research 1..9

(1)

Sodium Hypochlorite and Sodium Bromide

Individualized and Stabilized Carbon Nanotubes in Water

Item Type Article

Authors Xu, Xuezhu;Zhou, Jian;Colombo, Veronica;Xin, Yangyang;Tao, Ran;Lubineau, Gilles

Citation Xu X, Zhou J, Colombo V, Xin Y, Tao R, et al. (2017) Sodium Hypochlorite and Sodium Bromide Individualized and Stabilized Carbon Nanotubes in Water. Langmuir. Available: http://

dx.doi.org/10.1021/acs.langmuir.7b00850.

Eprint version Post-print

DOI 10.1021/acs.langmuir.7b00850

Publisher American Chemical Society (ACS)

Journal Langmuir

Rights This document is the Accepted Manuscript version of a Published Work that appeared in final form in Langmuir, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see http://pubs.acs.org/doi/abs/10.1021/

acs.langmuir.7b00850.

Download date 2023-12-03 19:25:48

Link to Item http://hdl.handle.net/10754/625831

(2)

1

Sodium Hypochlorite and Sodium Bromide Individualized and

2

Stabilized Carbon Nanotubes in Water

3

Xuezhu Xu,

^†

Jian Zhou,

^†

Veronica Colombo,

^†^,^‡

Yangyang Xin,

^†

Ran Tao,

^†

and Gilles Lubineau*

^,^†

4^†Physical Science and Engineering Division (PSE), COHMAS Laboratory, King Abdullah University of Science and Technology

5 (KAUST), Thuwal 23955-6900, Saudi Arabia

6^‡Chemical and Materials Engineering, University of Padua, Padova 35122, Italy

7 ABSTRACT: Aggregation is a major problem for hydro-

8 phobic carbon nanomaterials such as carbon nanotubes

9 (CNTs) in water because it reduces the eﬀective particle

10 concentration, prevents particles from entering the medium,

11 and leads to unstable electronic device performances when a

12 colloidal solution is used. Molecular ligands such as surfactants

13 can help the particles to disperse, but they tend to degrade the

14 electrical properties of CNTs. Therefore, self-dispersed

15 particles without the need for surfactant are highly desirable.

16 We report here, for theﬁrst time to our knowledge, that CNT

17 particles with negatively charged hydrophobic/water interfaces

18 can easily self-disperse themselves in water via pretreating the

19 nanotubes with a salt solution with a low concentration of

20 sodium hypochlorite (NaClO) and sodium bromide (NaBr). The obtained aqueous CNT suspensions exhibit stable and superior

21 colloidal performances. A series of pH titration experiments conﬁrmed the presence and role of the electrical double layers on the

22 surface of the salted carbon nanotubes and of functional groups and provided an in-depth understanding of the phenomenon.

23

■

INTRODUCTION

24CNTs are highly important materials with many applications.¹

25However, one of the most diﬃcult challenges associated with

26CNTs is their solubility because individual nanotubes naturally

27stack themselves into “ropes” due to van der Waals forces.²

28Thus, to fully exploit their nanoscale functions, it is especially

29critical to solubilize carbon nanotubes appropriately.

30 Solubilization can be realized by functionalizing the surface of

31the carbon nanotube covalently or by noncovalent interactions

32with a proper medium.³⁻⁷Covalent functionalization involves

33the use of chemical techniques, such as reﬂuxing and sonication

34for a long time.⁸ For example, CNT’s carboxylation typically

35requires an ineﬃcient and time-consuming process of reﬂux in

36concentrated nitric acid for 10−50 h.^9,10 Solubilization of

37CNTs in common solvents such as N-methyl-2-pyrrolidone

38(NMP) and N,N-dimethylformamide (DMF) works as well,

39owing to a match of surface free energy (68 mJ m⁻²) between

40the CNTs and the dispersant.³Noncovalent approaches include

41the use of dispersants such as anionic, cationic, and nonionic

42surfactants to wrap the nanotubes for eﬃcient dispersion in

43water.¹¹⁻¹⁴ Ionized CNT surfaces that help individualizing/

44dispersing/stabilizing CNT in solutions are also attracting

45certain attention.¹⁵⁻¹⁸However, studies mentioned above show

46that using covalent functionalization or a surfactant deteriorates

47the electrical properties of theﬁnal product, either because of

48the degradation of the CNT structure or by the introduction of

49an insulating interface around the CNTs. Thus, improving

50eﬃciency in solubilizing CNTs in water via a less destructive

approach with a high time eﬃciency would be considered a 51

signiﬁcant step toward fully exploring the unique properties and52

broadening the applications of CNTs. Aiming for this, we 53

report here a strategy to achieve excellent dispersibility of 54

CNTs without compromising their microstructure. In this 55

study, we dispersed CNTs in sodium hypochlorite (NaClO) 56

conjugated with sodium bromide (NaBr) aqueous solution. 57

Then, treated CNTs were dispersed in water again to obtain 58

CNT aqueous colloidal solutions. We also characterized their 59

colloidal properties to understand the dispersibility. 60

■

EXPERIMENTAL SECTION 61

We used a mixture of nonpolar single- and double-walled carbon 62

nanotubes (SDWNTs) in this study. It was purchased from Cheap 63

Tubes Company as prepared by the combustion chemical vapor 64

deposition (CCVD) process. The SDWNTs had outer diameters of 65

1−4 nm, purity >99 wt %, and an average length of 3−30μm. The 5% 66

sodium hypochlorite solution and sodium bromide powder were 67

purchased from RICCA Chemical Company. 37% pure hydrochloric 68

acid solution and sodium hydroxide pellets were both purchased from 69

Sigma-Aldrich Company. SDWCNT was used throughout the study 70

except the experiments demonstrated inFigure 3j−l. To conﬁrm the 71

eﬀectiveness of our dispersant, semiconductive CoMoCAT-type 72

singled-walled carbon nanotubes (SWCNT) were used for UV−vis− 73

NIR optical measurement (Figure 3j−l). The SWCNT (Product 74

#773735) was purchased from Sigma-Aldrich with a (6,5) chirality, 75

Received: March 14, 2017 Revised: September 10, 2017 Published: September 20, 2017

Article pubs.acs.org/Langmuir

(3)

76>95% carbon basis, and 0.78 nm average diameter. To disperse the

77SDWNTs, we mixed 0.064 g of SDWNTs, 30 mL of aqueous NaClO

78solution, and 2.073 g of NaBr and sonicated the mixture in a regular

79bath solicitor for 60 min. Then, we removed all the supernatant by

80using a centrifuge operated at 4000 rpm for 2 min, leaving behind

81SDWNT slurry, to which we then added 30 mL of pure water.

82Afterward, we sealed the resulting salted-SDWNT slurry in a bag made

83of Spectra/Por1 dialysis membrane (64 mm tubing diameter and 6−8

84kDa pore width) to reduce residual chemicals. The dialysis bag was

85immersed in 5000 mL of deionized water. Puriﬁcation completed after

86we changed the 5000 mL water every 12 h for 10 days. The

concentration of the puriﬁed salted-SDWNT aqueous colloidal87

solution was adjusted to its original value (2 mg mL⁻¹). To adjust88

the pH of various suspensions, we added 0.01 M HCl or NaOH 89

aqueous solution to the salted-SDWNT or neat-SDWNT colloidal 90

solution. Vacuum ﬁltration to obtain free-standing salted-SDWNT 91

ﬁlms was performed using a 47 mm diameter vacuumﬁlter assembly 92

(Wheaton Company). The ﬁlter we used was made of 47 mm 93

diameter polycarbonate (PC) ﬁlm with 0.05 μm pores (Whatman 94

Company). Afterﬁltration, the salted-SDWNTﬁlms were then dried 95

in a vacuum oven for 12 h at 100°C, which were used for FTIR and 96

Raman spectroscopy measurements. 97

Figure 1.(a) Schematic diagram of our approach to pretreat SDWNTs. The SDWNTs were dispersed in a salted solution followed by puriﬁcation and dispersion of the salted-SDWNTs in fresh water to obtain a highly concentrated CNT colloidal solution. (b) Scheme of pH titration experiments to study surface charges on salted-SDWNTs.

Figure 2.Preparation of salted-SDWNT aqueous colloidal solution at 2 mg mL⁻¹. Step 1: dispersion process of dissolving SDWNT powder into NaClO/NaBr salted solution at 2 mg mL⁻¹SDWNT concentration, by using bath sonication for 60 min. Step 2: the puriﬁcation process of removing NaClO/NaBr solution from dispersed SDWNTs slurry by dialysis. Step 3: demonstration of colloidal stability of the salted-SDWNT aqueous colloidal solution at 2 mg mL⁻¹, with photos taken every day for 1 week.

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX B

(4)

98 Scanning electron microscopy (SEM) was performed using a

99Quanta 3D (FEI Company). TEM images were taken using a Tecnai

100Twin microscope (FEI). FTIR measurements were taken on Thermo

101Scientiﬁc equipment coupled with a diamond smart accessory and

102measured on salted-SDWNT ﬁlms or neat-SDWNT powders. The

103Raman spectra were recorded using a Horiba/Jovin Yvon Lab Ram

104system equipped with a 473 nm laser source. Laser light scattering

105particle sizing on aqueous solutions of SDWNTs was done with a

106Malvern Zetasizer Nano system. Measurements for ζ-potential of

107various CNT colloidal solutions were performed on the same

108equipment with irradiation from a 633 nm He−Ne laser.

109Thermogravimetric analysis (TGA) was scanned by a NETZSCH

110TG, 208 F1 machine from 27 to 880°C at 20 K/min under a nitrogen

111atmosphere protection. X-ray photoelectron spectroscopy (XPS) was

112done in a Kratos Axis Ultra DLD spectrometer equipped with a

113monochromatic Al KαX-ray source (hν= 1486.6 eV) operating at 150

114W, a multichannel plate, and delay line detector under a vacuum of

115∼10−9 mbar. All spectra were recorded using an aperture slot of 300

116μm×700μm. Survey spectra were collected using a pass energy of

117160 eV and a step size of 1 eV. A pass energy of 20 eV and a step size

118of 0.1 eV were used for the high-resolution spectra. Samples were

119mounted in ﬂoating mode in order to avoid diﬀerential charging.

120Charge neutralization was required for all samples. Binding energies

were referenced to the sp²-hybridized (CC) carbon for the C 1s 121

peak set at 284.5 eV from carbon nanomaterials. Ultraviolet−visible122

(UV−vis) spectroscopy measurements were measured by a Cary100 123

ConC. 124

■

RESULTS AND DISCUSSION 125

126 f1

Figure 1a shows that by using regular bath sonication for 60 min in low-concentration dispersants of NaClO and NaBr, high 127

concentration up to 2 mg mL⁻¹ aqueous colloidal solutions128

with single- and double-walled carbon nanotubes (SDWNTs) 129

can be achieved. First, one of the dispersants, NaClO, was 130

dissolved in the aqueous solution at 5 wt %. NaBr powder was 131

then added at a molar ratio of 1 to 1 with respect to NaClO. 132

Figure 1b shows the pH titration experiment results on 133

obtained salted-SDWNTs, which will be used later to 134

understand the underlying phenomenology. 135

136 f2

Figure 2shows more details about the preparation of salted- SDWNTs. During the mixing step, we evidenced a solubiliza- 137

tion of the vast majority of SDWNTs in solution within 2 min. 138

After 60 min, the SDWNTs appeared well suspended in the 139

NaClO/NaBr salted solutions. Then, the salted solution with 140

Figure 3. (a−h) Dissolution process of dropping 5 mL of salted-SDWNT colloidal solution (2 mg mL⁻¹) into fresh water to observe its homogeneity in a diluting process. (i) UV−vis spectra of salted-SDWNT aqueous suspensions as treated with diﬀerent salting times. Pure-SDWNT suspended in water was also presented as a control. All samples were diluted to 0.002 mg mL⁻¹for the UV−vis measurement. (j) UV−vis−NIR spectra of aqueous“CoMoCAT”-type SWCNTs suspensions, which were previously dispersed by either NaClO/NaBr or each one of them as a dispersant. Samples were diluted 100 times for the measurements. (k, l) UV−vis−NIR spectra of aqueous UV−vis−NIR spectra of aqueous

“CoMoCAT”-type SWCNT suspensions. These SWCNT suspensions were dispersed by diﬀerent NaClO/NaBr concentrations, i.e., from 0.67 to 0.000 067 M (NaClO % in water). NaBr:NaClO ratio remained 1:1 in all samples.

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX C

(5)

141dissolved SDWNTs was puriﬁed (step 2). This process

142included dispersant removal by centrifuge to extract SDWNT

143slurry. The SDWNT slurry together with additional fresh water

144was then sealed in a dialysis bag, which was immersed in a

145container ﬁlled with a large volume of deionized water, to

146remove the majority of NaClO/NaBr chemicals. The dialysis

147took 10 days by changing the dialysis water twice a day. Finally,

148we adjusted the puriﬁed salted SDWNTs to the initial

149concentration of 2 mg mL⁻¹. As shown in step 3, the salted-

150SDWNTs remained stable for a week without phase separation

151or the formation of aggregates, implying that the salted-

152SDWNTs dispersed well in the aqueous solution.

153 Next, we diluted the salted-SDWNT aqueous colloidal

f3 154solution from 2 to 0.01 mg mL⁻¹ (Figure 3) to see its

155homogeneity. For this, 5 mL of the salted-SDWNT colloidal

156solution was dropped into 1000 mL of fresh water. A

157spontaneous diﬀusion completed in approximately 205 s,

until we saw a homogeneous SDWNT dispersion. A similar 158

phenomenon for the diﬀusion of CNTs in water was also159

reported.^1,9However, in that report, it only became possible by160

pretreating SDWNTs in oleum (>100% fuming sulfuric acid), 161

which caused an expansion of the tube−tube distance between 162

CNTs. Very likely, our salted-SDWNTs have been modiﬁed,163

which overcomes the van der Waals force attraction that tends 164

to attract SDWNT particles. This modiﬁcation ensures the free165

Brownian motion of the salted-SDWNT particles in water. 166

Good dispersion status of salted-SDWNT in pure water was 167

veriﬁed by UV−vis absorption measurement on various diluted168

suspensions (Figure 3i). All salted-SDWNT suspensions with 169

diﬀerent treatment time ranging from 5 to 60 min showed high170

absorption, suggesting the content of particles in solution was 171

higher than unmodiﬁed SDWNT suspended in water, which 172

was believed to be ascribed to a highly individualized dispersion 173

state of SDWNTs. 174

Figure 4.(a) AFM and (b) TEM images of salted-SDWNT particles. (c) XPS full spectra of salted-SDWNT showing absorption of additional ions.

(d) XPS C 1s spectrum on salted-SDWNT showing a relatively small amount of oxidation, compared to H₂SO₄/KMnO₄oxidized MWCNTs as reported by Kosynkin et al. (inset).²⁷The curves in the inset represented XPS carbon 1s spectra of oxidized nanoribbons from MWCNTs (red) and the reduced nanoribbons (blue). (e) SEM image and EDS (inset) of salted-SDWNT showing undamaged tube length. (f) TGA curves of pure- SDWNT and salted-SDWNT comparing their thermal stability. (g) FT-IR spectra of salted-SDWNTs compared with pure SDWNTs. (h) Raman spectra of original SDWNT, original SDWNT after sonication, and puriﬁed salted-SDWNTﬁlm after treatment.

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX D

(6)

175 To clarify the eﬀect of the salts on the dispersion of the

176carbon nanotubes, we performed two experiments. First, we

177observed the optical transition (S11, S22, etc.) measured by

178UV−vis−NIR on another type of tube with small diameter

179(Figure 3j). InFigure 3j, after disintegration, the blue-shift of

180peaks of SWNTs is observed, which behaves similarly to

181surfactant-dispersed ones.¹⁹Second, we measured the quality of

182the dispersion of the small diameter CNTs upon diﬀerent

183amounts of NaClO/NaBr addition (Figure 3k,l). We observed

184diﬀerent absorbance at all wavelengths (Figure 3k,l) when

185changing the concentration of NaClO/NaBr. Obviously, the

186concentration of NaClO/NaBr inﬂuenced the dispersion of

187carbon nanotubes in water for two reasons: (1) NaClO/NaBr

188are physically “soaked up” by tubes; the more NaClO/NaBr

189cations or anions in water, the longer the tube−tube distance is

190by the eﬀect of intercalation. (2) The higher light absorbance

191could be caused by a stronger oxidizing eﬀect (mainly ClO⁻).

192However, it seems that the diluted salt solution did not disperse

193the SDWNT well; see the 0.0067, 0.000 67, and 0.000 067 M

194samples inFigure 3l. There should be a critical amount of ions

195(threshold at 0.0067 M) surrounding the tubes for it to

196overcome the high surface energy of water and to be fully

197suspended in water.

198 We have not observed severe structural damage or

199morphology change in the salted-SDWNT. Both AFM and

f4 200TEM imaging (Figure 4a,b) prove SDWNT morphology has

201not been modiﬁed by the salt treatment. This is likely either

because the combination of NaClO (5%) and NaBr decreases 202

the oxidation capability of ClO⁻or because NaBr protects the 203

SDWNT. In the work of Yang et al., adequate sodium 204

hypochlorite concentration for preoxidization of multiwalled 205

carbon nanotubes showed that 7% NaClO treatment caused 206

damage to MWNT in 30 min.²⁰ They stated that an excess 207

dosage could destroy the CNT structure. Inversely, low dosage 208

did not induce any change to the CNTs structure. Moreover, 209

even if an occurred slight oxidation of the nanotubes can be 210

seen in the deconvolution of the C 1s spectra (Figure 4d, 211

COOH and CO bonds), this was negligible if compared to212

the one induced by strong acids such as H₂SO₄. The SEM/EDS 213

provided evidence that the salt layer was still there and was not 214

leached out during the cycles of dialysis. EDS data showed that 215

Na and Br weighted 3.5% and 6.4%, respectively, which attested 216

to the presence of a salt layer in the salted-SDWNT (Figure 4e, 217

inset). 218

The SDWNT dispersibility was retained even after the 219

dialysis, suggesting SDWNTs may have charged groups 220

provided by the oxidation using NaClO. Nevertheless, 221

according to our experimental evidence, the salt layer 222

(including either NaBr or NaClO or a conjugation) should 223

also be present during the dispersion and after dialysis, as 224

proved by X-ray photoelectron spectroscopy spectra which 225

showed the presence of Na, Br, and O (Figure 4c). In the TGA 226

curve, a burnout below 100°C for pure-SDWNT powder was 227

due to the evaporation of the moisture absorbed from the 228

Figure 5. (a) ζ-potential plots of neat-SDWNT aqueous suspensions. (b) ζ-potential plots of salted-SDWNT aqueous suspensions. (c) Mild oxidation caused by NaBrO and physical absorption of various ions on CNT surfaces. The electrical double-layer structure depicts the surface charge conﬁguration on a salted-SDWNT particle in water. (d)ζ-potential dependence on pH values. (e) Hydrodynamic particle sizes of both salted- SDWNT and neat-SDWNT aqueous suspensions against pH values. Green region marked the particle sizes <1000 nm.

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX E

(7)

229environment.²¹The mass loss of salted-SDWCNTs at around

230350°C can be attributed to the removal of the groups created

231on the nanotubes after the slight oxidation. The presence of

232absorbed salts can be discerned from the higher mass

233percentage left after heating up to 900 °C in salt-treated

234SDWCNTs (29.3%) as compared to pristine-SDWCNTs (less

235than 20%) (Figure 4f).

236 One important thing to be mentioned is that after treating

237the nanotubes in a solution with such a high molar ratio of salts

238(the molarity of the salted solution is equal to 1.34 M), no

239noticeable attenuation of dispersibility was seen after the

240attempt to remove the ClO⁻ ions from surface of the

241nanotubes, as proved from both the dialysis cycle and the

242dilution shown inFigure 3a−h. Therefore, we postulate that the

243removal of ClO⁻is diﬃcult. The reason why the ClO⁻aﬃnity

244to the SDWNTs is so high can be linked to the interactions

245existing between the SDWNTs (electron donor) and ClO⁻

246ions (electron acceptor) as postulated by Knorr et al.²²In their

247study, SWNTs were adsorbed to a small molecule (electron

248acceptor: o-nitrotoluene carrying =O and O⁻), and this

249changed the solubilization of SWNTs because the energy of

250interaction of the as-created complex was larger than the forces

251of π−π stacking, which in turn made the repulsion between

252nanotubes larger than the tube−tube attraction. Their study has

253similarity to ours, as in our salt system the electron-acceptor

254moieties are being used to induce a noncovalent functionaliza-

255tion of SDWNTs which, combined with the slight oxidation of

256the tubes, allows their eﬃcient and stable dispersion.

257 FT-IR (Figure 4g) measurements proved that the salt-treated

258SDWNTs (used powders were believed not reduced after

259drying at 100°C for overnight) experienced a certain degree of

260covalent surface modiﬁcation with the attachment of carboxylic

261groups. In fact, we can discern a small new absorbance band

262due to COOH stretching (approximately 3300 cm⁻¹) and

263various peaks attributed to the presence of COO⁻in the range

264between 2000 and 1500 cm⁻¹.²³⁻²⁵

265 Raman spectroscopy provides a semiquantitative measure of

266the degree of nanotube destruction by the relative size of the D

267band (disordered carbon) and the G band (graphite). Three

268Raman spectra including the neat-SDWNTs, the SDWNTs

269after only sonication, and the salted-SDWNTs after puriﬁcation

270and drying are shown in Figure 4h. First, the D band appears

271almost ﬂat at sonication, suggesting the starting material

272SDWNTs have a high crystallographic structure.⁸Its intensity

273increases with salt treatment, implying that the salted treatment

274causes a certain degree of modiﬁcation on the nanotubes. Note

275that this change in the D-band intensity is probably reinforced

276by an intercalation of salts into SDWNTs, which changes their

277graphitic perfection.²⁶ In order to explain why the salted-

278SDWNTs easily mixed in water, we proceeded by determining

279the role of the electrolytes.

280 A series of pH titration experiments were performed on these

281solutions to probe the electrical double layers (EDL) of the

282salted SDWNTs, and photos of CNT agglomerations upon pH

283titration are shown in Figure 1c; the experimental results on

f5 284EDL are plotted in Figure 5. As sketched in Figure 5c,

285stabilization of colloidal solutions is believed to be ascribed to

286oxidation and the electrostatic repulsion between colloidal

287particles caused by an EDL, originating from either ion

288enrichment or introduced by chemical bonding (yellow marked

289layer indicates chemical charges introduced by oxidation). Note

290that there is a term called the point of zero charge (PZC)

291characterizing the EDL.²⁸The PZC is the pH value at which a

solid material submerged in electrolyte exhibits zero electrical 292

charge on the particle surface. We use the pH titration 293

approach to probe the PZC values for both colloidal solutions 294

with pristine- and salted-SDWNTs. 295

The zeta (ζ)-potential values, which is the electric potential 296

in the interfacial EDL at the location of the slipping plane 297

relative to a point in the bulkﬂuid away from the interface, are 298

calculated from particle velocities by using the Helmholtz− 299

Smoluchowski approximation:²⁹ 300

μ ζ

= πηϵ V 4 D

m

whereμis the electrophoretic mobility,ηis the viscosity of the 301

solution,ϵmis the dielectrical constant of the medium, andDis302

the electrode separation. 303

Forζ-potentials,Figure 5d shows that a negative charge was304

observed for neat SDWNT aqueous solution at neutral pH∼7, 305

which behaved similarly to previous experimental reports.²⁸306

The presence of a negative charge at neutral or basic pH for 307

hydrophobe/water is universal at interfaces of oil/water, air/ 308

water, ice/water, etc.³⁰ Beattie et al. stated that a variety of 309

other experiments on, e.g., air bubbles, oil droplets in water, 310

and water droplets in air or oil report a negative surface 311

charge.³¹They also used pH-stat experiments in the absence of 312

surfactant to demonstrate quantitatively that hydroxide ions 313

charged and stabilized oil-in-water emulsion droplets.^32,33Most 314

importantly, they concluded that hydroxide was the single 315

largest contribution to the net negative charge on the surface at 316

near-neutral pH under a wide range of ordinary experimental 317

conditions.³⁴Nevertheless, some scholars hold a contradictory 318

point of view particularly for NaOH-titrated oil/water 319

emulsion; e.g., they justiﬁed that negatively charged hydro-320

phobic/water interface was caused by impurity due to presence 321

of byproduct from reaction of fatty acid in oil and 322

hydroxide.^35,36 However, for our colloidal solution containing323

pure electrolyte and SDWNTs, we believe that the enrichment 324

of negative charges at the SDWNTs/water interface plays a 325

critical role to generate the quick “miscibility” of the pristine 326

SDWNT powder into salted solution (step 1) and the good 327

dispersion of pretreated SDWNTs covered with salts in fresh 328

water (step 3), as depicted in Figure 1b. In conclusion, ion 329

absorption is believed to be one of the reasons causing the 330

change in the physical properties (surface tension, binding free 331

energy) at the interfaces as proven by Beattie et al.^34,37 332

In order to understand if each salt used separately could lead333

to a good dispersion, we investigated the quality of the 334

dispersion when using two monosalt solutions. The ﬁrst 335

solution corresponds to 5% aqueous NaClO (0.67 M). The 336

second one contains 2.073 g of NaBr dissolved in 30 g of 337

deionized water (0.67 M). The analysis done with the Zetasizer 338 339 t1

gave the results reported inTable 1. It is clear that while the original salt solution obtained mixing NaClO and NaBr had 340

proved to be an excellent dispersant, this was not true for the 341

Table 1. Average Size andζ-Potential of Particles in Three Diﬀerent Salt Solutions

solution size (nm) st dev (nm) ζ-potential

(mV) st dev (mV)

NaBr 98343 86174 0.76 0.17

NaClO 22497 7779 −1.439 0.94

NaClO + NaBr 588 38 −35.27 0.96

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX F

(8)

342other two solutions. In fact, the ζ-potential when using either

343NaBr or NaClO was almost zero. The low value ofζ-potential

344is consistent with the aggregation of SDWNTs that is detected

345by the increase in average size (which is very high at 98 343 and

34622 497 nm for NaBr and NaClO, respectively). Some studies

347reported that NaClO is an oxidizing agent with a oxidation

348capability higher than H₂O₂.²² However, it is clear from our

349results that the oxidation induced by NaClO alone is not

350enough if the SDWCNTs are treated with our method, as the

351value of ζ-potential is only slightly negative, and that its

352combination with NaBr plays a special role. The synergy

353between NaClO and NaBr is not reported for theﬁrst time. We

354adopted this combination inspired by the well-known TEMPO

355oxidation process (reagent: TEMPO oxidizer, NaClO, NaBr)

356for treating cellulose in paper industry.^38,39Formation of BrO⁻

357from NaClO and NaBr is unstable, which is well-known in the

358TEMPO oxidation mechanism for biopolymers (depicted in

359the drawing). Therefore, it is unlikely there would be BrO⁻

360surrounding the CNT. The mechanism for conversion of

361NaClO and NaBr into NaBrO and the ionic dissociation of

362them into cations and anions and physical absorption process

363onto CNT are shown inFigure 5c.

364 Back to the colloidal phenomenon, in Figure 5b,d, we

365noticed that the curve of salted-SDWNTs has a negative ζ-

366potential value of−50.5 mV, which likely implies presence of

367an intensive surrounding of a tightly bound layer of anions with

368“−”charges. Overall, the experiment to identify ζ-potential at

369neutral pH delivers two messages: First, neat-SDWNT particles

370have a negatively charged hydrophobic/water interface, which

371supports Beattie et al.’s view that a variety of other experiments

372on, e.g., air bubbles, oil droplets in water, and water droplets in

373air or oil report a negative surface charge.³¹ Second, salted-

374SDWNT particles have a very intense negatively charged

375interface resulting a good dispersion.

376 Where do these charges originate? In ourﬁrst step, dissolving

377NaClO/NaBr induces ionization of salts and introduces

378negative anions into the dispersing medium (electrolyte), e.g.,

379OH⁻, ClO⁻, and Br⁻; these ions are believed present in the

380solution when the system is in equilibrium:

+ ↔ + ↔ +

+ +

+ −

NaClO H O NaOH HClO Na OH H ClO

2

381 (1)

382The addition of sodium bromide into sodium hypochlorite

383generates no new ionic species. Based on ζ-potential experi-

384ments, there is likely a strong ionic enrichment that facilitates

385the formation of the electrical double layer (EDL) (Figure 5c).

386Up to now, we are unable to determine what ion species are in

387the double layer. From the literature, hydroxide ions certainly

388play a vital role in the EDL;³⁰ halogen group elements (Cl⁻,

389Br⁻) are proven also capable of entering the EDL.⁴⁰ In

390addition, ClO⁻behaves as a weak acid, which should hold the

391same mechanism as proposed by Dimiew et al.⁴¹They claimed

392a stable colloidal solution of CNTs dispersed in water was

393obtained by sonicating CNTs in hypophosphorous acid

394(H₃PO₂) for a short time. They also claimed that the

395dissociation and formation of anions H₂PO₂⁻ from H₃PO₂

396facilitate a good colloidal stability of CNTs. On top of the

397anion, to form an electrical double layer, there is a thin layer of

398cations, i.e., hydronium H⁺ and sodium Na⁺, which has to

399neutralize the negative surface charge. More speciﬁcally, the

400enrichment of anions on the surface of SDWNTs eventually

401overcomes the van der Waals forces: instead of naturally

aligning themselves together into ropes, the individual 402

nanotubes remain distanced owing to the anionic enrichment 403

and intercalation eﬀect of ions between CNTs.⁴² 404

This negative-charge-enrichment hypothesis was further405

supported by the changes of |ζ|values by pH titration of the 406

salted-SDWNT solution (Figure 5a,b,d). To adjust the pH, 0.1 407

M hydrochloric acid (HCl) or 0.1 M sodium hydroxide 408

(NaOH) solutions were added to the CNT solutions for 409

titration measurements. Here we characterize the system by 410

using IEP = 0 (Figure 5d, green zone). For the neat SDWNTs, 411

PZC = 5.0; at this point, the SDWNTs were surrounded by 412

zero charges. Above IEP, the ζ-potential decreases with an 413

increasing pH, implying that hydroxide ions dissociated from 414

NaOH electrolyte indeed accumulate on the SDWNT surface. 415

Below IEP, acidiﬁcation results in an accumulation of positive 416

charge on the neat SDWNTs. 417

On the contrary, in salted-SDWNTs theζ-potential becomes 418

more negative with increasing pH until a maximum is reached 419

and then starts to increase again. First, acidiﬁcation caused a 420

neutralization of the salts, as evidenced by an increased ζ- 421

potential from−50.5 mV (pH = 7) to−20.1 mV (pH = 1.8). 422

Below IEP (PZC = 0.8), extremely acidic pH environment 423

resulted in“zero” charge, suggesting the particles are intensive424

negatively “−” charged. Addition of a base into the salted- 425

SDWNTs still maintained a ζ-potential at ∼−40.0 mV. The426

upwarding potential at extremely high pH condition looks like 427

that of the surfactant-assisted CNT dispersion, which may be 428

associated with ion amount-induced coiling of the CNTs.²⁹ 429

Hydrodynamic particle sizes were analyzed for the salted- 430

SDWNT and the neat-SDWNT aqueous colloidal solutions, 431

using dynamic light-scattering measurement (Figure 5e). Neat- 432

SDWNT particles formed aggregates easily and precipitated 433

within a short time, forming a clear boundary between“coarse 434

particle” and “dispersant” (photo in Figure 1c). For salted- 435

SDWNTs, rapid coagulation occurred as pH became extremely 436

acidic at 1.6. The average aggregate radius came close to a high 437

of 10μm. This is due to the presence of a low surface charge in 438

acid-adjusted salted SDWNTs, which is not suﬃcient to avoid 439

strong aggregation and precipitation of SDWNTs during 440

measurement. The addition of a strong base to both the 441

neat-SDWNTs and the salted-SDWNTs maintained similar 442

particle sizes to that of neutral pH, seemingly from the no 443

phase separation, which strongly supports the view that anion 444

enrichment at the CNT/water interfaces facilitates colloidal 445

stability of carbon nanotubes. 446

■

CONCLUSIONS 447

To summarize, we demonstrated a simple approach to 448

solubilize high concentration SDWNTs in water. Suspensions 449

with a concentration up to 2 mg mL⁻¹were stable without the 450

use of any surfactants. A negatively charged hydrophobic/water 451

interface combined with NaClO-induced oxidation was proven 452

experimentally to induce a good colloidal stability of CNTs. 453

This dispersion technique for achieving SDWNTs at high 454

concentration can now inspire us to disperse similar types of 455

carbon nanomaterials in water, which has important implica- 456

tions for researchers working in otherﬁelds. 457

■

AUTHOR INFORMATION 458

Corresponding Author 459

*E-mail:[email protected](G.L.). 460

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX G

(9)

461ORCID

462Jian Zhou: 0000-0003-0144-5901

463Gilles Lubineau: 0000-0002-7370-6093

464Notes

465The authors declare no competingﬁnancial interest.

466

■

ACKNOWLEDGMENTS

467The research reported in this publication was supported by

468funding from King Abdullah University of Science and

469Technology (KAUST). We thank KAUST for its continuous

470support.

471

■

₍₁₎^REFERENCES

472 Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A. Carbon

473Nanotubes - the Route toward Applications.Science2002,297, 787−

474792.

(2)

475 Dai, H. Carbon Nanotubes: Synthesis, Integration, and

476Properties.Acc. Chem. Res.2002,35, 1035−1044.

(3)

477 Bergin, S. D.; Nicolosi, V.; Streich, P. V.; Giordani, S.; Sun, Z.;

478Windle, A. H.; Ryan, P.; Niraj, N. P. P.; Wang, Z. T. T.; Carpenter, L.;

479et al. Towards Solutions of Single-Walled Carbon Nanotubes in

480Common Solvents.Adv. Mater.2008,20, 1876−1881.

(4)

481 Zhou, J.; Lubineau, G. Improving Electrical Conductivity in

482Polycarbonate Nanocomposites Using Highly Conductive PEDOT/

483PSS Coated MWCNTs. ACS Appl. Mater. Interfaces2013,5, 6189−

4846200.

(5)

485 Zhou, J.; Xu, X.; Yu, H.; Lubineau, G. Deformable and Wearable

486Carbon Nanotube Microwire-Based Sensors for Ultrasensitive

487Monitoring of Strain, Pressure and Torsion. Nanoscale 2017, 9,

488604−612.

(6)

489 Zhou, J.; Yu, H.; Xu, X.; Han, F.; Lubineau, G. Ultrasensitive,

490Stretchable Strain Sensors Based on Fragmented Carbon Nanotube

491Papers.ACS Appl. Mater. Interfaces2017,9, 4835.

(7)

492 Xu, X.; Uddin, A. J.; Aoki, K.; Gotoh, Y.; Saito, T.; Yumura, M.

493Fabrication of High Strength PVA/SWCNT Composite Fibers by Gel

494Spinning.Carbon2010,48, 1977−1984.

(8)

495 Rosca, I. D.; Watari, F.; Uo, M.; Akasaka, T. Oxidation of

496Multiwalled Carbon Nanotubes by Nitric Acid. Carbon 2005, 43,

4973124−3131.

(9)

498 Liu, W. Bin; Pei, S.; Du, J.; Liu, B.; Gao, L.; Su, Y.; Liu, C.;

499Cheng, H. M. Additive-Free Dispersion of Single-Walled Carbon

500Nanotubes and Its Application for Transparent Conductive Films.Adv.

501Funct. Mater.2011,21, 2330−2337.

(10)

502 Chen, Z.; Kobashi, K.; Rauwald, U.; Booker, R.; Fan, H.;

503Hwang, W.; Tour, J. M. Soluble Ultra-Short Single-Walled Carbon

504Nanotubes.J. Am. Chem. Soc.2006,128, 10568−10571.

(11)

505 Yurekli, K.; Mitchell, C. A.; Krishnamoorti, R. Small-Angle

506Neutron Scattering from Surfactant-Assisted Aqueous Dispersions of

507Carbon Nanotubes.J. Am. Chem. Soc.2004,126, 9902−9903.

(12)

508 Matarredona, O.; Rhoads, H.; Li, Z.; Harwell, J. H.; Balzano, L.;

509Resasco, D. E. Dispersion of Single-Walled Carbon Nanotubes in

510Aqueous Solutions of the Anionic Surfactant NaDDBS.J. Phys. Chem.

511B2003,107, 13357−13367.

(13)

512 Islam, M. F.; Rojas, E.; Bergey, D. M.; Johnson, A. T.; Yodh, A.

513G. High Weight Fraction Surfactant Solubilization of Single-Wall

514Carbon Nanotubes in Water.Nano Lett.2003,3, 269−273.

(14)

515 Moore, V. C.; Strano, M. S.; Haroz, E. H.; Hauge, R. H.;

516Smalley, R. E.; Schmidt, J.; Talmon, Y. Individually Suspended Wingle-

517Walled Carbon Nanotubes in Various Surfactants.Nano Lett.2003,3,

5181379−1382.

(15)

519 Tune, D. D.; Blanch, A. J.; Shearer, C. J.; Moore, K. E.; Pfohl,

520M.; Shapter, J. G.; Flavel, B. S. Aligned Carbon Nanotube Thin Films

521from Liquid Crystal Polyelectrolyte Inks.ACS Appl. Mater. Interfaces

5222015,7, 25857−25864.

(16)

523 Li, X.; Jung, Y.; Sakimoto, K.; Goh, T.-H.; Reed, M. A.; Taylor,

524A. D. Improved Efficiency of Smooth and Aligned Single Walled

Carbon Nanotube/silicon Hybrid Solar Cells. Energy Environ. Sci. 525

2013,6, 879. 526

(17)Clancy, A. J.; Melbourne, J.; Shaffer, M. S. P. A One-Step Route 527

to Solubilised, Purified or Functionalised Single-Walled Carbon 528

Nanotubes.J. Mater. Chem. A2015,3, 16708−16715. 529

(18) Jiang, C.; Saha, A.; Xiang, C.; Young, C. C.; Tour, J. M.;530

Pasquali, M.; Martı, A. A.ACS Nano2013,7, 4503−4510. 531

(19)Ryabenko, A. G.; Dorofeeva, T. V.; Zvereva, G. I. UV−VIS− 532

NIR Spectroscopy Study of Sensitivity of Single-Wall Carbon 533

Nanotubes to Chemical Processing and Van-Der-Waals SWNT/ 534

SWNT Interaction. Verification of the SWNT Content Measurements 535

by Absorption Spectroscopy.Carbon2004,42, 1523−1535. 536

(20) Yang, J. C.; Yen, C. H.; Wang, W. J.; Horng, J. J.; Tsai, Y. P.537

Assessment of Adequate Sodium Hypochlorite Concentration for Pre- 538

Oxidization of Multi-Walled Carbon Nanotubes. J. Chem. Technol. 539

Biotechnol.2010,85, 699−707. 540

(21)Shi, Z.; Lian, Y.; Liao, F.; Zhou, X.; Gu, Z.; Zhang, Y.; Iijima, S. 541

Purification of Single-Wall Carbon Nanotubes. Solid State Commun. 542

1999,112, 35−37. 543

(22) Knorr, F. J.; Hung, W. C.; Wai, C. M. Aromatic Electron544

Acceptors Change the Chirality Dependence of Single-Walled Carbon 545

Nanotube Oxidation.Langmuir2009,25, 10417−10421. 546

(23)Hussain, S.; et al. Spectroscopic Investigation of Modified Single 547

Wall Carbon Nanotube (SWCNT).J. Mod. Phys.2011,2, 538−543. 548

(24) Cong, H.; Ren, X.; Wang, P.; Yu, S. Macroscopic Multifunc- 549

tional Graphene-Based Hydrogels and Aerogels by a Metal Ion 550

Induced.ACS Nano2012,6, 2693−2703. 551

(25) Luong, N. D.; Pahimanolis, N.; Hippi, U.; Korhonen, J. T.; 552

Ruokolainen, J.; Johansson, L. S.; Nam, J.-D.; Seppäla, J. Graphene/̈ 553

cellulose Nanocomposite Paper with High Electrical and Mechanical 554

Performances.J. Mater. Chem.2011,21, 13991. 555

(26) Cano-Márquez, A. G.; Rodríguez-Macías, F. J.; Campos- 556

Delgado, J.; Espinosa-Gonzalez, C. G.; Tristá n-Ló pez, F.; Ramírez-́ 557

González, D.; Cullen, D. A.; Smith, D. J.; Terrones, M.; Vega-Cantú, Y. 558

I. Ex-MWNTs: Graphene Sheets and Ribbons Produced by Lithium 559

Intercalation and Exfoliation of Carbon Nanotubes.Nano Lett.2009,560

9, 1527−1533. 561

(27)Kosynkin, D. V.; Higginbotham, A. L.; Sinitskii, A.; Lomeda, J. 562

R.; Dimiev, A.; Price, B. K.; Tour, J. M. Longitudinal Unzipping of 563

Carbon Nanotubes to Form Graphene Nanoribbons. Nature 2009,564

458, 872−876. 565

(28) Sun, Z.; Nicolosi, V.; Rickard, D.; Bergin, S. D.; Aherne, D.; 566

Coleman, J. N. Quantitative Evaluation of Surfactant-Stabilized Single- 567

Walled Carbon Nanotubes: Dispersion Quality and Its Correlation 568

with Zeta Potential.J. Phys. Chem. C2008,112, 10692−10699. 569

(29)White, B.; Banerjee, S.; O’Brien, S.; Turro, N. J.; Herman, I. P. 570

Zeta-Potential Measurements of Surfactant-Wrapped Individual 571

Single-Walled Carbon Nanotubes. J. Phys. Chem. C 2007, 111,572

13684−13690. 573

(30) Zimmermann, R.; Freudenberg, U.; Schweiß, R.; Küttner, D.; 574

Werner, C. Hydroxide and Hydronium Ion Adsorption - A Survey. 575

Curr. Opin. Colloid Interface Sci.2010,15, 196−202. 576

(31) Vacha, R.; Rick, S. W.; Jungwirth, P.; d eBeer, A. G. F.; de577

Aguiar, H. B.; Samson, J.-S.; Roke, S. Hydrophobic Oil Droplet-Water 578

Interface The Orientation and Charge of Water at the Hydrophobic 579

Oil Droplet-Water Interface. J. Am. Chem. Soc. 2011, 133, 10204− 580

10210. 581

(32)Beattie, J. K.; Djerdjev, A. M. The Pristine Oil/water Interface: 582

Surfactant-Free Hydroxide-Charged Emulsions.Angew. Chem., Int. Ed.583

2004,43, 3568−3571. 584

(33) Beattie, J. K.; Gray-Weale, A. Oil/water Interface Charged by 585

Hydroxide Ions and Deprotonated Fatty Acids: A Comment.Angew. 586

Chem., Int. Ed.2012,51, 12941−12942. 587

(34)Gray-Weale, A.; Beattie, J. K. An Explanation for the Charge on 588

Water’s Surface.Phys. Chem. Chem. Phys.2009,11, 10994−11005. 589

(35) Roger, K.; Cabane, B. Uncontaminated Hydrophobic/water 590

Interfaces Are Uncharged: A Reply.Angew. Chem., Int. Ed.2012,51,591

12943−12945. 592

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX H

(10)

(36)

593 Roger, K.; Cabane, B. Why Are Hydrophobic/water Interfaces

594Negatively Charged?Angew. Chem., Int. Ed.2012,51, 5625−5628.

(37)

595 Beattie, J. K.; Djerdjev, A. M.; Gray-Weale, A.; Kallay, N.;

596Lützenkirchen, J.; Preocanin, T.; Selmani, A. PH and the Surfacě

597Tension of Water.J. Colloid Interface Sci.2014,422, 54−57.

(38)

598 Isogai, A.; Saito, T.; Fukuzumi, H. Nanoscale.Nanoscale2011,

5993, 71−85.

(39)

600 Xu, X.; Zhou, J.; Xin, Y.; Lubineau, G.; Ma, Q.; Jiang, L. Alcohol

601Recognition by Flexible, Transparent and Highly Sensitive Graphene-

602Based Thin-Film Sensors.Sci. Rep.2017,7, 4317.

(40)

603 Drzymala, J.; Sadowski, Z.; Holysz, L.; Chibowski, E. Ice/Water

604Interface: Zeta Potential, Point of Zero Charge, and Hydrophobicity.J.

605Colloid Interface Sci.1999,220, 229−234.

(41)

606 Dimiev, A. M.; Gizzatov, A.; Wilson, L. J.; Tour, J. M. Stable

607Aqueous Colloidal Solutions of Intact Surfactant-Free Graphene

608Nanoribbons and Related Graphitic Nanostructures.Chem. Commun.

6092013,49, 2613−2615.

(42)

610 Ericson, L. M.; Fan, H.; Peng, H.; Davis, V. A.; Zhou, W.;

611Sulpizio, J.; Wang, Y.; Booker, R.; Vavro, J.; Guthy, C.; et al.

612Macroscopic, Neat, Single-Walled Carbon Nanotube Fibers. Science

613(Washington, DC, U. S.)2004,305, 1447−1450.

Langmuir Article

DOI:10.1021/acs.langmuir.7b00850 LangmuirXXXX, XXX, XXX−XXX I