Identification of Al13 on the Colloid Surface Using Surface-Enhanced

Subscriber access provided by University of Newcastle, Australia

Article

Identification of Al13 on the colloid surface using surface-enhanced Raman spectroscopy Ning Li, Chengzhi Hu, Xiaoning Fu, Xiufang Xu, Rui Liu, Huijuan Liu, and Jiuhui Qu Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b05721 • Publication Date (Web): 10 Feb 2017 Downloaded from http://pubs.acs.org on February 11, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

Environmental Science & Technology

1

Identification of Al13 on the colloid surface using surface-enhanced

2

Raman spectroscopy

3

Ning Lia, c, Chengzhi Hub, Xiaoning Fud, Xiufang Xud, Rui Liue, Huijuan Liu*a, c,

4

Jiuhui Qub, c

5 6

a

7

Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, 100085, P.R.

8

China

9

b

State Key Laboratory of Environmental Aquatic Chemistry, Research Center for

Key Laboratory of Drinking Water Science and Technology, Research Center for

10

Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, 100085, P.R.

11

China

12

c

13

d

14

e

15

Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing

16

100085, China

University of Chinese Academy of Sciences, Beijing, 100049, P.R. China Department of Chemistry, Nankai University, Tianjin, 300071, P.R. China

State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research

17 18

*Corresponding author: Tel.: 86-010-62849128; fax: 86-010-62923541; e-mail:

19

[email protected] (H. Liu).

1

ACS Paragon Plus Environment


20

Abstract: Al13 is the most active polymeric Al species responsible for coagulation at

21

the solid-liquid interface, whereas the detection techniques for Al13 at the interface are

22

currently limited. In this study, for the first time, the identification of Al13 on the

23

silicon dioxide-based colloid surface was realized by using surface-enhanced Raman

24

scattering (SERS), which is an ideal surface method sensitive for single-molecule

25

detection. The high purity Al13 salts were prepared by electrolysis procedure followed

26

by precipitation or metathesis. Al13-Cln was determined to be feasible for the Raman

27

detection as it exhibited more noticeable signals in comparison to Al13-(SO4)p and

28

Al13-(NO3)m. The peak of Al13-Cln at 635 cm-1 could be the major characteristic peak

29

of Al13, and the other two peaks at 300 and 987 cm-1 could be accessorial evidences

30

for the identification. Further, the identification of Al13 adsorbed on the surface of Ag

31

and gold-core/silica-shell colloids was confirmed by the SERS response at the above

32

three wavenumbers with a higher signal-to-noise ratio than the normal Raman

33

scattering. According to the least squares fitting computed Raman spectra, each of the

34

characteristic peaks was associated with specific vibrational modes.

2


Page 2 of 34

Page 3 of 34


35

TOC/Abstract Art

(b)

SERS spectrum (Al13 on Ag colloids)

(c)

Intensity (a. u.)

LSFC Raman (Al13 on Ag colloids)

(a) (a) (b) (c) SERS spectrum (Al13 on Au/SiO2)

(d)

LSFC Raman (Al13 on Au/SiO2)

(f)

(d) (e) 300

36

400

500

(e)

(f)

(g) (h)

(g)

600

(h)

700

800 -1

Wavenumber (cm )

37

3


900

1000

1100


38

Page 4 of 34

Introduction

39

Coagulation as a common water treatment technology is a reaction process of the

40

coagulants and particulates proceeding at the solid-liquid interface. During

41

coagulation, the aluminum (Al) hydrolyzed products can act as a highly effective

42

coagulant for the removal of charged colloidal particulates and organics, which would

43

aggregate into large flocs and could be removed by the following sedimentation or

44

filtration treatment.1-3 The Al13 polycation, [AlO4Al12(OH)24(H2O)12]7+ (Al137+) (i.e.,

45

Keggin-Al13), is confirmed to be the most active polymeric Al species responsible for

46

the coagulation.1 Particularly, the production and persistence of the active Al13 on the

47

colloid surface was assumed to play a dominant role in the coagulation.3, 4 However,

48

no direct evidence for the Al13 on the colloid surface was reported.1, 3 Therefore,

49

identification of the Al13 aggregates at the solid-liquid interface is desirable for

50

understanding the coagulation of Al salts.

51

It has been reported that in aqueous solutions, the aluminum-27 nuclear magnetic

52

resonance spectroscopy (27Al-NMR), ferron colorimetric method, and electrospray

53

ionization-mass spectrometry can be applied for the determination of Al species.5-8 As

54

for the flocs containing Al13, X-ray photoelectron spectra and solid-state

55

were used for the identification.2,

56

polymeric Al species adhered to the surface of colloids substantially affect the

57

coagulation mechanism.1 However, the detection technique for the Al13 formed on the

58

colloid surface is currently limited. Recently, Raman scattering techniques have

59

attracted vivid interests because they can distinguish different species and molecules

3

27

Al-NMR

In fact, it is recognized that the hydrolyzed

4


Page 5 of 34


60

based on the characteristic “fingerprint” spectra with minimal sample pretreatment.9-11

61

For instance, the monomer Al(OH)4- containing only tetrahedron coordinated Al (AlⅣ)

62

in the aluminate solution at high pH can be well identified by a strong band at 620

63

cm-1 corresponding to Al-O stretching mode, and the dimeric or oligomeric aluminum

64

species were characterized by two bands at 535 and 695 cm-1.12 In addition, flat Al13,

65

[Al13(OH)24(H2O)24]15+, is constructed upon a repeated set of cubane-like moieties

66

arrayed in a lattice of edge shared with an octahedra Al (AlⅥ) core (Figure S1). The

67

Raman study of flat Al13 revealed that the region from 200 to 900 cm-1 and 900 to

68

1300 cm-1 exhibited the AlⅥ-O vibrations and hydroxo group vibrations, respectively,

69

which defined the fingerprints for flat Al13.13, 14 Al137+, the most common polynuclear

70

species in the coagulation process, has not been studied by Raman techniques so far.15,

71

16

72

Particularly, surface-enhanced Raman spectroscopy (SERS) as a promising surface

73

analytical technique appeared more attractive for the ultrasensitive detection down to

74

the single-molecule level since it can provide an increased sensitivity (enhancement

75

factor as high as 108 to 1012) with the nanoscale metal substrates.17,

76

speculated that the electromagnetic field enhancement arising from localized surface

77

plasmon resonance, and the chemical enhancement due to charge transfer might be

78

responsible for the increased scattered intensity.19-21 Through SERS analysis, the

79

slight changes in molecular structure can be detected.22 The optimized geometry and

80

structure of Al13 are shown in Figure 1. It has an ε-Keggin-like structure of Td

81

symmetry, which is characterized by a AlO4 tetrahedron center surrounded by 12 5


18

It was


82

octahedrally coordinated Al sharing edges and oxygens (-O-, -OH and –OH2). The

83

unique vibrational structure of Al-O bonds makes Raman and SERS a promising

84

method for the detection of Al13 on the colloid surface. The nanoscale metals such as

85

Ag and Au have been widely used as substrates in the SERS analysis for a better

86

detection of the organic and inorganic contaminants, pathogens, and nanomaterials in

87

environmental samples.11 Considering that SiO2 is the dominant component of the

88

clay mineral widely dispersed in natural water,23 Au core was coated with an ultrathin

89

SiO2 shell which was used to simulate the natural particle surface. Both Ag and

90

Au-core/SiO2-shell (Au/SiO2) colloids are used as the substrates for investigating the

91

SERS spectrum and characteristic peaks of Al13.

92

In the present work, a SERS method for the detection of Al13 on the colloid surface

93

was established. The results showed that the Raman and SERS can act as an ideal

94

method to identify Al13 species in solids, solutions, and on the surface of colloids. A

95

detailed structural-mode assignment was given based on the geometry optimization

96

and frequency calculation of Al137+ using density functional theory (DFT) calculations.

97

The spectroscopic observations of Al13 on colloid surface may be helpful to further

98

understand the hydrolysis reaction of Al salts in some specific cases and optimize the

99

strategies for the productivity of Al13 during coagulation.

100

Materials and methods

101

Preparation of Al13 salts

102

PACl with the total Al concentration of 0.55 mol/L was prepared by the electrolysis

103

process as reported by Liu et al.24 Then, it was purified to obtain Al13 salts 6


Page 6 of 34

Page 7 of 34


104

(Al13-(SO4)p, Al13-(NO3)m, and Al13-Cln) by sulfate precipitation, nitrate metathesis,

105

and ethanol-acetone precipitation method (see Supporting Information for details).

106

The solid Al13 was obtained by air-drying at room temperature and then placed onto

107

the clear glass slides for the Raman detection. The Al13 solution with a concentration

108

of 0.50 mol Al/L was prepared by re-dissolving the solid Al13-Cln in ultrapure water

109

and was transformed onto the as-prepared silicon wafers, followed by the spectral

110

acquisition using a confocal Raman spectroscopy system. The silicon wafers were

111

chosen as the substrate for placing the samples during the detection procedure due to

112

that it provides less spectral overlaps than other substrates such as glass and mica

113

plate (see details in Figure S2 in the Supporting Information).

114

Preparation of Ag colloids and Au/SiO2 particles

115

The aqueous solution of Ag colloids was prepared by the reduction of silver nitrate

116

with hydroxylamine hydrochloride.25 The colloids were produced by the fast addition

117

of 10 mL of hydroxylamine hydrochloride/sodium hydroxide (1.5×10-2 /3×10-2 mol/L)

118

solution to 90 mL of 1.11×10-3 mol/L silver nitrate solution.

119

It was reported that the optimum size of coinage metals (Ag, Au, and Cu) for SERS

120

enhancement ranged from 10 to 100 nm.26 Au nanoparticles with a diameter of about

121

50 nm were synthesized as cores following Frens’ method, on which the ultrathin

122

SiO2 shells were coated to form Au/SiO2 nanoparticles.27-30 In a typical synthetic

123

procedure for Au/SiO2 nanoparticles, 1.25 mL of a freshly prepared 1 mmol/L

124

(3-Aminopropyl) trimethoxysilane aqueous solution was drop-wise added to the 250

125

mL of Au sol under vigorous magnetic stirring for 15 min to ensure the complete 7



126

complexation of the amino groups with the gold surface. Subsequently, 10 mL of 0.54

127

wt% sodium silicate solution, the pH of which was adjusted to 10-11 by the

128

successive addition of cation exchange resin, was added to the sol under vigorous

129

stirring. After stirring for another 15 min at the temperature of 90 °C, 100 mL of the

130

as-prepared solution was added into 400 mL ethanol. Growth of the additional shell

131

was accomplished by the drop-wise addition of 400 µL of ammonia and 60 µL of

132

tetraethyl orthosilicate. The reaction system was then stirred for 15 min and allowed

133

to remain undisturbed for at least 12 h. Finally, the product was washed and dried at

134

the room temperature to obtain Au/SiO2 particles.

135

Adhesion of Al13 on Ag colloids and Au/SiO2 particles

136

The Al13 solutions were prepared by dissolving Al13-Cln in ultrapure water. Ag

137

substrate was found to be unstable over time due to the aggregation of Ag colloids.31

138

Hence, 10 µL of the as-prepared Ag colloid solution was mixed with isovolumic Al13

139

solution (0.01 mol Al/L) transferred on the prepared silicon wafers, and then was

140

dried for Raman analysis immediately. In the case of Au/SiO2, Au/SiO2 was used as

141

the model colloid for the simulated adsorption of Al13 onto the colloid surface in

142

natural water. Hence, a certain amount of as-prepared particles was added to the flasks

143

containing 40 mL Al13 solution with different concentration (10-2, 10-3, 10-4, 10-5, 10-6

144

and 0 mol Al/L), which covers the common dosage of Al coagulant in the coagulation

145

process (several tens to thousands of micromoles per liter). After shaking for 24 h

146

(stirring rate, 160 rpm; temperature, 25±1 °C) for a sufficient adsorption process, the

147

mixture was filtered through an ultrafiltration membrane, and then the dried Au/SiO2 8


Page 8 of 34

Page 9 of 34


148

with a thin layer was spread onto the silicon wafer for Raman detection. The

149

adsorption efficiency of Al on the Au/SiO2 surface was determined by the difference

150

in Al13 concentration in solution before and after adsorption.

151

Analytical methods

152

Raman spectra were obtained by using a confocal Raman spectroscopy system

153

(Renishaw InVia Raman microscope, Renishaw PLC) equipped with a Leica

154

microscope with 10, 50, and 100 X objectives, and a 532 nm laser lines as the

155

excitation source with a 5 mW laser power. Every signal was measured by taking 6

156

accumulations with an exposure time of 10 s. To increase signal-to-noise ratio, spectra

157

were acquired using SERS technique. The total Al concentration is determined by the

158

inductively coupled plasma optical emission spectrometer (ICP-OES, PerkinElmer

159

Co.), and the purity of Al13 product is determined by 27Al-NMR (Avance III 500 MHz,

160

Bruker) and the solid-state

161

morphology and mean size of the synthesized core-shell particles are measured by

162

transmission electron microscopy (TEM, FEI F20), and scanning electron microscope

163

(SEM, JSM 6301), respectively.

164

Computational Methods

27

Al NMR (Avance III 400 MHz, Bruker). The

165

The geometry optimization for Al(H2O)63+, Al(OH)4-, Al2Cl6 and Al13, and the

166

frequency analysis for Al13 were carried out using DFT on B3LYP level with a

167

6-31G(d) basis set.32-35 The Cartesian coordinates of the models are listed in Table

168

S3-6 in the Supporting Information. The frequencies were corrected by using

169

frequency scaling factor 0.9614.36 All calculations were carried out with Gaussian 9



Page 10 of 34

170

09.37 To identify the individual vibrational modes consisting each of the peaks in the

171

SERS spectra of Al13, the least squares fitting computed Raman (LSFC-Raman)

172

spectra were created from the computed frequencies and intensities of Al13 model.13, 38

173

More details on the calculation and the method for selecting Gaussian functions are

174

listed in the Supporting Information.

175

Results and discussions

176

Verification of pure Al13 by NMR

177

For the Raman detection of Al13, the high purity Al13 salts, Al13-(SO4)p, Al13-(NO3)m,

178

and Al13-Cln were chosen as the target analytes due to that they can inhibit the

179

interference of monomeric, dimeric, oligomeric and other Al species. Firstly, the

180

purity of Al13 in PACl, Al13-(NO3)m, and Al13-Cln was characterized by the liquid

181

27

182

ppm in PACl is assigned to the monomeric Al species, which disappeared in the

183

samples of Al13-(NO3)m and Al13-Cln, revealing that no monomeric Al species existed

184

after purification. The peaks at 63 and 80 ppm are attributed to the resonance of

185

central Al atom in the structure of Al13, and the inner standard Al(OH)4-, respectively.

186

The yield of Al13 in PACl is determined to be 65.7%, while the contents of Al13 in

187

both Al13-(NO3)m and Al13-Cln are calculated to be more than 98% of the total Al

188

concentration after purification. The purity of Al13-(SO4)p cannot be measured by the

189

liquid

190

believed to have the same purity with Al13-(NO3)m due to Al13-(NO3)m was prepared

191

from Al13-(SO4)p by a metathesis method with Ba(NO3)2. The solid-state

Al-NMR, and the results are shown in Figure 2. The peak at the chemical shift of 0

27

Al-NMR directly because it was an insoluble polymer. Whilst Al13-(SO4)p is

10


27

Al NMR

Page 11 of 34


192

spectrum of Al13-(SO4)p is provided in Figure S3. The purified Al13 solution was

193

freeze-dried into the solid, and re-dissolution of the solid exhibited no significant

194

decrease in Al13 purity, which was consistent with the previous research.39 The results

195

suggest that other than Al13, there is no trace of other Al species during Raman

196

detection.

197

Identification of the characteristic peaks for Al13 in solid and aqueous solution by

198

Raman

199

Considering that the anions of Al13 salts may affect the identification of the

200

vibrational modes in Al13, the Raman spectra of solid Al13-(SO4)p, Al13-(NO3)m, and

201

Al13-Cln are compared with those of solid Na2SO4, NaNO3, and AlCl3, respectively.

202

Al13-Cln is compared with AlCl3 rather than NaCl due to that the Cl is monatomic and

203

there is only one kind of Raman-active mode in NaCl, the external lattice mode,40

204

which is helpless for the assignment of the Raman signals of Al13-Cln. It is observed

205

that the spectrum of Al13-Cln is quite different from that of AlCl3 (Figure 3a). As for

206

the solid Al13-Cln salts, no investigation on the bonding mode of Al13 with Cl atom

207

was reported. However, the Br, which also belongs to the halogen family, has been

208

reported to intersperse among interstices between the tetrahedrons and octahedrons of

209

the polyaluminium, and has fixed positions after the formation of strong hydrogen

210

bonds between OH and Br.41 Additionally, the Al atoms in Al13 have a saturated

211

bonding structure with OXs (X stands for –Al, -H, and –H2) as shown in Figure 1, so

212

no bonds between Al and the anionic groups can be formed. Hence, it can be

213

rationally deduced that Cl atoms are combined with Al13 by forming a hydrogen 11



214

bonds with OH instead of Al-Cl bond. Thereby no signal associated with Cl exists in

215

the spectrum of the solid Al13-Cln. The three bands in Al13-Cln, a sharp one at 300

216

cm-1, a broader at 635 cm-1, and a smaller at 987 cm-1 can be accordingly assigned to

217

the vibrational modes of Al-OXs or hydroxo group. Although these Raman peaks are

218

not too strong, their signal-to-noise ratios are statistically significant. By contrast,

219

only one prominent band located at 305 cm-1 appears in the spectrogram of solid

220

AlCl3, which can be attributed to the stretching vibration of Al-Cl bond.42, 43 The

221

Raman spectrum of the solid AlCl3 which is characteristic by a centrasymmetric

222

Al2Cl6 molecule is calculated and provided in Figure S4. The strong peak appears at

223

314 cm-1 in the calculated Raman spectrum, which is nearly identical to 305 cm-1 in

224

the experimental spectrum and 308 cm-1 in the study of Edwards et al.43

225

As for NaNO3, the three peaks centered at 724, 1066, and 1384 cm-1 correspond to

226

the v4, v1, and v3 modes of NO3, respectively (Figure 3b).14 In the Raman spectrum of

227

Al13-(NO3)m, the Raman peaks of Al13 at 300 and 635 cm-1 could be observed clearly.

228

On the basis of the peak fitting results, the weak peak at 1014 cm-1 and the strong

229

sharp peak at 1053 cm-1 are assigned to Al13 (hydroxo group bending vibrations at 987

230

cm-1 in Al13-Cln) and NO3 (ν1 NO3 symmetric stretching mode at 1066 cm-1 in

231

NaNO3), respectively. The shifts of the two peaks may arise from the interaction of

232

the hydroxo group in Al13 with the NO3 group.

233

In the spectrum of Na2SO4 (Figure 3c), four bands at 454, 645, 991, and 1150 cm-1

234

are characteristics of the v2, v4, v1, and v3 modes of SO4, respectively.44 While in the

235

spectrum of Al13-(SO4)p, four bands at the similar wavenumber values of 454, 616, 12


Page 12 of 34

Page 13 of 34


236

988 and 1087 cm-1 are observed, and the band at 300 cm-1 is considered to be related

237

with the vibration mode of Al13. For Al13-(SO4)p, the characteristic peaks of Al13 at

238

635 and 987 cm-1 are close to the v4 and v1 modes of SO4, respectively. It was reported

239

that SO4 had a large Raman cross-sections which would result in the characteristic

240

peaks with high intensity,21, 45 whereas the intensity of Al13 spectra was relatively

241

feeble owing to the high symmetry and small polarizabilities of Al13 nanocluster.

242

Therefore, the peaks of Al13 at 635 and 987 cm-1 are overlapped by the modes of SO4

243

due to their similar positions and the high intensity of the peaks related to SO4. The

244

applicability for Al13-(NO3)m and Al13-(SO4)p is excluded in this study, because

245

Al13-Cln exhibits more noticeable characteristic peaks at 300, 635, and 987 cm-1,

246

which could be successfully used for the identification of Al13. The subsequent

247

detection of Al13 was thereby based on the Al13-Cln.

248

Raman analysis of Al13-Cln aqueous solution transferred onto the silicon wafer was

249

investigated, and the results are presented in Figure 4. No obvious changes were

250

observed in the band frequencies between the solid and aqueous solution of Al13. For

251

Al13-Cln in the aqueous solution, it is believed that Cl is in the bulk solution or the

252

electric double-layer of Al13 as a free anion, which would not contribute to the Raman

253

signals. The strong band at 520 cm-1 and the broad band near 987 cm-1 are related to

254

the silicon wafer (shown in inset). As mentioned above, AlCl3 has a vibrational band

255

at 305 cm-1, and silicon wafer has one around 987 cm-1, so it is unreliable to

256

distinguish Al13 based on the two bands. The Raman band at 635 cm-1 could be the

257

major characteristic peak for the effective identification of Al13, and the signals at 300 13



258

and 987 cm-1 could be accessorial evidences.

259

Identification of Al13 on the surface of colloids by SERS

260

To investigate the feasibility of SERS for detecting Al13 on the colloid surface, two

261

nanoparticles (Ag and Au/SiO2) were chosen as the substrates for evaluation due to

262

that the nanoscale metal substrates (e.g. Ag and Au) are almost essential for enhancing

263

the Raman signals of the target molecules.19,

264

obvious bands at 300, 635, and 987 cm-1 corresponding to Al13 are observed after

265

adding Al13 into Ag colloids, suggesting that SERS exhibited an excellent response for

266

Al13. The peak area at 635 cm-1 is more than half of the strongest peak at 520 cm-1,

267

which is roughly considered as the spectral feature belonging to silicon wafer. The

268

signal-to-noise ratio of Al13 species is substantially increased in the SERS compared

269

to that in the normal Raman. Since the Raman bands of Al13 in the SERS spectrum

270

using Ag colloids and the normal Raman spectrum are similar in their positions, and

271

no chemical bonds or charge-transfer were validated between the Ag and Al13, the

272

chemical enhancement cannot be supported. The electromagnetic enhancement, which

273

should be a nonselective amplifier for Raman scattering by all molecules adsorbed on

274

the surface of substrates, is likely to be responsible for the enhanced Al13 spectra. In

275

addition, two new weak bands at 475 and near 520 cm-1 appeared in SERS, which

276

would be analyzed subsequently.

27

As illustrated in Figure 5a, three

277

The identification of various amounts of Al13 adsorbed on the surface of Au/SiO2

278

colloids by SERS is shown in Figure 5b. The mean size of Au/SiO2 nanoparticles is

279

approximately 50 nm (Figure S5a), and the thickness of silica shell is about 2-4 nm 14


Page 14 of 34

Page 15 of 34


280

(Figure S5b). The adsorption efficiencies of Al13 on Au/SiO2 (dosage of 1g/L) are

281

determined, and the amounts of Al13 on per unit weight of Au/SiO2 are calculated to

282

be 0.13×10-2, 0.98×10-3, 1.0×10-4, 1.0×10-5, and 1.0×10-6 mol/g for the initial Al13

283

concentration of 10-2, 10-3, 10-4, 10-5, and 10-6 mol Al/L, respectively (Table S1). For a

284

better comparison, the Raman spectrum of Au/SiO2 treated by the ultrapure water

285

placed on silicon wafer is also provided as a blank control. The bare Au/SiO2 (0 mol

286

Al/g) yielded no signals in SERS except the peaks related to the silicon wafer. In the

287

presence of Al13 species, three obvious vibrational bands at 300, 635, and 987 cm-1

288

can be found, demonstrating the successful identification of vibrational modes of Al13

289

adsorbed onto the particle surface with enhanced Raman signals. These Raman

290

features, especially the peaks at 300 and 635 cm-1, strongly increased in intensity with

291

increasing the amounts of Al13 adsorbed on Au/SiO2 particle surface, and were still

292

observable with the low Al13 amounts of 10-6 mol Al/g. Considering that the common

293

dosage of Al salts is higher than 10-5 mol Al/L,3 the concentration range investigated

294

in this study can cover most cases in coagulation treatment processes. Therefore, the

295

SERS approach has the potential to be applied in the identification of the Al13 species

296

on the colloid surface. To further examine the quantitative ability of this method, the

297

linear calibration curves are presented in Figure S6. The result indicates a

298

less-than-ideal linear correlation, which is probably due to the nonuniform

299

distribution of the solid samples on the silicon wafer as observed by the microscopy,

300

inhomogeneous hot spot densities, analyte polarizability, or the inhomogeneity and

301

irreproducibility of the noble-metal substrate. In the case of Au/SiO2, the Au core is 15



302

used to generate a large field enhancement, and the inert silica shell prevents the core

303

from the direct contact with Al13 molecules, ensuring that there are no chemical

304

interactions between Al13 and Au.46 The adsorption of Al13 on the Au/SiO2 surface

305

depends on the electrostatic attraction.47, 48 Hence, the electromagnetic enhancement

306

should be the enhancement mechanism for Au/SiO2 colloids. On the basis of the

307

obtained results, SERS is proved to be a reliable method to identify Al13 on the

308

surface of Ag and Au/SiO2 colloids with a higher peak intensity.

309

Theoretical analysis for Al13 identification

310

It is known that the vibrational modes are related to the species and structures.49

311

The characteristic Raman signals of Al13 can be assigned to the vibration of bonds

312

related to the Al13 molecule structure. The molecule of Al13 ions is considered to be

313

symmetrical, assuming that Al13 group is free with Cl-. As shown in Figure 6a, AlⅣ-O,

314

AlⅥ-OAlⅣ, AlⅥ-OH, AlⅥ-OH2, and O-H (or -H2) are believed to be the dominant

315

chemical bonds. The LSFC-Raman spectra of Al13 adsorbed on the Ag and Au/SiO2

316

colloid surface were created to associate all the Raman peaks with specific vibrational

317

modes. During the hydrolysis process of Al coagulants, different kinds of monomers

318

such as Al(H2O)63+ and Al(OH)4- may polymerize to dimeric or oligomeric aluminum

319

species, and further to form multimeric species like Al13 and flat Al13, although their

320

reaction conditions are generally different.13,

321

Al(H2O)63+ (Figure 6b) and the AlⅣ-O bond in Al(OH)4- (Figure 6c) are the structural

322

units of Al13 or flat Al13. The Raman signals of the Al-O bonds in Al(H2O)63+,

323

Al(OH)4-, and flat Al13 were used as the assistance and supplementary of LSFC results

50

As a result, the Al Ⅵ -O bond in

16


Page 16 of 34

Page 17 of 34


324

for the assignment of vibrational peaks of Al13.

325

The experimental SERS spectra, associated LSFC-Raman spectra, and constituent

326

vibrational modes of Al13 adsorbed on Ag and Au/SiO2 colloids are shown in Figure

327

6d. It is observed that the spectra of Al13 on the two kinds of colloids are similar, as

328

well as their corresponding vibrational modes. The LSFC-Raman spectra are

329

composed of 156 unique vibrational modes (the coefficients of the component

330

Gaussian functions are listed in Table S2 in the Supporting Information). The LSFC

331

Raman spectra of Al13 samples, incorporating the vibrations of the aforementioned

332

various bonds, correlate well to the experimental spectra over the entire frequency

333

range of 260-1100 cm-1. The peaks from 260 to 900 cm-1 in the Raman spectra of Al13

334

are found to be mainly associated with the Al-O vibrational modes. And the positions

335

of these bands are in agreement with recent Raman studies of the Al-O bands in the

336

flat Al13 which contains only AlⅥ.13, 14 The band at 300 cm-1 is accordingly assigned to

337

the AlⅣ-O, AlⅥ-OAlⅣ bending, and Al Ⅵ-OH2 stretching vibrations. The previous

338

study of the Na2O-Al2O3-H2O system also supported the assignment of the region

339

below 350 cm-1 to the Al-O bending vibrations.12 The low-intensity band at 475 cm-1

340

in the SERS spectra of Al13, which is insensitive for the normal Raman method, is

341

associated with the AlⅥ-OH stretching vibrations. This result is consistent with the

342

previous report that the peak at 478 cm-1 was attributed to the metal-oxygen (AlⅥ-O)

343

symmetric stretch of the flat Al13 cluster.14 For the peak near 520 cm−1, it is observed

344

that AlⅣ-O, AlⅥ-OAlⅣ, AlⅥ-OH, and bonds in silicon wafer all contribute spectral

345

intensity to this region. Because they partially overlap with the intense peak at 520 17



346

cm−1 of silicon wafer, there is considerable uncertainty in the precise identification of

347

Al13 on the basis of this region.

348

The most dominant features in the region from 550 to 1100 cm-1 are the peaks

349

located around 635 and 987 cm-1. The peak at 635 cm−1 is assigned to the stretching

350

vibrations of AlⅣ-O and AlⅥ-OH, and bending vibrations of AlⅥ-OH2. The AlⅣ-O

351

stretching vibration have been reported to be located at 620 cm-1 in the Raman

352

spectrum of Al(OH)4-.51, 52 While in this study, the peak corresponding to the AlⅣ-O

353

vibrational mode in Al13 is resolved to be located around 660 cm-1. It is recognized

354

that the position of a Raman band, which relates to the corresponding bond force

355

constant, basically depends on electronic structure and bonding.53 The 40 cm-1 blue

356

shift of the AlⅣ-O band in Al13 reveals an overall increase in the extent of AlⅣ-O

357

bonding. Due to that the electron donor H-atom was replaced by the Al-octahedrons,

358

the electron density of AlⅣ-O in Al13 is more uniform than that of Al(OH)4-, resulting

359

in the decrease in the polarity of AlⅣ-O bond, and thereby the blue shift in the

360

frequency. Jackson and co-workers recently reported that the region from 900-1300

361

cm-1 exhibited resolved bridging hydroxo group vibrations that collectively define

362

fingerprints for the flat Al13 clusters.

363

hydroxo group bending vibrations dominate in the region from 900-1000 cm-1. These

364

give rise to a broad signal around 987 cm-1. The spectra of the parent monohydrate

365

salts and other polymeric species with the similar structure have been reported in the

366

previous studies.13, 14, 54-56 By contrast, the obtained results show that Al13 exhibits

367

unique vibrational characteristics, and reveal the reliability of Raman spectrum for the

14

Computations in this study also reveal that

18


Page 18 of 34

Page 19 of 34


368

detection of Al13 molecule.

369

Potential Application

370

The interactions between coagulants and colloids are known to play a vital role in

371

the coagulation process. Particularly, the production and persistence of active Al13

372

species on the colloid surface have a significant impact on the treatment performance.

373

In this study, SERS is proposed as an effective analytical method for the identification

374

of Al13 species on the surface of colloids. The findings may shed some light on

375

understanding of the real coagulation process by clarifying the behaviors of Al13 at the

376

solid-liquid interface. It may also help to optimize the realistic water purification

377

process upon the optimization of the selection, dosing mode, and dosage of the

378

coagulants, and the regulation of reaction conditions (pH, basicity, temperature, etc).

379

Moreover, the established method is considered to have the application potential in

380

the study of geochemical cycle of Al, due to that Al13 plays an important part in the

381

process in the nature system. In some cases, the applications of the detection method

382

may be limited due to the interferences by some anions, thus the inhibition of the

383

interferences needs to be further investigated. However, the foreseeable applications

384

can be realized in the simulated water systems for investigating the migration and

385

transformation of Al13 on the colloid surface, and for providing some references for

386

the Al coagulation in a realistic water treatment process.

387

Associated content

388

Supporting information. Preparation of Al13-(SO4)p, Al13-(NO3)m, and Al13-Cln,

389

SEM and TEM images of the Au/SuO2 particles, Raman spectra of silicon wafers, 19



27

390

glass, and mica plate, solid-state

Al NMR spectrum of the pure Al13-(SO4)p,

391

calculated Raman spectrum of the solid AlCl3, discussion on the enhancement factor,

392

quantitative detection of Al13 adsorbed on the Au/SiO2 surface, Least-squares fitting

393

procedure and the Cartesian coordinates for Al(H2O)63+, Al(OH)4- and Al137+, as well

394

as the coefficients of all the component Gaussian functions of the LSFC-Raman of

395

Al13.

396

Acknowledgment

397

This work was financially supported by the National Science Fund for

398

Distinguished Young Scholars of China (Grant No. 51225805). The authors sincerely

399

thank Prof. Jingfu Liu, and Ms Fanglan Geng at RCEES for their generous

400

experimental assistance.

401

Reference

402

(1) Lin, J. L.; Chin, C. J. M.; Huang, C. P.; Pan, J. R.; Wang, D. S. Coagulation

403

behavior of Al13 aggregates. Water Res. 2008, 42, 4281-4290.

404

(2) Lin, J. L.; Huang, C. P.; Chin, C. J. M.; Pan, J. R. The origin of Al(OH)3-rich and

405

Al13-aggregate flocs composition in PACl coagulation. Water Res. 2009, 43,

406

4285-4295.

407

(3) Hu, C. Z.; Liu, H. J.; Qu, J. H.; Wang, D. S.; Ru, J. Coagulation behavior of

408

aluminum salts in eutrophic water: Significance of Al13 species and pH control.

409

Environ. Sci. Technol. 2006, 40, 325-331.

410

(4) Zhao, H.; Liu, H. J.; Hu, C. Z.; Qu, J. H. Effect of Aluminum speciation and

411

structure characterization on preferential removal of disinfection byproduct 20


Page 20 of 34

Page 21 of 34


412

precursors by aluminum hydroxide coagulation. Environ. Sci. Technol. 2009, 43,

413

5067-5072.

414

(5) Bertsch, P. M.; Thomas, G. W.; Barnhisel, R. I. Characterization of

415

hydroxy-aluminum solutions by aluminum-27 nuclear magnetic resonance

416

spectroscopy. Soil Sci. Soc Am. J. 1986, 50, 825-830.

417

(6) Buffle, J.; Parthasarathy, N.; Haerdi, W. Importance of speciation methods in

418

analytical control of water treatment processes with application to fluoride

419

removal from waste waters. Water Res. 1985, 19, 7-23.

420 421

(7) Parker, D. R.; Bertsch, P. M. Identification and quantification of the “Al13” tridecameric polycation using ferron. Environ. Sci. Technol. 1992, 26, 908-914.

422

(8) Zhao, H.; Hu, C. Z.; Liu, H. J.; Zhao, X.; Qu, J. H. Role of aluminum speciation

423

in the removal of disinfection byproduct precursors by a coagulation process.

424


425 426

(9) Kusic, D.; Kampe, B.; Rosch, P.; Popp, J. Identification of water pathogens by Raman microspectroscopy. Water Res. 2014, 48, 179-189.

427

(10) Ghosal, S.; Macher, J. M.; Ahmed, K. Raman microspectroscopy-based

428

identification of individual fungal spores as potential indicators of indoor

429

contamination and moisture-related building damage. Environ. Sci. Technol. 2012,

430

46, 6088-6095.

431 432 433

(11) Halvorson, R. A.; Vikesland, P. J. Surface-enhanced Raman spectroscopy (SERS) for environmental analyses. Environ. Sci. Technol. 2010, 44, 7749-7755. (12) Johnston, C. T.; Agnew, S. F.; Schoonover, J. R.; Kennedy, J. W.; Page, B.; 21



434

Osborn, J.; Corbin, R. Raman study of aluminum speciation in simulated alkaline

435

nuclear waste. Environ. Sci. Technol. 2002, 36, 2451-2458.

436

(13) Wang, W.; Liu, W. M.; Chang, I. Y.; Wills, L. A.; Zakharov, L. N.; Boettcher, S.

437

W.; Cheong, P. H. Y.; Fang, C.; Keszler, D. A. Electrolytic synthesis of aqueous

438

aluminum nanoclusters and in situ characterization by femtosecond Raman

439

spectroscopy and computations. Proc. Natl. Acad. Sci. USA 2013, 110,

440

18397-18401.

441

(14) Jackson, M. N.; Wills, L. A.; Chang, I. Y.; Carnes, M. E.; Scatena, L. F.; Cheong,

442

P. H. Y.; Johnson, D. W. Identifying nanoscale M13 clusters in the solid state and

443

aqueous solution: vibrational spectroscopy and theoretical studies. Inorg. Chem.

444

2013, 52, 6187-6192.

445

(15) Akitt, J. W.; Farthing, A. Aluminium-27 nuclear magnetic-resonance studies of

446

the hydrolysis of aluminum (III). Part 4. Hydrolysis using sodium carbonate. J.

447

Chem. Soc., Dalton Trans. 1981, 1617-1623.

448

(16) Hu, C. Z.; Liu, H. J.; Qu, J. H. Preparation and characterization of

449

polyaluminum chloride containing high content of Al13 and active chlorine.

450

Colloid Surf. A: Physicochem. Eng. Asp. 2005, 260, 109-117.

451 452 453 454 455

(17) Le Ru, E. C.; Etchegoin, P. G. Quantifying SERS enhancements. MRS Bull. 2013, 38, 631-640. (18) Stiles, P. L.; Dieringer, J. A.; Shah, N. C.; Van Duyne, R. R. Surface-Enhanced Raman Spectroscopy. Annu. Rev. Anal. Chem. 2008, 1, 601-626. (19) Haynes, C. L.; McFarland, A. D.; Van Duyne, R. P. Surface-enhanced Raman 22


Page 22 of 34

Page 23 of 34


456

spectroscopy. Anal. Chem. 2005, 77, 338A-346A.

457

(20) Kneipp, K.; Kneipp, H.; Itzkan, I. Dasari, R. R.; Feld M. S. Ultrasensitive

458

chemical analysis by Raman spectroscopy. Chem. Rev. 1999, 99, 2957-2975.

459

(21) Campion, A.; Kambhampati, P. Surface-enhanced Raman scattering. Chem. Soc.

460

Rev. 1998, 27, 241-250.

461

(22) Guo, H. Y.; Zhang, Z. Y.; Xing, B. S.; Mukherjee, A.; Musante, C.; White, J. C.;

462

He, L. L. Analysis of silver nanoparticles in antimicrobial products using

463

surface-enhanced Raman spectroscopy (SERS). Environ. Sci. Technol. 2015, 49,

464

4317-4324.

465

(23) Liew, Y. M.; Heah, C. Y.; Mond, M. A. B.; Kamarudin, H. Structure and

466

properties of clay-based geopolymer cements: A review. Prog. Mater. Sci. 2016,

467

83, 595-629.

468

(24) Liu, H. J.; Qu, J. H.; Hu, C. Z.; Zhang, S. J. Characteristics of nanosized

469

polyaluminum chloride coagulant prepared by electrolysis process. Colloids Surf.

470

A: Physicochem. Eng. Asp. 2003, 216, 139-147.

471

(25) Leopold, N.; Lendl, B. A new method for fast preparation of highly

472

surface-enhanced Raman scattering (SERS) active silver colloids at room

473

temperature by reduction of silver nitrate with hydroxylamine hydrochloride. J.

474

Phys. Chem. B 2003, 107, 5723-5727.

475 476 477

(26) Moskovits, M. Surface-enhanced Raman spectroscopy: a brief retrospective. J Raman Spectrosc. 2005, 36, 485-496. (27) Lu, Y.; Yin, Y.; Li, Z. Y.; Xia, Y. Synthesis and self-assembly of Au@ SiO2 23



478

core-shell colloids. Nano Lett. 2002, 2, 785-788.

479

(28) Li, J. F.; Huang, Y. F.; Ding, Y.; Yang, Z. L.; Li, S. B.; Zhou, X. S.; Fan, F. R.;

480

Zhang, W.; Zhou, Z. Y.; Wu, D. Y.; Ren, B.; Wang, Z. L.; Tian, Z. Q.

481

Shell-isolated nanoparticle-enhanced Raman spectroscopy. Nature 2010, 464,

482

392-395.

483 484 485 486

(29) Stöber, W.; Fink, A.; Bohn, E. Controlled growth of monodisperse silica spheres in the micron size range. J. Colloid Interf. Sci. 1968, 26, 62-69. (30) Frens, G. Controlled nucleation for the regulation of the particle size in monodisperse gold suspensions. Nature 1973, 241, 20-22.

487

(31) Ingram, J. C.; Pemberton, J. E. Investigation of the quenching of surface

488

enhanced Raman scattering from pyridine on copper and silver electrodes by

489

underpotential deposition of lead. Langmuir 1992, 8, 2040-2048.

490 491

(32) Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648-5652.

492

(33) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti

493

correlation-energy formula into a functional of the electron density. Phys. Rev. B

494

1988, 37, 785-789.

495 496

(34) Becke, A. D. A new mixing of Hartree-Fock and local density-functional theories. J. Chem. Phys. 1993, 98, 1372-1377.

497

(35) Qian, Z.; Feng, H.; Yang, W.; Bi, S. Theoretical investigation of water exchange

498

on the nanometer-sized polyoxocation AlO4Al12(OH)24(H2O)127+ (Keggin-Al13) in

499

aqueous solution. J. Am. Chem. Soc. 2008, 130, 14402-14403. 24


Page 24 of 34

Page 25 of 34


500

(36) Scott, A. P.; Radom, L. Harmonic vibrational frequencies: an evaluation of

501

Hartree-Fock, Møller-Plesset, quadratic configuration interaction, density

502

functional theory, and semiempirical scale factors. J. Chem. Phys. 1996, 100,

503

16502-16513.

504 505

(37) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; et. al. Gaussian 09, Revision D.01,Gaussian, Inc., Wallingford CT, 2013.

506

(38) Wang, W.; Chang, I. Y.; Zakharov, L.; Cheong, P. H. Y.; Keszler, D. A.

507

[Sc2(µ-OH)2(H2O)6(NO3)2](NO3)2: aqueous synthesis and characterization. Inorg.

508

Chem. 2013, 52, 1807-1811.

509

(39) Shi, B. Y.; Li, G. H.; Wang, D. S.; Tang, H. X. Separation of Al13 from

510

polyaluminum chloride by sulfate precipitation and nitrate metathesis. Sep. Purif.

511

Technol. 2007, 54, 88-95.

512

(40) Pasternak, A.; Coche, E.; Gilat, G. Calculation of second-order Raman

513

scattering for KBr, NaCl, and MgO crystals. Phys. Rev. B 1973, 9, 4584-4592.

514

(41) Sun, Z.; Wang, H; Feng, H.; Zhang, Y.; Du, S. Crystal structure of

515

[Al4(OH)6(H2O)12][Al(H2O)6]Br12: A new polyaluminum compound. Inorg. Chem.

516

2011, 50, 9238-9242.

517

(42) Alves, C. C.; Campos, T. B. C.; Alves, W. A. FT-Raman spectroscopic analysis

518

of the most probable structures in aluminum chloride and tetrahydrofuran

519

solutions. Spectrochim Acta, Part A 2012, 97, 1085-1088.

520

(43) Edwards, H. G. M.; Farwell, D. W.; Johnson, A. F. FT-Raman spectroscopic

521

study of aluminum(III) chloride in acetonitrile and dichloromethane solutions 25



522

Page 26 of 34

containing water. J. Mol. Struct. 1997, 344, 37-44.

523

(44) Dong, J. L.; Li, X. H.; Zhao, L. J.; Xiao, H. S.; Wang, F.; Guo, X.; Zhang, Y. H.

524

Raman observation of the interactions between NH4+, SO42-, and H2O in

525

supersaturated (NH4)2SO4 droplets. J. Phys. Chem. B 2007, 111, 12170-12176.

526

(45) Dudik, J. M.; Johnson, C. R.; Asher, S. A. Wavelength dependence of the

527

preresonance Raman cross-section of CH3CN, SO42-, ClO4- and NO3-. J. Chem.

528

Phys. 1985, 82, 1732-1740.

529

(46) Li, J. F.; Tian, X. D.; Li, S. B.; Anema, J. R.; Yang, Z. L.; Ding, Y.; Wu, Y. F.;

530

Zeng, Y. M.; Chen, Q. Z.; Ren, B.; Wang, Z. L.; Tian, Z. Q. Surface analysis

531

using

532

Protocols. 2013, 8, 52-65.

shell-isolated

nanoparticle-enhanced

Raman

spectroscopy.

Nature

533

(47) Lin, J.; Huang, C.; Chin, C. M.; Pan, J. R. Coagulation dynamics of fractal flocs

534

induced by enmeshment and electrostatic patch mechanisms. Water Res. 2008, 42,

535

4457-4466.

536

(48) Ma, X.; Zeng, G.; Zhang, C.; Wang, Z.; Yu, J.; Li, J.; Huang, G.; Liu, H.

537

Characteristics of BPA removal from water by PACl-Al13 in coagulation process.

538

J. Colloid Interf. Sci. 2009, 337, 408-413.

539

(49) Jin, X. Y.; Yan, Y.; Shi, W. J.; Bi, S. P. Density functional theory studies on the

540

structures and water-exchange reactions of aqueous Al(III)-oxalate complexes.

541


542

(50) Tang, H. X.; Xiao, F.; Wang, D. S. Speciation, stability, and coagulation

543

mechanisms of hydroxyl aluminum cluster formed by PACl and alum: A critical 26


Page 27 of 34


544 545 546

review. Chem. Rev. 2015, 226, 78-85. (51) Moolenaar, R. J.; Evans, J. C.; McKeever, L. Structure of the aluminate ion in solutions at high pH. J. Phys. Chem. 1970, 74, 3629-3636.

547

(52) Watling, H. R.; Fleming, S. D.; van Bronswijk, W.; Rohl, A. L. Ionic structure in

548

caustic aluminate solutions and the precipitation of gibbsite. J. Chem. Soc.,

549

Dalton Trans. 1998, 3911-3917.

550

(53) Schlucker, S.; Singh, R. K.; Asthana, B. P.; Popp, J.; Kiefer, W.

551

Hydrogen-bonded pyridine-water complexes studied by density functional theory

552

and Raman spectroscopy. J. Phys. Chem. A 2001, 105, 9983-9989.

553 554

(54) Watling, H. Spectroscopy of concentrated sodium aluminate solutions. Appl. Spectrosc. 1998, 52, 250-258.

555

(55) Rudolph, W. W.; Mason, R.; Pye, C. C. Aluminium(III) hydration in aqueous

556

solution. A Raman spectroscopic investigation and an ab initio molecular orbital

557

study of aluminium(III) water clusters. Phys. Chem. Chem. Phys. 2000, 2,

558

5030-5040.

559

(56) Lu, B.; Jin, X.; Tang, J.; Bi, S. DFT studies of Al-O Raman vibrational

560

frequencies for aquated of aluminium species. J. Mol. Struct. 2010, 982, 9-15.

561

27



562

Figure captions

563

Figure 1. (a) The optimized geometry of Al13 and (b) the structure of ε-Keggin-like

564

Al13 shown in polyhedral representation.

565

Figure 2.

566

(total Al concentration: 0.50 mol/L), and Al13-(NO3)m (total Al concentration: 0.50

567

mol/L).

568

Figure 3. (a) Raman spectra of solid AlCl3 and Al13-Cln. (b) Raman spectra of solid

569

NaNO3 and Al13-(NO3)n and the fitting curve for Al13-(NO3)n. (c) Raman spectra of

570

solid NaSO4 and Al13-(SO4)n.

571

Figure 4. Raman spectrum of Al13-Cln (Al13) aqueous solution (total Al concentration:

572

0.50 mol/L) on silicon wafer. Inset shows the Raman spectrum of silicon wafer.

573

Figure 5. (a) SERS spectra of Ag colloids and Al13 (0.01 mol Al/L) on Ag colloids.

574

The inset shows the fitted peak areas of two samples. (b) SERS spectra of Au/SiO2

575

and Al13 adsorbed on the surface of Au/SiO2 particles.

576

Figure 6. (a) Schematic diagram of the bonds in Al13, the optimized geometry of (b)

577

Al(H2O)63+ and (c) Al(OH)4-, and (d) LSFC-Raman spectra that depict the

578

contributions of each type of vibrational modes of Al13 on the surface of Ag and

579

Au/SiO2 colloids (concentration of Al13 with Ag colloids: 0.01 mol Al/L; amounts of

580

Al13 absorbed onto Au/SiO2: 0.13×10-2 mol Al/g).

27

Al-NMR spectra of PACl (total Al concentration: 0.55 mol/L), Al13-Cln

581

28


Page 28 of 34

Page 29 of 34


582 583

Figure 1

584

29



Page 30 of 34

Al(H2O)63+ PACl Al137+ Al13-Cln

Al13-(NO3)m 100 90

80

70

60

50

40

30

20

10

δ (ppm) 585 586

Figure 2

587

30


0

-10 -20


Intensity (a. u.)

Page 31 of 34

987

635

(a)

Al13-Cln

305

Al-Cl

200

AlCl3

400

(b) Intensity (a. u.)

Al-OX or hydroxo group

300

600

800

1000

Experimental Fitting curve Baseline

300

635

1200

1052

722

1400

Al13-(NO3)m 1383

1014

NaNO3 400

600

800

1000

1200

1400

Intensity (a. u.)

(c) Al-OX or hydroxo group

Al13-(SO4)p v2 SO4

v4 SO4

v1 SO4

v3 SO4

Na2SO4 400

600

800

1000

1200

-1

588 589

Wavenumber(cm ) Figure 3

590

31


1400


Intensity (a. u.)

Relative Intensity

1.0x103

Page 32 of 34

7.5x102

Si

5.0x102 200

400

600

800

1000

1200

1400

-1

Wavenumber (cm

2.5x102

300

635

)

Al13

987

0.0 200 591 592

400

600

800

1000

Wavenumber (cm-1) Figure 4

593

32


1200

1400

Page 33 of 34


(a)

Al13 on Ag colliods

520

5

Relative Intensity

1.2x10

Ag colloids (Blank)

-1

Center(cm ) 520.691 986.82 1064.05 1352.13

8.0x104 Ag

4.0x104

Area 837381 211980 190492 791854

Al13-Ag

-1 Area Center(cm ) 300.85 402770 520.16 1003620 635.637 613195 987.429 326616

new peak new peak

300 635

987

0.0

520

300

450

1352

986 1064

600

750

900

1050

1200

1350

1500

-1

Wavenumber (cm ) 4

6x10

(b)

-2

0.13× ×10 mol Al/g -3

Relative Intensity

Al13 on Au/SiO2

0.98× ×10 mol Al/g -4

10 mol Al/g

5x104

520

-5

10 mol Al/g -6

10 mol Al/g 0 mol Al/g

2x104

1x104

300 635

987

0 300

400

500

600

700 -1

Wavenumber (cm ) 594 595

Figure 5

596

33


900 1000


Page 34 of 34

AlⅣ -O bending vibrations

(d)

AlⅣ -OH bending vibrations AlVI-OH2 stretching vibrations SERS spectrum of Al13 with Ag colloids LSFC Raman (Al13 with Ag colloids)

AlⅣ -OH stretching vibrations AlⅣ -OH2 bending vibrations AlⅣ -O stretching vibrations

Intensity (a. u.)

AlⅣ -OAlⅣ stretching vibrations hydroxo group bending vibrations

SERS spectrum of Al13 with Au/SiO2 LSFC Raman (Al13 with Au/SiO2)

300

597 598

400

500

600

700

800 -1

Wavenumber (cm )

Figure 6

34


900

1000

1100

Identification of Al13 on the Colloid Surface Using Surface-Enhanced

Recommend Documents