2018-12-11 12:00 - Maintenance complete.

Evaluating a fast pyrolysis and hydrodeoxygenation process for the production of jet fuel and jet-fuel range aromatics

McLaren, Suandrie Cornel (2015-12)

Thesis (MEng)--Stellenbosch University, 2015.


ENGLISH ABSTRACT: The aviation industry is considering alternative fuels to reduce its dependency on fossil-derived jet fuel, and to mitigate environmental impacts of the latter. Governing bodies have set ambitious goals for the implementation of renewable fuels into the aviation industry and infrastructure. This study aims to investigate the possibility of utilizing fast pyrolysis followed by bio-oil upgrading, via hydrotreating, to produce an aromatic-rich renewable jet fuel. The study consists of two sections: an experimental fast pyrolysis section where the influence of pre-treatment and processing factors on product yields, bio-oil quality and bio-oil characteristics were investigated and a modelling section, where literature data was used to develop a simulation for a fast pyrolysis and hydrodeoxygenation process, using Aspen Plus®. The mass and energy balances from the simulation were used to estimate product yields, utility requirements and waste products. Fast pyrolysis screening experiments utilized a 2³ experimental factorial design with different levels of particle size, biomass moisture content and pyrolysis reactor temperature over ranges of 0.25 -2mm, 3% - 10% (wet basis) and 450–500 oC respectively. Chars were analysed with ultimate analysis and proximate analysis. Bio-oils were analysed using ultimate analysis, Karl Fischer, viscosity analysis, quantified GC-MS, ¹³C NMR and ¹H NMR. Design Expert® was used to perform statistical analyses on the results to determine whether the process parameters had a significant effect on product yield and bio-oil quality. From the results, the optimal point of operation for transportation fuel production, that will require the least upgrading treatment, could be identified. Optimal conditions were identified based on a high organic yield and a high quality bio-oil best suited for fuel, with the most important criteria as: the higher heating value (HHV), a low acid content, a low aldehyde content and a low pyrolytic water content. Particle size had a significant influence on the organic liquid yield, with a small particle size increasing the organic liquid yield. Temperature did not have a statistically significant influence on the organic liquid yield. However, operation at a high temperatures and low biomass moisture contents did improve the quality of the bio-oil. Maximum bio-oil yield and quality was achieved at small particle sizes, low moisture contents and high temperatures. The screening experiment results were used in conjunction with literature ranges to guide to a selection of preferred operating conditions for the fast pyrolysis unit in the simulation. Literature reports on experimental studies producing hydrodeoxygenated bio-oils that are close to jet fuel specifications were limited, with details of processing conditions, especially temperature, pressure, yields and catalysts, often not reported fully within a particular study. Therefore, a combination of available literature was used to develop a simulation model to describe all the steps in the conversion process. The simulation consisted of a fast pyrolysis unit, followed by a hydrodeoxygenation and distillation unit. The simulation also included auxiliary units for hydrogen production, heat recovery, steam production and electricity generation. A pinch point analysis was performed for heat integration. The plant capacity was 92 dry tons/h of biomass feed. Mass balances showed that the aromatic yield in the jet fuel boiling range was only 0.76wt% (dry biomass basis). Furthermore, a jet fuel yield of 3.7wt% (dry biomass basis) was obtained, indicating that the process was not selective towards jet fuel production. The other fuels produced included naphtha, diesel and gas oil, at yields of 4.9wt%, 3.5wt% and 2.3wt% (dry biomass basis) respectively. A high thermal efficiency, 56.8%, compared to the liquid fuel efficiency, 34.1% considering liquid fuels as products, indicated that the energy content of the biomass was not concentrated into the fuel products, but was present in many other streams, including waste streams. The further decrease in liquid fuel efficiency to 11.7%, considering jet fuel as product, indicated that a significant amount of the biomass energy is transferred to the other liquid fuel products. The waste streams and by-products were exploited for thermal use to generate electricity and allowed the plant to produce three times more electricity than it consumes - i.e. 70.2 MW produced with a surplus of 46.4 MW. The simulated bio-oil`s properties were comparable to bio-oil properties determined from the fast pyrolysis experiments in this study, showing a similar HHV, density and moisture content. The distillation gradient of the pyrolysis to jet synthesized paraffinic kerosene (PTJ-SPK) were lower than the minimum specification requirement for SPK in ASTM International D7566 (ASTM D7566), however this is partially due to the limited selection of modelling components, all within a similar boiling range. The lower heating value of the PTJ-SPK was lower than minimum requirement for a SSJF. Both of these deviations from the specification can be addressed by blending PTJ-SPK with a conventional jet fuel with a high boiling range, or with a heavy Hydroprocessed Esters and Fatty Acids jet fuel SPK (HEFA-SPK). The high density and a high aromatic content (20vol%) of PTJ-SPK differentiates it from other SPK fuels. These properties provide an opportunity for producing a fully synthetic jet fuel that contains aromatics by blending 40vol% of PTJ-SPK with 60% of HEFA-SPK or Fischer Tropsch (FT) SPK to obtain a minimum of 8vol% aromatics. However, excessive biomass requirements, due to low PTJ-SPK yields, makes such an option unfavourable. The fast pyrolysis plant capacity in this study will produce sufficient PTJ-SPK to produce a 10vol% PTJ-SPK blend for a FT-facility producing 304 000 tonnes of jet fuel per annum, yet such a blend will not be able to comply with the minimum specification requirement of 8vol% aromatics. Increasing the bio-oil yield from fast pyrolysis and the hydrodeoxygenated oil from hydrotreating, will improve the viability of the process and might render this a feasible option in future. The process effectiveness can be improved by utilizing all the final fuel fractions as products and by exploiting the by-products for electricity generation. This will produce additional gasoline and diesel products, which will require upgrading to meet their final specification requirements. Speciality high value chemicals production from the remaining fuel fractions could also be considered.

AFRIKAANSE OPSOMMING: Die lugvaart industrie oorweeg alternatiewe brandstowwe om hul afhanklikheid van fossielbrandstowwe te verminder. Regulatoriese instansies het ambisieuse doelwitte gestel vir die implimentering van die gebruik van hernubare brandstowwe in die lugvaart-industrie en infrastruktuur. Hierdie studie beoog om die moontlikheid van die gebruik van snel piroliese, gevolg deur hidrodeoksigenering, te ondersoek om `n aromatiese-ryk, hernubare vliegtuig-brandstof te produseer. Hierdie studie bestaan uit twee afdelings: `n ekspirimentele snel piroliese afdeling waar die effek van biomassa behandeling en proses-toestande op die produk opbrengste, bio-olie kwaliteit en bio-olie eienskappe ondersoek is, sowel as `n modellerings afdeling, waar literatuur data gebruik is om `n Aspen Plus® simulasie te ontwikkel. Massa-en energie-balanse is gebruik om die produk opbrengste, utiliteit behoeftes en afval-produkte te bepaal. Snel piroliese siftings-ekspirimente is gedoen, deur gebruik te maak van `n 2³ faktoriaal ontwerp. Verskillende vlakke van partikelgrootte, biomassa vog-inhoud en piroliese reaktor temperatuur is ondersoek oor `n spektrum van 0.25 -2mm, 3% - 10% (nat basis) en 450–500 oC onderskeidelik. Die houtskool is geanaliseer deur elementele (C,H,N) analise en komposisie (“proximate”) analise. Die bio-olie is geanaliseer deur elementele (C,H,N) analise, Karl Fischer analise, viskositeits-bepaling, gekwantifiseerde GC-MS analise, ¹³C NMR analise en ¹H NMR analise. Design Expert® sagteware is gebruik om statistiese analises op die resultate te doen, om te bepaal of die proses toestande `n noemenswaardige invloed op die produk opbrengste en bio-olie kwaliteit het. Vanuit die resultate kan die optimum punt vir bedryf geïdentifiseer word om `n produk vir vervoerbrandstof te produseer wat minimale opgradering vereis. Optimale toestande is geïdentifiseer gebaseer op `n hoë organiese opbrengs en `n hoë kwaliteit bio-olie, met die belangrikste kwaliteit-kriteria as: die hoër warmte waarde (HHV), `n lae suur konsentrasie, `n lae aldehied inhoud en `n lae pirolitiese-water inhoud. Die partikelgrootte het `n noemenswaardige invloed op die organiese vloeistof opbrengs gehad, met klein partikelgroottes wat die organiese vloeistof opbrengs verhoog. Temperatuur het nie `n noemenswaardige effek op die organiese vloeistof obrengs getoon nie. Tog het hoë temperature en `n lae biomassa vog-inhoud die bio-olie kwaliteit verbeter. Die maksimum bio-olie opbrengs en kwaliteit is bereik by `n klein partikelgrootte, lae vog-inhoud en hoë temperature. Die siftings-ekspiriment resultate en literatuur data is saam as riglyn gebruik om die beste bedryfspunt te identifiseer vir die snel piroliese eenheid in die simulasie. `n Beperkte hoeveelheid ekspirimentele studies is beskikbaar waar die proses-toetande, veral temperatuur, druk, opbreng en katalis, gerapporteer word vir hidrodeoksigeneerde bio-olie wat naby aan die vliegtuig-brandstof spesifikasies is. Daarom is `n kombinasie van beskikbare literatuur bronne gebruik om die simulasie model vir die proses te ontwikkel. Die simulasie het bestaan uit `n snel piroliese eenheid, gevolg deur `n hidrodeoksigenerings en distillasie eenheid. Die simulasie het ook addisionele eenhede vir waterstof-produksie, hitte-integrasie, stoom-produksie en elektrisiteits-generering gehad. `n Knyp-punt analise is gedoen vir hitte-integrasie. Die aanlegkapasiteit was 92 droë ton biomassa/h. Massa balanse het aangedui dat die aromatiese opbrengs in die vliegtuig brandstof kook-reeks, slegs 0.76wt% (droë biomassa basis) was. Die opbrengs vir die lugvaart-brandstof kook-reeks, was slegs 3.7 wt% (droë biomassa basis), wat aandui dat die proses nie selektief is vir die produksie van lugvaart-brandstof nie. Ander brandstowwe wat ook geproduseer is, is nafta, diesel en gas-olie (brandstof-olie), met opbrengste van 4.9wt%, 3.5wt% en 2.3wt% (droë biomassa basis) onderskeidelik. Die hoë termiese effektiwiteit van 56.8%, in vergelyking met die vloeibare brandstof effektiwiteit van 34.1%, vir vloeibare brandstowwe as produkte, toon dat die energie wat in die biomassa teenwoordig is nie gekonsentreer word in die brandstowwe nie, maar versprei word in ander strome insluitend afval strome. `n Verdere afname in vloeibare brandstof effektiwiteit tot 11.7% vir slegs vliegtuigbrandstof as produk, dui aan dat `n noemenswaardige hoeveelheid van die biomassa se energie voorkom in ander vloeibare brandstowwe. Die afval strome en by-produkte is vir termiese gebruik benut om elektrisiteit op te wek. Dit het die aanleg in staat gestel om drie keer meer elektrisiteit te produseer as wat die elektrisiteit gebruik is; met `n totale produksie van 70.2 MW en `n surplus van 46.4 MW. Die bio-olie eienskappe in die simulasie en die bio-olie eienskappe wat ekspirimenteel bepaal is in die studie, vir die HHV, vog-inhoud en digtheid, was soortgelyk. Die distillasie gradiënt van die piroliese tot vliegtuigbrandstof gesintetiseerde paraffiniese keroseen (PTJ-SPK) was laer as die minimum spesifikasie wat vereis word vir SPK in ASTM Internasionaal D7566, alhoewel hierdie gedeeltelik veroorsaak is deur die beperkte keuse van model-komponente wat in die simulasie gebruik is, aangesien die komponente soortgelyke kookpunte het. Die laer warmte waarde van die gesimuleerde PTJ-SPK was laer as die minimm vereiste vir SSJF. Beide die afwykings vanaf die spesifikasie kan aangespreek word deur PTJ-SPK te meng met `n konvensionele vliegtuigbrandstof met `n hoë kook-reeks, of deur dit te meng met `n swaar gehidrogineerde esters en vetterige sure vliegtuig-brandstof SPK (HEFA-SPK). Die hoë digtheid en hoë aromatiese inhoud (20vol%) maak PTJ-SPK uniek relatief tot die ander SPK-brandstowwe. Hieride eienskappe skep `n geleentheid om `n volledige sintetiese vliegtuigbrandstof wat aromatiese komponente bevat, te vervaardig, deur 40vol% PTJ-SPK met 60vol% HEFA-SPK of Fischer Tropsch (FT) SPK te meng, sodat `n 8vol% aromatiese inhoud bereik word. Buitensporige hoë biomassa vereistes, wat veroorsaak word deur lae PTJ-SPK opbrengste, maak egter die proses ongunstig. Die snel piroliese aanleg in hierdie studie sal voldoene PTJ-SPK produseer om `n 10vol% PTJ-SPK mengsel vir `n 304 000 ton vliegtuigbrandstof per jaar FT-aanleg te produseer. Hierdie mengsel sal egter nie aan die minimum 8vol% vereiste in die spesifikasie voldoen nie. Deur die bio-olie opbrengs en die hidrodeoksigeneerde olie vanaf hidrogenering te verhoog, sal die lewensvatbaarheid van die proses verhoog en kan hierdie proses `n haalbare opsie in die toekoms raak. Verder kan die effektiwiteit van die proses verhoog word deur al die finale brandstowwe as produkte te benut en die newe-produkte vir elektrisiteit-generering te gebruik. Dit sal addisionele vervoer-brandstowwe, soos petrol en diesel, produseer wat egter verdere opgradering sal moet ondergaan om te voldoen aan die finale brandstof-vereistes. Hoë-waarde spesialiteits-chemikalieë kan ook geproduseer word vanaf die finale brandstof fraksies wat nie vir vliegtuig-brandstof gebruik word nie.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/98060
This item appears in the following collections: