UNDERSTANDING THE SINTERING OF DIGITAL INKJET PRINTED (DIP) CONTACTS TO ACHIEVE LOW- CONTACT RESISTANCE ON SILICON SOLAR CELLS Abasifreke Ebong, Veysel Unsur, Nian Chen and Ahrar Chowdhury Energy Production and Infrastructure Center (EPIC), Department of Electrical and Computer Engineering, University of North Carolina at Charlotte, 9201 University City Blvd, Charlotte, NC 28223-0001 ABSTRACT: Digital inkjet printing (DIP) is a precise and promising technology to create fine gridlines for silicon solar cell. It is based on drop on demand (DoD) technology, which drops only when it is properly aligned to avoid wide gridlines. The gridline spreading is controlled by heating the chuck (wafer holder) to ≥200 o C at the time of the ink droplets. This eliminates drying step that is common to the screen-printing technology and eradicate gridline spreading. The screen-printed technology (SPT) on the other hand cannot exclude the drying step but relies on the silver loading and organics to balance the paste viscosity and rheology, which are used to control the gridline spreading. The difference between the inkjet and screen-printed inks is the particle size of the paste or ink constituents. While the DIP inks are in the nano-particle range, the SPT counterparts are in the micro-regime. Therefore, understanding the sintering of the nano-particle sizes is needed to achieve low contact resistance and hence high fill factor (FF), which is influenced by the total series resistance of the device. In this paper the microstructural analysis of the inkjet contact system was used to optimize the peak firing temperature for DIP gridlines. The optimized temperature profile found was similar to the SPT. This led to FF of ~79.2% for mono crystalline cell and efficiency of ~19.3% and ~17.4% for multi-crystalline with FF of ~78.5%. It was also found that the series resistance of these cells was not dominated by contact but emitter resistance. Keywords: inkjet, solar cell, sintering 1 Introduction Screen-printing has been the dominant metallization technique for commercial silicon solar cell because of its high throughput and cost effectiveness. The front and back contacts to a solar cell involves the printing of Ag paste through wire mesh opening ranging from 40-100 µm and the Al through 200-290 mesh. The front Ag gridlines are printed on the antireflection (AR) coating, while the back contacts are printed directly on silicon substrate. Both contacts are fired simultaneously at elevated temperature of ~750-800 o C and dwell times of few seconds. Figure 1 shows the Ag paste system consisting (i) metallic powder of Ag (ii) glass frit which consists PbO, Al 2 O 3 , SiO 2 etc. and (iii) organic binder. For the front Ag contact, during the firing or sintering process, the glass frit start melting at low temperature (>380 o C), react with AR coating to be reduced to its metal. As the temperature increases, this molten metal then dissolves Ag, and silicon dissolves into the mix. At the same time the gridline sinters to form a low resistance conductor. As the system cools down, the three materials separate. The silver precipitates (known as Ag crystallite) so that the silicon regrowth embed part of it while the other part is covered by the reformed glass which bonds the gridline. The effectiveness of the gridline sintering and contact formation is a function of Ag particle size and morphology, and percentage of glass frit. The organic binder normally evaporates after 380 o C but plays a major role in the gridline print quality. Critical to front contact quality is the adhesion, which is a function of the glass content. While desiring a thin glass layer at the gridline/crystallite interface, it must be adequate to provide the required adhesion. Therefore, understanding the paste system as well as the sintering of the gridlines and contact is critical. Although screen-printed technology is advantageous in high throughput, the control over the gridline width and height is lacking. To overcome this drawback, digital inkjet approach, which is precise in gridline geometry definition has been proposed. However, to implement this requires building the inkjet machine suitable for (i) high throughput jetting as with the screen-printer, (ii) fire through AR coating and (iii) exhibit excellent adhesion. Figure 1: Front silver paste constituents The challenges of implementing digital inkjet metallization include (i) Designing a digital inkjet machine that dispenses full gridlines at jetting speed equal to the SPT (ii) formulation of Ag inks with nano- particle silver that will not sinter at room temperature, (iii) maintaining the wafer at >200 o C during printing to avoid finger spreading, (iv) keeping the nozzles unclogged and (v) sintering the gridlines and forming the contacts in the same fashion as the screen-printed contacts. To overcome these challenges, Xjet Solar designed a digital inkjet machine that can jet full gridlines on 6-inch wafer at the speed similar to the conventional screen- printer. Most importantly, the machine is so smart that the nozzle that is not aligned properly does not fire to avoid gridline broadening. In addition, the chuck temperature of >200 o C was achieved without clogging the nozzles. Also, a two layer approach was adopted to address the contact and gridline resistances, independently. Thus the first layer, called the contact layer, consists of high percentage of glass frit to facilitate the etching of AR coating to form contact with silicon and at the same time provide bonding to the second layer – the finger layer, which contains very little or glass frit. Although the two layer approach is great in optimizing the inks separately, bonding between the