1
During processing of dairy products, volatile aroma compounds may be lost due to evaporation or thermal degradation. Traditional methods of recovery of aroma compounds: distillation, gas stripping, solvent extraction. But they involve high temperature, high energy consumption or introduction of other chemicals. Pervporation - no damage to aromas, no additives, good selectivity to aromas. Mass transport mechanism: solution-diffusion model Pervaporation setup Conclusions Recovery of aroma compounds from dairy products by pervaporation Boya Zhang 1 and Xianshe Feng 1, * 1 Membrane Separation Laboratory, Department of Chemical Engineering, University of Waterloo, Waterloo, ON, Canada N2L 3G1 Acknowledgements Background Motivation 0 1 2 3 4 5 0 500 1000 1500 2000 2500 3000 Aroma flux, g/(m 2 .h) Aroma concentration in feed, ppm Nonanal Diacetyl Hexanoic acid Indole Dimethyl sulfone (b) 0.1 1.0 10.0 100.0 1000.0 10000.0 2.9 3.0 3.1 3.2 3.3 3.4 3.5 Permeation flux, g/(m 2 .h) 1000/T, K -1 50 500 1000 2500 Diacetyl concentration in feed, ppm } Water Diacetyl } (a) 0.1 1.0 10.0 100.0 Ratio of flux, driving force or permeability at 65 ℃ and 25 ℃ J65℃/J25℃ DF65℃/DF25℃ P65℃/P25℃ 0.0 0.3 0.6 0.9 1.2 1.5 0 200 400 600 800 1,000 Mass of aroma compound in the organic phase of the accumulated permeate, g Operating time, h 0% 20% 40% 60% 80% 100% 0 100 200 300 400 Recovery rate of diacetyl Operating time, h F 0 /A=226 kg/m 2 F 0 /A=50 kg/m 2 F 0 /A=15 kg/m 2 Organophilic membrane Investigate the effectiveness of pervaporation on recovery of aromas from dairy solutions. Polymer used as membrane material: PEBA 2533 Aroma compounds Formula Odor MW (g/mol) TB (°C) Solubility in water , 25 °C (g/kg) Esters Ethyl hexanoate Pineapple 144 228 0.46 Ethyl butanoate Pineapple 116 120 5.75 Ketones 2-Heptanone Banana 114 151 4.3 Diacetyl Buttery 86 88 250 Acid Hexanoic acid Goat 116 206 10.82 Aldehyde Nonanal Grass 142 191 0.096 Sulfur compound Dimethyl sulfone Hot milk 94 240 Miscible Aromatic compound Indole Jasmine 117 253 1.9 Characterization of PV performance Permeation flux (J i ), g/(m 2 .h) Enrichment factor (β) Ni: permeation rate, g/h X i : Aroma concentration in feed, g/g Y i : Aroma concentration in permeate, g/g = / 0 50 100 150 200 250 0 500 1000 1500 2000 2500 3000 Aroma flux, g/(m 2 .h) Aroma concentration in feed, ppm 2-Heptanone Ethyl butanoate Ethyl hexanoate (a) Effect of feed concentration 0 5 10 15 20 25 30 0 500 1000 1500 2000 2500 3000 Enrichment factor Aroma concentration in feed, ppm Diacetyl Hexanoic acid Indole Dimethyl sulfone (b) 0 50 100 150 200 250 300 0 500 1000 1500 2000 2500 3000 Enrichment factor Aroma concentration in feed, ppm Ethyl butanoate Ethyl hexanoate 2-Heptanone Nonanal (a) Effect of temperature 0 5 10 15 20 25 15 25 35 45 55 65 75 Enrichment factor Temperature, ºC 50 500 1000 2500 Diacetyl concentration in feed, ppm Arrhenius-type equation: J 0 : preexponential factor for flux, g/(m 2 .h); R: universal gas constant，J/(K.mol); T: absolute temperature, K. Reference = − : membrane thickness, μm i : activity coefficient : saturated pressure, Pa : permeate pressure, Pa C a , ppm Diacetyl-water E Ja E Pa E Jw E Pw 50 27.3 -16.4 45.2 1.6 500 46.6 3.1 43.3 -0.3 1000 52.0 8.6 42.4 -1.2 2500 55.0 11.6 43.5 -0.1 Batch pervaporation 0% 20% 40% 60% 80% 100% 0 3 6 9 12 15 Recovery rate Operating time, h Hexanoic acid Diacetyl Ethyl butanoate 2-Heptanone Ethyl hexanoate Feng and Huang’s model (well mixed binary feed): t: operating time, h; F 0 : initial feed amount, g; A: membrane area, m 2 ; X 0 : initial aroma concentration in feed, g/g Initial feed amount: 500 g; Initial feed conc.: 320 ppm for ethyl hexanoate; 2500 ppm for other aromas; Temperature: 36 ℃ S o : aroma solubility in water, g/g; S w : water solubility in aroma, g/g. When aroma concentration in accumulated permeate reaches aroma solubility in water, phase separation occurs. Ethyl butanoate 2-Heptanone Ethyl hexanoate The ratio of initial feed mass (F 0 ) and membrane area (A) affect recovery efficiency. To achieve a 100% recovery, F 0 /A = 226 kg/m 2 , t = 381 h; F 0 /A = 50 kg/m 2 , t = 84 h; F 0 /A = 15 kg/m 2 , t =17 h. Effect of non-volatile dairy components on PV Ethyl hexanoate Nonanal Ethyl butanoate 2-Heptanone Hexanoic acid Indole Dimethyl sulfone Diacetyl 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Normalized enrichment factor NaCl 2 wt% Lactose 2wt% Protein 2 wt% Fat 2 wt% 0 2 4 6 8 10 12 14 16 18 20 22 24 26 0 50 100 150 200 250 Without non-volatile components Whey protein 3.5 wt% Fat 3.5 wt% NaCl 2 wt% Lactose 5 wt% Total flux, g/(m 2 .h) Time, h 1. With an increase in feed aroma concentration or temperature, aroma fluxes increase rapidly. 2.The recovery of aroma compound from its aqueous feed solution is influenced by F 0 /A for a given period of batch operating time. 3. Lactose, whey protein and milk fat in feed negatively affect the pervaporative recovery of aroma compounds, due to the hydrophobic interaction between these non- volatile compounds and aromas. NaCl positively influences the recovery of more hydrophobic aroma compounds. Feng, X., Huang, R.Y.M., 1992. Separation of isopropanol from water by pervaporation using silicone-based membranes. J. Memb. Sci. 74, 171–181. Parliment, T.H., McGorrin, R.J., 2000. Critical flavor compounds in dairy products, in: Flavor Chemistry: Industrial and Academic Research. American Chemical Society, Washington, D.C., pp. 44–71. 0 2 4 6 8 10 12 14 16 18 20 22 24 26 0 40 80 120 160 200 240 Enrichment factor of ethyl butanoate Time, h In general, lactose, whey protein or milk fat in feed solutions negatively affected the pervaporative recovery of aroma compounds. NaCl positively influenced the recovery of more hydrophobic aroma compounds. No concentration polarization or membrane fouling with the presence of lactose, whey protein, fat and NaCl was observed.
