Challenges and Solutions to Meet Future Cost Targets

Since LED became an attractive alternative for general lighting, the market demand for higher brightness, higher efficiency and lower costs was the motivation for improving the LED tech – nology. Whereas an increase in brightness and efficiency is mostly a question of the LED chip design, a reduction of the costs is in the focus of the manufacturing technology. As Haitz stated in 2000, the cost per lumen falls by a factor of 10 every decade. This, later referred to as Haitz’s law [1] , is considered the LED counterpart to Moore‘s law, which states that the number of transistors in a given integrated circuit (IC) doubles every 18 to 24 months. Both laws rely on the process optimization in the production of semi – conductor devices. However, at the IC industry it is recognized that an increase in the number of transistors come along with higher costs as manufacturing processes then have to meet higher requirements. These performance related  expenses need to be compensated to keep the  overall costs low. Several factors could be key to  success, e.g. the introduction of standards, the  integration  of  several  functions  to  minimize  the   number of process steps or the increase of the  wafer size from >2″ to 300 mm or even 450 mm  in the near future.

Additive Reality: Drops of Technology

The solder mask application revolves around the solder mask material. Still, not all the attention should end on the material alone. The inkjet printed solder mask layer will be made of a collection of drops; next to the solder mask ink, a few other materials play a role in the shaping of these drops.

My previous column paid great heed to the positioning of the drop. However, at the same level as the position, what matters is how the flying drop, once on the substrate, will reach its final shape. The cross section of this shape (Figure 1), and especially the edges, are the result of an equilibrium. The contact angle theory describes it. The interesting catch here is that three states of matter are involved, and therefore, a large set of materials affects the drop. Even a virtual, digital state of matter will play a role.

luca_three_state_of_matter_700.jpg

Figure 1: Sketch showing drops on different materials.

The drop, in its liquid state, finds an equilibrium position on a solid—the substrate. The predictability of this equilibrium is an inkjet core parameter. It dominates the ability to reproduce a pattern with a high fidelity. For the physics-lovers at heart, there is plenty of literature that explains why, what, and how this equilibrium arises. For the scope of reproducibility, it is sufficient to review the materials involved: air, the ink material, and the substrate. The latter will need special care, since each substrate varies in topography and matter.

Air is the most straightforward material element to discuss and should not be taken for granted. The attention goes to suspended particles and air currents. First, when by chance, the drops precipitating on the solder mask layer give strong perturbation to the contact angle. One consideration and two action points prevent issues. Only large size particles of at least few micrometers actually create problems. This is a relief; decades of expertise from the semiconductor industry effort on controlling particles are available to handle such sizes.

From the consideration to the actions: The first action happens outside of the inkjet printer, a wet cleaning frees the incoming board surface of contaminants, such as particles and organic contamination. The second action ensures that the internal environment of the printer has at least ISO 5 (class 100). The absence of humans in the internal environment simplifies the functional design for this specification, since they are the biggest source of particles. Additionally, this specification implies that a slow, homogeneous flow of clean air is present at all time. This, together with a proper printhead array design and movement, delivers predictable air currents where the printhead jets the drops.

The solid (substrate) will present itself in several ways. Laminates and copper are by far the most common and therefore the focus here. Although, the occasional gold finishing or the polyimide substrate would follow similar considerations. Alas, the product design fixes these materials and their topography. The idea of accommodating a favorable contact angle by replacing them is unfeasible. However, the only point of care is the surface where the drop lands. The bulk properties of the substrate components can stay the same while the surface acquires the desired properties with pre-treatments. So far, the PCB industry improves adhesion with surface pre-treatments. Additionally, inkjet printing asks for a good match with the ink uniformly across the board. The skill of the manufacturer would be to require such match from his pre-treatment chemistry supplier and to have wet-chemical line setups that can deliver the uniformity needed. The pre-treatments out there, in their commercial state or in their development phase, have two flavors: either they care about the copper and focus on avoiding bleed (the tendency to form an almost zero contact angle) or they take upon the whole board and prepare a homogeneous surface independently from the material below, be it a laminate or a copper.

The first would leave to the inkjet technology the task of compensating contact angle differences between, for example, laminate and copper. Companies offer this technology today (examples in alphabetical order: Atotech and MEC COMPANY LTD). The second pre-treatment option, with their additional inner complexity, makes inkjet printing patterning task easier. Companies such as Taiyo offer such pre-treatments. In this initial phase of inkjet printing technology for solder mask, a clear winner for the pre-treatment strategy has not emerged yet. Indeed, a recent cooperation between two material suppliers was announced. Many collaborations between ink manufacturers, pre-treatment suppliers and equipment suppliers are ongoing. This is good news: awareness is at the right level in the material supply chain.

The liquid is a single material. A worthwhile exercise is to describe the common properties of the materials: The viscosity of inkjet printable solder mask is low, lower than the traditional solder mask. At room temperature, the ink would have a viscosity a bit higher than milk and a bit lower than linseed oil. Their rheology brings in other flow characteristics such as thixotropy. Inkjet inks are generally Newtonian liquids, unlike standard solder mask which is pseudo-plastic (shear thinning). Their fillers are, if present, one or two order of magnitude smaller than in traditional solder mask. The integrated photoinitiator enables a fast and complete polymerization of the material. Finally, the last noteworthy macro-property is the amount of volatile organic components (VOCs) that is close to zero or zero depending on the brand and type. This latter property is at the base of the environmentally friendly principle of these materials. Just to be clear, not only VOCs are low, also the absence of imaging and development steps are clear advantages in the low environmental impact of inkjet printing.

After a few years of close work with several inkjet material providers, my opinion is that their roadmap will be similar. First, these providers will strive to finalize the compliance list within the IPC-SM-840 and other customer driven standards; some suppliers are far ahead and some are preparing their starting move. After that, each one will acknowledge the strength of their developed material(s) and explore how much further these can reach. I do not exclude all-around ink.

Examples that concern inkjet printing might be: stable and high contact angles for finer features, stackability for high aspect-ratio features, appearance beyond color (matt / gloss levels) to favor optical automatic components placement, combination of tuned electrical properties and thickness control as provided by inkjet for extreme RF applications (5G and future radio frequencies), extreme low viscosity for under-filling purposes, etc.

So, you’ve stayed with me this far, and now you want your answer: Which is the best bet for my solder mask inkjet material? It is not the role of this column to make explicit commercial preferences. Instead, my aim is to provide a timeless guide on how to assess a material and its supplier. One essential trademark of a serious ink material supplier is their (in-house) ability to implement inkjet printing, either at a lab scale or at a pilot level. Only in this way, the ink supplier will understand the needs and challenges of their customers when implementing this technology for the first time. Equipment suppliers have also a support role, though having a common language is a base for good communication. Another characteristic is how they integrate inkjet in their roadmap. Is inkjet printing material an additional product on their brochure or a core technology that will eventually replace their present workhorse? Finally, how do they place themselves in the market? To my knowledge, more than a handful of ink supplier have commercial solutions (in alphabetical order: Agfa, Electra Polymers Ltd, MicroCraft K.K, Peters Group, Shenzhen RongDa Photosensitive Science & Technology Co. Ltd., Taiyo America) and several other material suppliers are considering their move from development to commercialization. The scene is therefore getting broader, which favors the progress of the technology.

The three states of matter are now through, though the discussion this far had one assumption: the contact angle equilibrium happens well before any solidification mechanism by polymerization and cross-linking induced by UV light. In the time between the contact angle formation and the solidification mechanism, approximately a hundred microseconds, the printed feature expands its edge front by a few micrometers. The inkjet equipment handles this information on material. This knowledge is the final touch to get the feature size exactly as intended. This is a part of the “what you see is what you get” art.

This article brought in several detailed aspects of the materials involved in inkjet printing. The supply chain is in its early age indeed, though, the critical years of initial uncertainty are past. Interested PCB manufacturers will find a good level of competence independently from the supplier of choice.

Luca Gautero is product manager at SUSS MicroTec (Netherlands) B.V. 

Partnership between SUSS MicroTec and SET to develop a combined equipment solution for 3D chip integration

SUSS MicroTec SE and SET Corporation SA announce a partnership in sequential die-to-wafer (D2W) hybrid bonding, a die-based interconnect technology. As part of the partnership, SUSS MicroTec and SET will provide a fully automated, customizable, highest-yield equipment solution to customers. This solution will accelerate the industry’s path towards advanced 3D multi-die solutions such as stacked memory and chiplet integration.

Garching, Germany and Saint-Jeoire, France, September 1, 2021 – SUSS MicroTec, a leading supplier of equipment and process solutions for the semiconductor industry announces a joint development agreement (JDA) with SET, a leading supplier of high precision flip-chip bonders. The main focus of the JDA is the development of a fully automated, customizable, highest-yield sequential die-to-wafer hybrid bonding equipment solution by combining SUSS MicroTec’s expertise with FEOL-compatible automated surface preparation of wafers and singulated dies which are populated on a wafer or frame and SET’s ultra-high accuracy die-placement technology, which will be further enhanced by high-performance metrology that offers closed-loop feedback to the bonding system.

At a time where traditional transistor-scaling is approaching its limit, 3D packaging and heterogeneous integration have already been widely adopted in the industry in order to further increase the performance and functionality of today’s semiconductor devices. However, today’s 2.5D and 3D packaging schemes are limited by the minimum interconnect density that traditional microbump technology can offer. Hybrid bonding solves this problem by bonding the direct contact between two metal pads (mostly copper) and surrounding dielectrics in one single bonding step. This bumpless bonding approach allows for substantially smaller pitches and higher interconnect density which are the key enablers for future generations of multi-die solutions.

Interconnect density scaling is driven by a number of fast-growing applications that include power computing, artificial intelligence (e.g. autonomous driving), 5G mobile, as well as a variety of additional More-than-Moore devices such as next-generation CMOS image sensors. To obtain high yields for devices with high interconnect densities customers not only require ultra-precise die-placement solutions, but also reliable surface activation and a process guaranteeing particle-free surfaces.

As part of this partnership, SUSS MicroTec’s high-efficiency surface preparation modules and throughput-optimized metrology solutions for post bond overlay verification will be combined with SET’s latest ultra-high accuracy D2W hybrid bonding platform. Closed-loop feedback between the metrology and the die-bonder module will help to continuously monitor and optimize the overlay performance, enabling consistent die-placement accuracy below 200 nm as well as interconnect pitch scales in the sub-micron region. The modular, highly flexible equipment concept allows for stand-alone surface preparation and hybrid bonding as well as a fully integrated equipment solution depending on the specific application and/or customer requirements. This concept also enables an integrated cluster approach that can support all individual hybrid bonding paths in a single platform: wafer-to-wafer (W2W), collective die-to-wafer (CoD2W) and/or sequential die-to-wafer (D2W).

Goetz M. Bendele, PhD, CEO of SUSS MicroTec, puts this into perspective: “Hybrid bonding is one of the main growth drivers of the advanced semiconductor backend equipment space, as well as one of the main growth levers for SUSS MicroTec. With our partnership with SET, we will be able to offer our customers a complete suite of both die-to-wafer and wafer-to-wafer hybrid bonding solutions for the broadest set of heterogeneous integration applications in the advanced backend space. Our die-to-wafer bonding solution, leveraging the combination of SET’s leading-precision die placement technology with SUSS’s proven surface activation, automation, and metrology capabilities, will deliver additional customer value through differentiation in terms of throughput and yield, while at the same time enabling friction-less integration into our customers’ fabrication sites.”

Pascal Metzger, PhD, CEO of SET: “Thanks to several partnerships we had & our more than 10 years experience in hybrid bonding, we have succeeded in taking hybrid bonding from a purely laboratory state to an industrial state. Thus in September 2019, SET launched a stand-alone machine – the NEO HB. Thanks to our new partnership with SUSS MicroTec, we are now going to accelerate the integration and automation phase of the process. This will permit to provide a complete industrial solution to our customers, for applications coming in the very near future such as HPC, IA, 5G and many other more, to diversify our offer and address new market segments.”

For more details on Hybrid Bonding at SUSS MicroTec please visit our website: https://www.suss.com/hybrid-bonding

About SUSS MicroTec
SUSS MicroTec is a leading supplier of equipment and process solutions for microstructuring in the semiconductor industry and related markets. In close cooperation with research institutes and industry partners SUSS MicroTec contributes to the advancement of next-generation technologies such as 3D Integration and nanoimprint lithography as well as key processes for MEMS and LED manufacturing. With a global infrastructure for applications and service SUSS MicroTec supports more than 8.000 installed systems worldwide. SUSS MicroTec is headquartered in Garching near Munich, Germany. For more information, please visit www.suss.com

About SET
Founded in 1975, based in France, SET is a world leading supplier of high accuracy flip-chip bonders (chip-to-chip and chip-to-wafer) and versatile Nanoimprint Lithography (NIL) solutions. SET accompanies laboratories and industries of semiconductor, which look for a high precision and an important reliability in the assembly of their components in their projects, and accelerate their developments of the chips of future. With equipment installed worldwide, SET is globally renowned for the unsurpassed accuracy and the flexibility of its flip-chip bonders. Ranging from manual loading to fully automated version, they adapt to all main bonding techniques: fluxless reflow, thermo-compression, adhesive joining compression, thermosonic, hybrid bonding. www.set-sas.fr

en_USEnglish
Product has been added to your cart