Where Einstein Meets Edison

Disrupting Coal with Modular Nuclear Reactors

Disrupting Coal with Modular Nuclear Reactors

May 30, 2010

Energy Secretary Steven Chu recently published an article in the Wall Street Journal advocating small nuclear reactors as a new source of electricity1.  The article emphasized that compared to traditional large reactors, small modular reactors are cheaper, use less water for cooling, and would allow utilities to buy only the capacity they need.

Modular reactors are about equally attractive as large reactors along the most important performance metrics: proliferation, safety, and waste.  However, modular reactors are a potentially disruptive technology because of their cost structure, which makes them especially attractive for uses not traditionally associated with nuclear energy such as power generation in remote locations and long-range rockets.

Several companies have recently announced modular nuclear reactor products, most notably Babcock & Wilcox, NuScale Power, and Hyperion Power Generation.  These manufacturers believe that their smaller reactors will cut the cost of constructing nuclear plants and also increase safety.  An added benefit is the flexibility of a plant to easily buy or sell modules if electricity demand changes.

Nuclear Power in the US and around the World

Nuclear power plants are an attractive alternative to coal-burning or gas-burning plants because nuclear plants are technically well understood, emit a negligible2 amount of greenhouse gas, have a minimal fuel cost, and do not suffer from intermittency issues, as do solar and wind.  Presently, nuclear power accounts for 20% of the electricity generation in the US3 and 6% of the electricity in the world4.

As a proven technology with low penetration, the expansion of nuclear power can create new business opportunities and reduce carbon emissions.  Since electricity generation accounts for nearly 40% of carbon emissions in the US5, a transformation to fully nuclear electricity generation would cut US carbon emissions by about 32%.

In the US, the approval process for new nuclear plants can take up to ten years, so little additional nuclear capacity is expected to be added in the foreseeable future.  Other issues associated with nuclear power include:


Unlike coal and gas plants, the capital cost of nuclear plants is the largest component of the total cost, as shown in Figure 1.  Because nuclear fuel is inexpensive, utilities that own nuclear plants view these plants as cash cows and are economically incentivized to extend their operating licenses for as long as possible.

The three main cost-determining factors for nuclear power are the initial capital cost, the length of the construction period, and interest rates6.  Per megawatt of electricity generated, building a single, large nuclear plant is generally cheaper than building several small plants. This is because with large plants, components such as containment walls and control rooms only need to be built once7.  Despite the lower cost per megawatt, the large investments (greater than $9 billion) and construction periods of five or more years can make investments in modular reactors more attractive.  Modular reactor manufacturers are therefore incentivized to emphasize shorter construction periods and increased financial flexibility as the key benefits of modular reactors.


One risk of commercial nuclear power is that the technologies and materials will be misused for making nuclear weapons.  Since 1945, the knowledge of how to make a nuclear device has spread quite widely.  Producing a nuclear weapon requires a sufficient quantity of plutonium or highly enriched uranium, knowledge of the principles of making a nuclear device, and the engineering effort to fabricate the device into a controllable weapon8.

Naturally existing uranium has a 0.71% concentration of uranium-238, the active isotope in nuclear reactors and bombs.  Uranium must be enriched to 4.5% to be reactor-grade, and at least 80% to be bomb-grade, but once technological hurdles to achieve 4.5% enrichment are overcome, few barriers exist to achieving 80% enrichment.  This fact allows nations seeking nuclear weapons to claim only the intent to generate electricity.

Nine nations are presently known to possess or suspected to possess nuclear weapons.  To date, no sub-national groups are known to have made nuclear weapons, and the likely path for these groups to acquire nuclear weapons is through theft or purchase.

Reprocessing of nuclear waste allows non-reacted fuel to be separated and later enriched into a form that can be reacted.  This process maximizes the energy extracted from a given amount of uranium-238, but the byproducts of waste reprocessing can be used to make nuclear weapons.  Proposals for reprocessing waste must therefore include protection of the reprocessed material from theft or black market sales.

Any proposed nuclear power reactor – large or modular – must address the issues of proliferation and plans to minimize this threat.


There are two main safety requirements for nuclear reactors.  First, the nuclear chain reaction must be quickly and reliably stoppable under all possible operating conditions.  This is typically ensured through the structures of control rod insertions and core materials.

Second, decay heat must be removed.  Even after a chain reaction is stopped, decay heat is continuously generated in large amounts. If this heat is not removed, it may cause overheating, melting, or a steam explosion.  In theory, passive cooling systems, which rely on natural heat removal mechanisms, are preferable because they do not require redundancy.  However, there are economic incentives to maximize the power density of nuclear reactors, and high power density reactors require active cooling.

The Three Mile Island accident in 1979 occurred because the water pump system failed and decay heat was not removed from the reactor core.  Although the incident resulted in no injuries and only minimal increases in the radiation exposure to the external population, it has been cited as one reason for the decline of the nuclear industry in the 1980s9.

In contrast, the Chernobyl accident in 1986 killed 56 people directly and an estimated 4,000 out of the 600,000 most highly exposed people through cancer10.  Initially, the investigation blamed operator error for the accident11, but a revised analysis in 1993, when more data became available, identified the reactor design as the main cause12.  Most notably, the reactor was designed so that the reaction rate would speed up if the coolant started overheating, leading to even greater overheating.  This positive feedback loop (called a positive void coefficient) caused minor overheating to result in a catastrophic explosion.

The risk of accidents can be minimized through well-designed reactors and detail-oriented operating procedures, as illustrated by the perfect safety record of the US Nuclear Navy.  The US Navy has operated nuclear submarines and aircraft carriers for over 50 years but has never suffered a single reactor-related casualty or escape of radiation13.  This is a remarkable track record, considering the numerous accidents suffered by the Soviet nuclear fleet and the fact that nuclear vessels in the US Navy are staffed mostly by young men and women fresh out of school.

In his book Chasing the Rabbit, MIT Professor Steven Spear credits the culture of the nuclear navy for this exceptional successful record.  Operators are all clear on how equipment is expected to operate, what to do if performance deviates from expectations, and how to incorporate new knowledge into improving these procedures.  Safety-related alarms went off more frequently at Three Mile Island than in the nuclear navy, but in contrast, operators often did not know how to react to these alarms or simply ignored them14.  The nuclear power industry can operate safely and match the exceptional record of the nuclear navy by creating a similar culture of problem discovery and continuous improvement.


Unlike the gaseous waste from fossil fuel burning power plants, nuclear waste is a radioactive, heat-emitting solid.  Because of the harm it can cause to biological organisms, the waste must be isolated from the biosphere.  Most long-term proposals involve burying it deep underground.

The economics of nuclear waste disposal sites favor building large-capacity sites, but large sites give rise to political problems.  The political problems arise mostly because populations living near a disposal site will feel they are carrying an unfairly large burden.  In 2002, Yucca Mountain in Nevada was chosen as the main site for storage of waste generated in the US.  However, in May 2009, Secretary Chu announced the appointment of a committee to look at other options.

Modular Reactors

Modular reactors are expected to reduce the initial capital cost of nuclear power through simplicity of design, economies of scale due to mass production, and reduced siting costs15.  Additionally, modular reactor designs commonly include passive cooling and a pre-loaded, multi-year supply of fuel to minimize the operating costs.

Babcock & Wilcox

Babcock & Wilcox is known for building submarine reactors for the Navy.  Their modular nuclear reactor, called the mPower reactor, will generate 135 to 140 megawatts.  The company is already working with the Tennessee Valley Authority to set up a site for the first plant using this technology, but this plant is not expected to be operational until 2018 at the earliest16.

Hyperion Power Generation

At 1.5 meters across, Hyperion power modules (HPM) are approximately the size of a hot tub.  These modules produce 25 megawatts of electricity, enough to power 17,000 American homes.  HPMs are ideal for communities seeking to be independent of their local utility’s power source, such as military bases, hospitals, and college campuses. 

NuScale Power

Unlike Hyperion’s small backyard reactors, NuScale expects its systems to be used in large nuclear plants.  The company claims it can cut construction cost and increase safety with its 45-megawatt modular reactors, largely due to its passive water-cooled system17.  NuScale plans to submit its design to the US Nuclear Regulatory Agency in 2011.  The company expects a three-year review process, so its first nuclear plant may not be ready until 201818.

Akme Engineering

Based in Russia, Akme’s goal is to produce a prototype of a 100-megawatt nuclear reactor small enough to fit into a typical American backyard by 201919.  Akme’s design uses molten lead rather than water to cool the reactor core, because the high boiling point of molten lead compared to water allows heat to be extracted from the core more effectively.  


Westinghouse is a maker of traditional reactors, having built the first nuclear reactor in the US in 1957.  The company entered the modular reactor market when it won a $5.3 billion contract with China to build four 1,000-megawatt reactors.  The first two are expected to go online in 2013, and Westinghouse intends to build even smaller reactors that it hopes will reach more markets20.

A Disruptive Technology

The early adopters of modular reactors will likely be remote communities who value portability and independence from the grid, rather than large utilities.  In many remote areas where regular delivery of fossil fuels is prohibitively expensive, a modular reactor could provide electricity for the community and power to pump fresh water from deep underground.

Flexibility is the most attractive feature of modular reactors compared to large-scale generators, which may not appear in an economic comparison model21.  Building a large generation plant with modular reactors keeps options open, because the plant can flexibly scale up or down to meet electricity demand.  These options are most valuable under conditions of high uncertainty and long expiration periods, and nuclear power plants can find themselves in precisely these conditions.  For example, when a new urban area is under development, the electricity demand is highly uncertain over the timescale of decades.  In such a situation, the option value of modular reactors makes them likely to disrupt traditional reactors in the marketplace. A master’s student in MIT’s Technology and Policy Program explored the value of options for nuclear power. The resulting thesis provides a detailed mathematical valuation22.

Like most new product introductions, the success of modular nuclear reactors will be highly dependent on proper marketing. Manufacturers must first target customers who value portability and flexibility, as there are no clear advantages in the traditional metrics (proliferation, safety, waste, and fuel cost).  Finding the right initial target group will give modular nuclear manufacturers both economies of scale and feedback for design improvement, affording them the opportunity to disrupt the low-carbon electricity market.


Figure 1: Breakdown of nuclear plant cost, compared to other sources23


Figure 2: Location of US commercial nuclear reactors, courtesy of Lara Pierpoint, MIT.


  1. Wall Street Journal, March 23, 2010. http://energy.gov/news/8782.htm
  2. Nuclear power generation theoretically does not emit carbon during the process of generating energy.  However, the term “carbon-free” is too strong, when considering the emissions associated with building a nuclear plant in compliance with stringent safety codes.
  3. Nuclear Energy Review, US Energy Information Administration, June 26, 2009.
  4. "Key World Energy Statistics 2007" (PDF). International Energy Agency. 2007. http://www.iea.org/textbase/nppdf/free/2007/key_stats_2007.pdf.
  5. “Carbon-emissions culprit? Coal”, The Seattle Times, June 3, 2007.
  6. “The New Economics of Nuclear Power”, World Nuclear Association, December 2005.
  7. “A Preassembled Nuclear Reactor”, Kevin Bullis.  Technology Review, June 16, 2009.
  8. Making Technology Work – Applications in Energy and the Environment, John Deutch and Richard Lester, Cambridge University Press, 2004.  pp. 154-156.
  9. Making Technology Work – Applications in Energy and the Environment, John Deutch and Richard Lester, Cambridge University Press, 2004.  p. 122.
  10. "IAEA Report". In Focus: Chernobyl. http://www.iaea.org/NewsCenter/Focus/Chernobyl/.
  11. IAEA Report INSAG-1 (International Nuclear Safety Advisory Group). “Summary Report on the Post-Accident Review on the Chernobyl Accident.” Safety Series No. 75-INSAG-1. IAEA, Vienna, 1986.
  12. NEI Source Book: Fourth Edition (NEISB_3.3.A1) http://www.insc.anl.gov/neisb/neisb4/NEISB_3.3.A1.html
  13. Chasing the Rabbit, Steven J. Spear, McGraw Hill 2009.  p.113
  14. Spear, pp. 124-125.
  15. “Small Nuclear Power Reactors,” World Nuclear Association, updated March 2010.
  16. “A Preassembled Nuclear Reactor”, Kevin Bullis.  Technology Review, June 16, 2009.
  17. “Nuclear Reactors the Lego Way”, Michael Kanellos. GreenTech Media, August 21, 2009
  18. “NuScale Wants to Bring Modular Nuclear Reactors to the Mainstream”, Ariel Schwartz.  Fast Company, Aug 21, 2009.
  19. Safety Issues Linger as Nuclear Reactors Shrink in Size”, Andrew E. Kramer.  The New York Times, March 18, 2010
  20. “Westinghouse at core of nuclear power trend toward smaller reactors”, Thomas Olson, Pittsburgh Tribune-Review, March 14, 2010.
  21. “Option value” is the finance term used for flexibility.
  22. Modular gas-cooled reactor economics : the application of continent-claims analysis, John Stephen Thomas.  Advised by Lawrence M. Lidsky and Henry D. Jacoby, Massachusetts Institute of Technology. Dept. of Nuclear Engineering http://dspace.mit.edu/handle/1721.1/13220
  23. Risto Tarjanne, Kari Luostarinen. “Competitiveness comparison of the electricity production alternatives” (price level March 2003)
Mark Chew


Mark Chew presently leads the distributed generation policy and strategy at Pacific Gas and Electric Company in San Francisco. He joined PG&E in 2010 as an internal consultant, and he has also worked on demand-side management programs and forecasting distributed generation penetration. Mark received his MBA and MS in Chemical Engineering graduated from MIT; he also holds MS and BS degrees in Electrical Engineering and Computer Science from UC Berkeley. While at MIT, Mark was a founding editor of the MIT Entrepreneurship Review and was a lead organizer for the MIT Energy Conference. Before MIT, Mark spent 4 years at Qualcomm designing RF chips now used in mobile devices, including the iPad 3 and iPhone 4, 4S, and 5.