Here is the abstract you requested from the imaps_2019 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.
|Ni Via Fill Metallization|
|Keywords: Nickel, Electrodeposition, Via Fill|
|As technology requires smaller and faster semiconductor devices, with more transistors and passives, so does this also increase the circuit density at all packaging levels. The resulting increases in wattage and thermal output places increased demands on all interconnections. It becomes critical for the semiconductor manufacturers to develop ways to make highly reliable interconnect in the most efficient and cost-effective way. The current material stack in wafer level interconnect is usually made with three layers which include, the copper pillar as the main conductor, a barrier layer and soldering material.1 A barrier layer is often required between Cu and Sn (or Sn alloy) because the adjacent layers can lead to the formation of an intermetallic phase between Cu and Sn during reflow. This intermetallic can result in voiding which may cause reliability issues. Nickel is one of most common elements used as a barrier layer. The increasing use of this barrier material then raises the question if possibly the copper conductor might be simply completely replaced by nickel. If nickel does not present any significant penalty in the replacement of copper then the potential benefits of simplicity, cost ultimate reduction, good mechanical properties and increased reliability would weigh in its favor. Is it possible to achieve similar bottom-up fill for nickel just as is seen with copper chemistry? In this work, the possibility of nickel via fill metallization by electrodeposition from aqueous solution is explored. Nickel chemistry has been developed by MacDermid Enthone to achieve bottom-up fill with results that are comparable with that of copper via fill. Initial testing shows that a via, ~25 �m wide and ~12 �m deep, was fully filled by this chemistry. These test substrates have demonstrated void-free filling of photo resist patterned substrates with a via openings in the passivate layer. This work exhibits a filling mechanism with a predictable bottom-up fill behavior and suggests that these observations align with a Convection Dependent Adsorption (CDA) model. In other words, MacDermid Enthone via fill chemistry can provide less suppression in metal deposition inside via than on the field. These differences in deposition rate show dependency on agitation which result in faster growth inside the via. While electrodeposited nickel can exhibit high internal stress, this property is controllable in this chemistry. By adjusting the composition of plating bath, the internal stress of the deposit can also be shifted from compressive to tensile. By adjusting bath make-up and operating window internal stress can be altered to help meet the different requirements of some applications. Boric acid is widely used in nickel plating bath formulations and is one of the most commonly used buffers. Nickel electrodeposition from aqueous solution has a common side reaction which is the reduction of H+ resulting in a rise in the pH at the cathode. Boric acid can play a critical role in buffering local pH. However, as the raising environmental concern of boric acid, regulation has been imposed in Europe and Asia in controlling boric acid. Also, boric acid powder can be challenging substance in clean room environment. The MacDermid Enthone via filling chemistry uses an alternative to boric acid. In this work, a boric acid free chemistry is investigated. It can achieve similar performance as boric acid chemistry.|
West Haven, CT