Highly precise alignment for the rapid fabrication of Plexiglas® microfluidic devices

G. Simone

University of Naples Federico II, Piazzale Tecchio 80, 80125 Napoli, Italy.


Why is this useful?

Most of microfluidic devices use channels with rectangular cross-sections. The microfabrication of rectangular shaped channels is straightforward with standard tools such as photolithography.

Fluid dynamics, rheology, soft matter and, recently, biology-based investigations need circular cross-section microchannels; indeed, pre-fabricated capillaries are normally used to carry out the studies. However, capillaries are impractical for some investigations requiring complicated designs.

For Plexiglas® (or other plastic) devices, microfabrication by micromilling is a low cost procedure, which, in the last decades, has gained popularity in the field of microfluidic applications.1-2 The fabrication and the sealing of Plexiglas® microchannels with circular cross-section can be challenging.

Here, a method of fabrication of plastic microfluidic devices with circular cross-section is presented. The method is low cost and it can be performed by minimally trained users. The alignment step builds on a procedure first introduced by Lu et al. that used circular magnets to align layers of polydimethylsiloxane (PDMS).3 The protocol is validated for circular cross-section channels, but it can be used for fabricating rectangular channels or special inlets as well.


What do I need?

  • Plexiglas® sheets (thickness 1mm) Rohm Italy
  • CNC MicroMill (Minitech, US)
  • Ball nose end-mills (0.001 inch, PMT Endmill)
  • Magnets 4 mm × 4 mm× 4 mm, DX
  • Microscope Bresser 58-02520
  • Clamps RS Italy
  • Ethanol
  • Microscope (or stereomicroscope)
  • Glass slides for microscopy (50 mm × 75 mm)


What do I do?

1. Computer aided design of the device (CAD).

  • Design the microfluidic channels with CAD software. Fig. 1a displays the top and bottom layer of the main channel and, in particular, aligned square through-holes in each layer.
  • Translate the CAD file to Computer Numerical Control (CNC) code for micromilling.

2. Micromilling

  • The endmill should be aligned to set the point z=0 by using a microscope.
  • Then, the microchannels (top and bottom) must be milled. Channel with depends on the endmill diameter.4 It is worth emphasizing that setting the correct work parameters (such as tool speed and depth) is crucial. Indeed, if the latter are not appropriate, the plastic workpiece develops internal stresses that during the bonding result in cracks and chip breakdown.
  • Finally, the through holes and the frame have to be milled. The number of through holes for locating the magnets depends on the dimension of the microfluidic device. It is good practice for the rectangular through-holes to have a pitch of roughly 10 mm (this dimension depends on the size of the microchannels and of the magnets).

3. Alignment

  • Align the top and bottom layers with a stereomicroscope (Fig. 1b). The bottom layer should be set on a solid surface and the magnets inserted into the through-holes. Then, the second layer should be placed on the first and a second set of magnets inserted in the square through-holes of the second layer.
  • The magnets located in the first and second layer naturally provide a good “first alignment” of the microchannels, while leaving freedom to slightly adjust the layers (i.e., this is a reversible sealing). Once the magnets are fixed, the quality of the alignment should be checked with a microscope.

    Fig. 1. The Fabrication. a. CAD design of a microfluidic channel with circular cross section. The length of the microfluidic device is 40mm, the width is 15mm. The depth of grooves is 50μm, width 25μm. b. Magnets employed for the bonding. C. Clamps used for sealing the microchannel.

4. Bonding

  • Clamp the Plexiglas® layers together and submerge assembly in ethanol for 15 minutes (Fig. 1c). For optimal bonding, the clamping should be done with clamps positioned at the edges of the Plexiglas®.5
  • After 15 minutes, the sealing of the microfluidic channels should be checked. If the Plexiglas® is sufficiently bonded, the magnets can be removed and glass slides placed on either side of the Plexiglas® layers. Clamp the glass slides and allow the assembly to rest for another 5-15 minutes.
  • The device is ready to be used for different applications as shown in Fig. 2a. Capillary tubing can be easily connected to the channel for modular design (Fig. 2b).

    Fig. 2. Examples of microfluidic devices. a. Perfusion of the samples through hole at the inlet. b. Perfusion through capillaries.

In conclusion, analyzing the protocol, the following advantages can be emphasized:

  1. The process is designed for different materials, but it fits perfectly with Plexiglas®
  2. The equipment necessary for the fabrication and assembly includes simply a micromilling machine and a (stereo)microscope
  3. The use of square magnets (instead of circular ones) allows for more precise alignment due to further restriction to the sliding of the top and bottom layers.


References

  1. G. Simone, G. Perozziello, J. Nanosc. Nanotech., 2010, 11, 2057.
  2. G. Simone, RSC Advances, 2015, 5, 56848.
  3. J-C Lu, W-H Liao, Y-C Tung, J. Micromech. Microeng. 2012, 22, 075006-075014.
  4. G. Perozziello, G. Simone, P. Candeloro, F. Gentile, et al. Micro and Nanosystems, 2010, 2, 227-238.
  5. G. Medoro, G. Perozziello, A. Calanca, G. Simone, N. Manaresi, 2010, US Patent App. 13/257,545.

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Direct Delivery of Reagents from a Pipette Tip to a PDMS Microfluidic Device

Hoon Suk Rho1, Yoonsun Yang2, Henk-Willem Veltkamp1, and Han Gardeniers1

1Mesoscale Chemical Systems Group, MESA+ Institute for Nanotechnology, University of Twente, The Netherlands.

2Medical Cell BioPhysics Group, MIRA Institute for Biomedical Technology and Technical Medicine, University of Twente, The Netherlands.


Why is this useful?

Most common way to handle and transport reagents in chemical or biological labs is by using a pipette. However, tubing connection is generally used for the delivery of reagents into a microfluidic device. Even though the connection with commercial tubing and connectors allows various choices on the sizes and materials of the tubing and easy connection, difficult sampling from stock solutions, dead volume in tubing and connectors, and extra sterilization on tubing and connectors for biomaterials, still remain challenges. Here we demonstrate a direct connection of pipette tips to a PDMS device and loading reagents by pressure driven flow.


What do I need?

  • Stainless still puncher (Syneo LLC)
  • Precision stainless steel tip (23 gauge, #7018302, Nordson Corporation)
  • Tygon tubing (0.020″ x 0.060″OD, #EW-06418-02, Cole-Parmer Instrument Company)
  • 3D printed plug
  • Pliers

What do I do?

Punching inlet and outlet on a PDMS device

1. Select the size of a puncher based on the size of a pipette tip will be connected. We achieved tight connections of pipette tips on a PDMS substrate when we punched holes by using punchers with outer diameter of 2.4 mm, 1.8 mm, 1.3 mm, and 1.0 mm for 50 – 1000 µl, 2 – 200 µl, 0.5 – 20 µl, and 2 – 200 µl (capillary) pipette tips, respectively.

Fig. 1 Direct connection of pipette tips on a PDMS device.

Plug connection preparation

1. Separate a pin from a precision stainless steel tip. The pin can be easily removed by twisting the plastic part while the pin is held by pliers.

2. 3D-print a plug. The outer diameter of the plug depends on the size of the pipette tip that will be connected. The detailed dimensions of the plug are shown in Fig. 2.

3. Connect a precision stainless steel tip, tubing, a pin, and a plug. Because the plastic part of the tip is luer tapped, it can be connected to male luer connectors and commercial plastic syringes.

Fig. 2 Tubing connection with a 3D-printed plug.

Solution loading

1. Pipette the sample, connect the pipette tip into the inlet of a PDMS device, and take off the pipette. When an empty pipette tip is connected into the outlet of the device, the sample from the outlet can be collected in the tip.

2. Insert the plug into the tip and connect the tip to a pressure source. When pressure is applied into the pipette tip, the solution in the tip is pushed into a microchannel (Fig. 3A). The luer tip can be connected to a syringe and a syringe pump can be used as a pressure source (Fig. 3B). The flow rate of the sample solution can be controlled by the syringe pump. The luer tip also can be connected to a pressure regulator with a luer fitting. Fig. 3C shows the flow rate of sample loading at various applied pressures controlled by a pressure regulator. In this case an external gas source is required. However, this system is cheaper than commercial microfluidic flow control systems. Also a digital pressure regulator can be used for accurate flow rate control at the low flow rate regime less than 1 µl/min.

Fig. 3 A. Loading blue food dye solution into a microchannel, B. Solution loading by a syringe pump, and C. Solution loading by using a pressure regulator.


What else should I know?

No leakage of the solution was observed in the connection of a pipette tip onto a 1mm thick PDMS substrate. However at least a thickness of 3 mm is highly recommended to achieve tight fitting and stable support for the pipette tip. The pin obtained from a precision stainless tip is very useful for tubing connections. For example two separated tubes can be connected by the pin and also the pin can be inserted into an inlet or outlet of a PDMS device punched by a puncher with an outer diameter of 1mm. Also the pin can be easily bent by using pliers for compact connection to a PDMS device.

Fig. 4 Tubing connection by using the pin from a precision stainless tip.

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Syringe funnels for facile loading of precious samples

Alexander Price, Wesley Cochrane and Brian Paegel
Department of Chemistry, The Scripps Research Institute, Jupiter FL 33458


Why is this useful?

Syringe pumps are the most popular tool for transporting fluids within microfluidic devices. In the process of loading sample into a syringe, air bubbles (derived from the syringe dead volume) frequently migrate into the barrel and require removal to achieve consistent flow. Ideally, a researcher would have a large excess of sample so that the barrel can be filled and evacuated multiple times. During loading, syringes are held vertically with the sample directly below the tip, necessitating forceful evacuation to dislodge rising bubbles, however this not feasible for low/intermediate-volume “precious” samples (50-500 µL). Here, we present a simple funnel to aid bubble removal during syringe loading.


What do I need?

Plastic transfer pipets (FisherBrand 13-711-7M)

Razor blade

Pipette and pipette tips

Syringe (we use Hamilton Gastight 1700 series w/ TLL tips)

What do I do?

1.  Using the razor blade, carefully cut the end off of the transfer pipet so that it fits snugly over the tip of your syringe (Fig. 1). It might take a couple tries, but you can use this as a template once you have found the right location to cut.

2.  Cut the transfer pipette again, roughly 2 inches up from the previous cut (Fig. 1). Your funnel is complete.

3.  Attach the funnel onto the tip of the syringe. Holding the syringe vertically (funnel up), load your sample into the bottom of the funnel (Fig. 2).

4.  Fill and evacuate the syringe barrel as needed to eliminate any air bubbles (Fig. 3).

5.  Dispose of the funnel.

Figure 1. Construction of a syringe funnel from a transfer pipet.

Figure 2. The funnel is attached to the syringe (left), and sample is loaded into the bottom of the funnel (middle and right).

Figure 3. Air bubbles in the barrel are expelled into the funnel and syringe is filled.

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Rapid technique for UV-curable adhesive bonding of glass coverslips to polystyrene microdevices

David J. Guckenberger,*a Jake Kanack,*a Loren Stallcop,b David J. Beebea

aDepartment of Biomedical Engineering, Wisconsin Institutes for Medical Research, University of Wisconsin-Madison, Madison, WI, USA
bDepartment of Materials Science and Engineering, University of Wisconsin
Madison, Madison, WI, USA
* Authors contributed equally


Why is this useful?


With several microfabrication techniques now available, including: 3D-printing,1 micromilling,2 and hot embossing,3 in-house fabrication of thermoplastic microdevices has become cheaper, faster, and easier. However, for many applications – such as cell culture and microscopy – these devices must be bonded to optically-transparent substrates such as glass. While bonding similar materials, such as Polystyrene (PS) to PS, is relatively simple, bonding dissimilar materials, such as PS to glass, presents a particular challenge. Current methods to circumvent these challenges include spin coating adhesives, such as polydimethylsiloxane (PDMS), onto sacrificial substrates4 and injecting adhesive directly into the bond interface.5 However, equipment requirements, associate long cure times, heterogeneity in glue uniformity, and complexity limit acceptance of these techniques.

Here we present simple technique for applying uniform layers of adhesive to enable rapid – less than a minute – bonding of PS to glass. Using UV-curable adhesives, readily accessible materials, and a simple techniques, we demonstrate how to apply thin uniform layers of adhesive to a microchannel. We provide design suggestions that will improve bonding repeatability, and additional information that may help apply this technique to materials beyond PS and glass.


What do I need?


  • Polystyrene                                            (1.2 mm, #ST313120, Goodfellow)
  • Glass coverslip                                      (#260450, Ted Pella, Inc.)
  • UV curable adhesive                           (Ultra Light-Weld 3025, Dymax)
  • Silicone foam (Textured)                   (#31943970, MSC Industrial Supply Co.)
  • Isopropyl alcohol                                 (#190764, Sigma-Aldrich)
  • Low particulate wipers                       (#TX609, Texwipe)
  • Wooden flat stick                                 (#704, Brightwood)
  • UV lamp                                                 (OmniCure S1000, Exfo)
  • Adhesives are often material-specific. Consult the manufacturer to determine the best adhesive for your application.

Tip: Some adhesives may require post-treatment / aging to reach a full cure.

  • We have tested this protocol with Ultra Light-Weld 3025 (Dymax) and Norland Optical Adhesive 68 (Thor Labs, Inc.). These adhesives had similar performance, however the protocol may need to be tailored for other adhesives.
  • If the adhesive is too viscous or does not adequately wick around the rib, heat may be applied to achieve thinner adhesive layers, or to improve the wicking of the adhesive.
  • This protocol is amenable to wide variety of materials, including: cyclic olefin copolymer (COC), glass, metal, PS, and various rapid-prototyping materials.
  • Creating the rib and allowing the adhesive to wick eliminates excess adhesive and prevents adhesive from squeezing into the microchannel.


What do I do?

Fig.1 Channel border design

Step 1: Fabricate the microdevice. To improve bonding repeatability and adhesive distribution we recommend fabricating a groove (thickness > 0.5mm) around the channel – leaving a rib (0.5 mm < thickness < 1.5 mm) around the perimeter of the channel.

Tip: Rib thickness may need to be tuned for individual adhesives
Tip: Extra caution while applying the adhesive may be necessary for channels shallower than 0.1 mm

Step 2: Thoroughly clean the surface of the microdevice, silicone foam sheet, and coverslip using isopropyl alcohol and low-particulate wipers. Remaining particulates can be blown off with compressed air. Ensure PS and glass surfaces remain clean throughout the bonding process.

Step 3: Apply a dollop of UV curable adhesive to the foam sheet.

Step 4: Use a tongue depressor to spread the adhesive into a uniformly thin layer across the foam. The area of the adhesive should be larger than the microdevice – add more adhesive if necessary.

Step 5: Position the device onto the adhesive bonding surface down. Press down gently; avoid sliding the microdevice to prevent build-up of adhesive within the channels. Pick up the device and repeat this step two or three times to ensure the bonding surface is completely covered with adhesive.

Tip: Take care to ensure no adhesive is transferred from gloves to surfaces of the device not intended to be bonded.
Tip: Minimize the delay between step 4 and step 5 to help ensure a uniform thickness of adhesive

Step 6: Position the microdevice above the coverslip, and gently lower it until it makes contact. Once contact is made, release the device, taking extra caution to avoid sliding the microdevice.
Tip: The adhesive may have a yellow color after bonding. If necessary, allow 24 hours for adhesive to clear

Step 7: Allow a few seconds for the adhesive to wick along the ribs, then cure device for 20 seconds with ~350 nm UV light at [Intensity]

Fig. 2 Process workflow



What else should I know?

Fig. 3 Cross sectional image of a PS microchannel bonded to a glass coverslip. Scale bar represent 0.5 mm

•          Adhesives are often material-specific. Consult the manufacturer to determine the best adhesive for your application.

Tip: Some adhesives may require post-treatment / aging to reach a full cure.

•          We have tested this protocol with Ultra Light-Weld 3025 (Dymax) and Norland Optical Adhesive 68 (Thor Labs, Inc.). These adhesives had similar performance, however the protocol may need to be tailored for    other  adhesives.

•          If the adhesive is too viscous or does not adequately wick around the rib, heat may be applied to achieve thinner adhesive layers, or to improve the wicking of the adhesive.

•          This protocol is amenable to wide variety of materials, including: cyclic olefin copolymer (COC), glass, metal, PS, and various rapid-prototyping materials.

•          Creating the rib and allowing the adhesive to wick eliminates excess adhesive and prevents adhesive from squeezing into the microchannel.





References


1. Au, A. K., Lee, W., Folch, A., Lab Chip, 2014, 14(7), 1294-1301.
2. Guckenberger, D. J., de Groot, T., Wan, A. M.-D., Beebe, D., & Young, E., Lab Chip, 2015, 15(11), 2364–2378.
3. Young, E. W. K., Berthier, E., Guckenberger, D. J., Sackmann, E., Lamers, C., Meyvantsson, I., Beebe, D. J., Analytical Chemistry, 2011, 83(4), 1408–1417.
4. Gu, P., Liu, K., Chen, H., Nishida, T., Fan, Z. H., Anal. Chem., 2011, 83(1), 446-452
5. Lu, C., Lee, L. J., & Juang, Y. J., Electrophoresis, 2008, 29(7), 1407–1414.


Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

DIY peristaltic pump

Shannon Faley, Bradly Baer, Matthew Richardson, Taylor Larsen, and Leon M. Bellan*

Vanderbilt University, Department of Mechanical Engineering, Nashville TN, 37235, USA

*leon.bellan@vanderbilt.edu

Why is this useful?

Figure 1: Fully assembled peristaltic pump


The majority of microfluidic applications require an external pumping mechanism.  Multi-channel, individually addressable pumps are expensive, often large, and prone to failure when operated inside cell culture incubators at 95% humidity.  The number of experiments that can be run at a given time is limited by the availability and expense of pumps.  Perfusing artificial tissue scaffolds containing engineered vasculature requires long-term (days to weeks) continuous flow at low rates.  We designed an inexpensive (~$100 for 2 pumps, ~$70 for each additional set of 2 pumps) peristaltic pumping system using an Arduino- controlled stepper motor fitted with a custom 3D-printed pump head and laser-cut mounting bracket. Each pump has a footprint roughly that of the NEMA 17 stepper motor and is easily controlled individually using open source software.  Up to 64 motor shields can be stacked for a given Arduino Uno R3, each capable of supporting two stepper motors, and thus has the expansion potential to control 128 pumps in parallel.  We have successfully implemented two stacked motor shields driving four independent stepper motors. Flow rate is dependent upon both tubing diameter and step rate.  We found flow rates to range between ~50-250 μl/min for 1/16” tubing and ~500-1500 μl/min for 1/4″ tubing.  We anticipate that this pump design will likely prove more resilient to incubator humidity compared to standard peristaltic pump powered by DC motors.  Since implementation, these pumps have functioned without fail for 3 months (intermittent) under humid conditions. In the event of failure, however, cost of motor replacement is an economical $14.

Figure 2

What do I need?


Materials:

  • Nema 17 stepper motor ($14, spec, vendor)
  • Arduino Uno R3 Controller ($25, spec, vendor)
  • Arduino Motor Shield ($20, spec, vendor)
  • M3 machine screws (4) & hex bolts (4) ($1, McMaster-Carr)
  • DB9 Male & Female Solder Connectors ($9, StarTech)
  • 18AWG 4C speaker cable ($10, Monoprice)
  • Spring steel
  • ABS Filament
  • 6-32 machine screws & square nuts (3) ($1, McMaster-Carr)

Equipment:

  • 3D printer
  • Laser/Metal cutter
  • Soldering iron & solder
  • Butane torch

What do I do?


Pump head fabrication:

  1. Using ABS filament, 3D print pump head from file pumphead.crt.9
  2. Cut three 15 mm (length) sections from rigid ¼” tubing to serve as rollers.
  3. Use the three 6-32 machine screws and square nuts to assemble the tubing to pump head as shown in Figure 3.

Figure 3

Mounting bracket fabrication:

  1. Using bracket template file (2000 Pump Mount v4) and laser cutting facilities, produce a mounting bracket from spring steel, or other appropriate metal.  Note that the score line bisecting the bracket is intended to be cut at a lower power.  This line is just a marker to show where to bend the bracket in the following step.
  2. Using handheld butane torch, heat mounting bracket along score line and bend with pliers.  Repeat until mounting bracket forms a right angle (see Figure 1).

Motor Electrical Wiring: (see figure 4 for example orientation)

  1. Solder motor wires to DB9 Male Connector
  2. Solder one end of speaker wire to DB9 Female Connector
  3. Connect opposite end of speaker wire to Arduino Motor shield

Figure 4 Example of connection scheme by wire color

Pump Assembly:

  1. Use M3 machine screws to attach mounting bracket to stepper motor, with corresponding hex nuts as spacers between motor and bracket.
  2. Press fit pump head onto rotor shaft.
  3. Connect motor to Arduino using DB9 connectors

Arduino/Motor Shield Assembly:

  1. Follow assembly instructions provided by adafruit.com  (https://learn.adafruit.com/adafruit-motor-shield-v2-for-arduino/stacking-shields).  See also Figure 5.

Figure 5

Computer Control:

1. See online resources for easy starter code (https://learn.adafruit.com/adafruit-motor-shield-v2-for-arduino/install-software)

2. Load example code to control 2 stacked motor shields running four independent pumps simultaneously. (foursteppers_v2.ino)

3. Start pumping!  See  video clip for multi-pump demonstration:



Please click here to download the DIY Peristaltic Pump Files

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Low Temperature Melting Metal Solders For Electrical Interconnects On Plastics

Abdul Wasay, Dan Sameoto
Department of Mechanical Engineering, University of Alberta, Edmonton, T6G 2R3, CANADA.
Email: sameoto@ualberta.ca

Why is it useful?

Electrodes have been used for driving electrochemical reaction on various Lab on chip applications, running electroosmotic pumping or simple sensing of current or voltages. With the recent thrust to move to thermoplastics, simple electrodes have been patterned at low temperatures using vacuum plasma sputtering via intermediate masks1 or ink electrodes2. While, wire-bonding or soldering have been the standard protocols for providing interconnections to the bond pads on silicon or glass chips, the temperatures in these cases are often beyond the glass transition or even melting temperatures of most thermoplastic substrates, which leads to defects and ineffective connection. Other techniques such as conductive epoxies can be problematic due to longer cure times, permanent fixtures and potentially low uniformity of conductive properties which are affected by sintering times and temperatures.

We present a simple soldering technique using low temperature melting metals that is compatible with nearly any thermoplastic substrate and thin-film electrodes.

Fig.1. Basic Apparatus

  • Field’s metal (eutectic alloy of 32.5% Bi, 51% In, 16.5% Sn) preferred for low toxicity, although lead based alternatives exist.
  • Syringe connected to tygon tube
  • Copper wire
  • Aluminum foil dish
  • Electrodes patterned on thermoplastic substrate (here, polystyrene with  ~ 20 nm thick Au electrodes)

Tools

  • Hot Plate
  • Shearing scissor
  • Hot glue gun (in this case a Mastercraft dual temperature glue gun, operated at low temperature setting)

What do I do?

To extrude the Fields metal into a thin filament, (Fig.2)

  • melt the Fields metal (ROTO1443;  Tm=62.2oC, typical resistivity ~520nΩ-m) in an aluminum foil dish.  The metal is very soft, so bolt cutters can easily remove smaller portions for melting if desired.
  • use a suitable diameter and length  tygon tube and pull the molten metal into it using a syringe – it will freeze within a few inches (1/32” inner diameter tube shown here).
  • allow it to cool to room temperature (less than a minute) and then pull out the filament.

Soldering Process,(Fig.3.)

  • Hold the Field’s metal filament with a hot glue gun used for lower melting point glues (measured tip temperature   ̴120 C) over the electrode patterned petri dish and melt it over the electrodes.
  • Solder the copper wire with the Field’s metal.  After soldering, excess metal can be removed from the hot glue gun tip with paper towel.

The resultant solder is strong enough to be used for mainstream lab on chip applications. They can be removed by a strong tug, under which case the thin electrodes on plastics could be stripped.

Fig.2. Fields metal extrusion process

Fig.3. Soldering process

Fig.4. Gecko adhesives based reversibly bonded4 capillary electrophoresis device with integrated electrodes

Fig.5. a) Demonstration of solder strength b) small electrode stripping as observed when the solder is removed by a strong tug

What else should I know?

While there are quite a few options of low melting temperature metals to choose from, most of them contain heavy metal like lead or cadmium, which may ideally be avoided. While Field’s metal is relatively expensive due to the indium content, the solders can be recovered for reuse in the lab and in general is a better option for electrically connecting thin-film electrodes to temperature sensitive substrates in a quick, reliable fashion.

Watch the video here: Low Temperature Melting Metal Solders For Electrical Interconnects On Plastics

References:

1 A. Toossi, M.Sc, University of Alberta, 2012.
2 C. E. Walker, Z. Xia, Z. S. Foster, B. J. Lutz and Z. H. Fan, ELECTROANALYSIS, 2008, 20, 663-670.
3 http://www.rotometals.com/product-p/lowmeltingpoint144.htm
4 A. Wasay and D. Sameoto, Lab Chip, 2015, (DOI:10.1039/C5LC00342C).

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Micromilling Techniques II: Tool Alignment

David J. Guckenberger1, Theodorus E. de Groot1, Alwin M.D. Wan2, David J. Beebe1 and Edmond W. K. Young2,*

1 Department of Biomedical Engineering, Wisconsin Institutes for Medical Research, University of Wisconsin-Madison, Madison, WI, USA

2 Department of Mechanical and Industrial Engineering, University of Toronto, 5 King’s College Road, MC313, Toronto, ON, Canada.

* Corresponding Author, E-mail: eyoung@mie.toronto.ca; Tel: +1(416) 978 1521

Why is this useful?


­­­­­­­­­­Micromilling is a highly efficient method for fabricating microfluidic devices directly in polymeric materials like thermoplastics. Please see the review article by Guckenberger and co-workers for a primer on micromilling.1 After securing your workpiece to the milling table,2 the next step in the milling process is to align the tool to the workpiece, thereby defining the coordinate origin. Many high-end mills designed for micromilling have automated tool alignment systems. However, lower cost mills that are also capable of milling microdevices may not have tool alignment systems, and the user must therefore manually align the tool to the workpiece to the desired accuracy. Here we present several alignment techniques that are low cost and can be performed by minimally trained users. We divide tool alignment into two separate directions: the vertical z-axis direction, and the planar x-y plane direction.

What do I need?


 

  • CNC Mill (PCNC 770, Tormach)
  • Tool, e.g. endmill of choice
  • Workpiece (properly secured to milling table, see ref. [2])

Specific tooling or materials required for each technique are referenced separately below, under the “Tooling Note”.

What do I do?

 

 


Z-axis Direction

Fig. 1. Four techniques for aligning the tool to the workpiece in the z-axis (from left to right): (i) reflection, (ii) chip, (iii) paper, and (iv) collet technique.

 

(i) Reflection Technique

Tooling Note: This technique requires a reflective surface (e.g., transparent materials such as PS or PMMA), but otherwise does not require any specific tooling.

 

 

Step 1: Start with the tip of the tool slightly above the workpiece, with the spindle turned off.
Step 2: Looking from a near planar location with respect to the workpiece, lower the tool until the tool itself comes in contact with its reflection.
Step 3: Set this location as z = 0.

Tip:         Placing a piece of paper behind the tool will improve contrast, making it easier to identify when the tool contacts the reflection. A magnifying glass can be used to improve visibility.

 

(ii) Chip Technique

Tooling Note: This technique does not require any specific tooling.

 

 

Step 1: Start with the tool slightly above the surface with the spindle running (i.e., the tool should be rotating).
Step 2: Lower the tool towards the surface until either (a) a chip is observed, (b) a mark is made on the surface, or (c) a sound is made from the tool cutting the material.
Step 3: Set this location as z = 0.

Tip:      This method works best for large and flat endmills, and can be more difficult with small endmills or any tool that is pointed or round at the tip. Note that this is a physical contact method, and will blemish the surface.

 

(iii) Paper Technique

Tooling Note: This technique requires a small piece of paper of known thickness.

 

 

Step 1: Start with the tool above the surface with the spindle turned off.
Step 2: Place a piece of paper (with known thickness) between the tool and the workpiece.
Step 3: While moving the piece of paper back and forth, lower the tool in stepwise increments.
Step 4: Continue lowering the tool until it causes resistance to the sliding piece of paper.
Step 5: Set this location as z = the thickness of the paper (e.g., .003”)

Tip:      Practice more to become comfortable with identifying when the endmill comes in contact with the surface.

(iv) Collet Technique

Tooling Note: This technique requires an ER20 tool holder (#31829, Tormach) from the Tormach Tooling System (TTS) and an ER20 1/8” collet (#30112, Tormach).

 

 

Step 1: Place tool in collet and secure using the setscrew on the side.
Step 2: Without the spindle running, lower the tool until it is just above the surface.
Step 3: Loosen the setscrew and allow the tool to fall into contact with the workpiece.
Step 4: Tighten the setscrew.
Step 5: Set this location as z = 0.

 

X-Y Plane Direction

Fig 2. Techniques for aligning the tool in the xy plane. (A) Illustrations of the (i) edgefinder, (ii) chip, and (iii) paper techniques. (B) Guide to offsetting a tool (left) and finding the center of an object (right).

 

(i) Edgefinder Technique

Tooling Note:
This technique requires an edgefinder (e.g., #02035186, MSC Industrial Supply).

 

 

Step 1: Place edgefinder in collet and start spindle (1000 rpm works well with the edgefinder).
Step 2: Deflect the tip of the edgefinder so that it lies eccentric to its initial axis.
Step 3: Starting with the x-axis, move the edgefinder toward a perpendicular surface. The tip of the edgefinder will become concentric upon contact with surface, and then return to an eccentric position immediately afterward. This sudden “jump” in eccentricity marks the edge.
Step 4: Set the current location to either plus or minus the radius of the tip. See Fig. 2B for more details on deciding a positive or negative bias.

Chip Technique

Tooling Note: This technique does not require any specific tooling.

 

 

Step 1: Start with the tool near the face of interest with the spindle running (i.e., the tool should be rotating).
Step 2: Step the tool towards the surface, until (a) a chip is observed, (b) a mark is made on the surface, or (c) a sound is made from the tool cutting the material.
Step 3: Set this location as (plus or minus) the radius of the endmill.

Tip:      This method works best for large diameter endmills. Note that this is a physical contact method, and will blemish the surface.

Paper Technique

Tooling Note: This technique requires a small piece of paper of known thickness.

 

 

Step 1: Start with the tool near the face of interest with the spindle running
Step 2: Gripping gently with thumb and fore-finger, or by pressing it against the surface place a piece of paper between the tool and the surface.
Step 3: Step the tool towards the surface until the tool pulls the paper.
Step 4: Set this location as (plus or minus) the sum of the endmill radius and paper thickness.

Tip:                  This method works best for large endmills.
Caution:          Be sure to keep fingers clear of the cutting tool.

References


1.   Guckenberger DJ, de Groot T, Wan AMD, Beebe DJ, Young EWK, “Micromilling: A method for ultra-rapid prototyping of plastic microfluidic devices”, Lab on a Chip, DOI:10.1039/c5lc00234f (2015).

2.   Guckenberger DJ, de Groot T, Wan AMD, Beebe DJ, Young EWK, “Micromilling Techniques I: Securing Thin Plastic Workpieces for Precise Milling”, Chips & Tips (2015).

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Micromilling Techniques I: Securing Thin Plastic Workpieces for Precise Milling

David J. Guckenberger1, Theodorus E. de Groot1, Alwin M.D. Wan2, David J. Beebe1 and Edmond W. K. Young2,*

1 Department of Biomedical Engineering, Wisconsin Institutes for Medical Research, University of Wisconsin-Madison, Madison, WI, USA

2 Department of Mechanical and Industrial Engineering, University of Toronto, 5 King’s College Road, MC313, Toronto, ON, Canada.

* Corresponding Author, E-mail: eyoung@mie.toronto.ca; Tel: +1(416) 978 1521

Why is this useful?


Micromilling is a highly efficient method for fabricating microfluidic devices directly in polymeric materials like thermoplastics. A recent review article highlights the use and relevance of micromilling in the field of microfluidics.1 While milling is most popular among machinists, the technique is becoming cheaper and more accessible to many others. Thus, there is a need for disseminating technical know-how for achieving quality micromilled parts. One common technical issue is ensuring flat work surfaces, and eliminating warping and bending of thin plastics during micromilling, which can lead to uneven milling and large errors in feature dimensions. Here we present a simple technique to achieve flat work surfacesleveled to within 40 μin / in (.04 μm / mm)) – and properly secure thin plastic workpieces to minimize warping. Please refer to the review article by Guckenberger and co-workers for a primer on micromilling

What do I need?


Tools:

  • CNC mill                                 (PCNC 770, Tormach)
  • 3/8” set screw holder       (#31820, Tormach)
  • Granite block                        (6” x 12” x 2” Grade AA, Standridge)
  • T-slot clamps                        (#32580, Tormach)
  • Drop test indicator             (#24-315-4, SPI)
  • Wrench                                    (choose appropriate size for clamps)

Materials:

  • Plastic workpiece                        (e.g., polystyrene, cyclo-olefin copolymer (COC), PMMA)
  • Surface protection tape            (#6092A24, McMaster-Carr)
  • Two-sided or transfer tape      (#9472LE, 3M)
  • Silicone rubber, 1/16” thick    (#8632K41 – Durometer Hardness: 40A, McMaster-Carr)

Fig. 1 Tooling and Materials. (A) Hardware necessary for setting up the granite block. Assembly of the T-slot clamps is shown in Fig. 2. (B) (Top) Components needed for the drop test indicator: (i) collet/tool holder, (ii) aluminum adapter, (iii) drop test indicator, (iv) screw to attach indicator to the adapter. (Bottom) Side view of the assembled components. (C) Materials needed for adhering workpiece to the block: (i) surface protection tape, (ii) silicone rubber, (iii) transfer tape, (iv) polystyrene sheet.

What do I do?


Preparation: The dial indicator must be affixed to the head of the mill. To do so, mounting brackets can be purchased or custom-made, as we have done using a block of aluminum. This aluminum serves as a simple adapter between the dial indicator and the collet.

Step 1: Place granite block on the milling table. Note that a stiff yet compliant material is necessary between the block and the milling table to allow for minor adjustments. The particle-board feet affixed to the granite block by the manufacturer (Standridge) are sufficient.

Step 2: Assemble the clamps (Fig. 2A), and position the clamps in either a 3-clamp (Fig. 2B) or 4-clamp (Fig. 2C) configuration. The front of the clamp (i.e., the portion that contacts the granite block) should be the same height as or lower than the rear of the clamp.

Step 3: Hand-tighten each clamp, then tighten an addition 1/8th turn using a wrench.

Step 4: Level the granite block by using the drop test indicator to measure the height across the entire block. The following steps are recommended:

a)      Lower the indicator onto the center of the block.
b)      Move the indicator along x-axis to the two ends of the block, and note the height of both ends.
c)      Position the indicator on the high side of the block and tighten the clamps until the height matches the low side.
d)     Repeat steps (b) and (c) for the y-axis.
e)      Repeat steps (b) to (d) until the block is level to desired tolerance.
f)       Tip: When using the 4-clamp configuration, clamps should be equally tightened in pairs to maintain levelness in the other direction.

Fig. 2 (A) T-slot clamp setup configuration. (B) Granite block configuration using three clamps. (C) Granite block configuration using four clamps.

Step 5: (Optional) Apply a protective adhesive film to the workpiece. This will simplify removal of the transfer tape from the workpiece (see Video).

a)      Peel a piece of protective adhesive film from the roll of film.
b)      Place a piece of silicone rubber on the roll of tape or other cylindrical roll.
c)      Place the protective adhesive film on top of the silicone rubber with the adhesive side face up, assuring the tape is flat.
d)     Roll the workpiece onto the piece of tape. The silicone rubber will provide compliance below the tape to prevent bubbles in the tape.
e)      Press the tape down to make sure it is fully adhered to the workpiece.

Step 6: Apply 2-sided tape or transfer tape to the workpiece

Step 7: Stick the workpiece onto the granite block.

Step 8: (Optional) Use the drop test indicator again to ensure the workpiece is flat.

References


1.   Guckenberger DJ, de Groot T, Wan AMD, Beebe DJ, Young EWK, “Micromilling: A method for ultra-rapid prototyping of plastic microfluidic devices”, Lab on a Chip, DOI:10.1039/c5lc00234f (2015).

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

A simple and inexpensive device to remove edge beads

Markus Ludwig and Seth Fraden

Martin A. Fisher School of Physics, Brandeis University, Waltham, Massachusetts 02454, United States

Why is this useful?


To achieve high resolution and high aspect ratio features in contact lithography it is necessary to have the photomask in direct contact with the photoresist during the exposure. This is impeded by the formation of edge beads on the edge of the wafer, which can measure multiples of the nominal resist thickness [1].

Dedicated spin coaters remove edge beads automatically just after spin coating. However, during soft baking, a new edge bead forms due to the coffee ring effect [2]. It is therefore recommended, especially for thicker resist films, to remove the edge bead not before, but after soft baking [3].

Edge bead removal is not critical on single layer devices. However, for multi level designs, edge bead removal can improve the feature resolution significantly. This is especially the case if the first layer of resist is thick and the second layer is thin.

This article presents the fabrication and use of a simple and inexpensive device to reliably remove edge beads by solvent spraying.

What do I need?


Materials:

  • 100 ml round media storage bottle GL45
  • Diba Labware bottle cap Q series GL45 2 x 1/4 -28 ports with valves (00945Q-2V)
  • acrylic sheet, 12” x 6”, 3/16” thick
  • polyethylene sheet, 1/16 “ thick
  • 3x AAA battery pack with on/off switch + batteries
  • Parker micro diaphragm pump T2-05 IC (3/32 ID)
  • 2x ¼-28 to barbed 3/32 ID fitting
  • silicone tubing  0.078” ID, 50 cm length
  • 2x flat head machine screw 4-40 3/8 + nuts
  • 2x socket head machine screw 6-32 3/8 + nuts + washers
  • stainless steel blunt needle with 27 gauge Luer polypropylene hub, 1/2″ length
  • cable tie 4”
  • polypropylene tube fitting, male Luer slip to barbed coupler, 3/32″ tube ID

Tools:

  • hot air gun (Aoyue 968a+)
  • laser cutter (40W/45W CO2 Hobby Laser by Full Spectrum Laser) or scroll saw, drill press and drill bits
  • soldering iron (Aoyue 968a+) and solder

Figure 1: Materials needed

What do I do?


To build the edge bead remover, first prepare all components as shown in figure 1.

Cut the acrylic sheet and the polypropylene sheet and drill the holes as indicated in the schematic1.pdf or schematic1.dwg (AutoCAD) file. Solder battery pack wires to the pump. Then follow the instructions in video 1.

1. The blunt needle has to be bent at a right-angle.

Figure 2: Assembled device

Edge bead removal (follow the instructions in video 2):

  1. Spin coat wafer, soft bake and cool down.
  2. Center the wafer on spin coater and spin at 700 rpm. The optional centering tool used in the video was 3D printed on a formlabs form one 3D printer. (Use centering tool.stl file to reprint)
  3. Turn on pump, open pump valve to pressurize bottle, position nozzle over edge bead.
  4. Open spray valve and dissolve edge bead
  5. Sweep outwards slowly and keep spraying for another 15 seconds.
  6. Turn off spray valve and ramp up to 2000 rpm for 10 sec.
  7. Turn off pump and close both valves. Developer may remain in the bottle and in the tubing.
  8. Wafer does not have to be baked again and can be exposed immediately.

Figure 2: SU8 coated 3” wafer before (left) and after edge bead removal (right).

Figure 3: Comparison of feature resolution

CAD drawing of a multi layer design

SU8 master fabricated with edge bead

SU8 master fabricated without edge bead

Video 1:

Video 2:

References


[1] ‪ Shaurya Prakash, Junghoon Yeom, Nanofluidics and Microfluidics: Systems and Applications ‪Micro and Nano Technologies‪, William Andrew, 2014

[2] Robert D. Deegan, Olgica Bakajin, Todd F. Dupont, Greb Huber, Sidney R. Nagel and Thomas A. Witten, Capillary flow as the cause of ring stains from dried liquid drops, Nature 389, 827-829 (23 October 1997), doi :10.1038/39827

[3] http://www.cns.fas.harvard.edu/facilities/docs/SOP031_r2_6_SU-8%20photolithography%20process.pdf

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

A simple and low cost method to fabricate NOA microfluidic chips

Gabriele Pitingolo1, Raffaele Vecchione1,3 and Paolo A. Netti1,2,3

1 Center for Advanced Biomaterials for Healthcare, Istituto Italiano di tecnologia (IIT@CRIB), Largo Barsanti e Matteucci, 53, 80125, Naples, Italy.

2 Dipartimento di Ingegneria Chimica, dei Materiali e della Produzione Industriale D.I.C.MA.P.I, Università di  Napoli  Federico II, Naples 80125, Italy.

3 Centro di Ricerca Interdipartimentale sui Biomateriali (CRIB), Università di Napoli Federico II, p.le Tecchio 80, Naples, 80125, Italy

Why is this useful?


Microfluidic chips are often made of silicon or glass which presents the drawbacks of being relatively expensive, time consuming and has limitations to the geometries that can be realized. PMMA is an optimal solution to overcome these aspects but it presents a low chemical resistance to organic solvents and aggressive chemicals

Norland Optical Adhesive 60 (“NOA60”) is a clear, colorless, liquid photopolymer that cures when exposed to ultraviolet light1. Surface bonding can be activated with light therefore monolithic and transparent devices especially useful for optical elements can be realized. In particular, the use of NOA 60 eliminates premixing, drying, or heat curing operations common to other optical adhesive systems. Curing time is a matter of minutes and is dependent upon the thickness applied and the energy of ultraviolet light available. Dupont and colleagues have recently developed a NOA microfluidic channel via a photolithography multistep method that presents a long time process2.

Here, we demonstrate the possibility to micromachine already cured NOA substrates by micromilling that is much easier and cheaper than photolithographic techniques for fabrication of microchannels or microstructures in general.  In addition, by micromilling it is possible to easily drill and make open channels in NOA substrates if needed. Also, in the case of microstructures on the two layers to be bonded if one layer presents microstructures with feature sizes below 25 micron and one substrate with feature sizes above this value then it is possible to prepare one substrate by photolithographic techniques and the other substrate by micromilling with following bonding, saving time and money.

What do I need?


  • Fully cured PDMS mold
  • Norland Optical Adhesive 60
  • UV light (E-Series Ultraviolet Hand Lamps)
  • Micromachining machine
  • Oxygen plasma machine
  • Clamp

What do I do?


1. Pour liquid photopolymer NOA into a preformed PDMS mold covering the entire surface (figure 1A). After a few minutes to stabilize the liquid polymer put the PDMS mold under UV light for 30 minutes at a  365 nm wavelength (fig.1B).

2. After the curing time, NOA substrate is ready to use; to fabricate a microfluidic chip two NOA substrates are prepared. NOA substrates are replicated onto flat PDMS surfaces exploiting the flexibility of the PDMS mold as shown in figures 2A and 2B.

3. Take the NOA substrate and mill a channel and related inlet/outlet holes using a micromilling machine (Minitech CNC Mini-Mill) (fig. 3A-3B), the certified positioning accuracy of the three-axis are 12″ / 300mm in x-axis, 9″ / 228mm in y-axis, and 9″ / 228mm in z-axis. To minimize the experimental uncertainty, the NOA substrates preformed in point 2 are smoothed before milling.

4. Prepare the NOA channel (clean with water and dry with an absorbent cloth) and  treat the channel and top layer by exposing to oxygen plasma for 60s, at a pressure less than 0.1 Torr and with a plasma power of 20 W (fig 4A). Clamp the two substrates and put the clamped channel under UV-light for 1 h finalizes the bonding process (fig 4B).

5. Your well-bonded NOA microfluidic chip (Figure 5) is now ready to use.


References

  1. https://www.norlandprod.com/literature/60tds.pdf
  2. E. P. Dupont, R. Luisier and M. A. M. Gijs, Microelectronic Engineering, 2010, 87, 1253-1255.
Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)