RE: Berkshire Energy Laboratory - Weekly Update

To: Noa

From: Tom

Date: September 26th, 2008

RE: Berkshire Energy Laboratory - Weekly Update

Summary

·  Found an academic paper from Japan that supports Bio-Petrol’s results. Waiting on a paper that should give some idea of the oil composition in terms aliphatics and aromatics.

·  Gathered information on local treatment plants and sludge handling processes. Plant visit tomorrow. How much sludge is really available to us locally and regionally?

·  The big questions for BEL are around validating the yields and quality of sludge liquefaction oils and advancing processing and upgrading methods.

·  FT Synthesis is has a number of process inputs that impact product quality and research would be directed toward process development and scale down.

Sludge to Diesel

Last week in our discussion I noted that Bio-Petrol has promised to forward test results of their fuel oil so that we can determine the composition in terms of the number of aliphatic hydrocarbons, aromatics and contaminants. They provide an elemental break down but claim only that the fuel oil is 88% aliphatics. I also mentioned that the academic literature is a bit sparse on sludge liquefaction. Bio-Petrol has not yet come through but this week I did have a bit more success finding some research papers that help characterize liquefaction products among other things. They’re posted here on the Berkshire Energy Page I created .

http://www.adktroutguide.com/greentogreen.html

Unfortunately, the most specific paper on the subject (“Analysis of Oil Derived from Liquefaction of Sewage Sludge”, Fuel, vol 71, issue 9, September, 1992) is not available electronically so I’ve requested it through interlibrary loan.

I also found that researchers in Japan built a 5 tpd pilot plant back in 1994, synthesizing a fuel oil with a heating value about the same as Bio-Petrol (85% of diesel). They’re energy balance suggest a yield of 1.5 tpd oil (466 gals or 11 barrels) from 60 tpd of dewatered sludge. For comparison, Bio-Petrol claims to get about 2 tpd from the same size plant.

Additionally, I’ve found some work around sludge gasification (H2, CO), liquefaction in solvents and a few other related topics.

To ground some of this research I spent some time characterizing waste water treatment and sludge removal in New York State. According to the NYS DEC the state has granted licenses to about 610 facilities to treat waste water, approximately 40 of which are within a 50 mile radius of Saratoga. There are a variety of treatment methods in use for the biosolids that we’re interested in. About half apply some method characterized as beneficial (land application, composting, pellitization and chemical stabilization(add lime to make a fertilizer)). The rest either incinerate or landfill it. The plant I’m visiting tomorrow is in Fort Edward and has a composting facility on site. I’ve contacted some of the other plants nearby to find out what they’re doing with their sludge.

Last week we also discussed what big questions we want to answer with our work at Berkshire. We’ve condensed on the opinion that converting sludge to diesel has tremendous potential in the current economy and we’ve been focusing on hydrothermal liquefaction as the optimal processing route. The first big question I have is

·  Is hydrothermal liquefaction the best processing route to convert sludge to oil?

o  My research tells me that it is but another route that has been studied is gasification (wet and dry) and in general I’d like to compare the energy and financial economics of these routes and other feed stocks.

·  What yield of oil can we ACTUALLY achieve on the bench and in pilot scale?

o  We can gather some information from the literature but some process will have to be physically duplicated.

·  What solvents and catalysts are ideal and which can lead to patentable process designs?

·  What is the highest quality oil that can be achieved in a one stage process and is it actually possible to upgrade sludge based oil to transportation grade and what are the yields?

·  What new reactor design concepts can be modeled and tested to improve first pass yields?

Wood to Diesel Research

I worked this week to advance my analysis of the FT process to the point that I can make some recommendations for project work at BEL. Because the reaction mechanisms are not well known, modeling of the FT synthesis is conducted using probabilistic methods. They start with the assumption that some estimate of the probability of carbon chain growth is available (α). In other words, for a given set operating conditions there is a finite probability that C-C bond will form continuing chain growth, versus the opposite probability that a C-H bond will form and terminate chain growth. That probability estimate is itself the subject of intensive research however one of the better papers relates it to the relative fraction of hydrogen and carbon monoxide in the syngas, and the reaction rates of hydrogen with C1, C5 and C6 bonds (known values).

The figure indicates that the maximum production of a gasoline/diesel fraction corresponds to α = 0.7 to 0.9. This is equivalent to a H2/CO ratio in the syngas of between 2 and 2.5. I caution here that this gives only an indication of direction and that a number of other factors are involved.

The reaction rates of hydrogen with C1, C5 and C6 bonds are used as constant values in the chain growth probability estimates but in reality are functions of temperature and other things. FT synthesis is carried out at either low temperature (200 – 240C) on cobalt catalysts or high temperature (300-350C) on either iron or cobalt catalyst. The lower temperatures favor the longer chain hydrocarbons and the higher temperatures the shorter. There is significant cross over however. In addition to temperature, chain growth is effected by pressure, catalyst type, syngas quality, degree of catalyst fouling and the reactor design itself. There are other considerations as well including undesirable side products (olefins, alpha olefins) leaving the research opportunities wide open.