Here we present a protocol to construct a pressure-controlled syringe pump to be used in microfluidic applications. This syringe pump is made from an additively manufactured body, off-the-shelf hardware, and open-source electronics. The resulting system is low-cost, straightforward to build, and delivers well-regulated fluid flow to enable rapid microfluidic research.
Microfluidics has become a critical tool in research across the biological, chemical, and physical sciences. One important component of microfluidic experimentation is a stable fluid handling system capable of accurately providing an inlet flow rate or inlet pressure. Here, we have developed a syringe pump system capable of controlling and regulating the inlet fluid pressure delivered to a microfluidic device. This system was designed using low-cost materials and additive manufacturing principles, leveraging three-dimensional (3D) printing of thermoplastic materials and off-the-shelf components whenever possible. This system is composed of three main components: a syringe pump, a pressure transducer, and a programmable microcontroller. Within this paper, we detail a set of protocols for fabricating, assembling, and programming this syringe pump system. Furthermore, we have included representative results that demonstrate high-fidelity, feedback control of inlet pressure using this system. We expect this protocol will allow researchers to fabricate low-cost syringe pump systems, lowering the entry barrier for the use of microfluidics in biomedical, chemical, and materials research.
Microfluidic tools have become useful for scientists in biological and chemical research. Due to the low volume utilization, rapid measurement capabilities, and well-defined flow profiles, microfluidics has gained traction in genomic and proteomic research, high-throughput screening, medical diagnostics, nanotechnology, and single-cell analysis1,2,3,4. Furthermore, the flexibility of microfluidic device design readily enables basic science research, such as investigating the spatiotemporal dynamics of cultured bacterial colonies5.
Many types of fluid injection systems have been developed to accurately deliver flow to microfluidic devices. Examples of such injection systems include peristaltic and recirculation pumps6, pressure-controller systems7, and syringe pumps8. These injection systems, including syringe pumps, are often composed of expensive precision engineered components. Augmenting these systems with closed-loop feedback control of pressure in the output flow adds to the cost of these systems. In response, we previously developed a robust, low-cost syringe pump system that uses closed-loop feedback control to regulate outputted flow pressure. By using closed-loop pressure control, the need for expensive precision-engineered components is abrogated9.
The combination of affordable 3D-printing hardware and a significant growth in associated open-source software has made the design and fabrication of microfluidic devices increasingly accessible to researchers from a variety of disciplines10. However, the systems used to drive fluid through these devices remain expensive. To address this need for a low-cost fluid control system, we developed a design that can be fabricated by researchers in the lab, requiring only a small number of assembly steps. Despite its low-cost and straightforward assembly, this system can provide precise flow control and provides an alternative to commercially available, closed-loop syringe pump systems, which can be prohibitively expensive.
Here, we provide protocols for the construction and use of the closed-loop controlled syringe pump system we developed (Figure 1). The fluid handling system is composed of a physical syringe pump inspired by a previous study11, a microcontroller, and a piezoresistive pressure sensor. When assembled and programmed with a proportional-integral-derivative (PID) controller, the system is capable of delivering a well-regulated, pressure-driven flow to microfluidic devices. This provides a low-cost and flexible alternative to high-cost commercial products, enabling a broader group of researchers to use microfluidics in their work.
1. 3D-printing and Assembly of Syringe Pump
2. Microfluidic Device Preparation
3. Feedback-controlled Syringe Pump System Assembly
4. Pressure Sensor Calibration
NOTE: Based on the amplifier chosen in this paper, the formula to calculate the gain is G = 5 + (200k/RG) with RG = R1 and G = amplifier gain. The amplifier gain here is approximately 606. This value can be changed by changing the resistance used for R1. In addition, as the logic level of the microcontroller board is 5 V and the instrumentation is powered with 10 V, a simple voltage divider circuit, R2 and R3, is used to safeguard the output signal to be no more than 5 V.
5. Capturing Images from the Microfluidic Device
6. Controlling Syringe Pressure Pumps
7. Tuning the PID Controller Parameters
NOTE: The ideal controller parameter values may vary depending on the application and the microfluidic device geometry. For example, for long-term studies (hours), a lower proportional constant (Kp) may be preferable to minimize overshoot at the expense of response time. These tradeoffs depend on experimental conditions and objectives.
Here, we present a protocol for the construction of a feedback-controlled syringe pump system and demonstrate its potential uses for microfluidic applications. Figure 1 shows the connected system of the syringe pump, pressure sensor, microfluidic device, microcontroller, pressure sensor circuit, and stepper motor driver. Detailed callouts for the syringe pump assembly are shown in Figure 2 and the electronic circuit schematic for pressure sensing is presented in Figure 3. The process of tuning the controlling parameters is shown in Figure 4. Finally, a representative result of controlling inlet pressure in a two-inlet Y-shaped microfluidic device is shown in Figure 5.
Figure 1: Setup of the feedback-controlled syringe pump system. This image shows the setup of the syringe pump system. The syringe contains the solution for injection and is actuated by the 3D-printed syringe pump. As A. the piezoresistive pressure sensor is connected with B. the syringe pump and C. the microfluidic device, the pressure from the device is detected and converted into an electrical signal to D. the pressure sensor circuit with instrumentation amplifier once the liquid is delivered through the tubing. The signal from the pressure sensor is read by E. the open-source microcontroller board which then transmits the necessary signal to F. the stepper motor driver to control the actuation of the syringe pump. G. A power supply and H. a laptop is needed to operate and program the system. Please click here to view a larger version of this figure.
Figure 2: Assembly photo for the 3D-printed syringe pump. This figure shows the step-by-step instructions for the 3D-printed syringe pump assembly, with photos corresponding to the procedure in step 1.2 of the protocol. A. This image shows the materials for the syringe pump assembly. B. This image shows how the stepper motor is connected to the threaded rod (step 1.2.1). C. This image shows how the part from step 1.2.1 of the protocol is connected to the part from step 1.2.2 of the protocol (step 1.2.3). D. This image shows the assembly of the traveler push piece (step 1.2.5). E. This image shows how the end stop is connected (step 1.2.10). F. This image shows how the syringe plunger female connector piece is connected to the assembled components (step 1.2.11). G. This image shows the assembly of the syringe plunger male connector piece (step 1.2.13). H. This image shows how the syringe clamp is connected (step 1.2.14). Please click here to view a larger version of this figure.
Figure 3: Illustration for the microcontroller and pressure sensor circuit. The circuit allows the microcontroller board to measure amplified pressure signals from the pressure sensor. A. This is the assembly photo for the circuit. B. This figure shows the circuit board layouts. The exposed wires from the pressure sensor are color-coded and should be connected as follows: red should connect to V+, black should connect to V-, green should connect to Signal+, and white should connect to Signal-. Please click here to view a larger version of this figure.
Figure 4: Tuning of control parameters. The PID controller used to regulate the syringe pump fluid pressure may be tuned by modifying the proportional (Kp), integral (Ki), and differential (Kd) parameters. Here, we show how tuning (using Kp) will help to reduce the response time. Further tuning (using Ki and Kd) can help to ensure a setpoint stability and reduce overshoot. In this protocol, controllers are primarily tuned using a manual trial-and-error approach. Please click here to view a larger version of this figure.
Figure 5: Control of inlet pressure for a laminar flow microfluidic device. A Y-shaped microfluidic device is fabricated following the procedure detailed in step 2 of this protocol. The device features two inlet ports and one outlet port. Two syringe pump systems are assembled to control the inlet pressures. One of the syringes is loaded with a blue dye and the other is loaded with water. A. These images of the fluid flow resulting from the same pressure provided by both pumps are captured using the approach detailed in step 6 of this protocol. B. This figure shows how the inlet pressures are monitored and controlled using the PID controller tuned in Figure 4. Close adherence to the set point can be observed. Shorter (s) and longer (h) experiments have demonstrated similar results. Please click here to view a larger version of this figure.
Supplemental Files. Please click here to download the files.
Here, we presented a new design for a syringe pump system with closed-loop pressure control. This was accomplished by integrating a 3D-printed syringe pump with a piezoresistive pressure sensor and an open-source microcontroller. By employing a PID controller, we were able to precisely control the inlet pressure and provide fast response times while simultaneously maintaining the stability about a set point.
Many experiments using microfluidic devices require a precise fluidic control and exploit a well-characterized laminar flow profile. Examples where a stable flow profile is important include experiments that explore temporal and spatial concentration gradients15 and generate precise fluidic encapsulations for further analysis16. By using a PID controller to maintain the high-performance response, the system described in this protocol produces the flow regulation and long-term stability necessary to study such laminar flow experiments.
However, it is important to recognize that microfluidic devices and experiments involving them have subtle variations and differences. For instance, different microfluidic geometries (channel width and height) may necessitate different flow profiles. As a result, the parameters for the PID controllers must be tuned accordingly. Additionally, some experiments may require a tight regulation of the pressure ranges. In these cases, the pressure overshoot may not be acceptable. As such, the PID control parameters must be tuned so that the overshoot is minimized, usually at the expense of response time.
Due to the low-cost production of this syringe pump system, researchers should be able to rapidly develop microfluidic experiments. The estimated cost for a 3D-printed syringe pump, microcontroller, and pressure sensor circuit is approximately US$130. In contrast to commercially available alternatives, such as peristaltic and recirculation pumps, this syringe pump system provides a flexible and straightforward platform that may be adapted to a variety of laboratory uses. Although not discussed here, simpler control strategies, such as the bang-bang controller, may be used for long-term microfluidic studies. Additionally, the syringe pump systems may be used to apply a vacuum pressure to a control volume.
One potential limitation of this syringe pump system using a PID controller is the reliance on a constant power supply. Because the PID control method requires the constant energization of the stepper motor, there is a relatively large power requirement. In contrast, the bang-bang controller only energizes the stepper motor when necessary, using substantially less power. This power requirement may be mitigated by developing a hybrid control structure that implements a PID controller to initially reach a set-point range, and then de-energizes the stepper motor coils once the pressure value is within a given set-point range. Alternatively, a simple bang-bang controller may be used as well.
Additionally, this syringe pump system allows for a flexible performance and control by altering the size of both the stepper motor and the syringe itself. In previous experiments, we have used syringes of 1 mL, 5 mL, 10 mL, and 30 mL. Naturally, each syringe pump may necessitate slightly different PID controller parameters and would, therefore, require individualized parameter tuning. However, this flexibility allows the syringe pump system described in this protocol to be used in a range of applications.
It should be noted that a common area of microdevice failure is an inability to effectively bond the PDMS to the cover glass. For the microfluidic device fabrication, the power of the plasma cleaner should be optimized if the binding is ineffective. Also, any lubricants or impurities on the cover glass' surface should be removed prior to the bonding to ensure a strong bond with the PDMS. Thoroughly washing and removing any dust from the PDMS component should help to ensure a good seal is formed between the PDMS and the glass.
The low-cost, feedback-controlled syringe pump system presented here allows researchers to manipulate the fluid profile with a high degree of stability in a flexible manner. By integrating the pressure sensing module with simple PID control methods, the system is able to provide high-performance pressure-driven flow control. This tool can be broadly applied across many research fields where microfluidics tools are in use.
The authors have nothing to disclose.
The authors acknowledge support from the Office of Naval Research awards N00014-17-12306 and N00014-15-1-2502, as well as from the Air Force Office of Scientific Research award FA9550-13-1-0108 and the National Science Foundation Grant No. 1709238.
Arduino IDE | Arduino.org | Arduino Uno R3 control software | |
Header Connector, 2 Positions | Digi-Key | WM4000-ND | |
Header Connector, 3 Positions | Digi-Key | WM4001-ND | |
Header Connector, 4 Positions | Digi-Key | WM4002-ND | |
Hook-up Wire, 22 Gauge, Black | Digi-Key | 1528-1752-ND | |
Hook-up Wire, 22 Gauge, Blue | Digi-Key | 1528-1757-ND | |
Hook-up Wire, 22 Gauge, Red | Digi-Key | 1528-1750-ND | |
Hook-up Wire, 22 Gauge, White | Digi-Key | 1528-1768-ND | |
Hook-up Wire, 22 Gauge, Yellow | Digi-Key | 1528-1751-ND | |
Instrumentation Amplifier | Texas Instruments | INA122P | |
Microcontroller, Arduino Uno R3 | Arduino.org | A000066 | |
Mini Breadboard | Amazon | B01IMS0II0 | |
Power Supply | BK Precision | 1550 | |
Pressure Sensor | PendoTech | PRESS-S-000 | |
Rectangular Connectors, Housings | Digi-Key | WM2802-ND | |
Rectangular Connectors, Male | Digi-Key | WM2565CT-ND | |
Resistors, 10k Ohm | Digi-Key | 1135-1174-1-ND | |
Resistors, 330 Ohm | Digi-Key | 330ADCT-ND | |
Stepper Motor Driver, EasyDriver | Digi-Key | 1568-1108-ND | |
USB 2.0 Cable, A-Male to B-Male | Amazon | PC045 | |
3D Printed Material, Z-ABS | Zortrax | A variety of colors are available | |
3D Printer | Zortrax | M200 | Printing out the syringe pump components |
Ball Bearing, 17x6x6mm | Amazon | B008X18NWK | |
Hex Machine Screws, M3x16mm | Amazon | B00W97MTII | |
Hex Machine Screws, M3x35mm | Amazon | B00W97N2UW | |
Hex Nut, M3 0.5 | Amazon | B012U6PKMO | |
Hex Nut, M5 | Amazon | B012T3C8YQ | |
Lathe Round Rod | Amazon | B00AUB73HW | |
Linear Ball Bearing | Amazon | B01IDKG1WO | |
Linear Flexible Coupler | Amazon | B010MZ8SQU | |
Steel Lock Nut, M3 0.5 | Amazon | B000NBKLOQ | |
Stepper Motor, NEMA-17, 1.8o/step | Digi-Key | 1568-1105-ND | |
Syringe, 10mL, Luer-Lok Tip | BD | 309604 | |
Threaded Rod | Amazon | B01MA5XREY | |
1H,1H,2H,2H-Perfluorooctyltrichlorosilane | FisherScientific | AAL1660609 | |
Camera Module | Raspberry Pi Foundation | V2 | |
Compact Oven | FisherScientific | PR305220G | Baking PDMS pre-polymer mixture and the device |
Dispensing Needle, 22 Gauge | McMaster-Carr | 75165A682 | |
Dispensing Needle, 23 Gauge | McMaster-Carr | 75165A684 | |
Fisherbrand Premium Cover Glasses | FisherScientific | 12-548-5C | |
Glass Culture Petri Dish, 130x25mm | American Educational Products | 7-1500-5 | |
Plasma Cleaner | Harrick Plasma | PDC-32G | Binding the cover glass with the PDMS device |
Razor Blades | FisherScientific | 7071A141 | |
Scotch Magic Tape | Amazon | B00RB1YAL6 | |
Single-board Computer | Raspberry Pi Foundation | Raspberry Pi 2 model B | |
Smart Spatula | FisherScientific | EW-06265-12 | |
Sylgard 184 Silicone Elastomer Kit | FisherScientific | NC9644388 | |
Syringe Filters | Thermo Scientific | 7252520 | |
Tygon Tubing | ColeParmer | EW-06419-01 | |
Vacuum Desiccator | FisherScientific | 08-594-15C | Degasing PDMS pre-polymer mixture and coating fluorosilane on the master mold |
Weighing Dishes | FisherScientific | S67090A |