The U.S. Department of Energy’s Bioenergy Technologies Office (BETO) hosted a two-day webinar series, Biocarbon Incorporation into Transportation Fuels via Co-processing in Refineries, highlighting key takeaways from the Bio-oil Co-processing with Refinery Streams project. Below is a transcription of, “Webinar Day 1: Co-processing Fast Pyrolysis Bio-oils and Hydrothermal Liquefaction Bio-crudes in Fluid Catalytic Cracking and Hydroprocessing in Refineries,” held on September 20, 2023.
.....
Simone Hill-Lee, BETO
Good afternoon. This Zoom call is being recorded, and may be posted on to BETO’s website, shared, or used internally. If you do not wish to have your voice recorded, please do not speak during the call or disconnect. Now, if you do not wish to have your image recorded, please turn off your camera or participate only by phone. If you speak during the call or use a video connection, you are presumed to consent to recording and to the use of your voice or image. We're going to start our program. We'll begin shortly, but just before we do so…
We have a few housekeeping reminders for those of you in the audience. As you'll notice, this event is set up such that you will not have the ability to unmute or turn on your camera during today's webinar. You're encouraged to submit your questions using the Q&A function, you should see an icon at the top or bottom of your screen, including, if you're having any technical issues, please feel free to submit a question, using that as well. The chat will be disabled, so the primary way of communication is the Q&A function. You are welcome and invited to submit your questions at any point during the presentation today. Every effort will be made to try to address all of the questions that are submitted. So, with that, I would like to turn it over to my colleague, Robert Natelson, to get us started.
Robert Natelson, BETO
Thanks, Simone. Today's webinar will be on Co-processing Fast Pyrolysis Bio-oils and Hydrothermal Liquefaction Bio-crudes in Fluid Catalytic Cracking and Hydroprocessing Units in Refineries.
First, I'll provide some brief overview of the funding agency, and then you will hear an overview of the overall co-processing project and then individual talks from the laboratories on co-processing strategies. And if you have the time and availability, please join us again next week on the other part of this co-processing project on biogenic carbon tracking and measurement.
The work presented today is funded by the U.S. Department of Energy Bioenergy Technologies Office (BETO). BETO is an applied research and development office that partners with the DOE national laboratories, academia, nonprofits, and industry to support pre-commercial research. BETO’s mission focuses on reducing greenhouse gas emissions in the transportation and chemical sector by using the nation's biomass and waste resources. BETO’s priority is efficient and scalable technologies for low-carbon intensity fuels, and other chemicals and sustainable aviation fuel. This updated strategy has been published in multiple documents in the last year, including the 2023 multi-year program plan, the Sustainable Aviation Fuel Grand Challenge Roadmap, and the US National Blueprint for Transportation Decarbonization, all documents widely available online from the BETO website. BETO also supports the Billion Ton study. I'm sure many in this audience are familiar that there's been rapid growth in the last decade in the industry in the use of vegetable oils and biogenic fats, oils and greases for the production of renewable biofuels, often done in collaboration at petroleum refineries, with configurations such as co-processing with petroleum units. But conventional wisdom is these feedstocks will not be enough in coming years. New feedstocks in conversion technologies need to be unlocked to meet the demand for low carbon intensity fuels and products.
BETO is organized into four program areas, renewable carbon resources, conversion technologies, systems development and integration, and the cross-cutting data modeling and analysis. The work presented today is from the systems development and integration program. BETO’s four programs work together to collaborate on moving forward is the goals of the BETO office, including supporting the production of three billion gallons of SAF by 2030, supporting the production of thirty five billion gallons of SAF by 2050, and supporting pathways for that SAF, as well as other strategic fuels and chemicals in methods and processes that are cost-effective and with at least seventy percent greenhouse gas reductions compared to conventional processes.
Our first presenter today will be Dr. Reinhard Seiser from the National Renewable Energy Laboratory or NREL. Dr. Seiser received his M.S. and Ph.D. degrees in chemical engineering at the University of Graz, Austria. He subsequently worked as a researcher at the University of California, San Diego, in the fields of combustion and fuels. He joined NREL in 2020 to work on pyrolysis and catalytic pyrolysis of biomass and co-processing of these liquid bio-intermediates with vacuum gas oil in the laboratory scale fluid catalytic cracker. Dr. Seiser's talk is on co-processing fast pyrolysis bio-oil and catalytic fast pyrolysis bio-oil in the fluid catalytic cracking unit.
Reinhard Seiser, National Renewable Energy Laboratory
Thank-you. I will start by giving an overview of the co-processing project, and I will do that by first talking about the feeds we are considering. On the left, you can see relatively dry biomass, such as forest waste, or other lignocellulosic biomass, such as agricultural residues, and they would be converted by either fast pyrolysis or catalytic fast pyrolysis to bio-oil, and then that bio-oil would be shipped to refineries and co-processed in either fluid catalytic cracking or hydro-processing. On the right, you can see a second feedstock that would be mostly wet biogenic wastes, for example, slurries from wastewater treatment sludge or from algae, and they would be converted to a bio-crude by hydrothermal liquefaction, and then the resulting bio-crude would be co-processed in refineries in hydro-processing.
This is an example of a petroleum refinery, and as you can see, it contains many unit operations to convert crude oil into final fuels and chemicals, and this graphic doesn't include a lot of other infrastructure, such as the production of hydrogen, steam, or the treatment of wastewater or exhaust gases. So, there's a lot of infrastructure and knowledge at refineries that one can take advantage of by co-processing, and in this case, we are targeting the fluid catalytic crackers, hydrotreaters, or hydrocrackers, and the current US capacity of those units is in excess of one hundred billion gallons per year. So even co-processing biogenic feeds at five percent would amount to billions of gallons of biofuels that could enter the transportation sector.
And a special point, especially in those units, that I mentioned, for biofuels, fluid catalytic cracking and hydrotreating and hydroprocessing. And what we want to do is, we want to look beyond fats, oils, and greases. Those are examples of biogenic fuels that are currently already co-processed by some refiners. But we are more interested in the lignocellulosic types of feeds like forestry wastes and agricultural residues, and algae that we see as having a much larger potential resource.
When one looks at the properties of the bio-oils like fast pyrolysis oil or catalytic fast pyrolysis oil, or bio-crude from hydrothermal liquefaction, one can see that the composition is quite different from petroleum. So, for the pyrolysis oils it's mostly the high oxygen content, and for the bio-crude it's the high nitrogen content, and all those bioliquids contain water. So, this is something that the refiners need to take in account when co-processing those liquids.
And from this we derive our challenges, and that is that there is not a lot of co-processing data available for those types of biofuels, and they can also upset processes and catalysts, and it's especially important for refiners to prove that bio-carbon from these bioliquids makes it into the transportation fuels.
And from this we derive our objectives. We basically would like to study different biogenic liquids in both FCC and hydroprocessing and look at the impact on processes and catalysts, and develop models for analysis, and tools for the measurement of biogenic carbon incorporation.
The project has been a collaborative effort between the three national labs, where we have looked at the production of the feeds, the co-processing, as well as the analysis of the products. We have collaborated with external partners, as well as an Industrial Advisory Board, where we have received important guidance and feedback.
I am now going to talk specifically about the co-processing of bio-oils in fluid catalytic cracking, and I would like to start with an overview of the different bio-oils.
For reference, I’ve shown fast pyrolysis oil here on the right, and you can see that it contains the most oxygen and is the furthest away from transportation fuels. So, it would really have to be processed quite a lot to get to over here, and especially that’s true for jet fuel that cannot contain any oxygen at all. And so, other pathways that we are considering are to upgrade fast pyrolysis vapors to catalytic fast pyrolysis oils. This is done with a catalyst, either in a fixed bed or in a fluidized bed, and you can see that the oxygen content is lowered, and this also improves other qualities, such as viscosity and water content. And this depends on the severity of the upgrading as well as the catalyst. And then another bio-oil we’re considering is a stabilized pyrolysis oil that we have received from BTG. This is an upgraded oil that was produced from fast pyrolysis oil via mild hydro-treating, and it has also improved properties due to the higher hydrogen content. So, for the stabilized pyrolysis oil, the high production yield is an advantage, while the catalytic fast pyrolysis oils have pretty good other properties and low oxygen contents; and there is also a number of references to some of our and others’ publications on the production of those types of oils.
This is an example how these catalytic fast pyrolysis oils could be produced. This is something we have available in-house. We could first produce pyrolysis vapors from a fluidized bed pyrolyzer and then those pyrolysis vapors, without condensing them first, would go directly into a circulating fluidized bed, and they would be upgraded by zeolite-based catalysts to produce low oxygen containing catalytic fast pyrolysis oils.
And on this slide I'm showing a summary of the characterization of those different types of oils. Again, I'm showing here fast pyrolysis oil as a reference, and then you can see, as we go to more severe upgrading, using zeolite catalysts, the resulting catalytic fast pyrolysis oils decrease in oxygen-containing groups as shown in some of the blue and green bars, and an increase in aromatic groups is shown in the yellow bars. Going the other direction, by mild hydro-treating fast pyrolysis oil, the stabilized pyrolysis oil, as you can see, reduces in aromatics, and increases in aliphatics.
Another method to improve the quality of bio-oils is to make sure there are no particulates left over from the pyrolysis step, because particulates from char or ash could cause polymerization of those oils when heated, and that could cause build up or plugging. This is by either acting as inception points or as catalysts. And so, it's really important to keep particulates out. Our solution, for many years, has been using hot-gas filtering right after the pyrolysis vapors are produced. So then, we're taking out the fine particulates right in the beginning, and this has generally served us well for producing good-quality oils. Another additional opportunity is to use the same filter housing by putting a catalyst right after the filter. Then we can do this in the same vessel by adding an additional mild upgrading or other chemical conversion steps.
I'm now going to co-processing of those biofuels. You can see here an example of a pilot-scale FCC reactor. This is licensed from GRACE and is our pilot plant here at NREL; and the purpose of fluid catalytic cracking is to reduce the size of large molecules into smaller fuel-range molecules. That is mostly gasoline, but also some jet fuel and diesel are produced in the process. We have added additional capabilities to our Davison Circulating Riser, or short “DCR”, by adding an additional cyclone after the stripper and additional condensation steps for collecting the liquid product, and a modified nozzle, so that we can co-feed bio-oils with vacuum gas oil. Those two are not miscible, but they are introduced through the same nozzle, and because of the activity of the bio-oils, these oxygenated streams can be challenging when feeding them into a high temperature reactor, such as the riser. And the riser here is the main reactor where the catalytic cracking happens, and it's operated around 521c.
These are some of the results of one of our campaigns, where we have looked at the three different bio-oils, that I’ve introduced earlier, at a five percent co-processing level with VGO. On the on the upper right, you can see the main product fractions compared to VGO only, and there is really not a significant change in the product compositions at this 5% co-processing level. But you can see that the bio-oils, because of the oxygen content, do have an increased aqueous fraction, and also slight increase in the coke fraction. We've also analyzed the liquid products for carbon 14, and the results are shown here. By putting the biocarbon measurement of the product in relation to the feed, we can compute a biocarbon yield that's shown here in the green bars, and that represents how many percent of the biocarbon in the feed ends up in the liquid. For comparison, in the gray bars, that would be how much of the total carbon in the feed ends up in the liquid product. So, the larger the green bars, the better the biocarbon incorporation, and you can see immediately that for the different oils, it's really either the one with the high hydrogen, the stabilized pyrolysis oil, or the catalytic fast pyrolysis oil with lower oxygen that show better biocarbon incorporations, compared to the CFP oil with the higher oxygen content. And we have done this for two different catalysts, and we have consistently found that when we add Johnson Matthey’s CP758 catalyst to E-Cat, that the bio-carbon incorporation was slightly improved compared to E-Cat.
And we have looked also at the mechanism for that biocarbon incorporation by doing an analysis with a carbon 13 labeled wood in a micro-pyrolyzer. Here you can see here how this works. We have obtained a purpose-grown oak stem that contains more than ninety seven percent carbon 13, while vacuum gas oil from petroleum mostly contains carbon 12. And those two get mixed in the micro-pyrolyzer, and then the wood pyrolyzes into a bio-oil, and then both of the vapors get sent over a catalytic bed, and with a mass spectrometer the vapors can be measured. That provides a good method of distinguishing between the molecules that are formed from either the fossil or the biogenic parts. And shown as an example of a product would be here toluene. If you only process VGO, the fossil fuel, you will see the signal on the mass spectrometer, as showing the carbon 12 origin. And if you look at only oak, you will see a toluene molecule containing carbon 13; and then, when you flow both VGO and oak over E-Cat, you see some of some of each: some of the fossil, and some of the biogenic toluene, but not much in terms of mixed molecules. But when we use the mixture of VGO and oak over the CP758 catalyst, you can really see all kinds of toluene isotopes ranging from one, two, three, four, five, and so on carbon 13 atoms inside the toluene molecule, and that shows that there is a strong interaction of both the biogenic and the fossil feeds. And interaction is important. Mostly, because that allows to direct some of the biocarbon into the liquid fraction versus going to tail gas or coke like commonly happens if you process bio-oils by themselves or independently of VGO.
And we have also routinely analyzed our products from co-processing, and we have several in-house methods here available because of the fuels containing oxygen. We're specifically looking for oxygenates in the products using a variety of methods, but at the smaller co-processing ratios it's difficult to see any remaining oxygenates. But one method that we have found worked well is that we first fractionate our products into different boiling point fractions shown here, and then analyzing those separately, for example with GC-MS. And you can see here that in the jet fuel fraction, there are phenols, carbonyls, and furans, and those kind of oxygenates would have to be converted in the subsequent jet fuel hydrotreater, which they usually are.
We have also performed techno-economic analysis on the co-processing in FCC. And our team has looked at various aspects and computed a minimum fuel selling price for co-processing those bio-oils. The analysis includes all kinds of by-products that are formed, such as light olefins and LPG. And collecting all the product fractions and using our yield structure that we have experimentally measured and are coming up with an assessment of the price. And we found that the results show promising improvements over time by improved measurements, improved methods, improved technology, and some of the details of this are shown here in our publication. And we are also investigating opportunities for the production of sustainable aviation fuels, because that's DOE's goal. And so, we're looking either at including heavy or cracking of heavier fractions, or operating FCC differently to increase the jet fuel fraction.
As a summary, we have produced several different bio-oils, using catalytic fast pyrolysis, with oxygen contents ranging from twelve to thirty six percent. We have then co-processed some of those bio-oils between one and ten percent, and we found that at five percent we get quite good results and a good biocarbon incorporation. And in principle all of the three bio-oils have shown to work well for co-processing. Except maybe the CFP oil with the high oxygen content that has a lower biocarbon incorporation, and that has a little bit more operational challenges when feeding. But in principle, all three bio-oils work well, and this gives refineries enough flexibility to choose the amount of co-processing or oxygen content of the bio-oil, depending on their operational conditions and considering some of the effects, like derating of the equipment or production of additional waste water.
And on my last slide I'm showing here the future work that we're looking at. So we would like to continue to use FCC for difficult-to-process streams like lignin, and looking at a variety of feeds, and how they would interact with each other; what kind of synergies they have, and over time we would also want to push the co-processing more towards one hundred percent and eventually only have a variety of biogenic feeds; and we would like to continue with analyzing the product fractions and track biocarbon for this.
And with that I'm handing it back to Robert.
Robert Natelson
Thank-you, Reinhard.
Our next talk will be on Co-processing of HTL Bio-crudes and Bio-oils and Hydroprocessing.
Our second speaker today will be Dr. Huamin Wang from Pacific Northwest National Laboratory.
Dr. Wang is a chief research engineer working in the Advanced Energy Systems group at PNNL. With a Ph.D. in physical chemistry, experience and postdoctoral training, Dr. Wang joined PNNL in 2011. His current research focuses on the innovative catalyst and process development for generating renewable fuels and chemicals from biomass and waste and the fundamental understanding of the catalytic reactions integral to those processes. Additionally, Dr. Wang holds a joint appointment as adjoint professor at Washington State University.
Huamin Wang, Pacific Northwest National Laboratory
Thank-you, Robert.
Today we'd like to also share with you the work in co-processing bio-oil and bio-crudes in hydroprocessing. As shown here, the hydro-processing involves a set of the unit operation in refinery and including hydrotreating and hydrocracking. In general, hydrotreating is doing the work of removing heteroatoms like sulfur, nitrogen, oxygen to really produce a clean product. While hydrocracking is converting heavy gas oils into lighter fuel blends by breaking the C-C bonds. For hydroprocessing we use hydrogen to do this chemistry, therefore, to prevent the carbon rejection. For hydroprocessing, we normally use a fixed bed operation, and therefore require long catalyst lifetime, and use high pressure hydrogen, and consuming hydrogen as well. Considering the similarity of the chemistry requirement, and also the product properties of biofuel, hydrotreating and hydrocracking can co-process bio-oils and bio-crudes to enable biofuel and biogenic carbon incorporation into those products.
Reinhard has already told you a lot about fast pyrolysis and catalytic fast pyrolysis. Here, I'd like to introduce hydrothermal liquefaction which can transform a way to waste into liquid products. Hydrothermal liquid factory is a conceptually simple process, and it can operate continuously. Basically, you feed your wet feedstock, such as wastewater sludge, manure, or food waste to a hydrothermal liquefaction reactor, which is a heated pipe, at a subcritical water condition. You can convert those slurry into a bio-crude and other byproducts. Sixty percent of carbon from the feedstock stays in bio-crudes. This bio-crude can go through hydroprocessing to remove heteroatoms, such as sulfur, nitrogen and oxygen, and then produce fuel products with more than ninety five percent carbon efficiency. Overall the hydrothermal liquefaction and hydroprocessing has high bio-carbon yield to liquid products and can tolerate dirty and wet feedstocks. And another benefit is that in the United States only through hydrothermal liquefaction we have potential to produce six billion gallons per year of fuel in the United States only. And also, hydrothermal liquefaction is an alternative process to dispose those wastes. At PNNL we have a hydrothermal liquefaction process development unit, a PDU, which can process slurry at twelve to eighteen liter per hour capacity. All the feedstocks used in this research are produced at PNNL in the PDU using real world feedstocks.
This is how we are doing the study on co-processing of the bio-oils and bio-crudes in hydroprocessing. We're working on co-processing bio-oils and HTL bio-crudes, here specifically a sewage sludge HTL bio-crude, at two to twenty percent of blending level with different petroleum feedstock, like VGO, straight run diesel, or kerosene, or fuel oil. We're using commercial catalyst extrudates in a lab-scale reactor. We're doing feed analysis. We monitor hydroprocessing performance at a more than three hundred hour time on stream with the steady-state operation. We carefully look at what happened to our catalyst and also conduct fuel analysis of the produced products.
We are working on co-processing of both fast pyrolysis bio-oil and hydrothermal liquefaction bio-crudes. And in some cases, we also tried to co-process a hydrotreated bio-crude with a target of removing nitrogen to meet the SAF requirement. We are co-processing them with different refinery operation units, like diesel hydrotreating, VGO hydrotreating or VGO hydrocracking and other unit operation as well. I will give you more detail about the issue of, or challenge of co-processing of fast pyrolysis bio-oil in hydrotreating, and also, I'll tell you that how CFP bio-oil will be less challenging when co-processing in hydrotreating and hydrocracking. Also, I will tell you later how nitrogen is an issue, and also something needed to address for bio-crudes, which really give a significant challenge for processing a bio-crude in diesel hydrotreater. However, after we manage the nitrogen issue, we can co-process the bio-crude with VGO in hydrotreating and hydrocracking which normally operate at a more severe hydroprocessing condition. We also tried to co-process the bio-crude with fuel oil, with a target of incorporating bio-crude to fuel oil for marine application. We found that it will enable nitrogen and sulfur reduction from both bio-crude and fuel oil to enable to produce ultra-low sulfur fuel oil. However, we do have some issue of fuel homogeneity, which is something ongoing and we're not going to talk more about it today. We tried to co-process hydrotreated HTL bio-crude. I'll show you later that bio-crude has the nitrogen challenge and SAF will require below 2 ppm nitrogen in their product. Therefore, we're trying to utilize the co-processing the hydrotreated bio-crude, especially its jet fuel fraction, with kerosene and to utilize synergy between them to enable a hydrodenitrogenation to meet below one ppm nitrogen in final a co-processed product to meet the requirement for SAF. In general, we learned that the feasibility of a co-processing bio-oil and bio-crude with different petroleum stream in hydroprocessing greatly depends on their heteroatom content and speciation. I will use my next couple slides to really dive into these co-processing scenarios.
Let's first look at the fast pyrolysis bio-oil. As we all know that fast pyrolysis bio-oil is not stable and direct co-processing raw fast pyrolysis bio-oil in hydroprocessing really lead to reactor plugging and, therefore, we have to stabilize bio-oil first. Through the hydrogenation approach developed at PNNL we were able to stabilize the bio-oil and enable stable coprocessing with VGO over several hundred hours’ time on stream. What we learned is that when we co-process this stabilized bio-oil we do see a minimal impact of bio-oil on the reaction of VGO and we do see the simultaneous conversion of bio-oil and VGO with reducing oxygen and sulfur. However, if we look more closely that there is some issue for the coprocessing bio-oil with VGO and diesel as well. Here, we normally start with VGO only to reach the steady state to establish the performance. Then we introduced bio-oil to the same reactor at maintained conditions to really check what happened after we introduced bio-oil into the system. As showing here, after the co-processing of twenty percent of fast pyrolysis bio-oil with VGO, we consume more hydrogen, as expected, and form a lot more gas, a lot more water, and also biogenic carbon incorporated into the fuel range. Based on the mass balance and also biogenic carbon measurement by isotope measurement, we find that seventy percent of biogenic carbon in bio-oil ended up to be in a fuel range. All of the other missing carbon is in the gas range. It really suggested bio-oil co-processing leads to challenges like high water formation, relative low biogenic carbon incorporation. There is unclear long-term impact on catalyst stability.
Now, we use catalytic fast pyrolysis bio-oils, as Reinhard mentioned that it significantly increased or enhanced the quality of the bio-oil produced. By using catalytic fast pyrolysis, we were able to incorporate ninety percent of carbon in CFP bio-oil into fuel products as shown here at varied bio-oil blending ratio. It always gives more than 90% biogenic carbon incorporation into the products. Of course, this incorporate will largely depend on the oxygen content. As showing here, this specific example is given by a relatively low oxygen content bio-oil. If you look at the product in more detail in diesel fraction and gas oil fraction after VGO hydrotreating, we learned that co-processing CFP bio-oil will have negative impacts on the diesel fraction quality by a reduced cetane number. Also, we see a significant biocarbon content in diesel fraction, which represent roughly fifty four percent of the total biogenic carbon incorporated into the system, which suggests that biogenic carbon incorporated more selectively into diesel range in the product. Similarly in gas oil fraction, we do see a maintained nitrogen content, and also the increased aromatic and also a relatively small incorporated biogenic carbon to this fraction. Similar to the fast pyrolysis bio-oil, we believe that there still is a challenge in co-processing CFP bio-oil including water formation, reduced diesel fuel quality, and also unclear long-term impact on catalyst stability.
Now let's switch to the bio-crude coprocessing with VGO and other petroleum feedstocks. The bio-crude, especially showing an example here the bio-crude derived from wastewater sludge. Now we see the major characteristic of this bio-crude is the high nitrogen content, which is as high as five percent. Oxygen and sulfur are relatively low compared to fast pyrolysis bio-oil. We know that for hydroprocessing and hydrocracking how nitrogen, especially basic nitrogen, can cause the inhibition for the reaction for heteroatom removal and hydrocracking. Therefore, for this high nitrogen containing bio-crude we have to go through a two step or two stage process to use the hydrotreating at the first step to knock down nitrogen and then to do the mild hydrocracking or deep hydrocracking, which I will show you in later slides. By doing that, we can remove nitrogen largely by first step and therefore enable the hydrocracking at the second step. If we do a detailed analysis, including mass balance and biogenic carbon measurements, we find that for the co-processing bio-crude we can achieve ninety five percent, more than ninety five percent of biogenic carbon incorporation into fuel products. More closely, we can look at the diesel fraction and we see an increased cetane number, suggesting an increased fuel quality, and maintained sulfur and nitrogen content of diesel fraction. Similar to the CFP bio-oil, we do see that the majority or 60% of biogenic carbon in bio-crudes are incorporated into diesel fraction which is really desired. As we see that the major challenge for co-processing bio-crude is that we have nitrogen containing species that we have to take care of them and also we have to maintain catalyst stability. I will spend next two slides to really talk about this nitrogen containing species and also catalyst stability.
If you look closely of the bio-crude, it will show a lot of a nitrogen or sulfur containing species. For some nitrogen or sulfur containing species which can also be seen in petroleum feedstock. However, for bio-crude, it will introduce some very unique nitrogen and oxygen containing species such like amide. The chemistry of this specific molecular is not known under the hydrotreating conditions. We spend a lot of time to really figure out the reaction kinetics and mechanism of those molecular. At the same time, we measured all the kinetics of all those different kinds of oxygenates and nitrogen containing species, together with the sulfur containing spaces that we can see in VGO to understand the chemistry and also kinetics. In general, no surprise that the hydrodenitrogenation is slow and we have to really take care of those kind of difficult nitrogen containing species introduced by bio-crudes. With that, we were able to generate all the kinetic and energetic data for all the different major components in bio-crudes and compare it to VGO. With these data, we were able to develop a kinetic based reactor model for co-processing. We use existing VGO hydrotreating kinetics and we use kinetic data we generated to have a bio-crude only model. We combine both model by using our co-processing data to tune the kinetic parameters that enables a validated kinetic-based co-processing hydrotreating model. We hope we use this model to provide protective capability and also condition for optimization for reactor configuration and operation.
Now let's look at how bio-crudes could potentially cause catalyst deactivation. We co-process a raw bio-crude with VGO to simulate that what can go wrong at a long-term co-processing scenario. We learned that after the run, we carefully cleaned the catalyst to remove all the possible absorption species and learned that catalyst actually maintained only half of this initial activity, measured by the kinetic measurements, suggesting that there is the irreversible catalyst deactivation. We conducted detailed catalyst characterization to understand what happened. There are two major things, one is that the inorganics in bio-crudes will be deposited on the external surface of catalyst extrudate and coat the catalysts. At a longer run, it will eventually cause a pressure drop increase in the reactor. And also, we learned that in atomic level that the bio-crudes contain some reactive oxygenates will form a special coke and it will deposit near the catalyst site, selectively block certain active sites, and actually shutting down one pathway for hydrotreating chemistry. We learned that all those catalyst deactivation mechanisms, we were able to mitigate this by doing bio-crude pretreatment, to remove some of the problematic species, and also use guard beds. By doing that, we can maintain the catalyst activity. As shown here, that after you use the catalyst, it shows similar activity as the fresh or diesel only catalysts. Of course, we are studying this deactivation in hundred hours’ time scale. Longer term impact still requires further investigation.
So, with nitrogen management in bio-crudes, we were able to conduct deep hydrocracking of VGO together with bio-crudes using a zeolite containing catalyst which is known to be sensitive to nitrogen containing species. We were able to vary the reaction conditions to increase or to tune the hydrocracking conversion. As shown here, we can monitor, by measuring biogenic carbon content by isotope method. We can see the biogenic carbon distribution at the different fractions and at different hydrocracking conditions. Generally, we learned that biogenic carbon is largely incorporated into the mid-distillate range, including jet and diesel. Also, bio-crudes are less sensitive than VGO on the hydrocracking severity.
We also conducted an economic analysis to understand what's the benefit of co-processing to biorefinery, especially for hydrothermal liquefaction process. What we learned that at different scenarios in general co-processing will give us a significant reduction of bio-crude upgrading cost compared to stand alone hydrotreating. Also, we conducted refinery impact analysis of co-processing to understand the benefit to using refineries. What we learn is that the model break even value of the HTL bio-crude will be greater in certain cases than the model minimal bio-crude selling price, meaning that refinery can also benefit from processing bio-crudes.
In summary, I hope I show you today that co-processing bio-oil and bio-crudes have great potential in hydroprocessing. We demonstrate a high biogenic carbon corporation through coprocessing high-quality CFP bio-oil and HTL bio-crude in hydroprocessing. We have to manage the competition between heteroatoms introduced to the reactor, especially the nitrogen introduced by HTL bio-crude. We need to ensure we focus on hydrodenitrogenation to enable co-processing in hydrocracking. We developed kinetics and reactor model, and also understand the deactivation and provide the mitigation strategy. And lastly, we showed that the co-processing can offer benefit to both biorefinery and refinery. Based on what we learned, we would like to continue focusing on catalyst stability over longer duration and larger scales. We’d like to establish feedstock specification, especially on contaminants and other factors. Also, we’d like to understand the impact of the co-processing beyond the catalytic reactor. And lastly, we would like to increase the blending level to incorporate more biogenic carbon into fuel products.
With that, I would like to acknowledge the great team working on this project in the last couple of years, the support from BETO office, and also the people who are working on this project from different angles from NREL, PNNL, and LANL. I would like to specially acknowledge our collaborators from industry for their great input and sometimes providing materials to help this research. With that, I would like to send back to Robert. Thank you.
Robert Natelson
Thank-you, Huamin. We have some time left for some Q&A. I think I'll start with some questions here towards Reinhard, since he had a little bit of a break. What are the high heating values? The HH values of the pyrolysis oils that you investigated?
Reinhard Seiser
Fast pyrolysis oil is the one that would have the lowest heating value, because it is almost fifty percent oxygen, and from that you can calculate the heating value, and it will be approximately half so, probably around twenty megajoules, per kilogram, instead of forty-two for hydrocarbon, and then, as we get to higher quality, CFP. Or else, like we have made CFP oils down to twelve percent oxygen. Yeah, you can assume that the heating value would gradually go from twenty towards forty megajoules per kilogram. So basically, you have a choice of how much you upgrade your bio-oil to get a better heating value. And another problem that we also see is, let's say you process fast parallels with a lot of oxygen, generally, that oxygen takes away some of the carbon, and it goes to CO or CO2, so you're not getting the full benefits of the carbon going into liquid fuels.
Robert Natelson
Here's another question. I'll send this for Reinhard regarding the amount of water in the bio-crudes. Have you considered any way to reduce it before getting to the coprocessing unit to avoid possible corrosion in salt permission?
Reinhard Seiser
Yeah, I think that's more for Huamin. I mean, maybe because I think he's dealing more with the water in the bio-crudes.
Huamin Wang
Yeah, Robert. The water in bio-crudes would depend on feedstock and also depends on our processing condition can vary and also depend on pretreatment or bio-crudes post-treatment after hydrothermal liquefaction. It's in the range of two to ten percent. Of course, bio-crudes also contain oxygen and after the hydroprocessing, it produces water. But considering the relatively low blending ratio we're working with, two to five percent in our testing scale, the water yield is really low. But of course, you can imagine that it will have water forming during co-processing in hydroprocessing.
Robert Natelson
Another question maybe for Huamin. Any comments on impact of metal contaminants from the biogenic feed from the catalyst performing yield.
Huamin Wang
It's definitely yes. As showed in some of the work for co-processing bio-crudes or unpretreated bio-crudes, which contain a lot of inorganics, as we showed that those inorganics will deposit on external surface of our hydrotreating catalysts, especially on the top layer of the catalyst bed. The impact of the inorganics introduced by woody feedstock and also waste feedstock in the catalyst is kind of some ongoing topic in our other project as well. But in general, we believe that certain inorganics will be more anatomically interacting with catalyst active sites like potassium, causing more trouble. Certain inorganics will be more like physically staying on the external surface of catalyst extrudates. But definitely yes, inorganics will have a role in catalyst deactivation and deactivate the catalyst in different mechanisms.
Robert Natelson
Thanks. Our next question. It's a great question, because it's a clarification question but what is the basis of why you use the term bio-crude instead of HTL and bio-oil when discussing pyrolysis oil, most in industry consider the terms to be synonyms.
I know we are aware of that, any comments?
Reinhard Seiser
Yeah. So, we apologize for the confusion. But we have adopted that nomenclature for ourselves because we, ah, for example, fast pyrolysis oil and calorie fast pyrolysis oil. We don't want to exclude one or the other, and so by saying, by or else it would sort of use both of them, and it is a very oily product, and as opposed to the HTL product that has more water. That's why we call it bio-crude. But even if you use them interchangeably, we will sort of be able to deal with that.
Robert Natelson:
Next question. Have you tried co-processing above five percent? And if so, was it successful, or what issues were encountered?
Reinhard Seiser
Yeah. So, for FCC we have tried larger amounts. We have done, for example, ten percent. And we're trying shortly fifteen percent. And, mostly at our laboratory scale. Feeding was the biggest problem. If you had more bio-oil, it's polymerizing inside the nozzle and building up back pressure. So that's one limitation. But in general, we are interested in doing that because we will be better able to see any negative effects or product composition.
But the reason why five percent was really our center was because of the availability of bio-oils. So, at the scale of refinements, even five percent would be a quite a large scale, and bringing in that much bio-oil from different plants.
And then the question is, also, what types of feeds are we looking at? I mean, I presented to you the sort of like CFP oils, stabilized oils, and eventually you get too much oxygen, or water into your system. It derates your system. It's not good for the catalyst, but there are other feeds that have less of those problematic compounds. So, we have also co-processed feeds from Fischer-Tropsch synthesis up to forty percent vegetable oils easier. We have done it up to one hundred percent.
So that's why, when we say we see a combination of those different ones that could go quite a bit higher than ten percent.
Huamin Wang
For hydrotreating and hydroprocessing, in this project, we're working in five to twenty percent blending level. For hydrotreating or hydrocracking of both CFP bio-oil or HTL bio-crude, the blending level can be as high, we can run pure bio-crude or bio-oil in hydrotreating as a stand-alone processing. So, it means that it can be varied from very low to one hundred in stand-alone hydrotreating. The reason we are working on the relatively low blending, as Reinhard mentioned, we are considering the overall feedstock availability, and at the same time we would like to operate at a condition that has minimal impact on the existing refinery operation, so that they will have minimal modification or interruption of the operation of the current refinery. But again, if needed, we can increase blending level to 50%, or even operate the hydrotreater at a stand-alone mode to do pure bio-oil and bio-crude processing.
Robert Natelson:
Thanks. So, next I’ll ask what form is the nitrogen in after removal to Huamin for any opportunity to return this back to the farm as a fertilizer to reduce carbon intensity of fuel?
Huamin Wang
There are two answers. One is that for the nitrogen in bio-crude is in all kind of nitrogen containing species that you can imagine, such as amide, amine, or these cyclical amine or aromatic amine, for instance quinoline. Normally after hydrotreating nitrogen will be removed as ammonia and the rest of nitrogen, or maybe those hard-to-be-removed nitrogen, mostly like quinoline type nitrogen require deeper or harder denitrogenation. But nitrogen removal products from this chemistry is ammonia.
Robert Natelson
This question, perhaps either or both of you could answer. Has it been considered to co-process the HTL bio-crude in the FCC unit?
Reinhard Seiser
We haven't looked much into that mostly because we think there is too much water, but maybe Huamin has also an answer for that.
Huamin Wang
But actually, you know, water's kind of relatively a minor issue. But a major thing is five percent nitrogen which may have a inhibition impact on the zeolite catalyst used for FCC. A second reason is also considering the species in bio-crudes. Its normally containing pretty nice fatty acid or fatty acid amides. We believe that those carbon, if we can primely process in hydroprocessing condition, can make high quality diesel or jet products and similar to what we are doing for fat and grease oil as well.
Robert Natelson
Next question. Did folks work on HTL biofuels for marine fuels at all? Or is that a pathway that doesn't need much processing other than hydrogenation?
Huamin Wang
Yeah, that's a good question. We do have a separate program working specifically on the opportunity of using bio-oils and bio-crudes, including hydrothermal liquefaction bio-crudes, for the marine sector. We have a recent publication on that one as well, including the team from PNNL, NREL, and also Oak Ridge. So yes, there are significant research on utilizing bio-crudes for the marine sector. And definitely we need to understand how we can process the bio-crudes to make it feasible for that and hydrotreating is one of the approaches to process bio-crudes.
A lot more questions, a little bit more time. Have you had some foaming issues in CFP bio-oil hydrotreating?
Yes, actually, this is really depending on whether we are operating at the relatively small scale and lab-scale reactor. The way we handle the product may be different from what refinery is handling because we just take sample out for analysis. But yes, depending on the operation condition, if we didn't do a good job on deep oxygen removal, in that case, we will see some foaming issues. But again, as I mentioned, our observation is on the lab-scale reactor and may not be applicable in refinery operation.
It’s about the end of the hour and one last question… How much additional expense will result from confirming biogenic carbon content of fuel. Is there the same concern or information is relevant to other products coming out of the oil refinery? I think that’s a great question…
If you can join us again next week, at the same time to hear about the rest of the portion. That's a project on tracking biogenic content in co-processing.
Thank-you.