Introduction

Historically, it has been a running quip that ‘fusion is always 30 years away.’ This is a phrase, or some variation of it, any researcher in fusion energy must have heard of at least once in their career. In fact, fusion energy’s promise of clean, safe, and virtually unlimited energy has driven researchers of many generations toward the realization of this future energy source. However, in recent years, efforts toward commercialization have significantly accelerated worldwide, which is exemplified by recent national initiatives like the Bold Decadal Vision for Commercial Fusion Energy in the U.S. [1] and the UK’s electricity-generating STEP plant that plans to be completed by 2040 [2]. Further, the private sector is proposing even more ambitious timelines than the UK government, where 3 out of 4 companies are expecting to achieve fusion power generation during the 2030s [3].

Thus arises the following question: is the age-long sarcasm of “fusion is always 30 years away” still valid in 2023? This paper answers this question through a literature review of researchers' expectations about when fusion energy will be “ready” for over the past 40 years.

Results

Figure 1 summarizes the evolution of the researchers' remarks on the estimated timing of fusion realization published between 1985 and 2022. The screening protocol for literature is described in the “Appendix”. The horizontal axis of Fig. 1 shows the year of publication, and the vertical axis plots the remarks within the given paper about the number of years needed before fusion is realized, i.e., the “fusion is X years away” factor.

Fig. 1
figure 1

Researchers’ expectations on fusion realization over the past 40 years. (N.B. Drawn by the author based on Tables 1 and 2. JET: Joint European Torus; NIF: National Ignition Facility; JT-60: Japan Torus-60; FIA: Fusion Industry Association.)

Full size image

The authors then performed regression analyses of the review result. Here, the realization of fusion was categorized into two phases: the first transmission of fusion power to the grid, i.e., the realization of a DEMO plant or equivalent, and the introduction of fusion to the commercial market, i.e., the realization of the first commercial plant. Linear regression models, which assume constant progress over the years, are shown as dotted lines (the Constant Progress Model). The model assuming non-constant progress, regression models fitted with cubic functions, are drawn as solid lines (the Dynamic Progress Model). Regression results are summarized in the “Appendix” as Table 2.

Discussions and Conclusion

Progress Towards Fusion Realization

The result illustrated in Fig. 1 is insightful. Referring to the Constant Progress Model, which assumes constant progress toward fusion realization over the past four decades, the slope of the regression line is − 0.4319 for first fusion power generation and − 0.2521 for commercialization. In other terms, scientists’ expectations for the first electricity generation plant have shortened by 2.5 years every 10 years since 1985; similarly, expectations for the first commercial plant have shortened by 4.3 years every 10 years. This indicates that the scientists’ collective expectation for fusion realization has progressed over the last four decades, as opposed to widespread cynicism.

A further intriguing trend becomes apparent under the Dynamic Progress Model, which assumes a dynamic change in expectations toward fusion. Observing the solid lines in Fig. 1, it is clear that the expectations for fusion energy went downhill from 1985 until 2000s, and then improved afterward. This is a quantitative indication that scientists were optimistic about the realization of fusion until the 1980s, but became pessimistic from the 1990s to the 2000s, and then their confidence returned after the 2010s. The peak of the pessimism was in 2006, with “fusion being 29.2 years away” for demonstration, and in 2001, with “fusion being 53.0 years away” for commercialization.

Accelerated Expectations for Fusion

Comparing the R-squared values of the two models (Table 2), it is clear that the Dynamic Progress Model fits the evolution of the expectations for fusion better than the Constant Progress Model. This implies that the change in the expectation for fusion has, in fact, not been constant; rather, it has been accelerated since the 2010s.

What caused these dynamic changes in expectations? Comparing the timelines of the key events in the fusion community and the shifts in expectations, it would seem that the major scientific achievements in the late 90 s (including the successful JET and JT-60 experiments) created the turning point for the expectations for commercialization. Then, milestone achievements of the public programs (including NIF and ITER) as well as the advent of the private sector around the year 2010 coincided with the turning point in the expectation for the first fusion power generation.

It would be difficult to objectively conclude if private enterprises (or the public programs, for that matter) contributed to the accelerated progress or if they are simply benefitting from this trend from this result alone. Personally, judging from Fig. 1, the author would like to believe that private companies are both contributors and beneficiaries of this trend. In the end, the turning in expectations did happen before the advent of the private sector; at the same time, there would be no denying that the accelerated expectations for fusion around the year 2020 owe significantly to the private sector.

Limitations

Before moving into conclusions, it should be noted there are some limitations to the study. The most notable limitation of this review is that, due to the ambiguous nature of the review criteria, the authors could not design a systematic review process. As a result, the authors could not ensure that every published literature that mentions the timing of fusion realization is included in this study. Further, the “fusion is X years away” statements are often subject to biases and opinion of the scientists, rather than scientific assessments. Notwithstanding, the authors made every effort to be comprehensive and impartial in their screening process as outlined in “Appendix”, and the extracted literature (N = 45) was sufficient to produce statistically significant regression models. As a result, the authors believe that the result is sufficient to discuss the general trend of the scientists' expectations of fusion energy over the past four decades.

How Many Years Away is Fusion?

In conclusion, according to the collective remarks by scientists, the popular phrase “fusion is always 30 years away” is proven wrong, technically speaking. To be precise, we should now say “fusion was said to be 19.3 years away 30 years ago; it was 28.3 years away 20 years ago; 27.8 years away 10 years ago.” And now, scientists believe fusion energy is only 17.8 years away. So there is a progress, and it is accelerating toward the realization of this ultimate clean energy.