The following post is by Earl Killian, guest blogger at Climate Progress.


The Science Behind the HeadlinesPart 1 of this book review looked at the (mis)handling of climate science in two books by Professor Richard A. Muller — his textbook and general public book, which, confusingly, are both named Physics for Future Presidents. Here I turn to portions of the general public book, such as the chapters on climate solutions, his treatment of terrorist nukes, and even his unsubstantiated dissing of the Toyota Prius.

There are two solution chapters: “The Fruit on the Ground” and “New Technologies.” The first lists comfortable conservation methods: compact fluorescent lightbulbs, cool colors, the fizzling population bomb, and automobile efficiency. New technology solutions are biofuels, concentrating solar, safe nukes, clean coal, carbon offsets, and renewable energy. Please see earlier carbon offset posts for information on one of the listed solutions.

The “Nonsolutions” (to global warming) chapter lists hydrogen, electric cars, fusion, solar, recycling, and Kyoto. At least he got two right. There are a several mistakes here; this review will concentrate on just one.

Electric cars

Plug-in cars (primarily plug-in hybrids, but also pure battery electric cars) are one wedge of the solution to global warming and petroleum woes. Unfortunately the books attempt to discredit electric cars using bogus data and arguments. Let’s start with page 67 of the general reader:

The fundamental physics reason for our addiction (or marriage) to oil is the same reason that oil was the weapon of choice for the 9/11 terrorists: it carries huge amount of energy. Consider that our best rechargeable batteries hold only 1% of the energy of gasoline. Now ask, Why don’t we drive electric autos? Where they killed by a conspiracy? Conspiracy or not, it hardly matters. This energy storage discrepancy provides a huge physics barrier. Batteries simply do not store much energy-not when compared to gasoline.2 We love gasoline because it is so energetic!

The footnote says, 281 pages later:

2. Electric energy can be used more efficiently than the heat energy from gasoline, at least for turning the wheels of a car, and that increased efficiency reduces the disadvantages of batteries by about a factor of three. Thus, gasoline actually outcompetes expensive computer batteries for automotive energy by a factor of only 30. We’ll return to this when discuss electric cars in more detail.

Professor Muller is distorting, exaggerating, and cherry-picking again. First, the energy density of gasoline and batteries is not the most important consideration in electric vs. gasoline cars. It is distortion and exaggeration to state say that there is “fundamental physics reason” for gasoline and that battery energy density is “a huge physics barrier.” The energy density of batteries is a relevant engineering consideration, but is not a huge barrier, and moreover the data given is wrong. The energy density of Li-Ion batteries is 160 Wh/kg. The energy density of gasoline is 11,972 Wh/kg.

We can get a fairly good estimate of the efficiency difference between a gasoline vehicle and an electric vehicle by comparing instances of each that are otherwise very similar. Fortunately, in 2002 Toyota offered their RAV4 SUV as both a 2WD Automatic and as an EV, and the EPA assigned a mpg rating to each. The RAV4-EV averages 112 mpg(where the electrical wall plug energy is converted to an equivalent number of gallons of gasoline using the lower heating value of gasoline). The non-EV RAV4 averages 23 mpg. The gasoline-powered vehicle uses 4.9 times as much energy to go the same distance as the electric-powered vehicle. Taking this efficiency into account, gasoline has 15.4 times the energy density of batteries. This still sounds like a large difference, the mass of the fuel or battery does not dominate car mass, and so the difference basically tells us little about the practicality of electric cars.

If you then further compensate for the fact that your garage doesn’t have a gas pump, but it does have an electric plug. So you want enough range in a gasoline car so you only detour to the gas station once a week to refill, as opposed to charging every night, then the factor falls to just 2.2 times, and that difference is swamped by the additional mass items in a gasoline vehicle not present in a pure electric car (engine, radiator and antifreeze, oil and oil filter, transmission, catalytic converter, muffler, alternator, and so on).

Interestingly, there is a fundamental physics reason why electric motors are more efficient than internal combustion engines (the Carnot limit from thermodynamics), but Professor Muller does not want to teach physics that disagrees with his opinions.

In his second attack on electric cars (page 306), Professor Muller writes:

High-performance batteries are very expensive and need to be replaced after typically 700 charges. Here is a simple way to calculate the numbers. The computer battery for my laptop (on which I am writing this) stores 60 watt-hours of electric energy. It can be recharged about 700 times. That means it will deliver a total of 42,000 watt-hours, or 42 kilowatt-hours, before it has to be replaced for $130.

This attack is based upon economics, not physics, and here also Professor Muller has gotten his facts wrong. No automaker proposes buying laptop replacement parts to build EVs, and they are certainly not going to pay the $2.17/Wh that he estimates. In fact 18650 cells (the batteries that laptop batteries are typically made from) cost in bulk somewhere between $0.200/Wh to $0.350/Wh. The claim of 700 charges should be compared to the testing done and reported by EPRI [PDF], “battery durability testing sponsored jointly by EPRI and Southern California Edison demonstrate that current lithium-ion batteries are likely to retain sufficient capacity for more than 3000 dynamic deep-discharge cycles — about 10-12 years of typical driving.”

Manufacturers understand that plug-in vehicle need long lifetime batteries, so it should be no surprise then that when Sanyo announced a product for the automotive market, they targeted 10,000 cycles. The combined errors make his analysis pessimistic by a factor 26 to 154. To call this distortion or exaggeration is kind.

Professor Muller goes on from electric cars to diss hybrids:

Even the Prius suffers from battery limits, although few Priuses have been driven far enough to require battery replacement yet … Eventually the batteries will need to be replaced, which may cause the owners consternation when they learn the cost.

No reference is given for this FUD. It appears to be the author’s personal speculation. Had he done a little checking, he would have found data:

With more than 100,000 Honda hybrids on the road, the automaker told Newsweek that fewer than 200 had a battery fail after the warranty expired. That’s a 0.002 likelihood. Toyota says its out-of-warranty battery replacement rate is 0.003 percent — or one out of 40,000 Priuses — for the second generation Prius. Based on this rate, and the fact that very few of the second-generation Priuses have been driven beyond the warranty period, perhaps fewer than a dozen have had battery failures after the warranty expired. Replacement rates for the first generation Prius was closer to 1 percent.

There are number of Priuses operating in taxi fleets, and they rack up very high mileage quickly. For example, see the report, Toyota Prius taxi tops 340,000mi, dispels battery myth.

Terrorist nukes

While chapter two, “Terrorist Nukes” (in Part I Terrorism), is not about climate, an example from that chapter further illustrates the problems with the books. The primary message of the chapter is that nuclear bombs are hard to design — especially small portable ones — and terrorists are unlikely to pull it off. The damage from a portable (“suitcase”) nuke, or a larger device that fizzles, is smaller than is generally known. As a counter-point to the first point, the reader might find John McPhee’s book The Curve of Binding Energy illuminating, especially since it is founded on interviews with a person cited in Future Presidents. Professor Muller illustrates the latter point with an example of the North Korean 20-kiloton weapon that fizzled and produced perhaps a 1-kiloton explosion. (A better example might have been the U.S. W54 warhead with a mass of 68 kg.) He then posits this fizzle occurring in San Francisco and shows the expected damage on a map. He writes,

Figure 2.2 shows the sort of devastation that would occur. The inner circle identifies the blast of destruction. The outer circle shows the area that would have included substantial numbers of deaths from flying debris. The figure was calculated using the Federation of American Scientists’ online Special Weapons Primer (

Are you surprised at how small the circle is? If so, it’s because you are used to thinking about large doomsday bombs, the kind carried in our nuclear warheads and bombs. A 1-kiloton blast radius is about 450 feet. If the bomb were detonated in the middle of New York City’s Central Park, the blast would not destroy buildings outside the park. Nuclear radiation has a longer reach, though, and might reach the surrounding buildings, although it would not penetrate beyond the first row. Most deaths might, in fact, be caused by shattered glass on the periphery of the park.

The page cited above no longer gives a blast radius, but a sister page has an estimate of blast effects for 1 kiloton. It is 700m (2,300 ft), which represents and area 26 times larger than the value cited in the book. Perhaps he got his data from an older version of the page he cites. The Internet Archive’s Wayback Machine found a copy from 2005, which gives:

Ionizing radiation (50% immediate transient ineffectiveness) 500m
Ionizing radiation (50% latent lethality) 800m
Blast (50% casualties) 140m
Thermal radiation (50% casualties, second degree burn under fatigue uniform) 369m

The 140m there finally matches up with his 450 feet, but what about 50 percent latent lethality at 800m? He ignored that in the text.

Which would you want your Future President to know, the distance at which buildings are destroyed, or the distance at which people are killed and hurt? The 450 feet blast radius statistic to be cherry-picking or downplaying the data. Consider Wikipedia’s data for a 1 kiloton explosion at 200 m altitude:

Urban areas almost completely leveled (20 PSI) 200m
Destruction of most civilian buildings (5 PSI) 600m
Moderate damage to civilian buildings (1 PSI) 1700m
Conflagration 500m
Third degree burns 600m
Second degree burns 800m
First degree burns 1100m
Lethal total dose (neutrons and gamma rays) 800m
Total dose for acute radiation syndrome 1200m

It seems to me that Muller is downplaying suitcase nuclear bombs when he stresses a blast radius of 450ft when there devastating effects — acute radiation syndrome — at 8.8 times that radius (77 times the area).

If this is what are Future Presidents are being taught, we’re going to continue to have problems in this country.

This post was created for, a project of the Center for American Progress Action Fund.