Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Feb 14;96(8):3284–3290. doi: 10.1021/acs.analchem.3c03593

A Quantitative Approach to Characterize the Surface Modification on Nanoparticles Based on Localized Dielectric Environments

Taketo Mochizuki †, Shota Sampei †, Keishi Suga †,*, Kanako Watanabe †, Tom A J Welling ‡, Daisuke Nagao †,*
PMCID: PMC10902806  PMID: 38355104

Abstract

graphic file with name ac3c03593_0006.jpg

Nanoparticles (NPs) are utilized for the functionalization of composite materials and nanofluids. Although oxide NPs (e.g., silica (SiO2)) exhibit less dispersibility in organic solvents or polymers due to their hydrophilic surface, the surface modification using silane coupling agents can improve their dispersibility in media with low dielectric constants. Herein, SiO2 NPs were functionalized using octyltriethoxysilane (OTES, C8) and dodecyltriethoxysilane (DTES, C12), wherein the degrees of surface modification of SiO2@C8 and SiO2@C12 were quantitatively evaluated based on the ratio of modifier to surface silanol group (θ) and the volume fraction of organic modifier to total particle volume (ϕR). The variations of surface properties were revealed by analyzing the Hansen solubility parameters (HSP). Particularly, the surface modification using OTES or DTES significantly affected the polarity (δP) of NPs. The local dielectric environments of surface-modified SiO2 NPs were characterized using a solvatochromic dye, Laurdan. By analyzing the peak position of the steady-state emission spectrum of Laurdan in a NP suspension, the apparent dielectric environments surrounding NPs (εapp) were obtained. A good correlation between ϕR and εapp was observed, indicating that ϕR is a reliable quantity for understanding the properties of surface-modified NPs. Furthermore, the generalized polarization (GP) of NPs was investigated. The surface-modified SiO2 NPs with higher ϕR (≥0.15) exhibited GP > 0, suggesting that the modifiers are well-organized on the surface of NPs. The localized dielectric environment surrounding NPs could be predicted by analyzing the volume fraction of nonpolar moieties derived from modifiers. Alternatively, εapp and GP can be utilized for understanding the properties of inorganic–organic hybrid NPs.

Introduction

Oxide nanoparticles (NPs), typically with a diameter of about 100 nm or smaller, exhibit unique optical, electrical, magnetic, and catalytic properties that are different from those of bulk materials.1−5 These NPs, such as silica (SiO2), are also used as filler materials to improve the properties of nanocomposites.6−9 Recently, nanocomposite films loaded with SiO2 NPs were focused on as gas separation membranes,10−12 where the dispersibility of NPs in the polymer matrix played important roles in permeation selectivity. Importantly, the aggregation or agglomeration of such NPs should be avoided, as it easily leads to a loss of their surface functionalities. However, most of oxide NPs are hydrophilic,13−16 and hence have poor dispersibility in nonpolar media such as organic solvents and polymers. The surface modification of NPs is a promising way to improve the dispersibility of filler NPs in polymer.7 Therefore, a precise surface modification should be performed to control the dispersibility of NPs in nonpolar media.

NPs modified with a nonpolar ligand exhibit superior dispersibility in nonpolar solvents (e.g., toluene and n-hexane).17−20 Owing to chemisorbed organic modifiers on NPs, the dispersibility of NPs could be drastically altered.21,22 By using such surface-modified NPs, polymer nanocomposites with better NP dispersibility have been developed. Particularly in the field of electronics, a higher dispersibility of nanofillers (e.g., barium titanate (BaTiO3)) is required to improve the electrochemical performance and transparency of nanocomposites.23−26 On the other hand, some organically modified NPs are less dispersible in polar organic solvents such as DMSO, acetone, etc.22,27 Possible approaches to tune the dispersibility of NPs in solvents, via surface modification, are (1) a combined use of modifiers28 and (2) a quantitative control of the ratio of modifier to surface silanol group.14 In most cases, however, an excess amount of modifier was used in the surface modification reaction, which made it difficult to have a quantitative discussion of how the degree of surface modification contributes to altering the surface properties of NPs. Hence, the level of surface modification on a nanoparticle should be quantitatively investigated and its effect on the NP dispersibility should be comprehensively studied.

As characterization of NP dispersibility, Hansen solubility parameters (HSP) or Hansen dispersibility parameters (HDP) have been adopted to obtain a deeper insight into the surface properties of NPs.16,22,29,30 Fujiwara et al. reported the surface modification of copper NPs using silane coupling agents and estimated the alteration of their surface properties based on HSP.16 Although the correlation between the degree of surface modification and surface properties was unclear, the HSP value of surface-modified NPs became closer to that of the modifier. Stauch et al. reported a correlation between the degree of surface modification and the polarity of SiO2 NPs.14 They reported a stepwise variation of HSP values of surface-modified NPs, which could be because of the limited numbers of solvents used for HSP analysis. Although HSP has nonetheless proved a powerful characterization tool for predicting NP dispersibility, the HSP characterization is a time-consuming process because it generally requires ca. 20 kinds of solvents for the dispersibility test.16,22,30 As another approach to obtain the local surface properties of materials, solvatochromic dyes are useful. Pyrene derivatives, betaine derivatives, and naphthalene derivatives have been utilized to investigate the micropolarity in blend polymer,31 silica NPs,14 and lipid self-assemblies,32−35 respectively. However, the surface properties and particle dispersibility have not yet been linked sufficiently, and the prediction of NP dispersibility in different media is still a challenge.

The degree of surface modification is usually described by the ratio of modifier to surface silanol group (θ, surface coverage) or by the volume fraction of the organic modifier to total particle volume (ϕR). When comparing different types of modifiers such as octyltriethoxysilane (OTES, C8) and dodecyltriethoxysilane (DTES, C12), ϕR is beneficial to understand how much amount of organic modifiers are attached on the particle surface. In this study, we synthesized surface-modified NPs with different amounts of modifier and investigated how the properties of NPs were altered by surface modification. Stöber silica NPs (SiO2, diameter: 35, 58, 132, and 150 nm) were selected as model NPs and were modified using OTES or DTES. HSP values of NPs were evaluated (Scheme S1). To investigate the localized hydrophobicity surrounding NPs in the solvent, fluorescence emission spectroscopic analysis using 6-dodecanoyl-2-dimethylaminonaphthalene (Laurdan) was performed. The local dielectric environments were analyzed by using the apparent dielectric constant (εapp) and generalized polarization (GP), and the correlation between NP dispersibility and these parameters was discussed.

Experimental Section

Synthesis of Surface-Modified SiO2 NPs

Stöber silica NPs were synthesized under base-catalyst conditions.36,37 A NH3/H2O/EtOH mixture was stirred at 55 °C for 30 min, and after that, an aliquot amount of tetraethyl orthosilicate (TEOS) was injected to initiate the reaction. The reaction was carried out at 55 °C for 2 h. Microscopic images and nitrogen adsorption–desorption isotherms of synthesized SiO2 NPs are shown in Figure S1. Using as-prepared SiO2, surface modification reactions were conducted at 70 °C for 1 h with varied OTES concentrations (1.4–11 mM) and with varied DTES concentrations (1.6–3.2 mM). After the reaction, the particles were washed by centrifugation and vacuum-dried at 60 °C overnight. Particle characterizations (HSP, Laurdan analysis) were conducted within 4 weeks after synthesis.

Characterization of Synthesized Particles

The synthesized SiO2 NPs were observed by using FE-TEM (Hitachi, HD-2700). The colloidal stability of NPs was evaluated using dynamic light scattering (DLS; Otsuka Electronics, ELS-Z2), at the particle concentrations of approximately 0.1 vol %. N2 adsorption–desorption isotherms of SiO2 particles were measured by a BELSORP-mini II (Bel Japan Inc.). Thermogravimetric analysis (TGA, TG/DTA 7200, Seiko Instruments Inc., Japan)) was carried out to estimate the amount of modifier molecules introduced on the surface of SiO2 particles. The fluorescence spectra of Laurdan in suspensions of SiO2 NPs were measured at an excitation wavelength of 340 nm at 22 °C by using a fluorescence spectrophotometer (Hitachi High-Tech Corporation, F-7000).

The ratio of modifier to surface silanol group (θ) and the volume fraction of modifier to total particle volume (ϕR) are defined as follows;

graphic file with name ac3c03593_m001.jpg 1
graphic file with name ac3c03593_m002.jpg 2

where Nmod and NSiOH represent the amount of modifier and silanol group, respectively. VR and Vcore represent the volume fractions of modifier and core particles, respectively.

Evaluation of HSP Values

The HSP values of SiO2 NPs were determined by the Hansen solubility sphere method, according to a previous report.16 The solvents used in this study are given in Table S1. SiO2 NPs (2.0 mg/mL) were incubated for 20 h at room temperature (22 °C). Via visual inspection, the tested solvents including sedimentation were regarded as poor solvent (score 0). Then, DLS measurements were performed to evaluate the dispersed particle size (dDLS). In cases where dDLS ≤ 2dprim, the tested solvents were good solvents (score 1). The HSP values of NPs were analyzed using a software (HSPiP ver. 5.4.08, https://www.hansen-solubility.com/). The HSP distance (Ra) is calculated as follows:

graphic file with name ac3c03593_m003.jpg 3

where δD,i, δP,i, and δH,i indicate the HSP values of component i (i = 1, 2).

Results and Discussion

Evaluation of the Degree of Surface Modification

SiO2 NPs of diameter 35, 58, 132, and 150 nm, with different degrees of surface modification, were used in this study. The surface-modified SiO2 NPs using the OTES (SiO2@C8) and DTES (SiO2@C12) were characterized based on TGA (Figure 1a and Figure S2). Generally, the alkoxide group of the silane coupling agent is hydrolyzed (R-Si(OH)3), and subsequently, a condensation reaction between silanol groups proceeds.38,39 The weight loss observed in pristine SiO2 particles (Figure 1a and Figure S2) originated from the dehydration and condensation of silanol groups on the surface of silica.40 Although the weight loss of pristine SiO2 particles varied (1.8–4.2%), surface-modified SiO2 particles exhibited more significant weight loss, revealing successful introduction of organic modifiers on the particle surface. The weight loss of surface-modified SiO2 NPs increased in proportion to the concentration of the OTES, suggesting an increase in surface modification degree. Similarly, the weight loss became significant by an increase in applied modifier, in the cases of SiO2 NPs of dprim = 35, 132, and 150 nm (Figure S2). A correlation between θ and ϕR is shown in Figure 1b. As described, the ϕR values efficiently increase with θ when the particle size becomes smaller. For stable NPs such as silica, θ can be quantitatively investigated by comparing the TGA curves of pristine NPs and modified NPs. On the other hand, some NPs are not stable without modifiers, which makes the quantification of θ difficult. For these, it is also possible to calculate the ϕR value based on the primary diameter (observed by TEM) and TGA curve of surface-modified NPs.41−43 Thus, ϕR is more appropriate when comparing NPs of different particle sizes (dprim) and those using different modifier types.

Figure 1.

Figure 1

Open in a new tab

(a) TGA analyses for SiO2@C8 (dprim = 58 nm). (b) Experimental values of θ and ϕR for SiO2@C8. Circles: dprim = 58 nm, triangles: dprim = 132 nm. Dotted lines are theoretical values of ϕR calculated from θ. For details, see the Supporting Information.

HSP Values of Surface-Modified SiO2 NPs

The dispersibility of NPs was evaluated from two perspectives, namely, dispersion stability and hydrodynamic diameter measured using DLS (dDLS) (Scheme S1). Using the solvents listed in Table S1, the HSP values of SiO2@C8 and SiO2@C12 were investigated according to the HSP sphere method (Figures S3 and S4, Table S2). Details of dispersibility are shown in Tables S3–S15. Pristine SiO2 were not dispersible in nonpolar solvents (e.g., n-hexane), while SiO2@C8 (dprim = 58 nm and θ = 2.1 (ϕR = 0.19)), SiO2@C8 (dprim = 132 nm and θ = 2.1 (ϕR = 0.091)), and SiO2@C12 (dprim = 150 nm and θ = 2.0 (ϕR = 0.11)) exhibited monodispersity in n-hexane. The alteration of HSP values as a function of ϕR is summarized in Figure 2. For SiO2 NPs of dprim = 132 and 150 nm, δD values were almost constant (Figure 2a), δP values decreased with ϕR (Figure 2b), and δH reached a plateau at ϕR ≥ 0.1 (Figure 2c). A similar trend was observed for the SiO2 NPs of dprim = 58 nm, while the impact of ϕR on the variations of δ was different as compared to the SiO2 particles larger than 100 nm. Consequently, δtot values dose-dependently decreased with ϕR (Figure 2d).

Figure 2.

Figure 2

Open in a new tab

ϕR-dependency of HSP values for SiO2@C8 and SiO2@C12. Circles, triangles, and diamonds indicate SiO2@C8 of dprim = 58 nm, SiO2@C8 of dprim = 132 nm, and SiO2@C12 of dprim = 152 nm, respectively. (a) δD, (b) δP, (c) δH, and (d) δDtot. (e) (left) Chemical structure of OTES derivatives. (right) Ra values between OTES derivatives and SiO2@C8. Open bars: dprim = 58 nm and θ = 2.1 (ϕR = 0.19), filled bars: dprim = 132 nm and θ = 2.1 (ϕR = 0.091).

In the surface modification reactions using alkoxysilane, it is expected that the silanol groups derived from modifiers exist on the particle surface. To reveal this, the HSP distance (Ra) between the SiO2@C8 (dprim = 58 nm and θ = 2.1 (ϕR = 0.19), dprim = 132 nm and θ = 2.1 (ϕR = 0.91)) and OTES (0) and its derivatives (octyldiethoxysilanol (ODES–OH: 1), octylethoxysilanediol (OES-(OH)2: 2), and octylsilanetriol (OS-(OH)3: 3 (Figure 2e)) are compared. The Ra value toward SiO2@C8 of dprim = 58 nm becomes a minimum with the use of ODES-(OH), whereas the Ra value toward SiO2@C8 of dprim = 132 nm is minimized with the use of OES-(OH)2 (Figure 2f). It is suggested that the physicochemical properties of NPs after modification with OTES could be regarded as ODS–OH or OES-(OH)2, when the ratio of modifier to surface SiOH is sufficiently high (ϕR ≈ 0.1). Fujiwara et al. reported on surface-modified Cu NPs using several kinds of silane coupling agents,16 wherein the δH values ranges from 0.3 to 14.1. Also, Stauch et al. reported the surface-modified SiO2,14 wherein the δH values ranges from 10.9 to 23.3, depending on the applied modifier amounts in the surface modification reaction. It is revealed that the surface modification of NPs using trialkoxysilane (R-triethoxysilane, R-trimethoxysilane) can alter the δP of NPs that can be quantitatively tuned by ϕR (or θ), while the δH of modified NPs tends to be maintained, due to the persistence of silanol groups.

Localized Dielectric Environment Surrounding the NP

The number density of silanol groups at the silica particle surface is almost constant (α = 4.9–5.6 units/nm2),44,45 which brings an idea that the properties of surface-modified SiO2 NPs only depends on θ. But actually, the HSP values of SiO2@C8 varied with both θ and particles size (dprim). This could be due to the differences in the structure (or orientation) of modifiers on the particle surface. To clarify this point, the local dielectric environment surrounding SiO2 NPs was investigated based on the steady-state fluorescence emission spectrum of Laurdan. In the dispersion medium of water/EtOH = 9:1 (ε ≈ 7346), Laurdan emits negligible fluorescence, thus, the fluorescence observed in surface-modified SiO2 suspensions can be derived from the hydrophobic environment of the surface of NPs. Original emission spectra of Laurdan obtained in SiO2@C8 and SiO2@C12 suspensions (dprim = 58 nm, 132 nm, 150 nm) are shown in Figure S5, and normalized emission spectra are shown in Figure 3 and Figure S5. With an increase of ϕR, the emission peak area increased, and the emission peak wavelength (λem) blue-shifted. As shown in Figure 3, the superimposed spectra exhibited an isoemissive point at around 465–470 nm. This suggests that the particle surface can be understood by a two-state model: the hydrophilic surface of pristine silica and the hydrophobic surface covered by modifiers. By increasing ϕR, the transition from a hydrophilic surface to a hydrophobic one can occur. To better understand the peripheral hydrophobicity of particles, we also investigated the local dielectric constant (εapp) and the generalized polarization (GP).

Figure 3.

Figure 3

Open in a new tab

Steady-state Laurdan emission spectra for SiO2@C8 of dprim = 58 nm. Measurements were performed at 22 °C, particle concentration: 3.0 mg/mL, dispersion medium: water/EtOH 9:1. Arrow indicates an isoemissive point observed at ca. 470 nm. Original spectra are shown in Figure S5.

Peripheral Permittivity

Laurdan exhibits solvatochromism in solvents,33,47 where λem is proportional to the dielectric constant of solvent (εsol). Using 1,4-dioxane/water and EtOH/water mixtures as solvent, a linear correlation between λem and εsol was found (Figure S6). Based on a linear correlation between λem and ε in solvent system, the local hydrophobicity of NPs, εapp, herein described as “peripheral permittivity”, can be calibrated:

graphic file with name ac3c03593_m004.jpg 4
graphic file with name ac3c03593_m005.jpg 5

The correlation between ϕR and εapp is shown in Figure 4a. Laurdan fluorescence intensity observed in pristine SiO2 suspension was quite weak (Figure S5a), revealing that a smaller number of hydrophobic sites exist on pristine SiO2 particles. With an increasing surface modification ratio, the peak intensity increased but a peak shift did not occur (ϕR < 0.070). In this state, the density of modifiers is insufficient to decrease the peripheral permittivity due to loosely packed chains. When ϕR > 0.070, the density of modifier chains increases, and the modifiers start to have lateral interactions with each other. This leads to the exclusion of solvent molecules from the peripheral region and a decrease in the peripheral permittivity. The εapp asymptotically approaches to the level of a nonpolar solvent (ε ≈ 2) when ϕR ≥ 0.15. This indicates that ϕR becomes an indicator to predict NP’s dispersibility in nonpolar solvents. Arita et al. reported on hexanoic acid (C6)-modified cerium oxide (CeO2) NPs (dprim: 6.5 nm, ϕR: 0.70):41 considering the ϕR value, these particles seemed to be sufficiently hydrophobized. Notably, the C6-modified CeO2 NPs exhibited a good dispersibility only in CHCl3, THF, and toluene. Tomai et al. reported on decanoic acid (C10)-modified CeO2 NPs (dprim: 5.8 nm, modifier density: 3.9 nm, estimated ϕR: 0.50) that exhibited a good dispersibility in n-hexane.22 Single-nm sized NPs possess high curvature, which might affect the lateral chain packing of modifiers. Lysenko et al. carried out a comprehensive study of magnetic NP coagulation in various solvents.48 They reported that the osmotic attraction could be a driving force for particle coagulation, where the depletion interaction between NPs could be induced by precipitant. Particle coagulation can be induced by adding polymer NPs as depletant,49 where the size of depletant is much smaller than the particles. Considering that the surfactant layer has a weak depletion effect,50,51 an excess of surface modifier might induce agglomeration/flocculation.

Figure 4.

Figure 4

Open in a new tab

Localized dielectric environments of surface-modified SiO2 NPs. (a) ϕR-dependency of peripheral permittivity (εapp) for SiO2@C8 and SiO2@C12. (b) ϕR-dependency of generalized polarization (GP) for SiO2@C8 and SiO2@C12. GP < 0 and GP > 0 indicate looser alkyl chain packing and higher alkyl chain packing, respectively. Circles: SiO2@C8 of dprim = 58 nm; triangles: SiO2@C8 of dprim = 132 nm; squares: SiO2@C8 of dprim = 35 nm; and diamonds: SiO2@C12 of dprim = 150 nm.

Generalized Polarization

Based on Laurdan analysis, generalized polarization (GP) is also used to investigate the packing state of alkyl chains.52 Lipid membranes in densely packed states exhibited a blue-shifted Laurdan emission spectrum (λem ≈ 440 nm), while lipid membranes in loosely packed states exhibited a red-shifted one (λem ≈ 490–500 nm) (Figure S7).53,54 Given that the packing density of the modifier depends on the degree of surface modification, the packing state of modifiers can be considered as a two-state model: densely packed and loosely packed. Because the isoemissive point was observed, the Laurdan spectrum obtained in the surface-modified SiO2 suspension was analyzed based on GP as follows:

graphic file with name ac3c03593_m006.jpg 6

where the Iblue and Ired represent the emission peak intensities at around 440 and 490 nm, respectively. As shown in Figure 4b, GP increased with an increase in ϕR. In addition, the NPs tested in this study maintained their dispersed state during fluorescence measurements. Thus, the localized hydrophobicity observed in NP suspensions was due to the organized modifiers on the particle surface. It is assumed that the GP value reflects the packing state of modifiers, which can be altered depending on ϕR. Although the chain structure of modifiers in less polar solvents is unknown, Yeh et al. indicated that the shorter saturated chains exhibited a higher tolerance toward styrene.55 The calculated HSP values of styrene are δD = 17.8, δP = 2.5, δH = 3.5, δtot = 18.3, and ε = 2.4. It is expected that our surface-modified SiO2 NPs maintain organized chain structures in such less polar environments.

The ϕR, the volume fraction of organic modifiers to total particle volume, could be suitable to compare nanoparticles with different sizes and those modified with different chain lengths. As shown in Figure 2d, a linear relationship between δtot and ϕR (dprim = 132 nm, 150 nm) was observed, suggesting that the increase in the total amount of organic modifiers contributes to decreasing the δtot of particles. In Figure S5, the emission peak wavelength (λem) of SiO2@C12 (dprim=150 nm and θ = 2.0 (ϕR = 0.11)) blue-shifted as compared to that of SiO2@C8 (dprim=132 nm and θ = 2.1 (ϕR = 0.092)), despite a similar surface modification ratio (θ). This suggests that differences in the lengths of the modifier chains affect the polarity surrounding NPs. From this perspective, ϕR is a useful parameter because it encompasses these differences. Although we have not examined single-sized nanoparticles, the CeO2 nanoparticles, reported by Arita et al.41 or Tomai et al.,22 exhibited a good dispersibility in nonpolar solvents, which could be revealed by higher ϕR values. Therefore, ϕR is suitable for quantitative understanding of the effect of surface modification by organic modifiers.

Factors Relevant to Particle Dispersibility

Based on the theory of Hildebrand’s regular solution, the solubility (monodisperse particle concentration) can be discussed as follows:

graphic file with name ac3c03593_m007.jpg 7

where ΔG is the Gibbs free energy, X2, V2, and ϕ2 are the molar fraction, molar volume, and volume fraction of component 2, and δ1 and δ2 are the solubility parameters (δtot) of components 1 and 2, respectively. For understanding the miscibility between polymer and solvent, the difference in δ is also utilized to obtain Flory–Huggins χ parameters. These theories support the idea that the solubility of solute increases with a decrease of the solubility parameter difference between solute–solvent, which is also applicable to the NP system.

For solvents, the correlations between ε and δ are described (Figure S8). δD does not depend on εsol, while δP increases linearly with εsol. The dependence of δH can be classified based on the type of solvent: nonpolar solvent (only CH), polar protic solvent (including −OH or -NH), or polar nonprotic solvent (including O or N atoms, but neither OH nor NH bonds). Focusing on the polar protic solvents and nonpolar solvents, a good correlation between εsol and δtot (Hildebrand solubility parameter) was seen (Figure S8d). This indicates that δtot can be identified based on εsol, where the value of δtot depends on the solvent type. Comparing the solvents with similar εsol values, polar nonprotic solvents tend to exhibit a lower dielectric constant value than polar protic solvents.

Considering the δtot–εsol correlation, the miscibility (solubility) of the two solvents can be discussed based on the difference of not only δ but also ε (Figure S8d). Assuming that NPs are similar to molecules, the solubility of NPs could be discussed based on δtot or εsol. According to our results, good dispersibility was observed when the values of εapp and εsol are close. For polar nonprotic solvents, e.g., DMSO and acetone, dispersion stability was fine; however, agglomeration was observed via DLS (Tables S3–S15). 1,4-Dioxane is a nonprotic solvent, while its level of dielectric constant is as low as nonpolar solvents. In addition, 1,4-dioxane is miscible with water. Therefore, 1,4-dioxane/water mixtures and EtOH/water mixtures are used to obtain the λ–ε calibration curve (Figure S6). This calibration curves only cover the δtot–εsol correlation of nonpolar solvents and protic solvents. The NP dispersibility in polar nonprotic solvents should be approached with caution in this case.

By employing ϕR, the peripheral permittivity εapp and generalized polarization (GP) could be estimated. Considering the δtot–ε correlation in solvents, εapp can be used to predict the δtot values of NPs. By analyzing the GP value, a parameter relevant to the chain packing states in a lipid self-assembly system, the structure of nonpolar chains on the NP surface can be discussed (Figure 5). With a lower degree of surface modification, chains are loosely packed, which enables flexible conformation of chains (amorphous-like42). With a higher degree of surface modification, i.e., ϕR ≥ 0.15, the chain packing is sufficiently tight, and the accessibility of solvent molecules could be reduced. Owing to the tightly packed structures of alkyl chains derived from modifiers, the surface of NPs becomes hydrophobic, resulting in a decrease of δtot.

Figure 5.

Figure 5

Open in a new tab

Schematic illustration of organically modified silica nanoparticles with different surface modification degrees.

Conclusions

The surface of silica NPs was modified by using OTES, where the degree of surface modification was described by ϕR. The dependency of ϕR on HSP values was investigated. A linear correlation between ϕR and δP was found, whereas the value of δH did not reach to zero with increased ϕR. This could be due to the persistence of silanol groups derived from hydrolyzed OTES. With ϕR ≥ 0.1, the HSP values of SiO2@C8 NPs could be regarded as that of ODS–OH or OES-(OH)2, which differs from the HSP of modifier (OTES). We revealed that the number of hydrolyzable groups in the modifier should be considered to estimate the HSP value of NPs after surface modification. To predict the surface properties of NPs, ϕR is a useful parameter to understand their properties as a function of the physical property values of a solvent (δ, ε).

Employing Laurdan, we quantitatively investigated the localized dielectric environments surrounding the NPs. With an increase in the surface modification, the NPs’ surroundings turned out to be hydrophobic. The generalized polarization (GP), indicating the balance of hydrophilic and hydrophobic regions at NP surface, could be utilized as a comprehensive parameter to understand the hydrophobicity of NPs. In cases of SiO2@C8 and SiO2@C12 NPs, the ratio of modifier volume to total particle volume (ϕR) can be a regulating factor for the hydrophobicity of SiO2 NPs.

The degree of surface modification could be an indicator of the surface properties of organically modified NPs. Particularly, Hildebrand’s solubility parameter, δtot, is important to understand the dispersibility of NPs in a target medium (solvent, polymer, etc.). We developed a method to evaluate the local hydrophobicity surrounding NPs in dispersion, which brings an insight into the comprehensive design of NP surfaces and enables a comparative study of reported NPs to understand how the surface modification alters the NPs’ dispersibility. Interestingly, in some solvents, e.g., DMSO and acetone, the dispersions were stable while NPs were agglomerated. Such a phenomenon could be applied to control the clustering of NPs in a medium.

Acknowledgments

The authors thank the technical support staff in the Department of Engineering, Tohoku University, for STEM observation. This research was supported by the Ministry of Education, Culture, Sports, Science and Technology: JSPS KAKENHI grants (19K15338, 20K21097, 21K14491, 22H01843, 23H00242) and Materials Processing Science project (“Materealize”) of MEXT, grant JPMXP0219192801. We thank Dr. Nozomi Watanabe (Osaka University) and Prof. Dr. Hiroshi Umakoshi (Osaka University) for their kind suggestions for this research.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.3c03593.

  • Experimental details (calculation methods for θ and ϕR, HSP evaluation, particle characterization); analytical scheme of dispersibility; TEM image and N2 adsorption/desorption isotherms of pristine SiO2; TGA analyses of SiO2@C8 (dprim = 132 nm, 35 nm) and SiO2@C12 (dprim = 150 nm); HSP spheres for SiO2@C8 (dprim = 58 nm); HSP spheres for SiO2@C8 (dprim = 132 nm) and SiO2@C12 (dprim = 150 nm); original and normalized Laurdan spectra; calibration curves for λ–ε correlation; Laurdan spectra for high-packing and loose-packing membranes; relationship between δ and ε for solvents; δ and ε values of solvents; HSP values of SiO2@C8, SiO2@C12, and modifiers; and dispersion stability and DLS measurement results for SiO2@C8 and SiO2@C12 (PDF)

Author Contributions

§ T.M. and S.S. equally contributed as first authors.

Author Contributions

T.M., S.S., K.S., and D.N designed the study. T.M., S.S., and K.S. performed the experiments. T.M, S.S., and K.S. wrote the first draft of the manuscript. K.S., K.W., T.A.J.W., and D.N. revised and supervised the manuscript. All authors discussed the results and approved the final version of the paper.

The authors declare no competing financial interest.

Supplementary Material

ac3c03593_si_001.pdf (1.9MB, pdf)

References

  1. Murphy C. J.; Sau T. K.; Gole A. M.; Orendorff C. J.; Gao J.; Gou L.; Hunyadi S. E.; Li T. Anisotropic Metal Nanoparticles: Synthesis, Assembly, and Optical Applications. J. Phys. Chem. B 2005, 109 (29), 13857–13870. 10.1021/jp0516846. [DOI] [PubMed] [Google Scholar]
  2. Torres Galvis H. M.; Bitter J. H.; Khare C. B.; Ruitenbeek M.; Dugulan A. I.; De Jong K. P. Supported Iron Nanoparticles as Catalysts for Sustainable Production of Lower Olefins. Science (80-.). 2012, 335 (6070), 835–838. 10.1126/science.1215614. [DOI] [PubMed] [Google Scholar]
  3. Wang F.; Feng L.; Qin Y.; Zhao T.; Luo H.; Zhu J. Dual Functional SiO2@TiO2 Photonic Crystals for Dazzling Structural Colors and Enhanced Photocatalytic Activity. J. Mater. Chem. C 2019, 7 (38), 11972–11983. 10.1039/C9TC03426A. [DOI] [Google Scholar]
  4. Wang X.; Li Z.; Ding Y.; Wang K.; Xing Z.; Sun X.; Guo W.; Hong X.; Zhu X.; Liu Y. Enhanced Photothermal-Photodynamic Therapy for Glioma Based on near-Infrared Dye Functionalized Fe3O4 Superparticles. Chem. Eng. J. 2020, 381, 122693 10.1016/j.cej.2019.122693. [DOI] [Google Scholar]
  5. Yoko A.; Naito H.; Seong G.; Tomai T.; Adschiri T. Nucleation and Coalescence of BaTiO3Using a Continuous Flow Reactor with Water-Ethanol Mixed Solvents. J. Phys. Chem. C 2021, 125 (35), 19489–19496. 10.1021/acs.jpcc.1c04914. [DOI] [Google Scholar]
  6. Zhang Q.; Huang W.-X.; Zhong G.-J. Towards Transparent PMMA/SiO2 Nanocomposites with Promising Scratch-Resistance by Manipulation of SiO2 Aggregation Followed by in Situ Polymerization. J. Appl. Polym. Sci. 2017, 134, 44612. 10.1002/app.44612. [DOI] [Google Scholar]
  7. Kim H.; Kim H. G.; Kim S.; Kim S. S. PDMS-Silica Composite Membranes with Silane Coupling for Propylene Separation. J. Membr. Sci. 2009, 344 (1–2), 211–218. 10.1016/j.memsci.2009.08.004. [DOI] [Google Scholar]
  8. Liu J.; Shi B.; Su Y.; Li X.; Zhang Z. Preparation of Reactive Nano-Silica/Polydimethylsiloxane Composites with Multiple Crosslinked Network Structure and Investigation of Their Mechanical Properties. Compos. Commun. 2020, 19, 37–41. 10.1016/j.coco.2020.02.003. [DOI] [Google Scholar]
  9. Liu J.; Yao Y.; Chen S.; Li X.; Zhang Z. A New Nanoparticle-Reinforced Silicone Rubber Composite Integrating High Strength and Strong Adhesion. Compos. Part A Appl. Sci. Manuf. 2021, 151, 106645 10.1016/j.compositesa.2021.106645. [DOI] [Google Scholar]
  10. Khanbabaei G.; Vasheghani-Farahani E.; Rahmatpour A. Pure and Mixed Gas CH4 and N-C 4H10 Permeation in PDMS-Fumed Silica Nanocomposite Membranes. Chem. Eng. J. 2012, 191, 369–377. 10.1016/j.cej.2012.03.033. [DOI] [Google Scholar]
  11. Kudo Y.; Mikami H.; Tanaka M.; Isaji T.; Odaka K.; Yamato M.; Kawakami H. Mixed Matrix Membranes Comprising a Polymer of Intrinsic Microporosity Loaded with Surface-Modified Non-Porous Pearl-Necklace Nanoparticles. J. Membr. Sci. 2020, 597, 117627 10.1016/j.memsci.2019.117627. [DOI] [Google Scholar]
  12. Wang J.; Wang G.; Zhang Z.; Ouyang G.; Hao Z. Effects of Mesoporous Silica Particle Size and Pore Structure on the Performance of Polymer-Mesoporous Silica Mixed Matrix Membranes. RSC Adv. 2021, 11 (58), 36577–36586. 10.1039/D1RA05125C. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Mathioudaki S.; Barthélémy B.; Detriche S.; Vandenabeele C.; Delhalle J.; Mekhalif Z.; Lucas S. Plasma Treatment of Metal Oxide Nanoparticles: Development of Core-Shell Structures for a Better and Similar Dispersibility. ACS Appl. Nano Mater. 2018, 1 (7), 3464–3473. 10.1021/acsanm.8b00645. [DOI] [Google Scholar]
  14. Stauch C.; Süß S.; Luxenhofer R.; Binks B. P.; Segets D.; Mandel K. Quantifying Surface Properties of Silica Particles by Combining Hansen Parameters and Reichardt’s Dye Indicator Data. Part. Part. Syst. Charact. 2018, 35 (12), 1800328. 10.1002/ppsc.201800328. [DOI] [Google Scholar]
  15. Wieneke J. U.; Kommoß B.; Gaer O.; Prykhodko I.; Ulbricht M. Systematic Investigation of Dispersions of Unmodified Inorganic Nanoparticles in Organic Solvents with Focus on the Hansen Solubility Parameters. Ind. Eng. Chem. Res. 2012, 51 (1), 327–334. 10.1021/ie201973u. [DOI] [Google Scholar]
  16. Fujiwara N.; Imai S.; Yamamoto H. Evaluation of the Influence of Fine Particle Surface Modification with the Hansen Solubility Parameters. Mater. Chem. Phys. 2019, 229, 139–148. 10.1016/j.matchemphys.2019.02.091. [DOI] [Google Scholar]
  17. Xu Z.; Shen C.; Hou Y.; Gao H.; Sun S. Oleylamine as Both Reducing Agent and Stabilizer in a Facile Synthesis of Magnetite Nanoparticles. Chem. Mater. 2009, 21 (9), 1778–1780. 10.1021/cm802978z. [DOI] [Google Scholar]
  18. Saita S.; Takeda S.; Kawasaki H. Hansen Solubility Parameter Analysis on Dispersion of Oleylamine-Capped Silver Nanoinks and Their Sintered Film Morphology. Nanomaterials 2022, 12 (12), 2004. 10.3390/nano12122004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Suga K.; Yoshida T.; Ishii H.; Okamoto Y.; Nagao D.; Konno M.; Umakoshi H. Membrane Surface-Enhanced Raman Spectroscopy for Sensitive Detection of Molecular Behavior of Lipid Assemblies. Anal. Chem. 2015, 87 (9), 4772–4780. 10.1021/ac5048532. [DOI] [PubMed] [Google Scholar]
  20. Faried M.; Suga K.; Okamoto Y.; Shameli K.; Miyake M.; Umakoshi H. Membrane Surface-Enhanced Raman Spectroscopy for Cholesterol-Modified Lipid Systems: Effect of Gold Nanoparticle Size. ACS Omega 2019, 4 (9), 13687–13695. 10.1021/acsomega.9b01073. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Yamashita S.; Sudo T.; Kamiya H.; Okada Y. Colloidal Stability of TiO 2 Nanoparticles: The Roles of Phosphonate Ligand Length and Solution Temperature. Chem. - Eur. J. 2022, 28 (50), e202201560 10.1002/chem.202202558. [DOI] [PubMed] [Google Scholar]
  22. Tomai T.; Tajima N.; Kimura M.; Yoko A.; Seong G.; Adschiri T. Solvent Accommodation Effect on Dispersibility of Metal Oxide Nanoparticle with Chemisorbed Organic Shell. J. Colloid Interface Sci. 2021, 587, 574–580. 10.1016/j.jcis.2020.11.014. [DOI] [PubMed] [Google Scholar]
  23. Ait Laasri H.; Fasquelle D.; Tachafine A.; Carru J. C.; Rguiti M.; Elaatmani M. Ferroelectric BT–PVDF Composite Thick Films for Electrical Energy Storage. J. Electron. Mater. 2021, 50 (3), 1132–1139. 10.1007/s11664-020-08657-y. [DOI] [Google Scholar]
  24. Zheng M. S.; Zheng Y. T.; Zha J. W.; Yang Y.; Han P.; Wen Y. Q.; Dang Z. M. Improved Dielectric, Tensile and Energy Storage Properties of Surface Rubberized BaTiO3/Polypropylene Nanocomposites. Nano Energy 2018, 48, 144–151. 10.1016/j.nanoen.2018.03.049. [DOI] [Google Scholar]
  25. Hu P.; Gao S.; Zhang Y.; Zhang L.; Wang C. Surface Modified BaTiO3 Nanoparticles by Titanate Coupling Agent Induce Significantly Enhanced Breakdown Strength and Larger Energy Density in PVDF Nanocomposite. Compos. Sci. Technol. 2018, 156, 109–116. 10.1016/j.compscitech.2017.12.025. [DOI] [Google Scholar]
  26. Sato M.; Ishii H.; Sueda Y.; Watanabe K.; Nagao D. A Reinforced, High-κ Ternary Polymer Nanocomposite Dielectrics of PVDF, Barium Titanate Nanoparticles, and TEMPO-Oxidized Cellulose Nanofibers. Compos. Part C Open Access 2021, 5, 100163 10.1016/j.jcomc.2021.100163. [DOI] [Google Scholar]
  27. Iijima M.; Kobayakawa M.; Yamazaki M.; Ohta Y.; Kamiya H. Anionic Surfactant with Hydrophobic and Hydrophilic Chains for Nanoparticle Dispersion and Shape Memory Polymer Nanocomposites. J. Am. Chem. Soc. 2009, 131 (45), 16342–16343. 10.1021/ja906655r. [DOI] [PubMed] [Google Scholar]
  28. Iijima M.; Kobayakawa M.; Kamiya H. Tuning the Stability of TiO2 Nanoparticles in Various Solvents by Mixed Silane Alkoxides. J. Colloid Interface Sci. 2009, 337 (1), 61–65. 10.1016/j.jcis.2009.05.007. [DOI] [PubMed] [Google Scholar]
  29. Bapat S.; Kilian S. O.; Wiggers H.; Segets D. Towards a Framework for Evaluating and Reporting Hansen Solubility Parameters: Applications to Particle Dispersions. Nanoscale Adv. 2021, 3 (15), 4400–4410. 10.1039/D1NA00405K. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Fujiwara N.; Nishida T.; Yamamoto H. Adaptation of Hansen Solubility Parameter in Evaluating Transparency of Composite Materials. Heliyon 2019, 5 (12), e02833 10.1016/j.heliyon.2019.e02833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Park Y.; Jeong D.; Jeong U.; Park H.; Yoon S.; Kang M.; Kim D. Polarity Nano-Mapping of Polymer Film Using Spectrally Resolved Super-Resolution Imaging. ACS Appl. Mater. Interfaces 2022, 14 (40), 46032–46042. 10.1021/acsami.2c11958. [DOI] [PubMed] [Google Scholar]
  32. Parasassi T.; Gratton E. Membrane Lipid Domains and Dynamics as Detected by Laurdan Fluorescence. J. Fluoresc. 1995, 5 (1), 59–69. 10.1007/BF00718783. [DOI] [PubMed] [Google Scholar]
  33. Watanabe N.; Goto Y.; Suga K.; Nyholm T. K. M.; Slotte J. P.; Umakoshi H. Solvatochromic Modeling of Laurdan for Multiple Polarity Analysis of Dihydrosphingomyelin Bilayer. Biophys. J. 2019, 116 (5), 874–883. 10.1016/j.bpj.2019.01.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Watanabe N.; Suga K.; Slotte J. P.; Nyholm T. K. M.; Umakoshi H. Lipid-Surrounding Water Molecules Probed by Time-Resolved Emission Spectra of Laurdan. Langmuir 2019, 35 (20), 6762–6770. 10.1021/acs.langmuir.9b00303. [DOI] [PubMed] [Google Scholar]
  35. Wakileh W.; Watanabe N.; Suga K.; Ikushima N.; Kajimura N.; Mitsuoka K.; Okamoto Y.; Umakoshi H. Dispersibility and Surface Properties of Hydrocortisone-Incorporated Self-Assemblies. Colloids Surfaces A Physicochem. Eng. Asp. 2023, 665, 131217 10.1016/j.colsurfa.2023.131217. [DOI] [Google Scholar]
  36. Stöber W.; Fink A.; Bohn E. Controlled Growth of Monodisperse Silica Spheres in the Micron Size Range. J. Colloid Interface Sci. 1968, 26 (1), 62–69. 10.1016/0021-9797(68)90272-5. [DOI] [Google Scholar]
  37. Watanabe K.; Kuroda K.; Nagao D. Polyethylenimine-Assisted Synthesis of Hollow Silica Spheres without Shape Deformation. Mater. Chem. Phys. 2021, 262, 124267 10.1016/j.matchemphys.2021.124267. [DOI] [Google Scholar]
  38. Guo W.; Guo X.; Li J.; Li Y.; Sun M.; Dai W. Assessing the Effect of Nano Hydrophobic Silane Silica on Aggregate-Bitumen Interface Bond Strength in the Spring-Thaw Season. Appl. Sci. 2019, 9 (12), 2393. 10.3390/app9122393. [DOI] [Google Scholar]
  39. Isoda T.; Maeda R. Development of an Interaction Assay between Single-Stranded Nucleic Acids Trapped with Silica Particles and Fluorescent Compounds. J. Funct. Biomater. 2012, 3 (3), 601–614. 10.3390/jfb3030601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wang Z.; Liu M. C.; Chang Z. Y.; Li H. B. Study on the Graft Modification Mechanism of Macroporous Silica Gel Surface Based on Silane Coupling Agent Vinyl Triethoxysilane. RSC Adv. 2021, 11 (41), 25158–25169. 10.1039/D1RA04296C. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Arita T.; Ueda Y.; Minami K.; Naka T.; Adschiri T. Dispersion of Fatty Acid Surface Modified Ceria Nanocrystals in Various Organic Solvents. Ind. Eng. Chem. Res. 2010, 49 (4), 1947–1952. 10.1021/ie901319c. [DOI] [Google Scholar]
  42. Arita T.; Yoo J.; Adschiri T. Relation between the Solution-State Behavior of Self-Assembled Monolayers on Nanoparticles and Dispersion of Nanoparticles in Organic Solvents. J. Phys. Chem. C 2011, 115 (10), 3899–3909. 10.1021/jp110172s. [DOI] [Google Scholar]
  43. Usune S.; Ando M.; Kubo M.; Tsukada T.; Sugioka K.-I.; Koike O.; Tatsumi R.; Fujita M.; Takami S.; Adschiri T. Numerical Simulation of Dispersion and Aggregation Behavior of Surface-Modified Nanoparticles in Organic Solvents. J. Chem. Eng. JAPAN 2018, 51 (6), 492–500. 10.1252/jcej.17we197. [DOI] [Google Scholar]
  44. Zhuravlev L. T. Concentration of Hydroxyl Groups on the Surface of Amorphous Silicas. Langmuir 1987, 3 (3), 316–318. 10.1021/la00075a004. [DOI] [Google Scholar]
  45. Rahman I. A.; Vejayakumaran P.; Sipaut C. S.; Ismail J.; Chee C. K. Size-Dependent Physicochemical and Optical Properties of Silica Nanoparticles. Mater. Chem. Phys. 2009, 114 (1), 328–332. 10.1016/j.matchemphys.2008.09.068. [DOI] [Google Scholar]
  46. Moriyoshi T.; Ishii T.; Tamai Y.; Tado M. Static Dielectric Constants of Water + Ethanol and Water + 2-Methyl-2-Propanol Mixtures from 0.1 to 300 MPa at 298.15 K. J. Chem. Eng. Data 1990, 35 (1), 17–20. 10.1021/je00059a005. [DOI] [Google Scholar]
  47. Viard M.; Gallay J.; Vincent M.; Meyer O.; Robert B.; Paternostre M. Laurdan Solvatochromism: Solvent Dielectric Relaxation and Intramolecular Excited-State Reaction. Biophys. J. 1997, 73 (4), 2221–2234. 10.1016/S0006-3495(97)78253-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Lysenko S. N.; Astaf’Eva S. A.; Kornilitsina E. V.; Yakusheva D. E.; Morozov K. I. Osmotic Attraction: A New Mechanism of Nanoparticle Aggregation. Langmuir 2022, 38 (46), 14313–14325. 10.1021/acs.langmuir.2c02379. [DOI] [PubMed] [Google Scholar]
  49. Watanabe K.; Tajima Y.; Shimura T.; Ishii H.; Nagao D. Depletion-Interaction-Driven Assembly of Golf Ball-like Particles for Development of Colloidal Macromolecules. J. Colloid Interface Sci. 2019, 534, 81–87. 10.1016/j.jcis.2018.08.117. [DOI] [PubMed] [Google Scholar]
  50. Yang Y. J.; Corti D. S.; Franses E. I. Effect of Triton X-100 on the Stability of Titania Nanoparticles against Agglomeration and Sedimentation: A Masked Depletion Interaction. Colloids Surfaces A Physicochem. Eng. Asp. 2017, 516, 296–304. 10.1016/j.colsurfa.2016.12.026. [DOI] [Google Scholar]
  51. Dederichs T.; Möller M.; Weichold O. Temperature-Dependent Colloidal Stability of Hydrophobic Nanoparticles Caused by Surfactant Adsorption/Desorption and Depletion Flocculation. Langmuir 2009, 25 (18), 10501–10506. 10.1021/la901216g. [DOI] [PubMed] [Google Scholar]
  52. Parasassi T.; De Stasio G.; Ravagnan G.; Rusch R. M.; Gratton E. Quantitation of Lipid Phases in Phospholipid Vesicles by the Generalized Polarization of Laurdan Fluorescence. Biophys. J. 1991, 60 (1), 179–189. 10.1016/S0006-3495(91)82041-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Suga K.; Otsuka Y.; Okamoto Y.; Umakoshi H. Gel-Phase-like Ordered Membrane Properties Observed in Dispersed Oleic Acid/1-Oleoylglycerol Self-Assemblies: Systematic Characterization Using Raman Spectroscopy and a Laurdan Fluorescent Probe. Langmuir 2018, 34 (5), 2081–2088. 10.1021/acs.langmuir.7b04044. [DOI] [PubMed] [Google Scholar]
  54. Izza N.; Suga K.; Okamoto Y.; Watanabe N.; Bui T. T.; Wibisono Y.; Fadila C. R.; Umakoshi H. Systematic Characterization of Nanostructured Lipid Carriers from Cetyl Palmitate/Caprylic Triglyceride/Tween 80 Mixtures in an Aqueous Environment. Langmuir 2021, 37 (14), 4284–4293. 10.1021/acs.langmuir.1c00270. [DOI] [PubMed] [Google Scholar]
  55. Yeh V.; Goode A.; Johnson D.; Cowieson N.; Bonev B. B. The Role of Lipid Chains as Determinants of Membrane Stability in the Presence of Styrene. Langmuir 2022, 38 (4), 1348–1359. 10.1021/acs.langmuir.1c02332. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ac3c03593_si_001.pdf (1.9MB, pdf)

Articles from Analytical Chemistry are provided here courtesy of American Chemical Society

RESOURCES