Dear all MEMS-talkers,

I’m working with the production and control of droplets at a t-junction
geometry. I’m having problems with the channel walls acting
hydrophilic. The water, which is supposed to break off as individual
droplets, sticks to the wall and produces large plugs far downstream. I
have been thermally bonding two pieces of PMMA to generate my
microfluidic chip. The procedure for thermal bonding:

1.The structure was milled in PMMA.
2.Both the structure and a PMMA lid were annealed in a oven at about 80
degrees Celsius for 90 min. This was done to avoid stress cracks in the
structure.
3.In order to obtain a tight bond, the surfaces were first cleaned with
ethanol and then they were immersed in ethanol for 10 min.
4.The ethanol was wiped off first with a lint-free tissue and then dried
by Nitrogen gas.
5.The parts were then mounted in a press and placed in an oven for 90
min at 80 degrees Celsius.
6.At last the bonded structure was left to cool down.

The bonding process was successfully and there were no cracks in the
structure.

I’m using Decane and water respectively as the continuous (2mm*300µm)
and dispersed (250*250µm) phase flow. PMMA is normally hydrophobic, but
I’m afraid the ethanol treatment have chemically bonded ethanol to the
PMMA material and then given hydrophilic properties.
I have cleaned the channels with the solvent Toluene to avoid any rests
of ethanol. This didn’t give any improvement.

I’m wondering if there is anything which can be infused to the channels
to prevent it from acting hydrophilic?  (I do not want to use coating,
for example Sigmacote, thus it can reduce the channels.)

Cheers
Ingvild Aker