- Email: [email protected]
Refractive index measurement of nanoparticles by immersion refractometry based on a surface plasmon resonance sensor
Chemical Physics Letters 654 (2016) 72–75
Contents lists available at ScienceDirect
Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett
Research paper
Refractive index measurement of nanoparticles by immersion refractometry based on a surface plasmon resonance sensor Hiroshi Kano a, Ayumu Iseda a, Katja Ohenoja b, Ilpo Niskanen c,d,⇑ a
Department of Information and Electronic Engineering, Muroran Institute of Technology, Mizumoto 27-1, Muroran, Hokkaido 050-8585, Japan Faculty of Technology, Fibre and Particle Engineering Group, University of Oulu, P.O. Box 4500, FI-90014 Oulu, Finland c Faculty of Technology, University of Oulu, P.O. Box 4500, FI-900014 Oulu, Finland d National Institute of Industrial Science and Technology, Namiki 1-2, Tsukuba, Ibaraki 305-8564, Japan b
a r t i c l e
i n f o
Article history: Received 28 January 2016 In final form 7 May 2016 Available online 9 May 2016 Keywords: Nanoparticles Refractive index Surface plasmon resonance technique
a b s t r a c t Accurate determination of the refractive index of nanoparticles has important ramifications for applications, such as pharmaceuticals, cosmetics, paints, textiles, and inks. We describe a new method to determine the refractive index of nanoparticles by immersion refractometry with a surface plasmon resonance sensor. With this method, the refractive index of the nanoparticles is perfectly matched with that of the surrounding liquid. We demonstrate this method for calcium fluoride nanoparticles that have an average diameter of 100 nm; the results achieve an accuracy of better than 0.002 refractive index units. Ó 2016 Published by Elsevier B.V.
1. Introduction Nanoparticles are one of the most-researched materials today due to their range of potential applications in biomedical, optical, and electronic fields. The refractive index of nanoparticles is one of the most important physical parameters used to inspect the quality of particle-based solids. Materials technology strives to tune the effective refractive index of nanoparticles to manipulate their matrix properties at a molecular level [1]. Tuning the optical, magnetic, and electrical properties of refractive index can be tailored nanoparticles for specific applications, thus increasing their effectiveness. It is difficult to precisely measure the refractive index of nanoparticles with current techniques [2]. In general, the refractive index is not only related to the chemical composition, crystal structure, and symmetry of each particle but is also affected by the temperature, wavelength of incident light and internal stresses in the sample [3]. Several techniques have been developed to measure the refractive index of nanoparticles. Total internal reflection has been used to determine the refractive index of ZnO, TiO2, and ZrO2 nanoparticles; this technique is based on measuring the critical angle of total internal reflection [4]. The refractive index is measured with the experimental uncertainty of 1%. The critical-angle method has long been applied in the Abbe refractometer; the advantages ⇑ Corresponding author at: Faculty of Technology, University of Oulu, P.O. Box 4500, FI-900014 Oulu, Finland. E-mail address: [email protected] (I. Niskanen). http://dx.doi.org/10.1016/j.cplett.2016.05.013 0009-2614/Ó 2016 Published by Elsevier B.V.
include easy experimental implementation and a simple calibration [5]. The critical-angle method is precise to about ±0.002 refractive index units, RIU. Turbid liquids contain a high concentration of solid particles that scatter light toward the detector, combining with the specular reflection to cause an error [6]. Ellipsometry has been used to measure the refractive index of gold and TiO2 nanoparticles; it is based on the change in the polarization state of light reflected from the surface of a sample [7,8]. Interferometry has been used to determine the refractive index of CdSe nanoparticles [9]. Interferometry and ellipsometry are accurate techniques (0.0001–0.00001 RIU), but they are expensive and require samples with very smooth surfaces [10]. Optical methods and electron microscopy have been used to determine the size distribution and complex refractive index of magnetite (Fe3O4) nanoparticles [11]. Immersion matching has been used to measure the refractive indices of various types of nanoparticles [12,13]. The principle of the immersion method is to match the refractive index of solid particles with that of the immersion liquid. In the event of a perfect match between the index of the non-absorbing pigments and the immersion liquid, the transmittance will be 100% and the scattering intensity will approach zero. The advantages of immersion matching are that the technique is independent of the shape or size of a particle, and the measurement is easy and relatively fast to perform [14]. The refractive index of an immersion liquid is generally in the range of 1.4–1.8 [15]. As to the accuracy of the immersion methods, various authors assign it an error of 0.001–0.005 as usually performed. Dynamic holography techniques have been used to measure the refractive index of
H. Kano et al. / Chemical Physics Letters 654 (2016) 72–75
carbon nanoparticles [16]. In-line holographic microscopy of micrometer-scale techniques have been measured refractive index of silica particles with a resolution of 0.002 RIU [17]. Nanoparticle tracking analysis (NTA) is a relatively new technique for determining the refractive index of suspended nanoparticles based on the rate of their Brownian motion [18,19] which an uncertainty in the range is 0.021–0.046 RIU. NTA currently works best for particles having a diameter of approximately 10–1000 nm. The lower detection limit depends on the refractive index of the nanoparticles; characterization is only possible for nanoparticles composed of materials with a high refractive index, such as gold or silver. The upper detection limit is due to the limited Brownian motion of large nanoparticles. The rate of particle diffusion is related to the viscosity and temperature of the liquid [20,21]. Surface plasmon resonance (SPR) sensors can also perform refractive index measurements [22]. In this technique, surface plasmons (SPs), the quanta of collective electron oscillations excited on a metallic surface, are used as a measurement probe because the propagating constant of the SPs strongly depends on the refractive index on the metallic surface. In the most typical setup, p-polarized incident light is given to a metallic thin film through a prism coupler. The light having a proper incident angle resonates with the SPs that consume light energy. Therefore, the resonance angle found in angular dependency of reflectance tells the refractive index on the metal surface. The electric field produced by the SPs shows an evanescent decaying property and remarkable field enhancement against incident light. These properties provide surface sensitive and high-sensitive measurements. The theoretical resolution in the refractive index measurement with a typical setup reaches 5 107 RIU. The high sensitivity contributes to detection of an adsorbed monomolecular layer on the metallic surface [23]. A variation of the SPR sensor enables a microscopic measurement while maintaining the high resolution in the refractive index measurement [24,25]. In this case, an excitation beam is tightly focused onto a metallic thin film coated on a glass substrate by using an oil immersion objective lens with high numerical aperture; interference of excited SPs propagating in many directions localize the SPs in the optical diffraction limit region. In the measurement of nanoparticle, the localized SPs are useful for probing a region with high particle density. Furthermore, the localized-SPR sensor can be combined with the immersion technique while retaining the advantages described above, viz., refractive index measurements that are independent of the shape and size of particles. In this case, we examine a condition showing perfect index matching between the nanoparticles and the immersion liquid by using localized-SPR sensor. The aim of this study is to get the refractive index of a nanoparticle by immersion liquid method based on a localized-SPR sensor. The high sensitivity of the localized-SPR sensor enhances index mismatch between the sample and the immersion liquid, resulting in higher precision.
73
1.2. Grinding proceeded with smaller beads (320 lm for 150 min), resulting in a median particle size of 104 nm and WPSD = 1.06. Further grinding broadened the particle size distribution, indicating that the first two steps had achieved the grinding limit [27,28]. Samples processed by the first two steps were used in this study. The Field Emission Scanning Electron Microscope (FESEM) image taken with a Zeiss Ultra Plus is presented in Fig. 1, along with the particle size distribution of a ground CaF2 suspension measured by laser diffraction. Fig. 2 shows a schematic of the localized-SPR sensor. The beam width of linearly polarized light with a wavelength of 632.8 nm was enlarged by a beam expander; the polarization was converted from linear to radial by using a radial polarization converter (ARCoptix, Switzerland) to obtain the optimal localization of SPs [25]. The position of the converter is imaged to the entrance pupil of an oil immersion objective lens with a numerical aperture of 1.65 (Olympus, Japan). The light converged from the objective lens and illuminated a coverslip coated with a metallic thin film. The reflected light from the coverslip was relayed to an image sensor located at an optically conjugate plane of the exit pupil of the objective lens. The image sensor recorded the spatial frequency distribution of the reflected light. A typical measurement recorded by the image sensor is shown in Fig. 3. An absorption ring appeared in the image; its radius is used to calculate the propagating constant of the surface plasmons. The absorption pattern is fit by computational processing to quantify the radius and coordinates of the ring. The radius that corresponds to the real part of the propagating constant of excited SPs can be approximately expressed by the following equation [22,25,29]:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi! n2 n2 qsp ¼ Reðksp Þ ffi Re k0 2 m s 2 nm þ ns
ð1Þ
2. Materials and methods We examined this technique by measuring the refractive index of prepared nanoparticles. To produce the nanoparticles, CaF2 powder (Merck, 97%) was purchased; it contained trace amounts of Fe, Mg, Na, and SiO4, according to the product description. The median particle size (on a volumetric basis) was 7.0 lm, measured by a Beckman Coulter LS 13 320 laser diffraction analytical device. Nanogrinding was carried out for the CaF2 powder via a two-step process with a stirred media mill (90 AHM hydro mill, Hosokawa Alpine). The operation of the mill has been previously described in detail [26]. After the first grinding (370 min with 560 lm Zirconia (YSZ) beads), the median particle size was 110 nm and the width of the particle size distribution WPSD = ((d90 d10)/d50) was
Fig. 1. (a) FESEM image and (b) particle size distribution of ground CaF2 nanoparticles. The scale bar in the FESEM image represents 300 nm.
74
H. Kano et al. / Chemical Physics Letters 654 (2016) 72–75
best-fitting ring in pixel units to spatial frequency) by measuring the refractive index of air, which is 1.000. The samples were measured at room temperature. After finding the coordinate center, we can obtain the relationship between the reflected intensity and the spatial frequency of light. We include the plot in Fig. 3 for readers’ information. The position of absorption dip varies by the effective index on the metal surface. 3. Results and discussion
Fig. 2. Schematic of the localized-SPR sensor. Radially polarized light is focused onto the metallic thin film by the oil immersion objective lens. Surface plasmons excited in the diffraction-limited-region (180 nm) [28] interact with a sample (nanoparticles and immersion liquid). The spatial-frequency distribution of reflected light is recorded by an image sensor. The propagating constant is determined from the excited surface plasmons, enabling calculation of the effective refractive index on the metal surface.
We used vacuum evaporation to prepare substrates coated by silver with a thickness of 54 nm. The substrates in each batch were secured onto the rotating stage in the evaporator at the same distance from the rotation center to ensure uniform quality. We also prepared glycerol (Prolabo Company) solutions diluted with MilliQ water to concentrations of 44, 54, 65, and 100 wt% glycerol. We drop cast these glycerol solutions onto substrates to prepare samples without nanoparticles. Samples with nanoparticles were prepared by drop casting CaF2 nanoparticles dispersed in ethanol onto substrates; the substrates were dried in air and different concentrations of glycerol solutions were drop cast on top. Two separate evaporation batches were used to produce (1) the substrates without nanoparticles and (2) substrates with nanoparticles. The effective refractive indices of these substrates were found using Eq. (1) with nm = 0.0666 + 4.045i [30]. A plot of effective refractive index versus glycerol concentration is shown in Fig. 4. Each measured point represents the mean value from 10 measurements. The least squares linear fits for the measurements with and without nanoparticles are given by nilo(c) = 1.109 103 c + 1.344 and nwnp(c) = 8.421 104 c + 1.365, respectively, where c denotes the concentration. In Fig. 4, the linear fits are shown with confident bands. The refractive index of nanoparticles is 1.431, calculated from the intersection of the two lines. This value is slightly lower than the literature value of 1.433 for CaF2 at 632.8 nm [31]. Refractive index differences can typically be explained by differences in the structure or purity of the material or by different measurement temperatures. Khlebtsov et al. have studied the effect of particle size on the refractive index of silica nanoparticles [32];
Fig. 3. (a) An image recorded by the image sensor. The radius of the ring-shaped absorption pattern is obtained from computational image processing. The substrate holds nanoparticles immersed by a 65 wt% glycerol solution. (b) A plot of reflected intensity versus spatial frequency is shown for readers’ information. The plot value represents averaged intensity of the reflected light along circumference of a concentric ring.
where qsp, ksp, k0, nm, and ns denote the radius of the absorption pattern, the propagating constant of the surface plasmon, the wavenumber of light in vacuum, the refractive index of the metal, and the refractive index of the sample, respectively. Prior to measuring samples, the radius was calibrated (correlating the
Fig. 4. Measured refractive indices of substrates with or without nanoparticles as a function of glycerol concentration. The substrates were immersed into glycerol/ water solutions. The plot points with error bars represent averages and standard deviations form 10 measurements. The least squares linear fits for both series of substrates are shown with confident bands; the refractive index of the nanoparticles can be found from the intersection of these lines.
H. Kano et al. / Chemical Physics Letters 654 (2016) 72–75
they found that the refractive index of silica changes from 1.474 to 1.472 as the size of the particle decreases from 210 to 100 nm, respectively. One of the reasons for lesser value of the estimated effective index of refraction could be due to a high hygroscopic property of glycerol [33]. Although our results show the same basic trend, further experiments are necessary to rule out spurious effects using, e.g., nanoparticles with a narrower size distribution, finer temperature control, and a variety of immersion liquids. 4. Conclusions In this study, we have developed and demonstrated a localized SPR sensor based immersion refractometry technique for measuring the refractive index of CaF2 nanoparticles. Major advantages of localized-SPR sensor, compared to other techniques, are the sensitivity and the small amount of sample (6 atto liters) [34]; timeconsuming measurements are a disadvantage at present. This method is accurate up to 0.002 refractive index units; such rigorous characterization is necessary to understand and control the optical properties of nanoparticles. Nanoparticles with absorption that appears in the imaginary part of the refractive index can be also characterized by measuring the width of the absorption pattern in the reflected spatial frequency distribution because damping of SPs due to absorption on the metallic surface broadens the propagating constant [35]. We believe that determination of the refractive index of nanoparticles with localized SPR sensor will become a power tool in pharmaceuticals, cosmetics, paints, inks as well as other fields wherein the optical properties of nanoparticles must be accurately characterized. Acknowledgment Niskanen wishes to express his gratitude as an International Research Fellow of the Japan Society to the Promotion of Science (JSPS) for the grant that made also possible a part of this study. References [1] Y. Cheng, C. Lü, B. Yang, Recent Pat. Mater. Sci. 4 (2011) 15. [2] G. Knöner, S. Parkin, T.A. Nieminen, N.R. Heckenberg, H. Rubinsztein-Dunlop, Phys. Rev. Lett. 97 (2006) 157402. [3] Z. Wang, N. Wang, D. Wang, Z. Tang, K. Yu, W. Wei, Opt. Lett. 39 (2014) 4251.
75
[4] I. Bodurov, T. Yovcheva, S. Sainov, J. Phys: Conf. Ser. 558 (2014) 012062. [5] H.H. Lim, M.S. Kwon, H.J. Choi, B.-J. Kim, M. Cha, J. Opt. Soc. Korea 12 (2008) 210. [6] G.H. Meeten, A.N. North, Meas. Sci. Technol. 6 (1995) 215. [7] S. Kubo, A. Diaz, Y. Tang, T.S. Mayer, I.C. Khoo, T.E. Mallouk, Nano Lett. 7 (2007) 3418. [8] S. Auvinen, M. Alatalo, H. Haario, E. Vartiainen, J.-P. Jalava, R.J. Lamminmäki, J. Phys. Chem. C 117 (2013) 3503. [9] M.G. Feeney, R. Ince, M.H. Yukselici, C. Allahverdi, Appl. Opt. 50 (2011) 3259. [10] J.G. Webster, The Measurement, Instrumentation and Sensor Handbook, CRC, Florida, 1999. [11] E.Y. Levitin, N.G. Kokodiy, V.A. Timanjuk, I.O. Vedernikova, T.M. Chan, Neorg. Mater. 50 (2014) 817. [12] Z. Wang, N. Wang, D. Wang, Z. Tang, K. Yu, W. Wei, Opt. Lett. 39 (2014) 4251. [13] R. Márquez-Islas, C. Sánchez-Pérez, A. García-Valenzuela, Opt. Lett. 39 (2014) 559. [14] E.S. Larsen, H. Herman, The Microscopic Determination of the Nonopaque Minerals, second ed., United State Government Printing Office, Washington, 1934. [15] A.S. Andrushchaka, B.V. Tybinkaa, I.P. Ostrovskija, W. Schranzb, A.V. Kitykc, Opt. Lasers Eng. 46 (2008) 162. [16] N.V. Kamanina, S.V. Serov, N.A. Shurpo, N.N. Rozhkova, Tech. Phys. Lett. 37 (2011) 949. [17] H. Shpaisman, B.J. Krishnatreya, D.G. Grier, Appl. Phys. Lett. 101 (2012) 091102. [18] E.V.D. Pol, F.A.W. Coumans, A. Sturk, R. Nieuwland, T.G.V. Leeuwen, Nano Lett. 14 (2014) 6195. [19] C. Gardiner, M. Shaw, P. Hole, J. Smith, D. Tannetta, C.W. Redman, I.L. Sargent, J. Extracell. Vesicles 24 (2014) 2536. [20] E. Patois, M.A.H. Capelle, C. Palais, R. Gurny, T. Arvinte, Sci. Technol. 22 (2012) 427. [21] V. Filipe, A. Hawe, W. Jiskoot, Pharm. Res. 27 (2010) 796. [22] J. Homola, S.S. Yee, G. Gauglitz, Sens. Actuators, B 54 (1999) 3. [23] J. Homola, Surface Plasmon Resonance Based Sensors, Springer, Berlin, 2006. [24] H. Kano, S. Mizuguchi, S. Kawata, J. Opt. Soc. Am. B 15 (1998) 1381. [25] K. Watanabe, N. Horiguchi, H. Kano, Appl. Opt. 46 (2007) 4985. [26] K. Ohenoja, M. Illikainen, Powder Technol. 283 (2015) 254. [27] K. Ohenoja, J. Saari, M. Illikainen, S. Breitung-Faes, A. Kwade, J. Niinimäki, J. Chem. Eng. Technol. 37 (2014) 833. [28] C. Knieke, M. Sommer, W. Peukert, Powder Technol. 195 (2009) 25. [29] H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings, Springer, Berlin, 1988. [30] W. Wolfe, Handbook of Optics, McGraw-Hill, New York, 1978. [31] E.D. Palik, Handbook of Optical Constants of Solids III, Academic Press, San Diego, 1998. [32] B.N. Khlebtsov, V.A. Khanadeev, N.G. Khlebtsov, Langmuir 24 (2008) 8964. [33] G. Terentyuk, E. Panfilova, V. Khanadeev, D. Chumakov, E. Genina, A. Bashkatov, V. Tuchin, A. Bucharskaya, G. Maslyakova, N. Khlebtsov, B. Khlebtsov, Nano Res. 7 (2014) 325. [34] J. Ning, K. Nagata, A. Ainai, H. Hasegawa, H. Kano, Jpn. J. Appl. Phys. 53 (2013). 082402-1. [35] H. Kano, S. Kawata, Appl. Opt. 33 (1994) 5166.
Contents lists available at ScienceDirect
Chemical Physics Letters journal homepage: www.elsevier.com/locate/cplett
Research paper
Refractive index measurement of nanoparticles by immersion refractometry based on a surface plasmon resonance sensor Hiroshi Kano a, Ayumu Iseda a, Katja Ohenoja b, Ilpo Niskanen c,d,⇑ a
Department of Information and Electronic Engineering, Muroran Institute of Technology, Mizumoto 27-1, Muroran, Hokkaido 050-8585, Japan Faculty of Technology, Fibre and Particle Engineering Group, University of Oulu, P.O. Box 4500, FI-90014 Oulu, Finland c Faculty of Technology, University of Oulu, P.O. Box 4500, FI-900014 Oulu, Finland d National Institute of Industrial Science and Technology, Namiki 1-2, Tsukuba, Ibaraki 305-8564, Japan b
a r t i c l e
i n f o
Article history: Received 28 January 2016 In final form 7 May 2016 Available online 9 May 2016 Keywords: Nanoparticles Refractive index Surface plasmon resonance technique
a b s t r a c t Accurate determination of the refractive index of nanoparticles has important ramifications for applications, such as pharmaceuticals, cosmetics, paints, textiles, and inks. We describe a new method to determine the refractive index of nanoparticles by immersion refractometry with a surface plasmon resonance sensor. With this method, the refractive index of the nanoparticles is perfectly matched with that of the surrounding liquid. We demonstrate this method for calcium fluoride nanoparticles that have an average diameter of 100 nm; the results achieve an accuracy of better than 0.002 refractive index units. Ó 2016 Published by Elsevier B.V.
1. Introduction Nanoparticles are one of the most-researched materials today due to their range of potential applications in biomedical, optical, and electronic fields. The refractive index of nanoparticles is one of the most important physical parameters used to inspect the quality of particle-based solids. Materials technology strives to tune the effective refractive index of nanoparticles to manipulate their matrix properties at a molecular level [1]. Tuning the optical, magnetic, and electrical properties of refractive index can be tailored nanoparticles for specific applications, thus increasing their effectiveness. It is difficult to precisely measure the refractive index of nanoparticles with current techniques [2]. In general, the refractive index is not only related to the chemical composition, crystal structure, and symmetry of each particle but is also affected by the temperature, wavelength of incident light and internal stresses in the sample [3]. Several techniques have been developed to measure the refractive index of nanoparticles. Total internal reflection has been used to determine the refractive index of ZnO, TiO2, and ZrO2 nanoparticles; this technique is based on measuring the critical angle of total internal reflection [4]. The refractive index is measured with the experimental uncertainty of 1%. The critical-angle method has long been applied in the Abbe refractometer; the advantages ⇑ Corresponding author at: Faculty of Technology, University of Oulu, P.O. Box 4500, FI-900014 Oulu, Finland. E-mail address: [email protected] (I. Niskanen). http://dx.doi.org/10.1016/j.cplett.2016.05.013 0009-2614/Ó 2016 Published by Elsevier B.V.
include easy experimental implementation and a simple calibration [5]. The critical-angle method is precise to about ±0.002 refractive index units, RIU. Turbid liquids contain a high concentration of solid particles that scatter light toward the detector, combining with the specular reflection to cause an error [6]. Ellipsometry has been used to measure the refractive index of gold and TiO2 nanoparticles; it is based on the change in the polarization state of light reflected from the surface of a sample [7,8]. Interferometry has been used to determine the refractive index of CdSe nanoparticles [9]. Interferometry and ellipsometry are accurate techniques (0.0001–0.00001 RIU), but they are expensive and require samples with very smooth surfaces [10]. Optical methods and electron microscopy have been used to determine the size distribution and complex refractive index of magnetite (Fe3O4) nanoparticles [11]. Immersion matching has been used to measure the refractive indices of various types of nanoparticles [12,13]. The principle of the immersion method is to match the refractive index of solid particles with that of the immersion liquid. In the event of a perfect match between the index of the non-absorbing pigments and the immersion liquid, the transmittance will be 100% and the scattering intensity will approach zero. The advantages of immersion matching are that the technique is independent of the shape or size of a particle, and the measurement is easy and relatively fast to perform [14]. The refractive index of an immersion liquid is generally in the range of 1.4–1.8 [15]. As to the accuracy of the immersion methods, various authors assign it an error of 0.001–0.005 as usually performed. Dynamic holography techniques have been used to measure the refractive index of
H. Kano et al. / Chemical Physics Letters 654 (2016) 72–75
carbon nanoparticles [16]. In-line holographic microscopy of micrometer-scale techniques have been measured refractive index of silica particles with a resolution of 0.002 RIU [17]. Nanoparticle tracking analysis (NTA) is a relatively new technique for determining the refractive index of suspended nanoparticles based on the rate of their Brownian motion [18,19] which an uncertainty in the range is 0.021–0.046 RIU. NTA currently works best for particles having a diameter of approximately 10–1000 nm. The lower detection limit depends on the refractive index of the nanoparticles; characterization is only possible for nanoparticles composed of materials with a high refractive index, such as gold or silver. The upper detection limit is due to the limited Brownian motion of large nanoparticles. The rate of particle diffusion is related to the viscosity and temperature of the liquid [20,21]. Surface plasmon resonance (SPR) sensors can also perform refractive index measurements [22]. In this technique, surface plasmons (SPs), the quanta of collective electron oscillations excited on a metallic surface, are used as a measurement probe because the propagating constant of the SPs strongly depends on the refractive index on the metallic surface. In the most typical setup, p-polarized incident light is given to a metallic thin film through a prism coupler. The light having a proper incident angle resonates with the SPs that consume light energy. Therefore, the resonance angle found in angular dependency of reflectance tells the refractive index on the metal surface. The electric field produced by the SPs shows an evanescent decaying property and remarkable field enhancement against incident light. These properties provide surface sensitive and high-sensitive measurements. The theoretical resolution in the refractive index measurement with a typical setup reaches 5 107 RIU. The high sensitivity contributes to detection of an adsorbed monomolecular layer on the metallic surface [23]. A variation of the SPR sensor enables a microscopic measurement while maintaining the high resolution in the refractive index measurement [24,25]. In this case, an excitation beam is tightly focused onto a metallic thin film coated on a glass substrate by using an oil immersion objective lens with high numerical aperture; interference of excited SPs propagating in many directions localize the SPs in the optical diffraction limit region. In the measurement of nanoparticle, the localized SPs are useful for probing a region with high particle density. Furthermore, the localized-SPR sensor can be combined with the immersion technique while retaining the advantages described above, viz., refractive index measurements that are independent of the shape and size of particles. In this case, we examine a condition showing perfect index matching between the nanoparticles and the immersion liquid by using localized-SPR sensor. The aim of this study is to get the refractive index of a nanoparticle by immersion liquid method based on a localized-SPR sensor. The high sensitivity of the localized-SPR sensor enhances index mismatch between the sample and the immersion liquid, resulting in higher precision.
73
1.2. Grinding proceeded with smaller beads (320 lm for 150 min), resulting in a median particle size of 104 nm and WPSD = 1.06. Further grinding broadened the particle size distribution, indicating that the first two steps had achieved the grinding limit [27,28]. Samples processed by the first two steps were used in this study. The Field Emission Scanning Electron Microscope (FESEM) image taken with a Zeiss Ultra Plus is presented in Fig. 1, along with the particle size distribution of a ground CaF2 suspension measured by laser diffraction. Fig. 2 shows a schematic of the localized-SPR sensor. The beam width of linearly polarized light with a wavelength of 632.8 nm was enlarged by a beam expander; the polarization was converted from linear to radial by using a radial polarization converter (ARCoptix, Switzerland) to obtain the optimal localization of SPs [25]. The position of the converter is imaged to the entrance pupil of an oil immersion objective lens with a numerical aperture of 1.65 (Olympus, Japan). The light converged from the objective lens and illuminated a coverslip coated with a metallic thin film. The reflected light from the coverslip was relayed to an image sensor located at an optically conjugate plane of the exit pupil of the objective lens. The image sensor recorded the spatial frequency distribution of the reflected light. A typical measurement recorded by the image sensor is shown in Fig. 3. An absorption ring appeared in the image; its radius is used to calculate the propagating constant of the surface plasmons. The absorption pattern is fit by computational processing to quantify the radius and coordinates of the ring. The radius that corresponds to the real part of the propagating constant of excited SPs can be approximately expressed by the following equation [22,25,29]:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi! n2 n2 qsp ¼ Reðksp Þ ffi Re k0 2 m s 2 nm þ ns
ð1Þ
2. Materials and methods We examined this technique by measuring the refractive index of prepared nanoparticles. To produce the nanoparticles, CaF2 powder (Merck, 97%) was purchased; it contained trace amounts of Fe, Mg, Na, and SiO4, according to the product description. The median particle size (on a volumetric basis) was 7.0 lm, measured by a Beckman Coulter LS 13 320 laser diffraction analytical device. Nanogrinding was carried out for the CaF2 powder via a two-step process with a stirred media mill (90 AHM hydro mill, Hosokawa Alpine). The operation of the mill has been previously described in detail [26]. After the first grinding (370 min with 560 lm Zirconia (YSZ) beads), the median particle size was 110 nm and the width of the particle size distribution WPSD = ((d90 d10)/d50) was
Fig. 1. (a) FESEM image and (b) particle size distribution of ground CaF2 nanoparticles. The scale bar in the FESEM image represents 300 nm.
74
H. Kano et al. / Chemical Physics Letters 654 (2016) 72–75
best-fitting ring in pixel units to spatial frequency) by measuring the refractive index of air, which is 1.000. The samples were measured at room temperature. After finding the coordinate center, we can obtain the relationship between the reflected intensity and the spatial frequency of light. We include the plot in Fig. 3 for readers’ information. The position of absorption dip varies by the effective index on the metal surface. 3. Results and discussion
Fig. 2. Schematic of the localized-SPR sensor. Radially polarized light is focused onto the metallic thin film by the oil immersion objective lens. Surface plasmons excited in the diffraction-limited-region (180 nm) [28] interact with a sample (nanoparticles and immersion liquid). The spatial-frequency distribution of reflected light is recorded by an image sensor. The propagating constant is determined from the excited surface plasmons, enabling calculation of the effective refractive index on the metal surface.
We used vacuum evaporation to prepare substrates coated by silver with a thickness of 54 nm. The substrates in each batch were secured onto the rotating stage in the evaporator at the same distance from the rotation center to ensure uniform quality. We also prepared glycerol (Prolabo Company) solutions diluted with MilliQ water to concentrations of 44, 54, 65, and 100 wt% glycerol. We drop cast these glycerol solutions onto substrates to prepare samples without nanoparticles. Samples with nanoparticles were prepared by drop casting CaF2 nanoparticles dispersed in ethanol onto substrates; the substrates were dried in air and different concentrations of glycerol solutions were drop cast on top. Two separate evaporation batches were used to produce (1) the substrates without nanoparticles and (2) substrates with nanoparticles. The effective refractive indices of these substrates were found using Eq. (1) with nm = 0.0666 + 4.045i [30]. A plot of effective refractive index versus glycerol concentration is shown in Fig. 4. Each measured point represents the mean value from 10 measurements. The least squares linear fits for the measurements with and without nanoparticles are given by nilo(c) = 1.109 103 c + 1.344 and nwnp(c) = 8.421 104 c + 1.365, respectively, where c denotes the concentration. In Fig. 4, the linear fits are shown with confident bands. The refractive index of nanoparticles is 1.431, calculated from the intersection of the two lines. This value is slightly lower than the literature value of 1.433 for CaF2 at 632.8 nm [31]. Refractive index differences can typically be explained by differences in the structure or purity of the material or by different measurement temperatures. Khlebtsov et al. have studied the effect of particle size on the refractive index of silica nanoparticles [32];
Fig. 3. (a) An image recorded by the image sensor. The radius of the ring-shaped absorption pattern is obtained from computational image processing. The substrate holds nanoparticles immersed by a 65 wt% glycerol solution. (b) A plot of reflected intensity versus spatial frequency is shown for readers’ information. The plot value represents averaged intensity of the reflected light along circumference of a concentric ring.
where qsp, ksp, k0, nm, and ns denote the radius of the absorption pattern, the propagating constant of the surface plasmon, the wavenumber of light in vacuum, the refractive index of the metal, and the refractive index of the sample, respectively. Prior to measuring samples, the radius was calibrated (correlating the
Fig. 4. Measured refractive indices of substrates with or without nanoparticles as a function of glycerol concentration. The substrates were immersed into glycerol/ water solutions. The plot points with error bars represent averages and standard deviations form 10 measurements. The least squares linear fits for both series of substrates are shown with confident bands; the refractive index of the nanoparticles can be found from the intersection of these lines.
H. Kano et al. / Chemical Physics Letters 654 (2016) 72–75
they found that the refractive index of silica changes from 1.474 to 1.472 as the size of the particle decreases from 210 to 100 nm, respectively. One of the reasons for lesser value of the estimated effective index of refraction could be due to a high hygroscopic property of glycerol [33]. Although our results show the same basic trend, further experiments are necessary to rule out spurious effects using, e.g., nanoparticles with a narrower size distribution, finer temperature control, and a variety of immersion liquids. 4. Conclusions In this study, we have developed and demonstrated a localized SPR sensor based immersion refractometry technique for measuring the refractive index of CaF2 nanoparticles. Major advantages of localized-SPR sensor, compared to other techniques, are the sensitivity and the small amount of sample (6 atto liters) [34]; timeconsuming measurements are a disadvantage at present. This method is accurate up to 0.002 refractive index units; such rigorous characterization is necessary to understand and control the optical properties of nanoparticles. Nanoparticles with absorption that appears in the imaginary part of the refractive index can be also characterized by measuring the width of the absorption pattern in the reflected spatial frequency distribution because damping of SPs due to absorption on the metallic surface broadens the propagating constant [35]. We believe that determination of the refractive index of nanoparticles with localized SPR sensor will become a power tool in pharmaceuticals, cosmetics, paints, inks as well as other fields wherein the optical properties of nanoparticles must be accurately characterized. Acknowledgment Niskanen wishes to express his gratitude as an International Research Fellow of the Japan Society to the Promotion of Science (JSPS) for the grant that made also possible a part of this study. References [1] Y. Cheng, C. Lü, B. Yang, Recent Pat. Mater. Sci. 4 (2011) 15. [2] G. Knöner, S. Parkin, T.A. Nieminen, N.R. Heckenberg, H. Rubinsztein-Dunlop, Phys. Rev. Lett. 97 (2006) 157402. [3] Z. Wang, N. Wang, D. Wang, Z. Tang, K. Yu, W. Wei, Opt. Lett. 39 (2014) 4251.
75
[4] I. Bodurov, T. Yovcheva, S. Sainov, J. Phys: Conf. Ser. 558 (2014) 012062. [5] H.H. Lim, M.S. Kwon, H.J. Choi, B.-J. Kim, M. Cha, J. Opt. Soc. Korea 12 (2008) 210. [6] G.H. Meeten, A.N. North, Meas. Sci. Technol. 6 (1995) 215. [7] S. Kubo, A. Diaz, Y. Tang, T.S. Mayer, I.C. Khoo, T.E. Mallouk, Nano Lett. 7 (2007) 3418. [8] S. Auvinen, M. Alatalo, H. Haario, E. Vartiainen, J.-P. Jalava, R.J. Lamminmäki, J. Phys. Chem. C 117 (2013) 3503. [9] M.G. Feeney, R. Ince, M.H. Yukselici, C. Allahverdi, Appl. Opt. 50 (2011) 3259. [10] J.G. Webster, The Measurement, Instrumentation and Sensor Handbook, CRC, Florida, 1999. [11] E.Y. Levitin, N.G. Kokodiy, V.A. Timanjuk, I.O. Vedernikova, T.M. Chan, Neorg. Mater. 50 (2014) 817. [12] Z. Wang, N. Wang, D. Wang, Z. Tang, K. Yu, W. Wei, Opt. Lett. 39 (2014) 4251. [13] R. Márquez-Islas, C. Sánchez-Pérez, A. García-Valenzuela, Opt. Lett. 39 (2014) 559. [14] E.S. Larsen, H. Herman, The Microscopic Determination of the Nonopaque Minerals, second ed., United State Government Printing Office, Washington, 1934. [15] A.S. Andrushchaka, B.V. Tybinkaa, I.P. Ostrovskija, W. Schranzb, A.V. Kitykc, Opt. Lasers Eng. 46 (2008) 162. [16] N.V. Kamanina, S.V. Serov, N.A. Shurpo, N.N. Rozhkova, Tech. Phys. Lett. 37 (2011) 949. [17] H. Shpaisman, B.J. Krishnatreya, D.G. Grier, Appl. Phys. Lett. 101 (2012) 091102. [18] E.V.D. Pol, F.A.W. Coumans, A. Sturk, R. Nieuwland, T.G.V. Leeuwen, Nano Lett. 14 (2014) 6195. [19] C. Gardiner, M. Shaw, P. Hole, J. Smith, D. Tannetta, C.W. Redman, I.L. Sargent, J. Extracell. Vesicles 24 (2014) 2536. [20] E. Patois, M.A.H. Capelle, C. Palais, R. Gurny, T. Arvinte, Sci. Technol. 22 (2012) 427. [21] V. Filipe, A. Hawe, W. Jiskoot, Pharm. Res. 27 (2010) 796. [22] J. Homola, S.S. Yee, G. Gauglitz, Sens. Actuators, B 54 (1999) 3. [23] J. Homola, Surface Plasmon Resonance Based Sensors, Springer, Berlin, 2006. [24] H. Kano, S. Mizuguchi, S. Kawata, J. Opt. Soc. Am. B 15 (1998) 1381. [25] K. Watanabe, N. Horiguchi, H. Kano, Appl. Opt. 46 (2007) 4985. [26] K. Ohenoja, M. Illikainen, Powder Technol. 283 (2015) 254. [27] K. Ohenoja, J. Saari, M. Illikainen, S. Breitung-Faes, A. Kwade, J. Niinimäki, J. Chem. Eng. Technol. 37 (2014) 833. [28] C. Knieke, M. Sommer, W. Peukert, Powder Technol. 195 (2009) 25. [29] H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings, Springer, Berlin, 1988. [30] W. Wolfe, Handbook of Optics, McGraw-Hill, New York, 1978. [31] E.D. Palik, Handbook of Optical Constants of Solids III, Academic Press, San Diego, 1998. [32] B.N. Khlebtsov, V.A. Khanadeev, N.G. Khlebtsov, Langmuir 24 (2008) 8964. [33] G. Terentyuk, E. Panfilova, V. Khanadeev, D. Chumakov, E. Genina, A. Bashkatov, V. Tuchin, A. Bucharskaya, G. Maslyakova, N. Khlebtsov, B. Khlebtsov, Nano Res. 7 (2014) 325. [34] J. Ning, K. Nagata, A. Ainai, H. Hasegawa, H. Kano, Jpn. J. Appl. Phys. 53 (2013). 082402-1. [35] H. Kano, S. Kawata, Appl. Opt. 33 (1994) 5166.