Aqueous processable WO_3−x nanocrystals with solution tunable localized surface plasmon resonance

Abstract

Heavily doped tungsten oxide nanoparticles with localized surface plasmon resonances (LSPRs) were recently highlighted as potential substitutes for noble metals in the field of plasmonic applications. Herein, oxygen deficient, spherically shaped WO_3−x nanocrystals (NCs) were synthesized with a pronounced visible LSPR absorbance peak instead of a broadband tail usually observed for WO_3−x nanowires. Although the tuning of the plasmon resonances was achieved mainly by changing the nanocrystals composition or solvent refraction index, we demonstrate this via the interfacial charge donation/extraction. In an aqueous NCs dispersion, the LSPR peak was either blue shifted in an acidic solution up to 80 nm or bleached by a basic solution making the NCs appropriate for sensing applications.

---

1. Introduction

Colloidal nanocrystals (NCs) exhibiting localized surfaces plasmon resonances (LSPRs) generally represented by various noble-metals^1,2 have recently been complemented by heavily doped semiconductors.^3–5 LSPR semiconductor NCs are usually self-activated either by cation vacancies (in Cu_2−xCh,^6–8 Cu_xIn_yCh₂,^9,10 Ch = S, Se, Te) or anion vacancies in metal oxides.^11,12 The defect concentration is engineered to be high enough for producing hole or electron degenerated materials. Owing to the tunable carrier concentration/light absorbance, semiconductor plasmonic NCs are highly desirable for many application ranging from photothermal therapy,^13,14 engineering of conductive parts of microchips,¹⁵ sensing¹⁶ or catalyses¹⁷ to the realization of absolute blackbody conditions.¹⁸ Recent developments in solvothermal methods of particle synthesis for heavily doped metal-oxides NCs^19,20 have made it possible to obtain aqueous dispersible LSPR nanoparticles. Moreover, these methods did not require the use of complex surfactant molecules, which usually affected plasmons in metal chalcogenides.²¹ Among the various plasmonic oxides, WO_3−x seems to be closest to metallic nanoparticles in terms of the LSPR properties such as spectral position of absorbance and carriers concentration.⁵ In particular, in W₁₈O₄₉ NCs, the surface plasmons can enhance Raman scattering to the same level as noble metals.²² Herein, we pursued the goal to tune the LSPRs in W₁₈O₄₉ NCs by interfacial charge transfer between the NCs and aqueous media of different pH, resulting in a scalable LSPR shift. A similar task for noble metal plasmonic NCs was realized by changing the NCs surrounding media.²³ This can include several types of surfactant, sensitivity to each other as well to the pH, and shrinkage or elongation under specific chemical conditions. As mentioned above, a main intrinsic LSPR characteristic (carrier concentration) is locked on the synthesis stage for such NCs. In contrast to other semiconductor LSPR NCs, wherein the defect concentration can be varied continuously by the synthesis procedures, WO_3−x is characterized by several discrete stable defect stoichiometries only, namely, W₁₈O₄₉ has the highest possible oxygen deficiency and only the tungsten suboxide is obtained in isolated specific monoclinic crystalline phase.²⁴ The successful synthesis procedures implied the fabrication of nanocrystalline W₁₈O₄₉, including solution combustion,²⁴ thermal evaporation,²⁵ microwave-assisted²⁶ and solvothermal procedures.^{19,20,22,27,28} The plasmonic absorbance in such particles is expected in the visible region and should not be tunable only by the refractive index of the surrounding medium, but also by interfacial charge transfer. This study demonstrates later for the W₁₈O₄₉ NCs, providing a tool for pH sensing by tuning the LSPR band.

2. Materials and methods

Tungsten hexacarbonyl (W(CO)₆, 97%) was purchased from Alfa Aesar. Absolute ethanol, methanol, 2-propanol, sodium hydroxide, pH buffer solutions (all of the ACS grade) and perchloric acid (70%) were ordered from Merck and Fluka, respectively. The water used was of Milli-Q quality. All chemicals were used without purification.

For the synthesis of WO_3−x nanocrystals, we modified solvothermal procedure recently developed for nanowires by the Xue group.²⁰ The purpose of such modification was to grow spherically shaped particles with eliminated longitudinal plasmon frequencies. The last ones in the case of nanowires, broadly distributed by aspect ratio, transform into broadband vis-NIR tail, which is unsuitable for LSPR tuning. An alternative way to observe the LSPR band is a chemical surface modification of WO_2.72 nanorods being rather complicated and comprising surfactant amine functionalization followed by the stepwise oxidation of nanoparticles, as shown by Grauer and Alivisatos.²⁹ The synthesis procedure used in this study was as follows: 50 mg of W(CO)₆ (0.14 mmol) handled in a glovebox was added to absolute ethanol with an amount of 30 mL. After 30 min stirring at room temperature, the mixture yielded a slightly yellow but transparent solution. The solution was then transferred to a 40 mL polyetheretherketone lined stainless steel autoclave, heated up to 160 °C and maintained at this temperature for 20 hours. The NCs were isolated from solution by centrifugation at 4.5 krpm for 7 min followed by re-dispersion in ethanol. The washing procedure was repeated at least three times and was the same for the transfer of NCs to other alcohols or aqueous solvents.

The UV-vis-NIR absorbance was measured in a 10 mm optical pathlength cuvette using a Thermo Scientific Evolution 60S Spectrophotometer. Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDXS) were carried out on a ZEISS ULTRA 55 machine, operating at 5/15 kV. X-ray diffraction patterns and X-ray photoelectron spectra were recorded using Siemens D5000 diffractometer and SES 2002 spectrometer with Cu Kα and Al Kα X-rays sources, respectively. For SEM, XRD and XPS, the NCs solution was drop casted onto Si wafers and plastic slides. pH dispersions were prepared using perchloric acid, sodium hydroxide or standard ACS buffer solutions.

3. Results and discussions

The synthesized NCs were broadly size-distributed in the range of 100–1400 nm, with a mean size of 440 nm and were almost all of a regular spherical shape (Fig. 1a–c). Similar W₁₈O₄₉ nanospheres were obtained solvothermally¹⁹ from tungsten ethanolate, elucidating the precursor concentration to be a key factor in size/shape control. The NCs composition was determined by EDX (Fig. 1c and d). The W to O atomic ratio was close to 2.7, corresponding to an oxygen deficiency in a sub-stoichiometric tungsten oxide WO_3−x with x = 0.3. In addition, EDX also detected some carbon and silicon originating from substrate and residuals of the precursors or solvents. The elemental maps (Fig. 1c) show the homogeneity of NCs in composition. The XPS data were in rather good agreement with EDX. The spectrum of the core W 4f level (Fig. S1†) exhibited distinct peaks of tungsten ions, W⁴⁺ and W⁶⁺, which are typical for W₁₈O₄₉ stoichiometry.²⁸ A ratio of W 4f peaks area, corresponding to different tungsten oxidation states determines the relative concentrations of W and O without the need for O 1s line and ASF data and confirms the oxygen deficiency in WO_3−x: x = 0.265 ± 0.005. This value is in slight disagreement with the one obtained by EDX because XPS tests just a surface layer of about 1.5 nm in thickness coinciding with an average electron escape depth. In addition, O 1s line reveals hydroxyl signal originated either from residuals of the solvent or from the adsorbed species. As it will be shown later, hydroxyl groups can cure oxygen vacancies and affect the oxidation state of tungsten. W₁₈O₄₉ crystallinity was monitored by XRD. All the diffraction peaks (Fig. 1e) were perfectly matched with those of monoclinic W₁₈O₄₉ (P2/m space group, JCPDS no. 84-1516). In the case of anisotropic nanowires,²² the diffraction peaks resulting from the planes of the preferable growth direction (i.e., (010) plane) were narrower. Herein, we observed an opposite trend; all the peaks were diffused showing almost isotropic growth of NCs.

Fig. 1 Representative (a) and detailed (b) SEM images, EDX elemental maps (c) and spectrum (d), XRD pattern (e) of WO_3−x NCs.

One of the distinctive manifestations of semiconductor plasmons is their tunable optical absorbance (Fig. 2). The optical density minimum at about 350–360 nm followed towards the blue area by an absorption onset, which correlates with W₁₈O₄₉ nanowires data²⁹ and originated by the band gap of W₁₈O₄₉. The slightly weaker absorbance (λ > 500 nm) is due to oscillations of the free electrons, formed in the conduction band by anion vacancies. Thus, we tested the optical absorbance of the NCs dispersed in different polar protic solvents (Fig. 2). For the sample, dispersed in ethanol a well-resolved LSPR peak appears at a wavelength of 556 nm. The position of the LSPR maximum in W₁₈O₄₉ NCs correlates almost linearly with the solvent refraction index (Fig. 3 and Table 1) compared to that of Cu_2−xS.⁶ That could be explained by the absence of surfactants, affecting the refraction on NC-solvent boundary. The Milli-Q water dispersion of NCs does not match the value of that tendency, showing up to 20% of free electrons losses (Table 1). The effect of the different pH on the LSPR properties of NCs is presented in Fig. 4. The addition of acids to a NCs solution gradually changes the LSPR peak, which was found to lose some part of the resonance quality and was blue-shifted up to 80 nm. This shift was enhanced by decreasing the pH to 4. NCs dispersed in acidic solutions substantially vary neither in size nor in shape elucidating the insignificance of the etching effects (Fig. S2†). In addition, the resonance wavelength is slightly tuned back to the red by the higher pH values, but we could not demonstrate the complete reversibility of the tuning. The blue shift in the extinction frequencies for the Au nanoparticles deposited on the electrode under a negative potential has recently been reported in ref. 32. The metal oxide surface acquires positive charge at pH values lower than the pH at the point of zero charge (PZC) and negative at pH values larger than the PZC. In the literature, different PZC values have been reported for WO₃. In one study³³ the PZC is given at pH value of 2, whereas in another study,³⁴ it is given at pH = 4–5. This means that at pH values >2 or >5 the surface of WO_3−x will be negatively charged. Therefore, the blue shift observed in the absorption frequencies in this study most probably originates from the negatively charged surface of tungsten oxide. However, we would like to mention that the data on the surface charging of WO_3−x is scarce in the literature and more study needs to be performed.

Fig. 2 UV-vis-NIR absorbance of WO_3−x nanocrystals – LSPRs in different protic polar solvents.

Fig. 3 UV-vis absorbance of WO_3−x nanocrystals – shifts of LSPR maximum with changing of solvent refractive index.

Table 1 Parameters of WO_3−x nanocrystals plasmon resonances^a

Sample	λsp (nm)	ωsp (s^−1)	ωp (s^−1)	Nn (cm^−3)
a -surface plasmon, -bulk plasmon frequencies; γ-spectral line width, Ne and me – concentration and effective mass of electrons,^30 respectively. The dielectric constants were taken as εm = n^2 for the corresponding wavelength.
WO3−x–C3H7OH	542	3.5 × 10^15	6.00 × 10^15	1.76 × 10^22
WO3−x–C2H5OH	556	3.4 × 10^15	5.98 × 10^15	1.75 × 10^22
WO3−x–CH3OH	575	3.3 × 10^15	5.96 × 10^15	1.71 × 10^22
WO3−x–HOH	613	3.0 × 10^15	5.42 × 10^15	1.43 × 10^22

---

Fig. 4 UV-vis-NIR absorbance of an aqueous solution of WO_3−x nanocrystals – LSPRs tuning versus pH of the solvent. The band in the NIR at about 975 nm was not connected with LSPR as being an intrinsic overtone/combination water absorbance.³¹

The addition of a base such as NaOH (Fig. 4) resulted in NCs solution bleaching and a red shift of the absorption band offset was observed simultaneously (λ > 400 nm). As a result, the concentration of free electrons at the bottom of the conductivity band was lowered enough not to manifest Moss–Burstein effect for degenerated materials.⁶ As a consequence, it was assumed that the base solution either traps electrons from NCs conduction band or cures oxygen intrinsic defects in tungsten suboxide. In both cases, the electron density decreases to the level inappropriate for LSPR observance. By changing the pH to more acidic, the overall tuning of LSPR almost matches the absorbance of NCs dispersed in methyl alcohol characterized by the similar water refractive index (Fig. 3). We assume that the increase in free electrons concentration limited to that one of 1.7 × 10²² cm⁻³ (Table 1). The addition of protons just cures the trap states appeared during the NCs surface–water interaction, altering the carrier concentration by 18% to the value of the as-synthesized sample.

4. Conclusions

We demonstrate a simple and facile modification of hydrothermal method employed to share control of plasmonic WO_3−x nanospheres. In contrast to well-studied WO_3−x nanowires with a broad vis-NIR plasmonic absorbance tail, spherical NCs possess an easily tunable LSPR band. Considering that the LSPR absorbance is of reasonable optical density and its essential blue shift in acidic pH solutions, we suggest the possible applicability of such plasmonic particles for the quantitative sensing of acidic media and the qualitative elucidation of the presence of bases.

Acknowledgements

One of the authors (O.A.B.) acknowledges the support from the Swedish Institute in form of a Scholarship for Postdoctoral research.

References

1. K. A. Willets and R. P. Van Duyne, Annu. Rev. Phys. Chem., 2007, 58, 267–297.
2. S. Eustis and M. A. El-Sayed, Chem. Soc. Rev., 2006, 35, 209–217.
3. X. Liu and M. T. Swihart, Chem. Soc. Rev., 2014, 43, 3908–3920.
4. M. V. Kovalenko, L. Manna, A. Cabot, Z. Hens, D. V. Talapin, C. R. Kagan, V. I. Klimov, A. L. Rogach, P. Reiss, D. J. Milliron, P. Guyot-Sionnnest, G. Konstantatos, W. J. Parak, T. Hyeon, B. A. Korgel, C. B. Murray and W. Heiss, ACS Nano, 2015, 9, 1012–1057.
5. T. M. Mattox, X. Ye, K. Manthiram, P. J. Schuck, A. P. Alivisatos and J. J. Urban, Adv. Mater., 2015, 27, 5830–5837.
6. J. M. Luther, P. K. Jain, T. Ewers and A. P. Alivisatos, Nat. Mater., 2011, 10, 361–366.
7. P. L. Saldanha, R. Brescia, M. Prato, H. Li, M. Povia, L. Manna and V. Lesnyak, Chem. Mater., 2014, 26, 1442–1449.
8. H.-J. Yang, C.-Y. Chen, F.-W. Yuan and H.-Y. Tuan, J. Phys. Chem. C, 2013, 117, 21955–21964.
9. J. S. Niezgoda, M. A. Harrison, J. R. McBride and S. J. Rosenthal, Chem. Mater., 2012, 24, 3294–3298.
10. J. S. Niezgoda, E. Yap, J. D. Keene, J. R. McBride and S. J. Rosenthal, Nano Lett., 2014, 14, 3262–3269.
11. K. Manthiram and A. P. Alivisatos, J. Am. Chem. Soc., 2012, 134, 3995–3998.
12. T. R. Gordon, M. Cargnello, T. Paik, F. Mangolini, R. T. Weber, P. Fornasiero and C. B. Murray, J. Am. Chem. Soc., 2012, 134, 6751–6761.
13. X. Liu, C. Lee, W.-C. Law, D. Zhu, M. Liu, M. Jeon, J. Kim, P. N. Prasad, C. Kim and M. T. Swihart, Nano Lett., 2013, 13, 4333–4339.
14. C. M. Hessel, V. P. Pattani, M. Rasch, M. G. Panthani, B. Koo, J. W. Tunnell and B. A. Korgel, Nano Lett., 2011, 11, 2560–2566.
15. K. Miszta, F. Greullet, S. Marras, M. Prato, A. Toma, M. Arciniegas, L. Manna and R. Krahne, Nano Lett., 2014, 14, 2116–2122.
16. M. Guo, W.-C. Law, X. Liu, H. Cai, L. Liu, M. T. Swihart, X. Zhang and P. N. Prasad, Plasmonics, 2014, 9, 893–898.
17. X. Wang, Y. Ke, H. Pan, K. Ma, Q. Xiao, D. Yin, G. Wu and M. T. Swihart, ACS Catal., 2015, 5, 2534–2540.
18. J. Huang, C. Liu, Y. Zhu, S. Masala, E. Alarousu, Y. Han and A. Fratalocchi, Nat. Nanotechnol., 2016, 11, 60–67.
19. C. Guo, S. Yin, M. Yan, M. Kobayashi, M. Kakihana and T. Sato, Inorg. Chem., 2012, 51, 4763–4771.
20. Z. Lou, Q. Gu, L. Xu, Y. Liao and C. Xue, Chem.–Asian J., 2015, 10, 1291–1294.
21. O. A. Balitskii, M. Sytnyk, J. Stangl, D. Primetzhofer, H. Groiss and W. Heiss, ACS Appl. Mater. Interfaces, 2014, 6, 17770–17775.
22. S. Cong, Y. Yuan, Z. Chen, J. Hou, M. Yang, Y. Su, Y. Zhang, L. Li, Q. Li, F. Geng and Z. Zhao, Nat. Commun., 2015, 6, 7800.
23. Y. Zhao, L. Cao, J. Ouyang, M. Wang, K. Wang and X.-H. Xia, Anal. Chem., 2013, 85, 1053–1057.
24. P. Chen, M. Qin, Y. Liu, B. Jia, Z. Cao, Q. Wan and X. Qu, New J. Chem., 2015, 39, 1196–1201.
25. Y. Feng, H. Kai, N. Shibing, W. Qi and H. Deyan, Nanoscale Res. Lett., 2010, 5, 416–419.
26. V. Hariharana, M. Parthibavarmana and C. Sekara, J. Alloys Compd., 2011, 509, 4788–4792.
27. H. Zhou, Y. Shi, Q. Dong, L. Wang, H. Zhang and T. Ma, RSC Adv., 2014, 4, 42190–42196.
28. J. Song, Z.-F. Huang, L. Pan, J.-J. Zou, X. Zhang and L. Wang, ACS Catal., 2015, 5, 6594–6599.
29. D. C. Grauer and A. P. Alivisatos, Langmuir, 2014, 30, 2325–2328.
30. A. Polaczekt, M. Pekala and Z. Obuszko, J. Phys.: Condens. Matter, 1994, 6, 7909–7919.
31. C. E. Lund Myhre and C. J. Nielsen, Atmos. Chem. Phys., 2004, 4, 1759–1769.
32. A. M. Brown, M. T. Sheldon and H. A. Atwater, ACS Photonics, 2015, 2, 459–464.
33. M. Anik and T. Cansizoglu, J. Appl. Electrochem., 2006, 36, 603–608.
34. L. M. Bertus and R. A. Carcel, Environ. Eng. Manage. J., 2011, 10, 1021–1026.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07938e

---