Digital Microfluidic Chip


#1

Hello!

I see everything besides the actual chip in the hardware list. Where can this be purchased or made? I have a synthetic biology club that would be interested in making a dropbot to use while working on the iGEM competition next summer.

I also wanted to know if you were familiar with DMF chips that had cross referencing capabilities to open up the area you can work with quite a bit? http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233

There are also methods to get the voltage down to less than 15 volts: http://www.sciencedirect.com/science/article/pii/S0038110108001275

I was wondering if you have already considered these technologies in future designs?

Thanks,
–Ryan George


#2

Hi Ryan,

Thanks for your email. These questions actually come up a lot, so it's
great to get this info out there for everyone. You've nailed a really
big problem with DMF as it currently stands. I think that the DropBot
system makes it easy for anyone to build a DMF control system, but
unless you can get/make devices, it's not incredibly useful. Our lab
has published lots of DIY-friendly, rapid prototyping techniques for
DMF devices (http://microfluidics.utoronto.ca/publications.php)
including micro-contact priting, PCB-based devices, dielectric films,
etc., but none of them are perfect, and certainly the devices made
using glass-substrate/photolithography have the best performance in my
experience.

We fabricate our own chips in the cleanroom facility at the University
of Toronto. Fabrication details can be found in any the publications
on our website (e.g.,
http://microfluidics.utoronto.ca/papers/DropBot%20paper.pdf).

I am familiar with the cross-referencing scheme (I think this was
originally published by CJ Kim's group several years back, before I
started). Some of the issues, if I remember correctly, were:

1. electrode cells in the same row/column as the target will be at
half potential
2. your software requires logic to avoid conflicts when actuating
multiple electrodes in parallel
3. probably requires 2 HV amplifiers (180 degrees out of phase)

In any case, it would be possible to build a DropBot system that could
support cross-referenced devices without too much trouble. But if you
require two amplifiers, any cost savings from limiting the number of
switches may be negligible. I do think this is a promising scheme
though, especially because it would really allow your device
geometries to scale up.

I'm also familiar with some of the efforts to make low voltage
devices. I think this could be really important for reducing the cost
of the control electronics and would also have some other nice
benefits (safety, portable, battery-powered systems). Right now the
biggest expenses from the electronics side are the high-voltage
amplifier and the high-voltage solid state relays. If drops could be
manipulated with <20 V, the whole control system could be much
cheaper.

We've mostly been focused (so far) on improving device robustness and
reliability and not so much on these other issues. To reduce the
voltage, you need to either (a) increase the dielectric constant of
your insulating layer and/or (b) reduce the thickness of the
dielectric. Previous students in the Wheeler lab have experimented
with some alternative dielectrics (e.g., barium titanate, silicon
nitride) that have higher a dielectric constant, but with our
equipment, we can only coat devices in small batches (i.e., <4),
whereas using Parylene-C, we are able to coat >100 at a time. Also,
with thinner dielectrics, you can run into problems because it is more
common to get small imperfections. Any failure point in the
dielectric, will cause breakdown and electrolysis of your drop.

I'm planning to setup some kind of a wiki or blog on the DropBot
website so that we can organize some info on potential low-cost device
fabrication strategies, future development ideas, etc. Right now, I'm
mainly focused on writing hardware assembly documentation as well as
my other research papers so that I can eventually graduate :wink:

But it is a near term goal to at least be able to support people who
want to experiment with different device fabrication strategies. I
suspect that there are people out there in the community who can
figure out a great way to make cheap, reliable, and maybe even
low-voltage devices.

-Ryan

···

On Thu, Sep 12, 2013 at 11:31 AM, Ryan George <ryang...@knights.ucf.edu> wrote:

Hello!

I see everything besides the actual chip in the hardware list. Where can
this be purchased or made? I have a synthetic biology club that would be
interested in making a dropbot to use while working on the iGEM competition
next summer.

I also wanted to know if you were familiar with DMF chips that had cross
referencing capabilities to open up the area you can work with quite a bit?
http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233

There are also methods to get the voltage down to less than 15 volts:
http://www.sciencedirect.com/science/article/pii/S0038110108001275

I was wondering if you have already considered these technologies in future
designs?

Thanks,
--Ryan George

--
You received this message because you are subscribed to the Google Groups
"dropbot-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to dropbot-dev...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.


#3

My Senior Design Project team in currently building a DropBot for one of our professors. We will use micro-fabrication strategies to build our own DMF chip. We elected to use transparency masks and silicon wafers because we have a clean-room which is capable of handling our custom design.

···

On Thursday, September 12, 2013 8:31:46 AM UTC-7, Ryan George wrote:

Hello!

I see everything besides the actual chip in the hardware list. Where can this be purchased or made? I have a synthetic biology club that would be interested in making a dropbot to use while working on the iGEM competition next summer.

I also wanted to know if you were familiar with DMF chips that had cross referencing capabilities to open up the area you can work with quite a bit? http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233

There are also methods to get the voltage down to less than 15 volts: http://www.sciencedirect.com/science/article/pii/S0038110108001275

I was wondering if you have already considered these technologies in future designs?

Thanks,
–Ryan George


#4

Hi all,

I just found out about the project and I find it really fascinating!

I’m in a lab where we do very repetitive experiments on ALS patient fibroblasts and I think the DropBot would be a great step forward.

I’ve built 3d printers in the past, soldered a little bit and can program somewhat in python, so I should be able to build it as cheap as possible and get a lot of functionality out of it.

I want to convince my boss, most of my questions were already answered here but I still have a few:

Are chips reusable, or you throw them out all the time?

Does teflon bind RNA/DNA from the droplets? This would be important for example for PCR, where very little cross contamination can cause problems.

You mentioned the chromium plated glass and I saw the jove publication (http://www.jove.com/video/1603/digital-microfluidics-for-automated-proteomic-processing).

Do you always make them, or is there a company that sells them? I think the litography and the spin coating would be doable in our lab.

Also in the jove I don’t understand why the protein precicpitate stays in place while the droplet moves away. I saw that the droplet is moved very slow. Is this not a tricky step? Do you have to design it in a way that you have a lot of precipitate? Did you ever try to precipitate DNA or RNA?

When doing magnetic separation do you use an inductor to generate a field controlled by the electronics or put a magnet under the chip?

You mentioned somewhere that you are working on inkjet printable chips, is it by AgIC(http://shop.agic.cc/collections/circuit-printer)? Do you spincoat the paper?

Can parts of the chip be heated (like not more than 100C) or do the droplets evaporate?

If there answers for these in your papers than sorry, don’t bother to answer i still need to go through them.

Thank you and bests,

Csaba

···

On Thursday, September 12, 2013 at 1:42:13 PM UTC-4, Ryan Fobel wrote:

Hi Ryan,

Thanks for your email. These questions actually come up a lot, so it’s

great to get this info out there for everyone. You’ve nailed a really

big problem with DMF as it currently stands. I think that the DropBot

system makes it easy for anyone to build a DMF control system, but

unless you can get/make devices, it’s not incredibly useful. Our lab

has published lots of DIY-friendly, rapid prototyping techniques for

DMF devices (http://microfluidics.utoronto.ca/publications.php)

including micro-contact priting, PCB-based devices, dielectric films,

etc., but none of them are perfect, and certainly the devices made

using glass-substrate/photolithography have the best performance in my

experience.

We fabricate our own chips in the cleanroom facility at the University

of Toronto. Fabrication details can be found in any the publications

on our website (e.g.,

http://microfluidics.utoronto.ca/papers/DropBot%20paper.pdf).

I am familiar with the cross-referencing scheme (I think this was

originally published by CJ Kim’s group several years back, before I

started). Some of the issues, if I remember correctly, were:

  1. electrode cells in the same row/column as the target will be at

half potential

  1. your software requires logic to avoid conflicts when actuating

multiple electrodes in parallel

  1. probably requires 2 HV amplifiers (180 degrees out of phase)

In any case, it would be possible to build a DropBot system that could

support cross-referenced devices without too much trouble. But if you

require two amplifiers, any cost savings from limiting the number of

switches may be negligible. I do think this is a promising scheme

though, especially because it would really allow your device

geometries to scale up.

I’m also familiar with some of the efforts to make low voltage

devices. I think this could be really important for reducing the cost

of the control electronics and would also have some other nice

benefits (safety, portable, battery-powered systems). Right now the

biggest expenses from the electronics side are the high-voltage

amplifier and the high-voltage solid state relays. If drops could be

manipulated with <20 V, the whole control system could be much

cheaper.

We’ve mostly been focused (so far) on improving device robustness and

reliability and not so much on these other issues. To reduce the

voltage, you need to either (a) increase the dielectric constant of

your insulating layer and/or (b) reduce the thickness of the

dielectric. Previous students in the Wheeler lab have experimented

with some alternative dielectrics (e.g., barium titanate, silicon

nitride) that have higher a dielectric constant, but with our

equipment, we can only coat devices in small batches (i.e., <4),

whereas using Parylene-C, we are able to coat >100 at a time. Also,

with thinner dielectrics, you can run into problems because it is more

common to get small imperfections. Any failure point in the

dielectric, will cause breakdown and electrolysis of your drop.

I’m planning to setup some kind of a wiki or blog on the DropBot

website so that we can organize some info on potential low-cost device

fabrication strategies, future development ideas, etc. Right now, I’m

mainly focused on writing hardware assembly documentation as well as

my other research papers so that I can eventually graduate :wink:

But it is a near term goal to at least be able to support people who

want to experiment with different device fabrication strategies. I

suspect that there are people out there in the community who can

figure out a great way to make cheap, reliable, and maybe even

low-voltage devices.

-Ryan

On Thu, Sep 12, 2013 at 11:31 AM, Ryan George > > ryan...@knights.ucf.edu wrote:

Hello!

I see everything besides the actual chip in the hardware list. Where can

this be purchased or made? I have a synthetic biology club that would be

interested in making a dropbot to use while working on the iGEM competition

next summer.

I also wanted to know if you were familiar with DMF chips that had cross

referencing capabilities to open up the area you can work with quite a bit?

http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233

There are also methods to get the voltage down to less than 15 volts:

http://www.sciencedirect.com/science/article/pii/S0038110108001275

I was wondering if you have already considered these technologies in future

designs?

Thanks,

–Ryan George

You received this message because you are subscribed to the Google Groups

“dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an

email to dropbot-dev...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.


#5

Hi Csaba,

Thanks for your email! I will try to answer your questions as best I can.

Hi all,

I just found out about the project and I find it really fascinating!
I'm in a lab where we do very repetitive experiments on ALS patient
fibroblasts and I think the DropBot would be a great step forward.
I've built 3d printers in the past, soldered a little bit and can program
somewhat in python, so I should be able to build it as cheap as possible and
get a lot of functionality out of it.

I want to convince my boss, most of my questions were already answered here
but I still have a few:

Are chips reusable, or you throw them out all the time?

The time your chips last will depend a lot on how good your
fabrication process is and what you're doing with the DMF chips. If
you're moving liquids with a lot of proteins in them, they will tend
to foul over time and drop movement becomes increasingly difficult. If
you are moving protein-free liquids, devices can last a very long
time. Oil-filled devices will also last longer relative to those
operated in air.

There are also some tricks that can partially recover the function of
devices that have been fouled by proteins (e.g., reflowing Teflon by
heating your devices on a hot-plate).

Does teflon bind RNA/DNA from the droplets? This would be important for
example for PCR, where very little cross contamination can cause problems.

Personally, I don't have a lot of experience in moving DNA/RNA, but
there are certainly many examples in the literature of people doing
PCR. My understanding is that moving DNA/RNA is less challenging as
compared to proteins since DNA/RNA are usually hydrophilic, which
means that they will have less of a tendency to stick to the
hydrophobic Teflon surface.

If you are interested in PCR, I think the bigger challenge is
preventing drop evaporation. Most of the people doing PCR on DMF use
an oil filler.

You mentioned the chromium plated glass and I saw the jove publication
(http://www.jove.com/video/1603/digital-microfluidics-for-automated-proteomic-processing).
Do you always make them, or is there a company that sells them? I think the
litography and the spin coating would be doable in our lab.

We used to chromium plate our own glass slides by e-beam, but now we
mainly purchase them from a company called Telic
(http://www.teliccompany.com/). That jove paper is quite old and our
fabrication methods have improved since then; it's better to look at
any of our more recent publications for the latest procedures.

Also in the jove I don't understand why the protein precicpitate stays in
place while the droplet moves away. I saw that the droplet is moved very
slow. Is this not a tricky step? Do you have to design it in a way that you
have a lot of precipitate? Did you ever try to precipitate DNA or RNA?

Proteins with hydrophobic domains will stick to the Teflon, which in
most cases is a bad thing because it impedes droplet movement, but in
this case it will help to anchor the precipitated proteins to the
surface. Moving a droplet over the surface produces very low shear
stress, which is probably why the proteins remain in place. I don't
think that this step is particularly tricky. Nobody in our lab has
worked with precipitating DNA/RNA as far as I know...

When doing magnetic separation do you use an inductor to generate a field
controlled by the electronics or put a magnet under the chip?

We use a permanent magnet. See the following paper for details:

Choi, K.; Ng, A.H.C.; Fobel, R.; Chang-Yen, D.A.; Yarnell, L.E.;
Pearson, E.L.; Oleksak, C.M.; Fischer, A.T.; Luoma, R.P.; Robinson,
J.M.; Audet, J.; Wheeler, A.R. "Automated Digital Microfluidic
Platform for Magnetic-Particle-Based Immunoassays with Optimization by
Design of Experiments" Analytical Chemistry 2013, 85, 9638-9646.
(available on our website:
http://microfluidics.utoronto.ca/publications.php)

You mentioned somewhere that you are working on inkjet printable chips, is
it by AgIC(http://shop.agic.cc/collections/circuit-printer)? Do you spincoat
the paper?

The inkjet printed devices are described in this paper:

Fobel, R.; Kirby, A.E.; Ng, A.H.C.; Farnood, R.R.; Wheeler, A.R.
"Paper Microfluidics Goes Digital" Adv. Mat. 2014, 26, 2838-2843.
(available on our website:
http://microfluidics.utoronto.ca/publications.php)

Can parts of the chip be heated (like not more than 100C) or do the droplets
evaporate?

The temperature limitation is mainly set by the dielectric. Parylene-C
is rated for a continuous service temperature of 80C and a short term
service temperature of 100C. We exceed this temperature during the
Teflon reflow step of our fabrication process, and it doesn't result
in catastrophic failure, but exceeding 80C for extended periods of
time may damage your device over the long term (Parylene data sheet:
http://vsiparylene.com/pdf/ParyleneProperties2013.pdf).

Evaporation will definitely be a problem if you plan to heat your
drops to anywhere close to 100C. You can ameliorate this problem
through drop replenishment and/or using an oil filler.

Hope that helps,

-Ryan

···

On Sat, Mar 7, 2015 at 6:32 PM, Csaba Konrád <csaba....@gmail.com> wrote:

If there answers for these in your papers than sorry, don't bother to answer
i still need to go through them.

Thank you and bests,
Csaba

On Thursday, September 12, 2013 at 1:42:13 PM UTC-4, Ryan Fobel wrote:

Hi Ryan,

Thanks for your email. These questions actually come up a lot, so it's
great to get this info out there for everyone. You've nailed a really
big problem with DMF as it currently stands. I think that the DropBot
system makes it easy for anyone to build a DMF control system, but
unless you can get/make devices, it's not incredibly useful. Our lab
has published lots of DIY-friendly, rapid prototyping techniques for
DMF devices (http://microfluidics.utoronto.ca/publications.php)
including micro-contact priting, PCB-based devices, dielectric films,
etc., but none of them are perfect, and certainly the devices made
using glass-substrate/photolithography have the best performance in my
experience.

We fabricate our own chips in the cleanroom facility at the University
of Toronto. Fabrication details can be found in any the publications
on our website (e.g.,
http://microfluidics.utoronto.ca/papers/DropBot%20paper.pdf).

I am familiar with the cross-referencing scheme (I think this was
originally published by CJ Kim's group several years back, before I
started). Some of the issues, if I remember correctly, were:

1. electrode cells in the same row/column as the target will be at
half potential
2. your software requires logic to avoid conflicts when actuating
multiple electrodes in parallel
3. probably requires 2 HV amplifiers (180 degrees out of phase)

In any case, it would be possible to build a DropBot system that could
support cross-referenced devices without too much trouble. But if you
require two amplifiers, any cost savings from limiting the number of
switches may be negligible. I do think this is a promising scheme
though, especially because it would really allow your device
geometries to scale up.

I'm also familiar with some of the efforts to make low voltage
devices. I think this could be really important for reducing the cost
of the control electronics and would also have some other nice
benefits (safety, portable, battery-powered systems). Right now the
biggest expenses from the electronics side are the high-voltage
amplifier and the high-voltage solid state relays. If drops could be
manipulated with <20 V, the whole control system could be much
cheaper.

We've mostly been focused (so far) on improving device robustness and
reliability and not so much on these other issues. To reduce the
voltage, you need to either (a) increase the dielectric constant of
your insulating layer and/or (b) reduce the thickness of the
dielectric. Previous students in the Wheeler lab have experimented
with some alternative dielectrics (e.g., barium titanate, silicon
nitride) that have higher a dielectric constant, but with our
equipment, we can only coat devices in small batches (i.e., <4),
whereas using Parylene-C, we are able to coat >100 at a time. Also,
with thinner dielectrics, you can run into problems because it is more
common to get small imperfections. Any failure point in the
dielectric, will cause breakdown and electrolysis of your drop.

I'm planning to setup some kind of a wiki or blog on the DropBot
website so that we can organize some info on potential low-cost device
fabrication strategies, future development ideas, etc. Right now, I'm
mainly focused on writing hardware assembly documentation as well as
my other research papers so that I can eventually graduate :wink:

But it is a near term goal to at least be able to support people who
want to experiment with different device fabrication strategies. I
suspect that there are people out there in the community who can
figure out a great way to make cheap, reliable, and maybe even
low-voltage devices.

-Ryan

On Thu, Sep 12, 2013 at 11:31 AM, Ryan George >> <ryan...@knights.ucf.edu> wrote:
> Hello!
>
> I see everything besides the actual chip in the hardware list. Where can
> this be purchased or made? I have a synthetic biology club that would be
> interested in making a dropbot to use while working on the iGEM
> competition
> next summer.
>
> I also wanted to know if you were familiar with DMF chips that had cross
> referencing capabilities to open up the area you can work with quite a
> bit?
>
> http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233
>
> There are also methods to get the voltage down to less than 15 volts:
> http://www.sciencedirect.com/science/article/pii/S0038110108001275
>
> I was wondering if you have already considered these technologies in
> future
> designs?
>
> Thanks,
> --Ryan George
>
> --
> You received this message because you are subscribed to the Google
> Groups
> "dropbot-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an
> email to dropbot...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

--
You received this message because you are subscribed to the Google Groups
"dropbot-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to dropbot-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


#6

Ryan,

Thank you for the prompt reply, your answers are very useful! I think if I can make a chip my boss will be convinced.

Thanks again!

Best,

Csaba

···

On Monday, March 9, 2015 at 11:31:54 AM UTC-4, Ryan Fobel wrote:

Hi Csaba,

Thanks for your email! I will try to answer your questions as best I can.

On Sat, Mar 7, 2015 at 6:32 PM, Csaba Konrád csab...@gmail.com wrote:

Hi all,

I just found out about the project and I find it really fascinating!

I’m in a lab where we do very repetitive experiments on ALS patient

fibroblasts and I think the DropBot would be a great step forward.

I’ve built 3d printers in the past, soldered a little bit and can program

somewhat in python, so I should be able to build it as cheap as possible and

get a lot of functionality out of it.

I want to convince my boss, most of my questions were already answered here

but I still have a few:

Are chips reusable, or you throw them out all the time?

The time your chips last will depend a lot on how good your

fabrication process is and what you’re doing with the DMF chips. If

you’re moving liquids with a lot of proteins in them, they will tend

to foul over time and drop movement becomes increasingly difficult. If

you are moving protein-free liquids, devices can last a very long

time. Oil-filled devices will also last longer relative to those

operated in air.

There are also some tricks that can partially recover the function of

devices that have been fouled by proteins (e.g., reflowing Teflon by

heating your devices on a hot-plate).

Does teflon bind RNA/DNA from the droplets? This would be important for

example for PCR, where very little cross contamination can cause problems.

Personally, I don’t have a lot of experience in moving DNA/RNA, but

there are certainly many examples in the literature of people doing

PCR. My understanding is that moving DNA/RNA is less challenging as

compared to proteins since DNA/RNA are usually hydrophilic, which

means that they will have less of a tendency to stick to the

hydrophobic Teflon surface.

If you are interested in PCR, I think the bigger challenge is

preventing drop evaporation. Most of the people doing PCR on DMF use

an oil filler.

You mentioned the chromium plated glass and I saw the jove publication

(http://www.jove.com/video/1603/digital-microfluidics-for-automated-proteomic-processing).

Do you always make them, or is there a company that sells them? I think the

litography and the spin coating would be doable in our lab.

We used to chromium plate our own glass slides by e-beam, but now we

mainly purchase them from a company called Telic

(http://www.teliccompany.com/). That jove paper is quite old and our

fabrication methods have improved since then; it’s better to look at

any of our more recent publications for the latest procedures.

Also in the jove I don’t understand why the protein precicpitate stays in

place while the droplet moves away. I saw that the droplet is moved very

slow. Is this not a tricky step? Do you have to design it in a way that you

have a lot of precipitate? Did you ever try to precipitate DNA or RNA?

Proteins with hydrophobic domains will stick to the Teflon, which in

most cases is a bad thing because it impedes droplet movement, but in

this case it will help to anchor the precipitated proteins to the

surface. Moving a droplet over the surface produces very low shear

stress, which is probably why the proteins remain in place. I don’t

think that this step is particularly tricky. Nobody in our lab has

worked with precipitating DNA/RNA as far as I know…

When doing magnetic separation do you use an inductor to generate a field

controlled by the electronics or put a magnet under the chip?

We use a permanent magnet. See the following paper for details:

Choi, K.; Ng, A.H.C.; Fobel, R.; Chang-Yen, D.A.; Yarnell, L.E.;

Pearson, E.L.; Oleksak, C.M.; Fischer, A.T.; Luoma, R.P.; Robinson,

J.M.; Audet, J.; Wheeler, A.R. "Automated Digital Microfluidic

Platform for Magnetic-Particle-Based Immunoassays with Optimization by

Design of Experiments" Analytical Chemistry 2013, 85, 9638-9646.

(available on our website:

http://microfluidics.utoronto.ca/publications.php)

You mentioned somewhere that you are working on inkjet printable chips, is

it by AgIC(http://shop.agic.cc/collections/circuit-printer)? Do you spincoat

the paper?

The inkjet printed devices are described in this paper:

Fobel, R.; Kirby, A.E.; Ng, A.H.C.; Farnood, R.R.; Wheeler, A.R.

“Paper Microfluidics Goes Digital” Adv. Mat. 2014, 26, 2838-2843.

(available on our website:

http://microfluidics.utoronto.ca/publications.php)

Can parts of the chip be heated (like not more than 100C) or do the droplets

evaporate?

The temperature limitation is mainly set by the dielectric. Parylene-C

is rated for a continuous service temperature of 80C and a short term

service temperature of 100C. We exceed this temperature during the

Teflon reflow step of our fabrication process, and it doesn’t result

in catastrophic failure, but exceeding 80C for extended periods of

time may damage your device over the long term (Parylene data sheet:

http://vsiparylene.com/pdf/ParyleneProperties2013.pdf).

Evaporation will definitely be a problem if you plan to heat your

drops to anywhere close to 100C. You can ameliorate this problem

through drop replenishment and/or using an oil filler.

Hope that helps,

-Ryan

If there answers for these in your papers than sorry, don’t bother to answer

i still need to go through them.

Thank you and bests,

Csaba

On Thursday, September 12, 2013 at 1:42:13 PM UTC-4, Ryan Fobel wrote:

Hi Ryan,

Thanks for your email. These questions actually come up a lot, so it’s

great to get this info out there for everyone. You’ve nailed a really

big problem with DMF as it currently stands. I think that the DropBot

system makes it easy for anyone to build a DMF control system, but

unless you can get/make devices, it’s not incredibly useful. Our lab

has published lots of DIY-friendly, rapid prototyping techniques for

DMF devices (http://microfluidics.utoronto.ca/publications.php)

including micro-contact priting, PCB-based devices, dielectric films,

etc., but none of them are perfect, and certainly the devices made

using glass-substrate/photolithography have the best performance in my

experience.

We fabricate our own chips in the cleanroom facility at the University

of Toronto. Fabrication details can be found in any the publications

on our website (e.g.,

http://microfluidics.utoronto.ca/papers/DropBot%20paper.pdf).

I am familiar with the cross-referencing scheme (I think this was

originally published by CJ Kim’s group several years back, before I

started). Some of the issues, if I remember correctly, were:

  1. electrode cells in the same row/column as the target will be at

half potential

  1. your software requires logic to avoid conflicts when actuating

multiple electrodes in parallel

  1. probably requires 2 HV amplifiers (180 degrees out of phase)

In any case, it would be possible to build a DropBot system that could

support cross-referenced devices without too much trouble. But if you

require two amplifiers, any cost savings from limiting the number of

switches may be negligible. I do think this is a promising scheme

though, especially because it would really allow your device

geometries to scale up.

I’m also familiar with some of the efforts to make low voltage

devices. I think this could be really important for reducing the cost

of the control electronics and would also have some other nice

benefits (safety, portable, battery-powered systems). Right now the

biggest expenses from the electronics side are the high-voltage

amplifier and the high-voltage solid state relays. If drops could be

manipulated with <20 V, the whole control system could be much

cheaper.

We’ve mostly been focused (so far) on improving device robustness and

reliability and not so much on these other issues. To reduce the

voltage, you need to either (a) increase the dielectric constant of

your insulating layer and/or (b) reduce the thickness of the

dielectric. Previous students in the Wheeler lab have experimented

with some alternative dielectrics (e.g., barium titanate, silicon

nitride) that have higher a dielectric constant, but with our

equipment, we can only coat devices in small batches (i.e., <4),

whereas using Parylene-C, we are able to coat >100 at a time. Also,

with thinner dielectrics, you can run into problems because it is more

common to get small imperfections. Any failure point in the

dielectric, will cause breakdown and electrolysis of your drop.

I’m planning to setup some kind of a wiki or blog on the DropBot

website so that we can organize some info on potential low-cost device

fabrication strategies, future development ideas, etc. Right now, I’m

mainly focused on writing hardware assembly documentation as well as

my other research papers so that I can eventually graduate :wink:

But it is a near term goal to at least be able to support people who

want to experiment with different device fabrication strategies. I

suspect that there are people out there in the community who can

figure out a great way to make cheap, reliable, and maybe even

low-voltage devices.

-Ryan

On Thu, Sep 12, 2013 at 11:31 AM, Ryan George > > >> ryan...@knights.ucf.edu wrote:

Hello!

I see everything besides the actual chip in the hardware list. Where can

this be purchased or made? I have a synthetic biology club that would be

interested in making a dropbot to use while working on the iGEM

competition

next summer.

I also wanted to know if you were familiar with DMF chips that had cross

referencing capabilities to open up the area you can work with quite a

bit?

http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233

There are also methods to get the voltage down to less than 15 volts:

http://www.sciencedirect.com/science/article/pii/S0038110108001275

I was wondering if you have already considered these technologies in

future

designs?

Thanks,

–Ryan George

You received this message because you are subscribed to the Google

Groups

“dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send

an

email to dropbot-dev...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.

You received this message because you are subscribed to the Google Groups

“dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an

email to dropbot-dev...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.


#7

No problem. Good luck, and let us know how you make out!

···

On Mon, Mar 9, 2015 at 2:01 PM, Csaba Konrád <csaba....@gmail.com> wrote:

Ryan,

Thank you for the prompt reply, your answers are very useful! I think if I
can make a chip my boss will be convinced.

Thanks again!

Best,
Csaba

On Monday, March 9, 2015 at 11:31:54 AM UTC-4, Ryan Fobel wrote:

Hi Csaba,

Thanks for your email! I will try to answer your questions as best I can.

On Sat, Mar 7, 2015 at 6:32 PM, Csaba Konrád <csab...@gmail.com> wrote:
> Hi all,
>
> I just found out about the project and I find it really fascinating!
> I'm in a lab where we do very repetitive experiments on ALS patient
> fibroblasts and I think the DropBot would be a great step forward.
> I've built 3d printers in the past, soldered a little bit and can
> program
> somewhat in python, so I should be able to build it as cheap as possible
> and
> get a lot of functionality out of it.
>
> I want to convince my boss, most of my questions were already answered
> here
> but I still have a few:
>
> Are chips reusable, or you throw them out all the time?

The time your chips last will depend a lot on how good your
fabrication process is and what you're doing with the DMF chips. If
you're moving liquids with a lot of proteins in them, they will tend
to foul over time and drop movement becomes increasingly difficult. If
you are moving protein-free liquids, devices can last a very long
time. Oil-filled devices will also last longer relative to those
operated in air.

There are also some tricks that can partially recover the function of
devices that have been fouled by proteins (e.g., reflowing Teflon by
heating your devices on a hot-plate).

> Does teflon bind RNA/DNA from the droplets? This would be important for
> example for PCR, where very little cross contamination can cause
> problems.

Personally, I don't have a lot of experience in moving DNA/RNA, but
there are certainly many examples in the literature of people doing
PCR. My understanding is that moving DNA/RNA is less challenging as
compared to proteins since DNA/RNA are usually hydrophilic, which
means that they will have less of a tendency to stick to the
hydrophobic Teflon surface.

If you are interested in PCR, I think the bigger challenge is
preventing drop evaporation. Most of the people doing PCR on DMF use
an oil filler.

> You mentioned the chromium plated glass and I saw the jove publication
>
> (http://www.jove.com/video/1603/digital-microfluidics-for-automated-proteomic-processing).
> Do you always make them, or is there a company that sells them? I think
> the
> litography and the spin coating would be doable in our lab.

We used to chromium plate our own glass slides by e-beam, but now we
mainly purchase them from a company called Telic
(http://www.teliccompany.com/). That jove paper is quite old and our
fabrication methods have improved since then; it's better to look at
any of our more recent publications for the latest procedures.

> Also in the jove I don't understand why the protein precicpitate stays
> in
> place while the droplet moves away. I saw that the droplet is moved very
> slow. Is this not a tricky step? Do you have to design it in a way that
> you
> have a lot of precipitate? Did you ever try to precipitate DNA or RNA?

Proteins with hydrophobic domains will stick to the Teflon, which in
most cases is a bad thing because it impedes droplet movement, but in
this case it will help to anchor the precipitated proteins to the
surface. Moving a droplet over the surface produces very low shear
stress, which is probably why the proteins remain in place. I don't
think that this step is particularly tricky. Nobody in our lab has
worked with precipitating DNA/RNA as far as I know...

>
> When doing magnetic separation do you use an inductor to generate a
> field
> controlled by the electronics or put a magnet under the chip?

We use a permanent magnet. See the following paper for details:

Choi, K.; Ng, A.H.C.; Fobel, R.; Chang-Yen, D.A.; Yarnell, L.E.;
Pearson, E.L.; Oleksak, C.M.; Fischer, A.T.; Luoma, R.P.; Robinson,
J.M.; Audet, J.; Wheeler, A.R. "Automated Digital Microfluidic
Platform for Magnetic-Particle-Based Immunoassays with Optimization by
Design of Experiments" Analytical Chemistry 2013, 85, 9638-9646.
(available on our website:
http://microfluidics.utoronto.ca/publications.php)

> You mentioned somewhere that you are working on inkjet printable chips,
> is
> it by AgIC(http://shop.agic.cc/collections/circuit-printer)? Do you
> spincoat
> the paper?

The inkjet printed devices are described in this paper:

Fobel, R.; Kirby, A.E.; Ng, A.H.C.; Farnood, R.R.; Wheeler, A.R.
"Paper Microfluidics Goes Digital" Adv. Mat. 2014, 26, 2838-2843.
(available on our website:
http://microfluidics.utoronto.ca/publications.php)

> Can parts of the chip be heated (like not more than 100C) or do the
> droplets
> evaporate?

The temperature limitation is mainly set by the dielectric. Parylene-C
is rated for a continuous service temperature of 80C and a short term
service temperature of 100C. We exceed this temperature during the
Teflon reflow step of our fabrication process, and it doesn't result
in catastrophic failure, but exceeding 80C for extended periods of
time may damage your device over the long term (Parylene data sheet:
http://vsiparylene.com/pdf/ParyleneProperties2013.pdf).

Evaporation will definitely be a problem if you plan to heat your
drops to anywhere close to 100C. You can ameliorate this problem
through drop replenishment and/or using an oil filler.

Hope that helps,

-Ryan

> If there answers for these in your papers than sorry, don't bother to
> answer
> i still need to go through them.
>
> Thank you and bests,
> Csaba
>
>
> On Thursday, September 12, 2013 at 1:42:13 PM UTC-4, Ryan Fobel wrote:
>>
>> Hi Ryan,
>>
>> Thanks for your email. These questions actually come up a lot, so it's
>> great to get this info out there for everyone. You've nailed a really
>> big problem with DMF as it currently stands. I think that the DropBot
>> system makes it easy for anyone to build a DMF control system, but
>> unless you can get/make devices, it's not incredibly useful. Our lab
>> has published lots of DIY-friendly, rapid prototyping techniques for
>> DMF devices (http://microfluidics.utoronto.ca/publications.php)
>> including micro-contact priting, PCB-based devices, dielectric films,
>> etc., but none of them are perfect, and certainly the devices made
>> using glass-substrate/photolithography have the best performance in my
>> experience.
>>
>> We fabricate our own chips in the cleanroom facility at the University
>> of Toronto. Fabrication details can be found in any the publications
>> on our website (e.g.,
>> http://microfluidics.utoronto.ca/papers/DropBot%20paper.pdf).
>>
>> I am familiar with the cross-referencing scheme (I think this was
>> originally published by CJ Kim's group several years back, before I
>> started). Some of the issues, if I remember correctly, were:
>>
>> 1. electrode cells in the same row/column as the target will be at
>> half potential
>> 2. your software requires logic to avoid conflicts when actuating
>> multiple electrodes in parallel
>> 3. probably requires 2 HV amplifiers (180 degrees out of phase)
>>
>> In any case, it would be possible to build a DropBot system that could
>> support cross-referenced devices without too much trouble. But if you
>> require two amplifiers, any cost savings from limiting the number of
>> switches may be negligible. I do think this is a promising scheme
>> though, especially because it would really allow your device
>> geometries to scale up.
>>
>> I'm also familiar with some of the efforts to make low voltage
>> devices. I think this could be really important for reducing the cost
>> of the control electronics and would also have some other nice
>> benefits (safety, portable, battery-powered systems). Right now the
>> biggest expenses from the electronics side are the high-voltage
>> amplifier and the high-voltage solid state relays. If drops could be
>> manipulated with <20 V, the whole control system could be much
>> cheaper.
>>
>> We've mostly been focused (so far) on improving device robustness and
>> reliability and not so much on these other issues. To reduce the
>> voltage, you need to either (a) increase the dielectric constant of
>> your insulating layer and/or (b) reduce the thickness of the
>> dielectric. Previous students in the Wheeler lab have experimented
>> with some alternative dielectrics (e.g., barium titanate, silicon
>> nitride) that have higher a dielectric constant, but with our
>> equipment, we can only coat devices in small batches (i.e., <4),
>> whereas using Parylene-C, we are able to coat >100 at a time. Also,
>> with thinner dielectrics, you can run into problems because it is more
>> common to get small imperfections. Any failure point in the
>> dielectric, will cause breakdown and electrolysis of your drop.
>>
>> I'm planning to setup some kind of a wiki or blog on the DropBot
>> website so that we can organize some info on potential low-cost device
>> fabrication strategies, future development ideas, etc. Right now, I'm
>> mainly focused on writing hardware assembly documentation as well as
>> my other research papers so that I can eventually graduate :wink:
>>
>> But it is a near term goal to at least be able to support people who
>> want to experiment with different device fabrication strategies. I
>> suspect that there are people out there in the community who can
>> figure out a great way to make cheap, reliable, and maybe even
>> low-voltage devices.
>>
>> -Ryan
>>
>> On Thu, Sep 12, 2013 at 11:31 AM, Ryan George >> >> <ryan...@knights.ucf.edu> wrote:
>> > Hello!
>> >
>> > I see everything besides the actual chip in the hardware list. Where
>> > can
>> > this be purchased or made? I have a synthetic biology club that would
>> > be
>> > interested in making a dropbot to use while working on the iGEM
>> > competition
>> > next summer.
>> >
>> > I also wanted to know if you were familiar with DMF chips that had
>> > cross
>> > referencing capabilities to open up the area you can work with quite
>> > a
>> > bit?
>> >
>> >
>> > http://biomedicaloptics.spiedigitallibrary.org/article.aspx?articleid=1306233
>> >
>> > There are also methods to get the voltage down to less than 15 volts:
>> > http://www.sciencedirect.com/science/article/pii/S0038110108001275
>> >
>> > I was wondering if you have already considered these technologies in
>> > future
>> > designs?
>> >
>> > Thanks,
>> > --Ryan George
>> >
>> > --
>> > You received this message because you are subscribed to the Google
>> > Groups
>> > "dropbot-dev" group.
>> > To unsubscribe from this group and stop receiving emails from it,
>> > send
>> > an
>> > email to dropbot...@googlegroups.com.
>> > For more options, visit https://groups.google.com/groups/opt_out.
>
> --
> You received this message because you are subscribed to the Google
> Groups
> "dropbot-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an
> email to dropbot...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups
"dropbot-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to dropbot-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


#8

Hello DMF community,

I’m happy to see that more and more people try out this technology. And I really appreciate that we have a plattform here to discuss our problems (and snitch other inventions, hehe :wink:

Some background:
We buy chromium coated glass wafer, which were wet-etched to get the electrode structures. We have a cleanroom and a lot experience with lithography (LiGA, UV, eBeam…), so this part is quite easy (CAD design, mask, resist…etching). The last part is the Teflon, all you need is a spincoater, a hotplate and this very expensive Teflon-Granulate or solution. Quite easy, too.
But, in my (short) experience the dielectrics are the biggest problem while producing a DMF chip by yourself.

What should I use to get perfect dielectric layer…
Aaron told me after a conference: “The key is no poinholes - the thicker the dielectric is, the better your chances of this. Good Luck”

At the moment I’m using SU-8 (3-10 microns thickess). All my devices are working quite nice, but I got some little bubbles or at least some uneven spots all the time. I guess there are 2 or 3 of those spots per device. No pinholes, but a droplet likes to stick there a little bit. We could apply parylene-c, too. But our machine is made for 10-300 nm layer. Instead I could also buy some silicon-nitride layers in industrial quality, which I used in my previous work - quite expensive, too. But here we also talking about only 300-1000nm and a external(!) production step.

I can accept the disadvantages of parylen-c (and also SU-8) when it becomes to heating experiments. If we need something special, then we pay for it, as soon as we need it.

I want to keep the device production in our house, or at least on the campus.

My questions:
Do you have an idea, which stuff can be used as a good dielectric layer and can be found in a well-equipped cleanroom or chemicals-cupboard?
Which parylen-c coating mashine are you using? I thought, I read it before in a paper, but cannot find it atm.

Yes, I allready did the saran wrap dielectrics :wink:

Thanks for reading and any helpfull thoughts or discussions!

Sebastian


#9

Hi both,

This just made me realize you vapor deposit the parylene. Shoot! I also found it doesn’t dissolve in anything, so i guess that answers why. Is there a way to get away with spin coating only? I looked up SU-8 it seems it is an option. Sebastian, do you have a detailed protocol? Do you guys know of other alternatives?

Best,

Csaba

···

On Tuesday, March 10, 2015 at 12:51:33 PM UTC-4, Sebastian vdE wrote:

Hello DMF community,

I’m happy to see that more and more people try out this technology. And I really appreciate that we have a plattform here to discuss our problems (and snitch other inventions, hehe :wink:

Some background:
We buy chromium coated glass wafer, which were wet-etched to get the electrode structures. We have a cleanroom and a lot experience with lithography (LiGA, UV, eBeam…), so this part is quite easy (CAD design, mask, resist…etching). The last part is the Teflon, all you need is a spincoater, a hotplate and this very expensive Teflon-Granulate or solution. Quite easy, too.
But, in my (short) experience the dielectrics are the biggest problem while producing a DMF chip by yourself.

What should I use to get perfect dielectric layer…
Aaron told me after a conference: “The key is no poinholes - the thicker the dielectric is, the better your chances of this. Good Luck”

At the moment I’m using SU-8 (3-10 microns thickess). All my devices are working quite nice, but I got some little bubbles or at least some uneven spots all the time. I guess there are 2 or 3 of those spots per device. No pinholes, but a droplet likes to stick there a little bit. We could apply parylene-c, too. But our machine is made for 10-300 nm layer. Instead I could also buy some silicon-nitride layers in industrial quality, which I used in my previous work - quite expensive, too. But here we also talking about only 300-1000nm and a external(!) production step.

I can accept the disadvantages of parylen-c (and also SU-8) when it becomes to heating experiments. If we need something special, then we pay for it, as soon as we need it.

I want to keep the device production in our house, or at least on the campus.

My questions:
Do you have an idea, which stuff can be used as a good dielectric layer and can be found in a well-equipped cleanroom or chemicals-cupboard?
Which parylen-c coating mashine are you using? I thought, I read it before in a paper, but cannot find it atm.

Yes, I allready did the saran wrap dielectrics :wink:

Thanks for reading and any helpfull thoughts or discussions!

Sebastian


#10

Hi Csaba,

I have a quite simple protocol for my SU-8 (to be honest, it is “mr-L”, a derivate of SU-8, but kind of the same): For a 4,2 µm layer we spincoat at 3000rpm with a ramp of 1500 for 60 seconds. After that heat it at 95°C for 10 minutes.

You can modify the thickness by just lowering the RPM to 1500, then you should get a layer between 6-8µm for example.

In the next days I want to check, how to get rid of the bubbles (vacuum maybe). They CANNOT be seen after spincoating, but they appear while heating… as far as I know.

Sebastian


#11

Hi Sebastian,

Thank you! Where do you buy it? I found its made by Micro Resist, but cant find it among their products.

What is the solvent? Maybe your bubbles are solvent vapor trapped in the film. Did you ever try heating the plate slowly or heating after letting it dry at room temp for a while (altough I did find that other people also bake it at 90 C)? I think the vacuum would probably favor bubble formation.

It could also be gas dissolved in the solvent, or if its a mixture of solvents they could be droplets of the remain of the higher boiling point one. Just ideas

Best,

Csaba

···

On Thursday, March 12, 2015 at 5:32:18 AM UTC-4, Sebastian vdE wrote:

Hi Csaba,

I have a quite simple protocol for my SU-8 (to be honest, it is “mr-L”, a derivate of SU-8, but kind of the same): For a 4,2 µm layer we spincoat at 3000rpm with a ramp of 1500 for 60 seconds. After that heat it at 95°C for 10 minutes.

You can modify the thickness by just lowering the RPM to 1500, then you should get a layer between 6-8µm for example.

In the next days I want to check, how to get rid of the bubbles (vacuum maybe). They CANNOT be seen after spincoating, but they appear while heating… as far as I know.

Sebastian


#12

oh, and a stupid question: how do you measure layer thickness? by absorbance?

···

On Thursday, March 12, 2015 at 10:48:24 AM UTC-4, Csaba Konrád wrote:

Hi Sebastian,

Thank you! Where do you buy it? I found its made by Micro Resist, but cant find it among their products.

What is the solvent? Maybe your bubbles are solvent vapor trapped in the film. Did you ever try heating the plate slowly or heating after letting it dry at room temp for a while (altough I did find that other people also bake it at 90 C)? I think the vacuum would probably favor bubble formation.

It could also be gas dissolved in the solvent, or if its a mixture of solvents they could be droplets of the remain of the higher boiling point one. Just ideas

Best,

Csaba

On Thursday, March 12, 2015 at 5:32:18 AM UTC-4, Sebastian vdE wrote:

Hi Csaba,

I have a quite simple protocol for my SU-8 (to be honest, it is “mr-L”, a derivate of SU-8, but kind of the same): For a 4,2 µm layer we spincoat at 3000rpm with a ramp of 1500 for 60 seconds. After that heat it at 95°C for 10 minutes.

You can modify the thickness by just lowering the RPM to 1500, then you should get a layer between 6-8µm for example.

In the next days I want to check, how to get rid of the bubbles (vacuum maybe). They CANNOT be seen after spincoating, but they appear while heating… as far as I know.

Sebastian


#13

Q: Where do you buy it?

···

A: I guess micro resist is our source of SU-8 and mr-L, but I don’t buy it, i just “misuse” the resist as a dielectric layer.

Q: Get rid of the bubbles?
A: A lot good point, the pre-heating and the slow-heating might be worth a try! The vacuum is my next step anyway, I will try to use our vacuum oven: heating and vacuum at once. I copy the protocol from our cleanroom team, they are working with SU-8 and mr-L since years, I’m quite new with it. So their protocol should be kind of optimized.

Q: Layer thickness measurement?
A: The numbers I gave you are from our protocol, that means, I didn’t measure them myself. At the moment the chips are stable (dielectrics are thick enough) and the bubbles are my problem. But if I want to measure, I can use the ellipsometer, the white light interferometer or just scratch it and measure the uneveness. Depends on the expected thickness and if I want to destroy the probe or not.


#14

Hi guys,

We deposit parylene-C using vapor deposition
(http://scscoatings.com/equipment/parylene_coating_equipment/scs_labcoater2.aspx).

Another interesting liquid-processable dielectric that we have been
playing with is Cyanoethyl Pullulan (CEP). It's cheap and has a high
dielectric constant (16-20). There have been a few recent publications
mentioning it:

Yang et al., “A Robust and Inexpensive Composite Insulation Layer for
Digital Microfluidic Devices.” Sensors and Actuators A: Physical 219
(2014): 6–12.
http://dx.doi.org/10.1016/j.sna.2014.06.004

Chen et al., “Study of Cyanoethyl Pullulan as Insulator for
Electrowetting.” Sensors and Actuators B: Chemical 199 (2014): 183–89.
http://dx.doi.org/10.1016/j.snb.2014.03.112

-Ryan

···

On Thu, Mar 12, 2015 at 11:03 AM, Sebastian vdE <sebasti...@kit.edu> wrote:

Q: Where do you buy it?
A: I guess micro resist is our source of SU-8 and mr-L, but I don't buy it,
i just "misuse" the resist as a dielectric layer.

Q: Get rid of the bubbles?
A: A lot good point, the pre-heating and the slow-heating might be worth a
try! The vacuum is my next step anyway, I will try to use our vacuum oven:
heating and vacuum at once. I copy the protocol from our cleanroom team,
they are working with SU-8 and mr-L since years, I'm quite new with it. So
their protocol should be kind of optimized.

Q: Layer thickness measurement?
A: The numbers I gave you are from our protocol, that means, I didn't
measure them myself. At the moment the chips are stable (dielectrics are
thick enough) and the bubbles are my problem. But if I want to measure, I
can use the ellipsometer, the white light interferometer or just scratch it
and measure the uneveness. Depends on the expected thickness and if I want
to destroy the probe or not.

--
You received this message because you are subscribed to the Google Groups
"dropbot-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to dropbot-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


#15

Thanks!

···

On Thursday, March 12, 2015 at 11:03:33 AM UTC-4, Sebastian vdE wrote:

Q: Where do you buy it?
A: I guess micro resist is our source of SU-8 and mr-L, but I don’t buy it, i just “misuse” the resist as a dielectric layer.

Q: Get rid of the bubbles?
A: A lot good point, the pre-heating and the slow-heating might be worth a try! The vacuum is my next step anyway, I will try to use our vacuum oven: heating and vacuum at once. I copy the protocol from our cleanroom team, they are working with SU-8 and mr-L since years, I’m quite new with it. So their protocol should be kind of optimized.

Q: Layer thickness measurement?
A: The numbers I gave you are from our protocol, that means, I didn’t measure them myself. At the moment the chips are stable (dielectrics are thick enough) and the bubbles are my problem. But if I want to measure, I can use the ellipsometer, the white light interferometer or just scratch it and measure the uneveness. Depends on the expected thickness and if I want to destroy the probe or not.


#16

Hi,

I asked for the CEP at Biddle sawyer, its 100$ for 100 grams, not so bad. The dimethyl formamide and the plasticoat 70 are also cheap (https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=) My only problem is that Dupont does not sell in small quantities.
If I get some CEP would any of you guys be interested in trading some for a small amount of teflon AF powder or solution with which i can test out if i can make my own devices?

Best,

Csaba

···

On Thursday, March 12, 2015 at 12:08:07 PM UTC-4, Csaba Konrád wrote:

Thanks!

On Thursday, March 12, 2015 at 11:03:33 AM UTC-4, Sebastian vdE wrote:

Q: Where do you buy it?
A: I guess micro resist is our source of SU-8 and mr-L, but I don’t buy it, i just “misuse” the resist as a dielectric layer.

Q: Get rid of the bubbles?
A: A lot good point, the pre-heating and the slow-heating might be worth a try! The vacuum is my next step anyway, I will try to use our vacuum oven: heating and vacuum at once. I copy the protocol from our cleanroom team, they are working with SU-8 and mr-L since years, I’m quite new with it. So their protocol should be kind of optimized.

Q: Layer thickness measurement?
A: The numbers I gave you are from our protocol, that means, I didn’t measure them myself. At the moment the chips are stable (dielectrics are thick enough) and the bubbles are my problem. But if I want to measure, I can use the ellipsometer, the white light interferometer or just scratch it and measure the uneveness. Depends on the expected thickness and if I want to destroy the probe or not.


#17

Sigma-Aldrich sells Teflon AF in smaller quantities:

PTFE AF 1600 1G = $372 USD

http://www.sigmaaldrich.com/catalog/product/aldrich/469610?lang=en&region=US

as well as

PTFE AF 2400 1G = $387.59.

http://www.sigmaaldrich.com/catalog/product/aldrich/469629?lang=en&region=US

–Michael

···

On Thursday, March 12, 2015 at 12:52:46 PM UTC-7, Csaba Konrád wrote:

Hi,

I asked for the CEP at Biddle sawyer, its 100$ for 100 grams, not so bad. The dimethyl formamide and the plasticoat 70 are also cheap (https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=) My only problem is that Dupont does not sell in small quantities.
If I get some CEP would any of you guys be interested in trading some for a small amount of teflon AF powder or solution with which i can test out if i can make my own devices?

Best,

Csaba

On Thursday, March 12, 2015 at 12:08:07 PM UTC-4, Csaba Konrád wrote:

Thanks!

On Thursday, March 12, 2015 at 11:03:33 AM UTC-4, Sebastian vdE wrote:

Q: Where do you buy it?
A: I guess micro resist is our source of SU-8 and mr-L, but I don’t buy it, i just “misuse” the resist as a dielectric layer.

Q: Get rid of the bubbles?
A: A lot good point, the pre-heating and the slow-heating might be worth a try! The vacuum is my next step anyway, I will try to use our vacuum oven: heating and vacuum at once. I copy the protocol from our cleanroom team, they are working with SU-8 and mr-L since years, I’m quite new with it. So their protocol should be kind of optimized.

Q: Layer thickness measurement?
A: The numbers I gave you are from our protocol, that means, I didn’t measure them myself. At the moment the chips are stable (dielectrics are thick enough) and the bubbles are my problem. But if I want to measure, I can use the ellipsometer, the white light interferometer or just scratch it and measure the uneveness. Depends on the expected thickness and if I want to destroy the probe or not.


#18

Hi,

Cytonix Fluoropel solutions are a cheaper option for Teflon AF. See
http://www.cytonix.com/fluorosolpoly.html

We got 200 g of Fluor0pel 1604V (4% solids in solution) for 1000 $.

Br,

Markus

12.3.2015, 22:46, Michael Connolly kirjoitti:

···

Sigma-Aldrich sells Teflon AF in smaller quantities:

PTFE AF 1600 1G = $372 USD

http://www.sigmaaldrich.com/catalog/product/aldrich/469610?lang=en&region=US

as well as

PTFE AF 2400 1G = $387.59.

http://www.sigmaaldrich.com/catalog/product/aldrich/469629?lang=en&region=US

--Michael

On Thursday, March 12, 2015 at 12:52:46 PM UTC-7, Csaba Konrád wrote:

    Hi,

    I asked for the CEP at Biddle sawyer, its 100$ for 100 grams, not
    so bad. The dimethyl formamide and the plasticoat 70 are also
    cheap
    (https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=
    <https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=>)
    My only problem is that Dupont does not sell in small quantities.
    If I get some CEP would any of you guys be interested in trading
    some for a small amount of teflon AF powder or solution with which
    i can test out if i can make my own devices?

    Best,
    Csaba

    On Thursday, March 12, 2015 at 12:08:07 PM UTC-4, Csaba Konrád wrote:

        Thanks!

        On Thursday, March 12, 2015 at 11:03:33 AM UTC-4, Sebastian > vdE wrote:

            Q: Where do you buy it?
            A: I guess micro resist is our source of SU-8 and mr-L,
            but I don't buy it, i just "misuse" the resist as a
            dielectric layer.

            Q: Get rid of the bubbles?
            A: A lot good point, the pre-heating and the slow-heating
            might be worth a try! The vacuum is my next step anyway, I
            will try to use our vacuum oven: heating and vacuum at
            once. I copy the protocol from our cleanroom team, they
            are working with SU-8 and mr-L since years, I'm quite new
            with it. So their protocol should be kind of optimized.

            Q: Layer thickness measurement?
            A: The numbers I gave you are from our protocol, that
            means, I didn't measure them myself. At the moment the
            chips are stable (dielectrics are thick enough) and the
            bubbles are my problem. But if I want to measure, I can
            use the ellipsometer, the white light interferometer or
            just scratch it and measure the uneveness. Depends on the
            expected thickness and if I want to destroy the probe or not.

--
You received this message because you are subscribed to the Google Groups "dropbot-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dropbot-dev...@googlegroups.com <mailto:dropbot-dev...@googlegroups.com>.
For more options, visit https://groups.google.com/d/optout.


#19

A number of people here have also started using Fluoropel instead of Teflon-AF. I haven’t done a thorough characterization yet, but there doesn’t seem to be much of a difference in performance.

-Ryan

···

On Thu, Mar 12, 2015 at 5:05 PM, Markus Haapala markus...@gmail.com wrote:

Hi,

Cytonix Fluoropel solutions are a cheaper option for Teflon AF. See

http://www.cytonix.com/fluorosolpoly.html

We got 200 g of Fluor0pel 1604V (4% solids in solution) for 1000 $.

Br,

Markus

12.3.2015, 22:46, Michael Connolly kirjoitti:

Sigma-Aldrich sells Teflon AF in smaller quantities:

PTFE AF 1600 1G = $372 USD

http://www.sigmaaldrich.com/catalog/product/aldrich/469610?lang=en&region=US

as well as

PTFE AF 2400 1G = $387.59.

http://www.sigmaaldrich.com/catalog/product/aldrich/469629?lang=en&region=US

–Michael

On Thursday, March 12, 2015 at 12:52:46 PM UTC-7, Csaba Konrád wrote:

Hi,



I asked for the CEP at Biddle sawyer, its 100$ for 100 grams, not

so bad. The dimethyl formamide and the plasticoat 70 are also

cheap

([https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=](https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=)

<[https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=](https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=)>)


My only problem is that Dupont does not sell in small quantities.

If I get some  CEP would any of you guys be interested in trading

some for a small amount of teflon AF powder or solution with which

i can test out if i can make my own devices?



Best,

Csaba



On Thursday, March 12, 2015 at 12:08:07 PM UTC-4, Csaba Konrád wrote:



    Thanks!



    On Thursday, March 12, 2015 at 11:03:33 AM UTC-4, Sebastian

    vdE wrote:



        Q: Where do you buy it?

        A: I guess micro resist is our source of SU-8 and mr-L,

        but I don't buy it, i just "misuse" the resist as a

        dielectric layer.



        Q: Get rid of the bubbles?

        A: A lot good point, the pre-heating and the slow-heating

        might be worth a try! The vacuum is my next step anyway, I

        will try to use our vacuum oven: heating and vacuum at

        once. I copy the protocol from our cleanroom team, they

        are working with SU-8 and mr-L since years, I'm quite new

        with it. So their protocol should be kind of optimized.



        Q: Layer thickness measurement?

        A: The numbers I gave you are from our protocol, that

        means, I didn't measure them myself. At the moment the

        chips are stable (dielectrics are thick enough) and the

        bubbles are my problem. But if I want to measure, I can

        use the ellipsometer, the white light interferometer or

        just scratch it and measure the uneveness. Depends on the

        expected thickness and if I want to destroy the probe or not.

You received this message because you are subscribed to the Google Groups “dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an email to dropbot-dev+unsubscribe@googlegroups.com mailto:dropbot-dev+unsub...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

You received this message because you are subscribed to the Google Groups “dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an email to dropbot-dev+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.


#20

Thanks!

···

On Thursday, March 12, 2015 at 5:09:28 PM UTC-4, Ryan Fobel wrote:

A number of people here have also started using Fluoropel instead of Teflon-AF. I haven’t done a thorough characterization yet, but there doesn’t seem to be much of a difference in performance.

-Ryan

On Thu, Mar 12, 2015 at 5:05 PM, Markus Haapala mark...@gmail.com wrote:

Hi,

Cytonix Fluoropel solutions are a cheaper option for Teflon AF. See

http://www.cytonix.com/fluorosolpoly.html

We got 200 g of Fluor0pel 1604V (4% solids in solution) for 1000 $.

Br,

Markus

12.3.2015, 22:46, Michael Connolly kirjoitti:

Sigma-Aldrich sells Teflon AF in smaller quantities:

PTFE AF 1600 1G = $372 USD

http://www.sigmaaldrich.com/catalog/product/aldrich/469610?lang=en&region=US

as well as

PTFE AF 2400 1G = $387.59.

http://www.sigmaaldrich.com/catalog/product/aldrich/469629?lang=en&region=US

–Michael

On Thursday, March 12, 2015 at 12:52:46 PM UTC-7, Csaba Konrád wrote:

Hi,



I asked for the CEP at Biddle sawyer, its 100$ for 100 grams, not

so bad. The dimethyl formamide and the plasticoat 70 are also

cheap

([https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=](https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=)

<[https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=](https://www.whitworths.com.au/main_itemdetail.asp?cat=123&item=33738&intAbsolutePage=)>)


My only problem is that Dupont does not sell in small quantities.

If I get some  CEP would any of you guys be interested in trading

some for a small amount of teflon AF powder or solution with which

i can test out if i can make my own devices?



Best,

Csaba



On Thursday, March 12, 2015 at 12:08:07 PM UTC-4, Csaba Konrád wrote:



    Thanks!



    On Thursday, March 12, 2015 at 11:03:33 AM UTC-4, Sebastian > > >  > > >         vdE wrote:



        Q: Where do you buy it?

        A: I guess micro resist is our source of SU-8 and mr-L,

        but I don't buy it, i just "misuse" the resist as a

        dielectric layer.



        Q: Get rid of the bubbles?

        A: A lot good point, the pre-heating and the slow-heating

        might be worth a try! The vacuum is my next step anyway, I

        will try to use our vacuum oven: heating and vacuum at

        once. I copy the protocol from our cleanroom team, they

        are working with SU-8 and mr-L since years, I'm quite new

        with it. So their protocol should be kind of optimized.



        Q: Layer thickness measurement?

        A: The numbers I gave you are from our protocol, that

        means, I didn't measure them myself. At the moment the

        chips are stable (dielectrics are thick enough) and the

        bubbles are my problem. But if I want to measure, I can

        use the ellipsometer, the white light interferometer or

        just scratch it and measure the uneveness. Depends on the

        expected thickness and if I want to destroy the probe or not.

You received this message because you are subscribed to the Google Groups “dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an email to dropbot-dev...@googlegroups.com mailto:dropbot-dev+unsubscribe...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

You received this message because you are subscribed to the Google Groups “dropbot-dev” group.

To unsubscribe from this group and stop receiving emails from it, send an email to dropbot-dev...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.