Article Electrochemistry, 83(2), 1–7 (2015)

Sodium Alginate Decorated Carbon Nanotubes-Graphene Composite Aerogelfor Heavy Metal Ions DetectionDafu WANG,a,b Feifei ZHANG,b,c,* and Jie TANGd,e,*a Department of Materials Science, College of Chemical Science and Engineering, Qingdao University,Shandong 266071, P.R. China

b State Laboratory of Fiber Materials and Modern Textile, Qingdao University,Qingdao, Shandong 266071, P.R. China

c Shandong Sino-Japanese Center for Collaborative Research of Carbon Nanomaterials, Qingdao University,Qingdao, Shandong 266071, P.R. China

d National Institute for Materials Science, 1-2-1 Sengen, Tsukuba 305-0047, Japane Doctoral Program in Materials Science and Engineering, University of Tsukuba, Tsukuba 305-8577, Japan

*Corresponding authors: zhangfeifei00921@163.com, tang.jie@nims.go.jp

ABSTRACTSingle-walled carbon nanotubes and graphene (SWNTs/G) composite aerogel have been prepared by ahydrothermal method with aid of sodium alginate (SA) decoration and used for modification of glassy carbonelectrode. The modified electrode was employed for the sensitive determination of metal ions using differentialpulse anodic stripping voltammetry method. The results showed that the introduction of SA not only improved theuniformity of SWNTs/G composite, but also exhibited synergistic effect for detection of several kinds of heavy metalion. The SA decorated SWNTs were inserted into graphene interlayers, resulting in more active sites and reactivesurface area. As a result, the modified electrode exhibited attractive inter-electrode reproducibility and highsensitivity for detection of Pb2+, Cd2+ and Cu2+ ions.

© The Electrochemical Society of Japan, All rights reserved.

Keywords : Graphene, Single-walled Carbon Nanotubes, Sodium Alginate, Heavy Metal Ions

1. Introduction

Graphene has been demonstrated as one of the most promisingmaterials in electrochemistry as novel electrode material for variouspurposes, due to its unique properties such as fast electrontransportation, high thermal conductivity and excellent mechanicalflexibility, etc.1–3 A number of studies have shown that graphenesheets synthesized by hydrothermal method usually retained wide-spread structural defects, such as vacancies and non-planar structurevacancies on the surface.4,5 Some oxygen-containing functionalgroups such as hydroxyl and epoxyl groups can be introduced onthe graphene sheets. These structure defects and functional groupsare favorable for many electrochemical applications.6,7 However,graphene sheets tend to aggregate irreversibly through the strongO–O interactions during the reduction process, thereby decreasingthe specific surface area. When single-walled carbon nanotubes(SWNTs) were used as spacers to prepare the graphene/SWNTscomposite, the re-stacking and agglomeration of graphene couldbe effectively avoided. In the meanwhile, the unique electronicproperties of SWNTs may help to accelerate electron transferreactions with the electro-active species in solution.8

Graphene, carbon nanotubes (CNTs) and their composites basedelectrodes have drawn increasing attentions in the field of metal ionsdetection due to their unique physical and chemical propertiesincluding high surface area, strong mechanical strength and electricconductivity, et al.9 There are some reports on the determinationof heavy metal ions using carbon nanotube modified electrode10

and graphene-modified electrode.11 In addition, some carbon-basednano-composites containing metal ions receptors such as cysteine12

and chitosan,13 which can react with metal ions and produce stablecomplexes, have been successfully prepared and employed in theheavy metal ions analysis. Sodium alginate (SA), a naturalpolysaccharide product extracted from the cell wall of brown

seaweed, show excellent performance for adsorption of heavy metalions (Pb2+, Cd2+, Cu2+, Hg2+)14–16 and could enhance the analyticalsensitivity of the as-prepared electrode. As a surface dispersant,SA has been reported to improve the dispersibility of SWNTs byadsorption and surface twine.17

In our opinions, the decoration of carbon nanotubes and SA tographene should create new nano-composite materials with theintegrated properties of the three components, which offers newopportunities for the development of sensors with improvedperformance. In the meanwhile, when SA was exploited as thesurface dispersant, the electrochemical stability of electrodematerials can be improved.17,18 In the present work, a new nano-composite of single-walled carbon nanotubes and graphene(SWNTs/G) was prepared by a hydrothermal method with aid ofSA. A hypothesis was proposed that SWNTs would be wrappedby the SA chains and worked as spacers between different layersof graphene nanosheets (Fig. 1). The polar functionalities on SAmay interact with the oxygen-containing groups on graphene viahydrogen bonding to enhance the stability of the as-preparedcomposite structure.

2. Experimental

2.1 Apparatus and ReagentsGraphene oxide (GO) was synthesized from natural graphite

flakes by a modified Hummers-Offeman method.19 The purity ofgraphite is 99%, and the average size is about 20 µm. Graphite,sodium alginate (SA) and urea were from Nacalai Tesque, Japan.SWNTs with a diameter of 0.8–1.2 nm and length of 100–1000 nmwere purchased from Super Purified HiPco SWNTs, Nanointegris.In the experimental section, all chemicals used were of analyticalgrade and all solutions were prepared with deionized water with theresistivity not less than 18.2M³ (Milli Q, Yamato WR700) and

Electrochemistry Received: August 11, 2014Accepted: December 8, 2014Published: February 5, 2015

The Electrochemical Society of Japan http://dx.doi.org/10.5796/electrochemistry.83.XXX

14-E00085

1

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

were degassed with nitrogen for 20min prior to electrochemicalmeasurements.

2.2 CharacterizationMorphology and structure of the material were characterized

using scanning electron microscopy (SEM, JEOL JSM-7001F).X-ray photoelectron spectroscopy (XPS) measurements were madewith a PHI Quantera SXM (ULTRAVAC-PHI). The Raman spectrawere obtained with a RAMAN-11 (Nanophoton) with a 532 nm lasersource. The specific surface area were measured with Autosorb-1(Quantachrome Instruments).

2.3 Preparation of the compositeThe suspension of GO at concentration of 1mg/mL in water and

SWNTs at concentration of 1mg/mL in 0.5wt% SA water solutionwere mixed together, followed by addition of 5mg urea. Theresulting mixture was sealed in a 200mL Teflon-lined autoclave for4 h at 180°C to obtain the SA decorated SWNTs and graphenecomposite aerogel (SA/SWNTs/G) and then freezing dried for 48 h.For comparison studies, hydrothermally reduced GO without addingSWNTs was also prepared using the same process.

2.4 Electrochemical measurementsElectrochemical measurements were performed using an EC-Lab

VSP-300 potentiostat (Biologic). Bare or modified glassy carbonelectrode (GCE) of 3mm in diameter was used as the workingelectrode, platinum wire as the counter electrode, and saturatedcalomel electrode (SCE) as the reference electrode. GCE waspolished with alumina slurry (0.05 and 0.3 µm) on a polishing cloth,then cleaned in a water bath with sonication for 30 s and finallyrinsed with water. For activation, the bare GCE was scanned for 20cycles in 0.5M H2SO4 with potential range from ¹1.0–0V at scanrate of 100mV/s.

The clean electrode was dried under an infrared lamp. 10mg ofSWNTs, graphene and SA/SWNTs/G composite were dispersed in10mL SAwater solution (with concentration of 0 to 1wt%) with theaid of ultrasonic agitation for 30min to afford a suspension withconcentration of 1mg/mL. The SA decorated SWNTs modifiedelectrode (SA/SWNTs-GCE), SA decorated hydrothermal graphenemodified electrode (SA/G-GCE), and SA/SWNTs/G modifiedelectrode (SA/SWNTs/G-GCE) were prepared by dropping 4µLof the solution with a fine pipette on the GCE and then driedunder an infrared lamp. During the electrochemical analysis usingdifferential pulse anodic stripping voltammetry (DPASV), the pHof Pb2+ in acetate buffer solution was maintained at 4.6 all thetime. The deposition was performed in 1 © 10¹6mol/L Pb2+, Cd2+

and Cu2+ solutions at a potential of ¹1.20V for 5 minutes, anoptimal depositing time was determined by the pre-experiment.20,21

Following the pre-concentration step, the solutions were quiet for

30 s, then the differential pulse voltammograms (DPV) wererecorded during the potential sweep from ¹1.20V to 0V (with astep increment of 5mV, amplitude of 80mV, and pulse period of0.2 s). The electrode was rinsed in clean acetate buffer solution byholding the potential at +0.3V for 1 minute while stirring, followedby potential scanning from +0.3V to 0V for complete removal ofthe previous deposits.

3. Results and Discussion

3.1 Characterizations of the SA/SWNTs/G composite3.1.1 Morphology of SA/SWNTs/G composite

Figure 2 shows the morphologies of (a) SWNTs dispersed in0.5wt% SA (SA/SWNTs), (b) hydrothermally reduced grapheneoxide previously dispersed in water (G) and (c) in 0.5wt% SA(SA/G), (d) SA/SWNTs/G composite. As shown in Fig. 2(a), theSWNTs could be dispersed well in the SA solution (0.5wt%).Figures 2(b) and 2(c) show the structure of the hydrothermallyreduced graphene before and after being modified with SA(CSA = 0.5wt%). It could be observed that the surface morphologyof graphene become rough, which may because of the uniformadhesion of SA on the surface of graphene due to the formation ofhydrogen bonds between SA and oxide groups on graphene22 or ananion-O type of interaction in between SA and graphene.23 As can beseen in Fig. 2(d), the SA/SWNTs/G composite has a well-definedand interconnected three-dimensional porous network, indicatingthat the SWNTs were inserted into the interlayer space of grapheneand the accumulation or re-stacking of graphene layers could beeffectively prevented. Subsequently, intra-pores would be created toincrease the specific surface area and the analytic molecules wouldbecome more accessible to the electrode surface.3.1.2 Brunauer-Emmett-Teller (BET) specific surface areas

By Nitrogen-adsorption and -desorption isotherms shown inFig. 3, the multipoint BET specific surface areas for hydrothermallyreduced graphene oxide dispersed in water (a, G), in 0.5wt% SA(b, SA/G) without SWNTs, and in SA decorated SWNTs solution(c, SA/SWNTs/G) were obtained as 558m2/g, 510m2/g, and702m2/g, respectively. It should be noted that, the BET specificsurface area of the SA/SWNTs/G composite is much larger thanthat of G and SA/G. This indicates that SA decorated SWNTs actedindeed as spacers between the graphene layers, thereby preventingtheir agglomeration and restacking, as revealed in Fig. 1.3.1.3 XPS characterization

XPS was applied to determine the elemental composition as wellas chemical and electronic states of elements on the surface of the

Figure 2. SEM images of (a) SA/SWNTs, (b) G, (c) SA/G and(d) SA/SWNTs/G composite.

Figure 1. (Color online) The schematic illustration for thestructure of sodium alginate (SA) decorated single-walled carbonnanotubes and graphene composite.

Electrochemistry, 83(2), 1–7 (2015)

14-E00085

2

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

material. The surface chemistry of modified electrodes is known toaffect the electrochemical processes.24

Figure 4(a) shows the wide spectra for GO and SA/SWNTs/Gcomposite. According to the C1s and O1s signal intensities, the C:Oatomic ratios were calculated to be 3:2 for GO and 14.7:1 for thecomposite. The results clearly revealed that GO was effectivelyreduced in the hydrothermal process.

The high resolution XPS spectra for C1s and O1s are shown inFigs. 4(b) and 4(c). The fitting curves of the C1s and O1s spectrawere performed using the Gaussian-Lorentzian Function after

performing a background correction. The XPS spectra displayedseveral types of functional groups on the surface of the compositedstructure. The C1s spectrum of the composite can be deconvolutedinto three peaks at 284.6, 287.0 and 291.0 eV, respectively. The peakat 284.6 eV corresponds to the sp2 hybridized graphitic carbon[Fig. 4(b)].25,26 In an adjacent region of higher binding energy,the chemical shifts around 291.0 eV were likely to be due to theO-O* shaking up. In addition, GO prepared by the Hummer methodusually comes with some epoxide groups (like C-O-C). Thoseepoxide groups would have a C1s binding energy similar to C-O andC-OH, which also triggered the chemical shifts in the nearby regionsof lower binding energy to 287.0 eV.26,27

The O1s spectra in Fig. 4(c) have a main peak at 531.1 eV, whichwas accompanied with another peak at 532.7 eV. The O1s photo-electron kinetic energies are lower than those of the C1s and the O1ssampling depth was smaller, therefore the O1s spectra were slightlymore surface specific.27 The O1s peak at 531.1 eV should beassigned to the vibration of oxygen-containing groups on the surfaceof graphene, meanwhile the peak at 532.7 eV indicating theexistence of H2O and the -OH functional groups.

During the process of GO being restored to graphene by thehydrothermal method, CO(NH2)2 has been used as the reducingagent. CO(NH2)2 is alkaline, which means CO(NH2)2 may initiatereactions with some functional groups on the surface of graphene,such like carboxyl.7 The mechanism of reduction of CO(NH2)2 onthe surface of graphene is likely to be similar to Wolff-KishnerReduction: aldehydes or ketones under basic conditions could havereactions with CO(NH2)2, and then the carbonyl groups would bereduced to methylene.28 The nitrogen has not been removed by

Figure 3. Nitrogen adsorption and desorption isotherms of (a) G,(b), SA/G and (c) SA/SWNTs/G.

Figure 4. (Color online) (a) The XPS wide spectra of GO and SA/SWNTs/G composite, and (b-d) high resolution C1s, O1s and N1s XPSspectra of SA/SWNTs/G composite. (For interpretation of the references to color in this figure legend, the reader is referred to the webversion of this article.)

Electrochemistry, 83(2), 1–7 (2015)

14-E00085

3

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

hydrolysis or any other methods, so there was still a small portionremaining on the surface of graphene. That is one of the reasonswhy the chemical shifts of N1s around 400.0 eV were displayed inthe XPS wide spectra of G/SWNT composite, which may be causedby -CN, -NH4

+, -NO, NH3 or -N=N-. Furthermore, the deviation ofXPS during the detection should also be considered.3.1.4 Raman characterization

The Raman spectrum of SA/SWNTs/G composite is shown inFig. 5. The four peaks corresponding to D, G, 2D, S3 (from left toright) can be clearly observed. Three of them (D, G, 2D) are likelyfrom the vibration and formant on graphene. Specifically, the G peakat 1595.6 cm¹1 should be resulted from the vibration of graphenecarbon atoms, which is also sensitive to stress.29 The D peak at1337.6 cm¹1 comes from the disordered vibration due to the latticemovement away from the center of the Brillouin zone, whichindicates the existence of both the structural and crystal defects ongraphene surface. The I(D)/I(G) intensity ratio of SA/SWNTs/Gsuggests a possible decrease in the average size of the sp2 domains.30

The 2D peak (also known as GB peak) at 2688.7 cm¹1 is the two-phonon Raman formant, which shows the existence of pile structureof graphene.29,31 The combination of the D and G peaks gives rise toan S3 peak at 2921.2 cm¹1 as a consequence of lattice disorders.32

Thus, the SEM, XPS and Raman results have revealed the mostprobable structure of the SA/SWNTs/G composite. SWNTs shouldbe wrapped by the SA chains, and work as the support structurebetween different layers of graphene nanosheets (Fig. 1). The polargroups of SA may interact with the oxygen-containing groups ongraphene via hydrogen bonding to enhance the stability of the as-prepared composite structure.

3.2 Electrochemical conditions for detection of Pb2©

3.2.1 Effect of SA on SWNTsSA can be an effective dispersant for the preparation of SWNTs

suspension.17 As observed by SEM (Fig. 2), the degree of dispersionwas obviously increased upon the addition of SA. On the otherhand, based on the difference between the peak and background ofthe stripping current by DPV, the sensitivity of electrode materialwas improved after being modified with SA in an appropriateconcentration (Fig. 6).

The structure of the as-prepared SA/SWNTs composite shouldbe very stable since SA tends to spontaneously wind around theSWNTs, resulting in a compact helical conformation driven by vander Waals force and hydrogen bonding interactions.17 As shown inFig. 6, SA not only influenced the peak position and intensity, butalso affected the background current of DPV graph. The addition ofSA could unclose the SWNTs bundles, accelerating the dissolutionof Pb metal; thereby influence the stripping potential peak.

However, too much SA could impact the conductivity of theSA/SWNTs modified electrodes. It can be seen that the optimalconcentration of SA for modifying SWNTs-GCE to afford themaximum peak intensity was 0.5wt%.3.2.2 Effect of SA on graphene

As shown in the DPV of G-GCE (Fig. 7), there is a strongbackground current with the blank sample. After compositing withSA, the shape of the curve has changed greatly, that is the peak wasnarrowed and the background current was reduced and stabilized.SA may help to stabilize the electrode surface, and correspondinglyinhibit some electrochemical reactions. As a result, the blank currentof graphene electrode was reduced and the signal-noise-ratio (SNR)was increased.3.2.3 Influence of deposition time

The effect of the deposition time on the peak current of1 © 10¹6mol/L Pb2+, Cu2+ and Cd2+ on the SA/SWNTs/Gcomposite modified electrode had been tested in the range from 1to 18min. Figure 8 shows that the response of the stripping peakcurrent of Pb2+ enhanced from 1min up to 5min. This is becausethat more metal ion was accumulated on the surface of the SA/SWNTs/G-GCE with increasing deposition time. However, whenthe deposition time is beyond 5min, the increasing of the strippingpeak current slowed down. Thus, a deposition time of 5min was

Figure 5. Raman spectrum of SA/SWNTs/G composite meas-ured with a 532 nm laser source.

Figure 6. DPV of SA/SWNTs (CSA = 0 to 1.0wt%) modifiedelectrode in 1 © 10¹6mol/L Pb2+ solution at pH = 4.6 at adepositing potential of ¹1.2V for 5min, DPV scanning range:¹1.20V to 0V. With a step increment of 5mV, amplitude of 80mV,and pulse period of 0.2 s.

Figure 7. DPVof SA/G (CSA = 0 to 0.5wt%) modified electrodein 1 © 10¹6mol/L Pb2+ solution at pH = 4.6 at a depositingpotential of ¹1.2V for 5min, DPV scanning range: ¹1.20V to 0V.With a step increment of 5mV, amplitude of 80mV, and pulseperiod of 0.2 s.

Electrochemistry, 83(2), 1–7 (2015)

14-E00085

4

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

chosen for all experiments in order to ensure the complete reductionof lead and other metal ions on SA/SWNTs/G-GCE. The effect ofthe deposition time on the peak current of Cd2+ and Cu2+ were alsoshown in Fig. 8.

3.3 Electrochemical performance of SA/SWNTs/G compositemodified electrode

To evaluate the electrochemical performance of SA/SWNTs/Gcomposite for detection of metal ion, DPASV method was usedwith DPV scanning range from ¹1.2V to 0V in the solution of1 © 10¹6mol/L Pb2+ (pH = 4.6) at a depositing potential of ¹1.2Vfor 5min (With a step increment of 5mV, amplitude of 80mV, andpulse period of 0.2 s). The DPVs of bare GCE, SA modified GCE(SA-GCE, CSA = 0.5wt%), SWNTs modified GCE (SWNTs-GCE),hydrothermal reduced graphene oxide modified GCE (G-GCE) andSA/SWNTs/G-GCE (CSA = 0.5wt%) are shown in Fig. 9.

As can be seen, the modification of SA and SWNTs led to anincreasing stripping peak [Figs. 9(b) and 9(c)]. The reason for theincrease of SWNTs-GCE is the nanoscale dimensions of the SWNTsbundle structure and the electronic structure. For SA, it is due to thechelation with metal ions. Obviously, the highest stripping peak

currents for Pb2+ were obtained on the SA/SWNTs/G-GCE[Fig. 9(d)], which means that the concentration of Pb metal atSA/SWNTs/G-GCE was higher than that at SWNTs-GCE, SA-GCEand G-GCE. It can be explained that the effective insertion of SWNTspacers in the graphene/SWNTs composite reduced restacking ofgraphene sheets and also produced additional intra-pores, resultinglarger surface area and more porosity in the structure (As obtained inBET results). In addition, as a result, more functional groups, suchas carboxylic and hydroxyl groups and graphene surface area werealso exposed to the absorption of metal ions.33,34

The DPV curves were recorded at SA/SWNTs/G compositemodified electrode in acetate buffer (pH = 4.6) containing 1 © 10¹6

mol/L for Cd2+, Pb2+ and Cu2+ respectively. As shown in Fig. 10,compared to GCE, SWNTs-GCE and G-GCE, SA/SWNTs/G-GCEgave the highest and completely separated peaks for three heavymetal ions, which indicated that this composite modified electrodecan also be used for simultaneous detection of multi-metal ions.

Under the optimum experimental conditions selected, the SA/SWNTs/G-GCE was applied for the determination of Pb2+, Cd2+

and Cu2+ by DPASV. The stripping peak current increased withincreasing concentrations of Pb2+, Cd2+ and Cu2+ as shown inFig. 11. The linear ranges were found to be 1 © 10¹9 to 1 ©10¹5mol/L, 1 © 10¹7 to 8 © 10¹6mol/L and 2 © 10¹7 to 2 © 10¹6

mol/L for Pb2+, Cd2+ and Cu2+, respectively. The calibration curvesand correlation coefficients were y = 1.7721 + 8.75331 © 106x,R = 0.9960, y = 0.1154 + 1.7223 © 106x, R = 0.9988 and y =¹0.7007 + 4.7506 © 106x, R = 0.9928 for Pb2+, Cd2+ and Cu2+,respectively (x: concentration/mol/L, y: peak current/µA). Thelimits of detection were 2.0 © 10¹11mol/L for Pb2+, 7.5 © 10¹10

mol/L for Cd2+ and 6.2 © 10¹9mol/L for Cu2+ (S/N = 3).The SA/SWNTs/G/GCE retained 95% of its initial response

after 2 weeks of storage in dry conditions. Such stability isacceptable for most practical applications. The intra-reproducibilitywas measured with the relative standard deviation (RSD) of 1.2%(n = 5). While the inter-reproducibility for five different modifiedelectrodes was found to be very well with a RSD of 1.8%.

The analytical performances of the proposed SA/SWNTs/Gmodified glassy carbon electrode were compared with other heavymetal ions sensing systems based on self-assembly of graphene,which are summarized in Table 1. Compared with the literaturereports on the determination of Pb2+, Cd2+, Cu2+, the SA/SWNTs/G composite modified electrode showed excellent performance witha wider linear range and lower detection limit.

Figure 8. Effect of the deposition time on stripping peak currentof 1 © 10¹6mol/L Pb2+, Cu2+ and Cd2+ on the SA/SWNTs/G(CSA = 0.5wt%) composite modified GCE. Deposition potential is¹1.2V, DPV scan interval: ¹1.20V to 0V in pH = 4.6 acetatebuffer solution. With a step increment of 5mV, amplitude of 80mV,and pulse period of 0.2 s.

Figure 9. DPV of different modified electrodes: (a) bare GCE,(b) SA-GCE (c) SWNTs-GCE, (d) SA/SWNTs/G-GCE and (e) G-GCE. DPV scanning range is from ¹1.2V to 0V in the solution of1 © 10¹6 mol/L Pb2+ (pH = 4.6) at a depositing potential of ¹1.2Vfor 5min. With a step increment of 5mV, amplitude of 80mV, andpulse period of 0.2 s.

Figure 10. DPV response of 1 © 10¹6mol/L each of Cd2+, Pb2+

and Cu2+ in acetate buffer (pH = 4.6) at (a) GCE, (b) SWNTs-GCE,(c) SA/SWNTs/G-GCE and (d) G-GCE. Deposition potential:¹1.20V, deposition time: 5min, scanning range is from ¹1.2V to0V. With a step increment of 5mV, amplitude of 80mV, and pulseperiod of 0.2 s.

Electrochemistry, 83(2), 1–7 (2015)

14-E00085

5

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

4. Conclusions

In summary, graphene/SWNTs composite with SA (SA/SWNT/G) has been successfully prepared for modification of glassy carbonelectrode. SA/SWNT/G-GCE displayed excellent performance inthe detection of Pb2+, Cd2+, Cu2+ ions. Structural characterizationsusing SEM, Raman and XPS suggested that the SA modifiedSWNTs could act as effective support structure between thegraphene layers. The stripping current response of DPASV was

improved 77% compared with the GCE blank electrode. It has beenfound that the composite structure of SA/SWNT/G played apositive role for both the electrochemical properties and thesensitivity of the electrode materials. The SA/SWNT/G-GCEshowed excellent performance for detection of Pb2+, Cd2+, Cu2+

ions with wider linear range and lower detection limit. Thereproducibility measurement with the same electrode gave a RSDof 1.2% (n = 5), while the one for different electrodes was found tohave a RSD of 1.8%.

Table 1. Comparison of some typical electroanalysis methods for determination of heavy metal ions.

Electrode and method Linear detection range Limit of detection Experimental condition References

Differential pulse anodic strippingvoltammetry (DPASV) at the Bimodified graphene-poly (sodium4-styrenesulfonate) composite filmscreen printed electrode (GR/PSS/Bi/SPE)

0.5–120µg/Lfor Pb2+ and Cd2+

0.089 µg/L for Pb2+

0.042 µg/L for Cd2+

2min accumulation at the potentialof ¹1.3V in acetate buffer solution(pH = 4.5) with 500 µg/L Bi3+ anddifferent concentrations of metalanalyte.

35

Square wave anodic stripping voltam-metry (SWASV) at the glassy carbonelectrode modified with Bi, Nafionand the 2-mercaptoethanesulfonate(MES)–tethered polyaniline (PANI)

0.1–30µg/L for Pb2+

0.1–20µg/L for Cd2+0.05 µg/L for Pb2+

0.04 µg/L for Cd2+

5min accumulation at the potentialof ¹1.2V in acetate buffer solution(pH = 4.0)

36

Cyclic voltammetry (CV) at novelpoly(crystal violet)/graphene-modified glassy carbon electrode(PCV/G/GCE)

2.00 © 10¹8–1.95© 10¹5mol/L

for Pb2+

4.00 © 10¹8–5.58© 10¹5mol/L

for Cd2+

2.00 © 10¹8mol/Lfor Pb2+

4.00 © 10¹8mol/Lfor Cd2+

Cyclic scans at the potential of¹2.0V Vs. Ag/AgCl in acetatebuffer solution (pH = 4.6) between¹1.6 and 0.4V until stable voltam-mograms that were obtained

37

DPASV at a glassy carbon electrodeby single-step electrodeposition ofgraphene (GR), gold nanoparticles(AgNPs), and chitosan (CS)

0.5–100 µg/L for Pb2+ 1 ng/L for Pb2+10min accumulation at the potentialof ¹1.2V in acetate buffer solution(pH = 4.5)

13

SWASV at simultaneous determinationof Pb2+ in siliceous spicules of marinesponges, directly in the hydrofluoricacid solution

Up to at least 4 µg/Lfor Pb2+ and Cd2+

Up to at least 20 µg/Lfor Cu2+

3.6 ng/L for Pb2+

5.8 ng/L for Cd2+

4.3 ng/L for Cu2+

5min accumulation at the potentialof ¹1.1V in HF solution of spicules(pH = 1.9)

38

SWASV at Graphene thin film elec-trodes synthesized by thermallytreating co-sputtered nickel–carbonmixed layers

0.3–1.7 µM for Pb2+,Cd2+ and Cu2+

0.3 µM for Pb2+,Cd2+ and Cu2+

3min accumulation at the potentialof ¹1.0V in 0.1M acetate buffersolutions modified with bismuthions (pH = 5.3)

39

SWASV at Graphene ultrathin filmelectrode

7–1200 nM for Pb2+ 7 nM for Pb2+3min accumulation at the potentialof ¹1.0V in acetate buffer solutions(pH = 5.3) containing 0.1M KNO3

40

SWASV at the in situ bismuth-modified graphene-carbon pasteelectrode (Bi-GCPE)

0.10–50.0 µg/Lfor Pb2+ and Cd2+

0.07 µg/L for Pb2+

0.04 µg/L for Cd2+

3min accumulation at the potentialof ¹1.2V in HCl (0.01–0.05mol/L)vs. Ag/AgCl

11

SWASV at a gold nanoparticle-graphene-cysteine compositemodified bismuth film electrode

0.50–40.0 µg/Lfor Pb2+ and Cd2+

0.05 µg/L for Pb2+

0.10 µg/L for Cd2+

13.3min accumulation at the poten-tial of ¹1.2V in measured in aceticbuffer solution with ABS and Bi3+

(pH = 4.5)

12

DPASV at the SA/SWNTs/Gmodified glassy carbon electrode

1 © 10¹9 to1 © 10¹5mol/L

for Pb2+,1 © 10¹7 to

8 © 10¹6mol/Lfor Cd2+,2 © 10¹7 to

2 © 10¹6mol/Lfor Cu2+

2.0 © 10¹11mol/Lfor Pb2+

7.5 © 10¹10mol/Lfor Cd2+

6.2 © 10¹9mol/Lfor Cu2+

5min deposition at the potential of¹1.2V in pH 4.6 acetate buffersolutions

This work

Electrochemistry, 83(2), 1–7 (2015)

14-E00085

6

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

Acknowledgment

We greatly appreciate the financial support JST ALCA Program,JSPS Grants-in-Aid for Scientific Research 22310074, and the

Nanotechnology Network Project of the Ministry of Education,Culture, Sports, Science and Technology (MEXT), Japan, NationalNatural Science Foundation of China (21275082, 21405086,21475071, 21203228 and 81102411), the National Key BasicResearch Development Program of China (973 special preliminarystudy plan, Grant no. 2012CB722705), the Natural ScienceFoundation of Shandong (ZR2011BZ004, ZR2011BQ005).

The authors also appreciate the sincere help from Dr. JianguoTang, Dr. Zonghua Wang, Dr. Jixian Liu, Dr. Jingquan Liu and theInternship Program 2013 of the National Institute for MaterialsScience, Japan.

References

1. M. Pumera, Chem. Rec., 9, 211 (2009).2. M. Pumera, Mater. Today, 14, 308 (2011).3. D. A. C. Brownson and C. E. Banks, Analyst, 136, 2084 (2011).4. B. J. Sanghavi, S. M. Mobin, P. Mathur, G. K. Lahiri, and A. K. Srivastava,

Biosens. Bioelectron., 39, 124 (2013).5. K. Schachl, H. Alemu, K. Kalcher, J. Jezkova, I. Svancara, and K. Vytras, Analyst,

122, 985 (1997).6. Y. Y. Shao, J. Wang, H. Wu, J. Liu, I. A. Aksay, and Y. H. Lin, Electroanalysis, 22,

1027 (2010).7. F. Banhart, J. Kotakoski, and A. V. Krasheninnikov, ACS Nano, 5, 26 (2011).8. F. F. Zhang, J. Tang, Z. H. Wang, and L. C. Qin, Chem. Phys. Lett., 590, 121

(2013).9. A. J. S. Ahammad, J. J. Lee, and M. A. Rahman, Sensors, 9, 2289 (2009).10. A. R. Sadrolhosseini, A. S. M. Noor, A. Bahrami, H. N. Lim, Z. A. Talib, and

M. A. Mahdi, PLoS ONE, 9, e93962 (2014).11. W. Wonsawat, S. Chuanuwatanakul, W. Dungchai, E. Punrat, S. Motomizu, and O.

Chailapakul, Talanta, 100, 282 (2012).12. L. Zhu, L. L. Xu, B. Z. Huang, N. M. Jia, L. Tan, and S. Z. Yao, Electrochim.

Acta, 115, 471 (2014).13. Z. Z. Lu, S. L. Yang, Q. Yang, S. L. Luo, C. B. Liu, and Y. H. Tang, Microchim.

Acta, 180, 555 (2013).14. V. V. Kobak, M. V. Sutkevich, I. G. Plashchina, and E. E. Braudo, J. Inorg.

Biochem., 61, 221 (1996).15. T. Gotoh, K. Matsushima, and K.-I. Kikuchi, Chemosphere, 55, 135 (2004).16. G. R. Seely and R. L. Hart, Macromolecules, 7, 706 (1974).17. Y. Z. Liu, C. Chipot, X. G. Shao, and W. S. Cai, J. Phys. Chem. B, 114, 5783

(2010).18. J. Li and I. Zhitomirsky, J. Mater. Process. Technol., 209, 3452 (2009).19. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 80, 1339 (1958).20. C. L. Hong, A. Wong, and M. Pumera, RSC Adv., 2, 6068 (2012).21. T. Kuila, S. Bose, P. Khanra, A. K. Mishra, N. H. Kim, and J. H. Lee, Biosens.

Bioelectron., 26, 4637 (2011).22. M. Ionita, M. A. Pandele, and H. Iovu, Carbohydr. Polym., 94, 339 (2013).23. N. Thayumanavan, P. Tambe, G. Joshi, and M. Shukla, Compos. Interfaces, 21,

487 (2014).24. C. H. An Wong, A. Ambrosi, and M. Pumera, Nanoscale, 4, 4972 (2012).25. T. C. Chiang and F. Seitz, Ann. Phys.-Berlin, 10, 61 (2001).26. S. Yumitori, J. Mater. Sci., 35, 139 (2000).27. C. L. Weitzsacker and P. M. A. Sherwood, Surf. Interface Anal., 23, 551 (1995).28. S. Chattopadhyay, S. K. Banerjee, and A. K. Mitra, J. Indian Chem. Soc., 79, 906

(2002).29. A. Das, B. Chakraborty, and A. Sood, Bull. Mater. Sci., 31, 579 (2008).30. D. Yang, A. Velamakanni, G. Bozoklu, S. Park, M. Stoller, R. D. Piner, S.

Stankovich, I. Jung, D. A. Field, C. A. Ventrice, and R. S. Ruoff, Carbon, 47, 145(2009).

31. J. R. Dennison, M. Holtz, and G. Swain, Spectroscopy, 11, 38 (1996).32. T. V. Cuong, V. H. Pham, Q. T. Tran, S. H. Hahn, J. S. Chung, E. W. Shin, and

E. J. Kim, Mater. Lett., 64, 399 (2010).33. D. Du, J. Liu, X. Y. Zhang, X. L. Cui, and Y. H. Lin, J. Mater. Chem., 21, 8032

(2011).34. B. Wang, Y. H. Chang, and L. J. Zhi, New Carbon Mater., 26, 31 (2011).35. H. F. Chao, L. Fu, Y. Z. Li, X. L. Li, H. J. Du, and J. S. Ye, Electroanalysis, 25,

2238 (2013).36. L. Chen, Z. H. Su, X. H. He, Y. Liu, C. Qin, Y. P. Zhou, Z. Li, L. H. Wang, Q. J.

Xie, and S. Z. Yao, Electrochem. Commun., 15, 34 (2012).37. M. F. Chen, M. Y. Chao, and X. Y. Ma, J. Appl. Electrochem., 44, 337 (2014).38. C. Truzzi, A. Annibaldi, S. Illuminati, E. Bassotti, and G. Scarponi, Anal. Bioanal.

Chem., 392, 247 (2008).39. Z. M. Wang, P. M. Lee, and E. Liu, Thin Solid Films, 544, 341 (2013).40. Z. M. Wang and E. J. Liu, Talanta, 103, 47 (2013).

Figure 11. DPV curves by SA/SWNTs/G composite modifiedelectrode at different concentration of (a) Pb2+, (b) Cd2+ and(c) Cu2+ ions. Insert are linear fitting graphs of the DPV peakcurrent by SA/SWNTs/G composite modified electrode at differentconcentration of Pb2+ from 1 © 10¹9 to 1 © 10¹5mol/L, Cd2+ from1 © 10¹7 to 8 © 10¹6mol/L and Cu2+ from 2 © 10¹7 to 2 © 10¹6

mol/L, respectively. DPV scan interval is from ¹1.20V to 0V inpH = 4.6 acetate buffer solution at a depositing potential of ¹1.2Vfor 5min, with a step increment of 5mV, amplitude of 80mV, andpulse period of 0.2 s.

Electrochemistry, 83(2), 1–7 (2015)

14-E00085

7

PROOF 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66