The basic use of a pipette is to transport a measured volume of liquid with high precision for the preparation of solutions, biological, biomedical and chemical tests with regards to the quantity of the chemical or reagent being used. Rather than use a single channel pipette many times, a multichannel pipette is a scientific laboratory device used to increase the output and productivity of transferring a measured volume of liquid. Pipetting can be a tedious task and these multichannel devices are often used in laboratories to speed up routine pipetting. Unfortunately, due to their relatively low-volume of manufacture as compared to consumer goods and precision requirements, which reduce market competition, multichannel pipettes are expensive. For example, a 4-channel VOYAGER adjustable tip spacing pipette costs over USD$2,000  and a budget 8-channel Thermoscientific Finnpipettes cost around USD$1,000 or more . Even Amazon’s best-selling budget multipippette costs over USD$235 . High costs of pipettes has been an issue dating back to the 1970s  and the costs of such scientific tools limit accessibility in some laboratories and educational labs . In addition, during emergencies when a surge of medical equipment is needed (e.g. testing for COVID-19 pandemic) and global supply chains is disrupted, there is a clear need for an alternative supplies for medical and scientific devices [6, 7, 8, 9]. A new method gaining traction during the COVID-19 pandemic is distributed manufacturing of open source medical hardware [10, 11, 12, 13, 14] using digital fabrication technologies like 3D printing [15, 16, 17, 18]. Most low-cost desktop 3D printing used during the pandemic was RepRap-class [19, 20, 21] fused filament fabrication of personal protection equipment [22, 23] like face shields [24–27]. This technique, however, was also used for more sophisticated devices like modifying helmets , making air-purifying particulate respirators , fabricating ventilator components  and even printing most of the components of emergency ventilator devices [31–33]. This approach has been long-established to provide custom state-of-the-art scientific hardware [34, 35, 36, 37] at lower costs [38, 39, 40, 41, 42]. For example, open source 3D printing of scientific hardware has been shown to save research funds while making high-quality custom hardware from chemical mixing [43, 44, 45] to centrifuges [46, 47, 48, 49, 50]. Most importantly, there have been several valuable investigations of open source pipettes including a 3D-printed 1000 μL micropump  and an adjustable micro-pipette that meets ISO standards for accuracy .
To build on both of these areas of prior work and democratize [53, 54] multichannel pipette-based lab processing, this study uses the open source hardware design approach [55, 56, 57] to make a parametric mostly 3D printable multichannel pipette. The device is designed to be widely accessible, easily manufactured, easy to maintain with interchangeable parts and includes isolation of fluids in individual tips. These multichannel pipette designs are parametric and thus are also able to provide an additional feature of easing customization by making changes to a few parameters in the source code. The multichannel pipette is fabricated, evaluated, and validated using ISO standards.
Overall Implementation and Design
The overall design of the open source multichannel pipette is shown as fabricated in Figure 1. Note the ergonomic handle that makes it easy to dispense fluids because of the hand bridge. The design is based around individual syringes controlled with a master plunger as shown in the x-ray view of Figure 2. The design follows Brennan et al.’s single pipette closely  as it has a high degree of functionality, is open source and easily accessible. The design features utilized from  are the main plunger locking mechanism and the same connector for the pipette tips, while the main plunger design was built upon the single plunger but with important modifications to incorporate the reciprocating action of a 4-individual-cylinder-plunger assembly. The multichannel pipette consists of four inline nozzles with a distance of 18 mm between the adjacent nozzles. This design choice was made because of the size of the selected components and enables users to fill the alternate wells in a multi-well plate. The model consists of three main 3D printable parts shown in Figure 3. Detailed renderings for the enclosure (Figure 4), syringe holder (Figure 5), and plunger (Figure 6), which produces a plunger action motion with the help of compression springs. The designed multichannel pipette works on the principle of positive air displacement, where the length of syringe plunger displaced within each individual syringe is equal to the amount of liquid moved by the nozzle connected to that syringe. The mechanism is designed in such a way that the main plunger (Figure 6) actuates the individual syringe plungers, which are retracted back by spring action. The critical component of this model is the syringe holder (Figure 5) due to its intricate shape and need for space compaction. The syringe holder enables syringes to be inserted and then twisted to be locked into place at appropriate spacing. The plunger and enclosure are designed based on this critical components. The material used for 3D printing the parts was polylactic acid (PLA) as it is easy to post-process and is the most widely available 3D printing filament. All of the other parts are widely accessible off-the-shelf components and are interchangeable.
Hardware documentation and file location
All of the documentation to recreate this device can be found on the Open Science Framework at https://osf.io/9tn6e/.
The documentation is organized into three sub-folders. In the ‘Design Files’ folder the raw computer aided design (CAD) files are made available in FCStd format for FreeCAD (an open source CAD package) as well as in SLDPRT format for Solid Works users. The ‘Documentation’ folder contains a detailed bill of materials (BOM) that includes prices of components and hyperlinks to available suppliers as well as a validation data set. Finally, in the ‘STL files’ folder the four 3D printable part files are made available for easy slicing and 3D printing.
(2) Quality control
Calibration of the multichannel pipette is done using gravimetric testing. During the gravimetric testing , each channel/syringe of multichannel instrument is regarded as an individual pipette. For the calibration test a Metlier Toledo XS205DU analytical scale with a resolution 0.01 mg was used. Commercial analytical scales with this resolution are available for a few thousand dollars, but can be purchased for less on the used market.
It should be noted that for gravimetric testing there is a bias from the air pocket expanding as the liquid is pulled by gravity as detailed in . This phenomenon increases with fluid volume, but can be avoided by tilting the pipette at 45 degrees while drawing liquid in, rather than keeping the pipette straight up and down as is done during calibration of commercial pipettes. All experiments performed here were at 45 degrees, which appears to be a more natural way to operate the pipette for most users. In this study the system was tested for ISO standards for 100 μl and 200 μl nominal volume, but if larger volumes are needed the calibration should be carefully repeated. Another approach taken to solve this problem, to adjust the graduation scale, is presented in , and may be a good adjustment for larger pipetting volumes.
The use of the multichannel pipette does not present any additional safety issues to the user than with conventional pipettes. In addition, although multichannel pipettes reduce effort and the overall time required to pipette multiple samples, the main concern with these pipettes in some laboratory environments is the potential cross contamination of adjacent specimens. This could, in the worst case scenario, lead to an unwanted chemical reaction and harm to the user. Some transfer pipettes in the market do have common actuating air (e.g. the Cole Parmer Scienceware Transpette 8-Channel Transfer Pipettors). These pipettes are commonly used to transfer solution from a single reservoir to separate wells. In the current design, not only do the cylinders have higher volumetric capacity, they also be used as transfer pipettes with appropriate pipette tips and be much more accurate than the Transpette. In addition, many multichannel pipettes have a common housing for cylinders (e.g. the Thermo Scientific Matrix Electronic Pipette). The design may consist of a single compact part (cylinder housing) in which numerous holes have been drilled that act like cylinders. If one of the cylinder hole gets contaminated, depending on the solution that was being used, the user may have to sterilize/replace the entire housing. A single cylinder housing design also usually has a piston assembly. This piston assembly consists of several pistons that are fixed to a common base plate. If one of the pistons gets damaged, the user may need to replace the entire assembly. This is not a problem in the open source design detailed in this study as each cylinder has its separate piston that is not attached to the actuating base plate. To avoid this, each nozzle tip is operated by a separate syringe. There is thus, no direct contact of the multiple specimens or the actuating air inside the syringes with each other. If a syringe does get contaminated due to human error, there is no need to throw away or replace the entire pipette system. Instead, the contaminated syringe can easily be replaced with a fresh one without disturbing the assembly of other syringes for a marginal cost.
Validation and Testing
Validation of the multichannel pipette was done considering every channel of multichannel instrument regarded as an individual pipette (ISO 8655) [59, 60]. Validation of the multichannel pipette was done using gravimetric testing . Each syringe is tested separately for different set volumes. A total of 10 tests were conducted for each set volume and for each syringe, separately. For the calibration test deionized (DI) water was used as pipetting fluid and was measured using an analytical scale having a resolution of 0.00001 g.
A method of modularity was built into the design could prove useful for different versions of the device as different 3D printed multichannel pipettes may have different elevation of the locking mechanism (the precision screw allows users to adjust the device so a set amount of fluid is released as seen in Figure 1) and hence the offset can easily be calculated using the syringe’s graduations. For the designs tested here, due to the different elevation of the locking mechanism, the piston inside the syringes gets locked at 30 µL instead of 0 µL. So, during setting up the pipetting volume, a positive offset of 30 µL has to be added to the volume. For example: To pipette a volume of 200 µL, the pipette was set for 230 µL using the syringe’s graduated marks. The ability to change the lock allows for greater flexibility of the design. For labs only using one system this offset could be designed to be zero in future iterations to make the device easy to use.
A total of 10 readings must be taken for each syringe for a particular set volume. Each weight obtained during a single test cycle can be converted to the specific volumes by multiplying it with the fluid’s density (Z). Equation 1 represents this formula, where mi is the mass measured in each test cycle and Z is the density of the fluid.
As 10 tests are to be conducted for a particular set volume for a single syringe, average of these tests are calculate using equation 2.
As this is a measuring device, accuracy and precision play an important role in determining the usefulness of the multichannel pipette. Systematic error (es) represents the accuracy of the device and can be calculated by subtracting set volume (Vs) from average volume (V̅) calculated during testing as shown in equation 3.
The random error (Sr) represents the precision of the device and can be calculated using equation 4. Where n is the number of measurements taken for each channel (here n = 10).
By adjusting the precision screw and by using the graduations on the syringes, the pipette was set to take two sets of volumetric readings at 200 µL and 100 µL.
As there are a total of four channels/syringes, each channel was tested 10 times to measure the accuracy and precision at both the volumetric settings. The results of these tests are displayed in Table 1. In this experiment, the density of the DI water was calculated to be 1.0025 g/ml. As this pipette was designed to measure a maximum volume of just under 300 µL, according to ISO 8655 standards, the systematic error should not be more than 4 µL and the random error should not cross 1.5 µL . As can be seen in Table 1, all the 4 channels of the open source multichannel pipette met the ISO 8655 standards. To clarify the quality of the device this data is shown in Figure 7 (200 µL) and 8 (100 µL), where the box represents the ISO 8655. Figures 7 and 8 plots show an inclusive median and the center represents the 50th percentile of the data set, which is derived using the lower and upper quartile values. For both Figures 7 and 8 the median value is displayed inside the box and the maximum and minimum values are displayed with vertical lines as the whiskers connecting the points to the center box. The plane horizontal line represents the set volume of the pipette, while the line with a circle in its middle represent the mean volumes calculated during the validation tests. All measurements were replicated 10 times and processed upright as in the ISO standard.
|VOLUME||CHANNEL||MEAN (µL)||SYSTEMATIC ERROR (µL)||RANDOM ERROR (µL)|
Growth of the drug and most recently (and urgently because of COVID-19) the vaccine discovery market as well as techniques such as molecular biology, immunology assays, high throughput screening and polymerase chain reaction (PCR) have expanded the use and applications of microtitre plates (also called microplate, microwell plates, and multiwells). Google Scholar provides some evidence of this as it shows about 150,000 articles using microplates between 2000 and 2020 (~7,500/year) having expanded from only 58,000 in the previous 20 years (~2,900/year). By May 8 of 2021 there have already been 6,130 papers published mentioning the term. Most scientific experimental work in these areas starts in individual test tubes for which single channel pipettes are ideal. As the work progresses, there is a need for increased productivity, which scales up the research and runs larger-scale pilot experiments or assay runs. These are often performed in 96 well microtitre plates. If a researcher continues to use a single channel pipettes the time costs suddenly become a real issue. For example, if a graduate student researcher costing US$25/hour spends 1.5 hours every working day single channel pipetting the labor costs involved in the pipetting parts of the experiments amount to (7.5 hrs/week × 52 weeks/year = 390 hrs/year) US$9,750. Note, this is only pipetting for a fraction of the overall workday and some lab workers may spend even more time on this task some days. These costs can be cut to roughly 1/4th to US$2,437 by using a four channel pipette and decreasing the researcher time to only 22.5 minutes per day or less than 100 hours per year. The fact that the time spent pipetting dropped below 300 hours is important for the health of the researcher as will be discussed in section 5. This work must not sacrifice accuracy and precision, while at the same time there is a need to reduce repetitive strain injuries of the researchers and save the time of highly qualified personnel (HQP). There is thus a demand for ergonomic and highly accurate and precise multichannel pipettes. As the results of this study show the open source multichannel pipette can be used for any chemical, biological or medical application that demands ISO 8655 for manual pipetting. As tested, the open source device speeds pipetting by a factor of four compared to single channel pipettes.
The main highlight of the design of the open source multichannel pipette is its modularity and the ease to adjust the amount of liquid processed simultaneously. Each and every part of this assembly can be replaced for a nominal cost and without the need of any special training. Each syringe assembly acts a single independent unit, which can be replaced in the case of breakdown or contamination. At this point the device should be re-calibrated. Due to the use of independent syringe assembly the potential for sample contamination is essentially eliminated. In addition, the model can be modified with increased number of nozzles in multiple of 4 (i.e. 6) as well as the size of the syringe according to the needs of the researcher. This would still be an advantage although filling of the 96 well plate would need to be done length wise because of the spacing between the tips only allow every other well to be filled at a time.
Reuse potential and adaptability
There has been a considerable quantity of research in the development of open source syringe pumps [61, 62, 63] based largely on 3D printing mechanical components and the use of the Arduino open source microcontoller  or Raspberry Pi single-board computer [61, 65]. These systems have been adapted to do multiple volumes  and even large volumes [67, 68] for extrusion-based additive manufacturing systems. In addition, multi-syringe systems have been built [61, 69] and open source syringe pumps have been demonstrated for high precision applications like occular drug delivery . The development of the open source multichannel pipette system shown here can be coupled to the existing work on syringe pumps to drive the main piston. This would enable further automation and acceleration of science using a tool like Osman, an open source microsyringe autosampler  or the conversion of a 3D printer to a fluid handling robot . Finally, open source syringe pumps have been augmented with feedback sensing and monitoring , which could be applied to the device here to for example automatically detect a clog in a single syringe during rapid pipetting. This would bring the scientific community one step closer to enabling scientists to fully digitally fabricate their open source labs [5, 73].
(4) Build Details
Availability of material and method
|PART NO.||PART NAME||QUANTITY||MATERIAL||SOURCE|
|1a||Enclosure (Body 1)||1||PLA||Amazon|
|1b||Enclosure (Body 2)||1||PLA||Amazon|
|4||1ml Syringe||4||Disposable plastic||Allegro Medical|
|5||Compression Spring||4||Stainless steel||McMaster-Carr|
|6||Female adapter||4||Disposable Plastic||Cole-Parmer|
|7||Universal pipette tips||4||Disposable Plastic||Amazon|
|8||Silicon medical tubing 3/16” OD||1 foot||Silicon||Freshwatersystems.com|
|9||M3x 35mm bolt||1||Stainless steel||McMaster-Carr|
|10||M3 Nut||1||Stainless steel||McMaster-Carr|
There are 10 parts in total out of which the three main parts are designed in the FreeCAD  open source designing software. In this study they are 3D printed on an open source RepRap-based Lulzbot Taz 6 printer  after slicing the model in open source Lulzbot CURA in PLA. It should be pointed out, however, that any RepRap-class printer could be used to fabricate this device. As such printers can be purchased for under US$200 now it is possible that fabricating a single multichannel pipette could justify the purchase of a 3D printer. In addition, the costs of the device as shown could be reduced by fabricating it with other plastics including waste plastic using a distributed recycling and additive manufacturing (DRAM) model [77, 78, 79, 80, 81]. The remaining parts are widely available from the vendors like Amazon and McMaster-Carr in the U.S. and AliExpress or others depending on availability in the researchers’ home country. The BOM and the links for each part are provided in the BOM file in the repository (https://osf.io/khupy/). The total cost including 3D printing materials was under US$24 at the time of this writing. These costs would vary on the availability of commercial components and 3D printing filament. The costs, could for example be reduced to under US$20 by using recycled plastic filament manufactured by an open source recyclebot (waste plastic extruder) . The validation of the pipette is carried out using a high precision analytical scale, but can be done using an open source scale for large volume syringes . The design is assembled following Figure 10 and is relatively straight forward.
First as shown in Figure 10a the four syringes (part 4) have the compression springs (part 5) inserted between the plunger and the body of the syringe. Then female adapters (part 6), a short section of the silicon medical tubing (part 8) and the universal pipette tips (part 7) are placed on the end the syringe tip. This syringe assembly is inserted into the syringe holder (part 3). Figure 10a shows the syringes being assembled individually with springs, tips and tip holders and being inserted into the syringe holder, which calls for a twisting motion to lock them into place (Figure 10b). The M3 nut (part 10) and bolt (part 9) are inserted into the enclosure as shown in Figure 10c. The one challenging part is the screw nut has a seating slot that holds the nut tight once it is assembled. The nut must be inserted (part 10) into the slot such that it acts like a nut trap. Then, the user tightens the nut on the screw through the slot and pull it outwards to fix it in its seating space. Next, the main plunger (part 2) in inserted into the enclosure following Figure 10d. The device is now ready for the final assembly. The enclosure is tilted up so that the sliding surface that will hold the syringe holder is exposed (Figure 10e) and the main plunger is pushed up into the enclosure as far as possible (ensure the M3 bolt has not been tightened down). Then the syringe holder is slid into the enclosure by depressing the syringes against the springs (Figure 10f). The syringe holder should be pushed to the end of the slide in the enclosure (Figure 10g). Finally the remaining enclosure component is slid into place providing a completed open source multichannel pipette as shown in Figure 10h.
Hardware documentation and files location:
Archive for hardware documentation, modifiable design files, software and build files.
Name: Open Science Framework
Licence:GNU GPL v3 for documentation and CERN OHL v2 for the hardware
Publisher:Ketan Mowade, Hrishikesh Kadam, Shubham Chinchane, Joshua Pearce
Date published: 17/01/2021
Discussion and Conclusions
The design of the open source multichannel pipette is comfortable to use for an average sized hand and has a finger hold (as seen at the index finger in Figure 11) for compression of the syringes. Considering the multiple syringes it is subjectively easy to maneuver. Designing of this pipette is done in FreeCAD, which is an open-source free designing software and will allow other researchers to easily adapt it to their needs. Even though the multichannel pipette consists of several different parts, there is no use of binding material such as glue or screws to hold these parts together. The pipette is designed in such a way that it can be deconstructed easily so that the parts could be replaced in case of any damage. Even though the main body is 3D printed, there are certain parts of the pipette that cannot be 3D printed such as the syringes using fused filament fabrication at this time (although future work, particularly with open source SLA may make this feasible). The current design consists of just four 3D printed parts and can be mass produced in a distributed fashion in case of emergencies as has been observed with PPE during the COVID pandemic. This 3D printed multichannel pipette does satisfy ISO 8655 standards as seen in validation experiments and can be used in professional laboratories. It is also inexpensive enough, with the full BOM costing less than $24, that it could also enable students to have a ‘real’ research lab experience even in the undergraduate college and high-school levels. For example, it can be used to accelerate research by a factor of 4 to fill a 96 well plate as shown in Figure 11 as compared to a single channel pipette. This not only reduces research labor costs as detailed above, but can also have secondary benefits to the users. Björksten et al. have shown in a study of plunger operated pipettes that the prevalence of hand ailments among female laboratory assistants was found to be twice that among female state employees in general . In their cohort a nested case-control study indicated that an increased risk of hand (OR = 5, 0) and shoulder (OR = 2, 4) ailments was associated with more than 300 hours of pipetting per year . In the economic example above, using the 4 channel open source system provided here not only allows the research to while cutting labor costs but it also drops the user under the 300 hour recommended ceiling on pipetting time without compromising on quality. This is also a useful device to teach students about various fluid measuring devices, about 3D printing and how it can be used to fabricate precision products. Finally, the resolution of the pipette can still be improved as the resolution of the syringe graduation is 10 µL (e.g. the smallest markings on the used syringes were 10 µL per graduation marks).
This design met the specifications for the intended project but could be improved. Due to the spacing of the syringes that was necessitated by their volume and the critical component users must pipette into every other well in a standard 96-well plate. This is adequate for some applications, but if pipetting with multiple colorless reagents in small volumes this could result in experimental error. This could be fixed in the future using smaller diameter commercially available syringes. For the target use scenarios, here larger 1mL syringes were used, however. Future work could also focus on removing the flexibility caused by the medical tubing in the tips by attaching the tips to the syringes directly with non-flexible couplings as well as enabling a fast removal of the tips. This may be possible by 3D printing couplings between different types of syringes and the common pipette tips. Future work can also focus on optimization and increasing the accuracy of the pipette. This design of pipette still uses 1CC syringes, which increases the size of the pipette. Use of smaller syringes is encouraged given that they can be used repeatedly for applications with smaller volumes needed. Future work is needed to investigate the impact of pipetting techniques using these multichannel pipettes  on precision and accuracy as well as new methods of calibration . Future work is also needed to determine the devices function over long-term continuous use (e.g. to determine how many cycles is the calibration stable and is it dependent on fluid viscosity). In addition, future work can focus on increasing the resolution of the scale as the current resolution of the syringes is just 10 µL. Use of a custom scale on the syringes similar to  or using precision screw gauges instead of the volume adjusting screw (part 9 in Figure 9) may be viable methods to improve the performance of the device. In addition, development of a stepped plunger will allow users to pipette different volumes of fluid in each syringe depending on the step size such as has been done with the Biropette . Various iterations of the design can be made like 8 channel pipette, 16 channel pipettes, etc. manufacturing with high precision 3D printer to get a better finish and better tolerances or the use of an open source high-temperature 3D printer  would allow for materials that can be sterilized with an autoclave or boiling. Lastly, as these devices are manually assembled with 3D printed parts, a future study could explore the differences in precision between devices that are printed on various types of printers and assembled by various individuals with different levels of fabrication experience to determine if the devices are still able to conform with ISO 8655.