Food Chemistry 128 (2011) 562–568

Contents lists available at ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Analytical Methods

Ultrasound-assisted hydrolysis and gas chromatography–mass spectrometricdetermination of phenolic compounds in cranberry products

Chengjun Wang a,b, Yuegang Zuo a,⇑a Departments of Chemistry and Biochemistry, University of Massachusetts Dartmouth, 285 Old Westport Road, North Dartmouth, MA 02747, USAb College of Chemistry and Materials Engineering, Wenzhou University, Wenzhou, Zhejiang 325035, China

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

Article history:Received 8 November 2010Received in revised form 27 February 2011Accepted 11 March 2011Available online 17 March 2011

Keywords:UltrasonicationHydrolysisGC–MSConjugated phenolic compoundsAntioxidantCranberry

0308-8146/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.foodchem.2011.03.066

⇑ Corresponding author. Tel.: +1 508 999 8959; faxE-mail address: yzuo@umassd.edu (Y. Zuo).

An ultrasound-assisted hydrolysis and gas chromatography–mass spectrometric (GC–MS) method hasbeen developed for determination of phenolics in cranberry products. Prior to GC–MS separation andcharacterisation, the phenolics in samples were hydrolysed by hydrochloric acid with ultrasound-assis-tance, extracted with ethyl acetate, and derivatised with N,O-bis(trimethylsilyl)trifluoroacetamide(BSTFA) + 1% trimethylchlorosilane (TMCS) reagents. The application of ultrasonication significantlyaccelerated the acidic hydrolysation of the conjugated phenolics. A baseline separation of the 20 pheno-lics and internal standard was achieved in 25 min. Standard calibration curves were linear over the con-centration range of 0.0–50 lg/mL and detection limits were 0.06–0.70 lg/mL. Twenty phenolics wereidentified in cranberry samples and all of them occurred mainly in conjugated forms. Of those, the ben-zoic acid, quercetin, and myricetin were most abundant phenolics. The total phenolics were 12.4 mg/g incranberry fruits, 9.1 mg/mL in 100% cranberry juice, and 11.1 mg/g in cranberry sauces, respectively.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

The American cranberry is a prominent agricultural food cropproduced in Massachusetts, Wisconsin, Michigan, Canada, New Jer-sey, Oregon, and Washington. The crop yield is approximately 500million pounds annually, which is processed into three basic cate-gories: fresh fruit (5%); sauce products, concentrate juice, and var-ious applications (35%); and juice drinks (60%) (Zuo, Wang, &Zhang, 2002). The phenolic components are believed to be theprincipal ingredients responsible for the beneficial health effectsof cranberry. Thus, it is important to identify and quantitate phen-olics in natural cranberry products. A number of analytical meth-ods have been proposed for the separation and determination ofthese compounds. Most of the reported methods are based on highperformance liquid chromatographic (HPLC) techniques with UVspectrophotometry. Häkkinen et al. (1999) determined flavonoids(quercetin and myricetin) and phenolic acids (p-coumaric, caffeic,ferulic, p-hydroxybenzoic, gallic and ellagic acids) simultaneouslyin cranberry fruit using HPLC–UV method. Chen, Zuo, and Deng(2001, 2007) reported on a solid-phase extraction (SPE) followedby subsequent HPLC separation to analyse phenolic compoundsand their glycosides in cranberry fruits. These authors found thatbenzoic acid, quercetin, and myricetin were the major phenoliccompounds and that phenolic compounds exist predominantly inconjugated forms, such as glycosides and ester. Recently, mass

ll rights reserved.

: +1 508 999 9167.

spectrometry (MS) has becomes a more popular detection tech-nique due to its identifying power over the UV–vis spectrophotom-etry and other detection techniques (Chen et al., 2001; Ferguson,Kurowska, Freeman, Chambers, & Koropatnick, 2004; Wang, Du,& Francis, 1978; Zuo, Wang, Lin, Guo, & Deng, 2008). The GC tech-nique enables complete separation and quantification of manycompounds and their derivatives simultaneously within compara-tively short times. Zuo’s group has developed the GC–MS methodfor simultaneous determination of phenolic compounds in cran-berry fruits and juice (Zuo et al., 2002), and the developed methodalso effectively simplified the extraction and purification processand was successfully applied to separate, characterise, and quan-tify the potential phenolic antioxidant components in cranberryproducts. However, the hydrolysis of the conjugated phenolic com-pounds was time-consuming and often requires more than 16 hduring the pretreatment process to release the conjugated pheno-lics. In the past decades, ultrasonic radiation, a type of low fre-quency (20 kHz–1 MHz) energy, has been extensively studied forimproving almost all sample preparation and pretreatment pro-cesses such as cleaning, degassing, digestion, leaching, crystallisa-tion, precipitation, and extraction (Bermejo, Capelo, Mota, Madrid,& Camara, 2004; Capote, 2006; Jiao & Zuo, 2009; Martinez-Ortega,Garcia-Parrilla, & Troncoso, 2004; Zuo et al., 2008; Zuo, Zhang, Wu,Fritz, & Rego, 2004; Zuo, Zhang, Wu, Rego, & Fritz, 2008; _Ilker Mis-tik & Müge Yükseloglu, 2005). Since Moon, Duchin, and Cooney(1979) first reported that ultrasound application lead to anincrease of biphasic hydrolysis of methyl benzoate in alkalinemedium, the capabilities of ultrasound technique to accelerate

C. Wang, Y. Zuo / Food Chemistry 128 (2011) 562–568 563

hydrolysis processes have aroused enthusiasm in many research-ers (Chen et al., 2008; Liu, Quan, Fu, Li, & Yang, 2004; Liu et al.,2008; Tuulmets and Salmar, 2001). However, no study has been re-ported on the effect of ultrasonication on hydrolysis of conjugatedphenolic compounds in cranberry products.

The aims of this study were: (i) to investigate the effects ofultrasonication on the hydrolysis process of conjugated phenolics;(ii) to provide an effective hydrolysis method which will be suit-able for release of conjugated phenolic compounds in cranberryproducts; (iii) to optimise a GC–MS method for the separation,characterisation, and quantification of phenolic compounds incranberry products.

2. Experimental

2.1. Chemicals

Phenolic and benzoic acids, m-methyl benzoic acid (internalstandard), benzoic, o-hydroxybenzoic, p-hydroxybenzoic, trans-cinnamic, m-hydroxybenzoic, p-hydroxybenzoic, p-hydroxyphe-nylacetic, o-phthalic, 2,3-dihydroxybenzoic, vanillic, o-hydroxycin-namic, 2,4-dihydroxybenzoic, p-coumaric, ferulic, caffeic, andsinapic acids, trans-resveratrol, (�)-epicatechin, catechin, querce-tin, and myricetin were purchased from Acros Organics (Spring-field, NJ). Methanol, n-hexane, and ethyl acetate were obtainedfrom Pharmcoproducts (Brookfield, NJ) and N,O-bis(trimethyl-silyl) � trifluoroacetamide (BSTFA) + 1% trimethylchlorosilane(TMCS) reagent was purchased from Supelco (Supelco Park, PA).All chemicals were of analytical reagent grade and were used with-out further purification.

2.2. Preparation of calibration standards and curves

Twenty phenolic stock standard solutions were prepared freshdaily at concentration of 1.00 mg/mL and m-methylbenzoic acid(internal standard) stock solution was prepared at 50.0 lg/mL inanhydrous methanol. The working standard solutions containing20 phenolic standards were prepared in triplicate at concentrationsof 0.00, 5.00, 10.0, 25.0, and 50.0 lg/mL by combining and dilutingthe individual stock standard solutions in methanol. In a separateanalysis, 100 lL of each of the stock standard solutions and 20 lLof internal standard stock solutions were pipetted into a 2 mLminivial and dried under a stream of nitrogen gas prior to the der-ivatisation step. Calibration curves were constructed by linearregression of the peak area ratio of individual standard to the inter-nal standard (10.0 lg/mL) versus the concentration.

2.3. Extraction of phenolic compounds from cranberry samples

All cranberry products samples were purchased from a localsupermarket including dry cranberry fruit (Ocean Spray Craisens;Middlebore, MA), 100% cranberry juice (R.W. Knudsen-Family; Orr-ville, OH) and whole berry cranberry sauce (Wegmans; Rochester,NY) and stored at 4 �C until used in this study.

2.4. Extraction of free phenolics in cranberry products

Sixty grams (60.0 g) of cranberry fruits craisens was thoroughlyground in distilled-deionised water (200 mL) with an electricalhigh-speed blender and then filtered through 8 lm of Whatmanfilter paper. One millilitre of filtrate was acidified by adding0.05 M HCl to maintain pH value at 2 and extracted with 5 mLethyl acetate three times. The combined ethyl acetate extracts(15 mL) were centrifuged and then decanted into a 20-mL glassvial followed by passing through anhydrous MgSO4 packed in a

Pasteur pipet to remove the residual water and then completelydried under a stream of nitrogen gas. Sixty grams of whole berrycranberry sauce was placed in 200 mL volumetric flask andbrought up to the mark with distilled water, then vigorously sha-ken for 30 s and filtered through 8 lm of Whatman filter paper.One millilitre of cranberry sauce filtrate or 1.00 mL of 100% cran-berry juice diluted solution was acidified, extracted, centrifuged,transferred, and dried as described above in the pretreatment ofcranberry fruit craisens samples.

2.4.1. Ultrasound-assisted hydrolysis and extraction of conjugatedphenolics in cranberry fruit

An ultrasound-assisted technique was developed for hydrolysa-tion of conjugated phenolic compounds before extraction. Twentymilligrams of ascorbic acid was dissolved in 9.00 mL distilleddeionised water in a 30 mL glass test tube and 1.00 mL of preparedfiltered cranberry juice from cranberry fruit craisens was added. Tothis solution, 5 mL of 6 N HCl were added slowly within 5 min intothe tube. Pure nitrogen gas was used to remove air in the tube andthen the tube was capped tightly. The cranberry fruit samples wereseparately prepared with eight equivalent parts. One part was usedfor conventional heating hydrolysis and others were employed forultrasonic-assisted hydrolysis trials. In traditional hydrolysis, thesample solution was heated and kept at 35 �C on heating platefor 16 h with a magnetic stirrer under the N2 atmosphere. In ultra-sonic-assisted hydrolysis trials, seven equivalent sample solutionswere individually placed in ultrasonic bath (Signature UltrasonicCleaner Model 75, 90 W, with digital timer, heat and power con-trol, VWR, South Plainfield, NJ, USA) at 35 �C for 15, 30, 45, 60,90, 120, 150 min. The temperature was maintained by heatingand water circulation. All sample solutions after hydrolysis wereallowed to cool and extracted with 5 mL ethyl acetate three times.The ethyl acetate extracts were dehydrated over MgSO4 and driedunder a stream of nitrogen gas.

2.4.2. Extraction of total phenolics after ultrasound-assisted hydrolysisin cranberry juice and sauce samples

Whole berry cranberry sauce or 100% cranberry juice samples,prepared by using 1.00 mL of cranberry sauce filtrate or 1.00 mLof 100% cranberry juice solution, was hydrolysed under ultrasonicirradiation for 90 min at 35 �C, and then extracted with ethyl ace-tate solvents, and dried under a stream of nitrogen gas.

2.5. Derivatisation of standards and samples

The dried residues of ethyl acetate extracts of all cranberry prod-ucts were dissolved in 200 lL of methanol and transferred into a2 mL mini-vial. 20 lL of m-methyl benzoic acid internal standardstock solutions were pipetted into the mini-vial and dried by astream of nitrogen. For analysis of total phenolics and some freephenolics with high concentration such as benzonic acid, quercetin,and myricetin, dilution was made before derivatisation. The calibra-tion standards and the cranberry sample residues were silylated bythe addition of 50 lL of pyridine and 50 lL of the derivatisation re-agent BSTFA + 1% TMCS with heating at 70 �C in sealed 2 mL mini-vials for 4 h. The silylated samples were analysed using GC–MS.The large excess of BSTFA + 1% TMCS ensured that the derivatisationwas complete. Care was taken to ensure anhydrous conditions dur-ing the preparation and derivatisation process because of the highsensitivity of trimethylsilyl (TMS) derivatives towards moisture.

2.6. Gas chromatography–mass spectrometry analysis

GC–MS analysis was carried out on a Hewlett–Packard (HP)model GC 5890 Series II gas chromatography equipped with anHP 5971 series mass selective detector and an HP 7673 GC

Fig. 1. GC chromatogram of TMS derivatives of phenolic compounds standards (25.0 lg/mL) with internal standard (10.0 lg/mL) using TIC mode, injection volume: 1 lL.

564 C. Wang, Y. Zuo / Food Chemistry 128 (2011) 562–568

autosampler. Samples were separated on 30 m � 0.32 mm i.d.,0.25 lm film DB-5 fused-silica capillary column (J&W Scientific,Folsom, CA, USA) on GC–MS. The column temperature was initiallyheld at 80 �C for 1 min, then programmed to 220 �C at a rate of10 �C per min, from 220 to 310 �C at a rate of 20 �C per min andwith a final hold time of 6 min. Helium was used as the carriergas and the column head pressure maintained at 13 psi. The injec-tor and detector temperature were maintained at 280 and 305 �C,respectively, and the injection volume was 1 lL in splitless mode.The interface temperature was held at 280 �C. Mass spectra werescanned from m/z 50 to 650 at a rate of 1.5 scans/s. Electron impactionisation energy was 70 eV.

2.7. Identification and quantitation

The phenolics in the cranberry samples were identified bymatching retention times and mass spectral data with the calibra-tion standards. All quantitations were performed by the method ofinternal standardisation using m-methyl benzoic acid at a concen-tration of 10.0 lg/mL. The quantitation of these compounds wasbased on the ratio of the peak area of the compound interestedover the peak area of the internal standard. All the calibration stan-

Table 1Retention times, characteristic mass ions, calibration curves, and detection limits of silyla

Standards of phenolics Retention time (min) Characteristic ions*(m/z)

Benzoic acid 6.95 194, 179, 135, 105, 77o-Hydroxybenzoic acid 10.56 267, 209, 135, 73trans-Cinnamic acid 11.04 220, 205, 161, 131, 103, 77m-Hydroxybenzoic acid 11.28 282, 267, 223, 193, 73p-Hydroxybenzoic acid 12.03 282, 267, 223, 193, 73p-Hydroxyphenyl acetic acid 12.19 296, 281, 252, 179, 164, 73o-Phthalic acid 12.70 310, 295, 221, 147, 732,3-Dihydroxybenzoic acid 13.24 355, 193, 147, 73Vanillic acid 13.57 312, 297, 267, 223, 193, 73o-Hydroxycinnamic acid 14.00 308, 293, 147, 732,4-Dihydroxybenzoic acid 14.08 355, 281, 73p-Coumaric acid 15.34 308, 293, 249, 219, 179, 73Ferulic acid 16.79 338, 323, 308, 249, 219, 73Caffeic acid 17.22 396, 381, 219, 73Sinapic acid 18.02 368, 353, 338, 249, 161, 73trans-Resveratrol 20.48 444, 429, 267, 179, 147, 73(–)-Epicatechin 21.02 368, 355, 267, 179, 73Catechin 21.12 368, 355, 267, 179, 73Quercetin 22.22 647, 559, 355, 207, 73Myricetin 22.42 647, 576, 559, 473, 207, 73m-Methyl benzoic acid (I.S.) 8.51 208, 193, 149, 119, 91

* Characteristic ions of silylated derivatives of phenolic compounds can be used to idena y Is the ratio of peak area of TMS derivatives of standards to that of internal standab Detection limits were measured as three times of the backgroundnoise.

dards and the cranberry fruit, juice, sauce samples were run intriplicate.

3. Results and discussion

3.1. Chromatographic separation and identification of phenoliccompounds

Prior to determining the phenolic compounds in cranberryproducts, the efficiency of the separation and detection of pheno-lics using the GC–MS–TIC technique was tested on a standard mix-ture of the TMS derivatives of phenolic compounds. All 20 phenolicand benzoic compounds studied in this work contain polar hydro-xyl functional group, and have relatively low volatility and are notsuitable for direct capillary GC analysis. BSFTA + 1% TMCS was cho-sen as the derivatisation reagent to convert analytes into volatiletrimethylsilyl derivatives. Pyridine, which is an aprotic organic sol-vent and a Lewis base, proved to be able to facilitate the silylationof hydroxyl groups with BSTFA + 1% TMCS since pyridine can acti-vate the hydroxyl groups and remove the acid formed in the sily-lation reaction and then drive the derivatisation of hydroxyl

ted derivatives of phenolic compounds in standards and samples by GC–MS.

Calibration curve equationa Coefficient (R2) Detection limitsb (lg/mL)

y = 0.0504x + 0.0500 0.9965 0.060y = 0.0485x � 0.0067 0.9986 0.062y = 0.0407x + 0.0576 0.9926 0.074y = 0.0481x + 0.0151 0.9989 0.062y = 0.0580x � 0.0283 0.9894 0.052y = 0.0486x + 0.0109 0.9980 0.062y = 0.0494x + 0.0667 0.9976 0.061y = 0.0504x + 0.0171 0.9992 0.060y = 0.0427x � 0.0190 0.9979 0.070y = 0.0395x + 0.0360 0.9980 0.076y = 0.0414x + 0.0016 0.9999 0.072y = 0.0364x + 0.0297 0.9985 0.082y = 0.0300x � 0.0083 0.9991 0.100y = 0.0364x � 0.0036 0.9999 0.082y = 0.0216x + 0.0013 0.9991 0.139y = 0.0245x + 0.0272 0.9950 0.122y = 0.0292x + 0.0243 0.9983 0.103y = 0.0282x + 0.0099 0.9997 0.106y = 0.0076x + 0.0012 0.9999 0.395y = 0.0043x + 0.0020 0.9990 0.697

tify the phenolic compounds.rd; x is the concentration of standards.

Fig. 2. MS spectrum of TMS derivatives of trans-cinnamic acid.

C. Wang, Y. Zuo / Food Chemistry 128 (2011) 562–568 565

group to the completeness (Zuo & Zhang, 2005). The instrumentparameters, such as the temperature program and carrier gas flowrate, were optimised to obtain good resolution between all 20 phe-nolic compounds in a short elution time. The GC–MS–TIC chro-matogram of TMS derivatives of phenolic compounds withinternal standard is illustrated in Fig. 1. A baseline separation ofthe 20 phenolic compounds and the internal standard wasachieved in 25 min using method described. The identities of the20 components were established from GC retention times and fromrecorded mass spectra. The retention times and characteristic massions of the TMS derivatives of phenolic compounds studied werelisted in Table 1. As an example, the mass spectra of silylatedtrans-cinnamic acid was illustrated in Fig. 2.

3.2. Quantitative analysis

The quantitative evaluations of the phenolic compounds werebased on peak area ratio to internal standards. Standards mixtures

0

100

200

300

400

500

600

Con

cent

ratio

ns o

f ph

enol

ics

(ug/

g) 15min 30min 45min 60min 90

Fig. 3. Total phenolic compounds in cranberry fruit samp

of the 20 phenolic compounds in the concentration range of 0.00–50.0 lg/mL were prepared for calibration curves. As presented inTable 1, all calibration curves were linear over the concentrationranges tested with correlation coefficients >0.99. The detectionlimits measured as three times the background noise were0.052–0.697 lg/mL for the TMS derivatives of phenolic com-pounds, and the detection limits increased with the increasingmolecular weight of the phenolic compounds since BSTFA + 1%TMCS derivatisation efficiency varied with the quantity and thelocation of hydroxyl functional group in phenolic compounds(Zhang & Zuo, 2005).

3.3. Effects of ultrasound on hydrolysis of conjugated phenoliccompounds

Plenty of studies have shown that the free phenolic compoundsare rarely present in plant-derived foods. More often, they occur asesters, glycosides and bound complexes (Chen & Zuo, 2007;

min 120min 150min Traditional hydrolysis

les by ultrasound-assisted and traditional hydrolysis.

566 C. Wang, Y. Zuo / Food Chemistry 128 (2011) 562–568

Fenton, Leung, & Clandinin, 1980; Krygier, Sosulski, & Hogge, 1982;Maillard & Berset, 1995). Most of the procedures described in theliterature to analyse bound phenolic compounds in food takeadvantage of acid hydrolysis to release bound phenolic compoundsusing a modified Hertog method (Chen et al., 2001; Hertog, Holl-man, & Venema, 1992; Nardini & Ghiselli, 2004; Zuo et al., 2002).Under normal conditions, almost no phenolic glycosides could behydrolysed in water due to the strong glycosidic bonds and thelow pKa values of phenolic acids. In general, strong acids used tobe added to achieve hydrolysis where water has no effect. Theacids are considered a catalyst to trigger and speed up the hydro-lysis reaction. Although our previous studies have shown that allconjugated phenolic compounds were hydrolysed under conven-tional acidic hydrolysis conditions, a long hydrolysis time of about16 h was still required to complete hydrolysis (Chen et al., 2001;Zuo et al., 2002). To examine the possible accelerative effects ofultrasound on the hydrolysis of the conjugated phenolic com-pounds, comparative studies of hydrolysis of cranberry fruits sam-ples were carried out with and without ultrasound-assistance.Ascorbic acid was added to prevent the oxidation of the phenolicacids. After hydrolysis, the released phenolic acids were extracted

Fig. 4. GC chromatograms of free and total phenolic compounds in

and analysed by GC–MS. The efficiency (measured by the meanconcentrations of the hydrolysed phenolic compounds) of theultrasound-assisted hydrolysis was directly compared with thatof the conventional hydrolysis (16 h at 35 �C). The effects of ultra-sonication time on the release of bound phenolic compounds incranberry fruit samples were presented in Fig. 3. The data indicatedthat ultrasonication of 90 min or longer resulted in similar resultsobtained by the conventional hydrolysis, and the degradation ofphenolics was not observed. It is not difficult to understand thatthe ultrasonic irradiation can accelerate the break of the conjugat-ing bond in phenolic esters, glycosides and bound complexes sincethe high energy released through implosive collapse of acousticcavitation bubble in the reaction medium, and a homogeneoussonochemical reaction is usually localised in the region of cavita-tion, that is either inside the cavity or in the liquid shell surround-ing the cavitation bubble (microreactor). As the sonochemicalacceleration of a polar homogeneous reaction presumably occursin the vicinity of collapsing bubbles, the intrinsic rate enhance-ment of the reaction in the site is most informative for the investi-gation and understanding of sonication effects (Mason & Lorimer,1988). In addition, heating employing ultrasonic radiation is based

cranberry fruit samples with internal standard (10.0 lg/mL).

Table 2Free and total phenolic compounds in fresh cranberry fruits, juice, and sauces samples.a

Identified phenolics Concentration in fruits (lg/g) Concentration in juice (lg/mL) Concentration in fresh sauces (lg/g)

Free Total Free Total Free Total

Benzoic acid 416.5 ± 32.4 5575.8 ± 363.4 234.6 ± 15.6 3256.7 ± 109.8 358.8 ± 44.9 4658.5 ± 478.4o-Hydroxybenzoic acid ND* 29.03 ± 2.32 14.12 ± 2.41 34.64 ± 3.43 2.33 ± 0.32 24.73 ± 3.53trans-Cinnamic acid 1.08 ± 0.31 18.29 ± 1.33 0.78 ± 0.14 7.68 ± 1.23 0.87 ± 0.21 9.07 ± 2.21m-Hydroxybenzoic acid 6.34 ± 0.34 14.82 ± 2.87 0.92 ± 0.21 3.72 ± 0.23 2.31 ± 0.12 6.35 ± 1.54p-Hydroxybenzoic acid 5.77 ± 0.24 27.83 ± 3.89 0.37 ± 0.09 3.82 ± 0.43 1.09 ± 0.13 12.89 ± 3.76p-Hydroxyphenylacetic acid ND* 4.95 ± 0.43 ND* 2.69 ± 0.21 ND* 3.64 ± 0.98o-Phthalic acid 6.34 ± 0.65 19.28 ± 2.43 ND* 23.96 ± 3.56 ND* 20.45 ± 4.742,3-Dihydroxybenzoic acid 1.63 ± 0.22 5.02 ± 0.89 7.99 ± 1.23 39.65 ± 4.63 0.93 ± 0.09 10.73 ± 1.89Vanillic acid 8.98 ± 0.98 33.86 ± 3.11 0.95 ± 0.76 23.99 ± 2.45 3.32 ± 0.87 43.02 ± 6.43o-Hydroxycinnamic acid 16.56 ± 1.21 92.43 ± 3.89 4.68 ± 1.01 140.68 ± 21.4 0.87 ± 0.23 90.93 ± 10.322,4-Dihydroxybenzoic acid 1.27 ± 0.11 24.32 ± 1.23 0.78 ± 0.11 7.83 ± 2.5 5.78 ± 1.09 15.68 ± 3.52p-Coumaric acid 31.63 ± 2.43 329.6 ± 23.5 10.88 ± 2.32 210.88 ± 23.6 25.86 ± 4.87 289.76 ± 33.8ferulic acid 7.92 ± 1.08 94.23 ± 10.9 3.86 ± 0.87 13.78 ± 1.78 3.23 ± 0.77 32.43 ± 5.78Caffeic acid 6.94 ± 1.11 202.5 ± 23.4 0.95 ± 0.21 40.75 ± 4.82 1.24 ± 0.13 31.64 ± 4.65Sinapic acid 0.88 ± 0.13 28.34 ± 1.32 18.87 ± 3.56 218.87 ± 34.7 0.52 ± 0.12 20.72 ± 5.23trans-Resveratrol 0.76 ± 0.11 47.03 ± 3.56 0.45 ± 0.23 40.43 ± 7.54 ND* 20.87 ± 4.34(�)-Epicatechin 48.7 ± 3.98 122.79 ± 13.67 6.12 ± 1.21 36.92 ± 3.26 35.2 ± 3.7 72.77 ± 6.23Catechin 3.62 ± 0.33 58.23 ± 7.55 2.86 ± 0.43 42.89 ± 6.87 3.66 ± 0.54 39.96 ± 4.53Quercetin 1399.7 ± 34.7 3445.8 ± 108.6 193.8 ± 16.5 2983.5 ± 187.3 689.4 ± 78.4 3675.8 ± 387.9Myricetin 789.6 ± 93.9 2108.4 ± 243.8 86.9 ± 3.4 1966.8 ± 194.3 426.8 ± 39.7 2026.3 ± 213.7

* Not detected.a Values are reported as means ± SD of triplicate analysis.

C. Wang, Y. Zuo / Food Chemistry 128 (2011) 562–568 567

on the direct effect of ultrasonic waves on molecules by dipolerotation and ionic conduction. Polar molecules absorb ultrasonicradiation strongly because they have a permanent dipole moment(Tuulmets & Raik, 1999).

3.4. Determination of free and total phenolic compounds in cranberryproducts

The separation and quantitation of phenolic compounds incranberry products are difficult because of their similar chemicalstructures and complexity of cranberry samples. A pre-cleanupand fractionation process prior to chromatographic analysis is usu-ally necessary for reliable identification and quantification. In thisstudy, phenolic compounds in cranberry juice, sauces, and fruitsamples were extracted with ethyl acetate. Total phenolic com-pounds (free + bound) in cranberry samples were determined afterhydrolysis by using developed ultrasound-assisted technique.Fig. 4 illustrates the GC separation chromatogram of the TMSderivatives of the free and total phenolic compounds in a cranberryfruit sample. The quantitative results of free and total phenoliccompounds are presented in Table 2. All 20 phenolic compoundsexcept p-hydroxyphenylacetic acid were identified in cranberryproducts occurring as free form, which is in good agreement withour previous published results (Chen et al., 2001; Zhang & Zuo,2004; Zuo et al., 2002). The total phenolic compounds, sum of freeand conjugated forms of all phenolic compounds, were 12.4 mg/gin cranberry fruits, 9.1 mg/mL in 100% cranberry juice, and11.1 mg/g in cranberry sauces, respectively. All quantitative resultsof both free and total phenolic compounds in cranberry productsindicate that only minor amounts of phenolic compounds occurin the free state in cranberry products. All phenolic compoundsare present predominantly in conjugated forms, which can be lib-erated by hydrolysis. The benzoic acid, quercetin, and myricetinwere much more abundant than other phenolic compounds, ando-hydroxycinnamic, p-coumaric, caffeic, and ferulic acids, and epi-catechin were present in moderate amounts in cranberry products.The results presented here indicate that the content of both freeand bound benzoic acid in cranberry is higher than those previ-ously reported (Chen et al., 2001; Zuo et al., 2002). This may bedue to differences in fruit source, ripeness and length of storage

time as well as differences in the procedures used for obtainingsamples. The consumption of cranberry fruit products can repre-sent a significant amount of the total dietary intake of phenolicantioxidants.

4. Conclusions

An ultrasonic-assisted hydrolysis, ethyl acetate extraction,BSTFA + 1% TMCS derivatisation followed by GC–MS separationand characterisation method has been optimised for determinationof phenolic compounds in cranberry fruits, juice, and sauces sam-ples. The results obtained in this study indicate that ultrasoundcan speed up the hydrolysis of conjugated phenolics in cranberryproducts and shorten the hydrolysation time from 16 h requiredby conventional hydrolysis technique to less than 1.5 h. The anal-ysis time is considerably reduced. Twenty phenolic compounds,including benzoic, o-hydroxybenzoic, p-hydroxybenzoic, trans-cin-namic, m-hydroxybenzoic, p-hydroxybenzoic, p-hydroxyphenyl-acetic, o-phthalic, 2,3-dihydroxybenzoic, vanillic, o-hydroxycinnamic, 2,4-dihydroxybenzoic, p-coumaric, ferulic, caf-feic, and sinapic acids, trans-resveratrol, (�)-epicatechin, catechin,quercetin, and myricetin, have been simultaneously identified andquantified on the basis of GC retention times and simultaneouslyrecorded mass spectra. The method has reported a high sensitivity,selectivity and a good repeatability and can be used as a standardmethod for the quality control of cranberry products. The methodcan also be applied in the determination of phenolics in otherplant-derived foods and in human fluids for their bioavailabilityand pharmacokinetic studies.

Acknowledgements

The authors would like to thank Drs. E. Ojadi, E. Barry, and D.Ryan for their contributions to this work. This research was partlysupported by the Massachusetts Agriculture Research Fund.

References

Bermejo, P., Capelo, J., Mota, A., Madrid, Y., & Camara, C. (2004). Enzymatic digestionand ultrasonication: A powerful combination in analytical chemistry. Trends inAnalytical Chemistry, 23, 654–663.

568 C. Wang, Y. Zuo / Food Chemistry 128 (2011) 562–568

Capote, F. (2006). Ultrasound in analytical chemistry. Analytical and BioanalyticalChemistry, 387, 249–257.

Chen, L., Li, B., Li, D., Gan, J., Jiang, W., & Kitamura, Y. (2008). Ultrasound-assistedhydrolysis and acidogenesis of solid organic wastes in a rotational drumfermentation system. Bioresource Technology, 99, 8337–8343.

Chen, H., & Zuo, Y. (2007). Identification of flavonol glycosides in Americancranberry fruit. Food Chemistry, 101, 1357–1364.

Chen, H., Zuo, Y., & Deng, Y. (2001). Separation and determination of flavanoids andother phenolic compounds in cranberry juice by high-performance liquidchromatography. Journal of Chromatography A, 913, 387–395.

Fenton, T. W., Leung, J., & Clandinin, D. R. (1980). Phenolic components of rapeseedmeal. Journal of Food Science, 45, 1702–1705.

Ferguson, P. J., Kurowska, E., Freeman, D. J., Chambers, A. F., & Koropatnick, D. J.(2004). A flavonoid fraction from cranberry extract inhibits proliferation ofhuman tumor cell lines. Journal of Nutrition, 134, 1529–1535.

Häkkinen, S., Heinonen, M., Kärenlampi, S., Mykkänen, H., Ruuskanen, J., & Törrönen,R. (1999). Screening of selected flavonoids and phenolic acids in 19 berries. FoodResearch International, 32, 345–353.

Hertog, M. G., Hollman, C. H., & Venema, D. P. (1992). Optimization of a quantitativeHPLC determination of potentially anticarcinogenic flavonoids in vegetablesand fruits. Journal of Agricultural and Food Chemistry, 40, 1591–1598.

_Ilker Mistik, S., & Müge Yükseloglu, S. (2005). Hydrogen peroxide bleaching ofcotton in ultrasonic energy. Ultrasonics, 43, 811–814.

Jiao, Y., & Zuo, Y. (2009). Ultrasonic extraction and HPLC determination ofanthraquinones, aloe-emodine, emodine, rheine, chrysophanol, and physcionein Radix Polygoni multiflori. Phytochemical Analysis, 20, 272–278.

Krygier, K., Sosulski, F., & Hogge, L. (1982). Free, esterified, and insoluble-boundphenolic acids. 1. Extraction and purification procedure. Journal of Agriculturaland Food Chemistry, 30, 330–334.

Liu, Z., Quan, X., Fu, H., Li, X., & Yang, K. (2004). Effect of embedded-silica onmicrostructure and photocatalytic activity of titania prepared by ultrasound-assisted hydrolysis. Applied Catalysis B: Environmental, 52, 33–40.

Liu, Y., Jin, Q., Shan, L., Liu, Y., Shen, W., & Wang, X. (2008). The effect of ultrasoundon lipase-catalyzed hydrolysis of soy oil in solvent-free system. UltrasonicsSonochemistry, 15, 402–407.

Maillard, M. N., & Berset, C. (1995). Evolution of antioxidant activity during kilning:Role of insoluble bound phenolic acids of barley and malt. Journal of Agriculturaland Food Chemistry, 43, 1789–1793.

Martinez-Ortega, M., Garcia-Parrilla, M., & Troncoso, A. (2004). Comparison ofdifferent sample preparation treatments for the analysis of wine phenolic

compounds in human plasma by reversed phase high-performance liquidchromatography. Analytica Chimica Acta, 502, 49–55.

Mason, T. J., & Lorimer, J. P. (1988). Sonochemistry: Theory, applications and uses ofultrasound in chemistry. Chichester: Ellis Horwood.

Moon, S., Duchin, L., & Cooney, J. (1979). Application of ultrasound to organicreactions: Ultrasonic catalysis on hydrolysis of carboxylic acid esters.Tetrahedron Letters, 41, 3917–3920.

Nardini, M., & Ghiselli, A. (2004). Analytical, nutritional and clinical methods:Determination of free and bound phenolic acids in beer. Food Chemistry, 84,137–143.

Tuulmets, A., & Salmar, S. (2001). Effect of ultrasound on ester hydrolysis in aqueousethanol. Ultrasonics Sonochemistry, 8, 209–212.

Tuulmets, A., & Raik, P. (1999). Ultrasonic acceleration of ester hydrolyses.Ultrasonics Sonochemistry, 6, 85–87.

Wang, P., Du, C., & Francis, F. (1978). Isolation and characterization of polyphenoliccompounds in cranberries. Journal of Food Science, 43, 1402–1404.

Zuo, Y., Wang, C., Lin, Y., Guo, J., & Deng, Y. (2008). Simultaneous determination ofanthraquinones in radix Polygoni multiflori by capillary gas chromatographycoupled with flame ionization and mass spectrometric detection. Journal ofChromatography A, 1200, 43–48.

Zuo, Y., Wang, C., & Zhang, J. (2002). Separation, characterization and quantitation ofbenzoate and phenolic antioxidants in American cranberry fruit by GC–MS.Journal of Agricultural and Food Chemistry, 50, 3789–3794.

Zhang, K., & Zuo, Y. (2005). Pitfalls and solution for simultaneous determination ofestrone and 17a-ethinylestradiol by gas chromatography–mass spectrometryafter derivatization with N,O-bis(trimethylsilyl)trifluoroacetamide. AnalyticaChimica Acta, 554, 190–196.

Zhang, K., & Zuo, Y. (2004). GC–MS determination of flavonoids and phenolic andbenzoic acids in human plasma after consumption of Cranberry juice. Journal ofAgricultural and Food Chemistry, 52, 222–227.

Zuo, Y., & Zhang, K. (2005). Discussion: Suitability of N,O-bis(trimethylsilyl)trifluoroacetamide as derivatization reagent for the determination of theestrogens estrone and 17a-ethinylestradiol by gas chromatography–massspectrometry. Journal of Chromatography A, 1095, 201–202.

Zuo, Y., Zhang, K., Wu, J., Rego, C., & Fritz, J. (2008). An accurate and non-destructiveGC method for determination of cocaine on US paper currency. Journal ofSeparation Science, 31, 2444–2450.

Zuo, Y., Zhang, L., Wu, J., Fritz, J., & Rego, C. (2004). Ultrasonic extraction andcapillary gas chromatography determination of nicotine in pharmaceuticalformulations. Analytica Chimica Acta, 526, 35–39.