The 2015 Paris climate agreement aims to put the world on a pathway that restricts global warming to well below 2°C above pre-industrial levels. Translated into emission trajectories, this means that net global greenhouse gas emissions have to decrease to zero in the latter half of this century (United Nations 2015). In terms of realizing the Paris ambitions, the recent decade has seen very positive developments in the form of rapid cost reductions and increases in deployment volumes for several central technologies, including solar photovoltaics, wind turbines and Li-ion batteries for battery electric vehicles (IRENA 2019; Nykvist et al. 2019)....

The central challenge when it comes to any form of innovation process is how to take a solution that has been proven in a laboratory setting and implement it in a real-world setting. While successful tests at a lab-scale may be an indication that a technology could become a successful innovation, they are by no means a guarantee. In fact, most technologies that have passed lab tests do not become commercialized whereas others take many decades from lab to industrial application (Hellsmark, Frishammar, et al. 2016; Moore 2014). As the phrase “valley of death” suggests, this formative phase of technology...

Figure 5 presents an overview of the original HYBRIT project time plan as it was laid out in 2018, with the pre-feasibility study (2016—17), pilot phase (2018—24) and demonstration phase (2025—35) in sequence.

As background, at this point it is worth reviewing the current status of the different phases, based on information available in early 2020.

In the summer of 2018 the HYBRIT pilot phase was initiated, as a SEK 528 million grant for the pilot phase was secured from the Swedish Energy Agency, with the remainder of the total pilot phase cost of SEK 1.4 billion...

Both the HYBRIT consortium and the literature on innovation systems agree that a demonstration plant should be “sub-scale” and, in terms of capacity, lie somewhere in between a pilot plant and a commercial plant. However, this is still a quite broad range, as the H-DR pilot facility in the HYBRIT case has an iron ore reduction capacity that is a mere 1% of that of a commercial blast furnace.

In the concluding discussion, we focus on how the scale of a HYBRIT demonstration plant affects both its value in terms of learning and on the characteristics of risks associated with...

In this working paper, we have drawn on research from the innovation systems literature to discuss how decisions that relate to scale and capacity of the HYBRIT demonstration plant influence a) the learning value of the demonstration phase towards full commercialization, and b) the risks associated with the deployment of the demonstration plant itself.

With the aim of making this discussion more tangible, we have centred our analysis around two different hypothetical HYBRIT demonstration plant alternatives: one “small”, which is more of a large pilot plant (80 MW electrolysis capacity) and one “large” that is of a size (320 MW...