Electrical Field-Flow Fractionation for Metal Nanoparticle

Article pubs.acs.org/ac

Electrical Field-Flow Fractionation for Metal Nanoparticle Characterization Wilaiwan Somchue,† Atitaya Siripinyanond,† and Bruce K. Gale*,‡ †

Department of Chemistry and Center for Innovation in Chemistry, Faculty of Science, Mahidol University, Rama VI Rd., Bangkok 10400, Thailand ‡ Department of Mechanical Engineering, University of Utah, 50 S. Central Campus Drive, Rm. 2122, Salt Lake City, Utah 84112, United States ABSTRACT: The potential of electrical field-flow fractionation (ElFFF) for characterization of metal nanoparticles was investigated in this study. Parameters affecting separation and retention such as applied DC voltage and flow rate were examined. Nanoparticles with different types of stabilizers, including citrate and tannic acid, were investigated. Changes to the applied voltage showed a significant influence on separation in ElFFF, and varying flow rate was used to improve plate heights in the experiments. For nanoparticles of a fixed size, the separation was based primarily on electrophoretic mobility. Particles with low electrophoretic mobility elute earlier. Therefore, citrate stabilized gold nanoparticles (−2.72 × 10−4 cm2 V−1 s−1) eluted earlier than tannic acid stabilized gold nanoparticles (−4.54 × 10−4 cm2 V−1 s−1) of the same size. In addition, ElFFF can be used for characterization of gold nanoparticles with different particle sizes including 10, 20, and 40 nm with a fixed stabilizing agent. For a specific separation condition, the separation of 10, 20, and 40 nm gold nanoparticles was clearly based on the particle size as opposed to the electrophoretic mobility, as the elution order was in order of decreasing mobility for 10 (−4.54 × 10−4 cm2 V−1 s−1), 20 (−3.97 × 10−4 cm2 V−1 s−1), and 40 (−3.76 × 10−4 cm2 V−1 s−1) nm particles, respectively. here was obtained under small applied fields. The separation is based on the ratio of the electrophoretic mobility and size of the particles (or particle diffusivity). The balance that exists between the electrophoretic mobility and the diffusion of the particles can determine the final location of the particles with respect to the channel wall resulting in the relative retention of the particles in the channel. Particle groups that maintain a unique average distance from the channel walls will eventually be separated. ElFFF theory is generally based on the concept of FFF presented by Giddings and electrical double-layer concepts presented by Smoluchowski. The characteristic equation of ElFFF is

fter the concept of field-flow fractionation (FFF) was first demonstrated by J. Calvin Giddings, FFF was extended to many fields of research such as food research, environmental observation, and nanoscience research.1,2 On the basis of the field applied in the system, FFF can be categorized into many subtechniques such as sedimentation FFF (SdFFF), flow FFF (FlFFF), thermal FFF (ThFFF), and electrical FFF (ElFFF).3 ElFFF was first demonstrated for protein separations,4 but the instrumental as well as theoretical background was improved many years later.5−12 With the improvement of ElFFF, there is an opportunity to extend the utility of this technique to new and important areas of research. Currently, nanotechnology is attracting significant attention and has become a substantial research area due to the excellent properties of nanomaterials. In nanotechnology, particle sizes typically range from 1 to 100 nm. Particles of this size are expected to have a significant impact in many scientific fields, including chemistry, material sciences, biology, medicine, and electronics. The application of ElFFF for nanoparticle characterization is still a challenge, and the separation of nanoparticles smaller than 20 nm has not been reported in previous studies of ElFFF techniques. In this paper, we employed a typical ElFFF system and demonstrated separation of nanoparticles with different sizes and differences in electrophoretic mobilities. In electrical field-flow fractionation (ElFFF), an elutionbased separation technique, a few volts of electrical potential is applied perpendicular to the direction of separation across the flow channel. Thus, the high resolution separation reported

A

© 2012 American Chemical Society

λ = (kT /3πηd)(1/μEeff )(1/w)

(1)

where the retention parameter (λ) is dependent on the diameter, d, or charge (electrophoretic mobility, μ) of the particles. T is the temperature; k is the Boltzmann constant. η is the carrier viscosity. Eeff is the effective electric field, and w is the thickness of the channel. The retention ratio (R) for FFF in the Brownian mode is related to the retention parameter as Received: March 6, 2012 Accepted: May 3, 2012 Published: May 3, 2012 4993

dx.doi.org/10.1021/ac300662b | Anal. Chem. 2012, 84, 4993−4998

Analytical Chemistry

Article

R = 6λ[coth(1/2λ) − 2λ]

Instrumentation. The ElFFF channel used in all experiments was identical to what was used in the previous publications.5,13,14 The channel system consists of two graphite plates separated by a Mylar spacer, which was modified to form a channel between two graphite plates. The two graphite plates play a dual role as electrodes as well as channel walls. The ElFFF channel was rectangular in shape with the dimensions being 64 cm long, 2 cm wide, and 178 μm thick. A T-shaped connector was placed at the inlet of ElFFF channel for injecting nanoparticles and carrying carrier liquid into the ElFFF channel. A sample volume of 50 μL was introduced into the ElFFF channel for all experiments. A high pressure liquid chromatography (HPLC) pump (Alltech model 426, Alltech Associates, Inc., IL, USA) was used to deliver the channel flow. An online UV/visible absorbance detector (Model 520, ESA, Inc., MA, USA) was set at 391 nm for detection of AgNPs and 520 nm for detection of AuNPs samples. Data from the UV/visible detector were collected by Labview 8.6 software from National Instruments. The required DC voltage applied to the ElFFF system was controlled by the power supply (Agilent, model E36424A). A small resistance (5 Ω) was added in series with the system. The voltage drop across this resistance was measured, and the effective field of the system was calculated using Ohm’s law and the measured conductivity of the buffer. The current across the small resistance was measured, and the data was collected using a digital multimeter (Hewlett-Packard, model HP 34401A). A digital conductivity meter (Fisher Scientific, Traceable Digital Conductivity Meter) was used to continuously measure the conductivity of the carrier that exited from the channel. The conductivity meter was calibrated using traceable conductivity standard certified reference materials purchased from Fisher Scientific (Pittsburgh, PA, USA).

(2)

For highly compressed zones, this expression approaches 6λ. For constant field and flow rate runs, the retention ratio is determined from measurements of the channel void volume, V0, and the sample elution volume, Vr (or replaced by the void time and the retention time t0 and tr) of a sample component.

R = V 0/Vr = t 0/tr

(3)

In order to determine a technique’s resolution in chromatography and FFF, channel efficiency is governed by the quantitatively measured plate height (H)

H = L /5.54(tr /W1/2)2

(4)

where L is the length of the channel and W1/2 is the width of the peak at half-height.10

■

EXPERIMENTAL SECTION Chemicals. The reagent used for preparation of the carrier liquid in this study included sodium carbonate (Na2CO3) purchased from Sigma-Aldrich, Inc., MO, USA. The samples were commercial spherical nanoparticles. Tannic acid stabilized 0.05 mg mL−1, 10, 20, and 40 nm spherical gold nanoparticles (AuNPs) and citrate stabilized 0.02 mg mL−1, 10 nm spherical silver nanoparticles (AgNPs) were purchased from NanoComposix, CA, USA. Citrate stabilized AuNPs (0.047 mg mL−1, 10 nm) were purchased from Sigma-Aldrich, Inc., MO, USA. All nanoparticles were diluted 2 times in DI water prior to injection into the ElFFF channel. The electrophoretic mobility of all nanoparticles was measured by a Zetasizer Nano series instrument (Malvern Instruments Ltd., UK). The mean electrophoretic mobilities of 10 nm citrate stabilized AgNPs, citrate stabilized AuNPs, and tannic acid stabilized AuNPs were −3.26 × 10−4, −2.72 × 10−4, and −4.54 × 10−4 cm2 V−1 s−1, respectively. For AuNPs of 20 and 40 nm, the measured electrophoretic mobility was −3.97 × 10−4 and −3.76 × 10−4 cm2 V−1 s−1, respectively (Table 1).

■

RESULTS AND DISCUSSION To begin, the use of sodium carbonate as carrier instead of fresh DI water was found to avoid the contamination of CO2. Thus, experimental repeatability was improved significantly, and the bulk resistance decrease and the effective field were found to be consistent across the many experiments reported here. Previous ElFFF papers have reported significant issues with repeatability using DI water without any buffer. There are many parameters that can be tuned to optimize a separation. In many FFF techniques, the relaxation step is one of the important steps because it can reduce the effect of band broadening. However, relaxation times of 0, 30, 60, and 90 s showed the same retention results (results not shown), so relaxation was generally not performed with these nanoparticle samples. Nevertheless, other parameters affecting the separation and retention on ElFFF were investigated and found to be significant. Effect of Applied Voltage. Figure 1 shows a series of fractograms of 10 nm AgNPs run under various applied DC voltages while the flow rate was kept constant at 1 mL min−1. It was noted in previous publications that applied voltages above 1.7 V can cause significant electrolysis and bubble formation in the system at low flow rates.9,10 Therefore, voltage ranging from 1 to 1.7 V was applied in the preliminary studies to avoid electrolysis and bubble formation. The fractograms show an increase in retention time as the applied DC voltage is increased because, as the voltage increases, the particles are forced to accumulate closer to the walls and remain in the lower velocity portions of the flow

Table 1. Measured Electrophoretic Mobility of Various Metal Nanoparticles samples AuNPs citrate AuNPs tannic acid AuNPs tannic acid AuNPs tannic acid AgNPs citrate

diameter (nm)

zeta potential (V)

measured electrophoretic mobility (×10−4 cm2 V−1 s−1)

10.5 ± 0.8

−34.8 ± 1.8

−2.72

9.1 ± 0.8

−58.0 ± 3.8

−4.54

20.0 ± 1.8

−50.6 ± 0.9

−3.97

39.5 ± 3.8

−48.1 ± 1.7

−3.76

9.1 ± 2.0

−41.7 ± 4.3

−3.26

Deionized water (18.2 MΩ cm−1) obtained from a waterpurification system (Barnstead International, Dubuque, IA, USA) was used to prepare all chemical reagents. A 7.84 μM Na2CO3 solution, pH 6.5, was prepared and was used as a carrier and used instead of deionized water to reduce contamination by atmospheric carbon dioxide and a subsequent change in conductivity. 4994



Article

Figure 2. Fractograms of 10 nm AgNPs run under an applied voltage of 1.5 V with different flow rates of 1 (open circle), 1.5 (square), 2 (triangle), and 2.5 (cross) mL min−1 monitored at 391 nm.

Figure 1. Fractograms of 10 nm AgNPs at increasing applied voltages monitored at 391 nm. The flow rate was kept constant at 1 mL min−1.

experience, on average, a lower effective field than particles that elute sooner. Thus, higher flow rates that leave particles in the channel for shorter periods may experience an overall higher field. This higher field appears to overcome any nonequilibrium effects and provide for increased retention, though this effect appears to be limited as the increased retention did not continue for all particles above 2 mL min−1. In addition, the fractogram of AgNPs at 2.5 mL min−1 was broader than the fractogram at 2 mL min−1 even though they eluted at the same elution volume. The increased band broadening observed at the 2.5 mL min−1 flow rate might be due to the inability of even these small particles, which have a high diffusion coefficient, to reach equilibrium in the high flow velocity gradient that occurs in these small channels as the flow rate increases.14 Thus, there may be an optimum flow rate for every ElFFF channel depending on the geometry (which leads to electrical properties) and operating conditions. Though some of the particles showed better retention at 2.5 mL min−1, 2.0 mL min−1 appears to be the optimum flow rate for this experimental setup. Retention of metal nanoparticles such as AuNPs and AgNPs with the same particle size of 10 nm but different stabilizers was investigated. Figure 3 shows the retention volume of 10 nm

profile, which leads to increased retention. Thus, the retention time of the particles increases with applied voltage. Note that the retention of these particles is significant in that retention of metal nanoparticles of this size has not been previously demonstrated in the literature. At low voltages, the current across the channel is very small as the system exhibits a nonlinear, diode-like current−voltage relationship. As the current is a good indicator of the effective field in the channel, a low current suggests a low effective field; therefore, particle retention or separation was not strongly observed in this region. As the turn-on voltage is crossed, the current rises and better separations are obtained.15 After the applied field is turned off after 900 s, the elution of AgNPs was observed. The higher applied voltages show more adsorption at the channel wall and the largest peaks after the field is removed. As shown in Figure 1, the largest peak was observed at 1.7 V as the higher applied voltage appears to trap particles on the accumulation walls, resulting in no elution of the AgNPs until the field is removed and they are released from the wall. Lower voltages also appear to trap some particles but in substantially less numbers based on peak area. Interestingly, particle adsorption on the accumulation wall was less significant for AuNPs using the same field conditions. This adsorption might be related to the known tendency of colloidal silver to deposit on many electrode surfaces.16 In any case, there is a potential loss of particles on the channel walls under high field conditions. Effect of Flow Rate. As reported in the literature, the carrier flow rate can affect the separation and plate heights of FFF systems.10,17,18 In this experiment, flow rate was varied from 1 to 2.5 mL min−1 in order to find the flow rate that provides the best separation efficiency of metal nanoparticles. The retention time of the fractogram decreased with increasing elution flow rate, as is typical (result not shown). Figure 2 shows the fractograms of 10 nm AgNPs for a plot of the retention volume at different flow rates. As the flow rate increased from 1 to 2 mL min−1, the retention volume of 10 nm AgNPs increased but did not increase significantly as the flow rate became 2.5 mL min−1. The increase in retention volume with flow rate is somewhat unexpected and opposed to typical FFF theory, which predicts a decrease in retention volume due to nonequilibrium effects, meaning that diffusion is unable to bring the particle cloud to equilibrium when the flow rate is too high. This failure to keep equilibrium leads to band broadening as particles remain in fast flow lines longer than they do in an equilibrium condition. The increase in retention may be due to the decaying electric field that drives retention in ElFFF. Particles that spend a longer time in the ElFFF channel will

Figure 3. Relationship between flow rate and retention volume of citrate stabilized AgNPs, tannic acid stabilized AuNPs, and citrate stabilized AuNPs run under an applied voltage of 1.5 V at various flow rates while being monitored at 391 nm for AgNPs and 520 nm for AuNPs.

AuNPs and AgNPs stabilized with citrate and tannic acid. The results show that citrate stabilized 10 nm AgNPs and AuNPs consistently show the same retention volume at a variety of flow rates. Because of their similar electrophoretic mobility, the elution times are quite close, as expected. These results suggest there is no “material” effect that impacts retention in the ElFFF channel. Interestingly, tannic acid stabilized 10 nm AuNPs 4995



Article

effective field decrease is not clear but could be due to a variety of factors such as the onset of electrolysis, which saps some of the electrical current that could generate a higher field in the bulk of the channel, or some steric or wall repulsion effect that limits the ultimate approach of particles to the channel walls. The flat retention ratio for higher voltages suggests the latter is involved, though electrolysis is likely to be present. The retention ratio decreased as expected when the applied field was increased before appearing to stabilize at about 0.2 for increasing applied voltage. The reason for the inability to further decrease the retention ratio above 1.9 V is not clear, but the reasons are likely the same as those mentioned for the decreasing effective field at the same voltage levels. Thus, for this particular ElFFF system, it appears that increasing the applied voltage indefinitely is not helpful and could possibly be disruptive; therefore, an applied voltage of 1.9 V at 2 mL min−1 was considered optimal and used for further investigation. Channel Efficiency. The channel efficiency, the factor that allows comparison of ElFFF to other techniques, was determined by measuring the plate height. In order to evaluate channel efficiency, plate heights at varying flow rates of the carrier were investigated. A plot of plate heights for 10 nm of tannic acid stabilized gold nanoparticles at various flow rates is shown in Figure 5. Smaller values of plate height indicate a better separation system. Plate heights in FFF systems typically increase with increasing flow rates.

show a higher retention volume when compared with the nanoparticles that are stabilized with citrate because tannic acid stabilized particles have a higher electrophoretic mobility than the AgNPs and AuNPs that are stabilized with citrate. The reason for the higher mobility with tannic acid is that tannic acid is a multidentate capping ligand and should provide a greater surface charge resulting in higher electrophoretic mobility when compared to the citrate stabilized nanoparticles. The effect of the stabilizer persists even in the highly dilute conditions found in the ElFFF system. Another important effect that was found related to flow rate is that, using a flow rate of 2 mL min−1, we can increase the applied voltage up to 2.1 V without electrolysis or bubble formation in the channel, leading to increased field strengths and even greater retention. Retention Ratio and Effective Field. From the experimentally obtained fractograms, we can determine the retention ratio of particles from the retention time and calculate the effective field that generated that retention. For a range of applied fields, the correlation between the applied field and the retention ratio was investigated. Figure 4 shows the calculated

Figure 4. Plot showing the relationship between the retention ratios of tannic acid stabilized AuNPs under different applied voltages and the dependence of the effective field on the applied voltage. Particle retention was monitored at 520 nm. The flow rate was kept constant at 2 mL min−1. The squares relate to the y-axis on the left while the circles relate to the y-axis on the right.

retention ratio plotted against the applied voltage for results obtained using 10 nm of tannic acid stabilized AuNPs. The voltage was applied from 0.8 to 2.1 V, and the flow rate was kept constant at 2 mL min−1. Normally, the actual electric field experienced by particles suspended in ElFFF channels (referred to as the effective field) is smaller than that calculated on the basis of the applied voltage because the application of voltage to the channel walls quickly generates polarization layers in the channels that shield the interior of the channel from the full field, so the effective field is typically less than 3% of the applied electric field.10 Note that experimental measurements indicate that the effective fields dramatically decrease within 50 s of the DC voltage application to the ElFFF channel with 7.84 μM Na2CO3 solution as a carrier. The effective field used to generate separation of nanoparticles in the channel was calculated and shown in Figure 4 for the range of voltages tested. The applied voltages of 0.8−2.1 V across the channel give rise to effective field strengths on the order of 20−600 V m−1. In Figure 4, with increasing applied voltage, the effective field increases nonlinearly, in accordance with the diode-like behavior of the electrochemical system, until the highest effective field was observed at 1.9 V. For an applied field larger than 1.9 V, the effective field decreased. The reason for the unexpected

Figure 5. Calculation of plate height obtained from ElFFF by injection of 10 nm of tannic stabilized AuNPs under an applied field of 1.9 V at different flow rates.

The results regarding plate height for the ElFFF system showed that a variety of competing phenomena were affecting plate height and it is difficult to draw a clear conclusion. With increasing flow rate from 1 to 1.5 mL min−1, plate height increases. However, a flow rate of 2 mL min−1 gives the smallest plate height, and then it increases again when applied at 2.5 mL min−1. As noted before, the faster flow rates lead to nonequilibrium effects, which cause increased plate heights for increasing flow rates, which may be the general trend here. As noted, the field also tends to decrease with time, so particles at faster flow rates may experience higher fields. The combination of these conflicting effects may be what causes a lack of an apparent trend. Separation of 10 nm Nanoparticles. In separate experiments, both Au and Ag NPs of 10 nm diameter were injected into the channel under the optimal retention conditions of an applied voltage of 1.9 V and a flow rate of 2 mL min−1. This experiment was designed to eliminate particle size effects. For a fixed size of nanoparticles, the separation is 4996



Article

of 1.9 V, and the flow rate was kept constant at 2 mL min−1 (Figure 7). All of the particles had the same stabilizing agent.

expected to be based primarily on electrophoretic mobility. Particles with lower electrophoretic mobility elute earlier. The citrate stabilized gold nanoparticles (−2.72 × 10−4 cm2 V−1 s−1) eluted earlier than tannic acid stabilized gold nanoparticles (−4.54 × 10−4 cm2 V−1 s−1) as expected. In an attempt to demonstrate the separation of the citrate and tannic acid stabilized 10 nm AuNPs, the two samples were mixed together in a 1:1 ratio and subsequently injected into the channel. The results shown in Figure 6 provide fractograms for

Figure 7. Fractograms for 10, 20, and 40 nm of tannic acid stabilized AuNPs both individually (line) and as a mixture (open triangle). The experiments were carried out using an applied voltage of 1.9 V, and the detector was set at 520 nm for AuNPs for detection. The flow rate was kept constant at 2 mL min−1.

With the selected separation condition, the separation of 10, 20, and 40 nm gold nanoparticles was observed. The elution order was from 10 (−4.54 × 10−4 cm2 V−1 s−1), 20 (−3.97 × 10−4 cm2 V−1 s−1), and 40 (−3.76 × 10−4 cm2 V−1 s−1) nm, respectively. These measured results are consistent with electrophoretic mobility theory in which mobility is typically expected to increase with decreasing size, assuming a constant surface charge or even surface charge density. The results show an inversion of elution order if only particle mobility is considered. The smallest particles, which have highest electrophoretic mobility, elute faster than the larger particles. These results suggest that diffusion or particle size is a more important effect than electrophoretic mobility in particle retention for these experimental conditions. Certainly, the relative magnitudes of the changes in mobility and size are quite different. The electrophoretic mobility (starting with the smallest value) changes only about 25% between the different particles, while the particle size changes by up to 300%. Thus, the mobility difference appears to be mostly inconsequential. Therefore, we can conclusively state that the separation of nanoparticles in these conditions is primarily based on particle size rather than the electrophoretic mobility. The mixing of 10, 20, and 40 nm AuNPs was carried out in 1:1:1 ratio and subsequently injected into the ElFFF channel with the same conditions used to obtain as the individual AuNPs fractograms. As noted, the results show that the separation was mainly based on the particle size. The same elution order was observed as for the individual injections, and the particles clearly separate from each other even though the electrophoretic mobility of the mixture measures out at only one value on the zetasizer. However, the retention time of 10 and 20 nm particles in the mixture was slightly faster than the retention times measured for individual injections. The reason for the change is not clear, but it is clear that AuNPs can be separated by size in the ElFFF channel.

Figure 6. Fractogram of citrate stabilized 10 nm AgNPs, tannic acid stabilized 10 nm AuNPs, and citrate stabilized 10 nm AuNPs with measured differences in electrophoretic mobility. The separation was carried out using an applied voltage at 1.9 V, and the detector was set at 391 nm for AgNPs and 520 nm for AuNPs for detection. The flow rate was kept constant at 2 mL min−1.

the individual samples as well as the mixed samples. Interestingly, the mixture elutes between the retention times of the individual AuNPs. There are not two peaks, indicating separation, and the mixture does not take on the characteristics of either of the particles. Instead, it appears that the mixed sample takes on an average value for that of the citrate and tannic acid stabilized particles. When the gold nanoparticles are in solution, the stabilizing agents, which weakly associate with the nanoparticle surface, play an important role to stabilize the particles as well as prevent aggregation. Our hypothesis is that because the stabilizing agents are weakly bound to the surface, they can be displaced by other molecules. Therefore, the stabilizers can exchange between both of the AuNP samples leading to a change in electrophoretic mobility of the mixture. The mixed stabilizers cause the electrophoretic mobility to fall somewhere between the electrophoretic mobility of the two individual particles. This result was confirmed by measuring the electrophoretic mobility of the mixture, which was found to be −3.60 × 10−4 cm2 V−1 s−1, which is between the values measured for the two AuNPs samples separately. This experimental condition was also used for separation of AgNPs (−3.26 × 10−4 cm2 V−1 s−1), and the fractogram is shown in Figure 6. The AgNPs are also citrate stabilized. These results suggest that the stabilizers, which determine the surface properties of these nanoparticles, play an important role in their retention in ElFFF and may outweight even elemental composition. The results also suggest that mixtures of nanoparticles that take on similar surface coatings may be difficult to separate using ElFFF. Accordingly, the effect of the buffers and other agents in the samples will need to be carefully considered if separations are attempted. Separation of 10, 20, and 40 nm AuNPs. In order to estimate the effect of size-dependent properties of particles on separation in ElFFF, AuNPs with 10, 20, and 40 nm diameters were individually injected to the system with an applied voltage

■

CONCLUSION Retention of Au and Ag nanoparticles in the size range of 10− 40 nm was governed by normal mode electrical field-flow fractionation. The applied voltage shows a great influence on separations in ElFFF and can be tuned to achieve optimal separation results. Increasing the flow rate actually improved retention before the effect saturated. Efficiencies of retention in 4997



Article

(14) Gigault, J.; Gale, B. K.; Hecho, I. L.; Lespes, G. Anal. Chem. 2011, 83, 6565−6572. (15) Chang, M. H.; Dosev, D.; Kennedy, I. M. Sens. Actuators, B 2007, 124, 172−178. (16) Welch, C. M.; Compton, R. G. Anal. Bioanal. Chem. 2006, 384, 601−619. (17) Smith, L. K.; Myers, M. N.; Giddings, J. C. Anal. Chem. 1977, 49, 1750−1756. (18) Kantak, A.; Srinivas, M.; Gale, B. K. Lab Chip 2006, 6, 645−654.

the channels were measured and found to be somewhat dependent on flow rate and that there is an optimal flow rate for maximum retention and efficiency. A maximum effective field was found at 1.9 V applied, and a minimum retention ratio also was found for these particles. Retention times for essentially identical particles with different electrophoretic mobilities were found to be different and dependent on the electrophoretic mobility, but mixtures of particles could not be separated due to mobile stabilizing agents, which lead to all particles adopting the same mobility. Particles of different sizes were readily separated, though in reverse mobility order as particle size outweighed the mobility effect. Overall, the results showed clear success with regard to characterizing and separating metal nanoparticles. Nevertheless, the technique still has some limitations such as adsorption of AgNPs at high applied voltages and the technique cannot separate nanoparticles which have the same particle size but a difference in electrophoretic mobility, if the surface coatings are mobile.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 801-585-5944. Fax: 801585-9826. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge the research grants from the Thailand Research Fund (TRF) and Center for Innovation in Chemistry: Postgraduate Education and Research Program in Chemistry (PERCH-CIC), Commission on Higher Education, Ministry of Education, Thailand, and the Office of the Higher Education Commission and Mahidol University under the National Research Universities Initiative. Financial support was from the Thailand Research Fund through the Royal Golden Jubilee Ph.D. Program (Grant No. PHD/0336/2550) to W.S. B.K.G. was supported by the USDA (USDA-CSREES 200935603-05037) and NSF (CBET-0967037).

■

REFERENCES

(1) Poda, A. R.; Bednar, A. J.; Kennedy, A. J.; Harmon, A.; Hull, M.; Mitrano, D. M.; Ranville, J. F.; Stevens, J. J. Chromatogr., A 2011, 1218, 4219−4225. (2) Kammer, F.; Legros, S.; Larsen, E. H.; Loeschner, K.; Hofmann, T. TrAC, Trends Anal. Chem. 2011, 30, 425−436. (3) Schimpf, M.; Caldwell, K.; Giddings, J. C. Field-Flow Fractionation Handbook; John Wiley & Sons, Inc.: New York, 2000; pp 8. (4) Caldwell, K.; Kesner, L. F.; Myers, M. N.; Giddings, J. C. Science 1972, 176, 296−298. (5) Caldwell, K. D.; Gao, Y. S. Anal. Chem. 1993, 65, 1764−1772. (6) Dunkel, M.; Tri, N.; Beckett, R.; Caldwell, K. D. J. Microcolumn Sep. 1997, 9, 177−183. (7) Palkar, S. A.; Schure, M. R. Anal. Chem. 1997, 69, 3223−3229. (8) Palkar, S. A.; Schure, M. R. Anal. Chem. 1997, 69, 3230−3238. (9) Gale, B. K.; Caldwell, K. D.; Frazier, B. IEEE Trans. Biomed. Eng. 1998, 45, 1459−1469. (10) Tri, N.; Caldwell, K. D.; Beckett, R. Anal. Chem. 2000, 72, 1823−1829. (11) Gale, B. K.; Caldwell, K. D.; Frazier, B. Anal. Chem. 2001, 73, 2345−2352. (12) Biernacki, J. J.; Mellacheruvu, P. M.; Mahajan, S. M. Anal. Chem. 2006, 78, 4998−5005. (13) Gale, B. K.; Srinivas, M. Electrophoresis 2005, 26, 1623−1632. 4998


Electrical Field-Flow Fractionation for Metal Nanoparticle

Recommend Documents