Optimization of Surfactant Concentration in Carbon Nanotube Solutions for Dielectrophoretic Ceiling Assembly and Alignment: Implications for Transparent Electronics.

Surfactants such as sodium dodecyl sulfate (SDS) are used to improve the dispersity of carbon nanotubes (CNTs) in aqueous solutions. The surfactant concentration in CNT solutions is a critical factor in the dielectrophoretic (DEP) manipulation of CNTs. A high surfactant concentration causes a rapid increase in the solution conductivity, while a low concentration results in undesirably large CNT bundles within the solution. The increase in the solution conductivity causes drag velocity that obstructs the CNT manipulation process due to the electrothermal forces induced by the electric field. The presence of large CNT bundles is undesirable since they degrade the device performance. In this work, mathematical modeling and experimental work were used to optimize the concentration of the SDS surfactant in multiwalled carbon nanotube (MWCNT) solutions. The solutions were characterized using dynamic light scattering (DLS) and ultraviolet-visible spectroscopy (UV-Vis) analysis. We found that the optimum SDS concentration in MWCNT solutions for the successful DEP manipulation of MWCNTs was between 0.1 and 0.01 wt %. A novel DEP configuration was then used to assemble MWCNTs across transparent electrodes. The configuration was based on ceiling deposition, where the electrodes were on top of a droplet. The newly proposed configuration reduced the drag velocity and prevented the assembly of large MWCNT bundles. MWCNTs were successfully assembled and aligned across interdigitated electrodes (IDEs). The assembly of MWCNTs from aqueous solutions across transparent electrodes has potential use in future transparent electronics and sensor devices.

Introduction

Carbon nanotubes (CNTs) have attracted much attention in recent years due to their unique properties and usability in many scientific and research applications.[1] CNTs are produced in the form of agglomerates or bundles that are tightly attached to each other, making their dispersity and solubility in common solvents challenging.[2,3] Solution-based processing is the main route available in manufacturing and engineering CNT-based devices, making the dispersion of CNTs in solutions an important research area.[4]

Two main approaches are used to improve the CNT dispersity in solvents, which are chemical and physical approaches. In the chemical approaches, CNTs are treated with acids to covalently attach functional groups onto their surface.[5,6] The physical approaches (noncovalent treatment) use strong mechanical shear forces such as sonication to weaken the binding force between the CNTs with the help of agent materials (dispersants). Examples of dispersants are polymers,[7−9] proteins,[10,11] and surfactants.[12,13]

Surfactants are preferred in dispersing CNTs because they do not cause any structural damage to the CNTs compared with the chemical approaches. Furthermore, the removal of the surfactant is much easier than that of polymers and proteins. Surfactants attached to the CNTs can be easily washed away after use using deionized water (DIW). There are three types of surfactants, which are anionic surfactants such as sodium dodecylbenzene sulfonate (SDBS)[14] and sodium dodecyl sulfate (SDS),[15] cationic surfactants such as dodecyl tri-methyl ammonium bromide (DTAB),[16] and nonionic surfactants such as polyoxyethylene octyl phenyl ether (Triton X-100).[17] Among the mentioned surfactants, SDS is widely used due to its low cost and easy processing procedures. SDS was used to prepare CNT solutions for several applications, such as nanocomposites,[18] cement pastes,[19] nanofluids,[20] antibacterial agents,[21] and coating materials.[22]

Many studies address the role of the SDS surfactant in CNT solutions in terms of the temperature effect,[23] SDS concentration,[15] sonication power,[24] and binding energy perspective.[25] Yu et al. concluded that there are rules to disperse MWCNTs with the help of SDS and proper sonication.[26] The first rule concerns the SDS/MWCNT ratio range, where the minimum weight ratio of SDS to MWCNTs to homogeneously dispersed MWCNTs in aqueous solutions was 1.5–1, while the maximum concentration was about 1.4 wt %. The second rule concerns the minimum sonication energy required, where a sonication time of 90 min (100 000 J) was enough to disperse MWCNTs at concentrations lower than 1.4 wt %. However, limited studies addressed the suitable SDS concentration in CNT solutions that are used in electrokinetic manipulation systems, such as the dielectrophoretic (DEP) deposition of MWCNTs.

DEP is an electrokinetic phenomenon that can be utilized to manipulate CNTs within the solution using nonuniform electric fields.[27] An example of CNT manipulation is the deposition of CNTs across microelectrode structures in an aligned form.[28,29] Despite the importance of the surfactants in dissolving CNTs, understanding their role in altering the solution’s electrical conductivity is an essential factor to achieve successful deposition.[30]

Controlling the electrical conductivity of CNT solutions by adjusting the surfactant concentration is one of the current challenges to avoid the occurrences of electrothermal phenomena in DEP systems, such as the joule heating effect and medium circulation due to the heat convection.[31] Furthermore, solutions with high electrical conductivity allow more carriers to pass through the circuit, which might damage the microelectrodes at low frequencies or break the CNT connections across the electrode gaps.[30] Optimizing the SDS concentration in DEP solution is required to ensure successful deposition while maintaining strong CNT dispersity. Additionally, introducing a new DEP setup compatible with CNT-SDS solutions is desirable in fabricating CNT-based devices such as transparent sensors.[32]

In this article, the role of the SDS surfactant in dispersing MWCNTs in aqueous solutions and how it alters the electrical conductivity were studied. The study aims to optimize the concentrations of SDS in MWCNT solutions in order to use them in electrokinetic manipulation systems such as DEP systems. The article also presents a novel DEP configuration to assemble MWCNTs from the optimized solution and align them across microelectrodes.

Theory and Modeling

The electrical conductivity of CNT solutions plays a critical role in the manipulation of CNTs in DEP systems. Solutions with high electrical conductivity cause other forces to appear in the systems along with the DEP force. Examples of these forces are the electrothermal (ETH) and AC electroosmosis (ACEO) forces. In this section, the direct effect of the solution’s electrical conductivity on these forces is theoretically investigated. The DEP force is the motion of polarized particles in medium subjected to nonuniform electric fields. The magnitude of the DEP force depends on three main factors which are the particle geometry factor, v, particle polarizability, α̃, and electric field, E as described by eq .[33]

The CNT geometrical factor depends on the CNT radius, rcnt, and CNT length, lcnt, and given by v = (2πrcnt2lcnt)/3, while the effective polarizability factor depends on the complex permittivity of the CNT, ε̃cnt, the complex permittivity of the medium, ε̃̃m, and the depolarization factor, L, as expressed by eq .

The complex permittivity of the medium and CNT is described by eq

At high frequencies (ω → ∞), the effective polarizability factor can be approximated by α̃̃ = ε0εmRe[(εcnt – εm)/(εm)], while at low frequencies (ω → 0), the effective polarizability factor can be approximated by α̃̃ = ε0εmRe[(σcnt – σm)/(σm)]. This means that medium electrical conductivity directly affects the DEP force at low frequencies.

The ACEO is the second force present in the electrokinetic system. ACEO occurs due to the existence of charges (negative or positive) at the solid–liquid interface. These charges form an electric double layer (EDL) due to the tangential component of the electric field. The EDL causes nonzero time-average Coulombic force on the ions at the electrode surface. This force causes medium drag velocity above the electrodes. The time-averaged velocity due to ACEO, Uaceo, is expressed by the Smoluchowski formula (eq ).[34]where Λ, ε0, εm, Vp, ηm, and x are the EDL capacitance ratio, vacuum permittivity, medium relative permittivity, voltage potential, medium viscosity, and distance from the electrode gap center to the calculation point, respectively. Ω is a dimensionless frequency expressed by eq .

The dimensionless frequency also depends on the signal angular frequency, ω, medium conductivity, σm, and Debye length, λDe. The Debye length equals the square root of the product of diffusivity and medium permittivity conductivity ratio (λDe = √((ε0εmDif)/σm)). From the equations given above, medium electrical conductivity directly affects the Debye length and the dimensionless frequency. Thus, the ACEO velocity is a function of medium electrical conductivity.

The third electrokinetic force in the electrokinetic system is the ETH force. The ETH force occurs in the medium due to the nonuniform heating caused by the flow of the electric current in the medium. The ETH is expressed by eq .[34]where ρq and ρs are the charge and mass densities, respectively. The three terms at the right-hand side of the equation are the Coulomb force, the dielectric force, and the electrostriction pressure. The last term can be ignored since its gradient of a scalar quantity does not affect the incompressible fluid dynamics. Generally, the expression defines the electrical body force and fluid motion in terms of local variations in permittivity and conductivity. The time-averaged body force can be written in terms of temperature gradient as described by eq .[35,36]where indicates the complex conjugation of the electric field. The approximation values of α = (1/ε)(∂ε/∂T) and β = (1/σ)(∂σ/∂T) for aqueous solution are −0.4% and +2% K–1, respectively.[37,38] The body force equation has two terms; the first term represents the Coulomb force, which is dominant at low frequency and the second term represents the dielectric force and dominates at high frequencies. Thus, the electrical conductivity of the medium directly affects the ETH force, especially at low frequencies.

Unlike the first three forces, the gravitational force, Fgrav, does not depend on the electric field. It depends on the CNT volume, v, and the density difference between the medium and the CNT (ρcnt – ρm). The gravitational force acting on a CNT is described by eq .[36]where g is the gravitational acceleration. The magnitude of the CNT velocity induced by gravity is calculated by dividing the gravitational force described in eq by a friction factor; γcnt represents the CNT mass flow rate (γcnt = 3πηmlcnt/(ln(lcnt/rcnt)).[35] The CNT total velocity, UCNT, is the sum of the velocity induced by the DEP force, UDEP, velocity induced by the gravitational force, Ugrav, and medium drag velocity due to the ETH and ACEO, Udrag (eq ).

CNTs are required to be deposited (assembled and aligned) between ITO electrodes. Figure illustrates the conditions of the velocities to ensure successful deposition. The velocity induced by the DEP force is the only velocity in the direction toward the deposition area. Thus, the DEP velocity must be greater than the sum of the other velocities present in the DEP system. If the DEP velocity is less than the sum of the other velocities, the concentration of the SDS surfactant must be optimized again to reduce the medium conductivity. Details regarding the physical model and parameters used in the simulation can be found elsewhere.[39]

Figure 1

Flow chart illustrates the conditions under which the CNTs can be successfully deposited across ITO electrodes. The colors of the arrows in the medium correspond to the velocity boxes in the flow chart, which show the velocity direction.

Materials and Methods

Solution Preparation

Three groups of solutions were prepared. In the first group, different amounts of SDS were mixed with DIW in 20 mL vials. The solutions were stirred and heated on a hotplate at 500 rpm and 35 °C for 5 min. The mass of the SDS surfactant was varied from 0 to 300 mg, which is equivalent to concentrations between 0.0 and 1.5 wt %. The purpose of preparing these solutions is to experimentally measure their electrical conductivity as a function of SDS concentration. An electrical conductivity meter (PRIMO5, Hanna Instruments) was used for the conductivity measurements.

In the second group, surfactant solutions were prepared in 20 mL vials with SDS concentrations of 1, 0.5, 0.1, 0.05, and 0.01 wt %. The same procedure was followed as in the first group. Then, 0.1 mg of MWCNT powder was dropped in each vial to result in a MWCNT concentration of 0.001 wt %. The final solutions were sonicated for 15 min (see Table S1 in the Supporting Information).

In the third group, 100 mg of the SDS surfactant was added to 20 mL of DIW (0.5 wt %). The solution was sonicated for 5 min and then diluted by adding 180 mL of DIW to result in a new concentration of 0.05 wt %. MWCNT powder was prepared separately in five vials at different masses, which were 2, 1, 0.75, 0.5, and 0.25 mg. A total of 20 mL of the surfactant solution (0.05 wt %) was added to each vial, resulting in MWCNT concentrations from 0.01 to 0.00125 wt %. The solutions were further sonicated for 90 min (see Table S2 in the Supporting Information).

The dynamic light scattering (DLS) technique (Malvern Instruments Nano S) was used to measure the size distribution of MWCNTs in solutions. The solubility of MWCNTs and the quality of the solutions were determined by their absorbance to a specific wavelength using ultraviolet–visible (UV–Vis) spectroscopy (Perkin Elmer Lambda 35).

Electrode Fabrication

Figure illustrates the standard lithography method that was used to fabricate ITO electrodes on glass substrates. First, the ITO layer was covered and spin-coated with a positive photoresist. The substrate was then heated on a hotplate to harden the photoresist layer. Interdigitated electrodes (IDEs) with a spacing of 50 μm were printed on the photoresist layer by exposing the substrate to UV light through a polyester photomask. The exposed patterns of the photoresist were then developed using a positive developer. An acid mixture was used to etch the ITO layer before cleaning the photoresist remains using acetone and IPA. The lithography protocol used in this work was further explained elsewhere.[40]

Figure 2

Electrode fabrication protocol and the fabricated electrode geometry. (1) ITO-coated substrate was cleaned using acetone, IPA, and DIW. (2) Positive photoresist (AZ 5214E) spin-coated the ITO and then baked for 2 min at a temperature of 90 °C. (3) Photoresist was exposed to UV light through a polyester photomask. (4) Developing process followed by hard baking for 2 h at a temperature of 120 °C. (5) ITO was etched using a mixture of HCL and HNO3 (4:1). (6) Final product was cleaned using acetone and IPA.

Deposition Setup

The ITO-coated glass substrate was glued on a bigger microscopic glass slide to facilitate its movement. One drop of the MWCNT solution was pipetted on the electrodes, and then, the substrate was flipped upside down so that the electrodes were on the top of the droplet. The method only works for small-volume drops (∼30 μL). If the medium volume is more than 30 μL, a glass cover must be used at the droplet’s bottom side. The substrate was stabilized using a metal holder before applying the AC signal. The parameters of the AC signal used for the assembly and alignment were 20 Vpp and 1 MHz, respectively, applied for 10 min. This method is expected to reduce the heat convection flow because the medium with low density remains near the electrode area. Moreover, the gravitational force attracts undesirably large MWCNT bundles downward away from the electrode gaps and toward the medium surface. Figure S1 in the Supporting Information illustrates the setup of the ceiling assembly.

Results and Discussion

Solution Characterization

Figure a presents the variation in the electrical conductivity of the solutions as a function of SDS concentration. The solution’s electrical conductivity increased linearly when the concentration of the surfactant was increased. The fitting formula of the conductivity curve is expressed by eq

Figure 3

Conductivity measurements. (a) Measured electrical conductivities of DIW at different SDS concentrations. (b) Measured electrical conductivities of MWCNT solutions at different MWCNT concentrations and a fixed SDS concentration (0.05 wt %). Tables S3 and S4 in the Supporting Information show the raw data, mean value, and standard deviation of the measured conductivities.

Although SDS is a well-studied surfactant in terms of how it alters the conductivity of solutions, it is essential to experimentally measure the conductivity in the presence of MWCNTs. Figure b presents the solution conductivity at a constant SDS concentration (0.05 wt %) and varied MWCNT concentration (0.01–0.00125 wt %). The conductivity was in the same order (10–2 S/m) with standard deviations in the order of 10–4. This indicates that MWCNTs did not alter the medium electrical conductivity in the same way as the SDS did at the mentioned concentrations. The solutions with varied MWCNT concentrations are shown in Figure S2, along with field emission scanning electron microscopy (FESEM) and high-resolution transmission electron microscopy (HRTEM) images.

Figure a presents the intensity of the scattered light from MWCNTs suspended in DIW. The intensity peaked in a larger size range in the solution that does not contain SDS. However, when the SDS surfactant was added to the MWCNT solution, the intensity curve was shifted to the left, indicating strong solubility of large MWCNT bundles to individual tubes.

Figure 4

DLS analysis and results of MWCNT solutions with and without the SDS surfactant. (a) Intensity of the scattered light from suspended MWCNTs. (b) Size distribution of MWCNTs with and without SDS. The concentrations of SDS and MWCNTs were 0.01 and 0.001 wt %, respectively.

Figure b shows the size distribution of the dispersed MWCNTs. Before adding the MWCNTs, the particle size distribution of pure SDS solution was around 2–3 nm, representing the SDS micelle diameter. In the solution containing SDS and MWCNTs, the total volume percentage in small size ranges was more than the volume percentage in the solution with only MWCNTs. For example, the total volume percentage of tubes with sizes less than 350 nm in solution with SDS solution was 38%, which was 4% more than the total volume percentage of tubes in solution without SDS in the same size range. In conclusion, the difference in the particle size distribution when SDS was used with MWCNTs proves the success of the surfactant in dissolving large bundles into individual tubes.

Figure a shows the UV–Vis absorbance results of the solution at different SDS concentrations (the DIW curve was used as a baseline to compare different SDS concentration curves). The absorbance intensity was higher at higher SDS concentrations, which was expected for quantitative analysis. However, there was a fixed peak at 240–242 nm, and a concentration-dependent peak ranged from 208 nm at a concentration of 1.25 wt % to below 190 nm at concentrations less than 0.005 wt % nm.

Figure 5

UV–Vis analysis and results. (a) UV–Vis absorbance of the surfactant solution at different SDS concentrations. (b) Different SDS concentrations at an MWCNT concentration of 0.001 wt %. The inset figure shows the absorbance at wavelengths between 250 and 300 nm. (c) Different MWCNT concentrations at an SDS concentration of 0.05 wt %.

Figure b shows the absorbance due to the presence of MWCNTs at a fixed MWCNT concentration of 0.001 wt % and different SDS concentrations (curves in Figure a were used as baselines to subtract the absorbance due to the SDS). High absorbance peaks were observed at 260–264 nm, which entirely agrees with other studies showing that the absorbance peak of an individual MWCNT was around 260 nm.[6] The intensity peaks were convergent regardless of the concentration of the surfactant. The absorbance curve drops at concentrations of 0.5 and 1 wt %, which indicates an adverse effect of the surfactant at high concentrations in addition to its pre-effect in increasing the medium electrical connectivity. The absorbance also decreased at a concentration of 0.01 wt %, indicating low solubility of MWCNTs at an SDS concentration below 0.01 wt %. In Figure c, the SDS concentration was maintained at 0.05%, while the MWCNT concentration was varied from 0.00125 to 0.01 wt %. The higher concentration of MWCNTs results in more single tubes and thus a higher absorbance peak. The selection of the MWCNT concentration usually depends on the required density of the deposited MWCNT layer.

In conclusion, the conductivity of MWCNT solutions exponentially increased with the increase in the SDS concentration. On the other hand, DLS and UV–Vis analysis showed that the addition of the SDS surfactant improves the MWCNT dispersity and solubility in aqueous solutions. SDS concentrations of ≥0.5 wt % are not desirable due to their adverse effect on MWCNT solubility in addition to the massive increase in the medium electrical connectivity. SDS concentrations lower than 0.01 wt % were not inefficient in dispersing MWCNTs in DIW.

Simulation Results

The role of the SDS concentration in altering the electrokinetic forces can be realized by solving the equations discussed in the theory section. The resulting DEP force, ETH force, and ACEO velocity as a function of SDS concentration are discussed in the following paragraphs.

Figure a shows the magnitude of the ETH and DEP forces as a function of SDS concentration. Assuming that the permittivity of the medium is merely affected by SDS concentrations, the DEP force was almost constant as the SDS concentration increased from 0.001 to 1 wt %.[41] On the other hand, there was a significant increase in the ETH force from 102 to 106 N/m3 as the SDS concentration increased from 0.001 to 1 wt %. Figure b shows that the ACEO velocity increases linearly with the increase in the SDS concentration. For example, at a frequency of 105 Hz, the ACEO velocity increased from 4.67 × 10–7 m/s at a concentration of 0.01 wt % to 6.33 × 10–6 m/s at a concentration of 0.1 wt %. The ACEO velocity is also a function of signal frequency where the velocity can be decreased by 2 orders of magnitude by increasing the signal frequency by 1 order of magnitude at the same SDS concentration.

Figure 6

Simulation results of the DEP, ETH, and ACEO at different SDS concentrations. (a) DEP and ETH forces versus SDS concentration at a point located 10 μm below the electrode edge. The inset figure is the DEP force versus SDS concentration. (b) ACEO velocity vs SDS concentration at different frequencies at a point located 10 μm below the electrode edge.

Figure a shows the drag velocity and the DEP velocity at different depths using different SDS concentrations. The DEP was the dominant velocity near the electrode surface up to −60 μm depth. At depths beyond −60 μm, the drag velocity becomes significant at an SDS concentration above 1 wt %. This means that MWCNTs located below 60 μm are dragged away by the medium motion and cannot reach the deposition area. When the SDS concentration was reduced to 0.1 wt %, the drag velocity became lower than the DEP velocity at depths between −60 and −80 μm. Thus, the DEP velocity can attract MWCNTs from deeper locations. Further reduction in the SDS concentration has no effect as the DEP velocity attenuation is very strong. Figure b shows the velocities at depths below the electrode center. The velocity due to the DEP force was much weaker than the velocity at the electrode edge. However, the drag velocity dominated the DEP velocity at depths beyond −40 μm.

Figure 7

Simulation results of the DEP and drag velocity at different SDS concentrations. (a) At location below the electrode edge. The inset figure illustrates the drag velocity at depths between −50 and −100 μm. (b) At location below the electrode center. The inset figure illustrates the drag velocity at depths between −50 and −100 μm.

The results discussed in Figure were taken at a specific location of the system geometry (electrode edge and electrode center). Figure presents the velocity vectors at three different SDS concentrations below selected electrodes. Figure a–c shows that the intensity of the drag velocity increased with the increase in the SDS concentration. The increase in the velocity was significant at SDS concentrations of >1 wt %. Figure d,e shows that the SDS concentration does not affect the velocity due to the DEP force because no matter how concentrated the solution is, the electrical conductivity will not exceed that of the MWCNTs. The SDS concentration becomes critical in determining the DEP velocity direction (+DEP or −DEP) only when the manipulated particles have electrical conductivity in the same order as the solution.

Figure 8

Velocities induced on MWCNTs in the DEP system. (a) Drag velocity at an SDS concentration of 0.01 wt %. (b) Drag velocity at an SDS concentration of 0.1 wt %. (c) Drag velocity at an SDS concentration of 1 wt %. (d) DEP velocity at an SDS concentration of 0.01 wt %. (e) DEP velocity at an SDS concentration of 1 wt %. (f) Gravitational velocity at an SDS concentration of 1 wt %. The black arrows in the figures represent the direction of the velocity. Note that the simulation results in figure are at a location near the electrode surface. Further simulation results across the entire geometry can be found in Figure S3 (Supporting Information).

Figure f shows that the velocity resulting from the gravitational force was a constant velocity directed to the ground (opposite of the deposition direction). The gravitational velocity of a suspended MWCNT depends on the MWCNT structure and dimensions. The variation in the gravitational velocity was not significant at different MWCNT lengths and densities. However, the increase in the MWCNT diameter significantly increased the gravitational velocity (see Figure S4 in the Supporting Information). Individual tubes have a diameter in the range of a few nanometers up to a few hundred nanometers, while the diameter of MWCNT bundles is equal to the average diameter of a single tube multiplied by the number of the tubes that form the bundle. Large MWCNT bundles that SDS fails to dissolve experience a stronger gravitational force. Thus, ceiling deposition is expected to eliminate the deposition of large MWCNT bundles and result in clean and homogeneous MWCNT networks.

In conclusion, low-conductivity media are required to avoid undesirable electrothermal and electroosmotic flows. Surfactant concentrations higher than 0.1 wt % caused a massive increase in the drag velocity at depths near the electrodes, which obstructs the suspended MWCNTs from reaching the deposition area. Minimum SDS concentration must be used with the help of ceiling deposition to avoid the assembly of undissolved and large MWCNT bundles.

Deposition Results

Deposition of MWCNTs across ITO was successfully conducted using ceiling deposition, as shown in Figure . The solution used in the deposition process has the SDS surfactant at a concentration of 0.05 wt % and MWCNTs at a concentration of 0.001 wt %. MWCNTs were accumulated instantaneously at the electrode edges because of the high-intensity DEP force at the electrode edges (Figure a). The MWCNTs continued to chain and attach to each other until complete connections were formed across the electrode gap (Figure b,c).

Figure 9

Deposition of MWCNTs across ITO electrodes. (a) MWCNTs accumulated at the ITO edges after applying an AC signal of 20 Vpp and 1 MHz. (b) Complete MWCNT connections after a few minutes at the IDE finger head. (c) Complete MWCNT connections after a few minutes at different gaps. (d) MWCNT connections were broken during the removal process. Figure S5 in the Supporting Information shows the alignment on a large scale.

Figure d shows that the MWCNT connections were broken during the removal process. SDS molecules penetrate the gap between the tubes in the solution. However, these molecules break down while drying the medium, reducing the MWCNT–MWCNT contact force. This problem can be solved by diluting the droplet with DIW after forming the MWCNT connections, which could help in maintaining the quality of the aligned MWCNTs.

Conclusions

Surfactants such as SDS are used to improve the dispersity and solubility of MWCNTs in aqueous solutions to form MWCNT suspensions. MWCNT suspensions are used in many applications, including the manipulation of MWCNTs in a microfluidic channel using an electric field. Furthermore, the deposition of MWCNTs from a solution to an electrode structure is widely used in the fabrication of CNT-based devices such as transistors and sensors. In this work, we focused on optimizing the SDS concentration in MWCNT solutions used in DEP systems. The simulation results showed that SDS concentrations of more than 0.1 wt % were not desirable because they caused a massive increase in the medium drag velocity. SDS concentrations low than 0.01 wt % were inefficient in dispersing MWCNTs in DIW. Thus, the optimum SDS concentration in MWCNT solutions for DEP deposition was between 0.1 and 0.01 wt %. The proposed DEP setup successfully assembled MWCNTs from the optimized solution and aligned them across ITO electrodes using an AC signal of 20 Vpp and 1 MHz. Ceiling deposition was preferable in MWCNT assembly from solution with low SDS concentrations because long-duration deposition allows large bundles to move toward the drop surface away from the deposition area. The proposed method and optimized materials have potential use in the fabrication of future transparent wearable electronics such as sensors and detection devices.