Printing conductive inks directly as circuit boards using inkjet printers

[JonathanCline]

For the past week I have been reading up on this. Instead of creating
circuit boards using photolithography and using acids to etch away the
copper, the idea is to start with a non-conductive substrate and print
conductive traces directly on the substrate.

In summary,

- inkjets are versatile for many liquids (either piezo or thermal
heads)

- conductive nanoparticles (silver or copper or other) pass through
the heads as easily as ink

- high DPI printers can achieve good traces as low as 50um wide (the
densest surface mount currently uses 150um, standard printed circuits
use around 1.27mm/0.05" traces)

- circuits can be printed on a variety of substrates: FR4 (standard
circuit board), glass, different poly plastics, thin films, natural
fabrics, plastic fabrics, ...

- commercial "flat bed" inkjet printers are pricey (>$30,000) while
consumer inkjet "paper" printers are ridiculously cheap ($100)

- conductivity is mostly comparable to standard copper circuit boards

- silver solution works better than copper solution. Bake the printed
board at ~150C for 30 mins after printing.

- some of the inks are conductive polymers.


Re-purposing the print cartridges is easy, modifying a consumer inkjet
printer to allow rigid substrate is easy, some inks are commercially
available. It seems this is already very doable.

Anyone already doing this?


References:

Inkjet Printable Polyaniline Nanoformulations
Orawan Ngamna,† Aoife Morrin,‡ Anthony J. Killard,‡ Simon E.
Moulton,†
Malcolm R. Smyth,‡ and Gordon G. Wallace*,†


Conductivity of silver paste prepared from nanoparticles
Keunju Park, Dongseok Seo, Jongkook Lee∗


Limits to feature size and resolution in inkjet printing
J.Stringer∗, B.Derby


The characterization of electrically conductive silver ink patterns
on flexible substrates
S. Merilampi *, T. Laine-Ma, P. Ruuskanen


## Jonathan Cline
## jcl...@ieee.org
## Mobile: +1-805-617-0223
########################

[ben lipkowitz]

flexible kapton circuit tape and $1/lb paper phenolic board stock from
ebay cover 99% of projects I would want to build. is there something
special about directly printing the traces instead of the resist that is
somehow better?

in some high power electronics (VSR switch in a power drill for example) a
silver paste is screen printed onto alumina ceramic substrate, then baked
at high temperatures until the silver particles sinter together. This is
expensive both because of the price of the silver and the substrate, so
it's not used often.

there was also some craze a while ago on the 'net about doing direct
printing of etch resist with yellow "mispro archival ink", which
apparently has a polymer pigment that sticks nicely when heated to
melting. also I've heard you can use floor polish as etch resist and load
it directly into the ink cartridge.

were you thinking about using one of those CD printers or running some
kind of flexible film through a normal printer or what?

[David Treadwell]

This patent does a great job of describing photopolymerizable ink jet
inks.

When I get a place to do it, I'd like to formulate an inkjet 3D
printing ink along these lines.

--David

On Jun 23, 2009, at 8:50 PM, ben lipkowitz wrote:

[David Treadwell]

One of my current (actually, my ONLY current) consulting project is on
the formulation of silver-silicone composites as flexible conductors.
There are many ways of doing this that don't require pure silver
particles and don't require a post-printing sintering step.

I'll be happy to help on formulation, so long as it doesn't violate my
non-disclosure agreement. Considering how long this technology has
been around, there are plenty of ways to to it.

--David



On Jun 23, 2009, at 8:50 PM, ben lipkowitz wrote:

>

[JonathanCline]

On Jun 24, 3:57 pm, David Treadwell <i.failed.turing.t...@gmail.com>
wrote:

> One of my current (actually, my ONLY current) consulting project is on  
> the formulation of silver-silicone composites as flexible conductors.  
> There are many ways of doing this that don't require pure silver  
> particles and don't require a post-printing sintering step.
>
> I'll be happy to help on formulation, so long as it doesn't violate my  
> non-disclosure agreement. Considering how long this technology has  
> been around, there are plenty of ways to to it.
>
> --David

Great!
I have read a couple methods in research though I always doubt
research (as published) is "true". I have found 3 vendors so far of
silver inks "made for inkjet" (nanoparticle or organic), the question
is, which ones work in typical desktop inkjets (and thermal vs
piezo). i.e. I just got an EPSON with CD tray, so presumably I could
put a glass slide on the tray instead of a CD & get a small circuit
printed in silver. I don't care much about the substrate (rigid or
flexi or anthing): I just want a 50um trace with 50um gap, and I want
it cheap & easy to fab.
How about we discuss this, I am at UT in MBB regularly. Or give me a
call.

"flexible kapton circuit tape and $1/lb paper phenolic board stock
from
ebay cover 99% of projects I would want to build. is there something
special about directly printing the traces instead of the resist that
is
somehow better? "

1. resolution -> 50um "relatively easily"
2. speed (no etch time)
3. cost (no buying copper clad then wasting 90% of the copper in acid)
4. more environmentally friendly - no acid, less waste
5. no resist "transfer" problems which is a lossy/noisy method

It seems RF circuits need tuning to the substrate and process, however
once this is done it sounds (from the papers anyway which may be
exaggerated) to be fairly reliable/reproducible. I'm not making RF,
I'm making surface mount with MEMS, so I don't care about that tuning
issue.

## Jonathan Cline
## jcl...@ieee.org
## Mobile: +1-805-617-0223
########################

>

[Bryan Bishop]

On Thu, Jun 25, 2009 at 3:33 PM, JonathanCline wrote:
> How about we discuss this, I am at UT in MBB regularly.  Or give me a
> call.

What the hell man. I'm across the street from MBB on a daily basis.
How about we do lunch sometime?

- Bryan
http://heybryan.org/
1 512 203 0507

[Gene Hacker]

Jonathan, do you have any data on conductivities or the cost of this
stuff? You can also convert a non-flatbed printer into a flatbed
printer, in fact I have some instructions on how to convert an Epson
printer into a very crude(and I emphasize the very) flatbed printer
for printing onto t-shirts. It's a good thing that picked an Epson
printer because Epson printers use a non-replaceable piezoelectric
print head. Being non-replaceable, it's designed to be cleaned, so you
could probably put anything you want through it without worrying too
much about breaking it.

Conductive ink also lets you do some pretty cool things. For example,
using a 3d printer capable of printing plastic, conductive ink, and
integrating electrical components you could print off a working robot
(by integrating motors), a new remote control for your TV when it
breaks, or even some weird electronic gadget. You could also make
circuits in 3d configurations that would be impossible to make any
other way(you really should take a look at this 3d printed rolling
ball thing they have in ETC to get an idea of what I mean by
impossible). If the plastic and ink was recyclable and you used cheap
microcontrollers to do everything(like they do in fablabs) you could
do perfect recycling. Bored with your army of swarm robots?
Deconstruct them and make them into a glowing John Conway's game of
life coffee table! Or if you make a mistake on your design,
deconstruct and recompile it.

Also, I'm on the UT campus too, would love to meet up with you and
Bryan sometime.