Fabrication of tungsten nanocrystals and silver–tungsten nanonets: a potent reductive catalyst

Abstract

This communication reports the first synthesis of well-defined colloidal W nanocrystals, Ag–W heterostructures and their broad plasmon absorbance in the visible to near infrared region. Strong interparticle plasmonic interactions significantly improve the photocatalytic reduction performance of silver–tungsten heterostructures.

---

Nanometals are attractive to researchers because of their unique physical and chemical properties.¹ Among the various types of physiochemical properties, Localised Surface Plasmon Resonance (LSPR) of metal nanocrystals (NCs), which arises from the collective oscillation of free electrons on a metal surface coupled with an electromagnetic wave, is an intriguing phenomenon. LSPR is currently of high interest and has already been applied in different fields including plasmonic based light harvesting processes, non-linear optics, biosensing, surface enhanced Raman spectroscopy, photocatalysis and high-speed photovoltaic (PV) solar cells.² Plasmonic features in metal NCs can be tuned in the visible-near infrared (VIS-NIR) region by regulating the size,^3a,b shape,^3c–e degree of alloying^3f–h or by heterostructure formation.^3i–l High energy ultraviolet (UV) plasmon bands of coinage metal NCs (Au, Ag, Cu) are rarely reported due to their instability in small dimensions, rapid degradation of the plasmon bands and predomination of interband transitions. Recently,⁴ a high energy plasmon band tuneable up to the low energy region was found for aluminium nanocrystals but it faced instability in ambient conditions due to the formation of a thin Al₂O₃ layer over the surface of the NCs. Some theoretical and experimental studies have revealed that non-noble metals (such as In, W, Ga, Sn, Pb, Bi) might show interesting plasmonic properties⁵ and find an application in plasmonic-based catalysis. Tungsten nanoparticles have gained attention mainly due to their widespread applications such as light-emitting sources,^6a hard materials,^6b thermionic cathodes, high power batteries and catalysis,^6c etc. Tungsten exists in two crystalline forms, α and β (where α is the more stable phase). The high oxophilicity of W makes the synthesis of free-standing well defined W(0) NCs very difficult. In some recent studies W(0) NCs have been synthesized by a thermal decomposition process, using ionic liquids,^7a hazardous reduction processes,^7b sonochemical methods,^7c reduction of WO₃ ^7d or chemical vapor deposition.^7e But in most cases a non-uniform particle distribution or non-crystalline amorphous nanoparticles were found.

The theoretically calculated plasmonic absorbance band of spherical shaped 10 nm W(0) NCs was found at 310 nm ⁸ and it is expected that the band position can be tuned in the UV-VIS region by changing the size, shape, alloying or by heterostructure formation. Due to the high oxophilic nature of tungsten (E₀ = −1.1 V, NHE for W⁺⁶ → W⁰), it is capable of reducing organic nitro compounds that have suitable reduction potential values (say, E⁰ for nitrobenzene/aniline = 0.42 V, NHE) from a thermodynamical point of view. Many non-plasmonic transition metals like Pd, Pt, and Ni show catalytic activity in chemical reactions and in most of the cases this is driven by conventional heating processes which may have some negative effects like the formation of unwanted products. So the synthesis of heterostructured or alloyed NCs containing a plasmonic metal with a catalytically active transition metal is important as it could effectively harvest light energy through plasmonic absorption and could increase the catalytic ability of transition metals without heat treatment.

Noble metal nanoparticles (Ag, Au) exhibit good photocatalytic reduction^3g,9 using their surface plasmon resonance (SPR) effect driven by sunlight. In terms of available energy, visible (43%) and infrared (52%) light constitute most of the solar emission with a very low amount of UV (4%) light.¹⁰ The most important photocatalysis challenge is to fabricate new photocatalysts that can absorb a large amount of solar energy, while keeping its activity or specificity intact in ambient conditions. Plasmonic coupling among metal nanoparticles generates broad, tuneable plasmon bands for different plasmon modes (capacitive, charge transfer) with enhanced electromagnetic (EM) field. Of course, this depends upon the degree of contact between two nanostructures.¹¹ This enhancement in the EM field also increases the photonic absorbance efficiency over a wide region of the solar spectrum and might boost the reduction activity. The interaction of visible light with the plasmon resonance of metal nanoparticles can be controlled by the correct combination of dissimilar metals. Experimental and theoretical investigations indicate that the electromagnetic interaction between metallic nanoparticles is a complex function of size, inter-particle distance, composition and environment. In this communication, the objective is twofold: (i) the synthesis and optical characterization of W nanoparticles and (ii) the introduction of plasmonic coupling in a random assembly of W and Ag nanoparticles to modify the optical absorption spectra of individual metallic nanoparticles and maximise the catalytic reduction property of W nanocrystals. The plasmonic absorptions of W and Ag lie in the wavelength region of 300–400 nm. The large difference between their electronegativities, W (2.36) and Ag (1.93), may also play an important role in influencing the plasmonic coupling between metal nanoparticles. The above facts stimulated us to design W–Ag nanonets for better photocatalytic activity at room temperature.

W(0) NCs were synthesized by the thermal decomposition of tungsten hexacarbonyl [W(CO)₆] at high temperature in the non-polar solvent 1-octadecene (ODE) using oleylamine (OLAM) as an activating and stabilising agent. In a typical synthesis, 1 mM of W(CO)₆ and 0.4 mM of OLAM were dispersed in 5 ml octadecene (ODE), placed under vacuum at room temperature for 1 h and then the system was backfilled with dry Ar. The temperature of the reaction flask was slowly increased to 300 °C (10 °C min⁻¹) and the reaction was continued for 2 h. Fig. 1(A) shows the large area TEM image of loosely aggregated W(0) NCs. Fig. 1(C) depicts the closer view of an assembly of NCs with a size of 4 ± 0.5 nm. A high resolution TEM image (Fig. 1(D)) shows lattice fringes of 0.235 nm corresponding to the (111) plane of fcc tungsten. Fig. 1(E) shows the XRD pattern of as-synthesized W(0) NCs and the diffraction peaks observed at 36.83, 43.50, 62.3 correspond to (111), (200) and (220) planes of the fcc (α) phase (JCPDF no. 882339). The crystallite size calculated from the XRD pattern using the Scherrer formula is 3.6 nm and this compares well with the TEM result.

Fig. 1 (A) and (B) are TEM and STEM images of free-standing nearly monodispersed W(0) NCs over large areas. (C) A closer view of a W(0) NC assembly. (D) HRTEM image of a few NCs and the inset shows the corresponding FFT pattern. The major growth plane was found to be (111). (E) The XRD pattern of bcc W(0) NCs. (F) Time dependent absorbance spectra of W(0) NCs.

The synthesis of W(0) NCs from W(CO)₆ involved the gradual decomposition of the metal carbonyl complex and catalysis by the solvent and surfactant molecules. Decomposition of W(CO)₆ at a high temperature (∼350 °C) in the presence of octadecane and octadecylamine (the saturated forms of ODE and OLAM) does not generate crystalline W(0) NCs. Fig. S1† shows the FTIR spectra of the OLAM, ODE, OLAM–W(CO)_x and ODE–W(CO)_x intermediate complexes. The disappearance of the olefinic C–H bond stretching frequencies at 3008 cm⁻¹ and 3077 cm⁻¹ (for OLAM and ODE, respectively) of the intermediate complexes suggests a ‘π’ donation from OLAM and ODE to the metal centre.¹² A red-shifting and broadening of the N–H stretching band of OLAM at 3336 cm⁻¹ is observed for the OLAM–W(CO)_x intermediate. An increase in the intensity and broadening of the C–N stretching band at 1068 cm⁻¹ also suggests the involvement of the lone pair of OLAM in the complex formation.¹³ Heating the precursor in octadecane with OLAM also generates a W(0) cluster but the synthetic procedure creates a high degree of sublimation from ∼80 °C to 120 °C resulting in a large loss of the reactants. But decomposition in ODE decreases the degree of sublimation by forming a complex between the M(CO)_x–π bond. The decomposition of W(CO)₆ in ODE results in W(0) NCs but with a high degree of agglomeration and a polycrystalline nature which proves the point of using OLAM as a capping agent (Fig. S2†). Use of an oxygen-containing capping agent which contains oxygen in a binding group like oleic acid, stearic acid, octadecanol or trioctyl phosphine oxide (frequently used in colloidal synthesis) generates WO_3−x NCs (ESI, Fig. S3†).

The growth of W(0) free-standing NCs was studied by quenching the reaction mixture at different time intervals. Fig. S4† shows TEM images of the sample at 5, 15, 30 and 60 min of the reaction. The formation of very small-sized clusters (∼1–1.5 nm) was found in the 5 min and 15 min reaction products. The as-formed clusters undergo further growth and the TEM image of the 30 min product shows a cluster size of ∼2 nm with no such detectable crystalline phase. The crystallinity of the NCs was found to appear after 1 h of the reaction. The absorbance spectra of the product at different time intervals in Fig. 1(F) shows the presence of two types of absorbance in each set. The absorbance at 227 nm is believed to be for OLAM. This absorbance band is found to be more blue-shifted (∼30 nm) than pure OLAM (250–270 nm, Fig. S5†) which indicates the electron donation of the lone-pair of the nitrogen atom of OLAM to the metal surface. Another absorbance band at a longer wavelength was found to be red-shifted during the progress of the reaction and shows a strong localised surface plasmon resonance band centred at 273 nm for the product after 2 h of reaction, extending to the blue region. Previously, Creighton et al.⁸ predicated a plasma band at 310 nm for 10 nm spherical W colloids. Such a blue-shift of the LSPR band is due to the particle size in the quantum confinement region frequently observed for noble metal NCs.^3a

Ag–W nanonets (NNs) were synthesized by employing spherical monodispersed Ag NCs (size 9.6 ± 1.7 nm, ESI, Fig. S6†) as the seed material at the beginning of the reaction. Different compositions of Ag–W NNs have been synthesized by varying the Ag : W molar ratio. For Ag : W = 2 : 1 (AgW1), formation of W(0) NCs was found on the surface of Ag NCs (Fig. S7†). The reaction mixture quenched at 280 °C shows the presence of an amorphous shell around free-standing Ag NCs (ESI, Fig. S7B†). OLAM which was present on the surface of the Ag NCs as a capping agent might be involved in the complex formation with W(CO)₆. The amorphous moiety converted to crystalline W(0) NCs upon completion of the reaction at 300 °C for 2 h. Here, the heterogeneous nucleation of W(0) onto the solid Ag crystalline surface is favoured over the homogeneous process in solution, as the former has a lower Gibbs energy¹⁴ (ΔG) for the minimal lattice constant mismatch between Ag and W. Moreover, the presence of OLAM on the surface provides a platform for the intermediate complex formation. When the Ag : W molar ratio was lowered (1 : 2; AgW2), more surface coverage of Ag seeds by W(0) NCs was found (Fig. S8†). String-like W(0) NCs were formed on Ag seeds and extended into the solution during the growth process. When more W-precursor (Ag : W = 1 : 5, AgW5) was employed in the reaction, a crystalline W-network was found to form keeping the size and shape of the Ag NCs intact, as depicted in Fig. 2(A–D). The formation of a net-like morphology can be interpreted by heterogeneous nucleation followed by a delayed homogenous nucleation process. First W(0) NCs form along the crystallographic edges of Ag seeds and additional growth of existing W(0) areas on the seeds takes place. As the growth of W(0) is very slow, as observed for free-standing W(0) synthesis, we believe that a delayed homogeneous nucleation occurred in the solution from the reaction between free OLAM and the excess W(CO)₆ complex. The as-synthesized W(0) seeds in the solution and W(0) strings present on the Ag surface undergo coalescence to form a net-like structure. The HRTEM image in Fig. 2(D) shows the presence of W(0) NCs on the Ag surface and also at the near surface (<1 nm). Elemental mapping of the Ag–W nanonets (AgW5) in Fig. S10† confirmed the presence of Ag NCs without self-agglomeration and W NCs around the Ag moiety and extended network region. The presence of some free-standing W(0) NCs (Fig. S8 and S10†) also supports the delayed nucleation of W(0) in solution.

Fig. 2 (A) TEM image of AgW₅ nanonets. (B) The HRTEM image of AgW₅ shows the presence of W(0) on the surface of the Ag NCs and also in the close vicinity. (C) The STEM image of the nanonet structure shows the formation of a network over a micrometer region and the presence of Ag NCs (bright spots) without self-agglomeration. (D) Presence of the W(0) network around the Ag NCs and the connections to other neighboring Ag NCs. (E) Normalized absorbance spectra of different Ag–W heterostructures.

One important feature of the as-synthesised Ag–W heterostructures is that they exhibit a well-defined SPR absorbance band, tuneable over the whole solar spectrum region and dependent upon the relative amounts of Ag and W, as demonstrated in Fig. 2(E). Pure Ag NCs which were used as seeds show a strong plasmon band centred at 404 nm. The absorbance peak maximum was found to be red-shifted and broadened when W NCs formed around the surface and gradually formed a network. Another plasmon band in the UV region 260 nm to 275 nm appeared for the W NCs. The red-shift and broadening of the Ag plasmonic is not due to a morphology change or alloy formation as we found no change in shape of the Ag seeds or alloy (Ag–W) formation from TEM or STEM (element mapping) study. Here, plasmonic coupling between Ag and W NCs is expected as W NCs are present on the Ag surface or in a very close vicinity. The LSPR frequency position of the metal NCs depends on the dielectric constant of the dispersion medium. The presence of W NCs changes the dielectric environment of Ag NCs by forming heterostructures or nanonets. The dielectric constant value of bulk tungsten is much higher than that of silver metal.^15a,b So, the enclosing of the Ag NCs by the W NCs increases the dielectric constant value of the surroundings of the Ag NCs and results in a gradual red-shift of the Ag plasmonic peak as the W concentration increases. The broad peaks for Ag–W NNs, mainly composed of 2–3 humps, might be the overlap of different plasmon modes such as capacitive plasmon and charge transfer plasmon modes due to random inter-particle distances. The plasmonic sensitivity of the Ag–W heterostructures is found to decrease upon increasing the W concentration. Fig. S13† shows the experimental change of the LSPR peak maxima of the NCs/NNs with the refractive index of solvents. Pure Ag NCs show a plasmonic sensitivity of ∼145 nm per RIU and this decreases to 79 nm per RIU for the AgW5 nanonet sample. The Ag–W heterostructures (especially Ag–W nanonets) are very efficient in harvesting solar light energy (over the full solar spectrum) as they contain strong plasmonic bands in the UV, VIS and NIR regions.

The exclusive heterostructure and plasmonic optical properties of the synthesized bimetallic Ag–W composites provide an opportunity to study light-enhanced catalytic applications. The plasmonic properties of the synthesized Ag–W nanonets have been utilized by developing a new reduction method under mild conditions. Reduction of a nitro group to an amine has been well known chemistry since 1948.¹⁶ But the uniqueness of our strategy includes the Ag–W catalyzed kinetic acceleration of the reduction process under solar illumination. Aliphatic as well as aromatic nitro compounds (1) underwent reduction into their corresponding amine functionalities (2) in the presence of Ag–W nanonets (5 mol%) and hydrazine hydrate [N₂H₄·H₂O] (1.1 equiv.) (eqn (1), Scheme 1).

Scheme 1 Plasmonic Ag–W nanonet catalyzed reduction acceleration.

To compare the catalytic activities of different nanostructures we studied the reduction of p-nitrotoluene to p-toluidine under similar reaction conditions using a light irradiance of 0.65 W cm⁻² (Fig. 3(A)). The pure Ag NCs were found to be catalytically inactive whereas the W NCs showed poor catalysis. Formation of Ag–W heterostructures was found to be effective and the maximum activity was found for the AgW5 NN structure which shows the maximum plasmonic absorbance in the visible and NIR regions. To study the dependence of the catalytic rate on light irradiance we carried out the same reaction using the AgW5 catalyst at different light irradiances (standardised with a photometer). The almost linear dependence of the % conversion with the light power density clearly indicates that the SPR properties of the heterostructures play a vital role in solar light-driven catalytic processes. For a clearer understanding of the catalytic process, XPS was performed on the AgW5 catalyst and recovered catalyst after 15 min of reaction (Fig. S14†). The as-synthesized AgW5 catalyst shows two peaks at 31.58 eV and 33.68 eV for W 4f_7/2 and W 4f_5/2, respectively, owing to spin–orbit splitting. The recovered catalyst shows four peaks in the binding energy range 30–40 eV. Deconvolution results in two peaks in the higher energy region, 35.62 and 37.63 eV, assigned to W⁶⁺ and lower energy peaks at 31.66 and 33.67 eV corresponding to W(0).^17a,b So during the reduction process, W(0) might be oxidised to the W⁶⁺ state and then reduced back to W(0) by hydrazine.

Fig. 3 (A) Catalytic activity of different nanostructures. (B) Dependence of the catalytic activity of AgW5 on light irradiance.

An optimization study (ESI,† Table 1.1) revealed that only 5 mol% of the catalyst is sufficient for it to perform its activity and commonly used hydrazine was found to be a potential candidate for catalyst regeneration. Under the developed conditions, different varieties of nitro compounds were screened to illustrate the robustness and generalization of the procedure. Aromatic ring-possessing electron donating (entries 1 and 7, Table 1) and electron withdrawing substituents (entries 6, 8 and 9, Table 1), heterocyclic ring (entry 3), and aliphatic analogues (entries 2, 4 and 5) are well tolerated for the formation of valuable amino synthons (2a–j, Table 1). All of the structures were confirmed by spectroscopic analyses (ESI†) and compared with the literature. A surprising result was obtained for 1,2-dinitrobenzene (entry 6) as both of the nitro groups were reduced to o-phenylenediamine (2f) very easily. Thus, the developed strategy demonstrates a simple, non-hazardous reduction technique with fast reaction convergence.

Table 1 Substrate scope for the catalytic reduction process

Entry	Starting material (1)	Product (2)	Time (h)	Yield (%)
1			1.1	2a, 78
2			1.0	2b, 71
3			1.0	2c, 73
4			0.8	2d, 66
5			0.8	2e, 70
6			1.2	2f, 63
7			1.1	2g, 66
8			1.0	2h, 80
9			1.0	2i, 82
10			1.1	2j, 70

---

In Summary, we have successfully synthesized free-standing W(0) NCs by a thermal decomposition process and have fabricated bimetallic Ag–W heterostructures using Ag seeds with tunable SPR features. The Ag–W nanonet structure shows efficient reductive photocatalytic activity. Highly oxophilic W in the Ag–W heterostructure acts as a catalytic site as evidenced from the XPS study. The design of heterostructures comprising metal nanoparticles was found to be crucial for the enhancement of the photocatalytic process.

Acknowledgements

S. G. and M. S. acknowledge CSIR, India for providing the fellowship (SRF and JRF) during the tenure of the work.

Notes and references

1. (a) P. K. Jain, X. Huang, I. H. El-Sayed and M. A. El-Sayed, Acc. Chem. Res., 2008, 41, 1578; (b) M. E. Stewart, C. R. Anderton, L. B. Thompson, J. Maria, S. K. Gray, J. A. Rogers and R. G. Nuzzo, Chem. Rev., 2008, 108, 494; (c) S. Linic, P. Christopher and D. B. Ingram, Nat. Mater., 2011, 10, 911–921; (d) J. A. Schuller, E. S. Barnard, W. S. Cai, Y. C. Jun, J. S. White and M. L. Brongersma, Nat. Mater., 2010, 9, 193–204.
2. (a) Y. Zakharko, T. Nychyporuk, L. Bonacina, M. Lemiti and V. Lysenko, Nanotechnology, 2013, 24, 055703; (b) J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and R. P. V. Duyne, Nat. Mater., 2008, 7, 442; (c) J.-H. Lee, J.-M. Nam, K.-S. Jeon, D.-K. Lim, H. Kim, S. Kwon, H. Lee and Y. D. Suh, ACS Nano, 2012, 6, 9574; (d) P. D. Howes, S. Rana and M. M. Stevens, Chem. Soc. Rev., 2014, 43, 3835; (e) S. Linic, P. Christopher and D. B. Ingram, Nat. Mater., 2011, 10, 911; (f) X. Zhang, Y. L. Chen, R.-S. Liu and D. P. Tsai, Rep. Prog. Phys., 2013, 76, 046401; (g) S. Pillai, K. R. Catchpole, T. Trupke and M. A. Green, J. Appl. Phys., 2007, 101, 093105.
3. (a) J. A. Scholl, A. L. Koh and J. A. Dionne, Nature, 2012, 483, 421; (b) S. Peng, J. M. McMahon, G. C. Schatz, S. K. Gray and Y. Sun, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 14530; (c) H. Chen, L. Shao, K. C. Woo, T. Ming, H.-Q. Lin and J. Wang, J. Phys. Chem. C, 2009, 113, 17691; (d) A. Tao, P. Sinsermsuksakul and P. Yang, Nat. Nanotechnol., 2007, 2, 435; (e) B. Nikoobakht and M. A. El-Sayed, Chem. Mater., 2003, 15, 1957; (f) N. E. Motl, E. Ewusi-Annan, I. T. Sines, L. Jensen and R. E. Schaak, J. Phys. Chem. C, 2010, 114, 19263; (g) R. He, Y.-C. Wang, X. Wang, Z. Wang, G. Liu, W. Zhou, L. Wen, Q. Li, X. Wang, X. Chen, J. Zeng and J. G. Hou, Nat. Commun., 2014, 5, 4327; (h) M. P. Mallin and C. J. Murphy, Nano Lett., 2002, 2, 1235; (i) S. Mandal, P. R. Selvakannan, R. Pasricha and M. Sastry, J. Am. Chem. Soc., 2003, 125, 8440; (j) Y. Ma, W. Li, E. C. Cho, Z. Li, T. Yu, J. Zeng, Z. Xie and Y. Xia, ACS Nano, 2010, 4, 6725; (k) C. Liusman, H. Li, G. Lu, J. Wu, F. Boey, S. Li and H. Zhang, J. Phys. Chem. C, 2012, 116, 10390; (l) O. Pena-Rodríguez, U. Pal, M. Campoy-Quiles, L. Rodríguez-Fernandez, M. Garriga and M. I. Alonso, J. Phys. Chem. C, 2011, 115, 6410.
4. (a) M. W. Knight, N. S. King, L. Liu, H. O. Everitt, P. Nordlander and N. J. Halas, ACS Nano, 2014, 8, 834; (b) M. W. Knight, L. Liu, Y. Wang, L. Brown, S. Mukherjee, N. S. King, H. O. Everitt, P. Nordlander and N. J. Halas, Nano Lett., 2012, 12, 6000.
5. (a) C. Kind and C. Feldmann, Chem. Mater., 2011, 23, 4982; (b) J. M. McMahon, G. C. Schatz and S. K. Gray, Phys. Chem. Chem. Phys., 2013, 15, 5415.
6. (a) P. Heszler, L. Landstrom, M. Lindstam and J. O. Carlsson, J. Appl. Phys., 2001, 89, 3967; (b) J. C. Bailar and H. J. Emeleus, Comprehensive Inorganic Chemistry, 1973, 3, 742; (c) L. Xiong and T. He, Chem. Mater., 2006, 18, 2211.
7. (a) E. Redel, R. Thomannb and C. Janiak, Chem. Commun., 2008, 1789; (b) C. Schottle, P. Bockstaller, D. Gerthsen and C. Feldmann, Chem. Commun., 2014, 50, 4547; (c) H. Lei, Y.-J. Tang, J.-J. Wei, J. Li, X.-B. Li and H.-L. Shi, Ultrason. Sonochem., 2007, 14, 81; (d) H. H. Nersisyan, J. H. Lee and C. W. Won, Combust. Flame, 2005, 142, 241; (e) L. Landstrom, J. Lu and P. Heszler, J. Phys. Chem. B, 2003, 107, 11615.
8. J. A. Creighton and D. G. Eadon, J. Chem. Soc., Faraday Trans., 1991, 87, 3881.
9. (a) X. Ke, S. Sarina, J. Zhao, X. Zhang, J. Changa and H. Zhu, Chem. Commun., 2012, 48, 3509; (b) T. Wu, S. Liu, Y. Luo, W. Lu, L. Wanga and X. Sun, Nanoscale, 2011, 3, 2142; (c) D. Kumar, S. Kaur and D.-K. Lim, Chem. Commun., 2014, 50, 13481.
10. Z. Zou, J. Ye, K. Sayama and H. Arakawa, Nature, 2001, 414, 625.
11. J.-H. Lee, M.-H. You, G.-H. Kim and J.-M. Nam, Nano Lett., 2014, 14, 6217.
12. (a) S. Ghosh, K. Das, K. Chakrabarti and S. K. De, Dalton Trans., 2013, 3434; (b) P. J. Thistlethwaite and M. S. Hook, Langmuir, 2000, 16, 4993.
13. S. Ghosh, M. Saha and S. K. De, Nanoscale, 2014, 6, 7039.
14. (a) J. Park, J. Joo, S. G. Kwon, Y. Jang and T. Hyeon, Angew. Chem., Int. Ed., 2007, 46, 4630; (b) E. E. Finney and R. G. Finke, J. Colloid Interface Sci., 2008, 317, 351.
15. (a) L. V. Nomerovannaya, M. M. Kirillova and M. M. Noskov, Soviet Physics JETP, 1971, 33, 405; (b) P. B. Johnson and R. W. Christy, Phys. Rev. B: Condens. Matter Mater. Phys., 1972, 6, 4370.
16. J. Pearson, Trans. Faraday Soc., 1948, 44, 683.
17. (a) J. Li, Y. Liu, Z. Zhu, G. Zhang, T. Zou, Z. Zou, S. Zhang, D. Zeng and C. Xie, Sci. Rep., 2013, 3, 2409; (b) N. V. Alov, J. Anal. Chem., 2005, 60, 431.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures, additional characterisation and others. See DOI: 10.1039/c4ra16567e

---