Effect of pH on stability and plasmonic properties of cysteine-functionalized silver nanoparticle dispersion

Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Effect of pH on stability and plasmonic properties of cysteine-functionalized silver nanoparticle dispersion Edit Csapó a , Rita Patakfalvi b , Viktória Hornok a , László Tamás Tóth a , Áron Sipos c , Anikó Szalai c , Mária Csete c , Imre Dékány a,d,∗ a

Supramolecular and Nanostructured Materials Research Group of the Hungarian Academy of Sciences, University of Szeged, Aradi vt. 1, 6720 Szeged, Hungary Centro Universitario de los Lagos, Universidad de Guadalajara, Enrique Díaz de León 1144, 47463 Lagos de Moreno, Mexico c Department of Optics and Quantum Electronics, University of Szeged, Dóm t. 9, 6720 Szeged, Hungary d Department of Medical Chemistry, Faculty of Medicine, University of Szeged, Aradi vt. 1, 6720 Szeged, Hungary b

a r t i c l e

i n f o

Article history: Received 7 October 2011 Received in revised form 10 February 2012 Accepted 28 March 2012 Available online 28 April 2012 Keywords: Biofunctionalized nanosilver Plasmonic properties pH effect

a b s t r a c t Citrate-stabilized spherical silver nanoparticles (Ag NPs) with d = 8.25 ± 1.25 nm diameter were prepared and functionalized with l-cysteine (Cys) in aqueous dispersion. The nanosilver–cysteine interactions have been investigated by Raman and 1 H NMR spectroscopy. The effect of pH on stability of biofunctionalized Ag NPs was investigated. The cysteine-capped nanosilver dispersions remain stable at higher pH (pH > 7), while the degree of aggregation increased as the pH decreased. Below pH ∼7, the characbareAgNP teristic surface plasmon band of bare silver nanoparticles was back-shifted from measured = 391 nm to 1measured = 387–391 nm, while the presence of a new band at 2measured = 550–600 nm was also observed depending on pH. Finite element method (FEM) was applied to numerically compute the absorption spectra of aqueous dispersions containing bare and cysteine-functionalized Ag NPs at different pH. Both the dynamic light scattering (DLS) measurements, Zeta potential values and the transmission electron microscopic (TEM) images confirmed our supposition. Namely, electrostatic interaction arose between the deprotonated carboxylate (COO− ) and protonated amino groups (NH3 + ) of the amino acid resulting in cross-linking network of the Ag NPs between pH ∼3 and 7. If the pH is measurable lower than ∼3, parallel with the protonation of citrate and l-cysteine molecules the connection of the particles via l-cysteine is partly decomposed resulting in decrease of second plasmon band intensity. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Due to the size- and shape-dependent optical properties of nanosize noble metal particles like gold or silver, the interaction of these nanoparticles with biomolecules (e.g. amino acids, peptides or proteins) has been intensively studied in the last 10 years [1–7]. The nanosize noble metal–biomolecule conjugates are applied as biosensors (e.g. colorimetric sensor) for detection of molecules [8–12] or metal ions [13]. Special bio-detection methods were developed for following the protein conformational changes and for DNA and protein recognition [14], but thiol-stabilized gold NPs are also used in catalysis [15]. A detailed review on localized surface plasmon resonance (LSPR) based biosensor has been recently published as well [16]. Numerous groups have focused their research on the investigation of the interaction between biomolecules and nanometal

∗ Corresponding author at: Aradi vt. 1, 6720 Szeged, Hungary. Tel.: +36 62 544210; fax: +36 62 544042. E-mail address: [email protected] (I. Dékány). 0927-7765/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2012.03.036

surface using versatile spectroscopic techniques. In case of gold or silver nanoparticles the different thiol group-containing molecules like alkyl-/arylthiols or l-cysteine are well-studied compounds. Generally, these molecules can interact with the nanometal surface via sulphur donor resulting in the formation of covalent Au S and Ag S bonds. Various spectroscopic data (UV–vis, Raman, 1 H NMR, FTIR, SERS spectroscopy, etc.) confirm the formation of cysteinecapped gold [17] or silver nanoconjugates [18] via sulphur–metal interaction. Namely, the disappearance of the S H stretching vibration of amino acid at ∼2550–2560 cm−1 in the infrared and Raman spectra [17,18] or the presence of the Au(or Ag) S vibrational bands between ∼210 and 270 cm−1 in the Raman, SERS or far-infrared spectra [19,20] suggests the binding/capturing of these molecules on metal surface. In addition to the formation of Au(or Ag) S covalent bond, binding of this ligand via amino-N or carboxylate-O donors is also observed [19,21–24]. In most cases the results of spectral measurements have also been supported by theoretical calculations. Numerous theoretical studies (e.g. DFT calculations) were published on cysteine gold/silver adsorption [19,25–27] but number of articles providing to the explanation of the appearance and shifts of plasmon bands due to the aggregation and/or coupling

44

E. Csapó et al. / Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49

of nanoparticles is in focus of interest as well [28,29]. Sweatlock et al. presented results about that linear arrays of Ag nanoparticles with 10 nm diameters and 0–4 nm spacing embedded in soda-lime glass exhibit splitting of the collective plasmon extinction band, and 1.5 eV red-shift of their longitudinal resonance was observed [28]. In arrays of strongly coupled particles local intensity enhancement on the order of 5000 might be reached, and the dipolar interaction exceeds to over 10 particles. The local intensity might be further enhanced by tailoring the particles size distribution inside linear aggregates, which is the plasmonic nano-focusing concept [30]. On the other part, it is also well-known that the colloid stability of these nanometal-biomolecule conjugates depends on numerous factors like electrolyte and ligand concentrations, temperature and pH. Among other research groups, Aryal et al. prepared different amino acid-stabilized gold nanoparticles and their stability was studied under physiological conditions [21]. Due to the increase of electrolyte concentration and decrease of pH, the aggregation of gold nanoparticles occurred. Water dispersible lysine-stabilized Au NPs were also prepared previously [23] and the binding of amino acid to the surface was studied by 1 H NMR. They found that the lysine binds to NPs via the ␣-amino group, while the terminal amino moiety forms hydrogen bonds with the carboxylate groups of surface-bound molecules on neighboring gold nanoparticles at higher pH. However, numerous research groups have investigated gold–biomolecule conjugates [31] but the number of the publications relating to the stability measurements of nanosilver-amino acids is somewhat less [18,19,32]. Research of Jing and Fang indicated that the adsorption behavior of l-cysteine on the surface of gold and silver colloid nanoparticles is different. While the molecules are interacted with gold surface via S-, N- and O-donors by SERS spectroscopy, only the carboxylate and amino groups were attached to silver surface. In paper written by Mandal et al. surface modification of aqueous colloidal silver nanoparticles with l-cysteine was studied [18]. The aging of the cysteine-capped silver dispersion led to the aggregation of the particles via formation of hydrogen bond between amino acids located on neighboring silver particles surfaces but the effect of pH on the stability of dispersion and the kinetic of aggregation process was not studied. In this work, the effect of pH on the stability of cysteinefunctionalized silver NPs-containing dispersions was studied. Moreover, the Zeta potential values of the functionalized silver surface were also determined at different pH values using DLS. The degree of aggregation was followed by TEM, DLS and UV–vis techniques. The binding of the molecule on silver surface was confirmed by 1 H NMR and Raman spectroscopic measurements. Moreover, both the appearance and shifts of plasmon bands were explained by theoretical calculations as well.

Milli-Q water were mixed. Finally, during continuous stirring freshly prepared NaBH4 aqueous solution (0.8 cm3 of 5 × 10−2 M) was dropped to the Ag+ -containing sample at room temperature. The effect of pH was investigated in the range of pH = 2.0–10.0. The pH of the Ag NPs dispersions was adjusted using 0.10 M HNO3 and NaOH solutions. The aggregation processes of the Ag NPs dispersion in the presence of l-cysteine (cAg = 1.8 × 10−4 mol/dm3 , ccyst = 5.7 × 10−6 mol/dm3 ) were followed by Ocean Optics USB2000 (Ocean Optics Ins., USA) diode array spectrophotometer in the  = 200–850 nm range using 1 cm quartz cuvette. In all cases individual samples were prepared at different pH values and the reaction mixture was stirred with a micro-magnetic stirrer. TEM images were recorded on a Philips CM-10 instrument at 100 kV accelerating voltage. The microscope was equipped with a Megaview II digital camera and Formwar-coated copper grids were used for sample preparation. The size distribution of the particles was calculated by using UTHSCSA Image Tool 2.00 software. The Zeta potential values and the particles size distribution of the dispersions were determined between pH = 2.0 and 10.0 by using a Zetasizer Nano ZS ZEN 4003 apparatus (Malvern Ins., UK). Since the Zeta potential studies were performed in aqueous dispersion, the Smoluchowski approximation was used to calculate the potential values with the aid of the electrophoretic mobility. 1 H NMR studies for l-cysteine and cysteine-functionalized Ag NPs dispersions have been performed on Bruker Avance 500 instrument at 298 K. The spectra were acquired with the WATERGATE solvent suppression pulse scheme. Ag NPs to amino acid ratio was 1:5 and the analytical concentration of Cys was 1 × 10−3 M in all samples. The spectra were recorded between pH = 3 and 10. Raman spectra of the Ag-Cys nanopowder were recorded in the range of 3500–100 cm−1 on a Thermo Scientific DXR Dispersive Raman Microscope equipped with 780 nm frequency-stabilized single mode diode lasers and a cooled CCD camera. The applied laser power was 1 mW. In order to evaluate the measured spectra the OMNIC Atl␮sTM software was used. 2.3. Theoretical calculations by FEM Besides the spectral characterization, theoretical calculations were also performed to numerically compute the absorption spectra of different Ag NPs-containing dispersions by finite element method (FEM) using the RF module of COMSOL software (version 4.1). The purpose was to explain the experimental observation of bare Ag NP a single peak at measured = 391 nm for citrate-stabilized Ag NPs dispersed in water with concentration cAg = 1.8 × 10−4 M, and the splitting of the spectrum resulting in two maxima at 1measured = 387–391 nm and at 2measured ∼550–600 nm depending on pH (at 2,pH=2.98

During the synthesis the following materials were used: silver nitrate (AgNO3 , Reanal, 99.9%), sodium borohydride (NaBH4 , 98%, Sigma), sodium citrate dihydrate (Aldrich, 99%), l-cysteine (Reanal), sodium hydroxide (NaOH, Reanal) and nitric acid (HNO3 , Reanal).

e.g. pH = 2.98, measured = 567 nm), when the surface of Ag NPs was functionalized with l-cysteine having a concentration of ccys = 5.7 × 10−6 M. In the course of FEM calculation the average diameter of nanoparticles determined by TEM was used, Cysthickness of 0.45 nm was taken into account [33], while the pH dependence of the refractive indices were neglected. The wavelength dependent refractive index of water was determined with Sellmeier-equations [34], while the index of refraction in Cys shell was taken into by including a Cauchy formula determined by ellipsometry [35,36]. The complex dielectric parameter of silver was included based on the spline-fit of values from Palik-database [37].

2.2. Preparation and characterization of nanoparticles

3. Results and discussions

The citrate-stabilized Ag NPs were prepared in aqueous dispersion by sodium borohydride reduction using Ag+ :citrate = 1:5 molar ratio. Namely, 10.0 cm3 of silver nitrate solution (c = 4 × 10−4 M), 0.5 cm3 of sodium citrate solution (c = 4 × 10−2 M) and 8.7 cm3

Numerous Ag NPs-containing dispersions have been synthesized in our lab [38,39] and also in others’ [32]. In this article the well-known citrate-stabilized Ag NPs were investigated. The average particle diameter was d = 8.25 ± 1.25 nm determined by TEM

2. Materials and methods 2.1. Materials

E. Csapó et al. / Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49

Fig. 1.

1

45

H NMR spectra of the l-cysteine (a) and Ag-cysteine 1:5 systems (b) at different pH values (cAg = 1.5 × 10−4 M, ccys = 7.5 × 10−4 M).

images, while the DLS measurements indicate the presence of NPs with d = 7.85 ± 1.23 nm. The formation of citrate-stabilized Ag NPs was followed by UV–vis spectroscopy [40]. The observed absorpbare Ag NP tance maximum was at measured = 391 nm after the end of the reduction. The prepared citrate-stabilized Ag nanoparticles have negatively charged surface determined by Zeta potential measurement (at pH = 7.00,  = −42.53 mV). Due to the decrease of pH, the Zeta potential values of the silver nanoparticles were not changed significantly at pH = 3.02 the  = −35.80 mV. Most probably the citrate molecules cannot be binded to the surface of Ag NPs via chemical bond(s), only negligible electrostatic interaction may be existed. First the chemical interaction between amino acid and Ag NPs surface has been investigated by laser Raman and NMR spectroscopy. The infrared (FTIR) spectroscopy is only partly suitable to confirm the binding of l-cysteine on silver surface, but the Raman spectroscopy is more specific for the direct confirmation of Ag S covalent bond.1 The results of the Raman studies are summarized in Fig. S1. Comparison of the Raman spectra of the l-cysteine (black line (1)) and Cys-functionalized Ag NPs dispersion (gray line (2)) we found that the S H stretching vibration of amino acid at 2550 cm−1 were disappeared in the cysteine-containing sample, moreover the appearance of a new broad and asymmetric band was also observed between ∼210 and 245 cm−1 in the above mentioned sample. According to the data in the literature, the characteristic vibrational band of Ag S corresponds to this new band at ∼210–245 cm−1 , but the Ag O and Ag N vibrations are also located in this wavenumber region [29,41,42]. Besides the Raman measurements 1 H NMR technique has also been applied to investigate the interaction between the amino acid and nanosilver surface in aqueous dispersion. The results of 1 H NMR measurements are summarized in Fig. 1. Fig. 1(a) shows the 1 H NMR spectra of l-cysteine, while Fig. 1(b) represents the spectra recorded for Ag:Cys 1:5 system at different pH values. The chemical shifts of the CH2 (denoted with B) protons are assumed to be sensitive to the binding of the molecule via sulphur donor onto silver surface, while the CH proton (denoted with A) of the amino acid is expected to sense the electrostatic interaction of ammonium moiety with the silver surface.2 Comparison of the chemical shifts of the “A” and “B” protons of the l-cysteine in the absence

1 In the infrared spectrum only the disappearance of the S H stretching vibration of the molecule at ∼2550–2560 cm−1 can support the binding of amino acid. 2 In Fig. 1(b) “C” corresponds to the CH2 protons of citrate molecules.

and in the presence of Ag NPs the results suggested that the electron density changes around both the S- and N-donors of the ligand in the Ag NPs-containing sample. The presence of the silver NPs around the soft donors can create inhomogeneities in the magnetic field and thus the local chemical environment has changed [17,21]. Most probably both S- and N-donors sense the presence of metal surface at applied conditions but more exact conclusion is not established using only 1 H NMR. Application of e.g. diffusion NMR measurements may provide more useful information about the nanoparticle–biomolecule interaction. Theoretical calculations (DFT) were also started to study the alignment of l-cysteine on silver surface. The results of primary calculations correlate with the experimental findings of Raman and 1 H NMR studies. Namely, some results indicate the binding of both nitrogen and sulphur donors of l-cysteine to silver surface in different geometry. The detailed results are summarized in additional publication [43]. Before application of functionalized aqueous gold or silver nanodispersions in biological systems many important properties (e.g. effect of pH and salt concentration, surface charge, etc.) must be known. In this work the effect of pH on the colloid stability of the cysteine-functionalized Ag NPs dispersion was proved by UV–vis spectra registered at various pH values. It is well-known that the position and absorbance maximum of the surface plasmon band of noble metal nanoparticles depends on the shape, size, composition and aggregation state of the particles assemblies [5,22]. Namely, besides the characteristic surface plasmon band of spherbare Ag NP ical Ag NPs at measured = 391 nm, a new band, which appears at higher wavelength, always corresponds to the formation of larger and/or ellipsoidal and/or aggregated particles in the system but this observation also may indicate reduced interparticle distance [22,41]. In the spectral measurements Ag:Cys 31.5:1 molar ratio was used which corresponds to the monomolecular coverage of silver nanoparticles. For calculation of the monomolecular coverage of nanoparticles (nsm ) Eq. (1) was used. nsm = 6 × 103 / × d × Am × NA [mol/g]

(1)

The  is the density of silver at room temperature ( = 10.49 g/cm3 ), d is the average diameter of nanoparticles by TEM, Am is the area per one l-cysteine molecule (Am = 0.3 nm2 /molecule) measuring the adsorption isotherm on gold surface by QCM, while NA is the Avogadro’s number. Considering the average particle diameter d = 8.25 nm, the monomolecular coverage of nanoparticles is nsm = 3.85 × 10−4 mol/g. The recorded spectra at different pH values under cAg = 1.8 × 10−4 mol/dm3 and ccyst = 5.7 × 10−6 mol/dm3 are seen

46

E. Csapó et al. / Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49

Fig. 2. (A) Representative UV−vis spectra recorded for Ag-Cys system at different pH values (t = 2 min, cAg = 1.8 × 10−4 M, ccys = 5.7 × 10−6 M). The pH values are the follows: 9.63 (a), 7.05 (b), 5.70 (c), 4.92 (d), 2.98 (e), 2.36 (f). (B) The measured (continuous line) and calculated (dashed line). UV−vis absorbance spectra of citrate−stabilized Ag and Cys−functionalized Ag at pH = 2.98.

in Fig. 2(A). As it can be seen in Fig. 2(A) at pH = 9.63 (a) and pH = 7.05 (b) the appearance of the second plasmon band at higher wavelength is not observed, which confirms that above pH > 7, where the ammonium moiety starts to loose its dissociable proton [44], the cross-linking of the NPs (formation of aggregates) is not occurred. Most probably, the electrostatic repulsions between the negatively charged COO− groups of the surface-binded lcysteine molecules are existed and the formation of aggregates is hindered. In contrast with it, the second surface plasmon band results in a separated peak is dominant at pH = 5.70 (c) 2,pH=5.7 2,pH=4.92 (measured = 548 nm) and at pH = 4.92 (d) (measured = 588 nm). Fig. 2(A) clearly represents that further decrease in pH causes that the degree of the aggregation is considerably significantly 2,pH=2.36 below pH ∼3 (f) (measured = 515 nm) than in the case of pH = 4.90 or pH = 5.70. Our experimental results revealed that both above pH > 7 and significantly below pH < 3 the cross-linking of the Ag NPs via l-cysteine is not predominant due to the deprotonation of ammonium moiety and protonation of the carboxylate group, respectively [44]. Finite element method was used to explain the observed plasmon bands. First numerical calculations were performed on stand-alone bare Ag NPs, and the calculated spectrum (dashed gray line) is seen in Fig. 2(B). Comparison of the measured spectra bare Ag NP (Fig. 2(B) gray line, measured = 391 nm) with FEM curves (Fig. 2(B) bare Ag NP

dashed gray line, calculated = 398 nm) indicates that the calculated absorptance curve is in good agreement with the measured

spectrum. According to literature the absorbance maximum of spherical and spheroidal metal nanoparticles is determined by the εmetal /εdielectric ratio of the dielectric parameters’ real part inside the metallic NPs and in the surrounding dielectric materials and it is influenced by the nanoparticle shape too [45]. The εAg,real /εwater (398 nm) ≈ −2 ratio at the absorbance maximum determined by FEM is in accordance with the theoretical prediction based on spherical particles. Similarly to bare Ag NPs the registered plasmon bands of functionalized Ag dispersion were also compared with a numerically calculated spectrum at one representative pH value (pH = 2.98), which is also seen in Fig. 2(B). Based on the results of previously published articles [28,30] and also taking into account the simplest geometry of aggregates, during calculations different linear chains were built up from N = 3, 11, 17, 33, 63, etc. (N = number of particles) from Ag NPs surrounded by Cys shell. Finally the gap between the Ag NPs composing a linear chain was varied to fit maxima observed experimentally, while the thickness of Cys shell (0.45 nm) and g (0.6 nm distance) was the same. Our conclusion is that linear chains built up from N = 17 pieces of Ag nanoparticles, surrounded by 0.45 nm Cys shell, and arrayed at g = 0.6 nm distance (g refers to interparticle gap between 2,pH=2.98 nanoparticles) result in a secondary peak at calculated = 567 nm, which is in good agreement with the secondary peak measured at pH = 2.98. The full width half maximum (FWHM) of the theoretical curve is considerably smaller then the FWHM of the measured curves. The difference might be explained by the co-existence of

Fig. 3. The E-field distribution on single 8.25 nm Ag NP (a) and along the array of N = 17 (b) Ag NPs covered by 0.45 nm Cys shell, arrayed at g = 0.6 nm distance at 402 and 567 nm.

E. Csapó et al. / Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49

47

Table 1 Zeta potential values and average particle diameter of l-cysteine-functionalized Ag NPs aqueous dispersion with PDI at different pH (cAg = 1.8 × 10−4 M, ccys = 5.7 × 10−6 M). pH

 (mV)

2.36 2.98 4.08 4.92 5.70 7.05 9.63

31.43 24.47 0.02 −7.56 −20.70 −28.87 −30.37

d2 min (nm) ± ± ± ± ± ± ±

7.0 6.0 4.1 4.1 3.1 3.2 3.0

32.9 19.4 420.5 523 431.7 7.0 6.2

± ± ± ± ± ± ±

5.4 4.9 95 120 119 1.1 0.8

aggregates with different size parameters and spatial alignment. Fig. 3(a) represents the E-field distribution on single 8.25 nm Ag NPs bare Ag NP at calculated = 398 nm. Moreover, Fig. 3(b) shows coupled dipolar oscillations on the E-field distributions along the array of N = 17 Ag NPs covered by 0.45 nm Cys shell, arrayed at g = 0.6 nm distance 2,pH=2.98 1,pH=2.98 at calculated and calculated as well. Further calculations were also performed to numerically compute the registered absorbance spec2,pH=4.92 2,pH=5.70 tra at pH = 4.92 (measured = 588 nm) and at pH = 5.70 (measured = 548 nm) as well, the results are summarized in our previous article [46]. In order to confirm the results of both spectral studies and theoretical calculations, DLS measurements were also performed. In addition, the Zeta potential values of the functionalized silver NPs were determined as well. The measured Zeta potentials and the average particle diameters by DLS are summarized in Table 1. Fig. 4 demonstrates the changes of the Zeta potential values as a

PDI2 min

d75 min (nm)

0.23 0.24 0.32 0.34 0.31 0.23 0.19

410.3 241.4 1100.0 1316.8 1281.5 6.0 7.0

± ± ± ± ± ± ±

61 25 147 215 150 0.6 1.1

PDI75 min 0.33 0.37 0.40 0.45 0.43 0.19 0.25

function of pH. As it can be seen the isoelectric point, where the plot passes through the zero Zeta potential, was found at pH ∼4.10. At this pH the functionalized aqueous dispersion is the least stabile. These measurements are in good agreement with the DLS data and the spectroscopical results as well. Namely, above pH > 7, where the Zeta values do not change measurable ( ∼−30 mV) the second plasmon band was not observed (Fig. 2(A)) and the average particle diameter values do not indicate the formation of larger aggregates. Between pH ∼3.0 and 7.0 where the cross-linking of nanoparticles via cysteine is particularly occurred, the dispersion is substantially instable. Besides these experimental findings the particle size distribution values also suggest that the degree of aggregation is the greatest in the above described pH range (Fig. 4). However, the second plasmon band with lower intensity is appeared significantly 2,pH=2.36 below pH ∼3.0 (e.g. at pH = 2.36 meaured = 515 nm) and the system reaches the  ∼+30 mV, but the DLS data in Table 1 confirm

Fig. 4. Zeta potential values and average particle diameter (by DLS) of l-cysteine-functionalized Ag NPs dispersion as a function of pH (cAg = 1.8 × 10−4 M, ccys = 5.7 × 10−6 M).

Fig. 5. TEM image for citrate-stabilized Ag NPs at pH 6.10 (a) and for l-cysteine-modified Ag NPs at pH = 4.92 (b).

48

E. Csapó et al. / Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49

Fig. 6. Schematic representation of the structure of cysteine-functionalized Ag nanodispersion at pH > 7 (a), pH = 3–7 (b) and pH < 3 (c).

that the average particle size is relatively less than in the pH range 3.0–7.0. TEM images of nanosilver dispersion were also recorded before (Fig. 5(b)) and after (Fig. 5(c)) addition of l-cysteine. As it can be seen, before addition of amino acid the Ag dispersion is stabile, the average particle diameter is d = 8.25 ± 1.25 nm. In contrast with it, Fig. 5(c) clearly represents the formation of chain-like structure of the cysteine-functionalized system at pH = 4.92. Based on the experimental results and theoretical calculations, Fig. 6 demonstrates one possible schematic representation of the pH dependence cross-linking of nanoparticles. Above pH ∼7.0, the ammonium groups start to loose their dissociable protons, the molecules have negative charge. The electrostatic interaction between donor groups (NH2 /COO− ) is not formed, the interparticle distances are also higher and the dispersion remains stable (Fig. 6(a)). Between the negatively charged carboxylate groups of Cys electrostatic repulsion will be existed. Between pH ∼3.0 and 7.0, most of the Cys molecules are in zwitterionic form. Most probably electrostatic interactions exist between the deprotonated carboxylate and the protonated amino groups of the surface binded molecules. This interaction facilitates the cross-linking of the particles and chain-like network is mainly dominant (Fig. 6(b)). Numerical calculations relating to the explanation of plasmon bands by aggregation also confirm the chain-like structure of the aggregates. Finally, significantly below pH ∼3, the above mentioned dominant electrostatic attractive interactions are decomposed due to the protonation of carboxylate, but in time larger aggregates are also appeared at very acidic pH (Fig. 6(c)).

4. Conclusions We have demonstrated the effect of pH on the stability and also the plasmonic properties of cysteine-functionalized aqueous nanosilver dispersion. Depending on pH different aggregation state of the studied bio-nano system was observed. The chain-like structure of the aggregates between pH = 3 and 7 is confirmed by TEM and theoretical calculations as well. The aggregated Ag NP structures determined by DLS and also TEM are in good agreement with the appearance and shifts of plasmon bands. Parallel with the increase of aggregates the shift of second plasmon bands toward higher wavelength is occurred.

Acknowledgements This work was supported by the Hungarian Scientific Research Fund (OTKA K73307 and CNK 78549) and also CONACyT SNI2008/90534 financial support.

The Project named “TÁMOP-4.2.1/B-09/1/KONV-2010-0005Creating the Center of Excellence at the University of Szeged” is supported by the European Union and co-financed by the European Regional Fund. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.colsurfb.2012.03.036. References [1] Y. Tan, Y. Li, D. Zhu, in: H.S. Nalva (Ed.), Encyclopedia of Nanoscience and Nanotechnology, vol. 8, American Scientific Publisher, California, 2004, pp. 9–40. [2] R. Sardar, A.M. Funston, P. Mulvaney, R.W. Murray, Langmuir 25 (2009) 13840. [3] S.K. Ghosh, T. Pal, Chem. Rev. 107 (2007) 4797. [4] V.H. Perez-Luna, K. Aslan, P. Betala, in: H.S. Nalva (Ed.), Encyclopedia of Nanoscience and Nanotechnology, vol. 2, American Scientific Publisher, California, 2004, pp. 27–49. [5] L.M. Liz-Marzán, Langmuir 22 (2006) 32. [6] A. Majzik, L. Fülöp, E. Csapó, F. Bogár, T. Martinek, B. Penke, G. Bíró, I. Dékány, Colloids Surf. B 81 (2010) 235. [7] M. Iosin, F. Toderas, P.L. Baldeck, S. Astilean, J. Mol. Struct. 924–926 (2009) 196. [8] H. Huang, C. Tang, Y. Zeng, X. Yu, B. Liao, X. Xia, P. Yi, P.K. Chu, Colloids Surf. B 71 (2009) 96. [9] F.X. Zhang, L. Han, L.B. Israel, J.G. Daras, M.M. Maye, N.K. Ly, C.J. Zhong, Analyst 127 (2002) 462. [10] P.K. Sudeep, S.T. Shibu Joseph, K. George Thomas, J. Am. Chem. Soc. 127 (2005) 6516. [11] Z.P. Li, X.R. Duan, C.H. Liu, B.A. Du, Anal. Biochem. 351 (2006) 18. [12] H. Li, F. Li, C. Han, Z. Cui, G. Xie, A. Zhang, Sens. Actuators B 145 (2010) 194. [13] H. Li, Z. Cui, C. Han, Sens. Actuators B 143 (2009) 87. [14] S. Chah, M.R. Hammond, R.N. Zare, Chem. Biol. 12 (2005) 323. [15] J.J. Pietron, R.W. Murray, J. Phys. Chem. B 103 (1999) 4440. [16] B. Sepúlveda, P.C. Angelomé, L.M. Lechuga, L.M. Liz-Marzán, Nano Today 4 (2009) 244. [17] S. Aryal, B.K.C. Remant, N. Dharmaraj, N. Bhattarai, C.H. Kim, H.Y. Kim, Spectrochim. Acta 63 (2006) 160. [18] S. Mandal, A. Gole, N. Lala, R. Gonnade, V. Ganvir, M. Sastry, Langmuir 17 (2001) 6262. [19] C. Jing, Y. Fang, Chem. Phys. 332 (2007) 27. [20] J.K. Lim, S.-W. Joo, J. Electroanal. Chem. 605 (2007) 68. [21] S. Aryal, K.C. Remant Bahadur, N. Bhattarai, C.K. Kim, H.Y. Kim, J. Colloid Interface Sci. 299 (2006) 191. [22] Z. Zhong, S. Patskovskyy, P. Bouvrette, J.H.T. Luong, A. Gedanken, J. Phys. Chem. B 108 (2004) 4046. [23] P.R. Selvakannan, S. Mandal, S. Phadtare, R. Pasricha, M. Sastry, Langmuir 19 (2003) 3545. [24] S. Stewart, P.M. Fredericks, Spectrochim. Acta A 55 (1999) 1641. [25] R.D. Felice, A. Selloni, E. Molinari, J. Phys. Chem. B 107 (2003) 1151. [26] C. Gautier, T. Bürgi, J. Am. Chem. Soc. 128 (2006) 11079. [27] B. Morzyk-Ociepa, D. Michalska, Spectrochim. Acta A 55 (1999) 2671. [28] L.A. Sweatlock, S.A. Maier, H.A. Atwater, J.J. Penninkhof, A. Polman, Phys. Rev. B 71 (2005) 235408. [29] M. Quinten, U. Kreibig, Appl. Opt. 32 (1993) 6173. [30] W.-Y. Chien, T. Szkopek, Opt. Express 16 (3) (2008) 1820. [31] A. Majzik, R. Patakfalvi, V. Hornok, I. Dékány, Gold Bull. 42 (2009) 113. [32] Z. Khan, A. Talib, Colloids Surf. B 76 (2010) 164.

E. Csapó et al. / Colloids and Surfaces B: Biointerfaces 98 (2012) 43–49 ˝ [33] L. Korösi, S. Papp, I. Dékány, Chem. Mater. 22 (2010) 4356. [34] M. Daimon, A. Masamura, Appl. Opt. 46 (2007) 3811. [35] S. Kurunczi, A. Németh, T. Hülber, P. Kozma, P. Petrik, H. Jankovics, A. Sebestyén, F. Vonderviszt, M. Fried, I. Bársony, Appl. Surf. Sci. (2010) 319. [36] H. Arwin, Appl. Spectrosc. 40 (1986) 313. [37] E.D. Palik, Handbook of Optical Constants of Solids, Academic Press, San Diego, 1998. [38] E. Pintér, R. Patakfalvi, T. Fülei, Z. Gingl, I. Dékány, C. Visy, J. Phys. Chem. B 109 (2005) 17474. [39] R. Patakfalvi, A. Oszkó, I. Dékány, Colloids Surf. A 220 (2003) 45.

[40] [41] [42] [43] [44] [45]

49

R. Patakfalvi, S. Papp, I. Dékány, J. Nanopart. Res. 9 (2007) 353. H.I.S. Nogueira, Spectrochim. Acta A 54 (1998) 1461. G.-N. Xiao, S.-Q. Man, Chem. Phys. Lett. 447 (2007) 305. T. Körtvélyesi, O. Berkesi, E.A. Hoffman, I. Dékány, unpublished results. I. Sóvágó, T. Kiss, K. Várnagy, B. Decock-le Révérend, Polyhedron 7 (1988) 1089. J.A. Schuller, E.S. Barnard, W. Cai, Y. Jun, J.S. White, M. Brongersma, Nat. Mater. 9 (2010) 193. [46] A. Szalai, Á. Sipos, E. Csapó, V. Hornok, L.T. Tóth, M. Csete, I. Dékány, in: M.I. Stockman (Ed.), Plasmonics: Metallic Nanostructures and Their Optical Properties IX, vol. 80963B, 2011 (Proceedings Vol. 8096).