Study of concentric, eccentric and split type magnetic membrane micro-mixers
Shao Qiang Tang, King Ho Holden Li, Ze Ting Yeo, Wei Xuan Chan, Say Hwa Tan, Yong-Jin Yoon, Sum Huan Ng
2018
Sensing and Bio-Sensing Research
ABS TRACT Electromagneticall y dri ven elasti c magneti c mi crofluidi c mi xers were inves ti gated for thei r performance in ai r, water and gl ycerol filled chambers. They were fa bri cated by embedding flexible magnets in pol ydi methylsiloxane (PDMS) membrane. At a dri vi ng frequency of 100 Hz, os cillating fluid flow was induced and mi xing was achieved.. Three designs were fabri ca ted and s tudied: a) concentri c type wi th the ma gneti c ma terial in the center of the membrane, b)
more »
... ntri c type wi th the magneti c ma terial offset from the center of the membrane and c) split type wi th two regions of magneti c materials wi th opposing polari ties . The spli t configura tion provides addi tional fluid folding, fa cili tating mi xing of the 20 μL fluorescent dye in 60 μL of sol vent. Simula tion a nd experimental resul ts show that the eccentri c and split designs were able to a chieve a 20-30 % reduction in mi xing time compa red to the concentri c design. At the same ma gneti c flux density, the eccentri c type design exhi bited the grea tes t deflection, explaining the better mi xing a chieved over the concentri c type desi gn. The split type design, ha vi ng the lowes t deflection, was a ble to perform better and more consistentl y than the eccentri c type design by crea ting a "mi cro rocker mi xer" effect. We pos tula ted tha t the shapes of the deflecti on profiles in the eccentri c and split designs contributed to the mi xing effi ciency by promoting better chaotic advection than the concentri c design. Introduction M icrofluidic technology has generated significant interest over the past few years due to the inherent advantages in terms of small volume, response time and cost reduction. This trend is evident based on the increasing number of articles and patents published yearly [1] . An example of a successful microfluidic system application is the handheld patient-side blood testing device [2] . Such devices require mixing of fluids to obtain diagnostic test results, which allow healthcare professionals to make an accurate treatment decision [2]. However, due to the very low Reynolds number (from less than 1 to 10) [3, 4, 31] and omnipresence of laminar flow in microfluidics, fluids can only be mixed through convection and molecular diffusion. This often requires a long period of time (which is in one order magnitude) [31] to achieve complete mixing [3] . Hence, it is apparent that an efficient micro-mixer is needed for such applications. Typically, in microfluidics channels, the diffusion time is on the order of magnitude of 1 for 100 um of diffusion distance, as estimated with the relation -T = L 2 /D, where L is the diffusion distance and D is the diffusion coefficient for the solute in the fluid [30, 31] . For example, the diffusion of sucrose in water over a distance of 100 um, would occur in 20 s, given the diffusion coefficient of sucrose in water to be about 0.5 x 10 -9 m 2 /s [44] . Considering a typical flow rate of around 100 mm/s, adjacent streams of different fluids should flow over distances of 200 cm or more before mixing of sucrose is completed via diffusion. This constraint is undesirable for the miniaturization of the microfluidics devices. On the other hand, complex structures such as grooves and ribs [33, [39] [40] [41] [42] [43] can be used to enhance mixing in passive micro-mixer. However, these structures are often complex in designs which increases the difficulty in the fabrication of the devices. Hence, chaotic, advection-based active micro-mixers are often preferred over passive micro-mixer. Among the advection-based micro-mixers are the T-junction mixers. Researchers have shown that, with the use of oscillating flows confluent at a T -junction, the mixing lengths can be significantly reduced by 2 orders of magnitude, as compared to that required for passive mixing -from tens of centimeters to millimeters [28] . Also, Okkela and Tabeling have shown numerically and experimentally that violent folding of streams occur at a T -junction when the amplitude of flow disturbance normalized against the velocity of flow is greater than the product of the Strouhal number with 2π [29]. Spatial-temporal resonances of a folding quantity have been plotted by Okkels and Tabeling, relating a series of amplitude of flow disturbances and Strouhal number combinations at which folding of the fluid flows is maximized. For an open flow system, the T-junction mixers are unsuitable as the mixing activity is being limited to the dimensions of the junction. A long downstream channel is still required for mixing to be completed passively after the phase-to-phase interface has been stretched. Hence, in this paper, an oscillating flow is applied in a closed environment, such as in a temporarily isolated chamber, to complete the fluids mixing before channeling the mixed fluids to other segments of the lab-on-chip devices. This reduces the overall footprint of the micro-mixer required. There are various methods to gen erate an oscillating flow field. One of which, is to use magnetically driven actuators which are able to provide robust and wireless operations. These actuators are ideal in miniaturized applications [7] and could potentially be applied to several applications in microfluidics. The use of magnetic membranes was reported in many works, demonstrating various actuation purposes [8] , in the form of an elastomer with permanent magnets attached or embedded within [9] or by introducing nano-sized magnetic particles into the polymer matrix [10], which may be structured as a micropillar [11] . However, the implementation of permanent magnets into microsystems may result in cracking and face poor adhesion with the substrate which could affect the performance of the micro-mixers. Furthermore, this is often limited to the size and shape of permanent magnets available commercially. Introduction of magnetic nanoparticles into polymer may result in undesired agglomeration, thus affecting polymer elasticity and performance. Such magnetic composite polymers are also not able to produce large deflection, limiting the efficiency ACCEPTED MANUSCRIPT PDM S, and the structure is partially cured before, a second layer of PDM S was spin-coated. After that, the PDM S was allowed to cure at 80 °C for 60 minutes [11] . This lower-than-typical curing temperature is chosen to achieve a greater flexibility of the PDM S membrane, with a Young's M odulus of around 1.8 M Pa, as achieved also by Johnston et al [12]. This ensures sufficient displacement by the solenoid field, and effective mixing of fluid in the chamber. A typing curing temperature of around 125 o C would result in a Young's M odulus of 2.4 M Pa [12], which would reduce the membrane deflection and resulting pressure on the fluid by 25 %. In addition, Johnston's group presented experimental data that shows that the time for the core of a 2 mm thick PDM S film to reach a curing temperature of 100 o C, is around 13 min or less. Since the PDM S layer of the magnetic membrane in our device is only 25μm or less, such a curing time of 60 min is sufficient for completely curing the PDM S at 80 o C. A similar curing ACCEPTED MANUSCRIPT Dr Holden King Ho Li received the B.Eng. degree from the National University of Singapore, in 1997, and the M.S. and Ph.D. degrees from Stanford University, CA, USA, in 2002 and 2005, respectively, all in mechanical engineering. In 2013, he joined the School of M echanical and Aerospace Engineering, Nanyang Technological University, Singapore, where he currently focus on micro and nanofabrication met hods and M EM S reliability study. Besides, Holden is actively working in the area of BioM EM S. Wei Xuan Chan received his Ph.D. degree in M echanical Engineering from Nanyang Technological University, Singapore in 2016. His current research is in micro-scaled acoustics. Dr S ay Hwa Tan is an ARC DECRA fellow with Queensland M icro-and Nanotechnology Centre, Griffith University, Australia. He received his B.Eng., M .Eng. and Ph.D. degrees from the Nanyang Technological University, Singapore, and the Georg-August-Universität Göttingen/M ax Planck Institute for dynamics and self-organization (M PI-DS), Germany, in 2008, 2010 and 2014 respectively. In 2016, he was highlighted as one of the 18 emerging investigators in the journal of Lab on a Chip. Dr Tan has published more than 30 research works in microfluidics. His research has established and pioneered different approaches to manipulate droplets and bubbles using thermal, magnetic, acoustic and elect ric energy. Dr S um Huan Ng received his B.Eng. and M .Eng. in mechanical engineering in 1997 and 2000 respectively from the National University of Singapore. He obtained his Ph.D. from the Georgia Institute of Technology. He is currently a senior scientist and programme manager for microfluidics manufacturing with SIM Tech working in the areas of microfluidics, lab-on-a-chip, microfabrication techniques and biosensing. ACCEPTED MANUSCRIPT
doi:10.1016/j.sbsr.2018.04.001
fatcat:aydzplkuqrg77f6h2rpregrneq