Use of Pulsed Streaming Potential with a Prepared Cationic

Oct 4, 2016 - ... https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... Streaming potential (SP) measurement is a traditional electr...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Use of Pulsed Streaming Potential with a Prepared Cationic Polyelectrolyte Layer to Detect Deposition Kinetics of Graphene Oxide and Consequences of Particle Size Differences Lei Zhao,† Lizhi Zhao,† Shenghong Yang,† Xianglu Peng,† Jing Wu,† Lei Bian,† Xiayan Wang,‡ and Qiaosheng Pu*,† †

State Key Laboratory of Applied Organic Chemistry, Key Laboratory of Nonferrous Metals Chemistry and Resources Utilization of Gansu Province, Department of Chemistry, Lanzhou University, Lanzhou, Gansu 730000, China ‡ Beijing Key Laboratory for Green Catalysis and Separation, Department of Chemistry and Chemical Engineering, Beijing University of Technology, Beijing 100124, China S Supporting Information *

ABSTRACT: The deposition kinetics of graphene oxide (GO) onto poly(ethylene imine) (PEI) layer was characterized in situ with pulsed streaming potential (SP) measurement, and it was found that the initial rate constant (ki) was dependent on the size of GO with same surface charge density at a fixed concentration under controlled experimental conditions. Assuming the deposition was controlled by diffusion at the initial stage, ki is proportional to Rh−2/3, where Rh is the hydrodynamic radius. By flushing a GO solution through a capillary coated with PEI, the initial change rate of relative SP (dEr/dt) was obtained in 20 s and ki was measured with five different concentrations in about 2 min. Three GO samples of different sizes obtained from the same batch of raw material were characterized with pulsed SP to get ki values, and their sizes were verified with atomic force microscopy and dynamic light scattering. The experimental results are consistent with the predicted effects of the size of NPs on their deposition kinetics.

S

protein characterization can be used for NPs. Through monitoring the adsorption kinetics of proteins onto the surface of flow cells, Norde and Rouwendal1 showed that the adsorption rate at the initial stage was proportional to the diffusion coefficient of the proteins. Adamczyk et al.8 obtained similar conclusions and they showed that under diffusion controlled transport, based on the random sequential adsorption model, the mass transport rate in a cylindrical channel was proportional to the diffusion coefficient. Because the sizes of either proteins or synthetic polyelectrolytes are comparable to NPs, it is reasonable to assume that these conclusions are valid for NPs. According to the Stokes− Einstein equation,32,33 the diffusion coefficient is dependent on the hydrodynamic radius of particles, the deposition kinetics of NPs onto a polymer layer is therefore related to their sizes. The dependence of the deposition kinetics of NPs to oppositely charged surfaces on their sizes has also been confirmed by optical reflectometry34 and quartz crystal microbalance methods.35 Because of its simplicity, pulsed SP measurement can be a facial way to get quantitative information on deposition kinetics of NPs onto well controlled surfaces.

treaming potential (SP) measurement is a traditional electrokinetic technique for characterizing surfaces. Its application has been extended to in situ determination of the deposition kinetics of proteins,1−3 polyelectrolytes,4−7 and certain colloids onto various surfaces and the stability of the formed layers.8,9 Deposition of nanoparticles (NPs) and their monolayers on polyelectrolyte layers were also described by the Adamczyk group.10−13 The effort to use SP measurement as an analytical tool was made several decades ago. Mattiasson et al. proved that SP measurement could be a tool to quantify glycoproteins over an affinity column14 and to monitor hybridoma cell cultivations.15 The measurement can be much faster when a single microfluidic channel is used together with pulsed driving pressure16−19 due to the unique flow pattern inside the microchannels. The advantages of the SP measurement include the simple device, easy operation, and possibility of online monitoring, especially when pulsed SP measurement technique is adopted. Further application of this technique in nanomaterials can be reasonably expected. NPs have been extensively explored in many fields due to their excellent physical and chemical properties. Their applications cover areas such as catalysis,20,21 sensors,22−24 separation,25,26 biomedicine and biotechnology,27−29 and environmental industry.30,31 Because of the similarity of the sizes of NPs and proteins, some techniques that are used for © XXXX American Chemical Society

Received: June 17, 2016 Accepted: October 4, 2016

A

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

S1, S2, and S3 were freeze-dried on a Scientz-12N vertical freeze-drying machine (NingBo Scientz Biotechnology Co., Ltd., Ningbo, China) and then ultrasonically dispersed (100 W, 30 min) in deionized water with a concentration of 1.0 mg mL−1 as stock suspension solutions. Characterization of GO Samples. The atomic force microscopy (AFM) study was conducted using a NanoScope IIIa MultiMode AFM (Bruker Corporation, England) in tapping mode with a drop of GO solution on a freshly cleaned silicon plate. Dynamic light scattering (DLS) measurements were performed on a BI-200SM goniometer of Brookhaven Instruments (Holtsville, NY) with 10.0 μg mL−1 GO in 1.25 mM PB (pH = 7.2) solution at room temperature to get the size distribution of GO. The surface charge of GO was titrated with poly(diallyldimethylammonium chloride) (poly-DADMAC, 0.0001 n, supplied by AFG Analytic Gmbh, Germany) using a Mütek PCD-03 particle charge detector (BTG Instruments Gmbh, Germany) with 40.0 μg mL−1 GO in 1.25 mM PB (the total volume is 10.00 mL). Pulsed SP Measurement. The SP measurement was similar to that described earlier.19,49 The schematic illustration of pulsed SP device is shown in the Supporting Information (Figure S1). The operation is briefly described as follows: capillary (3.0 cm long, 75 μm i.d. × 365 μm o.d. with polyimide cladding) was glued through a hole at the bottom of a 0.5 mL centrifuge vial (2 mm into the vial). For the measurement, the centrifuge vial with capillary was attached to a push-to-connect adapter (diameter 8 mm, Delixi-air, Leqing, China) which was equipped with a platinum wire as a electrode. The other end of the push-to-connect adapter was connected to the vacuum source through a 3-way solenoid valve (WTB-3R-N4F12VDC, Takasago Electronics, Suzhou, China) that was controlled through the digital output of a USB-6009 data acquisition card (National Instruments, Austin, TX). The free end of the capillary was immersed into the solution (measuring buffer or coating solution) in another 0.5 mL centrifuge vial with another platinum wire. The potential difference between two platinum electrodes was fed through a voltage follower to the USB-6009 data acquisition card which was plugged into a computer for measuring SP value E. During the measurement of SP, the applied pressure was pulsed by switching the solenoid valve to get rid of the interference that is caused by the electrochemical drift from electrodes, which is referred to as reference potential. When the valve was switched to turn on the vacuum (2 s) the solution was sucked through the capillary, the measured potential is a sum of SP and the reference potential. When the valve was switched to turn off the vacuum, there was no flow in the capillary, only the reference potential was measured. The SP value was obtained through subtracting the potential value when the valve was turned off from that when the valve was turned on automatically by a program written in Labview (National Instruments, Austin, TX). All measurements were performed at room temperature. Formation of PEI Layer onto Bare Silica Capillary. Before the formation of the PEI layer, the bare silica capillary (3.0 cm long, i.d. 75 μm) was rinsed with deionized water, 1 M NaOH and deionized water sequentially for 20, 40, and 20 s. After washing, the bare silica capillary was treated as follows to get the PEI layer: (1) The bare silica capillary was flushed with 1.25 mM PB (pH = 7.2 ± 0.2) for 20 s under a pressure drop of 0.068 MPa to stabilize the surface charge. (2) The capillary was then flushed with PEI (MW 70 kDa, dissolved in 1.25 mM PB and the concentration was 200.0 μg mL−1) solution for 80 s

Unlike other engineered NPs, graphene oxide (GO), as a derivative of graphene, has the form of a nanosheet and has been used in many fields including drug delivery,36 cancer treatment,37 cell growth control,38,39 FET/FRET sensors,40,41 and mass spectrometry as an affinity extraction and detection platform.42 It is also a raw material for preparing graphene by chemical reduction.43,44 The lateral dimensions of GO sheets play an important role in controlling their properties and applications; larger sized GOs can be used to prepare threedimensional networks,45 and smaller sized GOs have been used in drug delivery and biosensing.46 Because of the reactive oxygen functional groups, such as epoxide, hydroxyl, and carboxyl on their basal planes and edges,47 the surface charge of GO is negative, so it can obtain stable GO suspensions of various concentrations, which is important for the measurement of the deposition kinetics of NPs. With the aid of the deposition kinetics of GOs of different sizes with uniform surface characteristics onto cationic polyelectrolyte, consequences of particle size difference on the deposition behavior can be quantitatively investigated. In the present work, the deposition kinetics of GO onto cationic polyelectrolyte layer within fused silica capillary was systematically detected. Deposition rate constant and initial rate constant of GO on poly(ethylene imine) (PEI) layer were measured, and the deposition kinetics of GO at the initial stage has been shown to be diffusion controlled. With the rapid determination of deposition kinetics of three GOs of different sizes onto PEI layer, the correlation between the initial rate constant ki and the hydrodynamic radius of GOs was evaluated. The correlation was also demonstrated with silica NPs of different sizes.

■

EXPERIMENTAL SECTION Chemicals and Materials. Poly(ethylene imine) (PEI, 50% in water) with a molecular weight of 70 kDa was purchased from Aladdin (Shanghai, China). Graphene oxide (GO, 6.2 mg mL−1 in water) was purchased from Graphene Laboratories Inc. (Calverton, NY). KH2PO4 and NaOH were purchased from Tianjin Guangfu Technology Development Co., Ltd. (Tianjin, China). All other reagents used were analytical pure or above. The fused silica capillary (i.d. 75 μm, o.d. 365 μm) was purchased from Sino Sumtech Co., Ltd. (Handan, China). Buffer solutions used for the SP measurement were diluted from a 0.2 M KH2PO4 stock solution and adjusted pH with proper concentration of NaOH if necessary. All solutions were prepared with deionized water from a MilliQ water purification system (Millipore). Preparation of GO of Different Sizes. GO nanosheets of three different sizes were obtained through repeated centrifugation-dispersion from the same batch of GO (referred to as S0) to avoid any inconsistency:48 (1) GO aqueous solution with a concentration of 0.1 mg mL−1 was prepared by diluting S0 (original concentration is 6.2 mg mL−1) with deionized water, and then the solid in the solution which could not be dispersed was separated by centrifugation at 5000 rpm for 5 min. (2) The aforementioned GO aqueous solution was centrifuged at 12 000 rpm for 40 min, the supernatant was collected and taken as S1. (3) The precipitate at the bottom of the centrifuge tube in procedure 2 was resuspended with deionized water for further centrifugation at 8000 rpm for 30 min, the supernatant was collected as S2. (4) The precipitate at the bottom of the centrifuge tube in procedure 3 was redispersed and taken as S3. After preparation, the samples of B

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

difference between the values obtained when pressure on and off represents the pure SP. With the pressure pulsing across a piece of capillary (3.0−20.0 cm), the measurement can be made within 2−5 s using a broad range of buffers and additives through inert electrodes such as platinum. The deposition kinetics can be obtained in less than 60 s. The deposition kinetics can tell a lot about either the depositing substances or the surfaces. Norde and Rouwendal1 have proposed a model for the proteins, with a low protein concentration, during the initial stage of the adsorption process, the adsorption rate can be expressed as

under the same pressure drop of 0.068 MPa. (3) After PEI flushing, the capillary was rinsed with blank buffer solution (1.25 mM PB, without PEI, same as that used for SP measurement) for 20 s under the same pressure drop of 0.068 MPa to remove the loosely attached PEI, then the SP value of the formed PEI layer was measured. Typical SP values changing along with time during the preparation of PEI layer are shown in the Supporting Information (Figure S2). Determination of Relative SP. Luna-Vera and Alvarez have used SP to monitor the adsorption kinetics of lysozyme (LYZ) onto blank surfaces.18 In the present experiment, relative SP (Er) was used to monitor the deposition kinetics of GO onto the PEI layer instead of SP to overcome the inconsistency among different capillaries. Er was calculated through Et/Eb, where Et is the SP value of capillary with GO deposited onto a PEI layer at a given time and Eb is the SP value of the capillary with the PEI layer. Figure 1 shows a typical curve of Er along with the time of the flushing of GO (S0) solution.

⎛ γ ⎞1/3 dΓ(y) = 0.81⎜ ⎟ Dc dt ⎝ yD ⎠

(1)

where γ is the shear rate, D is diffusion coefficient, c is the concentration of the protein, and y is a distance from the entrance. If other conditions are kept same, the adsorption rate is proportional to D2/3. Adamczyk et al.8 proposed an equation for the mass transport rate under diffusion controlled transport based on a random sequential adsorption model, which is also proportional to D2/3. Because of the similarity between proteins molecules and NPs, the relationship is expected to be applicable to NPs. Because of the distinct relationship between the diffusion coefficient and particle size (Stokes−Einstein equation) under a given condition,33 it is possible to account for the observed dependency of particle size of NPs on deposition kinetics. To show it, GO of different sizes were selected as a model. It should be mentioned that GOs are in a form of nanosheets, the radii obtained from the Stokes−Einstein equation are not the actual sizes, because this equation is frequently used for demonstrate the size of GO using DLS50,51 and it can be taken as an orientation averaged size of the nanosheet. On the basis of the theoretical analysis mentioned above, the deposition rate should be proportional to Rh−2/3. Characterization of GO of Different Sizes. GOs of different sizes that used for pulsed SP measurement were obtained through centrifugation of the same batch of GO to avoid any inconsistency caused by preparation procedure or undefined additives. Before characterization with pulsed SP measurement, AFM and DLS were used to verify the sizes and morphology of the prepared GO samples. The AFM images of three GO samples of different sizes (S1, S2, and S3) are shown in Figure 2a−c, and the results indicated that the thickness of GO were around 2 nm but the sizes of the GO flakes are different. Figure 2d shows the histogram of size distributions by analyzing seven sheets. GO supernatant obtained at higher centrifugation speed (S1, 12 000 rpm) shows a quite small average size of 458 nm, while S2 and S3 give average sizes of 993 and 1345 nm, respectively. The size distribution and morphology are consistent with that reported previously.48 DLS is a widely used method to measure the size and its distribution of spherical particles. Although it cannot give the actual size of GO due to its sheet-like shape, this technique was used to characterize the size of GO.50,51 As shown in Figure 3, the particle size distribution is approximately Gaussian. The average diameters were 242 ± 23 nm, 539 ± 10 nm, and 922 ± 34 nm for S1, S2, and S3, respectively. Because of the difference of the measuring mechanism of these techniques, size is slightly different. With DLS method, we actually obtained the equivalent diameters of GO flakes in

Figure 1. Example of Er change along with the time of the flushing of GO (S0). The concentration of GO (S0) is 124 μg mL−1 in PB, conductivity of the solution was 220 ± 10 μS cm−1, pH = 7.2 ± 0.2, ΔP = −0.068 MPa.

As shown in Figure 1, Er changed significantly when the solution of GO was flushed through the capillary coated with PEI layer. The initial time variation of Er (dEr/dt) could be determined within initial 20 s, as shown in the inset of Figure 1. Because dEr/dt is dependent on the properties of adsorbed substance,37 the properties of substances related to dEr/dt can therefore be characterized. In SP measurement, two different sets of GO solutions were prepared. One set contained GO (S0) at 62, 124, 186, 248, 310 μg mL−1 in 1.25 mM PB (conductivity is 220 ± 10 μS cm−1, pH = 7.2 ± 0.2), and the other set contained GO of different sizes (S1, S2, and S3) at 10.0−50.0 μg mL−1 in 1.25 mM PB (conductivity is 220 ± 10 μS cm−1, pH = 7.2 ± 0.2).

■

RESULTS AND DISCUSSION Rationale for Dependency of SP Results on Particle Size. The goal of the present work is to better understand deposition kinetics of NPs, as affected by their particle size. Different from the traditional SP measurement with continuous pressure-driven flow, pulsed SP measurement intermittently switches on and off the applied pressure, so that SP is only produced when the pressure is switched on and it is measured together with the reference potential. When the pressure is switched off, only the reference potential was measured. The C

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 2. (a−c) AFM images of GO samples: (a) S1, (b) S2, (c) S3, and (d) the histograms of size distribution of GO samples.

Figure 4. Time dependence of Er for GO (S0) with different concentrations deposited onto PEI layer. Conductivity of all solutions was 220 ± 10 μS cm−1, ΔP = −0.068 MPa, pH = 7.2 ± 0.2, solid lines represent the curve fit to eq 2, 0.969 ≤ R2 ≤ 0.992.

Figure 3. Particle size distribution of three GOs (S1, S2, and S3) monitored by the DLS method.

Table 1. Surface Charge of GOs Titrated with polyDADMAC GO

S1

S2

S3

q (× 10−4, eq g−1)

2.60 ± 0.08

2.79 ± 0.19

2.86 ± 0.12

with the SP method is the relatively stable surface charge of these particles. There are some literature evidence to show that surface charge of GOs with difference sizes does not vary significantly.53 To get clear clue, the surface charge of three GOs was tested with titrated with positively charged polyDADMAC (0.0001 n), results are listed in Table 1. The variation of the surface charge was within 10%. Deposition Kinetics of GO onto PEI Layer. SP measurement has been used for studying particle deposition or desorption processes with moderate surface coverage.9 In the present work, we tried to use pulsed SP for fast characterization of the deposition kinetics of GO flakes onto PEI layer and to

aqueous solution, while in AFM the size is actually the dimensions of the tiled GO flakes. Because of the flake morphology of GO, strictly, neither of them was the actual size of GO,52 but they can all be used to indicate the size difference of GOs. SP is directly dependent on the surface charge, a prerequisite for comparing the deposition kinetics of NPs of different sizes D

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry Table 2. Values of ka[GO] + kd for Different GO (S0) Concentrations by Fitting to Equation 2 CGO (μg mL−1)

62

128

186

248

310

ka[GO] + kd

7.84 × 10−3

2.79 × 10−2

3.28 × 10−2

5.09 × 10−2

6.64 × 10−2

Figure 5. (a) Time response of Er of GO (S0) onto PEI layer within the initial 20 s, the solid lines represent linear fit, 0.950 ≤ R2 ≤ 0.998, (b) the linear relationship of time variation of Er with the concentration of GO (S0), solid line represents fit to eq 3. Conductivity of all solutions was 220 ± 10 μS cm−1, ΔP = −0.068 MPa, pH = 7.2 ± 0.2.

Table 3. Radii of Silica NPs Obtained from TEM and DLS and the Initial Rate Constants ki Obtained from the Pulsed SP Measurement silica NPs

SS1

SS2

SS3

R (nm, TEM) R (nm, DLS) ki (L g−1 s−1)

13 ± 2 29 ± 1 0.90 ± 0.056

36 ± 5 66 ± 3 0.21 ± 0.059

65 ± 6 105 ± 1 0.091 ± 0.0079

determine the correlation between the size and the deposition kinetics of GO. If the deposition of GO is controlled by the mass transport, eq 2 can be used to describe the change of measured Er:54 Er = E∞(1 − e−(ka[GO] + kd)t ) + E0

Figure 6. Linear relationship between dEr/dt and the concentration of GO of different sizes (S1, S2, and S3), solid lines represent fit to eq 3, conductivity of all solutions was 220 ± 10 μS cm−1, ΔP = −0.068 MPa, pH = 7.2 ± 0.2.

(2)

where E∞ is relative SP at equilibrium (t = ∞), E0 is relative SP at the beginning, ka is the deposition rate constant, kd is the desorption rate constant. During the deposition of GO (S0) onto the PEI layer, the value of Er is getting smaller and gradually become negative (Figure 4). The time to attain the deposition equilibrium became shorter when the concentration of GO increased. ka[GO] + kd values were obtained through curve fitting according to eq 2. The values for five were listed in Table 2, and the calculated deposition rate constant ka was 0.24 L g−1 s−1. This deposition behavior was also checked on poly-DADMAC layer (shown in Figure S3), there was no evident difference compared with the results from the PEI layer, which implies that the surface of PEI layer under the selected condition is as stable as poly-DADMAC layer. These results indicated that the deposition rate constant can be derived from the pulsed SP measurement and used for characterization of GO easily. The good fitting of experimental data to eq 2 suggested that the deposition of GO onto the PEI layer could be described by the mass transport model, and similar results were obtained from the poly-DADMAC layer (Figure S3). With mass transport model, the signal at the initial stages of deposition is linearly related with the concentration of GO,18 which can be represented with eq 3.18,54

Figure 7. Correlation between the initial rate constants ki and the radii of GOs (S1, S2, and S3) obtained from the DLS method, and the solid line represents nonlinear curve fit, R2 = 0.994.

E

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 8. Correlation between the initial rate constants ki and (a) radii obtained from TEM method of silica NPs (SS1, SS2, and SS3) and (b) radii of silica NPs (SS1, SS2, and SS3) obtained from DLS method; solid lines represent nonlinear curve fit for (a) R2 = 0.996 and for (b) R2 = 0.993.

dE r = k i[GO] dt

from DLS method as shown in Figure 7. The equation can be described as ki = a(Rh + b)−2/3, where a and b are constants dependent on the conditions, they can be calibrated with the GO samples of known structures and sizes. The constants a and b are 22.98 nm2/3 L g−1 s−1 and 230.28 nm, R2 = 0.994. Because of the limitation of our size differentiation, only 3 GOs of different sizes were used here but the dependence of the initial rate constant ki on the hydrodynamic radius of GO is evident. The correlation between the deposition kinetics and the particle size is not only applicable to GO sheets. Silica NPs with spherical shape were also tested with the proposed procedure. The results of the size characterization and deposition of three different silica NPs (SS1, SS2, and SS3) are shown in the Supporting Information (Figures S5−S8). The initial rate constants for silica NPs of different sizes are listed in Table 3, these data indicated that the dependence of the initial rate constant ki on the particle size was also valid for silica NPs. The correlations between the initial rate constants of silica NPs and radii measured with TEM and DLS are illustrated in Figure 8. The correlation between the radius and initial rate constant ki can also be expressed as ki = a(Rh + b)−2/3. For radii obtained from DLS method, a is 2.23 nm2/3 L g−1 s−1, b is −25.09 nm, R2 = 0.993. For the radii obtained from TEM, a is 1.68 nm2/3 L g−1 s−1 and b is −9.94 nm, R2 = 0.996. The dependence of ki on the particle size of silica NPs is clear too.

(3)

where ki is the initial rate constant, it can be calculated from the data obtained in the initial 20 s (Figure 5). The change of Er along with time can be fitted linearly (Figure 5a), and the obtained dEr/dt (slope) values were plotted in Figure 5b. On the basis of eq 3, the initial rate constant ki was 0.37 ± 0.013 L g−1 s−1, slightly higher than that obtained with 60 s periods and curve fitting, 0.24 ± 0.011 L g−1 s−1 which was calculated from Figure 4. The surface charge of PEI layer can be titrated with GO too, the concentration of GO (S0) that gave zero SP in initial 20 s was 138.0 μg mL−1. It should also be mentioned that with the higher concentrations, the curves of Er vs time were not perfectly linear (Figure 5a), R2 for 62 μg mL−1 was 0.998, but R2 was only 0.950 for GO of 310 μg mL−1. To speed the measurement, we measured the Er within initial 20 s with GO concentration no higher than 50.0 μg mL−1 and eq 3 was used for the characterization of GOs of different sizes with pulsed SP measurement. Dependence of Deposition Kinetics on the Particle Size. On the basis of the above results, it could be concluded that the deposition kinetics of GO onto PEI layer was governed by mass transport, there is a linear relationship between Er and the time at the initial stage of deposition. Assuming mass transport is mainly diffusion controlled due to the laminar flow pattern inside the capillary, the relationship of deposition or mass transport rate and the diffusion coefficient (∝ D2/3) is also applicable to our system. The initial rate constant ki is related to the hydrodynamic radius Rh of NPs, as ki ∝ Rh−2/3. The deposition kinetics of GOs of different sizes (S1, S2, and S3) onto PEI layer were measured within 60 s (Figure S4) with the concentration from 10.0 to 50.0 μg mL −1 (Low concentrations was used to ensure the diffusion based transport), and the initial change rates dEr/dt (initial 20 s) of three GO samples of different sizes assembling onto PEI layers were measured as mentioned above. As shown in Figure 6, linear relationship between the initial change rate dEr/dt and the concentration of GO was obtained for all three GO samples, the initial rate constants (ki) were 0.46 ± 0.024, 0.37 ± 0.0043, and 0.29 ± 0.0029 L g−1 s−1 for S1, S2, and S3, respectively. And then we checked the correlation between the initial rate constant ki and the hydrodynamic radius Rh of GOs obtained

■

CONCLUSIONS On the basis of the linear change of Er along with time using diluted GO solutions, it can be deduced that the deposition kinetics of GO at the initial stage under low concentrations is mainly controlled by diffusion transport according to the model reported previously.1 The initial rate constant ki of GO onto PEI layer can be obtained with a series of GO solutions, the whole measurement was about 2 min if 5 different concentrations were used. There is no extra sample preparation process needed as long as the measurement condition was well controlled. In addition, the pulsed SP is an ideal method for online detection, and it can be a suitable way for the investigation the formation process of composites of proteins, water-soluble polymers, or their aggregates with NPs. It may also be used for monitoring the influencing factors of deposition kinetics of NPs such as surface charge and surface contamination. F

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

■

(16) Pu, Q.; Elazazy, M. S.; Alvarez, J. C. Anal. Chem. 2008, 80, 6532−6536. (17) Gupta, M. L.; Brunson, K.; Chakravorty, A.; Kurt, P.; Alvarez, J. C.; Luna-Vera, F.; Wynne, K. J. Langmuir 2010, 26, 9032−9039. (18) Luna-Vera, F.; Alvarez, J. C. Biosens. Bioelectron. 2010, 25, 1539−1543. (19) Wu, J.; Zhang, Y.; Zhao, L.; Zou, L.; Alvarez, J. C.; Pu, Q. Sens. Actuators, B 2012, 174, 18−23. (20) Crooks, R. M.; Zhao, M. Q.; Sun, L.; Chechik, V.; Yeung, L. K. Acc. Chem. Res. 2001, 34, 181−190. (21) Bell, A. T. Science 2003, 299, 1688−1691. (22) McFarland, A. D.; Van Duyne, R. P. Nano Lett. 2003, 3, 1057− 1062. (23) Wang, F. F.; Liu, S. Z.; Lin, M. X.; Chen, X.; Lin, S. R.; Du, X. Z.; Li, H.; Ye, H. B.; Qiu, B.; Lin, Z. Y.; Guo, L. H.; Chen, G. N. Biosens. Bioelectron. 2015, 68, 475−480. (24) Zhang, G. Y.; Deng, S. Y.; Cai, W. R.; Cosnier, S.; Zhang, X. J.; Shan, D. Anal. Chem. 2015, 87, 9093−9100. (25) Pumera, M.; Wang, J.; Grushka, E.; Polsky, R. Anal. Chem. 2001, 73, 5625−5628. (26) Yang, L.; Guihen, E.; Holmes, J. D.; Loughran, M.; O’Sullivan, G. P.; Glennon, J. D. Anal. Chem. 2005, 77, 1840−1846. (27) Guo, S. J.; Li, D.; Zhang, L. M.; Li, J.; Wang, E. K. Biomaterials 2009, 30, 1881−1889. (28) Mout, R.; Moyano, D. F.; Rana, S.; Rotello, V. M. Chem. Soc. Rev. 2012, 41, 2539−2544. (29) Berry, C. C. J. Phys. D: Appl. Phys. 2009, 42, 224003. (30) Wong, M. S.; Alvarez, P. J. J.; Fang, Y. L.; Akcin, N.; Nutt, M. O.; Miller, J. T.; Heck, K. N. J. Chem. Technol. Biotechnol. 2009, 84, 158−166. (31) Guo, J.; Wang, R. Y.; Tjiu, W. W.; Pan, J. S.; Liu, T. X. J. Hazard. Mater. 2012, 225-226, 63−73. (32) Leung, A. B.; Suh, K. I.; Ansari, R. R. Appl. Opt. 2006, 45, 2186− 2190. (33) Lattuada, M.; Sandkuhler, P.; Wu, H.; Sefcik, J.; Morbidelli, M. Adv. Colloid Interface Sci. 2003, 103, 33−56. (34) Hayes, R. A.; Bohmer, M. R.; Fokkink, L. G. J. Langmuir 1999, 15, 2865−2870. (35) Kubiak, K.; Adamczyk, Z.; Ocwieja, M. Langmuir 2015, 31, 2988−2996. (36) Chen, B.; Liu, M.; Zhang, L.; Huang, J.; Yao, J.; Zhang, Z. J. Mater. Chem. 2011, 21, 7736−7741. (37) Zhang, L.; Xia, J.; Zhao, Q.; Liu, L.; Zhang, Z. Small 2010, 6, 537−544. (38) Ruiz, O. N.; Fernando, K. A. S.; Wang, B.; Brown, N. A.; Luo, P. G.; McNamara, N. D.; Vangsness, M.; Sun, Y.-P.; Bunker, C. E. ACS Nano 2011, 5, 8100−8107. (39) Lee, W. C.; Lim, C.; Shi, H.; Tang, L. A. L.; Wang, Y.; Lim, C. T.; Loh, K. P. ACS Nano 2011, 5, 7334−7341. (40) Ohno, Y.; Maehashi, K.; Matsumoto, K. J. Am. Chem. Soc. 2010, 132, 18012−18013. (41) Sheng, L.; Ren, J.; Miao, Y.; Wang, J.; Wang, E. Biosens. Bioelectron. 2011, 26, 3494−3499. (42) Gulbakan, B.; Yasun, E.; Shukoor, M. I.; Zhu, Z.; You, M.; Tan, X.; Sanchez, H.; Powell, D. H.; Dai, H.; Tan, W. J. Am. Chem. Soc. 2010, 132, 17408−17410. (43) Eda, G.; Fanchini, G.; Chhowalla, M. Nat. Nanotechnol. 2008, 3, 270−274. (44) Li, D.; Mueller, M. B.; Gilje, S.; Kaner, R. B.; Wallace, G. G. Nat. Nanotechnol. 2008, 3, 101−105. (45) Zhao, J.; Pei, S.; Ren, W.; Gao, L.; Cheng, H.-M. ACS Nano 2010, 4, 5245−5252. (46) Liu, Z.; Robinson, J. T.; Sun, X.; Dai, H. J. Am. Chem. Soc. 2008, 130, 10876−10877. (47) Qu, Q.; Gu, C.; Hu, X. Anal. Chem. 2012, 84, 8880−8890. (48) Chen, J.; Zhang, X.; Zheng, X.; Liu, C.; Cui, X.; Zheng, W. Mater. Chem. Phys. 2013, 139, 8−11. (49) Zhao, L.; Peng, X.; Yang, S.; Zhang, Y.; Wu, J.; Wei, X.; Li, F.; Pu, Q. RSC Adv. 2015, 5, 78519−78525.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b02342. Detailed experimental procedures, description of instruments, the formation of the PEI layer, the deposition kinetics of GO (S0) onto poly-DADMAC layer, the deposition kinetics of GO of different sizes onto the PEI layer, the preparation of silica NPs of different sizes, and deposition kinetics of silica NPs onto the PEI layer (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: 86-931-8912582. Phone: 86931-8913813. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to acknowledge financial support from the National Natural Science Foundation of China (Grants 21175062, 21275014, and 21527808), the Natural Science Foundation of Gansu Province (Grant No. 145RJYA245), and the Importation and Development of High-Caliber Talents Project of Beijing Municipal Institutions (Grant CIT&TCD20140309). The authors also thank Dr. Huie Jiang of Shaanxi University of Science & Technology for her help in the measurement with the Mütek PCD-03.

■

REFERENCES

(1) Norde, W.; Rouwendal, E. J. Colloid Interface Sci. 1990, 139, 169− 176. (2) Elgersma, A. V.; Zsom, R. L. J.; Lyklema, J.; Norde, W. Colloids Surf. 1992, 65, 17−28. (3) Etheve, J.; Dejardin, P. Langmuir 2002, 18, 1777−1785. (4) Fadel, H.; Schwarz, S.; Lunkwitz, K.; Jacobasch, H. J. Angew. Makromol. Chem. 1998, 263, 79−84. (5) Adamczyk, Z.; Zembala, A.; Michna, A. J. Colloid Interface Sci. 2006, 303, 353−364. (6) Michna, A.; Adamczyk, Z.; Kubiak, K.; Jamrozy, K. J. Colloid Interface Sci. 2014, 428, 170−177. (7) Morga, M.; Adamczyk, Z.; Goedrich, S.; Ocwieja, M.; Papastavrou, G. J. Colloid Interface Sci. 2015, 456, 116−124. (8) Adamczyk, Z.; Sadlej, K.; Wajnryb, E.; Nattich, M.; EkielJezewska, M. L.; Blawzdziewicz, J. Adv. Colloid Interface Sci. 2010, 153, 1−29. (9) Adamczyk, Z.; Zaucha, M.; Zembala, M. Langmuir 2010, 26, 9368−9377. (10) Morga, M.; Adamczyk, Z.; Ocwieja, M. J. Colloid Interface Sci. 2012, 386, 121−128. (11) Morga, M.; Adamczyk, Z.; Ocwieja, M. J. Nanopart. Res. 2013, 15, 2076. (12) Bubniene, U.; Ocwieja, M.; Bugelyte, B.; Adamczyk, Z.; NattichRak, M.; Voronovic, J.; Ramanaviciene, A.; Ramanavicius, A. Colloids Surf., A 2014, 441, 204−210. (13) Ocwieja, M.; Adamczyk, Z.; Morga, M.; Kubiak, K. Adv. Colloid Interface Sci. 2015, 222, 530−563. (14) Mattiasson, B.; Miyabayashi, A. Anal. Chim. Acta 1988, 213, 79− 89. (15) Miyabayashi, A.; Mattiasson, B. Anal. Biochem. 1990, 184, 165− 171. G

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (50) Hong, B. J.; Compton, O. C.; An, Z.; Eryazici, I.; Nguyen, S. T. ACS Nano 2012, 6, 63−73. (51) Li, S. H.; Mulloor, J. J.; Wang, L. Y.; Ji, Y. W.; Mulloor, C. J.; Micic, M.; Orbulescu, J.; Leblanc, R. M. ACS Appl. Mater. Interfaces 2014, 6, 5704−5712. (52) Boyd, R. D.; Pichaimuthu, S. K.; Cuenat, A. Colloids Surf., A 2011, 387, 35−42. (53) Wang, X.; Bai, H.; Shi, G. J. Am. Chem. Soc. 2011, 133, 6338− 6342. (54) de Mol, N. J.; Plomp, E.; Fischer, M. J. E.; Ruijtenbeek, R. Anal. Biochem. 2000, 279, 61−70.

H

DOI: 10.1021/acs.analchem.6b02342 Anal. Chem. XXXX, XXX, XXX−XXX