Understanding the impact hopping rate has on the aroma quality and intensity of beer dry hopped with Cascade Scott Lafontaine Ph. D. Advisor: Tom Shellhammer Oregon State University 36th European Brewery Convention May 16 th, 217
Rationale for this study *Source: Brewers Association and IRI
Rationale for this study US Craft Hop usage (g/hl) 6 567 548 5 497 55 462 435 4 361 369 3 2 1 28 29 21 211 212 213 214 215 Year *Adapted from Brewer s Association
Rationale for this study US Craft Hop usage (g/hl) 6 567 548 5 497 55 462 57% increase! 435 4 361 369 3 2 1 28 29 21 211 212 213 214 215 Year *Adapted from Brewer s Association
Rationale for this study US Craft Hop usage (g/hl) 6 567 548 5 497 55 462 435 4 361 369 3 2 1 78 Typical kettle hopped rate for American light lager 28 29 21 211 212 213 214 215 Year *Adapted from Brewer s Association
Rationale for this study US Craft Hop usage (g/hl) 8 75 7 Very high craft hopping rate 6 5 4 3 2 1 28 29 21 211 212 213 214 215 Year *Adapted from Brewer s Association
Main objective The main goal of this project was to determine a dose response curve for Cascade hop aroma.
Main objective The main goal of this project was to determine a dose response curve for Cascade hop aroma. Does more hop material = more aroma?
Background 214 Harvest Region OSU Hop Oil (ml/1g) CAS_1_15 WA.6 CAS_18_15 WA.7 CAS_2_15 OR.7 CAS_11_15 WA.9 CAS_2_15 WA 1. CAS_6_15 WA 1. CAS_12_15 OR 1. CAS_5_15 OR 1.1 CAS_21_15 WA 1.2 CAS_14_15 WA 1.2 CAS_9_15 OR 1.3 CAS_3_15 OR 1.4 CAS_1_15 WA 1.5 CAS_4_15 OR 1.7 CAS_7_15 WA 1.7 CAS_13_15 WA 1.7 CAS_15_15 WA 1.7 CAS_16_15 WA 1.7 CAS_24_15 WA 1.8 CAS_8_15 OR 1.8 CAS_17_15 WA 1.9 CAS_22_15 WA 2. 215 Harvest Region OSU Hop Oil (ml/1g) CAS_12_16 WA.5 CAS_27_16 WA.6 CAS_21_16 ID.6 CAS_22_16 ID.6 CAS_24_16 WA.6 CAS_7_16 ID.7 CAS_9_16 ID.8 CAS_19_16 WA.8 CAS_4_16 WA.8 CAS_25_16 OR.8 CAS_26_16 WA.9 CAS_6_16 ID.9 CAS_5_16 WA 1. CAS_11_16 WA 1. CAS_16_16 WA 1.1 CAS_17_16 OR 1.1 CAS_15_16 WA 1.2 CAS_3_16 OR 1.2 CAS_23_16 WA 1.2 CAS_2_16 WA 1.3 CAS_28_16 WA 1.4 CAS_29_16 WA 1.4 CAS_2_16 OR 1.4 CAS_8_16 OR 1.5 CAS_13_16 WA 1.5 CAS_1_16 WA 1.5 CAS_1_16 OR 1.7 CAS_18_16 WA 1.7 CAS_14_16 WA 2.6 Thesis Project: Examining the raw material factors that describe the dry hop aroma performance of hops in beer Multi-year study: Dry hopped different lots of the same cultivar 22Sample lots in 214 Work performed by Dan Vollmer 29Sample lots in 215
Background Cascade from 215 Harvest 29 Sample lots 19 Farms 13Unique oil values Dry hopped at 3.8g/L Region Farm (coded) OSU Hop Oil (ml/1g) CAS_12_16 WA 2.5 CAS_27_16 WA 1.6 CAS_21_16 ID 1.6 CAS_22_16 ID 1.6 CAS_24_16 WA 9.6 CAS_7_16 ID 7.7 CAS_9_16 ID 14.8 CAS_19_16 WA 2.8 CAS_4_16 WA 5.8 CAS_25_16 OR 13.8 CAS_26_16 WA 12.9 CAS_6_16 ID 7.9 CAS_5_16 WA 5 1. CAS_11_16 WA 2 1. CAS_16_16 WA 15 1.1 CAS_17_16 OR 17 1.1 CAS_15_16 WA 16 1.2 CAS_3_16 OR 4 1.2 CAS_23_16 WA 21 1.2 CAS_2_16 WA 19 1.3 CAS_28_16 WA 1 1.4 CAS_29_16 WA 11 1.4 CAS_2_16 OR 4 1.4 CAS_8_16 OR 8 1.5 CAS_13_16 WA 2 1.5 CAS_1_16 WA 2 1.5 CAS_1_16 OR 4 1.7 CAS_18_16 WA 18 1.7 CAS_14_16 WA 2 2.6
Background Cascade from 215 Harvest 29 Sample lots 19 Farms 13Unique oil values Dry hopped at 3.8g/L Region Farm (coded) OSU Hop Oil (ml/1g) CAS_12_16 WA 2.5 CAS_27_16 WA 1.6 CAS_21_16 ID 1.6 CAS_22_16 ID 1.6 CAS_24_16 WA 9.6 CAS_7_16 ID 7.7 CAS_9_16 ID 14.8 CAS_19_16 WA 2.8 CAS_4_16 WA 5.8 CAS_25_16 OR 13.8 CAS_26_16 WA 12.9 CAS_6_16 ID 7.9 CAS_5_16 WA 5 1. CAS_11_16 WA 2 1. CAS_16_16 WA 15 1.1 CAS_17_16 OR 17 1.1 CAS_15_16 WA 16 1.2 CAS_3_16 OR 4 1.2 CAS_23_16 WA 21 1.2 CAS_2_16 WA 19 1.3 CAS_28_16 WA 1 1.4 CAS_29_16 WA 11 1.4 CAS_2_16 OR 4 1.4 CAS_8_16 OR 8 1.5 CAS_13_16 WA 2 1.5 CAS_1_16 WA 2 1.5 CAS_1_16 OR 4 1.7 CAS_18_16 WA 18 1.7 CAS_14_16 WA 2 2.6 Selected one Cascade lot and dry-hopped at 4 distinct levels g/l 2 g/l 3.8 g/l 8 g/l 16 g/l
Brewing unhopped beer Beer Specifications: Grist: 85% Pale 2-row 13.5% Carmel 1L.5% Carmel 12L Original Gravity: 1.6 P Real Extract: 3.16 P BU = 2 mg/l (iso-extract) ABV = 4.8 % ABV
Hop Preparation and Dry-Hopping Parameters Hop Preparation For Dry Hopping Blended whole cone hops of Cascade, Mixed and homogenized by grinding Weighted into different ratios *sample also taken for oil analysis Dry Hopping Parameters Added to 4 Lfinished beer at, 2, 3.8, 8 & 16 g/l for 24 hrs CO 2 blanket during addition 18 o C during dry-hopping
Blending & filtration All dry-hop events occur in duplicate (4 L beer each) During filtration 2 kegs are blended during filtration into 1 keg Oxygen monitoring DH Event 1 DH Event 2 3.8 g/l 3.8 g/l Filtration 2 events treatment level,, 2, 3.8, 8, and 16 g/l for 24 hrs Carbonated Vollmer, D. et. al. J. Am. Soc. Brew. Chem. 74(4):242-249, 216 3.8 g/l treatment DO DO Spec. 11 ppb
Evaluations using draft beer Minimized total package oxygen Great for sensory testing implementation
Sensory evaluation - descriptive analysis Sensory protocol 12 panelists (1 males, 2 female; 27-54 yrs. old) 4 training sessions in advance 33 products/treatments 4 replications Blocked by replication Sensory Attributes Assessed Overall Hop Aroma Intensity Citrus Herbal/Tea
Sensory evaluation - descriptive analysis Sensory protocol 12 panelists (1 males, 2 female; 27-54 yrs. old) 4 training sessions in advance 33 products/treatments 4 replications Blocked by replication Sensory Attributes Assessed Overall Hop Aroma Intensity Citrus Herbal/Tea
Sensory evaluation descriptive analysis external controls Attributes Base (No dry hop) 3.8g/L 16 g/l Ballast Point Grapefruit Sculpin Citrus Mistress Assess Descriptors Based on Aroma Only Overall Hop Aroma Intensity 8-9 14-15 14-15 7-8 Citrus 7-8 5-6 13-14 6-7 Herbal/Tea 5-6 12-13 1-2 6-7 Panelists came to consensus for attributes on commercial and internally made samples References were served to panelists at each DA session
Sensory Results
Panelist Performance Metrics with OHAI (on samples) Source DF Sum of squares Mean squares F Pr> F Products 32 2717 84.9 1.3 <.1 Assessors 9 1751 194.6 21.2 <.1 Sessions 3 5 1.9.3.81 Products*Assessors 288 2371 8.2 1.6 <.1 Products*Sessions 96 464 4.8 1..511 Assessors*Sessions 27 157 5.8 1.2.23 Error 864 4222 4.9 1 Panelists could discriminate using OHAI attribute Panelist * Sample Interaction (p <.1): For the response OHAI, the rating given to a product depended on the assessor providing the rating Issues with consensus among the panelists (or groups of panelists)
Hop dose response overall hop aroma intensity 14, Sensory Score (15pt scale) 12, 1, 8, 6, 4, 2,, 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) Overall Hop Aroma Intensity
Hop dose response overall hop aroma intensity Sensory Score (15pt scale) 14, 12, 1, 8, 6, 4, 2, a a b c, 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) c Overall Hop Aroma Intensity Panelists could discriminate the different dry hop rate samples (i.e., 2, 3.8, 8, & 16 g/l) *Letters represent Tukey s HSD groupings
Hop dose response overall hop aroma intensity Sensory Score (15pt scale) 14, 12, 1, 8, 6, 4, 2, a a b c, 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) c Overall Hop Aroma Intensity Panelists could discriminate the different dry hop rate samples (ie, 2, 3.8, 8, & 16) *Letters represent Tukey s HSD groupings
Hop dose response hoppy quality (citrus and herbal/tea) 14, Sensory Score (15pt scale) 12, 1, 8, 6, 4, 2,, 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) Overall Hop Aroma Intensity Herbal/ Tea Citrus
Hop dose response hoppy quality (citrus and herbal/tea) Sensory Score (15pt scale) 14, 12, 1, 8, 6, 4, 2,, a a a a a a b b b c c 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) Herbal/ Tea response is similar to OHAI b c c b Overall Hop Aroma Intensity Herbal/ Tea Citrus *Letters represent Tukey s HSD groupings
Hop dose response hoppy quality (citrus and herbal/tea) Sensory Score (15pt scale) 14, 12, 1, 8, 6, 4, 2,, a a a a a a b b b c c b 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) Citrus response seemed to be suppressed compared to OHAI and Herbal/Tea c c b Overall Hop Aroma Intensity Herbal/ Tea Citrus *Letters represent Tukey s HSD groupings
Aroma intensity/ quality as power functions (log-log plots) 1,3 Log(Sensory Score) 1,1,9,7 y =.3x +.72 R² =.992 y =.45x +.47 R² =.993,5,2,3,4,5,6,7,8,9 1 1,1 1,2 1,3 log(dry Hop rate (g/l)) Overall Hop Aroma Intensity Herbal/ Tea Citrus Overall hop aroma intensity and herbal/tea appear to be power functions Kamadia, V.V., et. al. Relationships between Odorant Concentration and Aroma Intensity. Journal of Food Science, 71: S193 S197
Aroma intensity/ quality as power functions (log-log plots) 1,3 Log(Sensory Score) 1,1,9,7 y =.3x +.72 R² =.992 y =.45x +.47 R² =.993,5,2,3,4,5,6,7,8,9 1 1,1 1,2 1,3 log(dry Hop rate (g/l)) Overall Hop Aroma Intensity Herbal/ Tea Citrus If citrus is important there is an optimal rate ~ 4-6 g/l
Hop dose response overall hop aroma intensity Sensory Score (15pt scale) 14, 12, 1, 8, 6, 4, 2, a a b c, 2 4 6 8 1 12 14 16 18 Dry Hop rate (g/l) Panelists scaled the samples randomly against 28 other samples in the DA panel Range of DA OHAI ratings (i.e. ~6-9.5) for 215 and 216 Cascade DA Panels *Letters represent Tukey s HSD groupings c Overall Hop Aroma Intensity
Principal Component Analysis- 215 Cascade Harvest Samples 3 Biplot (axes F1 and F2: 99.48 %) F2 (7.52 %) 2 1-1 -2 29 7 8 24 5 2 28 4 114 13 3.8 g/l 3121 22 26 12 25 19 23 2 11 1 27 9 3 6 18 16 Citrus Overall Hop Aroma Intensity Herbal/ Tea -3-4 -4-3 -2-1 1 2 3 4 5 F1 (91.97%) PCA Type: Pearson (n-1)
Principal Component Analysis- Hop dosage treatments 3 Biplot (axes F1 and F2: 99.48 %) 2 Citrus F2 (7.52 %) 1-1 -2 g/l 2 g/l 3.8 g/l 8 g/l Overall Hop Aroma Intensity Herbal/ Tea -3 16 g/l -4-4 -3-2 -1 1 2 3 4 5 F1 (91.97%) PCA Type: Pearson (n-1)
Chemistry Results
Solid phase micro extraction GC/MS Hop volatiles Dry hopped beer Internal Standard 4-octanol Target Hop Analytes Linalool Terpinen-4-ol α-terpineol Nerol Phenyl Acetate Geraniol Geranial-citral Methyl Geranate Geraniol Acetate β-caryophellyene α-humulene β-farnesene Gernyal Isobutyrate
Hop Compound Concentrations Terpene Alcohols Geraniol(ppb) 4 35 3 25 2 15 1 5 Linalool (ppb) 12 1 2 4 6 8 1 12 14 16 18 8 6 4 2 Dry hop rate (g/l) *Average of 4 SPME-GC/MS instrumental runs 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) Nerol (ppb) 1,8 1,6 1,4 1,2 1,8,6,4,2 # # 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) # Estimated values lower than LOQ
Hop compound concentrations terpene alcohols Geraniol(ppb) 4 35 3 25 2 15 1 5 Linalool (ppb) 12 1 8 6 4 2 Detection threshold 4 ppb a 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) *Average of 4 SPME-GC/MS instrumental runs Detection threshold 1 ppb a 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) Nerol (ppb) 1,8 1,6 1,4 1,2 1,8,6,4,2 # # a KishimotoJ. Agric. Food ChemVol 54, no 23. 26 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) # Estimated values lower than LOQ
Hop oil content and compositional analysis ASBC Hops 13 Steam Distillation Total oil content ASBC Hops 17 Hop Oil Compositional Analysis (GC-FID) for 215 samples (GC-MS) for 216 samples 42 hop analytes
Hop compound % extraction terpene alcohols % Linalool extracted from hop material 25 2 15 1 5 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) % Geraniolextracted from hop material 7 6 5 4 3 2 1 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) % Nerol extracted from hop material *Assuming 1% extraction based on hop oil hydrodistillation 7 6 5 4 3 2 1 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l)
Hop Compound % Extraction Terpene Alcohols % Geraniolextracted from hop material 7 6 5 4 3 2 1 % Linalool extracted from hop material 25 2 15 1 5 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) % Nerol extracted from hop material *Assuming 1% extraction based on hop oil hydrodistillation 7 6 5 4 3 2 1 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) Dry hop rate Terpenealcohol extraction *Could be an artifact of single dry hopping event
Hop Compound Concentrations SPME GC/MS Terpenes 3 β-caryophyllene (ppb) 2,5 2 1,5 1,5 α-humulene (ppb) 2,5 2 1,5 1 #,5 # # 2 4 6 8 1 12 14 16 18 # # 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) Dry hop rate (g/l) β-farnesene (ppb) 2,5 2 1,5 1 # #,5 # 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) *Average of 4 SPME-GC/MS instrumental runs # Estimated values lower than LOQ
Hop Compound Concentrations SPME GC/MS Terpenes β-caryophyllene (ppb) 2,5 2 1,5 1,5 α-humulene (ppb) 3 2,5 2 1,5 1 #,5 # # 2 4 6 8 1 12 14 16 18 # # 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) Dry hop rate (g/l) β-farnesene (ppb) 2,5 2 1,5 1 # #,5 # 2 4 6 8 1 12 14 16 18 Dry hop rate (g/l) Compound Solubility Extraction *Average of 4 SPME-GC/MS instrumental runs # Estimated values lower than LOQ
Principal component analysis hop chemistry 2 Biplot (axes F1 and F2: 96.39 %) 1,5 1 Geranyl Acetate 2 g/l Methyl Geranate Phenyl Acetate F2 (27.47 %),5 -,5 3.8 g/l 8 g/l a-terpineol Geraniol Nerol Linalool Terpinen-4-ol Geranial-citral Alpha Humulene Beta Farnesene Beta Caryophellyene -1 16 g/l -1,5-2 -1,5-1 -,5,5 1 1,5 2 2,5 PCA Type: Pearson Correlation (n-1) F1 (68.92 %)
Principal Component Analysis Just Chemistry 2 Biplot (axes F1 and F2: 96.39 %) 1,5 1 Geranyl Acetate 2 g/l Methyl Geranate Phenyl Acetate F2 (27.47 %),5 -,5 3.8 g/l 8 g/l a-terpineol Geraniol Nerol Linalool Terpinen-4-ol Geranial-citral Alpha Humulene Beta Farnesene Beta Caryophellyene -1 16 g/l -1,5-2 -1,5-1 -,5,5 1 1,5 2 2,5 PCA Type: Pearson Correlation (n-1) F1 (68.92 %)
Sensory + Chemistry Results
Principal Component Analysis Chemistry + Sensory 2 Biplot (axes F1 and F2: 96.23 %) 1,5 F2 (22.86 %) 1,5 -,5 Geranyl Acetate 2 g/l 3.8 g/l Methyl Geranate 8 g/l Phenyl Acetate a-terpineol Citrus Geraniol Nerol Linalool Overall Hop Aroma Intensity Terpinen-4-ol Geranial-citral Herbal/ Tea Beta Farnesene Alpha Humulene -1 Beta Caryophellyene 16 g/l -1,5-2 -1,5-1 -,5,5 1 1,5 2 2,5 F1 (73.36%) PCA Type: Pearson Correlation (n-1)
Principal Component Analysis Chemistry + Sensory 2 Biplot (axes F1 and F2: 96.23 %) 1,5 F2 (22.86 %) 1,5 -,5 Geranyl Acetate 2 g/l 3.8 g/l Methyl Geranate 8 g/l Phenyl Acetate a-terpineol Citrus Geraniol Nerol Linalool Overall Hop Aroma Intensity Terpinen-4-ol Geranial-citral Herbal/ Tea Beta Farnesene Alpha Humulene -1 Beta Caryophellyene 16 g/l -1,5-2 -1,5-1 -,5,5 1 1,5 2 2,5 F1 (73.36%) PCA Type: Pearson Correlation (n-1)
Dry hopping having other impacts on beer Real Extract w/w % 3,55 3,5 3,45 3,4 3,35 3,3 3,25 3,2 3,15 3,1 y =.22x + 3.1 R² =.97 2 4 6 8 1 12 14 16 18 Dry Hop Rate (g/l) ph 4,6 4,5 4,5 4,4 4,4 4,3 4,3 4,2 4,2 4,1 4,1 y =.23x + 4.14 R² =.98 2 4 6 8 1 12 14 16 18 Dry Hop Rate (g/l) BU 31 29 27 25 23 21 19 17 15 2 4 6 8 1 12 14 16 18 Dry Hop Rate (g/l) *ph and RE measured using Anton Paar Alcolyzer *BU measured using ASBC MOA Beer-23
Dry hopping having other impacts on beer Real Extract w/w % 3,55 3,5 3,45 3,4 3,35 3,3 3,25 3,2 3,15 3,1 Bottle Carbonation BU 31 29 27 25 23 21 19 17 15 y =.22x + 3.1 R² =.97 2 4 6 8 1 12 14 16 18 Dry Hop Rate (g/l) ph 4,6 4,5 4,5 4,4 4,4 4,3 4,3 4,2 4,2 4,1 4,1 Bitterness 2 4 6 8 1 12 14 16 18 Dry Hop Rate (g/l) Flavor Stability y =.23x + 4.14 R² =.98 2 4 6 8 1 12 14 16 18 Dry Hop Rate (g/l) *ph and RE measured using Anton Paar Alcolyzer *BU measured using ASBC MOA Beer-23
Conclusions More hops might not be better! For Cascade there seems to be a optimal dry hopping rate ~4-6 g/l to maintain citrusy quality Are there other ways to increase extraction efficiencies of desirable compounds? Double/ Triple dry hopping Dry hopping is having more of an impact on beer than just flavor
A special thanks to Barth Haas Research Grant Ph. D. Advisor Tom Shellhammer OSU Pilot Plant Manager Jeff Clawson OSU Graduate Students: Daniel Vollmer, Christina Hahn, Kaylyn Kirkpatrick OSU Undergraduates Brian Wagman Crosby Hop Farm Staci Wallace BridgePortBrewing Company Jeff Edgerton and Christian Engstrom
Keynote presentations Japan, Germany, Belgium, USA Roundtable discussions Dry-hopping, breeding/local, adv. products Oregon hops farm tour OSU breeding + Coleman Farms hopsflavor217.com
Panelist Performance Metrics with OHAI (on controls) Sum of Mean Source Type DF F Pr > F squares squares Products Fixed 4 277.498 519.374 41.725 <.1 Assessors Random 9 113.968 12.663 1.173.358 Sessions Fixed 3 5.74 1.913.448.721 Products*Assessors Random 36 448.113 12.448 2.99.2 Products*Sessions Fixed 12 54.599 4.55.767.683 Assessors*Sessions Random 27 115.418 4.275.721.835 Error 18 64.558 5.931 Panelist * Sample Interaction (p <.1): For the response OHAI, the rating given to a product depended on the assessor providing the rating Issues with consensus among the panelists (or groups of panelists) 9 Panelists could discriminate using OHAI attribute