desorbed in
Opposite direction
oven
Headspace Sampler Gas Chromatograph
19

20. Analyzing Beer
20 © 2009 PerkinElmer

21. Beer - Component Identification by Mass Spectrometry
Sample Size: 5 mL
Beer Flavors HS_GC_MS Scan EI+
16.59;70 Sample Temp: 70oC TIC
100
Sample Load: 1 cycle 2.95e10
Trap Load Temp: 25oC
Nice peak shapes
Isoamyl alcohol
Dry Purge: 6 min
Trap high Temp: 300oC
Good peak separation Needle Temp:
T Line Temp:
160oC
180oC
Required detection limits Column Flow:
Pressure Pulse: 2mL/min for 0.4min
Analytical Flow Rate 1mL/min
Repeatable response
Mass Range: 30 to 300 amu
Linear response
16.76 22.31
55 70
%
6.04
46 11.43 12.02
61 43 30.33
88
26.01
22.42 88
43
Propyl acetate
18.17 31.54
10.15 19.26 33.77
43 104
43 43 88
0 Time
6.50 8.50 10.50 12.50 14.50 16.50 18.50 20.50 22.50 24.50 26.50 28.50 30.50 32.50 34.50
21

22. Sensitivity is AMAZING by HS Trap/GC/MS
DMS at 10 ppb
Signal to Noise is 56940 to 1
Diacetyl at 10 ppb
Signal to Noise is 1067 to 1
22

23. Data of Beer Analysis from Long Trail Brewery
Thank you Long Trail and Bill Yawney
Hops Evaluation
Beer Comparisons
Supplier Information
Brewing Investigation
23 © 2009 PerkinElmer

24. Hop Volatile Comparison
Hop Volatile Compound Comparison
Average Amounts - Sample 1 vs Sample 2
Compound #
12
Compound Name (Longer
names have been 10
truncated)
1 1R-à-Pinene 8
2 β-Myrcene
3 β-Pinene
ppm
6
4 à-Phellandrene
5 2,6-Dimethyl-1,3,5,7-octatetra
4
6 Limonene
7 1,4-Cyclohexadiene, 1-methyl-4
2
8 Furan, 3-(4-methyl-3-pentenyl)
9 Copaene
10 1,6,10-Dodecatriene, 7,11-dime 0
1 2 3 4 5 6 7 8 9 10 11 12
11 Bicyclo[3.1.1]hept-2-ene, 2,6- Sample 1 Volatile Compounds
12 Caryophyllene Sample 2
24

25. Beer Volatile Comparison
Compound
Retention
Name
Time
1-Propanol 8.02
2-Butanone, 4-hydroxy 9.72
1-Propanol, 2-methyl 10.45
1-Butanol, 3-methyl 12.54
1-Butanol, 2-methyl 12.80
Propanoic acid ethyl ester 13.63
n-Propyl acetate 13.74
Mixture of methyl butanols 14.77
Mixture of methyl butanols 14.93
Acetic acid, 2-methylpropyl ester 16.28
Butanoic acid, ethyl ester 17.37
1-Butanol, 3-methyl-, acetate 20.42
1-Butanol, 2-methyl-, acetate 20.55
Hexanoic acid, ethyl ester 24.49
Acetic acid hexyl ester 24.84
Heptanoic acid, ethyl ester 26.93
Acetic acid, heptyl ester 27.22
Phenyl ethyl alcohol 27.45
Octanoic acid 28.15
Octanoic acid, ethyl ester 28.95
Acetic acid, 2-phenylethyl ester 30.12
Ethyl9-decanoate 32.11
Decanoic acid ethyl ester 32.27
Caryophyllene 33.93
Alpha caryophyllene 34.55
Decanoic acid ethyl ester 35.86
25

26. Research and Development Project
Name RT
Dimethyl sulfide 8.323
1- Propanol 9.358
1- Propanol 9.591
Acetic acid, anhydride with formic acid 10.131
1-Propanol, 2-methyl 12.104
Butanol, 3-methyl 13.004
Formic acid 13.987
Vinyl butyrate 14.437
1-Butanol, 3-methyl 14.700
1-Butanol, 2-methyl-, (S) 14.962
Acetaldehyde, O-methyloxime 15.487
Vinyl butyrate 17.528
Butanoic acid, ethyl ester 19.178
Butanoic acid, 3-methyl-,ethyl ester 21.406
Cyclobutanone, 2,2,3-trimethyl- 21.744
Cyclopentane 21.872
Propanoic acid, 2-methyl-,2-methylpropyl ester 23.642
Pentyl glycolate 25.938
β-Myrcene 26.133
Propanoic acid, 2-methyl-,2-methylbutyl ester 26.530
Limonene 27.446
Methyl 2-methyl hexanoate 28.286
Butanoic acid, 2-methyl-,2-methylbutyl ester 28.631
Decanoic acid, ethyl ester 28.924
Octanoic acid, ethtyl ester 30.251
Benzenebutanal 31.452
2,6-Octadienoic acid, 3,7-dimethyl-, metyl ester, (E) 32.525
Decanoic acid, ethyl ester 33.717
26

27. Supplier Comparison
27

28. American Pale Ale
12mL sample taken approximately every
eight hours starting from time zero (prior to
adding the yeast) during the brewing
process
PerkinElmer Fermentation
Experiment
Testing the process using
HS/GC/MS
28 © 2009 PerkinElmer

29. The ‘Profile’ Beer: American Pale Ale
Grains
 Maris Otter Pale Malt
 Munich Malt
 Crystal Malt
Hops
 Chinook
 Centennial
 Amarillo
 Nelson Sauvin
Yeast
 SafAle American Ale 05 dry yeast, no starter
O.G.
 1.058
IBU
 45
Process
 Single infusion mash at 67°C
 Fermentation at 19-20°C
29

30. Results of Components Changing with Time
30

31. Activity of Two Components over 111 Hours of Sampling
Dimethyl Sulfide (DMS) 2,3-Butanedione (Diacetyl)
Plot: Detector Response –vs- Time
Time Interval: Every Eight Hours
31

32. Interesting Component … 3-Methyl-1-Butanol
60000000
50000000
40000000
30000000
20000000
10000000
0
0 20 40 60 80 100 120
-1E+08
Plot: Detector Response –vs- Time
Time Interval: Every Eight Hours
32

33. Concentration of Four Components through Fermentation
Time DMS Diacetyl 2,3-Pentanedione trans 2-Nonenal
(hours) (PPB) (PPB) (PPB) (PPB)
Zero 16.5 81 15 4.7
7 18.8 1577 16 3.9
15 4.6 2779 37 3.8
23 3.0 2183 52 3.9
31 3.2 1658 59 3.8
39 3.0 862 43 3.9
47 4.0 715 85 4.0
53 3.6 422 93 3.9
66.5 3.3 249 41 3.9
70.75 3.7 341 129 4.0
79 3.4 109 72 3.9
87 3.4 95 91 3.9
95.5 3.6 74 65 3.9
103 3.3 44 49 3.9
111 3.2 48 64 4.0
33

34. Specific Gravity Measurements over 95 hours
Profile Beer Specific Gravity
Time, hrs S.G Attn %
0.0 1.058 0.0
7.0 1.058 0.0
15.0 1.054 6.9
23.0 1.052 10.3
31.0 1.050 13.8
39.0 1.042 27.6
47.0 1.035 39.7
55.0 1.030 48.3
64.5 1.025 56.9
70.8 1.023 60.3
79.0 1.020 65.5
87.5 1.016 72.4
95.0 1.014 75.9
34

35. Conclusion
Quantifies defects
Characterizes beer by
 Providing component identity – WHAT IS IT? Qualitative information
 Providing relative component ratio information
 Providing concentration (quantitative) information – How much?
Enables the investigation of other beers – What makes my neighbors
beer so very delicious?
Optimizes process control
35

36. lee.marotta@perkinelmer.com
andrew.tipler@perkinelmer.com
36