Friday, February 20, 2015

Vox is wrong: the bad actor in Bakken shale is the gases, not benzene

UPDATE: Brad Plumer has made the correction. Thanks, Brad!

Vox's Brad Plumer had a nice explainer on oil trains derailing and exploding (written in light of the latest West Virginia derailment) which contained this line (emphasis mine): 
1) The newer oil is more volatile: Crude oil from the Bakken formation in North Dakota — where much of the new oil-by-rail is coming from — often contains extra chemicals like benzene that make the crude more flammable. The trains in the Lac-M├ęgantic and West Virginia accidents were both carrying crude from this region.
(Extra chemicals! N.B. there's always benzene in crude, I think.)

His source for this is (ultimately) desmogblog, which is better known as a climate change blog, I think. Here's their explanation:
...“Hazardous Air Pollutant (HAP) emissions are expected from the proposed equipment,” explains the Marquis permit. “There will be evaporative losses of Toluene, Xylene, Hexane, and Benzene from the crude oil handled by the installation.” 
Benzene is a carcinogen, while toluene, xylene and hexane are dangerous volatiles that can cause severe illnesses or even death at high levels of exposure.   
Scientific Vindication 
In a December 31 Google Hangout conversation between actor Mark Ruffalo, founder of Water Defense, and the group's chief scientist Scott Smith, Mr. Smith discussed the oil samples he collected on a previous visit to North Dakota's Bakken Shale. 
“What I know from the testing I've done on my own — I went out to the Bakken oil fields and pumped oil from the well — I know there are unprecedented levels of these explosive volatiles: benzene, toluene, xylene,” said Smith. 
“And from the data that I've gotten from third parties and tested myself, 30 to 40 percent of what's going into those rail cars are explosive volatiles, again that are not in typical oils.”
First, to a lab chemist, calling xylene a volatile is sort of odd (it has a boiling point of 140°C!), but everything is relative. When you're used to doing most of your reactions in THF (boiling point of 66°C) or say, heptane (boiling point of 98°C), then 140°C sounds pretty high. That said, EPA counts these aromatic solvents as VOCs, so that seems reasonable.

(Also, when your chief scientist has a degree in economics... I digress.)

But that said, I think there are many, many, many more volatile compounds than benzene in Bakken crude. This has been covered extensively by the Wall Street Journal -- here's a some nice examples of some of their explanations of the chemistry. From a February 2014 article by Russell Gold:
The rapid growth in Bakken production has far outpaced the installation of pipelines, which traditionally had been relied on to move oil from wells to refineries. Most shale oil from Texas moves through pipelines, but about 70% of Bakken crude travels by train.
Bakken crude actually is a mixture of oil, ethane, propane and other gaseous liquids, which are commingled far more than in conventional crude. Unlike conventional oil, which sometimes looks like black syrup, Bakken crude tends to be very light. "You can put it in your gas tank and run it," said Jason Nick, a product manager at testing-instruments company Ametek Inc. "It smells like gasoline."
Here's a July 2014 where Russell Gold and Chester Dawson say the same thing*:
Stabilizers use heat and pressure to force light hydrocarbon molecules—including ethane, butane and propane—to form into vapor and boil out of the liquid crude. The operation can lower the vapor pressure of crude oil, making it less volatile and therefore safer to transport by pipeline or rail tank car.
And yet another great explanation* by Alison Sider and Nicole Friedman:
There are geologic reasons that the new oil is particularly gassy and volatile. Over millions of years, organic material turns into a brew of hydrocarbons: crude oil, natural gas and other gas-infused liquids. The longer that fossil-fuel mixture cooks underground—in intense heat and under tremendous pressure—the more molecules escape from their source rocks and migrate to reservoirs where there is room to move around, says Scott Tinker, the state geologist for Texas. 
In those reservoirs, the oil and gas separate into less-dense gas on top and heavier crude oil below, much like a shaken vinaigrette settles into distinct layers. 
But shale rock is so dense that much less oil and gas escapes from it. The energy industry must frack shale to create tiny fissures so that oil and gas can flow out. Those minuscule pathways let only the smallest molecules rise, which is why large volumes of gas and the lightest liquids are coming out of the ground. 
In most cases, ultralight oil doesn't look like black gold. In fact, it can be as clear as water and some oil from the Eagle Ford Shale in Texas brims with so much dissolved gas that it bubbles, giving the appearance of boiling at room temperature. 
That gas makes ultralight shale oil highly combustible in a way conventional crude is not. In the past year, derailments of trains carrying light crude have resulted in spectacular blowups, including an explosion that killed 47 people in Quebec last July.
Ethane, propane and butane have boiling points of -89°C, -42°C and -1°C respectively. It seems intuitive to this chemist that they'd be far more flammable and more likely to burn and explode than benzene, toluene and xylenes. Vox is wrong, I think, and they should correct this.

(I should note that WSJ itself (and me, I guess) initially fell into this trap. I guess benzene just sounds like a bad actor.)

*How to get around the WSJ paywall -- search for the title of the article.


  1. As your quote mentions, dissolved gases are removed as easily as letting crude settle in a holding tank for a couple days. Depending on point of processing, C1-4s are so volatile they can be largely removed prior to rail transport, leaving your C4 to C6 and light aromatics as the primary volatile/flammable component. This depends on what capabilities your nearby sites have, what your customers can handle coming out the other end of the pipe, how valuable the stream components are, viscosity, etc

    But who knows? My understanding is that there has been no widely disseminated standard process, which is why we're seeing this discussion.

  2. The definition of Volatile Organic Compound is not at all straightforward. The EPA has several definitions depending on context, indoor vs. outdoor maybe others. One is a definition that has to do with compounds that react photochemically with car exhaust (nitrogen oxides) to generate ozone. This list does not include compounds that have been specifically exempted by 40 CFR 51.100(s) due to negligible photochemical reactivity. This includes compounds that any chemist would call a VOC like methylene chloride, methyl chloroform and acetone. This list of exemptions also has nothing whatsoever to do with health and safety. I discovered this some years ago when I bought some paint from Sherwin Williams that claimed to be VOC free. Something like this Upon opening it I noticed it smelled like hydrocarbons, so I looked at the documentation they are using 40 CFR 63.801(a) definition of VOC which is the EPA ozone generation definition. What is in it? Glycols maybe, long chain hydrocarbons, esters. Who knows? Theoretically you could put methylene chloride in there and label it VOC free. I'm sure there are a lot of people who would think it is safe to inhale based upon labeling.

    1. I worked a lot with pheromones a long time ago, and in the lab we called anything with around 15 carbons 'volatile' since most of the pheromones would have just a hydrocarbon chain, with chiral methyl groups thrown in, or a ketone, chiral alcohol or epoxodie, plus a double bond somewhere. That meant that it would spread in air pretty efficiently and insects could detect them from far away. At the neighbourhood of 25-30 carbon atoms, you got to the 'contact' pheromones where the insect had to tough a surface with them. Often you need both as the insect uses the latter set to know that it was following the right trail in the air.

      Anyways, by that definition, xylene, which has low surface tension and no hydrogen bonds, and thus probably has a high vapor pressure at 80 degrees below boiling (relative to polar stuff with hydrogen bonds such as water), is definitely volatile.

  3. I agree that the hydrocarbons are likely the culprits for increased flammability than the aromatics.

    There are several definitions for VOCs, but the most common classification when journalists are using it as a term comes from the air quality area and is based on vapor pressure, not boiling point. While there is a relationship between the two, they are not 100% correlated. While hydrocarbons are more volatile than the aromatics, both are considered VOCs because their vapor pressures are high enough to make them evaporate completely if given the chance.

  4. "[...] unprecedented levels of these volatile explosives: [...]"

    The argument about whether aromatics qualify as volatiles aside (mostly because it's capably put to bed here), "unprecedented" is maybe applicable if you focus just on the oil typically found in the south SK/MB and ND fields. Scott Smith would soil himself if he ever saw an analysis of the wonderful tarry heavy oil we produce in AB.

    Frankly, even though I work in the industry, I think the whole shale fracking boom - including how the production is subsequently handled - is on extremely shaky ground, ethically-speaking.

    (Pun intended.)

  5. The customer had no choice in telecommunications alternatives and for quite many years had to put up with exorcist costs as well as the most serious customer alternatives you can think of. Somewhere along the way somebody saw light and Telkom was deregulated on 1 Feb 2005 which started out up the market for other telecommunications companies.

    Vox Mobile Price In Pakistan