This review is related to medium injection and medium perfusion solutions for microfluidics. It explains different methods and techniques (active and passive) as an introduction to perfusion for live-cell imaging.
Passive medium injection techniques in microfluidics
Capillary effects are key forces at small scales, in the range of millimeters to micrometers. They are present when one liquid is in contact with any other gas, liquid, or solid. That contact or interface rich to a surface tension which is the force created between both surfaces of the different components when they get in touch. The creation of a new interface has a cost of energy that can be expressed in the following equation: E=σA. Where σ is the surface tension measured in N/m while A is the area of the interface or contact between the two components and E is the force created. As an example, surface tension is the reason why liquid droplets are spherical because they tend to reduce their surface to reduce the forces needed to be stable (1).
As can be seen in figure 1, there are three different contacts between distinct components at different states with various behaviors. The liquid-gas contact, the solid-liquid one, and the solid-gas interface. All of them have different surface tension and the equilibrium of forces is the sum of all the components that at the equilibrium is zero. The horizontal forces are obtained using the Young-Dupré relationà σ (solid-liquid)+ σ cosθ= σ (solid-gas) (2). Taking θ, as a contact angle and having the vertical component asà σsinθ=N. Where N is the reaction of the solid. There is only contact line if -1<cosθ= (σ (solid-gas)– σ (solid-liquid))/ σ <1
Due to the previous condition, there are 4 possible cases:
- Total wetting: when the liquid spreads total and forms a film that covers all surfaces. This only happens when the (σ (solid-gas)– σ (solid-liquid))/ σ >1, so the surface tension between the solid and the gas is much bigger than the other two combined so it is very hard to have a structure over it.
- Partial wetting: when the initial condition is satisfied but there are possible subcases. If the surface is hydrophilic then the contact angle θ is between 0< θ<90°. In the second subcase the surface is hydrophobic so the angle is 90°< θ<180°.
- No wetting: The liquid droplet becomes a pearl and it is a perfect sphere. It only happens when (σ (solid-gas)– σ (solid-liquid))/ σ <-1, which means the solid-liquid surface tension is bigger than the other two combines and that the solid surface does no like to be covered. These surfaces can be natural for example the lotus leaf. (3)
Movement of droplets with surface tension gradient
External changes of the solid-liquid surface energy will provide the forces needed to move small droplets. The direction of the droplet will be given by a gradient in surface energy. Due to this energy, there are two types:
- Thermo-wetting: the contact line comes from the solid-liquid interface when the energy goes up with the temperature. Mathematically expressed as: dσ(solid-liquid)/DT >0. There is a case when the heating is done very locally that the hotter part of the droplet will be harder to move and it will tend to go to the cooler size.
- Electrowetting: It is possible to apply a voltage to a droplet if it is placed between two electrodes, touching them at the same time. The difference of voltage V charges the surface energy due to this equation: σ(EW. solid-liquid) = σ(solid-liquid)-(c*V^2)/2. Where c is the surface capacitance. If only one of the surfaces is in contact with the droplet, it will move towards the electrode, because surface energy is much lower there.
Moving droplet in channels acting on surface tension
Actually, it is possible to move droplets in microchannels, even if the liquid is totally wetting the solid. It can be done by modifying the liquid-gas surface tension.
- Electrocapillarity force: In the case that the droplet is charged on its surface after having been in contact with an electrode. This droplet can be moved if it is under a voltage gradient.
- Thermocapillary forcing: By increasing the temperature of a liquid it is possible to reduce its surface tension. This means that following this equation, dσ (liquid-gas)/DT < 0, a long drop inside a channel with a side hotter than the other will be a move towards the warmer place in order to reduce the total energy of the system.
Advantages and drawbacks of using capillarity
The strongest points of capillarity are its medium injection passivity which allows loading of liquids with pipettes or syringes. In addition, it leaves a very small amount of dead volume (4). On the other hand, the main drawback is that this kind of injection is not an adjustable one because the flow rate due to capillarity will be fixed and dependant on the materials and the measurements of the channels.
It is possible to use gravity to set a flow rate. It is exactly the same principle that is used in medicine to deliver drugs via intravenous using a dropper (5). The key is to calculate the flow rate generated by the gravity force that pushes the liquid down. This calculation can be very complex if the liquid is a mix of different compounds. Softwares such as CONSOL tend to be very helpful to simulate these calculations.
The main issue of this medium injection technique is when the user wants very low flow rates, without applying any kind of force or energy so that the liquid enters in a passive way. This is why many microfluidics devices use reservoirs to load the liquid on the chips. Gravity is more powerful than capillarity. A major drawback is a difficulty to reach high precision flow rate control when using small volumes (6).
Medium injection by osmosis-driven flow
Osmosis-driven flow is the third way to inject medium in a system that does not require any applied force (passive), similarly the same as gravity and capillarity.
The flow rate is induced thanks to a concentration difference between two compartments separated by a permeable membrane. As this membrane is only permeable for the solvent (and not the solutes), the liquid will passively flow through the membrane to reach the equilibrium. In this case, the flow rate generated is proportional to two aspects: the difference in the osmotic pressure across the membrane that is permeable and the contact area of the membrane(6).
The main aspect of this osmotic technique is that they can provide very low flow rates at very stable constant flow and they can work from hours to days, making osmolarity suitable for long-term experiments (7-8). The flow rate being dependent on the solute concentrations at both sides of the membrane, it can easily be tuned.
Active perfusion techniques in microfluidics
Medium injection using syringe pumps
The majority of syringe pumps use a syringe driven by a motor and a screw that can rotate to control the flow. The linear motion allows to control the speed at which the piston is driven As the diameter of the syringe is known, the flow rate follows this equation: Q=vSà, where Q, is the flow rate, v is the speed of the piston and S is the section area of the syringe.
Syringe pumps have been used in a great variety of research areas (10) and the market offers many options fitting various applications, for example, there is a group that used them to trace metals in open ocean seawaters (11) and another team used them to improve technique to inject drugs during dental intravenous sedation (12). However, this technique is a “closed” system, unlike the previous capillarity, gravity, or osmolarity which are “open” systems. For long-term experiments, this has to be taken into account as the experimentation length is directly dependant on the volume of the syringe.
Medium perfusion using vacuum pumps
This system works exactly the same principle as the syringe pumps but instead of pushing the liquid the vacuum pumps apply a negative pressure and pulls the liquid through a conduit. By removing the liquid from the system, the vacuum pump can create a flow rate through the system. This flow rate is very stable over time.
Similar to the syringe pumps, vacuum pumps create an open actuation system. As a consequence, the flow rate monitoring is crucial for long-term experimentations to avoid any issues with perfusion and consequent impacts on cells.
Medium injection using peristaltic pumps
Peristaltic pumps are widely used in laboratories. A positive displacement of a wheel (or rotor) induces pressure on a flexible conduit. Thanks to the wheel’s rollers, the liquid is pushed into the conduit creating an oscillating flow rate inside the tubing.
A key aspect of this injection system is that it can be used backward and forwards just by changing the rotation of the rotor but it has one main drawback. Indeed, the flow created by the rotor tends to be pulsatile or oscillatory instead of constant. Nevertheless, this injection system allows it to work in both configurations: “open” using a reservoir and “close” if the circuit is recirculated. Finally, these pumps are not suitable for direct cells (in their medium) injection as they
Both peristaltic pumps and syringe pumps have no direct contact with the liquid, preventing contamination. The drawback here is that if the cells are going inside the tube they can be compressed while they pass through the rotor.
Medium perfusion using diaphragm pumps
The flow generated by diaphragm pumps is not continuous. The flow is generated by combining a piezoelectric piece and a mobile membrane. The piezoelectric part creates a mechanical movement that is transmitted to the membrane generating a flow. In addition to the un-continuous flow, the membrane can get lost after a certain time.
Medium injection using magnetic stirrer-based micropumps
It is very similar to a macroscopic rotatory pump because it uses mechanical propellers and gives a continuous flow that is controlled by the rotational frequency of a magnetic bar.
As can be seen in this review, the perfusion or injection of the medium is more than important for the development of microfluidic devices applied to cell biology. It allows controlling the flow rate which is the most crucial parameter to avoid shear stress and provide a perfect environment for cells and tissues. This is crucial to handle sophisticated devices and models such as organs-on-chips. Here more information on Cherry Biotech’s Corpus project focusing on Organs-on-chips.
Discover our Medium Perfusion system
Long-term medium perfusion, change, and drug injection
1.Roman B, Bico J, Adamson A W GAP, de Gennes P G B-WFQD, J MM, H MCH and HC, et al. Elasto-capillarity: deforming an elastic structure with a liquid droplet. J Phys Condens Matter [Internet]. 2010 Dec 15 [cited 2017 Jan 20];22(49):493101.
2.Neukirch S, Antkowiak A, Marigo J-J. Soft beams: When capillarity induces axial compression. Phys Rev E [Internet]. 2014 Jan 2 [cited 2017 Jan 20];89(1):12401.
3.Extrand C, Moon S. Repellency of the lotus leaf: contact angles, drop retention, and sliding angles. Langmuir [Internet]. 2014 [cited 2017 Jan 20];
4.Yamada H, Yoshida Y, Terada N, Hagihara S, Komatsu T, Terasawa A. Fabrication of gravity-driven microfluidic device. Rev Sci Instrum [Internet]. 2008 Dec [cited 2017 Jan 20];79(12):124301.
5.Intravenous Medication Administration: What to Know from HealthLine.
6.Allen JS, Lee SH, Chang ST, Choi Y-K, Friedrich C, Choi CK. A novel miniature dynamic microfluidic cell culture platform using electro-osmosis diode pumping. Biomicrofluidics [Internet]. 2014 Jul [cited 2017 Jan 23];8(4):44116. Available from: http://aip.scitation.org/doi/10.1063/1.4892894
7.Sachan VK, Singh AK, Jahan K, Kumbar SG, Nagarale RK, Bhattacharya PK. Development of Redox-Conducting Polymer Electrodes for Non-Gassing Electro-Osmotic Pumps: A Novel Approach. J Electrochem Soc [Internet]. 2014 Oct 25 [cited 2017 Jan 23];161(13):H3029–34.
8.Slouka Z, Senapati S, Chang H-C. Microfluidic Systems with Ion-Selective Membranes. Annu Rev Anal Chem [Internet]. 2014 Jun 12 [cited 2017 Jan 23];7(1):317–35.
9.Byun CK, Abi-Samra K, Cho Y-K, Takayama S. Pumps for microfluidic cell culture. Electrophoresis [Internet]. 2014 Feb [cited 2017 Jan 23];35(2–3):245–57.
10.Nightingale AM, Beaton AD, Mowlem MC. Trends in microfluidic systems for in situ chemical analysis of natural waters. Sensors Actuators, B Chem [Internet]. 2015;221:1398–405.
11.Lagerström ME, Field MP, Séguret M, Fischer L, Hann S, Sherrell RM. Automated on-line flow-injection ICP-MS determination of trace metals (Mn, Fe, Co, Ni, Cu and Zn) in open ocean seawater: Application to the GEOTRACES program. Mar Chem. 2013;155:71–80.
12.Seo K-S, Lee K, Shim Y, Kim A, Jeon E, An S, et al. Smart syringe pumps for drug infusion during dental intravenous sedation. J Dent Anesth Pain Med [Internet]. 2016 [cited 2017 Jan 30];16(3):165.