RELIABLE BATCH MANUFACTURING OF MINIATURIZED VERTICAL VIAS IN SOFT POLYMER REPLICA MOLDING

(1)

http://www.diva-portal.org

This is the published version of a paper presented at 11th International Conference on Miniaturized Systems for Chemistry and Life Sciences (microTAS 2007), Paris, France, 7-11 Oct, 2007.

Citation for the original published paper:

Carlborg, C., Haraldsson, T., Stemme, G., Wijngaart, W. (2007)

RELIABLE BATCH MANUFACTURING OF MINIATURIZED VERTICAL VIAS IN SOFT POLYMER REPLICA MOLDING.

In: 11th International Conference on Miniaturized Systems for Chemistry and Life Sciences (microTAS 2007) (pp. 527-529).

N.B. When citing this work, cite the original published paper.

Permanent link to this version:

http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-49442

(2)

Eleventh International Conference on Miniaturized Systems for Chemistry and Life Sciences 7 – 11 October 2007, Paris, FRANCE

RELIABLE BATCH MANUFACTURING OF

MINIATURIZED VERTICAL VIAS IN SOFT POLYMER REPLICA MOLDING

Carl Fredrik Carlborg, Tommy Haraldsson, Göran Stemme and Wouter van der Wijngaart

Microsystem Technology Lab, KTH - Royal Institute of Technology

Stockholm, SWEDEN (e-mail: fredrik.carlborg@ee.kth.se) ABSTRACT

We introduce and have successfully tested an uncomplicated polydimethylsiloxane (PDMS) compatible method for batch manufacturing vertical microfluidic interconnects via a surface inhibition of cationic photopolymerization. The yield of the maskless method is 100% and it also enhances bond strength with subsequently laminated polymer layers.

Keywords: polymer replica molding, vertical vias, uv-curing, PDMS, microfluidics 1. INTRODUCTION

Previously shown 3D vertical microfluidic interconnect methods in soft polymer replica molding [1,2] suffer from residual membranes in the vias caused by difficulty in squeezing out the prepolymer at the mold interface [3] (Fig. 1 A).

2. PRINCIPLE

Our novel method (Fig. 1 B) provides reliable via formation with perfect yield by locally inhibiting polymerization of a thin layer of the prepolymer. We use an epoxyfunctionalized PDMS prepolymer that UV photopolymerizes rapidly by cationic reactive centers with aryliodonium salt as photoinitiator. The prepolymer is brought in contact with inhibitor locally at the top surface. The strong acid species formed by the aryliodonium salt upon UV irradiation react readily with the inhibitor, consisting of tertiary amines, to form ammonium ions, which are too stable to react further with epoxy groups. This chemistry provides a convenient way to inhibit polymerization at the vias.

3. EXPERIMENTS

The technology was demonstrated and evaluated successfully by fabricating a reliably perforated 50 ȝm thin membrane in PDMS-like polymer from Epoxy Technolgies (OG133) which already contains an amount of cationic photoinitiator. Additional aryliodonium cationic photoinitator was added to fine tune the polymerization. The process was also successfully tested with epoxy terminated PDMS, mixed with aryliodonium photoinitator (Gelest, Inc). A mold of densely packed pillars, 50 ȝm high and 40 ȝm in diameter was dry etched in silicon and passivated with fluoropolymer to facilitate demolding (Fig 2 a). The prepolymer mixture is then cast against the mold (b). A glass cover with spun-on inhibitor layer is then pressed onto the mold (c). UV irradiation forms acid species throughout the monomer solution as a part of the polymer curing process. Near the glass surface these are inactivated by the tertiary amine groups whereas deeper in the bulk polymerization is unaffected (d).

(3)

528

R O⁺ H

I^- N

R

CH3 CH3 O O O CH3CH₃CH3

R O I^- NH⁺ R

CH₃ CH₃ O O O CH₃CH3CH3

Glass cover (c)

(a)

(b) Master (d)

residual membrane

~100 nm thick

polymer

Glass cover

A B

Figure 1. A) Reliability problem of previous methods for creating 3D vertical interconnects. The polymer is clamped between the master and a top layer. A residual membrane is left in the vias opening and blocks the interconnects [3]. B) During the cationic photopolymerization process, the epoxy terminated monomer forms a reactive cationic intermediary (a) with a stable counterion (b). In the presence of tertiary amines (c) the reactive centers are deactivated by a proton transfer to the amine, leaving the epoxy unreacted (d).

glass cover inhibitor

UV curing

slide off glass cover a)

b)

c)

d)

e)

peel off membrane f)

Master

polymer

no residual membrane

Figure 2. Fabrication procedure for a reliably perforated thin polymer membrane via surface inhibition of cationic photopolymerization. A master of pillars is fabricated in silicon or SU8 (a), the UV-curable PDMS is cast against the master (b) and a glass slide treated with tertiary amine groups is pressed onto the master (c) and exposed to UV light (d). The glass slide is removed (e) and after washing away the thin unpolymerized layer the membrane can be peeled off (f). SEM photos of the master and membrane are shown to the right.

(4)

529 In this manner a thin layer of unpolymerized monomer resides adjacent to the glass surface, allowing removal of the glass slide. After rinsing the unmolded sheet with isopropanol, the pillars protrude through the polymer (e). The sheet can now be readily peeled off from the mold. Examination of the perforated membrane in a SEM showed that 100% of the vertical vias were free from blocking residual membrane (f).

P

glass slide bonding interface

blister tube a

b

Figure 3. Experimental setup for testing bond strength between two laminated polymer layers (a) and (b) of which the top surface of (a) has been previosly inhibited.

The layers are bonded together except for a circular region acting as a blister. The blister is pressurized and the delamination pressure registered.

4. EVALUATION

The depth of the inhibition can be controlled by adjusting the amount of photoinitiator in the prepolymer (Table 1). A profilometer surface scan of the polymer surface showed a thickness variation of less then ±1 ȝm at the areas where the membrane was not perforated.

The inhibition process leaves a layer of partially uncured polymer at the surface facilitating bonding to a subsequent layer. Bond test structures were fabricated by laminating two polymer sheets of which one surface was inhibited during previous polymerization. The

Table 1. The effect of adding aryliodonium photoinitiator to the UV curable polymer

OG133 on the film thickness.

bond was UV irradiated for 1 min. In a blister pressure test (Fig. 3) the bond remained intact at pressures up to 7 atmospheres (700 kPa), which was the limit of our setup. When rupturing the structure by mechanically pulling the tube connector, the mechanical failure of the PDMS occurred throughout the bulk of the material and not at the interface, indicating a monolithic bond.

0 %1%

3%

Aryliodonium salt in

OG133 (mass%) Depth of inhibited zone (µm) 152 5 5. CONCLUSIONS

In this paper we proposed a novel method for spatially controlling curing of PDMS like polymers, which allows for an uncomplicated, high-yield fabrication of vertical interconnecting channels and a high throughput stacking method for building 3D microfluidic systems by replica molding.

REFERENCES

[1] B. H. Jo, L. M. Van Lerberghe, K. M. Motsegood and D. J. Beebe, J.

Microelectromech. Syst., (2000), 9, 76.

[2] J. R. Anderson, D. T. Chiu, R. J. Jackman, O. Cherniavskaya, J. C. McDonald, H. K.

Wu, S. H. Whitesides and G. M. Whitesides, Anal. Chem., (2000), 72, 3158.

[3] U. Kloter, H. Schmid, H.Wolf, B. Michel, D. Juncker, 17th IEEE International Conference on. MEMS, 2004, 745- 748