Here is the abstract you requested from the imaps_2018 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Screen Printing Fine Pitch Stretchable Silver Inks onto a Flexible Substrate for Wearable Applications|
|Keywords: Flexible Hybrid Electronics, Screen Printing, wearable electronics|
|The lightweight and bendable features of printed flexible electronics are very attractive. However, the functionality, performance and reliability of fully printed electronics are still limited compared to CMOS devices. To take advantage of printed flexible electronics and CMOS technologies, flexible hybrid electronics (FHEs) integrating ultra-thin integrated circuits (ICs) to flexible printed circuits is a viable solution. Currently stretchable silver inks are formulated for wide traces, typically 2 mm or greater. To attach ultra-thin silicon chips that have fine pitch onto printed organic substrate, it is necessary to print fine trace width/space that matches the pitch of the chips, which may be less than 200 microns. This paper will present the development and optimization of the screen printing process for printing stretchable silver ink onto stretchable thermoplastic polyurethane (TPU) substrate. A test vehicle was designed including test chip patterns with pitch of 200 microns as well as 50 µm/5 mm (line width/line length) to 350 µm/35 mm lines (at 4 biases). 10mm squares were placed on all four corners, which allowed us to measure sheet-resistance and determine any “within piece” variation. A 2-mm trace was included around the entire pattern. Critical features were oriented parallel to the squeegee direction. The screen mesh was biased at 22.5°. A comprehensive experiment was performed to investigate the effect of process parameters on the quality of prints. The printing process parameters were characterized for producing fine trace/space as well as for achieving low resistance. The parameters include screen characteristics (mesh count, open area, thread material, emulsion materials, emulsion thickness) and printing parameters (print speed, squeegee material/hardness, squeegee pressure, snap-off distance, one print vs. double print, print-flood vs. flood-print). One stretchable silver ink and one stretchable TPU substrate were used for the experiment. A 2-level factorial design with three replicates was selected. The quality of the prints is characterized by 1) resistance of traces, 2) sheet resistance of large squares, 3) z-axis height, and 4) trace width/spacing. The experiment was done on a DEK Horizon 03i printer. A DEK squeegee (200 Blue) and a DEK 265 flood bar (200 mm) were used. Snap-off distance was fixed to a value of 3-mil. Printing was done at humidity of RH≈50%. The substrate used was Bemis TPU ST604. It is approximately 75 µm (3-mil) in thickness with a plastic liner or backing to provide structural support during the printing process. In order to handle the flexible and stretchable TPU substrate, a custom porous vacuum plate fixture was designed and manufactured. The stretchable ink selected was DuPont PE 873. Dupont’s PE 5025 ink (non-stretchable conductive flake silver for flexible electronics) was used as a “control” to baseline the printing process. A 9 in x 9 in 400 tpi calendered stainless steel mesh was mounted on a polyester mesh (“trampoline”) within a 29 in x 29 in frame. The two different emulsions applied onto the 420 tpi meshes were Sefar S24 ultra-resolution fine line emulsion and McDermid Autotype CP capillary film. We observed significant noise in the z-axis (printed silver ink height) measured by a profilometer. We concluded that measured z-axis height by a profilometer is not a useful response variable for characterizing screen printing stretchable silver ink on TPU substrate, mainly due to high roughness of the TPU substrate. We also concluded that sheet resistance of a pad measured by a four-point probe is not accurate because the printed ink on the pads was not uniform in thickness. Thus, four-point probe sheet resistance data is not meaningful either. A calculated sheet resistance based on measured resistance value along the length of the trace, trace width, and trace length was derived. It was found that the calculated sheet resistance can replace the trace height measurement. It is a more reliable quality characteristic for evaluating printed stretchable silver ink on TPU substrates. From the experiment results, we found that squeegee pressure and emulsion type have statistically significant effects on calculated sheet resistance for printed traces while print speed does not have statistically significant effects. Lower functional squeegee pressure results in lower calculated sheet resistance. Traces printed with the mesh covered by the S24 emulsion can achieve lower sheet resistance than the ones printed with CP emulsion. After optimizing the screen printing process, we were able to print 100 µm (4 mils) trace width consistently. TPU has a low glass transition temperature. The drying process of stretchable silver ink on TPU was optimized for the box oven, the convection oven, and photonic drying. With optimized drying parameters, we are able to achieve sheet resistance values of less than 45 mΩ/sq/25 µm for all printed samples regardless of drying method. The results were better than manufacturer’s specification. This material is based upon work supported, in part, by Air Force Research Laboratory under agreement number FA8650-15-2-5401. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of Air Force Research Laboratory or the U.S. Government.|
San Luis Obispo, CA