Microfluidic Injector Simulation With FSAW Sensor for 3 D Integration

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL 64, NO 4, APRIL 2015 849 Microfluidic Injector Simulation With FSAW Sensor for 3-D Integration Thu Hang Bui, Tung Bui Duc, and Trinh Chu Duc Abstract— This paper presents a possible creation of the optimized liquid sensors for the inkjet nozzles The proposed focused surface acoustic wave (FSAW) device utilizing aluminum nitride (AlN) single crystal as the piezoelectric substrate is based on the pressure variation due to the continuous droplet ejector The design, specification, and numerical simulation results are described Comparisons between the output response of the conventional and concentric structures indicate a more efficient operation of the multiple-segment focused interdigital transducer (FIDT) structure According to the angular spectrum of the plane wave theory, the amplitude field of FIDTs is calculated through that of straight interdigital transducers The 3-D integrated model of the FSAW device has a number of advantages, such as the enhancement of the surface displacement amplitudes and an easier fabrication It is able to detect the breakup appearance of the liquid in the droplet formation process For the piezoelectric substrate AlN, it is compatible with the CMOS fabrication technology, leading to an inexpensive and reliable system Moreover, for the proposed FIDTs with multiple straight segments, the acoustic energy is more optimized and focused near the center of the inkjet nozzle The droplet generation process begins at an output voltage of roughly 0.074 V within 0.25 µs, and the background level of the attenuation of both the mechanical and electrical energy Index Terms— Focused interdigital transducer (FIDT) device, level set method, liquid sensor, microfluidic injector, piezoelectric technology, surface acoustic wave (SAW) devices I I NTRODUCTION I NKJET technology has been applied for various devices such as printers and applications in life sciences (diagnosis, analysis, tissue synthesis, and drug discovery) [1], [2] Inkjet printers are feasible tools for printing texts and images because of their low cost and high resolution within acceptable droplet speed and volume In inkjet technology, factors including the ink droplet volume, head alignment, jet blockage, and resolution may affect the photo-quality image [3], [4] Manuscript received May 31, 2014; revised July 25, 2014; accepted October 12, 2014 Date of current version March 6, 2015 This work was supported by the Vietnam National Foundation for Science and Technology Development through the Nafosted Project under Grant 103.99-2012.24 The Associate Editor coordinating the review process was Dr Deniz Gurkan T H Bui is with the Delft Institute for Microelectronics and Submicron Technology, Delft University of Technology, Delft 2628 CN, The Netherlands; and also with the Department of MicroElectroMechanical Systems and Microsystems, Faculty of Electronics and Telecommunications, University of Engineering and Technology, Vietnam National University, Hanoi, Vietnam (e-mail: hangbt@vnu.edu.vn) T Bui Duc and T Chuc Duc are with the Department of MicroElectroMechanical Systems and Microsystems, Faculty of Electronics and Telecommunications, University of Engineering and Technology, Vietnam National University, Hanoi, Vietnam (e-mail: trinhcd@vnu.edu.vn) Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org Digital Object Identifier 10.1109/TIM.2014.2366975 Failure, however, often occurs in the droplet ejection process When the ink droplet volume and its movement are not controlled properly, bleed and blur might occur at regular break-off intervals in color [5] This causes visual disturbances due to dark blue lines or blurred solids Therefore, advanced technologies, such as inkjet systems with the closed-loop controls that quantify and monitor ejected droplet volumes at the orifice in real time, are needed In other words, the negative feedback of the closed-loop systems may increase accuracy performe between the real aperture and the equivalent aperture of the first input FIDT finger When the number of straight segments increases, the path difference decreases Moreover, FIDTs with multiple segments still have properties similar to concentric circular arc FIDTs C Integrated Injector System Fig shows the geometry of the integrated injector During the formation of a droplet, a generated fluidic pressure forces the nozzle wall The sensor is positioned at the nozzle, consisting of the transmitter and receiver FIDTs to detect the change of the liquid pressure by the deformation of the output response and thereby detect the droplet formation process By detecting the amplitude and attenuation of the mechanical and electrical output signal, the necessary information about 852 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL 64, NO 4, APRIL 2015 TABLE I D ESIGN PARAMETERS OF IDT Fig Position of the air/ink interface and velocity field at (a) t = 13 μs and (b) t = 14 μs Fig Inlet velocity is excited by one pulse within the first 14 μs the droplet generating process inside the ink channel can be extracted To reduce the leaky SAW effect into the ink motion, which may cause jetting failure, the applied electrical energy at transmitter FIDTs should be enough to still receive the output potential For FIDTs, the high-intensity and narrow SAW beam is mostly focused on a part of the nozzle This also improves the sensitivity of the SAW device as the charge distribution on FIDTs is caused by the change of efficiency for SAW detection as well as excitation III S YSTEM C ONFIGURATION A FSAW Configuration In this section, we focus on the SAW sensor configuration on the Aluminum Nitride single crystal, which is integrated into the injector To build a 3-D model of the integrated sensing of the droplet volumes during generation, the size of the 3-D domain of piezoelectric substrate is the rectangle of 500 × 300 μm and the substrate thickness equals the nozzle size of 25 μm The size of the piezoelectric substrate is excessive to decrease wave reflection occurring at the edges FIDTs made of Al film are deposited on the surface The microfluidic channel plays a role at the nozzle of the injector When the number of fingers Fig Positions of ink droplet at various times (a) t = μs (b) t = μs (c) t = μs (d) t = μs (e) t = 11 μs (f) t = 13 μs (g) t = 14 μs (h) t = 25 μs increases, the focusing properties become unstable Therefore, to investigate in steady environment, the model is designed by three pairs of fingers The other design parameters in Table I are as follows The piezoelectric substrate and inkjet of the developed models were meshed adaptively to adjust the scaling of the fields manually and reduce the computation time These parameters provided a much denser mesh at the nozzle boundary of the model, which is essential to achieve a high accuracy in simulations A sinusoidal voltage of frequency 1430 MHz is applied to the input FIDTs to generate the needed SAWs An input voltage of 0.1 V is applied to the receiver FIDTs Moreover, this also avoids receiving very small changes at the receiver because the influence of the liquid pressure that is compared with the input signal needs to be significant The output voltages in all cases are acquired at the alternating fingers of the output IDT Due to the vibration coming from the driving signal and the droplet formation signal, a cross-talk effect including electrical, direct, and pressure-induced crosstalk occurs when T H BUI et al.: MICROFLUIDIC INJECTOR SIMULATION WITH FSAW SENSOR Fig Effect of the piezoelectric substrate on the liquid Fig Sensitivity of the IDT device and two-segment FIDT device 853 the frequencies of these signals are close to each other In experiments, for piezoelectric actuators, passive devices are used to reduce the effective piezoelectric substrates Another way is to use thin foil, external electrodes for the ground and inner electrodes for voltage [20] It is possible to apply these methods for the piezoelectric sensors in experiments In addition, the cross-talk effect of the piezoelectric sensor is much smaller because its frequency is much more than that of the droplet formation signal Moreover, in simulations the use of few IDT fingers and low energy reduces the cross talk B Input Parameters of Ink at the Nozzle The inlet velocity in the z-direction increases from to the parabolic profile during the first μs vi (x, y, t) = 4.5 x + y + 0.1[mm] 0.2[mm] × 1− x + y + 0.1[mm] 0.2[mm] · v(t)(mm/s) (9) 10−6 ) 10−6 ), − u(t − 13 · as shown Here, (t) = u(t − · in Fig 4, and u(t) is the unit function Hence, the pulse Fig Total amplitude fields of IDTs with the conventional and concentric shapes on the surface (a) Conventional IDTs (b) FIDTs with circular arcs (c) FIDTs with two straight segments (d) FIDTs with three straight segments frequency of the droplet formation process is about 20 KHz The velocity is then v(x, y) within 10 μs and finally falls down to zero within another μs Therefore, the ink velocity at the nozzle throat is sought in the following form: v n (x, y) = v i (x y) R12 R22 (10) 854 Fig 10 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL 64, NO 4, APRIL 2015 Total displacement measured at a point after the inkjet nozzle Fig 12 Spectral content of the mechanical wave motion of the FSAW devices with (a) curve fingers, (b) two-straight-segment fingers, and (c) three-straight-segment fingers A Droplet States Fig 11 Mechanical attenuation of SAWs after propagating through the inkjet nozzle The surface tension of the ink generates a capillary pressure that is ignored due to its insignificant influence To cut off the droplet, the pressure at the entrance of the nozzle has to overcome steady and unsteady inertia and forces resulting from the surface tension of the ink [20] Therefore, pressure includes the positive and negative excitation pressure After the negative excitation, the ink deformation at the nozzle happens to separate the droplet from the liquid reservoir Fig shows the ink surface and the velocity field at t = 13 μs when the velocity magnitude of ink is still focused at the nozzle After 14 μs, the breakup phenomenon of the droplet generation occurs Fig shows the time evolution of the ink jetting from the nozzle To move to the outlet of the target, the jetted droplet from the inlet needs 200 μs During the first 13 μs, ink at the nozzle throat is extensively forced [Fig 6(a)–(f)] After the second actuation pressure, the breakup point occurs, as shown in [Fig 6(g) and (h)] In other words, the potential energy becomes strong enough to cut off the droplet Therefore, to detect the initial period of the droplet generation, the running time of the simulation only needs to be carried out within 25 μs to determine the correlation between the droplet generation and the output signal variation B Working Mechanism of the FSAW Device IV R ESULTS AND D ISCUSSION The proposed simulation methodology has been implemented using finite element method and COMSOL Multiphysics 4.2a Pressure produced by the piezoelectric substrate insignificantly affects the liquid (Fig 7) Moreover, it also indicates that due to the uniform IDT fingers of the conventional, the liquid is influenced more at the region far from the local line T H BUI et al.: MICROFLUIDIC INJECTOR SIMULATION WITH FSAW SENSOR Fig 13 855 Output potential at the receiver FIDT of the SAW sensors The sensitivity S is defined as the relative change of the output signal per unit of the applied pressure and the input voltage [21] Fig shows that that of the two-segment FIDT structure is better than that of the conventional IDT structure Fig shows that the total displacement fields of FIDTs have a narrow concentric SAW beam When the number of straight segments of the proposed structure increases, its SAW beam resembles that of curve FIDTs Moreover, the total displacement magnitude of FIDTs with multiple segments is close to that of FIDTs with circular arcs in Fig 10 In Fig 11, the attenuation of the mechanical waves is almost due to the leaky wave phenomenon and the ink pressure Simulation results for four structures also show that the mechanical energy of the FIDTs is lower In other words, the FSAW devices organize more efficiently than the conventional devices The mechanical waves of the FIDT structure are observed in frequency-time domain in Fig 12 The spectrum of the mechanical motion at the output fingers in all focused structure cases illustrates that the total mechanical energy mostly focuses at 5.5 μs and it has other subharmonics Hence, the performance of the proposed multiple-segment FIDTs is similar to that of FIDTs with circular arcs In Fig 13, the contour plot illustrates the output signals of the FSAW devices at times ranging from to 25 μs The output signal of the FIDT structures is larger than that of the conventional IDT structure When it is excited by the first actuation pressure, the maximum voltage value still achieves 0.128 V After 13 μs, its velocity is able to overcome the surface tension force and becomes strong enough to cut off the droplet The breakup point may occur at around 0.074 V in this duration of 0.25 μs window (ranging from 13.2 to 15.7 μs) Hence, the alteration of the electrical signal at different Fig 14 Insertion loss of the output signal of the conventional and focused SAW devices with (a) conventional fingers, (b) curve fingers, and (c) 3straight-segment fingers generated pressures positioned at the nozzle wall and throat depends on the ink state For each droplet formation period, the attenuation responses of conventional and concentric fingers are shown in Fig 14 When all attenuation results of the electrical energy reach the background level, the separation process begins The separated droplet process keeps on moving due to inertia although the excitation impact does not exist After generating the droplet, as inertia oscillates, the significant attenuation continues and reduces gradually until the liquid surface tension returns to its resting state As the power of the conventional structure is 856 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL 64, NO 4, APRIL 2015 dissipated around the medium and more absorption happens at the edges, the energy loss is highest Consequently, it is proved that the proposed FSAW devices not only keep the advantageous properties of circular arcs, but like conventional IDTs, they are also quite sensitive to the actuation pressures of the inkjet nozzle V C ONCLUSION This paper presented a novel sensor for discovering the pressure variation at the inkjet nozzle The relation between the ink pressure at the nozzle and the wave motion was found in the equation of motion for the piezoelectric medium Based on the voltage, output power, and attenuation response of the electrical and mechanical signal, it is able to detect the droplet formation at the inkjet orifice For the proposed FIDTs with multiple straight segments, the SAW beam is similar to that of the FSAW device with circular arcs The greater the number of straight segments they get, the more their properties resemble circular arc FSAW devices In addition, it influences insignificantly the flow rate at the nozzle due to the narrow SAW beam focused mostly on small arcs of the inkjet nozzle Moreover, because of its straight shape, the proposed device is easier to fabricate For the proposed FIDTs with multiple straight segments, based on the saturation state of the attenuation response of the electrical signal, it is still able to monitor the injected droplet process, such as estimating the beginning of the droplet generation process The output signal may achieve up to 128 mV for the positive excitation pressure and down to approximately 74 mV for the negative excitation pressure The breakup point keeps the potential value of 74 mV within 0.25 μs R EFERENCES [1] W Siqun and C Qiong, “Microarray analysis in drug discovery and clinical applications,” in Bioinformatics and Drug Discovery, vol 316 New York, NY, USA: Humana Press, 2006, pp 49–65 [2] G Wiederrecht, Handbook of Nanofabrication Amsterdam, The Netherlands: Elsevier, 2010 [3] W S Rone and P Ben-Tzvi, “MEMS-based microdroplet generation with integrated sensing,” in Proc COMSOL Conf., 2011 [4] M Elewenspoek and R Wiegerink, Mechanical Microsensors New York, NY, USA: Springer, 2001 [5] A Hladnik, T Muck, and G Novak, “Quality evaluation of ink-jet paper with principal components analysis,” Int J Syst Sci., vol 33, no 8, pp 677–687, 2002 [6] J Wei, Silicon MEMS for Detection of Liquid and Solid Fronts Zutphen, The Netherlands: Wöhrmann Print Service, 2010 [7] J Wei, P M Sarro, and C D Trinh, “A piezoresistive sensor for pressure monitoring at inkjet nozzle,” in Proc IEEE Sens Conf., Nov 2010, pp 2093–2096 [8] H C Wu and H J Lin, “Effects of actuating pressure waveforms on the droplet behavior in a piezoelectric inkjet,” Mater Trans., vol 51, no 12, pp 2269–2276, 2010 [9] T.H Bui, T Bui Duc and T Chu Duc, “Microfluidic injector simulation with SAW sensor for 3D integration,” in Proc IEEE Sens Appl Symp., Feb 2014, pp 213–218 [10] S Shiokawa and J Kondoh, “Surface acoustic wave sensors,” Jpn J Appl Phys., vol 43, no 5B, pp 2799–2802, 2004 [11] T.H Bui and T Chu Duc, “Multilayer SAW device for flow rate sensing in a microfluidic channel,” in Proc IEEE Sens Conf., Baltimore, MD, USA, Nov 2013, pp 487–490 [12] D S Ballantine et al., Acoustic Wave Sensors—Theory, Design and Physico—Chemical Applications New York, NY, USA: Academic, 1997 [13] R Singh and V R Bhethanabotla, “Design of mutually interacting multi-directional transducer configurations on a surface acoustic wave device for enhanced biosensing,” in Proc IEEE Sens Conf., Oct 2009, pp 1044–1047 [14] M Kirci and E Akcakaya, “Analysis of focused surface wave transducers,” IEE Proc G Circuits, Devices, Syst., vol 137, no 6, pp 467–469, Dec 1990 [15] P Marechal, N Felix, F Levassort, L.-P Tran-Huu-Hue, and M Lethiecq, “P3P-7 modeling of lens focused piezoelectric transducer for medical imaging,” in Proc IEEE Ultrason Symp., Oct 2006, pp 2341–2344 [16] T.H Bui, T Bui Duc and T Chu Duc, “An optimization of IDTs for surface acoustic wave sensors,” in Proc IWNA, 2013, pp 159–162 [17] H Wijshoff, Structure- and Fluid-Dynamics in Piezo Inkjet Printheads Paris, France: Hermann & Cie, 2008 [18] T.-T Wu, H.-T Tang, Y.-Y Chen, and P.-L Liu, “Analysis and design of focused interdigital transducers,” IEEE Trans Ultrason., Ferroelectr., Freq Control, vol 52, no 8, pp 1384–1392, Aug 2005 [19] D P Morgan, Surface-Wave Devices for Signal Processing New York, NY, USA: Elsevier, 1985, pp 129–155 [20] H Wijshoff, “The dynamics of the piezo inkjet printhead operation,” Phys Rep., vol 491, nos 4–5, pp 77–177, 2010 [21] M.-H Bao, Micro Mechanical Transducers—Pressure Sensors, Accelerometers and Gyroscopes Amsterdam, The Netherlands: Elsevier, 2000 Thu Hang Bui received the B.Eng degree in electronics and telecommunications from the Hanoi University of Science and Technology, Hanoi, Vietnam, in 2010, and the Master’s (Hons.) degree from the Department of Electronics and Telecommunications, University of Engineering and Technology, Vietnam National University (VNU), Hanoi, in 2013 She is currently pursuing the Ph.D degree from the Delft University of Technology, Delft, The Netherlands She has been an Assistant Lecturer with the University of Engineering and Technology, VNU Her current research interests include microfluidic sensor, actuator, and piezoelectric technology Tung Bui Duc received the B.S degree in electronics and telecommunications from the University of Engineering and Technology, Vietnam National University, Hanoi, Vietnam, in 2013, where he is currently pursuing the M.Sc degree in microelectromechanical systems with a focus on piezoelectric and piezoresistive sensors, and microsystem technology Trinh Chu Duc received the B.S degree in physics from the Hanoi University of Science, Hanoi, Vietnam, in 1998; the M.Sc degree in electrical engineering from Vietnam National University (VNU), Hanoi, in 2002; and the Ph.D degree from the Delft University of Technology, Delft, The Netherlands, in 2007 His doctoral research concerned piezoresistive sensors, polymeric actuators, sensing microgrippers for microparticle handling, and microsystems technology He is currently an Associate Professor with the Faculty of Electronics and Telecommunications, University of Engineering and Technology, VNU Since 2008, he has been the Vice Dean of the Faculty of Electronics and Telecommunications Since 2011, he has been the Chair of the Department of MicroElectroMechanical Systems and Microsystems He has authored or co-authored over 70 journal and conference papers and patents Dr Chu Duc was the recipient of the VNU Young Scientific Award in 2010 at the 20th Anniversary of the Delft Institute of Microsystems and Nanoelectronics, the Delft University of Technology Best Poster Award in 2007, and the 17th European Workshop on Micromechanics Best Poster Award in 2006 He was a Guest Editor of the Special Issue of the MicroElectroMechanical Systems, Vietnam Journal of Mechanics, in 2012  ... T.H Bui, T Bui Duc and T Chu Duc, Microfluidic injector simulation with SAW sensor for 3D integration, ” in Proc IEEE Sens Appl Symp., Feb 2014, pp 2 13 218 [10] S Shiokawa and J Kondoh, “Surface... signal and the droplet formation signal, a cross-talk effect including electrical, direct, and pressure-induced crosstalk occurs when T H BUI et al.: MICROFLUIDIC INJECTOR SIMULATION WITH FSAW SENSOR. .. Total amplitude fields of IDTs with the conventional and concentric shapes on the surface (a) Conventional IDTs (b) FIDTs with circular arcs (c) FIDTs with two straight segments (d) FIDTs with three