Ultrasonics Sonochemistry 21 (2014) 346–353

Contents lists available at ScienceDirect

Ultrasonics Sonochemistry

journal homepage: www.elsevier .com/ locate/ul tson

Effect of ultrasonic treatment on the polyphenol content and antioxidantcapacity of extract from defatted hemp, flax and canola seed cakes

1350-4177/$ - see front matter � 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.ultsonch.2013.08.002

⇑ Corresponding author. Tel.: +64 34797566; fax: +64 34797567.E-mail addresses: suesiang.teh@otago.ac.nz (S.-S. Teh), john.birch@otago.ac.nz

(E.J. Birch).

Sue-Siang Teh, Edward John Birch ⇑Department of Food Science, University of Otago, PO Box 56, Dunedin 9054, New Zealand

a r t i c l e i n f o a b s t r a c t

Article history:Received 7 August 2012Received in revised form 27 June 2013Accepted 5 August 2013Available online 13 August 2013

Keywords:UltrasonicPolyphenolsTotal phenolic contentFlavonoidsAntioxidant capacityDefatted seed cake

The effectiveness of ultrasonic extraction of phenolics and flavonoids from defatted hemp, flax and canolaseed cakes was compared to the conventional extraction method. Ultrasonic treatment at room temper-ature showed increased polyphenol extraction yield and antioxidant capacity by two-fold over the con-ventional extraction method. Different combinations of ultrasonic treatment parameters consisting ofsolvent volume (25, 50, 75 and 100 mL), extraction time (20, 30 and 40 min) and temperature (40, 50,60 and 70 �C) were selected for polyphenol extractions from the seed cakes. The chosen parametershad a significant effect (p < 0.05) on the polyphenol extraction yield and subsequent antioxidant capacityfrom the seed cakes. Application of heat during ultrasonic extraction yielded higher polyphenol contentin extracts compared to the non-heated extraction. From an orthogonal design test, the best combinationof parameters was 50 mL of solvent volume, 20 min of extraction time and 70 �C of ultrasonictemperature.

� 2013 Elsevier B.V. All rights reserved.

1. Introduction

Hemp (Cannabis sativa), Flax (Linum usitatissimum) and Canola(Brassica napus) seed oils have been favoured for human consump-tion due to their high amount of essential fatty acids namely a-lin-olenic acid (C18:3; n-3) and linoleic acid (C18:2; n-6). Theproduction of plant seed oils has generated tonnes of processingwastes called seed cakes. The seed cakes are then further processedinto animal feed due to their high protein and energy contents.Previous study has shown that calves fed with seed cakes had sim-ilar live weigh gain with calves fed a mixture of soybean meal andbarley [1]. However, the by-products of the oil production also con-tain secondary metabolites namely phenolic acids and flavonoidsthat have not been studied extensively. The polyphenols that exhi-bit functional and nutraceutical properties could be used in func-tional foods. Previous studies show that the antioxidantcompounds improve human health such as healing human chroniculceration [2], anti-allergenic, anti-atherogenic, anti-inflammatory,anti-microbial, antiviral, antioxidant, anti-thrombotic, cardiopro-tective and vasodilatory effects [3,4], inhibition of cancer cells,improving the condition of cardiovascular diseases and diabetes[5,6].

Polyphenols of seed cakes are enclosed in hard and insolublestructures such as vacuoles, lipoprotein bilayers, lignin, hull and

cell walls [7]. Phenolic acids occur as esters, glycosides and boundcomplexes in plants. These factors have hindered polyphenolextraction from seed cakes by conventional extraction methods.Ultrasonic treatment is an economic, environmentally friendly, en-ergy saving and sustainable technology used industrially to in-crease mass transfer phenomena in plant cellular tissues [8].Ultrasound generates low frequency energy (20 kHz to 1 MHz)for purposes such as cleaning, sterilizing, degassing, precipitation,leaching, digestion, crystallization and extraction [9–13]. The en-ergy generates cavitation force to increase mass transfer rateswhere resulting bubbles in the liquid/solid extraction explosivelycollapse. This phenomenon generates localized pressure to disruptthe plant tissue and release the intracellular bioactive compounds[14]. Various studies have incorporated ultrasound into oil extrac-tion of Jatropha curcas, Pongamia [15] and hemp seed [16] whereultrasonic treatment successfully increased the oil yield in shorteroperating times compared to solvent extraction. Other studies thatapplied ultrasound include anthocyanin extraction from grape by-products [7] and phenolics extraction from cranberry products [9],coconut shell powder [17] and Sclerocarya birrea kernel oil cake[18].

The application of ultrasound for optimization of yield of poly-phenols from seed cakes contributes to industrial applications eco-nomically and environmentally since it reduces the usage oforganic solvent and extraction time. Furthermore, the polyphenolsextracted from the seed cakes by ultrasonic treatment would be agood source in product development for nutraceuticals and func-tional foods that could extend product shelf-life. The objectives

Fig. 1. Total phenolic content of the seed cake extracts by ultrasonic treatment andthe control. Values are mean ± standard deviations of three (n = 3) measurements.Lower case letters (a and b) within bars of the same sample with differentextraction methods and capital alphabet letters (A, B and C) within bars of thedifferent samples are significantly different at p < 0.05.

S.-S. Teh, E.J. Birch / Ultrasonics Sonochemistry 21 (2014) 346–353 347

of this study are: (i) to investigate the differences of polyphenolextraction yield in seed cakes between ultrasonic and conventionalmethods; (ii) to identify the optimum parameters of ultrasonica-tion for maximum polyphenol yield from seed cakes; (iii) to eval-uate the antioxidant capacity of the ultrasonicated seed cakeextracts.

2. Materials and methods

2.1. Materials

Methanol and acetone were analytical grade and purchasedfrom Labserv™, Biolab (Aust) Ltd., Victoria, Australia. Defattedhemp (C. sativa) and flax seed (L. usitatissimum) cakes were ob-tained from Oil Seed Extractions Limited, Ashburton, New Zealandwhile canola (B. napus) seed cake was supplied by New ZealandVegetable Oil Limited, Canterbury, New Zealand. The seed cakeswere milled into powder using a Cemotec Sample Mill 1090 (FOSSTecator, Hoganas, Sweden) and the powders were sieved to pro-duce particles to pass a 450 lm sieve. The seed cake powders werevacuum-packed and stored at 4 �C prior to analysis.

2.2. Methods

2.2.1. Ultrasonic extractionSeed cake powders (5.00 g) were mixed with methanol:ace-

tone:water (MAW, 7:7:6 v/v/v) in a conical flask. The flask was cov-ered with petri film to avoid solvent evaporation beforeultrasonication. The experiments were conducted using an ultra-sonic bath (Elma�, Germany) with a fixed power (200 W). Variousparameters such as solvent volume (25, 50, 75, 100 mL), extractiontime (20, 30, 35 min) and temperatures (40, 50, 60, 70 �C) wereused in the study. The extracts were filtered through filter paper(0.45 lm, Whatman™) by vacuum. Filtrates were stored in a darkambient glass bottle at 4 �C prior to analysis.

2.2.2. Control extractionIn order to compare ultrasonic treatment, similar conditions are

needed for the setup of the control. Seed cake powders (5.00 g)were mixed with 50 mL of mixed solvent (MAW) in a conical flask.The mixtures were stirred with a magnetic stirrer (4.5 � 0.5 cm) at1000 rpm without application of heat for 30 min at room temper-ature (25 ± 1 �C). The extracts were filtered through filter paper(0.45 lm, Whatman™) by vacuum before they were stored at4 �C prior to analysis.

2.2.3. Determination of total phenolics in seed cake extractsThe total phenolic content in the extracts was determined based

on the method of Gutfinger [19]. Extract (0.1 mL) was made up to5 mL with distilled water in a 10-mL volumetric flask. Folin–Ciocal-teau’s phenol reagent (0.5-mL; 2 N) was added into the mixture.After 3 min, saturated (35% w/v) sodium carbonate solution(1 mL) was added into the mixture, following by topping up themixture to 10 mL with water. The mixture was measured at725 nm using a spectrophotometer against a reagent blank thatconsisted of all reagents without the extract. The standard curveconsisted of gallic acid within the concentration range of 0–400 lg/mL assay solution. Results were expressed as mg gallic acidequivalents (GAE)/100 g of fresh weight.

2.2.4. Determination of total flavonoids in seed cake extractsThe flavonoids were determined following the method of Oo-

mah, Mazza, and Kenaschuk [20]. Distilled water (3 mL) was mixedwith 1 mL of extract, followed by 100 lL of diphenylboric acid 2-aminoethyl ester solution (1% v/v) before the mixture was

measured spectrophotometrically at 404 nm. Luteolin (0–42 lg/3 mL assay solution in 80% methanol) was used as the standardfor the calibration curve. Results were expressed as mg luteolinequivalents (LUE)/100 g of fresh weight.

2.2.5. Determination of 1,1-diphenyl-2-picrylhydrazyl (DPPH) freeradical-scavenging assay

DPPH (1,1-diphenyl-2-picrylhydrazyl) is a stable radical that iswidely used in estimating the antioxidant properties where freeradical scavengers react quickly with DPPH radicals, causing a de-crease in the absorbance measured [20]. Thus, a lower absorbancerepresents a higher DPPH scavenging activity and vice versa. TheDPPH free radical-scavenging assay was based on the method ofde Ancos et al. [21] with some modification. Extract (10 lL) wasmixed with 3.99 mL of 25 mM DPPH� in MAW solution. The mix-ture was vortexed and kept in the dark for 30 min. The mixturewas measured spectrophotometrically at 515 nm, against a blanksolution without the presence of DPPH�. Results were expressedas percentage inhibition of the DPPH� as shown in the followingequation:

% inhibition of DPPH¼ðAbsorbance control

�Absorbance sampleÞ=Absorbance control

�100

where absorbance control is the absorbance of DPPH� solution with-out extract.

2.2.6. Determination of ferric reducing/antioxidant power (FRAP)assay

Ferric reducing/antioxidant power (FRAP) of bioactive com-pound extract is the determination of the antioxidant capacity toreduce ferric-tripyridyltriazine (Fe(III)-TPTZ) complex to the fer-rous (FeII) form, which is a dark blue color at the UV absorptionof 593 nm. In this case, the antioxidant compounds represent therole of electron donor in order to reduce the ferric complex to fer-rous form. FRAP assay was carried out following the method ofBenzie and Strain [22] with modification. FRAP reagent was madeby mixing 200 mL of acetate buffer (300 mM, pH 3.6), 20 mL of2,4,6-tri(2-pyridyl)-s-triazine (TPTZ) solution (10 mM), 20 mL ofFeCl3 solution (20 mM) and 24 mL distilled water. TPTZ and FeCl3

solutions were prepared freshly every day prior to analysis. Thestraw colored FRAP reagent was kept in 37 �C water bath prior to

Fig.2.Total

flavon

oidsof

the

seedcake

extractsby

ultrason

ictreatm

ent

and

the

control.

Valu

esare

mean

±stan

darddeviation

sof

three

(n=

3)m

easurem

ents.

Lower

caseletters

(aan

db)

with

inbars

ofth

esam

esam

plew

ithdifferen

textraction

meth

odsan

dcapital

alphabet

letters(A

,B

and

C)

with

inbars

ofth

edifferen

tsam

plesare

signifi

cantly

different

atp

<0.05.

Fig.3.

Percentage

(%)

inh

ibitionD

PPH�

ofth

eseed

cakeextracts

byu

ltrasonic

treatmen

tan

dth

econ

trol.Valu

esare

mean

±stan

darddeviation

sof

three

(n=

3)m

easurem

ents.Low

ercase

letters(a

and

b)w

ithin

barsof

the

same

sample

with

different

extractionm

ethods

and

capitalalphabet

letters(A

,Ban

dC

)w

ithin

barsof

the

different

samples

aresign

ifican

tlydifferen

tat

p<

0.05.

Fig.4.

Ferricredu

cing

antioxidan

tpow

er(FR

AP

value)

ofth

eseed

cakeextracts

byu

ltrasonic

treatmen

tan

dth

econ

trol.Valu

esare

mean

±stan

darddeviation

sof

three

(n=

3)m

easurem

ents.Low

ercase

letters(a

and

b)w

ithin

barsof

the

same

sample

with

different

extractionm

ethods

and

capitalalphabet

letters(A

,Ban

dC

)w

ithin

barsof

the

different

samples

aresign

ifican

tlydifferen

tat

p<

0.05.

Table 1Effect of solvent volume on the phenolic and flavonoid contents and antioxidant capacity of seed cakes extracts.*

Solvent volume(mL)

HempB FlaxC CanolaA

Totalphenolics

Totalflavonoids

DPPH� FRAP Totalphenolics

Totalflavonoids

DPPH� FRAP Total phenolics Totalflavonoids

DPPH� FRAP

25 521.67 ± 1.04d 8.39 ± 0.02d 9.53 ± 0.03d 4.29 ± 0.03d 475.40 ± 2.13d 5.61 ± 0.03d 7.24 ± 0.03d 1.28 ± 0.02d 1972.00 ± 1.35d 17.56 ± 1.72d 14.28 ± 1.69d 4.29 ± 0.03d

50 871.60 ± 0.87a 11.67 ± 0.03c 21.14 ± 0.03a 8.27 ± 0.03a 806.67 ± 1.81a 8.54 ± 0.02c 15.28 ± 0.03a 4.29 ± 0.03a 2205.60 ± 0.66a 25.37 ± 0.02c 36.46 ± 0.04a 15.47 ± 0.03a

75 854.56 ± 0.59b 15.26 ± 0.05b 19.35 ± 0.06b 7.68 ± 0.05b 786.24 ± 2.47b 10.53 ± 0.03b 14.75±0.03b 4.03 ± 0.03b 2186.38 ± 1.68b 30.64 ± 0.10b 35.78 ± 0.18b 13.57 ± 0.08b

100 805.13 ± 2.90c 28.18 ± 0.63a 17.20 ± 0.02c 6.32 ± 0.02c 752.60 ± 1.65c 15.64 ± 0.02a 13.65 ± 0.02c 3.89 ± 0.02c 2163.33 ± 1.20c 38.28 ± 0.02a 34.15 ± 0.03c 11.68 ± 0.02c

abc Values are mean ± standard deviations of three (n = 3) measurements with different superscript letters within columns are significantly different (p < 0.05).ABC Values are mean ± standard deviations of three (n = 3) measurements with different superscript lettersacross seed types are significantly different (p < 0.05).Total phenolics, mg GAE/100 g fresh weight; total flavonoids, mg LUE/100 g fresh weight; DPPH�, % inhibition of DPPH�; FRAP, lmol Fe(II)/g fresh weight.* Results were calculated on the same volume basis for DPPH� activity.

348S.-S.Teh,E.J.Birch

/Ultrasonics

Sonochemistry

21(2014)

346–353

Table 2Effect of extraction time on the phenolic and flavonoid contents and antioxidant capacity of seed cake extracts.

Time extraction(min)

HempB FlaxC CanolaA

Totalphenolics

Totalflavonoids

DPPH� FRAP Totalphenolics

Totalflavonoids

DPPH� FRAP Total phenolics Totalflavonoids

DPPH� FRAP

20 887.67 ± 1.39a 15.73 ± 0.13a 20.26 ± 0.01a 7.88 ± 0.02a 844.30 ± 2.11a 9.78 ± 0.01a 17.14 ± 0.03a 4.76 ± 0.04a 2237.23 ± 0.23a 28.38 ± 0.02a 34.68 ± 0.03a 12.34 ± 0.04a

30 869.93 ± 0.40b 11.56 ± 0.04b 17.21 ± 0.01b 6.53 ± 0.03b 804.03 ± 1.32b 8.52 ± 0.02b 13.60 ± 0.02b 3.92 ± 0.02b 2201.27 ± 1.17b 25.42 ± 0.02b 34.10 ± 0.02b 11.70 ± 0.02b

40 646.97 ± 0.81c 8.29 ± 0.05c 13.59 ± 0.02c 5.28 ± 0.02c 617.20 ± 0.87c 6.26 ± 0.01c 12.14 ± 0.03c 3.54 ± 0.02c 2183.60 ± 0.78c 22.64 ± 0.01c 33.68 ± 0.03c 10.25 ± 0.05c

abc Values are mean ± standard deviations of three (n = 3) measurements with different superscript letters within columns are significantly different (p < 0.05).ABC Values are mean ± standard deviations of three (n = 3) measurements with different superscript letters across seed types are significantly different (p < 0.05).Total phenolics, mg GAE/100 g fresh weight; total flavonoids, mg LUE/100 g fresh weight; DPPH�, % inhibition of DPPH�; FRAP, lmol Fe(II)/g fresh weight.

Table 3Effect of temperature on the phenolic and flavonoid contents and antioxidant capacity of seed cake extracts.

Temperature(�C)

HempB FlaxC CanolaA

Total phenolics Totalflavonoids

DPPH� FRAP Total phenolics Totalflavonoids

DPPH� FRAP Total phenolics Totalflavonoids

DPPH� FRAP

40 911.30 ± 4.56d 14.09 ± 0.09c 19.33 ± 0.02d 9.20 ± 0.02d 873.97 ± 1.40d 10.58 ± 0.04d 17.30 ± 0.04d 5.02 ± 0.03d 2276.83 ± 0.76d 27.46 ± 0.01d 34.22 ± 0.03d 17.40 ± 0.02d

50 1261.20 ± 4.97c 25.27 ± 0.30a 24.15 ± 0.02c 13.68 ± 0.02c 950.20 ± 2.05c 15.25 ± 0.04a 18.46 ± 0.03c 5.56 ± 0.04c 2307.27 ± 0.87c 35.31 ± 0.03a 36.64 ± 0.02c 18.45 ± 0.02c

60 1277.73 ± 5.09b 25.10 ± 0.10a 24.90 ± 0.02b 14.27 ± 0.02b 984.47 ± 0.86b 12.70 ± 0.02b 18.62 ± 0.02b 6.67 ± 0.03b 2357.37 ± 1.01b 32.20 ± 0.02b 37.68 ± 0.02b 20.56 ± 0.04b

70 1542.03 ± 2.97a 23.31 ± 0.05b 30.94 ± 0.02a 16.74 ± 0.02a 1257.37 ± 1.31a 12.17 ± 0.03c 22.54 ± 0.04a 8.67 ± 0.03a 2563.50 ± 0.87a 30.10 ± 0.03c 38.56 ± 0.02a 23.64 ± 0.04a

abc Values are mean ± standard deviations of three (n = 3) measurements with different superscript letters within columns are significantly different (p < 0.05).ABC Values are mean ± standard deviations of three (n = 3) measurements with different superscript letters across seed types are significantly different (p < 0.05).Total phenolics, mg GAE/100 g fresh weight; total flavonoids, mg LUE/100 g fresh weight; DPPH�, % inhibition of DPPH�; FRAP, lmol Fe(II)/g fresh weight.

S.-S.Teh,E.J.Birch/U

ltrasonicsSonochem

istry21

(2014)346–

353349

Table 4The orthogonal experiment design and the responses for the % DPPH� and FRAP value of hemp seed cake extract.*

Test No. Factor and level % DPPH� FRAP

A B CSolvent volume (mL) Extraction time (min) Temperature (�C)

1 50 20 70 30.92 16.702 50 30 60 24.92 14.123 50 40 50 23.05 13.524 75 20 60 18.82 12.455 75 30 50 16.42 10.276 75 40 70 17.55 11.527 100 20 50 10.36 9.388 100 30 70 14.58 10.489 100 40 60 12.64 9.63TD1 78.89 60.10 49.83TD2 52.79 55.92 56.38TD3 37.58 53.24 63.05kD1 26.30 20.03 16.61kD2 17.60 18.64 18.79kD3 12.53 17.75 21.02RD⁄ 13.77 2.28 4.41Excellent level A1 B1 C1

Influencing factors A > C > BSuperior combination A1 B1 C1

TE1 44.34 38.53 33.17TE2 34.24 34.87 36.20TE3 29.49 34.67 38.70kE1 14.78 12.84 11.06kE2 11.41 11.62 12.07kE3 9.83 11.57 12.90RE⁄ 4.95 1.27 1.84Excellent level A1 B1 C1

Influencing factors A > C > BSuperior combination A1 B1 C1

All values are presented as mean of triplicates experiments (n = 3). TD1, TD2, TD3 represent % DPPH scavenging activity for each factor level; kD1, kD2, kD3 represent mean of %DPPH scavenging activity for each factor level; RD⁄ represents effect of the factor on % DPPH scavenging activity. TE1, TE2, TE3 represent FRAP value for each factor level; kE1, kE2,kE3 represent mean of FRAP value for each factor level; RE⁄ represents effect of thefactor on FRAP value. % DPPH�, % inhibition of DPPH�; FRAP, lmol Fe(II)/g fresh weight.* Results were calculated on the same volume basis for DPPH�.

350 S.-S. Teh, E.J. Birch / Ultrasonics Sonochemistry 21 (2014) 346–353

analysis. Extract (40 lL) was mixed with 3 mL of FRAP reagent be-fore it was incubated at 37 �C for 4 min. The mixture was thenmeasured spectrophotometrically at 593 nm against the blank.The blank solution was made up of 40 lL MAW solution in 3 mLFRAP reagent, incubated at 37 �C for 1 h. Standard solutions con-sisted of FeSO4�7H2O in ascending concentration order of 0.1–1.0 mM. The results were expressed as micromoles Fe(II)/g freshweight.

2.3. Statistical analysis

All experiments were carried out three times. Results were re-ported as mean ± standard deviation of triplicate measurements.Significant difference (p < 0.05) within means was analyzed byanalysis of variance (ANOVA) and Tukey’s honestly significant dif-ference (HSD) test. An orthogonal design test was conducted toevaluate the optimum combination of parameters that yieldedmaximum antioxidant capacity of the seed cake extracts. Theorthogonal test results were analyzed using General Linear Model,UNIVARIATE and descriptive statistics. All the data analysis testswere assessed by the SPSS Statistics Software Version 20 (IBM,New York, USA).

3. Results and discussion

3.1. Comparison between ultrasonic extraction of seed cake polyphenolcontent and the control

In order to identify the effectiveness of polyphenol extractionyields by ultrasound, a control experiment was conducted. Forultrasonication, seed cake powder (5.00 g) mixed with 50 mL of

solvent was ultrasonicated for 30 min at 200 W of ultrasonic powerat room temperature (25 �C). The results showed that ultrasonica-tion significantly increased the polyphenols extraction yield fromthe seed cakes compared to the control (p < 0.05; Figs. 1 and 2).For example, total phenolics of ultrasonicated hemp, flax and canolaseed cake extracts were 123.70%, 117.96% and 112.18% higher thanthe controls respectively (Fig. 1). In addition, total flavonoids ofseed cake extracts extracted using ultrasonic treatment were signif-icantly higher than the control (p < 0.05; Fig. 2). The results showthat total flavonoids of ultrasonication extracts from hemp, flaxand canola seed cakes were 7.17 ± 0.03, 4.71 ± 0.02 and12.2 ± 0.05 mg LUE/100 g fresh weight higher than the control.The antioxidant capacity was in accordance with the polyphenolcontent in the extracts. The results show that % inhibition of DPPH�

was higher in the ultrasonication extracts compared to the control(Fig. 3). Ultrasonicated hemp, flax and canola extracts exhibited %inhibition of DPPH� of 17.18%, 13.62% and 34.17% respectively. Fur-thermore, ultrasonicated hemp, flax and canola seed cake extractsexhibited reducing power, 6.30 ± 0.02, 3.86 ± 0.02, 11.7 ± 0.02 lmolFe(II)/g fresh weight respectively, higher than the controls (Fig. 4).This indicated that either a longer extraction time may be neededfor conventional solid–liquid extraction to achieve higher polyphe-nols extraction [23,24], and hence antioxidant capacity, or a lowerfinal yield than was achieved by ultrasonic-assisted extraction.

3.2. Effect of solvent volume on the ultrasonication assisted yield ofpolyphenols and antioxidant capacity from hemp, flax and canola seedcake extracts

The selected solvent volumes used in this study were 25, 50, 75and 100 mL. Other parameters were kept constant such as 5.00 g of

Table 5The orthogonal experiment design and the responses for the % DPPH and FRAP value of flax seed cake extract.*

Test No. Factor and level % DPPH� FRAP

A B CSolvent volume (mL) Extraction time (min) Temperature (�C)

1 50 20 70 22.50 8.632 50 30 60 18.58 6.503 50 40 50 17.56 5.424 75 20 60 16.47 5.085 75 30 50 14.23 4.566 75 40 70 15.85 4.887 100 20 50 7.58 3.938 100 30 70 10.35 4.369 100 40 60 8.62 4.11TD1 58.64 46.55 39.37TD2 46.55 43.16 43.67TD3 26.55 42.03 48.70kD1 19.55 15.52 13.12kD2 15.52 14.39 14.56kD3 8.85 14.01 16.23RD⁄ 10.70 1.51 3.11Excellent level A1 B1 C1

Influencing factors A > C > BSuperior combination A1 B1 C1

TE1 20.55 17.64 13.91TE2 14.52 15.42 15.69TE3 12.40 14.41 17.87kE1 6.85 5.88 4.64kE2 4.84 5.14 5.23kE3 4.13 4.80 5.96RE⁄ 2.72 1.08 1.32Excellent level A1 B1 C1

Influencing factors A > C > BSuperior combination A1 B1 C1

All values are presented as mean of triplicates experiments (n = 3). TD1, TD2, TD3 represent % DPPH scavenging activity for each factor level; kD1, kD2, kD3 represent mean of %DPPH scavenging activity for each factor level; RD⁄ represents effect of the factor on % DPPH scavenging activity. TE1, TE2, TE3 represent FRAP value for each factor level; kE1, kE2,kE3 represent mean of FRAP value for each factor level; RE

⁄ represents effect of the factor on FRAP value. DPPH�, % inhibition of DPPH�; FRAP, lmol Fe(II)/g fresh weight.* Results were calculated on the same volume basis.

S.-S. Teh, E.J. Birch / Ultrasonics Sonochemistry 21 (2014) 346–353 351

seed cake powders, 30 min of extraction time and 200 W of ultra-sonic power at room temperature (25 �C). The results show solventvolume had a significant effect (p < 0.05) on the extraction of pol-yphenol contents and antioxidant capacity from the seed cake ex-tracts (Table 1). The best solvent volume that yielded the highestphenolic content was 50 mL, followed by 75 and 100 mL of solvent.Variation of the maximum recovery of different types of phenoliccompounds from similar plant material had different optimum so-lid to liquid ratio [25], even from the different plant materials orvarieties, due to plant morphology, structure and rheology [26].In the present study, it was found that flavonoid content of the ex-tracts increased proportionally to the solvent volume and resultedin the highest flavonoid recovery from the extracts in 100 mL ofsolvent. Previous study showed that increasing the solid to liquidratio resulted in higher recovery of flavonoids [27], as the water in-creases the contact surface area between the plant matrix and thesolvent [28] since increasing the mixed MAW solvent increasedwater percentage proportionally. The results show that antioxidantcapacity (DPPH & FRAP values) of the extracts was the highest in50 mL of solvent, which was in accordance with the phenolic con-tents of the extracts [24,29].

3.3. Effect of ultrasonication time on the polyphenol yield andantioxidant capacity from hemp, flax and canola seed cake extracts

To determine the effect of ultrasonication time (20, 30 and40 min), the condition of constant parameters was set as 5.00 gof seed cake powders in 50 mL of solvent volume with 200 W ofultrasonic power at room temperature (25 �C). A significant effect

(p < 0.05) of ultrasonication time was found on the polyphenolyield from the seed cakes (Table 2). The results showed that anultrasonication time of 20 min yielded the highest total phenolicand flavonoid contents from seed cakes. The polyphenol contentsof seed cake extracts started reducing at the ultrasonication timeof 30 min. Similar trends were found in DPPH and FRAP values,where 20 min of ultrasonication time resulted in the highest anti-oxidant capacity of seed cake extracts. Prolonging ultrasonicationtime was not able to increase the polyphenol contents and antiox-idant capacity of the seed cake extracts. This is because the ultra-sonic wave decomposes the polyphenols of the extracts in longerexposure periods [30]. Among all materials, canola seed cake ex-tracts had the highest (p < 0.05) recovery of total phenolics, totalflavonoids and antioxidant capacity while flax seed cake extractshad the lowest recovery of all measured responses.

3.4. Effect of temperature on the ultrasonication assisted yield of thepolyphenols and antioxidant capacity from hemp, flax and canola seedcake extracts

The selected temperatures for ultrasonication were 40, 50, 60and 70 �C. The constant parameters consisted of 5.00 g of seed cakepowders in 50 mL of solvent, 20 min of ultrasonication time and200 W of ultrasonic power. The results show that ultrasonic tem-perature had a significant effect (p < 0.05) on the polyphenol con-tents of the seed cake extracts (Table 3). Among all materials,canola and flax seed cake extracts had the highest (p < 0.05) andlowest recovery of total phenolics, total flavonoids and antioxidantcapacity respectively. Application of temperature to polyphenols

Table 6The orthogonal experiment design and the responses for the % DPPH and FRAP value of canola seed cake extract.*

Test No. Factor and level % DPPH� FRAP

A B CSolvent volume (mL) Extraction time (min) Temperature (�C)

1 50 20 70 38.54 23.602 50 30 60 37.23 20.283 50 40 50 35.36 18.574 75 20 60 34.55 16.525 75 30 50 31.28 14.186 75 40 70 33.65 15.487 100 20 50 25.24 10.528 100 30 70 28.64 12.659 100 40 60 27.51 11.81TD1 111.13 98.33 91.88TD2 99.48 97.15 99.29TD3 81.39 96.52 100.83kD1 37.04 32.78 30.63kD2 33.16 32.38 33.10kD3 27.13 32.17 33.61RD⁄ 9.91 0.61 2.98

Excellent level A1 B1 C1

Influencing factors A > C > BSuperior combination A1 B1 C1

TE1 62.45 50.64 43.27TE2 46.18 47.11 48.61TE3 34.98 45.86 51.73kE1 20.82 16.88 14.42kE2 15.39 15.70 16.20kE3 11.66 15.29 17.24RE⁄ 9.16 1.59 2.82

Excellent level A1 B1 C1

Influencing factors A > C > BSuperior combination A1 B1 C1

All values are presented as mean of triplicates experiments (n = 3). TD1, TD2, TD3 represent % DPPH scavenging activity for each factor level; kD1, kD2, kD3 represent mean of %DPPH scavenging activity for each factor level; RD⁄ represents effect of the factor on % DPPH scavenging activity. TE1, TE2, TE3 represent FRAP value for each factor level; kE1, kE2,kE3 represent mean of FRAP value for each factor level; RE⁄ represents effect of the factor on FRAP value. % DPPH�, % inhibition of DPPH�; FRAP, lmol Fe(II)/g fresh weight.

* Results were calculated on the same volume basis.

352 S.-S. Teh, E.J. Birch / Ultrasonics Sonochemistry 21 (2014) 346–353

ultrasonication was found to increase the polyphenol contents ofthe seed cake extracts compared to polyphenol extraction withoutheat application (as shown in Tables 1 and 2). It was found that therecovery of phenolic content was parallel to the increase of tem-perature from 40 to 70 �C, while % inhibition of DPPH� and Ferricreducing antioxidant power (FRAP) were correlated well with thephenolics content (Table 3). Total flavonoids of all seed cakes in-creased significantly (p < 0.05) from 40 to 50 �C. After 60 �C, totalflavonoids in flax and canola seed cake extracts decreased, whiletotal flavonoids in hemp seed cake extract decreased at 70 �C. Pre-vious studies documented that a higher temperature destroyed fla-vonoid compounds from citrus peel [31]. Although a highertemperature increased cavitation of ultrasonic assisted extractionby assisting in cell wall breaking in order to release the polyphenolto the extraction medium, a higher temperature, ultrasonic fre-quency and power could decompose flavonoids that were releasedinto the extraction medium. The effect of heating temperature tothe polyphenol extraction from plant material was associated tothe types and different bound forms of polyphenol that are pre-sented in plants depending on the species [32]. Hence, differentplant species resulted in different optimum extraction and recov-ery of extracted compounds. The ultrasonication temperature of70 �C was the optimum temperature for the maximum yield of pol-yphenol contents and antioxidant capacity of the seed cake ex-tracts. Previous studies have proved that application of heat(70 �C) increased mass transport phenomena by improving inter-nal liquid phase pressure, resulting in centrifugal circulation ofthe solutes through plant membranes [7,33]. Furthermore, heatapplication is able to break the bonding of the phenolic-matrixand impact the membrane chemical structure of plant tissuesand causing coagulation of lipoproteins [7].

3.5. Optimization of antioxidant capacity through best combinationparameters using orthogonal testing

In order to evaluate the optimum combination of parametersfor the antioxidant capacity of the seed cake extracts, orthogonaltesting was incorporated in the study based on the results fromthe single-factor experiments. The orthogonal experiment designwas developed by Dr. Genichi Taguchi in order to narrow the var-iation in a process for evaluating various combination parametersthat affect the results (Web reference [1]). The orthogonal test ismeant for minimizing experimentation which requires few param-eters that contribute significantly to results. Three factors consist-ing of three levels for each factor was selected for orthogonal test,L9 (34). The factors were solvent volume (50, 75, 100 mL), extrac-tion time (20, 30, 40 min) and temperature (50, 60, 70 �C), asshown in Tables 4–6. The most influencing factor for the maximumantioxidant capacity of the seed cake extracts was solvent volume,followed by temperature and extraction time. The results showthat optimum parameters for 200 W ultrasonic extraction con-sisted of 50 mL of solvent volume, 20 min of extraction time and70 �C of temperature (Tables 4–6).

4. Conclusions

The application of ultrasound is able to increase polyphenolextraction yields from the seed cakes. At the same time, ultrasonictreatment aided in enhancing antioxidant capacity of the seed cakeextracts. The incorporation of heat during ultrasonic extraction re-sulted in higher polyphenol yields from the seed cakes comparedto the ultrasonic treatment without heat. The effective parameters

S.-S. Teh, E.J. Birch / Ultrasonics Sonochemistry 21 (2014) 346–353 353

for optimum antioxidant capacity of seed cake extract was 1:10(solid:solvent ratio), 20 min of extraction time, 70 �C of heatingtemperature, 200 W ultrasonic power. Ultrasound is a low-costtechnology for polyphenol extraction from seed cakes because itrequires minimum usage of solvent and shorter extraction timescompared to the conventional extraction method. Furthermore,the application of ultrasound in enhancing polyphenol extractionfrom seed cakes will augment a sustainability approach environ-mentally and economically in developing functional food for healthpurposes.

Acknowledgments

The authors are grateful to the University of Otago DoctoralScholarship for financial support throughout the study. Theauthors are appreciative to Dr. Brian Niven from Department ofMathematics and Statistics for assistance in designing experimentand analysis with orthogonal design.

References

[1] A. Hessle, M. Eriksson, E. Nadeau, T. Turner, B. Johansson, Cold-pressedhempseed cake as a protein feed for growing cattle, Acta AgriculturaeScandinavica, Section A: Animal Science 58 (2008) 136–145.

[2] K. Skórkowska-Telichowska, M. Zuk, A. Kulma, A. Bugajska-Prusak, K.Gasiorowski, K. Kostyn, J. Szopa, New dressing materials derived fromtransgenic flax products to treat long-standing venous ulcers – a pilot study,Wound Repair and Regeneration 18 (2010) 168–179.

[3] O. Benavente-García, J. Castillo, F.R. Marin, A. Ortuño, J.A. Del Río, Uses andproperties of citrus flavonoids, Journal of Agricultural and Food Chemistry 45(1997) 4505–4515.

[4] I. Bravo, Polyphenols: chemistry, dietary sources, metabolism, and nutritionalsignificance, Nutrition Reviews 56 (1998) 317–333.

[5] M. _Zuk, L. Dyminska, A. Kulma, A. Boba, A. Prescha, J. Szopa, M. Maczka, A.Zajac, K. Szołtysek, J. Hanuza, IR and Raman studies of oil and seedcake extractsfrom natural and genetically modified flax seeds, Spectrochimica Acta, Part A:Molecular and Biomolecular Spectroscopy 78 (2011) 1080–1089.

[6] E.N. Frankel, A.L. Waterhouse, P.L. Teissedre, Principal phenolicphytochemicals in selected California wines and their antioxidant activity ininhibiting oxidation of human low-density lipoproteins, Journal of Agriculturaland Food Chemistry 43 (1995) 890–894.

[7] M. Corrales, S. Toepfl, P. Butz, D. Knorr, B. Tauscher, Extraction of anthocyaninsfrom grape by-products assisted by ultrasonics, high hydrostatic pressure orpulsed electric fields: a comparison, Innovative Food Science & EmergingTechnologies 9 (2008) 85–91.

[8] S. Toepfl, A. Mathys, V. Heinz, D. Knorr, Review: potential of high hydrostaticpressure and pulsed electric fields for energy efficient and environmentallyfriendly food processing, Food Reviews International 22 (2006) 405–423.

[9] C. Wang, Y. Zuo, Ultrasound-assisted hydrolysis and gas chromatography–mass spectrometric determination of phenolic compounds in cranberryproducts, Food Chemistry 128 (2011) 562–568.

[10] P. Bermejo, J.L. Capelo, A. Mota, Y. Madrid, C. Cámara, Enzymatic digestion andultrasonication: a powerful combination in analytical chemistry, TrAC –Trends in Analytical Chemistry 23 (2004) 654–663.

[11] Y. Jiao, Y. Zuo, Ultrasonic extraction and HPLC determination ofanthraquinones, aloe-emodine, emodine, rheine, chrysophanol andphyscione, in roots of Polygoni multiflori, Phytochemical Analysis 20 (2009)272–278.

[12] M.V. Martínez-Ortega, M.C. García-Parrilla, A.M. Troncoso, Comparison ofdifferent sample preparation treatments for the analysis of wine phenoliccompounds in human plasma by reversed phase high-performance liquidchromatography, Analytica Chimica Acta 502 (2004) 49–55.

[13] Y. Zuo, C. Wang, Y. Lin, J. Guo, Y. Deng, Simultaneous determination ofanthraquinones in radix Polygoni multiflori by capillary gas chromatography

coupled with flame ionization and mass spectrometric detection, Journal ofChromatography A 1200 (2008) 43–48.

[14] D. Knorr, B.I.O. Ade-Omowaye, V. Heinz, Nutritional improvement of plantfoods by non-thermal processing, Proceedings of the Nutrition Society 61(2002) 311–318.

[15] S. Kaul, G. Sharma, J. Porwal, N. Bisht, Effect of low frequency ultrasonicassisted extraction on the quality of seed oils of Indian origin, Fuel ProcessingTechnology 92 (2011) 1813–1820.

[16] J.-Y. Lin, Q.-X. Zeng, Q.I. An, Q.-Z. Zeng, L.-X. Jian, Z.-W. Zhu, Ultrasonicextraction of hempseed oil, Journal of Food Process Engineering 35 (2012) 76–90.

[17] S. Rodrigues, G.A.S. Pinto, Ultrasound extraction of phenolic compounds fromcoconut (Cocos nucifera) shell powder, Journal of Food Engineering 80 (2007)869–872.

[18] A.A. Mariod, B. Matthäus, K. Eichner, I.H. Hussein, Antioxidant activity ofextracts from Sclerocarya birrea kernel oil cake, Grasas y Aceites 57 (2006)361–366.

[19] T. Gutfinger, Polyphenols in olive oils, Journal of the American Oil ChemistsSociety 58 (1981) 966–968.

[20] D.W. Reische, D.A. Lillard, R.R. Eitenmiller, Antioxidants, in: C.C. Akoh, D.B.Min (Eds.), Food Lipids, CRC Press/Taylor & Francis Group, 2002, pp. 489–516.

[21] B. de Ancos, S. Sgroppo, L. Plaza, M.P. Cano, Possible nutritional and health-related value promotion in orange juice preserved by high-pressure treatment,Journal of the Science of Food and Agriculture 82 (2002) 790–796.

[22] I.F.F. Benzie, J.J. Strain, The ferric reducing ability of plasma (FRAP) as ameasure of ‘‘antioxidant power’’: the FRAP assay, Analytical Biochemistry 239(1996) 70–76.

[23] Z. Hromádková, Z. Košt’álová, A. Ebringerová, Comparison of conventional andultrasound-assisted extraction of phenolics-rich heteroxylans from wheatbran, Ultrasonics Sonochemistry 15 (2008) 1062–1068.

[24] M.K. Khan, M. Abert-Vian, A.-S. Fabiano-Tixier, O. Dangles, F. Chemat,Ultrasound-assisted extraction of polyphenols (flavanone glycosides) fromorange (Citrus sinensis L.) peel, Food Chemistry 119 (2010) 851–858.

[25] X. Wang, Y. Wu, G. Chen, W. Yue, Q. Liang, Q. Wu, Optimisation of ultrasoundassisted extraction of phenolic compounds from Sparganii rhizoma withresponse surface methodology, Ultrasonics Sonochemistry 20 (2013) 846–854.

[26] H. Li, L. Pordesimo, J. Weiss, High intensity ultrasound-assisted extraction ofoil from soybeans, Food Research International 37 (2004) 731–738.

[27] W. Huang, A. Xue, H. Niu, Z. Jia, J. Wang, Optimised ultrasonic-assistedextraction of flavonoids from Folium eucommiae and evaluation of antioxidantactivity in multi-test systems in vitro, Food Chemistry 114 (2009) 1147–1154.

[28] S. Hemwimon, P. Pavasant, A. Shotipruk, Microwave-assisted extraction ofantioxidative anthraquinones from roots of Morinda citrifolia, Separation andPurification Technology 54 (2007) 44–50.

[29] Y. Ma, X. Ye, Y. Hao, G. Xu, G. Xu, D. Liu, Ultrasound-assisted extraction ofhesperidin from penggan (Citrus reticulata) peel, Ultrasonics Sonochemistry 15(2008) 227–232.

[30] W. Jahouach-Rabai, M. Trabelsi, V. Van Hoed, A. Adams, R. Verhé, N. De Kimpe,M.H. Frikha, Influence of bleaching by ultrasound on fatty acids and minorcompounds of olive oil. Qualitative and quantitative analysis of volatilecompounds (by SPME coupled to GC/MS), Ultrasonics Sonochemistry 15(2008) 590–597.

[31] M.-L. Chen, D.-J. Yang, S.-C. Liu, Effects of drying temperature on the flavonoid,phenolic acid and antioxidative capacities of the methanol extract of citrusfruit (Citrus sinensis (L.) Osbeck) peels, International Journal of Food Science &Technology 46 (2011) 1179–1185.

[32] S. Settharaksa, A. Jongjareonrak, P. Hmadhlu, W. Chansuwan, S.Siripongvutikorn, Flavonoid, phenolic contents and antioxidant properties ofThai hot curry paste extract and its ingredients as affected of pH, solvent typesand high temperature, International Food Research Journal 19 (2012) 1581–1587.

[33] J. Weltin-Chanes, J.F. Vélez-Ruiz, G.V. Barbosa-Canovas, Transport Phenomenain Food Processing, CRC Press, Boca Raton, FL, 2003.

Web reference

[1] <http://www.me.mtu.edu/~jwsuther/doe2005/notes/orth_arrays.pdf> (citedon 25 July 2012).