Analytical Chemistry Research 5 (2015) 14–20

Contents lists available at ScienceDirect

Analytical Chemistry Research

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

Amperometric detection of carbohydrates based on the glassy carbonelectrode modified with gold nano-flake layer

http://dx.doi.org/10.1016/j.ancr.2015.06.0012214-1812/� 2015 The Authors. Published by Elsevier B.V.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

⇑ Corresponding author.E-mail address: hien7240238@yahoo.com (Q.H. Nguyen).

Huy Du Nguyen a, T. Thuy Luyen Nguyen a, Khac Manh Nguyen a, Anh Mai Nguyen a, Quoc Hien Nguyen b,⇑a Central Laboratory for Analysis, University of Science, Vietnam National University in Ho Chi Minh City, 227 Nguyen Van Cu Street, Ho Chi Minh City, Viet Namb Research and Development Center for Radiation Technology, Vietnam Atomic Energy Institute, 202A, Street 11, Linh Xuan Ward, Thu Duc District, Ho Chi Minh City, Viet Nam

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

Article history:Received 9 March 2015Revised 8 June 2015Accepted 9 June 2015Available online 12 June 2015

Keywords:Pulsed amperometric detectionGold nano-flakeElectrochemical cellCarbohydrate

An electro-deposition approach was established to incorporate the gold nano-flakes onto the glassy car-bon electrode in electrochemical cells (nano-Au/GC/ECCs). Using pulsed amperometric detection (PAD)without any gold oxidation for cleaning (non-oxidative PAD), the nano-Au/GC/ECCs were able to maintaintheir activity for oxidizing of carbohydrates in a normal alkaline medium. The reproducibility of peak areawas about 2 relative standard deviation (RSD,%) for 6 consecutive injections. A dynamic range of carbo-hydrates was obtained over a concentration range of 5–80 mg L�1 and the limits of detection (LOD) wereof 2 mg L�1 for fructose and lactose and 1 mg L�1 for glucose and galactose. Moreover, thenano-Au/GC/ECC using the non-oxidative PAD was able to combine with the internal standard methodfor determination of lactose in fresh cow milk sample.

� 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

The amperometric detection of carbohydrates following highperformance liquid chromatographic (HPLC) separations has sev-eral advantages over other methods [1,2]. As a result, electrochem-ical cells (ECCs) with gold working electrode for carbohydrateanalysis are commercially available. When operating in direct cur-rent (DC) mode, the loss of electrode activity arises. This can beascribed to the fouling of the electrode by the reactant or productadsorption [3]. Thus, the pulsed amperometric detections (PADs)have been developed. In the PAD, anodic detection (ca. 50 mV)was alternated with oxidative and reductive cleaning steps by step-ping the potential to a greater positive value (ca. 600 mV) and to amore negative value (ca. �600 mV), respectively. The electrochem-ical (EC) cleaning of adsorbed reactants or products reactivates thegold electrodes [2–5]. Gold electrode also forms stable oxide layeron the anodic polarization that can be fully reduced to the baremetal during the cathodic polarization [4]. Consequently, PADapproach normally leads to baseline drift [6]. Moreover, gold elec-trode is susceptible to corrosion in the presence of Cl� anions [7],which were abundantly in many samples. This can deteriorate theperformance of EC cells. Therefore, decreasing the gold oxidationin PAD without performance loss can offer the amperometric detec-tion some further advantages in carbohydrate analysis.

By dropping an aliquot of gold(III) solution onto glassy carbon(GC) surface and cycling in a suitable potential range after dryingthe GC surface in oven, Casella et al. created a removable gold layerin EC cell for the amperometric detection of carbohydrates [8].When using as HPLC detector, this gold layer was, however, onlystable with mobile phase containing 1.0 lM of AuCl4

�.The aims of this study are to use an electro-deposition for prepa-

ration of gold nano-flake layer on the GC working electrode in ECcell (nano-Au/GC/ECC) that is able to detect carbohydrates preciselywith a normal alkaline supporting electrolyte and based on thenano-Au/GC/ECC for decreasing the oxidation of gold in PAD.

2. Experimental

2.1. Chemical and reagents

HAuCl4�3H2O, ZnSO4�7H2O, K4[FeCN6]�3H2O, NaOH, HCl andH2SO4 were of PA-grade from Merck (Germany). All the agents (lac-tose monohydrate, galactose, glucose, and fructose) used in thiswork were of PA-grade from Sigma–Aldrich (Singapore).Deionized water (purified by Mili-Q, Millipore) was used through-out all experiments. Stock standard solutions of carbohydrates (ca.1000 mg L�1) were prepared in water and stored at 4 �C. Workingstandard solutions were prepared in water and used in each exper-iment. Carrez I solution was prepared by dissolving 3.60 g ofK4[FeCN6]�3H2O in 100 mL of water. Carrez II solution was pre-pared by dissolving 7.20 g of ZnSO4�7H2O in 100 mL of water.

Fig. 1. Current (A) and voltage (B) versus time of the Au-RD electrode in 100 mM NaOH solution. A 0.5 lg of glucose was added into CV cell at 8th cycle.

Fig. 2. Current (A) and voltage (B) versus time of the nano-Au/Au-RD electrode in 100 mM NaOH solution. A 0.5 lg of glucose was added into CV cell in 8th cycle.

H.D. Nguyen et al. / Analytical Chemistry Research 5 (2015) 14–20 15

Gold(III) solution was prepared from HAuCl4�3H2O in the mixturesolution of 0.5 M H2SO4 and 0.05 M HCl.

2.2. Apparatus

The Waters HPLC systems (Waters, USA) consisted of a binarypump (Waters 1525, USA), a 20 lL loop, a column oven, and an

EC detector consisting of an EC flow cell. The EC detector or poten-tiostat (Waters 2465, USA) can operate in DC, pulse and scan mode.The EC flow cell is in wall-jet structure consisting of a 2 mm diam-eter GC working electrode, an in-situ silver/silver chloride refer-ence electrode (diameter of 2 mm) and the platinum bodycounter electrode. A typical three-electrode cell consisted of a2 mm diameter gold rotating disk working electrode, an Ag/AgCl,

Fig. 4. The SEM images of the nano-Au/GC/ECCs deposited with CAu(III) = 1.2 mM (A)and CAu(III) = 4.8 mM (B).

16 H.D. Nguyen et al. / Analytical Chemistry Research 5 (2015) 14–20

3 M KCl reference electrode and the platinum wire counterelectrode.

2.3. Electrode preparations

2.3.1. Preparation of nano-Au/Au-RD electrodeThe nano-gold layer on gold rotating disk (nano-Au/Au-RD)

electrode is prepared by the potentiostat operating in DC modeand the typical three-electrode cell. Following the procedures fromprevious studies [9–13], the Au-RD working electrode was polishedwith the water slurry of 0.05 lm alumina particles. The residualpolishing material was removed using ethanol and then water inan ultrasonic bath. The Au-RDE was immersed into 1 mM HAuCl4

solution and deposition was conducted at the applied potentialof �200 mV (vs. Ag/AgCl, 3 M KCl) for 5 min. Finally thenano-Au/Au-RD electrode was washed with water and kept in100 mM NaOH solution until use.

2.3.2. Preparation of nano-Au/GC/ECCThe nano-Au/GC/ECC was prepared by the potentiostat operat-

ing in DC mode and the EC flow cell. Before gold nano-flake(nano-Au) deposition, the electrodes are polished to a mirror sur-face with the slurry of alumina as described above. The GC elec-trode was conditioned in 0.5 M HCl solution by applyingpotential as follows: �500 mV for 10 min and 800 mV for 1 min.After conditioning, the obtained currents were of ca. 30 lA at�500 mV and less than 1 lA at 800 mV if the EC cell was clean.The nano-Au deposition is performed at the potential (EP) of�200 mV for deposition time (tP) of 2.5 min. The 4.8 mM HAuCl4

solution is put in the EC cell at the flow-rate (vF) of 0.5 mL min�1

by a syringe pump (KDS 100, Scientific). After nano-Au deposition,the reference electrode was polished to a mirror surface, the EC cellwas assembled with two gaskets and the working electrode wasconditioned in 50 mM NaOH solution (1.0 mL min�1) by applyingpotential as follows: �500 mV in 30 min and 200 mV in 2.0 min.If the gold deposition was successful, the obtained current usuallyreached to �12 lA at �500 mV after conditioning.

2.4. Electrode characterization

Cyclic voltammetric (CV) experiments were performed by thepotentiostat operating in scan mode and the nano-Au/GC/ECC orthe typical three-electrode cell with the Au-RD or thenano-Au/Au-RD working electrode. The voltage and current versustime data was recorded by the Empower Pro software (Waters,USA). The surface of working electrode was first activated in

300

350

400

450

500

550

1 2 3 4 5

CAu (III) (mM)

SP (mV.s)

Fig. 3. The SP of glucose peak versus CAu(III).

100 mM NaOH solution stirred by linear potential sweep from�800 to +800 mV with scan rate of 10 mV s�1. After obtainingreproducible voltammograms, a 0.5 lg of glucose was added intothe cell to evaluate the electrode activity for the carbohydrateoxidation.

The surface morphology of the nano-Au layers was examined bya field emission scanning electron microscope (FESEM), Hitachi,S-4800, Japan. The nano-Au layer was taken out of the GC surfaceof the EC cell by adhesive carbon tape before SEM photograph.

2.5. Carbohydrate analysis

2.5.1. HPLC-PAD procedureThe HPLC-PAD for carbohydrates was performed on the HPLC

system with the column oven set at 80 �C and the EC detector oper-ating in pulse mode with the nano-Au/GC/ECC. The PAD was set asfollows: E1 at 50 mV (800 ms), E2 at 200 mV (100 ms), E3 at�400 mV (200 ms), sampling for 100 ms and time constant in 2 s.The Aminex HPX-87C, 300 � 7.8 mm (Bio-Rad) column is usedfor HPLC separation. The binary pump consists of channel A drivingwater through the analysis column at the flow rate of 0.5 mL min�1

to elute the carbohydrates and the channel B putting thepost-column additive of 100 mM NaOH solution at the flow-rateof 0.5 mL min�1 into the T-tubing connector between the analysiscolumn and the EC cell.

2.5.2. Sample preparationThe sample clean-up was carried out according to AOAC 984.15.

Briefly, ca. 2.0 g milk sample was accurately weighed into 100 mLvolumetric flask containing 60 mL water. Then 0.5 mL carrez Iand 0.5 mL carrez II solution were added into the flask, and madeup to 100 mL with water. The precipitate was filtered off and thefirst 20 mL of the filtrate was discarded. The filtrate was diluted20 times by water and resultant solution was filtered with

200

250

300

350

400

450

500

550

1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2tP (min)

SP (mV.s)

Fig. 5. The SP of glucose peak versus tP.

Fig. 6. Current (A) and voltage (B) versus time of the nano-Au/GC/ECC in flow injection manner with a 100 mM NaOH solution fed at 1.0 mL min�1.

Fig. 7. Standard calibration curves of typical carbohydrates (A) and non-oxidative PAD chromatogram of 4 mg L�1 lactose and 2 mg L�1 of other carbohydrates (B).

H.D. Nguyen et al. / Analytical Chemistry Research 5 (2015) 14–20 17

0.45 lm membrane filter. A 20 lL aliquot of the membrane filtratewas injected on the HPX-87C column.

3. Result and discussion

3.1. The activity of nano-Au layer

For the Au-RD electrode, the negative sweep yields an oxidativecurrent peak that is greater than the oxidative current peakobserved at the same potential in the positive sweep when0.5 lg of glucose was added into the cell at 8th cycle (Fig. 1).This phenomenon was also obtained in the study of Neuburgerand Johnson [14]. Thus, the increase of the activity of gold surface

for carbohydrate oxidation is the virtual aim of the alternation ofoxidative and reductive steps in PAD.

On the other hand, for the nano-Au/Au-RD electrode, the cur-rent peaks of the glucose oxidations were the same in both catho-dic and anodic process. Further, these current peaks were alsogreater than those of Au-RD electrode (Fig. 2). Thus, the nano-Aulayer may possess more activity for the oxidation of glucose thanthe gold mirror surface alternated with oxidative and reductivesteps.

Therefore, the alternation of oxidative and reductive steps inPAD for raising the activity of gold surface may be replaced bythe modification of working electrode with a nano-Au layer. Thisassumption will be further clarified in the following part.

18 H.D. Nguyen et al. / Analytical Chemistry Research 5 (2015) 14–20

3.2. The optimization of electro-deposition

For EC cell, the renewal of the nano-Au layer before each chro-matographic run is not able though this can improve the repro-ducibility [12]. Therefore, enhancing and preserving the activityof the nano-Au layer are two main goals of the optimization ofthe electro-deposition. The activity of the nano-Au layer involvesdirectly to the magnitude of the HPLC-PAD peak area of carbohy-drate (SP) and the active preservation is illustrated by reproducibil-ity of SP or relative standard deviation of SP (RSD,%) betweenchromatographic runs.

Because of the wall-jet structure, it is really convenient andeffective if the electro-deposition for the nano-Au/GC/ECC is con-ducted in flow injection manner with the HAuCl4 solution fed intothe EC cell by a syringe pump. So that, the investigating parametersconsist of the HAuCl4 concentration (CAu(III)), the feeding flow rate(vF), the deposition potential (EP), the deposition time (tP) and tem-perature (TP).

In our previous study [11], the electro-deposition of goldnano-flakes (nano-Au) on electrode surface of RD electrode fromHAuCl4 solution depends on the activity of electrode surface forthe reduction of HAuCl4 into nano-Au, the transferring rate ofHAuCl4 to electrode surface and the deposition time (tP). The activ-ity of electrode surface is conducted by the deposition potential(EP). The transferring rate of HAuCl4 is proportional to the HAuCl4

concentration (CAu(III)) and flow rate of HAuCl4 solution (vF).When optimizing electro-deposition of gold nano-flakes(nano-Au) on electrode surface of the electrochemical cell ofHPLC system, we kept the activity of electrode surface at the opti-mal EP of �200 mV according to our previous study [11] and inves-tigated the transferring rate of HAuCl4 to electrode surface bychanging CAu(III) while constant vF of 0.5 mL because CAu(III) affecteddirectly to the size of nano-Au [12]. The optimal CAu(III) is ableenough to show the optimal ratio between the reduction rate ofAu(III) on electrode surface and the transferring rate of HAuCl4 toelectrode surface. Thus, the vF is not necessary to investigate fur-ther. After discovering the optimal CAu(III), we have optimized thedeposition time (tP).

3.2.1. The concentration of HAuCl4Amount of the nano-Au electrochemically deposited on the GC

increases with the CAu(III) and tP. To obtain the nano-Au layers withthe same amount of gold, CAu(III) and tP were varied in the oppositedirections as follows. CAu(III) was of 1.2, 2.4 and 4.8 mM corre-sponding to tP of 10, 5 and 2.5 min. For each electro-deposition

Fig. 8. The PAD chromatograms of lactose in the fresh cow milk sam

condition, 6 consecutive runs of HPLC-PAD using the resultantnano-Au/GC/ECC were replicated to evaluate detector signal. Inour experiment, RSD decreased significantly with CAu(III) whilethe average of SP is nearly constant (Fig. 3). The best conditionwas found with 4.8 mM HAuCl4 for 2.5 min.

There is significant difference in the morphology of nano-Aulayer when varying the HAuCl4 concentration. Gold nano-flakeswas dominant with CAu(III) = 4.8 mM (Fig. 4B) while nano-flakesand nano-particles were observed with CAu(III) = 1.2 mM (Fig. 4A).Thus, high concentration of HAuCl4 was likely to give equal possi-bility for all gold clusters to grow into nano-flakes and only goldnano-flakes is able to maintain its activity in PAD for carbohy-drates. In our previous study, the antioxidant efficiency of goldnano-particles decreased with its size [15] as the explanation forthe stability of the gold nano-flakes in PAD condition. It shouldbe noted that the inclination of nano-flakes in Fig. 4 could beresulted from the sample preparation for SEM photograph.

3.2.2. The deposition timeThe thickness of the nano-Au layer increased with tP. When the

gold nano-flakes touched to the nozzle of the cell and resisted thefeeding flow, the back pressure raised immediately and the syringepump stopped automatically. When this phenomenon happened, itis supposed that tP reached the maximum. With the CAu(III) of4.8 mM, the maximum time is about 3 min. Before tP reached themaximum, the SP and the RSD decreased slightly with tP (Fig. 5).Thus, the longer tP was, the larger the average size of the goldnano-flakes attained. This made the nano-Au layer more stablebut the electrochemically active surface smaller, as the result ofthe gold nano-flakes were bigger by agglomeration with eachother.

However, at the maximal value of tP, RSD dramatically increasedwhile SP significantly declined. This behavior could be derived fromthe partially destroyed structure of the nano-Au layer by the highpressure from syringe pump.

3.3. Carbohydrate detection

Cyclic voltammetry of the nano-Au/GC/ECC was measured inflow injection manner with a 100 mM NaOH solution fed at1.0 mL min�1 (Fig. 6) to study the pulsed potential waveform ofthe PAD. If the E2 (oxidative cleaning step) does not exceed200 mV, the PAD cannot oxidize nano-Au layer. This is also truefor bulk-Au/ECC [4]. In our experiments, the optimalnon-oxidative PAD had the pulsed potential waveform as follows:

ple (A) and the sample spiked glucose as internal standard (B).

y = 0.491x + 0.012R² = 0.998

0.5

1.5

2.5

3.5

4.5

2 4 6 8 10CLac/CGlu

(A)

SPLac/SP

Glu

(B)

Fig. 9. Internal standard calibration curve for lactose with glucose as internal standard (A) and the PAD chromatograms of internal standard solutions (B).

H.D. Nguyen et al. / Analytical Chemistry Research 5 (2015) 14–20 19

E1 = 50 mV for 800 ms, E2 = 200 mV for 100 ms, E3 = �400 mV for200 ms.

An advantage of the nano-Au/GC/ECC is to require low poten-tials to clean up the surface (E2 and E3 were of 200 mV and�400 mV, respectively). However, the period of detection stephad to be extended to 800 ms for eliminating capacitive currentbecause the electrochemically active surface of nano-Au layerwas higher than that of bulk-Au electrode [8].

With the nano-Au/GC/ECC using the non-oxidative PAD, the lin-ear responses for carbohydrates were observed over the concentra-tion range of 5-80 mg L�1 with alkaline supporting electrolyte(100 mM NaOH) (Fig. 7). The detection limits of sugars (corre-sponding to S/N = 3) were of 2 mg L�1 for fructose and lactoseand 1 mg L�1 for glucose and galactose.

The high detection limits due to the background line with a sig-nificant drift and noise (Fig. 7B) which may be a result of thepost-column alkaline addition [16] or the rough surface ofnano-Au layer [17]. Although the nano-Au/GC/ECCs using thenon-oxidative PAD possessed the detection limits of carbohydrateswhich are about 50-fold higher than previously reported results[18], the reproducibility of peak areas was very good (RSD � 2%,n = 6) which could be a consequence of the complete eliminationof gold oxidation. Therefore, this method is suitable for food andbiological samples which contain high levels of carbohydratesand interfering substances.

3.4. Determination of lactose in fresh cow milk

Because of high levels of lactose in milk, a 100 mM NaOH solu-tion was used as post-column additive to make the sensitivity ofcarbohydrate detection lower. The adjusting of the sensitivity ofmethod by changing the concentration of NaOH in post-columnalkaline additive did not influence on the HPLC separation of carbo-hydrates, as the explanation for using the post-column alkalineadditive and the Aminex HPX-87C column with water eluent.

Fresh cow milk samples contain the high contents of proteinsand other interfering substances. The HPLC analysis usually hasto follow the clean-up techniques using precipitating agents e.g.,carrez to remove the proteins off milk samples [19–21]. By usingonly carrez agents similar to Cataldi et al. [21], the non-oxidativePAD and the nano-Au/GC/ECC were able to maintain their activitythroughout the following HPLC detection of lactose in this milksample. The chromatographic peak areas of lactose showed a goodreproducibility with �2% RSD for 6 consecutive injections (Fig. 8).The reproducibility was better than that of previous publication[21] using bulk-Au/ECC and PAD with the oxidation of gold. This

is a convincing evidence for the usability of the nano-Au/GC/ECCbased on the non-oxidative PAD. Thus, for carbohydrate analysis,the nano-Au/GC/ECC using the non-oxidative PAD was capable toreplace the bulk-Au/ECC using the PAD with the oxidation of goldin order to decrease the corrosion of gold working electrode whenthe samples contain the agents that are able to interact with AuOH.

According to the results of the previous study [21] and AOAC44.1.35A (2005), the internal standard (IS) was necessary to improvethe precision of analysis results in the case the standard solutionswere chromatographed far away from the samples when carbohy-drates are determined by PAD. In this work, glucose was used as ISfor determination of lactose in fresh cow milk samples.Chromatograms of the milk sample spiked glucose (Fig. 8B) also dis-played high reproducibility of peak areas for both lactose and glu-cose, which were also about 2% RSD for 6 consecutive injections.The IS calibration curve with good correlation coefficient(r2 = 0.998) was presented clearly on Fig. 9. Thus, thenano-Au/GC/ECC using the non-oxidative PAD was successful tocombine with the IS method for quantifying lactose in fresh milksample. Therefore, this technique was capable to apply in batchanalysis for carbohydrates.

4. Conclusions

To our best knowledge this work is one of the first study forelectro-deposition of the gold nano-flakes onto GC electrode byflow injection manner. The nano-Au/GC/ECCs using thenon-oxidative PAD was able to maintain its activity for oxidizingcarbohydrates in normal alkaline supporting electrolyte. With theIS method, lactose in fresh cow milk sample could be quantifiedwith high precision by this technique.

The significance of our work is to decrease positive potentials toclean up the electrode surface (E2 = 200 mV), in order to prevent ECcells from risk when determination of sugar is performed on com-plicated samples. Therefore, electrode modification with nano-Aulayer and using non-oxidation PAD are significant complimentsof electrochemical detection of sugars.

Conflict of interest

The authors declare that they have no competing interest.

Acknowledgement

This research is funded by Vietnam National University Ho ChiMinh City (VNU-HCM) under Grant number B2012-18-30.

20 H.D. Nguyen et al. / Analytical Chemistry Research 5 (2015) 14–20

References

[1] J.M. Mariolir, P.F. Luo, T. Kuwana, Nickel–chromium alloy electrode as acarbohydrate detector for liquid chromatography, Anal. Chim. Acta 282 (1993)571–580.

[2] C. Corradini, A. Cavazza, C. Bignardi, High performance anion-exchangechromatography coupled with pulsed electrochemical detection as apowerful tool to evaluate carbohydrates of food interest: principles andapplications, Int. J. Carb. Chem. 2012 (2012) 1–13.

[3] I.G. Casella, E. Desimoni, A.M. Salvi, Chemically modified electrode for thedetection of carbohydrates, Anal. Chim. Acta 243 (1991) 61–63.

[4] D.C. Johnson, W.R. LaCourse, Liquid chromatography with pulsedelectrochemical detection at gold and platinum electrodes, Anal. Chem. 60(1990) 589–597.

[5] K. Mopper, C.A. Schultz, L. Chevolot, C. Germah, R. Revuelta, R. Dawsong,Determination of sugars in unconcentrated seawater and other natural watersby liquid chromatography and pulsed amperometric detection, Environ. Sci.Technol. 26 (1992) 133–138.

[6] W.R. Lacourse, D.A. Mead Jr., D.C. Johnson, Anion-exchange separation ofcarbohydrates with pulsed amperometric using a pH-selective referenceelectrode, Anal. Chem. 62 (1990) 220–224.

[7] G.M. Schmid, M.E. Curley-Fiorino, Gold, in: A.J. Bard, M. Stratmann, G.S.Frankel (Eds.), Encyclopedia of electrochemistry of the elements, Vol. IV,Marcel Dekker Inc., New York, 1975, p. 87.

[8] I.G. Casella, A. Destradis, E. Desimonib, Colloidal gold supported onto glassycarbon substrates as an amperometric sensor for carbohydrates in flowinjection and liquid chromatography, Analyst 121 (1996) 249–254.

[9] B. Liu, L. Lu, M. Wang, Y. Zi, A study of nanostructured gold modified glassycarbon electrode for the determination of trace Cr(VI), J. Chem. Sci. 120 (5)(2008) 493–498.

[10] O. Abollino, A. Giacomino, M. Malandrino, G. Piscionieri, E. Mentasti,Determination of mercury by anodic stripping voltammetry with a goldnanoparticle-modified glassy carbon electrode, Electroanalysis 20 (1) (2008)75–83.

[11] H.D. Nguyen, T.T.L. Nguyen, K.M. Nguyen, T.A.T. Tran, A.M. Nguyen, Q.H.Nguyen, Determination of ppt level chromium(VI) using the gold nano-flakes

electrodeposited on platinum rotating disk electrode and modified with 4-thiopyridinium, Am. J. Anal. Chem. 6 (2015) 457–467.

[12] X. Dai, O. Nekrassova, M.E. Hyde, R.G. Compton, Anodic stripping voltammetryof arsenic(III) using gold nanoparticle-modified electrodes, Anal. Chem. 76(2004) 5924–5929.

[13] R. Ouyang, S.A. Bragg, J.Q. Chambers, Zi.-L. Xue, Flower-like self-assembly ofgold nanoparticles for highly sensitive electrochemical detection ofchromium(VI), Anal. Chim. Acta 722 (2012) 1–7.

[14] G.G. Neuburger, D.C. Johnson, Pulsed amperometric detection of carbohydratesat gold electrodes with a two-step potential waveform determination ofsugars in unconcentrated seawater and other natural waters by liquidchromatography and pulsed amperometric detection, Anal. Chem. 59 (1987)150–154.

[15] N.D. Nguyen, V.P. Dang, A.Q. Le, Q.H. Nguyen, Electron beam/c-ray irradiationsynthesis of gold nanoparticles and investigation of antioxidant activity, Adv.Nat. Sci. Nanosci. Nanotechnol. 5 (2014) 045002.

[16] K. Mopper, C.A. Schultz, L. Chevolot, C. Germah, R. Revuelta, R. Dawson,Determination of sugars in unconcentrated seawater and other natural watersby liquid chromatography and pulsed amperometric detection, Environ. Sci.Technol. 26 (1992) 133–138.

[17] R.W. Andrews, R.M. King, Selection of potentials for pulsed amperometricdetection of carbohydrates at gold electrodes, Anal. Chem. 62 (1990) 2130–2134.

[18] J.R. Welborn, D.T. Manahan, Direct measurements of sugar uptake fromseawater into molluscan larvae, Mar. Ecol. Prog. Ser. 65 (1990) 233–239.

[19] C. Brons, C. Olieman, Study of the high-performance liquid chromatographicseparation of reducing sugars, applied to the determination of lactose in milk,J. Chromatogr. 259 (1983) 79–86.

[20] W. Xinmin, Z. Ruili, L. Zhihua, W. Yuanhong, J. Tingfu, Determination ofglucosamine and lactose in milk-based formulae by high-performance liquidchromatography, J. Food Compos. Anal. 21 (2008) 255–258.

[21] T.R.I. Cataldi, M. Angelotti, S.A. Bufo, Method development for the quantitativedetermination of lactulose in heat-treated milks by HPAEC with pulsedamperometric detection, Anal. Chem. 71 (1999) 4919–4925.