Automatic method for determination of total antioxidant capacity using 2,2-diphenyl-1-picrylhydrazyl assay
Post on 26-Jun-2016
Analytica Chimica Acta 558 (2006) 310318
Automatic method for determination of tlhale
ber 2er 200
In the pre on antotal antioxi ds pr2,2-dipheny the cantioxidant compounds monitored spectrophotometrically at 517 nm.
The influence of initial DPPH concentration and sample dilution in the present methodology was studied. It was verified that the amount ofDPPH consumed by antioxidant standards (ascorbic and caffeic acids) was independent of the initial concentration of radical except for situationswhere DPPH/antioxidant molar ratio was lower than the stoichiometric value. Furthermore, the sample dilution factor plays an important rolefor achieving results comparable to those from end-point batch method since the exhausting of scavenging ability of the sample should take placeduring the p
The propocapacity (VCcomparableabout 13 h1 2005 Else
There itive damagin ageingatherosclerthermore, tthat the actbeverages,tect againstective effedants specicompounds
0003-2670/$doi:10.1016/jeriod of absorbance measurement.sed method was applied to several food products and the total antioxidant capacity was expressed as Vitamin C equivalent antioxidantEAC). The results obtained by the proposed method ranged from 1.1 to 318 mg of ascorbic acid/100 ml and they were statistically
to those provided by the batch method. The detection limit was 0.34 mg of ascorbic acid/100 ml and the determination frequency waswith an excellent repeatability (R.S.D. < 1%, n = 10).
vier B.V. All rights reserved.
tal antioxidant capacity; DPPH; Multi-syringe flow injection; Beverages
s recent evidence that free radicals induce oxida-e to biomolecules. This damage has been implicatedand in several human pathologies such as cancer,osis, rheumatoid arthritis and other diseases . Fur-here is a considerable amount of studies indicatingive dietary constituents of fresh fruit, vegetables andprevent these free radical-induced diseases and pro-t foodstuff oxidative deterioration . These pro-cts have been attributed, in large part, to the antioxi-es (Vitamins C and E, carotenoids and polyphenolic) which scavenge free radicals [5,6].
ding author. Tel.: +351 22 2078994; fax: +351 22 2004427.dress: firstname.lastname@example.org (M.A. Segundo).
Several methodologies, based on free radical capture orformation suppression, are used to measure the antioxidantcapacity of biological material and model compounds [7,8].The most commonly used for their ease, speed and sensitivityare those involving chromogen compounds of a radical natureto simulate radical oxygen and nitrogen species. The mostwidely used assays are based on the scavenging of radical cation2,2-azinobis-(3-ethylbenzothiazoline-6-sulphonate) (ABTS+assay)  or of the stable radical 2,2-diphenyl-1-picrylhydrazyl(DPPH assay) [10,11]. The presence of antioxidant speciesleads to the disappearance of these radical chromogens whichcan be followed by spectrophotometric methods.
Recently, the DPPH assay was implemented using automaticmethods based on flow injection analysis (FIA) [12,13], sequen-tial injection analysis (SIA)  or HPLC-FIA [15,16], that wereapplied for screening and evaluation of scavenging capacity ofseveral pure compounds and complex matrices such as plantextracts and beverages. In the HPLC-FIA method, the separated
see front matter 2005 Elsevier B.V. All rights reserved..aca.2005.11.013using 2,2-diphenyl-1-picryLus M. Magalhaes, Marcela A. Segundo , S
REQUIMTE, Servico de Qumica-Fsica, Faculdade deRua Anbal Cunha, 164, 4099-030 P
Received 10 October 2005; received in revised form 3 NovemAvailable online 19 Decemb
sent work, an automatic method based on multi-syringe flow injectidant capacity, measured as the cumulative capacity of the compounl-1-picrylhydrazyl (DPPH) reaction. The determination is based onotal antioxidant capacityydrazyl assaytte Reis, Jose L.F.C. Lima
acia, Universidade do Porto,Portugal005; accepted 4 November 20055
alysis (MSFIA) was developed for the determination ofesent in the sample to scavenge free radicals, using theolour disappearance due to the scavenging of DPPH by
L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318 311
analytes react postcolumn with the DPPH solution , and theinduced scavenging is detected as a negative peak. These meth-ods, combicapacity, pas bioassaya time conloss in activisolation an
Howevematrices dopropertiesthe synergidants implicompound. For thallowed thethat enablethe matrix.
Recentlflow analysanalysis (Min order toflexibility othe objectiautomaticcapacity exanalysis.
The meantioxidantabsorbancein order toconsumptioing. Furtheand sample
All chemfurther purfrom Milliwere used.
2,2-Dipwere purchobtained fr(5.80 10appropriateand protect
For the fl1.45 10stock solutto 200 ml wand protectas carrier s
Ascorbimol l1) w
of the respective solid in ethanol solution 50% (v/v). Workingstandard solutions containing either ascorbic or caffeic acid in
ncen 4 1by din 50the
.25 ng stioned arderesta
s weintog duore imetho attnd beese ss wemple3 andete
n [21stoc/v) inm.
is apes inothewer
ge osigneflower vae duver-
werening separation and evaluation of radical-scavengingresent a major advantage relatively to batch methods-guide fractionation of natural or food samples is
suming and labour-intensive process. Moreover, theity for antioxidants due to decomposition during thed purification procedures is avoided.r, the presence of antioxidant compounds in complexes not necessarily imply that the same antioxidant
are verified in the whole sample. On the other hand,stic effect that may exist between different antioxi-es that the sum of antioxidant capacities from eachisolated may not exactly reflect the overall actionis reason, automatic methods based on FIA or SIAevaluation of antioxidant capacity in an environment
d the interactions between all compounds present in
y, novel computer controlled techniques for automaticis were reported, namely multi-syringe flow injectionSFIA). This technique was proposed in 1999 ,
combine the multi-channel operation of FIA and thef multi-commutation flow systems . Therefore,
ve of the present work was the development of anflow system for the assessment of total antioxidantploiting the features introduced by MSFIA in flow
thodology was based on consumption of DPPH byspecies present in the sample by monitoring theat 517 nm. A stopped flow approach was chosen
follow the reaction development and to assess if totaln of antioxidant is attained during reaction monitor-
rmore, the influence of initial DPPH concentrationdilution are also addressed in the present work.
nts and solutions
icals used were of analytical-reagent grade with noification. For the preparation of all solutions water-Q system (resistivity > 18 M cm) and ethanol p.a.
henyl-1-picrylhydrazyl (DPPH) and ascorbic acidased from Sigma (St. Louis, USA). Caffeic acid wasom Aldrich (Milwaukee, USA). A stock solution4 mol l1) of DPPH was prepared by dissolving theamount in ethanol. This solution was kept at 4 C
ed from light, and it was stable during a week.ow system, the DPPH working solution containing
4 mol l1 was prepared by measuring 50 ml of theion, 50 ml of ethanol and the volume was made upith water. This working solution was prepared dailyed from light. Ethanol solution 50% (v/v) was usedolution.c and caffeic acid stock solutions (1.00 102ere prepared by dissolving the appropriate amount
In oon theconcen
The c0.375 0.840,
Befbatchorder twine aand thsampleThe sato 1:33
ForHansefrom a50% (vat 428
Solof a mFor thsyringin thevalvesexchanwas as
for thethe oththe timavoid o
A pQuickBation (the po
Asspectroa flow-heim/Bmentstration range 0.2510.00 10 mol l were pre-lution of the respective stock solution using ethanol% (v/v). These solutions were prepared daily.analysis using the batch method, a DPPH solu-104 mol l1) in ethanol (50%, v/v) was prepared.
andard solutions containing ascorbic acid in the con-range 0.252.00 104 mol l1 were prepared asbove.to evaluate the influence of DPPH concentrationblishment of calibration curves, different DPPHons were prepared in ethanol solution 50% (v/v).ntrations studied were 1.500, 0.938, 0.750 and4 mol l1 providing initial absorbance values of8, 0.426 and 0.208, respectively.ples were purchased at local markets. The green teare prepared by pouring 200 ml deionised water ata glass with tea bag (1.491.77 g of leaves) and byring 3 min .ntroduction into the flow system or analysis using theod, samples were first diluted 50% using ethanol inain a final concentration of 50% (v/v) in ethanol. Forer samples, the initial alcohol content was consideredamples were diluted with ethanol and water. Somere further diluted using ethanol solution 50% (v/v).dilutions used for the flow system varied from 1:2
d for the batch method varied between 1:2 and 1:200.rmination of dispersion coefficient of Ruzicka and], a bromothymol blue (BTB) solution was preparedk solution (0.50 g l1) by dilution in ethanol solutionn order to provide an absorbance value of about 0.520
s were propelled through the flow network by meanssyringe burette (Crison Instruments, Allela, Spain).plication, the multi-syringe was equipped with 5 mlpositions 1 and 2 while 10 ml syringes were placedr two positions (Fig. 1). Three extra commutatione included in the module used. For all valves, theptions were classified in on/off lines. The off lined to the solution flasks and the on line was reservednetwork in the valves placed at the multi-syringe. Forlves, the positions on/off were chosen to minimisering which the valves were switched on in order toheating problems.nal computer, running lab-made software written in
4.5 (Microsoft), controlled the multi-syringe oper-ber of steps and direction of piston displacement and
of all commutation valves).ction system, a Jenway 6105 (Essex, UK) UVvistometer equipped with a thermostatic cell holder andugh cell from Hellma (80l, ref. 178.710-QS, Mull-n, Germany) was used and the absorbance measure-carried out at 517 nm. The cell holder was connected
312 L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318
Fig. 1. MSFIA manifold used for the determination of total antioxidant capacityusing DPPH assay: MS, multi-syringe; Si, syringes; Vi, commutation valves(solid and dotted lines represent the position on and off, respectively); B1 and B2,confluences; HC, holding coil (200 cm); RC, reaction coil (120 cm); D, detector;C, carrier (ethanol solution 50%, v/v); R, 2,2-diphenyl-1-picrylhydrazyl reagentprepared in Cputer; W, was
to a Tectronacquisitionat 3 Hz, usflow systemsoftware.
The syscally in Figbridge, UKBel, Franceacrylic Y-s
The conwere 200 cThe tubingThe connethe same le
B2 was 50 cm long while the connection between the valve V7and confluence B2 was 10 cm long. The reaction coil (RC) was120 cm long.
The protocol sequence for the determination of total antiox-idant capacity using DPPH assay is listed in Table 1. Beforestarting the analytical procedure, syringe 4 was filled with car-rier solution and further propelled to detection system. Thus, theflow-through cell was filled with it and the absorbance signalwas adjusted to zero. After that, syringe 3 was filled with radicalsolution while the syringes 1 and 2 were filled with carrier solu-tion and further propelled to the detector. The absorbance valuemeasured corresponded to the absorbance of radical solution inthe absence of antioxidant compounds.
The procedure included six steps. The first step consistedof filling syringes with the respective solutions. Then, 50l ofstandard/sample was aspirated into the holding coil (HC). Aftera dummy step, applied to change the flow direction , thesample, carrier and reagent were sent towards the detection sys-tem. During this step the sample plug was pushed by carrier
conflyring solled uoppe0 s (s wa
the vn posolu
Table 1Protocol sequ
2 Sa3 Du
The indicated; AO, antioxidant standard solution or sample; PC, personal com-te.
S-543 (Altrincham, UK) thermostatic bath. The datawas performed through a PCL-711B interface card
ing the same software developed for controlling the. The data obtained were analysed using Origin 6.1
A manifold and procedure
tem components were arranged as shown schemati-. 1. All connections were made from Omnifit (Cam-) PTFE tubing (0.8 mm i.d.) with Gilson (Villiers-le-) end-fittings and connectors. Two laboratory-made
haped connectors were used as confluences.nections between the multi-syringe and the valve V7m long. The holding coil (HC) had the same length.
up tofrom sDPPHpropelwas sting 18cell wathe baa new
Forsame psteps,were icarrierperfor
Forflow cculatedlength between the valves V5 and V6 was 4 cm long.ction between this valve and the confluence B1 hadngth. The connection between confluences B1 and
solution inrespective58 mol1 l
ence for the determination of total antioxidant capacity using DPPH assay
scription Position of the commutation valves
1 2 3 4 5 6
ringes are filled with the respectiveutions
F F F F F F
mple is aspirated N F F F N Fmmy step to change the flowection
F F F F F F
mple, carrier and reagent are sentards detection system
N N N F F F
w stop for reaction monitoringth data acquisition
F F F F F F
rrier and reagent are sent to washsystem
N N N F F F
values for volume refer to syringe 3 (10 ml). N and F represent the positions on anduence B1, where it was diluted by carrier solutione 2. Subsequently, the diluted sample was mixed withution, after confluence B2. This mixture was furtherntil it reached the detection system. Then, the flowd and the absorbance at 517 nm was measured dur-temperature = 25 1 C). Finally, the flow-throughshed with carrier and reagent solution to re-establishe. After this last step, the flow system was ready forytical cycle.determination of intrinsic absorption of sample thedure was performed, except for steps 4 and 6. In thesealve V3 was in position off and the valves V4 and V7sition on in order to replace the DPPH solution bytion. Moreover, the absorbance measurements wereduring 30 s instead of 120 s.determination of DPPH/AO molar ratio in the
he DPPH concentration in the flow cell was cal-the ratio between absorbance value of radicalthe absence of antioxidant compounds and the
molar absorption coefficient at 517 nm ( = 11 071 cm1). The antioxidant concentration in the flow cell
Volume (l) Time (s)7
F 3000 22.50
F 100 3.00F 500 3.75
F 750 15.00
F 1850 37.00
L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318 313
was calculated as the ratio between the concentration of the solu-tion introduced in the system and the dispersion coefficient.
The DPwas applieto a micropThe absorbHence, 50of DPPHradical absof ethanoldant soluti200l ofsample. Thevery 20 sto 0.003 unquadruplic
3.1. Develtotal antiox
The conallow the dsamples usmetric deteenced bysample a sabsorbancetion was asof DPPHabsorbancecompounds
In multisample intowould takeeffect. Hento provide aThe valvedisturbingbeen necesthe baselinsumption.
As someof detectionmeasureme
to allow thorder to pesample. Inmixed withposition of
The conthe sampleto allow th
the content of the syringe 2. The reaction coil (RC) disposed inknitted form was set at 120 cm long.
dies ione pased aaphsthe
), pre dubleas hamentnoled i. Thd inum)s the, 100ientrminwere
s redse. Fntioxthe
lue wd DPFig.ivenowkinentinreme
ancesorbas (A0oxidoint batch method
PH method described by Brand-Williams et al. d with some modifications  and it was adaptedlate reader (Synergy HT, Bio-Tek, Vermont, USA).ance measurements were carried out at 517 nm.l of ascorbic acid solution or sample and 200lsolution were placed in each well. To evaluate theorbance in the absence of antioxidant species, 50lsolution 50% (v/v) was added in place of antioxi-ons. To evaluate the intrinsic absorption of sample,ethanol solution 50% (v/v) was added to 50l ofe reduction in DPPH concentration was monitoreduntil the absorbance decrease was inferior or equalits min1 . All experiments were performed in
ate and the temperature was kept at 25.0 0.1 C.
opment of the MSFIA system for determination ofidant capacity
figuration of the MSFIA system was designed toetermination of total antioxidant capacity in severaling DPPH scavenging reaction and spectrophoto-ction. Since the reaction kinetics is strongly influ-the type of antioxidant compound present in thetopped flow approach was implemented. Hence, thewas monitored during 180 s. The DPPH consump-sessed from the difference between the absorbancein the absence of antioxidant compounds (A0) andof DPPH measured in the presence of antioxidantafter a fixed period of time (Af).
-syringe flow systems it is not feasible to introduce thethe system through one of the available syringes, as ita long time of washing steps for avoiding carry-overce, in the present system the valve V5 was includedccess to the antioxidant standard solution or sample.
V6 was included to allow sample exchange withoutthe content of the flow cell. Otherwise, it would havesary to send DPPH solution in order to re-establishe after sample exchange, increasing the reagent con-
samples can absorb radiation at the same wavelength, the manifold was adapted to accommodate in-linent of sample blank. Therefore, valve V7 was includede replacement of the DPPH solution by carrier, inrform the determination of intrinsic absorption ofthis way, after confluence B2 the sample plug waseither DPPH or carrier solution depending on the
valve V7 (off or on, respectively).fluence B1 was included in the manifold to increasedilution. Moreover, this confluence could be used
e adjustment of pH or ionic strength by replacing
Stucentratof thestions uparagr
For200lthe timtor (Taflow wexperiin ethadescrib428 nmselectemaximtoward25, 50coefficto detefoundvolum
Twotion instudiedintermset at0.813 centrat10.00 180 s asolutio
Theidant cof absopounddecreawith a
Forthe vaacid anplace (by a gafter flslowerues co
establias a fuabsorbthe abpoundof anticoncerning the sample volume and the DPPH con-were carried out in order to evaluate the influencerameters in the analytical performance. The condi-nd the results obtained are discussed in the following.different sample volumes tested (25, 50, 100 and
eliminary studies were performed in order to adjustring which solutions were sent towards the detec-
1, step 4). This was necessary to guarantee that thelted after maximum absorbance was achieved. Theses were performed using bromothymol blue dissolvedsolution 50% (v/v), as sample, and the proceduren Table 1 was applied. The wavelength was set ate time applied in step 4 for each sample volume wasorder to attain at least 90% of the highest signal (peak. Therefore, the time during which solutions were sent
detector was set to 14.5, 15.0, 16.5 and 19.0 s forand 200l, respectively. Moreover, the dispersion
 for each sample volume was calculated in ordere the sample dilution inside the system. The values44.2, 22.3, 13.0 and 7.8 for the above-mentioned
ioxidant compounds with similar DPPH consump-ar basis, but with different kinetic behaviours werecorbic acid (rapid kinetics) and caffeic acid (rapid-te kinetics) . The DPPH concentration was 104 mol l1 providing an absorbance value of05 after dilution inside the flow system. The con-of antioxidant standards used varied from 0.25 to4 mol l1. The absorbance values obtained duringflow halting for 50l of ascorbic and caffeic acid
re depicted in Fig. 2.orbance of radical solution in the absence of antiox-ounds (Fig. 2A and F) was stable during the periodce measurement. As expected, the antioxidant com-uced the radical DPPH, resulting in absorbanceurthermore, the amount of radical reduced increasedidant concentration.period during which the absorbance was monitored,as constant because the reaction between ascorbicPH occurred before absorbance measurement took
2BE). Therefore, the absorbance decrease provokedascorbic acid concentration was equal at any time
halting. On the other hand, as caffeic acid presentstic behaviour than ascorbic acid, the absorbance val-
ued to decrease during the interval of absorbancent for some of the solutions tested (Fig. 2HJ). Sim-were obtained for other sample volumes studied.h sample volume studied, a calibration curve wasby plotting the absorbance decrease ( absorbance)on of antioxidant concentration (R 0.9991). Thedecrease was calculated from the difference betweennce of DPPH in the absence of antioxidant com-) and absorbance of DPPH measured in the presenceant compounds after 120 s (Af).
314 L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318
Fig. 2. Absor50l of ascofixed at 1.50104 mol l1:
It shoulantioxidantreaction maless, such sas they wouby the othetrations.
The sencurve (abscompound)obtained fvalue obtaof 50l wof reagent6) was inf(100 and 2enhanced.
Using thence of inabsorbanceusing diffe
Table 2Values of absorbance decrease obtained for ascorbic acid for different initial
ration of DPPH (n = 3; S.D. 0.004)ic aciol l1
The valthe flow
The vaalso in
Theindepesituatithan thThis oway obance measurements during 180 s after flow halting obtained forrbic and caffeic acid solutions. The DPPH concentration was 104 mol l1. The antioxidant concentration is expressed in(A, F) 0.00; (B, G) 1.00; (C, H) 2.00; (D, I) 4.00; (E, J) 6.00.
d be emphasised that for higher concentration ofwith slow kinetic behaviour, such as caffeic acid, they not be completed within 120 s (Fig. 2J). Neverthe-
ituations can be easily spotted in the calibration curveld originate points deviated from the line establishedr points corresponding to lower antioxidant concen-
sitivity, estimated through the slope of calibrationorbance decrease per concentration of antioxidant
increased with sample volume. In fact the valuesor 25, 50 and 100 were 17, 36 and 60% of theined for 200l. Nevertheless, the sample volumeas selected for further studies because the amountexpended to re-establish the baseline (Table 1, steperior to that obtained for larger sample volumes00l). Moreover, the determination frequency was
e experimental conditions described above, the influ-itial concentration of DPPH was assessed. Thedecrease values obtained for ascorbic acid solutions
rent concentrations of DPPH are listed in Table 2.
ture it is uas percenttion or qabsorbanceabsorbanceDPPH, thbe differenple, for anabsorbancecentration,21.2, 41.1 a0.750 andwhen evalurate to useof radicalof radicalenging abimolar ratvalue.
The prototal antiojuices, isowas expres(VCEAC)(mg) presenutritional100 g .decrease afmin C) wation variedd])
[DPPH] (104 mol l1)1.500 0.750 0.375
0.094 (16.9) 0.093 (8.6) 0.092 (4.2)0.135 (11.3) 0.133 (5.7) 0.133 (2.8)0.178 (8.5) 0.175 (4.3) 0.173 (2.1)0.367 (4.2) 0.359 (2.1) 0.196 (1.0)a0.560 (2.8) 0.388 (1.4)a 0.197 (0.7)a
parenthesis correspond to the DPPH/antioxidant molar ratio in
tioxidant molar ratio lower than the stoichiometric value.
of DPPH/antioxidant molar ratio in the flow cell areted.PH consumption by antioxidant compound wast of the initial concentration of radical except forwhere DPPH/antioxidant molar ratio was loweroichiometric value (which is 2 for ascorbic acid).vation may have some implications concerning thetioxidant properties are expressed. In the litera-sually found the expression of antioxidant capacityage of consumed DPPH, referred to as inhibi-uenching, which is calculated as the ratio between
decrease and initial absorbance of DPPH. Asdecrease is independent of initial absorbance of
e values of percentage of consumed DPPH willt for the same amount of antioxidant. For exam-tioxidant concentration of 2.00 104 mol l1 thedecrease was similar, independently of radical con-but the percentage of inhibition or quenching wasnd 83.2% when the DPPH concentration was 1.500,0.375 104 mol l1, respectively. For this reason,ating the total antioxidant capacity, it is more accu-the absorbance variation rather than the percentageconsumed. It is also important to assure an excessDPPH with the intention of exhausting the scav-lity of the antioxidant, i.e. the DPPH/antioxidantio should be superior to the stoichiometric
ation of the MSFIA method and its application toposed MSFIA method was applied to determine thexidant capacity of several samples including fruittonic and soft drinks, tea, beers and wines. Thissed as Vitamin C equivalent antioxidant capacity
calculated as the equivalent amount of ascorbic acidnt in 100 ml of sample. These units were chosen sincelabelling on food products are indicated/100 ml orTherefore, a calibration curve relating the absorbanceter 120 s and the concentration of ascorbic acid (Vita-s established (R 0.9996). The standard concentra-between 0.44 and 10.57 mg of ascorbic acid/100 ml
L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318 315
Fig. 3. Profileduction into tdark beer 1:2wine A 1:333
(corresponabsorbanceascorbic acVitamin Cby the resp
Some saabsorb at tsituations,underestimpresent in tcontributioFor this reacalculatingwork, sampfor all samtial DPPHwhether themination.
The abssome of thples such athe mainly
scavenged and the absorbance value stabilised after a few sec-onds of absorbance measurement. However, for samples such as
Fig.sibleundshe re, asthelly pdurVC
ed Msultsvalue scaanceresp
provued arefouenhe retratistedwine (responcompocases tsloweddue tooriginaformed
Theproposilar restablethat thabsorbkineticthosecontin
Thethe inflattain tconcen
ples teof DPPH consumption for some samples, diluted prior to intro-he flow system: (A) orange juice A 1:2; (B) green tea 1:50; (C)0; (D) white wine 1:10; (E) ethanol solution 50% (v/v); (F) red; (G) red wine A 1:100; (H) red wine A 1:50.
ding to 0.25 and 6.00 104 mol l1). Thus, thedecrease obtained for samples was related to that of
id and the total antioxidant capacity was expressed asequivalents in mg/100 ml. This result was multipliedective dilution factor.mples can absorb or originate compounds that alsohe wavelength of measurement (517 nm). In boththe absorbance decrease of DPPH solution can beated . On the other hand, the absorbing specieshe sample can also react with DPPH, changing itsn to the total absorbance measured along the reaction.son, the sample blank should not be included whenDPPH consumption. Nevertheless, in the presentle blank was measured and values
316 L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318
Table 3VCEAC values (mg of ascorbic acid/100 ml) obtained by MSFIA and by batch method (after 2 min of reaction and at end-point) applying different dilution factorsSample Dilution factor MSFIAa (2 min) Batchb (2 min) Batchb (end-point) End-point time (min)Isotonic drink A 10 10.8 0.2 9.7 0.3 10.0 0.3 5
6.6 10.1 0.1 9.5 0.1 9.7 0.1 55 10.1 0.2 9.7 0.1 9.9 0.2 5
Soft drink 40 31.9 0.7 25.6 1.3 27.4 1.3 520 28.0 1.1 23.7 0.6 26.8 0.4 1013.3 27.7 0.4 22.5 0.6 27.7 0.3 20
Red Wine A 333 215 6 155 16 221 16 15200 199 6 140 11 222 9 25100 165 1 131 6 217 4 40
50 148 3 Each value corresponds to the mean S.D.
a (n = 3).b (n = 4).
where DPP are the fithat form tto the fastDPPH1 anfor the simexponentia(DPPH1) aDPPHr isafter antiox
The DPence betweor from theand that coconcentratithe experimthree sampand the absand DPPHlated (Tablend-point bDPPH coverified whwhich provmay be du3 min of repoint. TheDPPH2 v
ding9]. Fon liad toof
Table 4VCEAC valu
White wineRed wine ARed wine B
Each value coa (n = 3).b (n = 4).H is the radical concentration at any time, and andrst-order rate constants for the simple uniexponentialhe double exponential equation, one correspondingstep and the other to the slow step. The parametersd DPPH2 are the radical concentrations at time zerople uniexponential equations that form the doublel, and represent the DPPH consumed in the fast stepnd that consumed in the slow step (DPPH2). Finally,the remaining DPPH concentration in the mediumidant depletion.
PH consumption can be calculated from the differ-en the initial concentration of DPPH and DPPHrsum of DPPH consumed in the fast step (DPPH1)nsumed in the slow step (DPPH2). Since DPPHon values were proportional to absorbance values,ental data (absorbance versus time) obtained for
les were fitted by non-linear regression (R2 > 0.996),orbance values corresponding to DPPHr, DPPH1,2 were obtained. The VCEAC values were calcu-e 4) and similar values to those obtained by theatch method were attained using estimate values for
nsumption in the fast and slow steps. This was noten using the estimate value for remaining DPPH,ided an underestimated consumption. This situatione to the utilisation of data concerning only the first
Theand byantioxIn ordogy, atwo m(0.02in par(R = 0.and in0, respestimaconsec
Theresponsy/x [2detectiples htration0.34 m
action while it took about 35 min to reach the end-refore, we propose the application of DPPH1 andalues for calculation of VCEAC when samples do
cycle is notperformedthe multi-s
es (mg of ascorbic acid/100 ml) obtained for samples presenting slow kinetic behavioMSFIAa
Single point (120 s) Kinetic (DPPHr) Kin13.0 0.3 (1:20) 21.3 1.7 (1:10) 24.8215 6 (1:333) 187 10 (1:50) 226183 9 (1:200) 209 11 (1:133) 318
rresponds to the mean S.D.t its scavenging capacity within 3 min of reaction.lts obtained by the proposed methodology (CMSFIA)batch method (Cbatch) for the determination of totalcapacity in all samples are summarised in Table 5.evaluate the accuracy of the proposed methodol-
ar relationship between the results provided by theds was established. The equation CMSFIA = 1.009batch 0.9 (2.9) was obtained, where the valuesses are the limits of the 95% confidence intervals). From these data, it is clear that the calculated slopept do not differ significantly from the values 1 andely . The precision of the MSFIA method wasy calculating the relative standard deviation from 10
e determinations of two standard ascorbic acid solu-and 10.57 mg/100 ml), providing values of 0.9 and
ectively.ection limit was calculated as the concentration cor-
to the intercept value plus three times the statisticor five different calibration curves, the calculatedmit was about 0.17 mg/100 ml. However, as the sam-
be diluted 1:2 in order to attain a final concen-ethanol 50% (v/v), the detection limit was in fact0 ml.ring that the time required for a complete analytical
merely the summation of the time taken for each stepbecause data transference between the computer andyringe must also be accounted, the whole analytical
ur using end-point batch method and MSFIA
End-point batch methodb
etic (DPPH1 + DPPH2) 0.6 (1:10) 22.6 0.5 (1:20) 8 (1:50) 217 4 (1:100) 2 (1:133) 311 7 (1:200)
L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318 317
Table 5VCEAC values (mg of ascorbic acid/100 ml) obtained by MSFIA methodology (CMSFIA) and by the batch method (Cbatch) for the determination of total antioxidantcapacity
Fruit juiceOrange juiceOrange juiceSoft drinkIsotonic drinkIsotonic drinkGreen teaDark beerLager beerWhite winecRed wine ARed wine Bc
Each value co n perdeviation betw
a (n = 3).b (n = 4).c MSFIA re
The autothe performtion condittemperaturand antioxiworking ento evaporat
Consideusing theexpensiveelectron spously [12,1through theabsorbanceconsumedtion and forperformed
The prebatch methkinetic behthe samplesamples coDPPH, suidant capameasureme
tions conceautomatic flmodel to thestimate thor originatibination ofa considera
. Mde A
6.La V. FraBub,ller,. Ric
36.2 0.3 (1:10)A 9.9 0.1 (1:2)B 27.6 0.3 (1:20)
27.7 0.4 (1:13.3)A 10.1 0.2 (1:5)B 1.1 0.1 (1:2)
111 1 (1:100)36.1 0.8 (1:66.6)10.2 0.3 (1:10)24.8 0.6 (1:10)215 6 (1:333)318 2 (1:133)
rresponds to the mean S.D. The values in parentheses are the sample dilutioeen the two methods.
sults based on kinetic information.
in Table 1 took 273 s. Therefore, in this case, theion frequency was approximately 13 h1.
matic system proposed in the present work allowedance of DPPH assay using strictly controlled reac-
ions with reduced handling of the sample, control ofe and excellent repeatability. The contact of radicaldant species with oxygen and other substances in thevironment is reduced. Moreover, the solvent loss dueion is also reduced when compared to batch methods.ring the automatic flow systems previously describedDPPH assay, the present system relies on lessand common detection system when compared toin resonance (ESR) spectrophotometer used previ-3]. Moreover, the antioxidant capacity was calculatedabsorbance decrease which is independent of initialof DPPH solution rather than the percentage of
 C. E.N A.
(19 M.radical . This aspect is essential for standardiza-obtaining comparable and reliable results for studies
in different laboratories.sent work offers a fast alternative to the end-pointod, also providing qualitative information about theaviour of antioxidant compounds initially present ins or formed during the reaction with DPPH. Forntaining compounds that rapidly scavenge the radicalch juices containing ascorbic acid, the total antiox-city can be determined using a single absorbancent after flow stop. Moreover, the controlled condi-rning time and mixture of reagents provided by theow system allowed the application of a mathematicale data collected within the first 3 min of reaction toe total DPPH consumption for samples containingng slow reacting compounds. In this case, the com-automation and mathematical modelling resulted inble reduction of the time taken for a single analysis,
Analyst D. Huan R. Re, N
Evans, F M.S. Blo W. Bran
28 (1995 H. Uked H. Uked M. Polas
(2004) 7 T. Yama
Biochem I.I. Kole
2323. Z. Jia, B
2 (1998) V. Cerda`
P. Sitjar, F.R.P. R
J.L.M. SCbatchb R.D. (%)36.2 1.9 (1:80) 0.0
9.9 0.4 (1:10) 0.025.2 1.5 (1:40) +9.527.7 0.3 (1:13.3) 0.0
9.9 0.2 (1:5) +2.01.2 0.1 (1:2) 8.3119 1 (1:100) 6.739.1 0.4 (1:20) 7.711.8 0.2 (1:10) 13.622.6 0.5 (1:20) +9.7217 4 (1:100) 0.9311 7 (1:200) +2.3
formed prior to introduction into the flow system. R.D. = relative
esults comparable to those attained using the end-method.
agalhaes thanks FCT and FSE (III Quadro Comu-poio) for the PhD grant SFRH/BD/12539/2003.
ell, J.M.C. Gutteridge, Free Radicals in Biology and Medicine,Oxford University Press, Oxford, 1999.
airfield, R.H. Fletcher, J. Am. Med. Assoc. 287 (2002)
ecchia, A. Altieri, A. Tavani, Eur. J. Nutr. 40 (2001) 261.nkel, Food Chem. 57 (1996) 51.B. Watzl, L. Abrahamse, H. Delincee, S. Adam, J. Wever, H.
G. Rechkemmer, J. Nutr. 130 (2000) 2200.e-Evans, N.J. Miller, G. Paganga, Free Radic. Biol. Med. 2033.lovich, P.D. Prenzler, E. Patsalides, S. McDonald, K. Robards,
127 (2002) 183.g, B. Ou, R.L. Prior, J. Agric. Food Chem. 53 (2005) 1841.. Pellegrini, A. Proteggente, A. Pannala, M. Yang, C. Rice-
ree Radic. Biol. Med. 26 (1999) 1231.is, Nature 181 (1958) 1199.
d-Williams, M.E. Cuvelier, C. Berset, Lebensm. Wiss. Technol.) 25.a, Y. Adachi, M. Sawamura, Talanta 58 (2002) 1279.a, Bunseki Kagaku 53 (2004) 221.ek, P. Skala, L. Opletal, L. Jahodar, Anal. Bioanal. Chem. 37954.guchi, H. Takamura, T. Matoba, J. Terao, Biosci. Biotechnol.. 62 (1998) 1201.
va, H.A.G. Niederlander, T.A. van Beek, Anal. Chem. 72 (2000)
. Zhou, L. Yang, L. Wu, Z. Liu, J. Chem. Soc. Perkin Trans.911.
, J.M. Estela, R. Forteza, A. Cladera, E. Becerra, P. Altimira,Talanta 50 (1999) 695.
ocha, B.F. Reis, E.A.G. Zagatto, J.L.F.C. Lima, R.A.S. Lapa,antos, Anal. Chim. Acta 468 (2002) 119.
318 L.M. Magalhaes et al. / Analytica Chimica Acta 558 (2006) 310318
 D. Majchrzak, S. Mitter, I. Elmadfa, Food Chem. 88 (2004) 447. J. Ruzicka, E.H. Hansen, Flow Injection Analysis, 2nd ed., John Wiley
& Sons, New York, 1988, p. 23. M.A. Segundo, H.M. Oliveira, J.L.F.C. Lima, M.I.G.S. Almeida,
A.O.S.S. Rangel, Anal. Chim. Acta 537 (2005) 207. K. Schlesier, M. Harwat, V. Bohm, R. Bitsch, Free Radic. Res. 36 (2002)
177. C. Sanchez-Moreno, J.A. Larrauri, F. Saura-Calixto, J. Sci. Food Agric.
76 (1998) 270.
 D. Kim, K.W. Lee, H.J. Lee, C.Y. Lee, J. Agric. Food Chem. 50 (2002)3713.
 M.B. Arnao, Trends Food Sci. Technol. 11 (2000) 419. P. Goupy, C. Dufour, M. Loonis, O. Dangles, J. Agric. Food Chem. 51
(2003) 615. J.C. Espn, C. Soler-Rivas, H.J. Wichers, J. Agric. Food Chem. 48 (2000)
648. J.N. Miller, J.C. Miller, Statistics and Chemometrics for Analytical
Chemistry, 4th ed., Pearson Education, Harlow, 2000, p. 126.
Automatic method for determination of total antioxidant capacity using 2,2-diphenyl-1-picrylhydrazyl assayIntroductionExperimentalReagents and solutionsApparatusMSFIA manifold and procedureEnd-point batch method
Results and discussionDevelopment of the MSFIA system for determination of total antioxidant capacityEvaluation of the MSFIA method and its application to samples