Journal of Colloid and Interface Science 460 (2015) 128–134

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

journal homepage: www.elsevier .com/locate / jc is

Plasmonic-polymer hybrid hollow microbeads for surface-enhancedRaman scattering (SERS) ultradetection

http://dx.doi.org/10.1016/j.jcis.2015.08.0470021-9797/� 2015 Elsevier Inc. All rights reserved.

⇑ Corresponding authors at: Centro de Tecnologia Quimica de Cataluña and Universitat Rovira i Virgili, Carrer de Marcel�lí Domingo s/n, 43007 Tarragona, SpaiE-mail addresses: luca.guerrini@ctqc.org (L. Guerrini), ramon.alvarez@urv.cat (R.A. Alvarez-Puebla).

Anna Trojanowska a,b, Nicolas Pazos-Perez b,c, Cinta Panisello b, Tania Gumi a, Luca Guerrini b,c,⇑,Ramon A. Alvarez-Puebla b,c,d,⇑aDepartament d’Enginyeria Química, Escola Tècnica Superior d’Enginyeria Química, Universitat Rovira i Virgili, Av. Països Catalans, 26, 43007 Tarragona, SpainbCentro de Tecnologia Quimica de Cataluña and Universitat Rovira i Virgili, Carrer de Marcel�lí Domingo s/n, 43007 Tarragona, SpaincMedcom Advance SA, Viladecans Business Park – Edificio Brasil, Bertran i Musitu 83-85, 08840 Viladecans – Barcelona, Spaind ICREA, Passeig Lluís Companys 23, 08010 Barcelona, Spain

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 15 July 2015Revised 20 August 2015Accepted 22 August 2015Available online 24 August 2015

Keywords:Surface-enhanced Raman spectroscopyPlasmonicMetallic nanoparticlesHybrid materialsPolysulfonePhase inversion precipitationHot-spotsAnalyte accumulationMicrobeadsUltradetection

a b s t r a c t

Hybrid composites are known to add functionality to plasmonic nanomaterials. Although these sub-strates can be produced by common synthetic methods, the percentage of metal loaded into the func-tional material is usually small. Herein, we exploit a phase inversion precipitation method toincorporate large amounts of silver nanoparticles inside the polymeric matrix of polysulfone microbeads.The composite material combines the high SERS activity resulting from the plasmonic coupling of highlyinteracting nanoparticles and the ability to accumulate analytes of the polysulfone porous support. Thisallows for the quantitative SERS detection down to the nanomolar level, with a liner response thatextends over an impressive concentration range of five orders of magnitude.

� 2015 Elsevier Inc. All rights reserved.

n.

A. Trojanowska et al. / Journal of Colloid and Interface Science 460 (2015) 128–134 129

1. Introduction

Surface-enhanced Raman spectroscopy (SERS) is an ultrasensi-tive technique for the identification of a variety of analytes [1–4].However, since its initial discovery [5], this technique has been lar-gely limited to qualitative and semi-quantitative applications dueto the intrinsic inhomogeneities derived from the fabrication ofactive plasmonic materials, as well as from their largely variableand molecular-specific binding affinity for different analytes. Inthe recent years several works described new plasmonic platformswith the ability of reporting quantitative results. These studies aremainly based in lithographic techniques including physical evapo-ration with electron or ion beam patterning, nanosphere lithogra-phy or direct imprinting of plasmonic colloidal particles [6].Despite their good performances, these methods require expensivesetups for fabrication, and cannot be easily industrialized. Addi-tionally, linearity in their optical response is limited to no morethan two orders of magnitude [7,8].

To address such issues, colloidal fabrication of plasmonic sub-strates for SERS has evolved in different directions, such as thecombination of metallic nanoparticles with non plasmonic materi-als acting as support or matrix. The principal goals of constructingthese composite materials include the formation of stable hotspots, the increase in size of the plasmonic platform up to themicrometer scale in order to make easier their integration intosensing devices; and the development of analyte-trapping capabil-ity to concentrate the target molecule. Materials selected as sup-port for plasmonic particles include inorganic oxides, mainlysilica [9] or iron oxides [10], and polymers such as polystyrene[11]. As a matrix, the preferred materials are usually polymersincluding agarose [12,13], polyacrylic acid [14], polyallylamine[14], polyurethane [15], poly(N-isopropyl acrylamide-acrylic acid-2-hydroxyethyl acrylate) [16] or polystyrene [15], using tech-niques such as microfluidic flow focusing [12], or suspension [17]and living polymerization [18]. Notably, these fabrication methodsgive rise, on the one hand, to dense polymer structures with lowporosity, thus hampering the analyte diffusion inside the matrixand, on the other hand, to low amount of trapped nanoparticles(around 1% as compared with the total weight of the material),which severely limits the extensive generation of hot spots (i.e.low SERS activity). Furthermore, the analyte diffusion to the innerparts of the microparticles may be undetected by SERS due to thelimited laser penetration across the thick composite structure,leading to an underestimation of the analyte detection.

These disadvantages mainly arise from the fact that most of thefabrication methods of nanoparticle-polymer composites requirethe mixing of liquid phases where the solubility of the polymersis much larger than that of the nanoparticles (i.e. the nanoparticleconcentration is low). On the other hand, a minimum quantity ofpolymer is required to form the polymer matrix, which remainsmuch larger than the maximum tolerated nanoparticle concentra-tion in the mixture. Furthermore, the combination of hydrophilicand hydrophobic solvents as required in some fabrication methods(i.e. microfluidics or suspension and dispersion polymerization),usually leads to colloidal instability and, thus, uncontrollednanoparticle aggregation.

As an alternative synthetic strategy to circumvent such prob-lems, one can relies on layer-by-layer protocols [14]. However,these methods are extremely time-consuming, especially for thefabrication of large batches of discrete microcapsules. On the con-trary, phase inversion precipitation methods [19] allow to increasethe nanoparticle content by reducing the amount of polymer farbelow the minimal concentration required by the commonmethods. Importantly, since the driving force to form the beadsis air, highly porous hollow microparticles can be produced, thus

favoring the diffusion of the analyte through the support and itspositioning close/onto the plasmonic surfaces.

Herein, we present a novel synthetic strategy for the incorpora-tion of silver nanoparticles (AgNPs) in high content within thepolysulfone microbeads (PSf) of high porosity, based on the phaseinversion precipitation method. As a result, these composite mate-rials (PSf/Ag) accommodate a densely packed collection of efficienthot spots while providing large accessibility for the target mole-cules to the metallic surfaces, allowing for the SERS quantitativedetection in a window of five orders of magnitude.

2. Experimental

2.1. Materials

Trisodium citrate, silver nitrate (99.9%, AgNO3),Polyvinylpyrrolidone (PVP, Mw = 40,000), N,N-Dimetilformamida(DMF), ethanol (EtOH), and Polysulfone (PSf, Mw = 16,000) werepurchased from Sigma Aldrich. All reactants were used withoutfurther purification. Milli-Q water was used in all aqueoussolutions.

2.2. Synthesis of silver nanoparticles

Silver nanoparticles (AgNPs) of ca. 50 nm diameter were syn-thesized as follows. Briefly, 1 L of milli-Q water was heated undervigorous stirring. 6.8 mL of an aqueous solution of trisodiumcitrate (0.1 M) and 0.996 mL of an aqueous solution of ascorbic acid(0.1 M) were consecutively added to the boiling water. After 1 min,0.744 mL of AgNO3 (0.1 M) were also added to the mixture. Thesolution was kept boiling for 1 h under stirring and then left to cooldown to room temperature. AgNPs were functionalized with PVPby mixing 1 L of colloids into an aqueous solution of PVP (5 g in20 mL) to yield polymer-stabilized nanoparticles (AgNPs@PVP).The suspension was maintained under continuous stirring during24 h at room temperature. Afterwards, the solution was cen-trifuged twice (7000 rpm, 30 min), the supernatant was discardedand the AgNPs@PVP sediment was redispersed in 1 mL of DMF.

2.3. Microbeads preparation

Hybrid polysulfone/AgNPs microbeads (PSf/Ag) were preparedby phase inversion precipitation method [20]. Briefly, a polymericsolution was prepared by dissolving 15% w/w of PSf in DMF(2.490 mL) followed by the addition of the AgNPs@PVP dispersionin DMF (1 mL). The final composition of the polymeric solution was82.3% of DMF, 15% of PSf and 2.7% of AgNPs (w/w). The mixturewas stirred during 24 h at room temperature. To form the microbe-ads, an airbrush device working in semi-continuous process with anozzle size of 800 lm was employed [20,21] to produce micro-droplets which precipitate in a coagulation bath containing500 mL of water. Finally, the microbeads were collected by filtra-tion using a 5 lm nylon filter and were stored under vacuum ina desiccator.

2.4. Characterization

Evolution 201 UV–Visible Spectrophotometer (ThermoScientific) with a Hg light source and a wavelength range from300 to 850 nm was employed to measure the absorption of silvercolloids. External morphology of PSf/Ag microbeads was analyzedby JEOL JSM 6400 Scanning Microscopy Series, with accelerationvoltage of 15–20 kV. Particles size distribution was determinedby analyzing the SEM micrographs. Internal morphology of PSf/

Fig. 1. Schematic representation of (A) the preparation of polysulfone/silver nanoparticle microbeads (Psf/Ag); and (B) their functionalization with the Raman probe for SERScharacterization.

Fig. 2. (A) Localized surface plasmon resonance of silver colloids and thecorresponding (B) nanoparticle size distribution (a representative TEM image ofthe dried colloids is also included).

Fig. 3. SEM micrographs of PSf/Ag microbeads and the corresponding sizehistogram.

130 A. Trojanowska et al. / Journal of Colloid and Interface Science 460 (2015) 128–134

Ag microbeads was analyzed via a Quanta 600 FEI Scanning Micro-scope with acceleration voltage of 20–30 kV. The wall thickness ofthe microbeads and their internal diameter were determined byESEM micrographs analysis.

The content of Ag within the capsules was determined throughan elemental composition analysis by energy-dispersive X-rayspectroscopy (EDS). Internal structural investigation of themicrobeads was performed by cross section analysis of the PSf/Ag microbeads micrographs through two different methods: resin

inclusion and cryogenic breaking. Both analysis were performedaccording to the procedure described by Torras et al. [22].Thermogravimetric analysis (TGA) was employed to determinethe inorganic content in PSf/Ag by registering the loss of weightof the organic material, associated to the polysulfone degradationand vaporization, upon temperature increasing. The microbeadssamples (about 10 mg) were introduced into an aluminum oxidecrucible and analyzed by TGA under controlled temperature anda constant flow of oxygen of 290 cm3/min by using a Perkin Elmermodel Thermobalance TGA7 device equipped with a microbalancewith an accuracy of 1 lg. The samples were heated from 30 to900 �C at a rate of 10 �C/min.

A. Trojanowska et al. / Journal of Colloid and Interface Science 460 (2015) 128–134 131

Textural characterization PSf/Ag microbeads was determinedthrough adsorption-desorption experiments using a MicrometricsASAP-2020 device. The analysis was performed with 0.0111 g ofPSf/Ag microbeads at 77.3 K. The pore size distribution was calcu-lated from the nitrogen adsorption data using the NLDFT model(Non Localized Density Functional Theory) based on a slit poreequilibrium model.

Surface-Enhanced Raman Scattering (SERS) spectra were col-lected in backscattering geometry with a Renishaw Invia Reflexsystem equipped with a 2D-CCD detector and a Leica confocalmicroscope. Samples for SERS analysis were prepared by adding10 lL of ethanolic solutions of thiophenol at different concentra-tions from 1 mM to 0.1 lM to 1 mL suspension of dispersedmicrobeads. Thin films were created by drop-casting 10 lL of thesemixtures on glass slides and then analyzed after solvent evapora-tion. A 785 nm diode laser was focused onto the dried PSf/Ag

Fig. 4. (A) Representative cross-sectional ESEM image of one microbead preparedby cryogenic break. (B) Textural characterization of Psf/Ag beads by N2 adsorption–desorption isotherm. Low N2 adsorption takes place in the relative pressure P/Po inthe range from 0 to 0.8. Above this threshold, adsorption rapidly arises due to theinteractions of N2 with previously adsorbed molecules. (C) Pore size distribution(red line) and cumulative pore volume (blue line) calculated from the N2

adsorption–desorption isotherm using the NLDFT model based on slit pore-equilibrium model. The data show pores of different sizes ranging from 2.4 to15.2 nm. (For interpretation of the references to color in this figure legend, thereader is referred to the web version of this article.)

microparticles by a �50 objective (0.75 NA) with accumulationtimes of 10 s.

The averaged SERS spectra and the spectral marker values (theSERS intensity ratio I999/I791) reported in the manuscript wereobtained as follows. First, 10 spectra were acquired on differentPSf/Ag capsules dried on a glass slide. The resulting spectra werebaseline corrected and normalized to the 791 cm�1 PSf band beforethe corresponding averaged spectrum was calculated. The SERSintensity ratio I999/I791 was acquired from such averaged SERSspectra. The same protocol was applied to the Raman characteriza-tion of the pure PSf beads.

3. Results and discussion

Fig. 1 schematically depicts the fabrication process and compo-sition of our composite material polysulfone/silver nanoparticlemicrobeads (PSf/Ag).

Controlling the geometry of the preformed nanoparticles is akey feature to obtain a homogeneous distribution of hot spots inthe hybrid material. Although gold nanoparticles can be preparedalmost in every size and shape [23,24], the presence of stronglydamped plasmon resonances in gold nanostructures when excitedwith green or more energetic lasers limits their application as effi-cient SERS materials because of the coupling to interband transi-tions [25,26]. Differently, other plasmonic materials such as

Fig. 5. (A) Elemental compositional analysis of PSf/Ag microbeads via energy-dispersive X-ray spectroscopy (EDS). The elemental spectrum clearly shows thepresence of Ag (red color) and S (green color) coming from the AgNPs andpolysulfone, respectively. (B) Thermogravimetric analysis (TGA) of Psf and PSf/Agmicrobeads. (For interpretation of the references to color in this figure legend, thereader is referred to the web version of this article.)

132 A. Trojanowska et al. / Journal of Colloid and Interface Science 460 (2015) 128–134

silver generate much efficient electromagnetic fields with thesubsequent increase in the SERS signal. Unfortunately, silver ismore reactive than gold and, thus, homogeneous preparation ofnanoparticles becomes difficult, especially as their size increases.Thus, for our composite material we selected spheroidal silverparticles with a diameter of 50 nm as a compromise in betweenhomogeneity and size. Fig. 2 shows a plasmon centered at402 nm with a narrow nanoparticle size distribution of ca. 48 nmdiameter. As a polymeric ingredient, polysulfone was selectedsince it provides a porous matrix with low SERS-cross sectionand is soluble in DMF. This latter property is important as ournanoparticles can be also concentrated in this solvent favouringthe loading of metal into the final material.

Thus, for the preparation of PSf/Ag both solutions were mixedand atomized using an airbrush to form microdroplets into a waterbath [20]. Fig. 3 shows SEM images of the resulting PSf/Agmicrobeads and the corresponding histogram which reveals amean diameter of 18 lm. The cross-sectional ESEM image of onebead (Fig. 4A) corroborates its hollow nature. The volume of theinternal cavity contributes on average to the 18.2% of the wholebead volume.

A careful examination of the cross section images reveals a porenetwork within the walls of the beads. Thus, the surfaces of thebeads were explored by means of nitrogen (77 K) adsorption(Fig. 4B). The adsorption–desorption isotherm can be clearly classi-fied as a IUPAC type III [27,28], indicating a material with low affin-ity for nitrogen. The surface area, evaluated by applying BET, gaverise to 30 m2/g with a total pore volume of 0.064 cm3/g. Notably,the corresponding values for pure PSf capsules are 27 m2/g and0.128 cm3/g, respectively [29]. Thus, incorporation of AgNPs withinthe polymer matrix generates a ca. 10% increase of surface area butwith a large reduction of the total pore volume. This fact might beassociated with the presence of the nanoparticles inside the pores.

Fig. 6. (A) Normalized averaged SERS spectra (N = 10) of TP on PSf/Ag capsules at differenconcentration dependence of the ratiometric peak intensities I999/I791. (C) Normalized ave900–1200 cm�1 spectral range and the corresponding. (D) Difference spectra obtained b

Pore size distribution was calculated by using the non-local densityfunctional theory (NLDFT) based on a slit pore equilibrium model[30]. The calculations (Fig. 4C) display an extended fluctuation ofthe pore size in the range from 2.4 to 15.2 nm, diameters largeenough for small and mediummolecules to diffuse into the matrix.

Composition of the beads was characterized by ESEM–EDS(energy-dispersive X-ray spectroscopy) and thermogavimetricanalysis (TGA) (Fig. 5). Notably, the EDS mapping shows a welldistributed amount of silver into the material while the TGAindicates contents of metal above 10%, considerably larger thanthose of similar materials produced by other methods.

The SERS performance of PSf/Ag beads was investigated by illu-minating the sample with a 785 nm laser and using thiophenol(TP) as a Raman probe. TP is a molecule with high Raman cross-section which strongly binds silver nanostructures providing acharacteristic vibrational fingerprint [31]. PSf/Ag capsules wereimmersed into TP solutions at increasing analyte concentrationand the corresponding Raman signals were acquired by focusingthe laser onto dried microbeads, deposited over a glass slide, usinga �50 objective. For each sample, at least 10 capsules were inves-tigated to yield the corresponding averaged spectra illustrated inFig. 6A. For the sake of comparison, the spectra were also normal-ized to the intensity of the strongest band. In the absence of TP,only polysulfone features appear in the spectra, such as them(C–S–C) mode at 791 cm�1, the symmetric and asymmetric SO2

�

stretching at 1073 and 1109 cm�1, respectively, and the symmetricC–O–C stretching at 1148 cm�1 [32]. As the TP concentration isprogressively increased, the characteristic bands of the molecularprobe gradually emerge such as, among others, the intense TPfeatures at 999 and 1022 cm�1, both ascribed to in-plane ringbreathing vibrations, and at 1074 cm�1, assigned to an in-planering breathing mode coupled with m(C–S) [33]. As it can be seen,TP bands can be clearly distinguished down to the nanomolar

t concentrations in the 700–1200 cm�1 spectral range and the corresponding. (B) TPraged Raman spectra (N = 10) of TP on PSf capsules at different concentrations in they subtracting the Raman signal of the pure PSf to each PSf + TP spectrum.

A. Trojanowska et al. / Journal of Colloid and Interface Science 460 (2015) 128–134 133

regime, with a limit of detection of ca. 1 nM and an estimatedanalytical enhancement factor of ca. 4 � 105 (see SupplementaryInformation for details about the calculation). It is important tostress that the 785 nm laser was selected as the optimum excita-tion source after comparing the SERS response of the PSf/Agmicrobeads for TP 10�5 M upon illumination with three differentlasers (532 nm, 633 nm and 785 nm). The corresponding SERSefficiencies decrease as follows: 785 nm > 633 nm� 532 nm. Thisresult suggests that silver nanoparticles included in the polymericmatrix are closely interacting to each other leading to a large redshift and broadening of plasmon resonance frequency to longerwavelengths.

To quantitatively monitor the SERS response of PSf/Ag sub-strates as a function of the TP concentration, we selected the PSffeature at 791 cm�1 as an internal standard and, then, calculatedthe corresponding ratiometric peak intensities I999/I791 which isplotted against the Raman probe concentration in Fig. 6B. Thecalibration curve shows a very good linear correlation (r2 > 0.98)over an impressive 5 orders of magnitude concentration range(from 1 mM to 10 nM).

As a control experiment, we performed an identical Ramanstudy on pure PSf capsules (i.e. no AgNPs) exposed to TP solutionsat different concentrations (Fig. 6C). As for PSf/Ag measurements,the samples were extensively washed prior to the SERS analysisto remove any unbound TP molecules from the dispersion. In thiscase, the data indicate that TP does accumulate inside the matrixyielding a normal Raman signal that can be observed at bulkconcentration as low as 0.1 mM. This is well revealed by the differ-ence spectra shown in Fig. 6D, which were obtained by digitallysubtracting the signal of PSf from those of the PSf + TP mixtures.

However, the overall spectral intensity remains extremelysmaller than that observed in the case of PSf/Ag indicating thatTP vibrational spectra acquired on PSf/Ag is almost entirely dueonly to the SERS contribution (i.e. TP molecules attached or in closeproximity to the metal surface). On the other hand, these findingsalso suggest that such ‘‘analyte accumulation” effect promoted bythe PSf matrix may be at the root of the remarkable extended linearrange of response of the PSf/Ag substrate. In fact, whereas in thecase of AgNPs in suspension the maximum number of detectableTP molecules is substantially limited by the surface silver areaavailable for the sulfur binding (i.e. first monolayer), for PSf/Aghybrid materials the molecular probe has the possibility to occupyand accumulate in the whole volume surrounding the metal. As aresult, a much larger number of analyte molecules can profit fromthe electromagnetic enhancement generated by the nanostruc-tures upon laser excitation.

4. Conclusions

In summary, we exploited the phase inversion precipitationmethod to incorporate large amounts of silver nanoparticles insidethe polymeric matrix of polysulfone microbeads. The synthesizedcomposite material combines the high SERS activity resulting fromthe plasmonic coupling of highly interacting nanoparticles and theability to accumulate analytes of the polysulfone porous support.This allows for the quantitative SERS detection down to thenanomolar level, with a linear response that extends over animpressive concentration range of five orders of magnitude. Thenovel synthetic strategy described herein overcomes the limita-tions of the fabrication methods normally employed in the prepa-ration of composite plasmonic/polymeric materials, such as thelow nanoparticle loading capacity (around 1% as compared withthe total weight of the composite) and the low porosity of the poly-meric matrix, which hampers the analyte diffusion inside themicrobead. We can foresee that future work will be directed

toward different goals. For instance, efforts could be devoted tothe improvement of the reported synthetic protocol such as toachieve larger nanoparticle loading, better control over bead sizedistribution, tuning of the polymeric porosity for additional size-exclusion selectivity, minimizing the extension of the PVP coatingof the nanoparticles to maximize analyte diffusion toward themetallic surfaces, etc. On the other hand, introduction of differentmetallic nanoparticles with enhanced plasmonic performance(i.e. nanorods, nanocubes, nanostars, etc.) may further increasethe sensitivity of the sensing platform.

Acknowledgments

This work was funded by the European Research Council(CrossSERS, FP7/2013 329131, PrioSERS FP7/2014 623527), theSpanish Ministerio de Economia y Competitividad (CTQ2014-59808R), the Generalitat de Catalunya (2014-SGR-612) and Med-com Advance SA. A.T. acknowledges the Departament d’EnginyeriaQuímica de la URV for funding and Dr. Renata Jastrzab (Coordina-tion Chemistry Department, Adam Mickiewicz University, Poznan,Poland).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jcis.2015.08.047.

References

[1] S. Schlücker, Surface-enhanced Raman spectroscopy: concepts and chemicalapplications, Angew. Chem. Int. Edit. 53 (2014) 4756–4795.

[2] R.S. Golightly, W.E. Doering, M.J. Natan, Surface-enhanced Raman spectroscopyand homeland security: a perfect match?, ACS Nano 3 (2009) 2859–2869

[3] F. Casadio, M. Leona, J.R. Lombardi, R. Van Duyne, Identification of organiccolorants in fibers, paints, and glazes by surface enhanced Ramanspectroscopy, Acc. Chem. Res. 43 (2010) 782–791.

[4] R.A. Alvarez-Puebla, L.M. Liz-Marzan, SERS detection of small inorganicmolecules and ions, Angew. Chem. Int. Ed. 51 (2012) 11214–11223.

[5] M.G. Albrecht, J.A. Creighton, Anomalously intense Raman-spectra of pyridineat a silver electrode, J. Am. Chem. Soc. 99 (1977) 5215–5217.

[6] H. Ko, S. Singamaneni, V.V. Tsukruk, Nanostructured surfaces and assembliesas SERS media, Small 4 (2008) 1576–1599.

[7] M. Sackmann, A. Materny, Surface enhanced Raman scattering (SERS) – aquantitative analytical tool?, J Raman Spectrosc. 37 (2006) 305–310.

[8] S. Kasera, L.O. Herrmann, J.d. Barrio, J.J. Baumberg, O.A. Scherman, Quantitativemultiplexing with nano-self-assemblies in SERS, Sci. Rep. 4 (2014) (articlenumber: 6785).

[9] D. Tsoutsi, J.M. Montenegro, F. Dommershausen, U. Koert, L.M. Liz-Marzán, W.J.Parak, R.A. Alvarez-Puebla, Quantitative surface-enhanced Raman scatteringultradetection of atomic inorganic ions: the case of chloride, ACS Nano 5(2011) 7539–7546.

[10] J. Lim, S.A. Majetich, Composite magnetic–plasmonic nanoparticles forbiomedicine: manipulation and imaging, Nano Today 8 (2013) 98–113.

[11] R. Ahijado-Guzman, P. Gomez-Puertas, R.A. Alvarez-Puebla, G. Rivas, L.M. Liz-Marzan, Surface-enhanced Raman scattering-based detection of theinteractions between the essential cell division FtsZ protein and bacterialmembrane elements, ACS Nano 6 (2012) 7514–7520.

[12] S. Abalde-Cela, B. Auguie, M. Fischlechner, W.T.S. Huck, R.A. Alvarez-Puebla, L.M. Liz-Marzan, C. Abell, Microdroplet fabrication of silver-agarosenanocomposite beads for SERS optical accumulation, Soft Matter 7 (2011)1321–1325.

[13] P. Aldeanueva-Potel, E. Faoucher, R.A. Alvarez-Puebla, L.M. Liz-Marzan, M.Brust, Recyclable molecular trapping and SERS detection in silver-loadedagarose gels with dynamic hot spots, Anal. Chem. 81 (2009) 9233–9238.

[14] S. Abalde-Cela, S. Ho, B. Rodríguez-González, M.A. Correa-Duarte, R.A. Álvarez-Puebla, L.M. Liz-Marzán, N.A. Kotov, Loading of exponentially grown LBL filmswith silver nanoparticles and their application to generalized SERS detection,Angew. Chem. Int. Edit. 48 (2009) 5326–5329.

[15] M. Yang, R. Alvarez-Puebla, H.-S. Kim, P. Aldeanueva-Potel, L.M. Liz-Marzán, N.A. Kotov, SERS-active gold lace nanoshells with built-in hotspots, Nano Lett. 10(2010) 4013–4019.

[16] J. Zhang, S. Xu, E. Kumacheva, Polymer microgels: reactors for semiconductor,metal, and magnetic nanoparticles, J. Am. Chem. Soc. 126 (2004) 7908–7914.

[17] A.A. Farah, R.A. Alvarez-Puebla, H. Fenniri, Chemically stable silvernanoparticle-crosslinked polymer microspheres, J. Coll. Interface Sci. 319(2008) 572–576.

134 A. Trojanowska et al. / Journal of Colloid and Interface Science 460 (2015) 128–134

[18] T. Von Werne, T.E. Patten, Preparation of structurally well-defined polymer-nanoparticle hybrids with controlled/living radical polymerizations, J. Am.Chem. Soc. 121 (1999) 7409–7410.

[19] M. Mulder, Membrane preparation | phase inversion membranes, in: I.D.Wilson (Ed.), Encyclopedia of Separation Science, Academic Press, Oxford,2000, pp. 3331–3346.

[20] C. Panisello, R. Garcia-Valls, Polysulfone/vanillin microcapsules productionbased on vapor-induced phase inversion precipitation, Ind. Eng. Chem. Res. 51(2012) 15509–15516.

[21] B. Peña, C. Panisello, G. Aresté, R. Garcia-Valls, T. Gumí, Preparation andcharacterization of polysulfone microcapsules for perfume release, Chem. Eng.Sci. 179 (2012) 394–403.

[22] C. Torras, L. Pitol-Filho, R. Garcia-Valls, Two methods for morphologicalcharacterization of internal microcapsule structures, J. Memb. Sci. 305 (2007)1–4.

[23] N. Pazos-Perez, F.J. Garcia de Abajo, A. Fery, R.A. Alvarez-Puebla, From nano tomicro: synthesis and optical properties of homogeneous spheroidal goldparticles and their superlattices, Langmuir 28 (2012) 8909–8914.

[24] M. Grzelczak, J. Perez-Juste, P. Mulvaney, L.M. Liz-Marzan, Shape control ingold nanoparticle synthesis, Chem. Soc. Rev. 37 (2008) 1783–1791.

[25] M. Moskovits, Rev. Mod. Phys. 57 (1985) 783–826.[26] D.-Y. Wu, X.-M. Liu, S. Duan, X. Xu, B. Ren, S.-H. Lin, Z.-Q. Tian, Chemical

enhancement effects in sers spectra: a quantum chemical study of pyridineinteracting with copper, silver, gold and platinum metals, J. Phys. Chem. C 112(2008) 4195–4204.

[27] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T.Siemieniewska, Reporting physisorption data for gas/solid systems withspecial reference to the determination of surface area and porosity(Recommendations 1984), Pure Appl. Chem. 57 (1985) (1984) 603–619.

[28] R.C. Bansal, M. Goyal, Activated Carbon Adsorption, CRC Press, 2010.[29] B. Peña, L.C. De Menorval, R. Garcia-Valls, T. Gumí, Characterization of

polysulfone and polysulfone/vanillin microcapsules by 1H NMRspectroscopy, solid-state 13C CP/MAS-NMR spectroscopy, and N2adsorption–desorption analyses, ACS Appl. Mater. Interfaces 3 (2011) 4420–4430.

[30] J. Landers, G.Y. Gor, A.V. Neimark, Density functional theory methods forcharacterization of porous materials, Colloid Surf. A – Physicochem. Eng. Asp.437 (2013) 3–32.

[31] B. Mir-Simon, I. Reche-Perez, L. Guerrini, N. Pazos-Perez, R.A. Alvarez-Puebla,Universal one-pot and scalable synthesis of SERS encoded nanoparticles,Chem. Mater. 27 (2015) 950–958.

[32] S.A. Gordeyev, G.Y. Nikolaeva, K.A. Prokhorov, R. Withnall, I.R. Dunkin, S.J.Shilton, P.P. Pashinin, Super-selective polysulfone hollow fibre membranes forgas separation: assessment of molecular orientation by Raman spectroscopy,Laser Phys. 11 (2001) 82–85.

[33] H.Y. Jung, Y.-K. Park, S. Park, S.K. Kim, Surface enhanced Raman scattering fromlayered assemblies of close-packed gold nanoparticles, Anal. Chim. Acta 602(2007) 236–243.