What is pyrolysis oil?
Pyrolysis oils are characterized by a high density and a viscosity which can vary as a function, inter alia, of the starting biomass. As they are acids and thus corrosive, their use requires the employment of specific corrosion-resistant materials, such as stainless steel, high density polyethylene, propylene…
Furthermore, pyrolysis oils are unstable chemically and thermally. The chemical instability of pyrolysis oils is reflected by the change over time in their physicochemical properties (viscosity, water content, solids content, . . . ), which can result in separation into two phases. The thermal instability of pyrolysis oils is reflected by a very rapid change in their properties when they are heated to temperatures greater than 80° C. Due to this instability, these products cannot be upgraded in a refinery in their crude form, except in a combustion application with a few modifications to current plants. For any other application in a refinery, it appears necessary to stabilize pyrolysis oils before use, for example by removing or converting the most reactive entities.
Pyrolysis oils exhibit a mean water content of 25% w/w, an oxygen content of the organic fraction of the order 35-40% w/w and a molecular structure of great complexity. The water content may in addition result in a partial phase separation, having an effect on their other physical properties. Finally, their ash and alkali metals contents may result in the formation of deposits and in the fouling of the plants.
Furthermore, due to their hydrophilic nature and their polarity, fast pyrolysis oils are immiscible with hydrocarbons. Thus, pyrolysis oils cannot be upgraded as such in a refinery as a mixture with hydrocarbon fractions of fossil origin.
Thus, due to their specific properties, the use of pyrolysis oils raises numerous problems.
Currently, the main upgrading routes studied are the combustion of the pyrolysis oil in boilers or gas turbines in order to produce heat and/or electricity, or the production of bases for chemistry.
How to make fuel from pyrolysis oil?
In order to be able to be used in refineries for the purpose of the production of fuel, pyrolysis oils have to undergo a pretreatment targeted at stabilizing them. Such a pretreatment can be a deoxygenation stage which can be:
total, in order to convert the bio-oils into fuel bases,
partial (>90%), in order to render the oils miscible with hydrocarbon fractions of petroleum origin, and thus to introduce it into the refining scheme,
partial (>50%), in order to stabilize the oils for the purpose of subsequent use.
This deoxygenation can be carried out in particular by hydrodeoxygenation (HDO), according to the simplified reaction:
However, the amount of hydrogen necessary is high (from 2 to 7% by weight per weight of pyrolysis oil) and has a major impact on the cost of the treatment of theses pyrolysis oils (approximately ⅓ of the cost price of the fuel bases produced) and on the life cycle analysis (LCA), which is a tool for evaluating the impacts of a system, in this instance the biofuel, on the environment, including all of the activities from the extraction of the starting materials up to the management of the final disposal of the waste. Thus, as regards the biofuel produced from the pyrolysis oil, it is estimated that ⅔ of the total fossil emissions are due to the production of the hydrogen necessary for the hydrodeoxygenation of the pyrolysis oil. In point of fact, an excessively high consumption of hydrogen during the HDO may be totally unacceptable for regarding the fuel bases produced as biofuel according to the criteria of sustainability established by European directives, such as Directive 2009/28/EC, which specifies that, from 2013, the reduction in the emissions of greenhouse gases resulting from the use of biofuels must be at least 35% with respect to fossil fuels. This reduction must be at least 50% from 2017. There thus exists a great need to reduce impact of the consumption of hydrogen during the treatment of pyrolysis oils for the purpose of producing fuel.
One solution might consist in use of the pyrolysis oils themselves to produce hydrogen, for example by catalytic steam reforming.
Steam reforming in a fixed bed has in particular on an “aqueous” fraction of the pyrolysis oils. The fraction was obtained by adding water in order to compel phase separation with the pyrolytic lignin fraction. However, coke deposition results in rapid deactivation of the catalyst, limiting the time of the test to a duration (4 hours) which is less than the time necessary for the regeneration of the catalyst (6 to 8 hours), demonstrating the limitations of the fixed bed for reactions of this type. These phenomena are even more marked on using the total oil as steam-reforming feedstock: under the same operating conditions, the maximum duration of a test reaches 45 minutes, whereas the regeneration time for the catalyst is 8 hours.
With the same operating conditions as in a fixed bed (850° C., S/C>5, GC1HSV of 800 to 1000 h−1) and the same feedstock (“aqueous” fraction obtained by phase separation after addition of water to the pyrolysis oil), the duration of the test under continuous conditions is more than 100 hours.
It is thus apparent that the steam reforming of the aqueous fraction of pyrolysis oil is technically realizable in a fluidized bed and makes it possible to reduce the formation of coke. However, the attrition of the catalyst which is brought about by operation in a fluidized bed requires the development of novel catalysts and of reactors dedicated to this type of treatment. Furthermore, the organic fraction of the pyrolysis oil cannot be upgraded in steam reforming as it comprises compounds which are too heavy for reactions of this type and require another pretreatment in order to be able to be upgradable in a refinery as this fraction is immiscible with hydrocarbons.
The production of hydrogen from pyrolysis oil by a steam reforming process thus can only be envisaged with regard to the aqueous fraction of the oil and does not make possible upgrading of the whole of the oil. As it is possible for said aqueous fraction to still comprise compounds which are precursors of coke following the type of separation used to fractionate the oil, the steam reforming reaction remains difficult to employ.
The invention is targeted at overcoming these disadvantages by providing a process for the upgrading of pyrolysis oils which makes it possible to produce hydrogen while upgrading the whole of the pyrolysis oils treated