In the global push toward sustainability, net-zero gains in greenhouse gas (GHG) emissions have been a primary target across industries. Chipmakers have been working extensively to reduce emissions from the fab activities (Scope 1) and electricity used for machinery, including heating and cooling equipment (Scope 2). Together, these areas represent about 65% of manufacturing GHG emissions for semiconductor companies. Now, manufacturers are faced with the challenging task of dealing with upstream emissions—or Scope 3.
Scope 3 emissions are those generated by the suppliers providing semiconductor companies with the silicon, other materials, and services they need for chip fabrication. This timing of this move is partly driven by the commitments of major customers to achieve net-zero GHG in their global operations within the next one to two decades. A good example is Intel, which has committed to net-zero emissions by 2040, with an interim target of using renewable electricity exclusively by 2030.
The vast majority of Scope 3 emissions generated by materials and services typically used in chip manufacturing come from three sources. Purchased materials account for the vast majority at 62%. Meanwhile, maintenance and equipment upgrades account for 22% while transportation and deliveries make up 6%. The remaining 10% comes from facilities and other business support. These are, of course, averages based on typical components, which can and do vary from manufacturer to manufacturer.
For example, the 62% attributed to purchased materials include chemicals such as solvents, silicon wafers, gases such as nitrogen, and metals like tantalum and aluminum, among other materials crucial to the manufacturing process.
Even between two manufacturers using the same services and materials, variances can make assessing upstream emissions accurately a challenge. For instance, the aluminum used in chip manufacturing requires 99.9% purity, compared to the 99.0% that passes muster in most industries. That small difference requires a tremendous increase in energy consumption (and consequently emissions) to produce pure aluminum.
Like their counterparts in other industries, semiconductor manufacturers usually procure needed materials and services from hundreds of suppliers. At first glance, that makes the idea of assessing and managing upstream emissions even more of a nightmare.
Fortunately, most semiconductor manufacturers can address at least half of their Scope 3 emissions by dealing with just a handful of their suppliers. On average, at least 50% of a chipmaker’s upstream emissions for chemicals, wafers, and gases are generated by its top ten suppliers. Another few suppliers are responsible for at least half of upstream emissions from equipment upgrades, spare parts, and maintenance.
Most chipmakers utilize a four-level strategy to achieve upstream decarbonization. In order of complexity to implement, these include supplier decarbonization, waste reduction, materials optimization, and product specifications.
Supplier decarbonization is a simple step that involves switching from current Tier 1 suppliers to others that offer lower carbon footprints. However, if preserving current relationships is the focus, working with suppliers to agree on emissions reduction targets, identify abatement levers, and define execution road maps is an effective strategy. Offering financial incentives to encourage the use of renewable energy is another option. Finally, encouraging Tier 1 suppliers to pressure Tier 2 suppliers to reduce emissions should not be overlooked.
Waste reduction inside the manufacturing facility itself is both an obvious strategy and highly effective. Calculating trade-offs is often involved, such as determining whether using chemical bath solutions longer before replacement would cause an unacceptable reduction in product yield.
Through R&D, quality control, and engineering, semiconductor manufacturers may find they can switch to lower-emission materials, chemicals, or gases used in production. Again, trade-offs will need to be analyzed, such as whether such a change in materials would cause an unacceptable decrease in product performance or yield.
Finally, chipmakers seeking to cut upstream emissions regularly review product specifications in search of those that can be relaxed without interfering with product quality. For example, reducing the required purity levels in a given metal could substantially reduce GHG emissions for the supplier.
Despite the large number of suppliers involved in generating upstream emissions, semiconductor manufacturers can make a substantial impact with reasonable effort. By pursuing just a few strategies targeted at a handful of key suppliers, chipmakers can make great strides toward reaching zero-emissions goals over the next decade and beyond.