Literature/202504281112 sdl approach to distributed discovery

• Source: [[@canty2025Science acceleration and accessibility with self-driving labs]]
• Tags: #SDL #self-driving-lab #democratization #material-discovery

The main difference between an SDL (self-driving labs) and High-throughput experimentation (as already defined) is "the judicious selection of experiments, the adaptation of experimental methods, and the development of workflows that can integrate the operation of multiple tools."

The core idea of this paper is that there is a dual approach towards SDL (self-driving labs): centralized or decentralized labs. The objective is to accelerate discovery of new materials and (perhaps) democratize the access to scientific results.

Centralized vs Decentralized

The effort of maintaining an SDL is large, therefore dividing the paradigm in centralized vs decentralized can help understanding how many resources should be allocated and what to expect. In this context, the authors divide between "shared facilities" and individual researchers' labs.

The idea is that "low-cost" systems can be used to develop workflows, testing, and troubleshooting before submitting requests to external facilities.

There's also some discussion about creating instruments that can be plugged into external facilities, allowing the development of custom solutions that can fit into a larger pipeline. In this context, large national laboratories can be the building ground for collaborations and to explore the best paths forward.

Whatever the approach, it is possible to extend the access to SDL's regardless of researchers' economic means. It's not a matter of design, but of deciding how they will be managed and for what outcome.

Financially, there's a chance centralized facilities have an easier time attracting funding (both public and private). They could avoid cutting on waste and can help prevent redundancies.

Centralized approaches pool resources to achieve more technologically advanced SDLs but face challenges in generalizability while distributed approaches provide flexibility but require greater coordination. As SDL-related technologies evolve, however, many of the capital and operating expenses (for the creators, managers, and users of SDLs) will change.