Utilizing a high precision microfluidic dead-end filtration setup to achieve global alignment of single walled carbon nanotubes

This application note on how to utilize high precision microfluidic for single walled carbon nanotube alignment was written by Christian Rust, Han Li, Georgy Gordeev, Manuel Spari, Markus Guttmann, Qihao Jin, Stephanie Reich & Benjamin S. Flavel.

A short introduction to the field of research

Single-wall carbon nanotubes (SWCNTs) are considered as quasi 1-dimensional (1D) carbon nanostructures, which are known for their outstanding anisotropic electronic, mechanical, thermal and optical properties. They can be understood as a rolled-up sheet of graphene, where the diameter of the nanotube is determined by the chiral vectors (n,m) of the hexagonal arrangement of carbon atoms [1]. Depending on the diameter, nanotubes can exhibit metallic or semiconducting properties and are consequently found in energy [2], photonics [3] and electronics [4].

Despite continuous research, difficulties to control large-scale thin film architectures are limiting their application range. Recently, alignment of SWCNTs with a vacuum filtration method has shown to facilitate high degrees of alignment over an area of A= 1 cm², but remains challenging to reproduce, due to a lack of control of the process parameters [5]. Hence, we developed a filtration setup, which is capable of producing large areas (A= 3.81 cm²) of aligned carbon nanotubes benefitting from the precision of microfluidic components.

Advantages

• Zero Flow condition in the beginning of the experiment, due to the use of valves
• Precise control of volume rate (with automatically optimized PI feedback loop)
• Precise inline measurement of transmembrane pressure, filtration resistance and membrane retention
• Reproducibility 

Applications 

How this experimental protocol could be used?

Due to its precision, this setup can be used to study the fouling of membranes with nanoparticles and the alignment of other nanotubes or other one-dimensional nanomaterials.

The Setup

In our custom-made filtration cell shown in Animation 1, the membrane is placed on a 300 µm fine stainless-steel mesh laser-welded to a thin metal ring. The inner diameter and thus the effective filtration diameter is 22 mm, while the remaining membrane area is used to pinch the membrane in, using Teflon cylinders from either side. The intermediate flange fixes the membrane in place and allows for the addition of feed solution through the top, whilst the second makes a gas tight seal for filtration. The inlet and outlet of the stainless-steel sleeve can be connected via 1/4”-28 UNF standard threads and we use 1/8” and 1/16” polytetrafluoroethylene (PTFE) tubing with an inner diameter of 2 mm and 1 mm to connect nitrogen gas and fluid, respectively. Positive pressure is provided through the inlet by electronic pressure regulator rated to 0 bar – 2 bar operation.  The outlet of the filtration cell is met by an arrangement of two 3/2 solenoid valves, which can be actuated to pass the permeate either through a MPS 2, MPS 1 pressure piezoelectric sensor or stop the flow completely.  After the pressure sensors, the permeate passes into a bubble trap and the BFS 1+ Coriolis flow sensor followed by a UV-diode array. .The gravity flow of the system without a membrane in place and with the solenoid valve open was adjusted to a constant value of 100 µL min-1 by raising or lowering the position of the waste container relative to the height of the filter holder.  In this way, the fluid flow driven by gravity in the setup was hindered by the backpressure of the water in the waste bottle with respect to its relative height to the filtration cell.  The complete system shown in Figure 1 (A) and (B) can be operated in constant inlet pressure or constant volume rate, while the latter is realized with a PI controller using an automated first order plus dead time (FOPDT)- model, which evaluates the controller parameters of the response of an inlet pressure step.

Animation 1. The filtration cell being assembled

Figure 1. (A) Filtration setup without UV diode array (B) Schematic of the complete microfluidic dead-end filtration setup.

The list of components used to perform the experimental protocol

• Gas supply: Nitrogen gas 99.999% (Alphagas)
• Regulator: OB 1 MK3
• Filter holder: Custom made filter holder
• Valves: 3/2 Solenoid Valve 299250 24 V DC (Bürkert)
• Valve controller: Arduino UNO with relays and 24 V power source
• Inline pressure sensors: MPS 1 and MPS 2
• Bubble trap: Bubble trap 44 µL
• Flow sensor: BFS 1 +
• UV array: Agilent G2258A UV diode array
• Exit: waste bottle on lift

Suspensions of SWCNTs

Electric arc SWCNTs (40 mg, lot no. 02-A011, Carbon Solutions) were dispersed in deionized water (18.2 MΩcm, pH = 6.93) from an Arium pro UV (Sartorius) using 2 wt% sodium deoxycholate (BioXtra, 98+%). After tip sonication for for 45 min (0.9 WmL-1) in an ice bath, the dispersion was centrifuged at 45,560g for 1 h (Beckman Optima L-80 XP, SW 40 Ti rotor). The obtained supernatant was than enriched  by filtering them through a 300 kDa membrane (NMWL Biomax polyethersulfon, Merck Milipore) in order to reduce the water and surfactant content. For a typical filtration experiment, we used 20 ml containing 1.2  μg mL-1 SWCNTs and 0.006 wt% sodium deoxycholate.

Membranes

Membranes were obtained from it4ip with a diameter of 47 mm, a porosity of 6 × 108 and a thickness of 25 µm.

The experimental protocol

Prior to filtration, the setup needs to be filled up with deionized water reaching the filter sieve in the filtration holder (cf. Animation 1 A). The membrane can then be placed on to the sieve and should remain slightly wetted. After reassembling the filter holder without the lid, a slight pressure due to compressed water below the membrane should be released by opening the valves for a short time. After the system is at equilibrium, the SWCNT dispersion is carefully added by using a pipette before closing the lid. At this point, a flow program is chosen, consisting of a slow filtration regime, where the carbon nanotubes can form crystallites or attach to the membranes and a fast filtration regime, which is used to freeze the alignment into position. 

For a 20 ml dispersion filtered onto an 80 nm pore sized membrane, the slow volume rate is usually set to 100 µL min-1 for 19,250 mL and then ramped up to 500 µL min-1 for the final pushing step, Figure 2. During filtration, the pressures of the regulator pin and chosen inline pressure sensor pout are recorded as well as the trans membrane pressure P_TMP = P_in– P_out . Using Darcy’s law in combination with the volume rate then yields the total resistance R_tot, whose curvature with respect to permeated volume shows how the carbon nanotubes film is growing on the membrane using filtration blocking-laws. The UV-diode array can be used additionaly to gather information on the retention of the membrane in place.

Figure 2. Experimental parameters recorded during a 20 mL filtration of an 1.2 μg mL-1 EA-SWCNT dispersion in 0.006 wt % DOC on a 80 nm pore sized membrane. The volume rates were set to 100 µL min-1 and 500 µL min-1, respectively, while the pressures (P_in, P_out and P_TMP= P_in– P_out) are measured and the total resistance R_tot is calculated.

Some glimpse of results

In Figure 3 (A), several cross-polarized light images of the carbon nanotube film on a membrane were stitched together, showing global alignment. Additionally a scanning electron microscopy image is provided in Figure 3 (B). Therefore, the SWCNT-film has been transferred to silicon by using chloroform to dissolve the membrane and 2-propanol to clean the surface.

Figure 3. (A) Cross-Polarized image of the global aligned membrane in bright (45 %) and dark (0%) orientation .                                      (B) Scanning electron microscopy of aligned single-walled carbon nanotubes utilizing our microfluidic dead-end filtration setup.

Acknowledgement

This application note is based on the paper Rust, C., Li, H., Gordeev, G., Spari, M., Guttmann, M., Jin, Q., … & Flavel, B. S. (2021). Global Alignment of Carbon Nanotubes via High Precision Microfluidic Dead‐End Filtration. Advanced Functional Materials, 2107411.

References