Emerging Technologies for the Conversion of Residues

2.5.1 Introduction

Around the world, there are major reserves of theso called non-conventional crude oils, that is to say,heavy crudes and bitumens that are recoverablefrom oil shale and oil sands bitumen, and whichform a strategic reserve to supplement the crudeswith which we are familiar and which areidentified as conventional crude oils. Even thoughthere is no universally recognized definition,normally these fossil sources are classified on thebasis of the API gravity and viscosity values at thereservoir conditions (Table 1). According to thisclassification, heavy crudes are defined as oilswith an API gravity of less than 25°. Among them,those with a viscosity greater than 10,000 mPa·sare classified as extra-heavy; in general, theirgravity is less than 10°API, which indicates thatthey have a higher density than water. Bitumensextractable from bituminous sand, better known bythe term oil sands bitumen, also fall into thiscategory, as well as the oil produced through heattreatment of oil shale.

From a geological point of view, a large part of theheavy crudes derive from mature oils which, after

having been expelled from the source rock, migratedinto permeable layers of rock where they were able toundergo a series of degrading processes, such as attackby micro-organisms, evaporation or washing out of thelight fractions, which resulted in the concentration ofthe heavier component of the oil. A commoncharacteristic of the greater part of heavy oils is theirpresence in fluvial basins relatively close to thesurface, as in the case of the Orinoco basin inVenezuela (Orinoco Belt).

The estimated reserves of heavy crudes and oilsand bitumens amount to around 5,000 Gbbl.Considering also that the technically recoverablefraction is in the range 15-20%, it is evident that weare talking about enormous quantities if one considersthat the whole of the Middle East has reserves of 2,000Gbbl, of which 683 is considered to be recoverable(IEA, 2004; Perrodon et al., 1998). The greater part ofthese reserves is concentrated in Canada in theprovince of Alberta and in Venezuela in theabove-mentioned Orinoco Belt. A third country whichis rich in non-conventional oil is Russia, even thoughin this case, the data for establishing the quantities ofthese reserves and the types of oils are much moreuncertain (Table 2).

137VOLUME III / NEW DEVELOPMENTS: ENERGY, TRANSPORT, SUSTAINABILITY

2.5

Emerging technologiesfor the conversion of residues

Table 1. Classification of non-conventional oils

Specific gravity(°API)

Viscosity(Pa�s)

Rheologiccharacteristics of oil

at reservoir conditions

Heavy crude oils 16-25 Mobile

Extra-heavy crudes �10 �10 Mobile

Bitumens from tar sand 7-12 Non-mobile

Oil shales Impermeability of the source rock

As far as oil shale is concerned, the worldwidereserves are in the order of 2,600 Gbbl; of these, about2,000 Gbbl are within the territory of the United Statesand, in particular, in the formations called Green River(Colorado), the Uinta Basin (Utah) and the WashakieBasin (Wyoming). Other significant reserves can befound in Brazil, Australia, China, Russia and Estonia(Dyni, 2004).

The exploitation of these fossil reserves is highlystrategic inasmuch as it would make it possible toincrease the known reserves without recourse toinvestment in new exploration. Moreover, theseresources help to diversify the sources of supply and,given their geographical distribution which locatesthem predominantly in areas other than the MiddleEast, to eliminate the geopolitical risks which havebeen a constant factor in the crude oil marketplace.

Production of heavy crudes and bitumens veryoften involves the use of special technologiesdeveloped specifically to handle products which arehighly viscous or, as in the case of bitumens, dispersedwithin sandy mineral groundmasses or, yet again,recoverable only through thermal processing of theorganic material contained in sedimentary rocks, ashappens with oil shale.

In the last 15-20 years, interest in developing thetechnologies for exploiting non-conventional oils hasgone through alternating phases, depending on thefavourability, or otherwise, of the macro-economicclimate for scheduling investments in this sector(forecasts of the price of crude and of the differentialbetween heavy and light oils). Moreover, expectationsabout overcoming various technological obstaclesconnected primarily to the compositionalcharacteristics of these resources often have not beenlived up to.

Nevertheless, the forecasts of the demand for oilover the next 20-30 years and the pointers whichemerge on the availability of reserves of conventional

crude, according to which production will peak overthe next ten years (Fig. 1), reinforce the idea that it willbe increasingly necessary to make recourse to non-conventional oils, drawing on the reserves of extra-heavy crudes and bitumens of Canada, Venezuela andRussia and, thereafter (beyond 2030), on oil shale.

These factors are behind a series of industrialinitiatives which, over the next decades, could bringsignificant amounts of synthetic and/or distilled crudesfrom non-conventional sources onto the market. This isalso due to the progressive reduction in productioncosts resulting from the development/optimization ofnew technologies, both upstream and downstream. Inthis regard, the most significant case is certainly that ofCanada where the efforts in developing ad hoctechnologies for exploiting the oil sands bitumenfields, which began in the 1970s, have made it possibleto reduce production costs by more than 50%, makingthis type of activity commercially viable. It is forecastthat in 2010, more than 60% of Canadian production

138 ENCYCLOPAEDIA OF HYDROCARBONS

HYDROCARBONS FROM NON-CONVENTIONAL AND ALTERNATIVE FOSSIL RESOURCES

0

25

50

75

100

125

1971 1980 1990 2000 2010 2020 2030

oil w

orld

pro

duct

ion

(Mbb

l/d)

existing capacities

development of new discoveries

enhanced oil recoveries

development of existing reserves

non-conventional oils

Fig. 1. Medium-term forecasts for worldwide oil production (IEA, 2004).

Table 2. Main deposits of bitumens (1) and heavy crude oils (2)

Deposits of bitumens and heavy crudes Reserves (Gbbl)Technicallyrecoverable

(Gbbl)Depth (m) °API

CanadaAthabasca1, Cold Lake1, Peace River1,Lloydminster2

1,630 315 0-750 8-20

Venezuela2

Orinoco Belt (Cerro Negro, Zuata, etc.),Bachaquero, Boscan

1,900 272 1,300 8-11

Russia1

Siberian platform, Malekess450

will come from oil sands bitumen in the form ofbitumen itself or synthetic crude oil (SCO), and thatwill enable it to increase its oil production from thecurrent 2.5 Mbbl/d to over 5 Mbbl/d, making it thefourth biggest producer after Saudi Arabia, Russia andthe United States.

2.5.2 Properties and chemicalcharacteristics of non-conventional oils

Heavy crude oils and bitumensAs far as their compositional characteristics are

concerned, as with petroleum, heavy crude oils andbitumens are made up of highly complex mixtures ofhydrocarbons. Their chemical and chemical-physic(molecular weight, H/C ratio, specific gravity,volatility, etc.) characteristics vary continuously, fromthe simplest paraffin structure (hydrocarbon gases) tothe macromolecules made up of scores of carbonatoms, as well as hetero atoms (sulphur and nitrogen)and metals.

The methodologies developed to determine theircharacteristics are related to the methods used inoilfields to simplify the mixture, operating in such away as to separate the fractions with the mosthomogeneous chemical-physic characteristics possible(Altgelt and Boduszynski, 1993). The primaryoperation is distillation which enables the fractions tobe separated based on their volatility. In this regard,the heavy oils and bitumens have a distillablehydrocarbon content (naphtha and gasoils) which isremarkably lower than that of traditional crudes, suchas for example, Arabian Light (Table 3).

The different distillation cuts can be furtherfractionated, based on criteria of polarity and/ormolecular weight, by processes with varying degreesof precision according to the complexity of themixture and depending on the type of informationrequired. In the case of light distillates (naphtha andatmospheric gasoils), chromatographic separation isused to sub-divide the saturated hydrocarbons from thearomatic. For the heavier fractions and the distillationresidues – which, in the case of the products underconsideration, account for the largest quantity of


EMERGING TECHNOLOGIES FOR THE CONVERSION OF RESIDUES

Table 3. Principal compositional characteristics of heavy crude oils and bitumens

ArabianLight

Zuata Boscan Maya Cold LakeAthabasca

Bitumen

Source Saudi Arabia Venezuela Venezuela Mexico Canada Canada

API gravity 33.6 8.5 10.5 21.5 10.2 7.4

Distillation yield (% by weight)

Naphtha 20.6 0.0 4.0 12.9 1.5 1.0

Atmospheric gasoil 36.0 14.1 11.6 21.7 14.9 13.0

Vacuum gasoil 23.2 31.0 20.2 22.2 38.8 34.0

Vacuum residue (VR) 20.2 54.9 64.2 42.2 44.8 52.0

Vacuum residue characteristics

TBP cut * 530°C� 500°C� 350°C� 500°C� 340°C� 300°C�

API gravity 8.3 2.5 7.2 1.5 7.2 7.8

Sulphur (weight %) 4.0 4.2 6.0 5.2 4.9 4.6

Nitrogen (weight %) 0.25 0.97 0.96 0.81 0.70 0.48

Nickel (ppm) 30 154 119 132 107 70

Vanadium (ppm) 110 697 1,473 866 210 186

C7 Asphaltenes (weight %) 5.3 19.7 18.2 30.3 N/a 12.4

CCR (weight %) 18.0 22.1 18.3 29.3 20.8 13.6

* The mnemonic TBP (True Boiling Point) indicates that the cut has been carried out in accordance with the procedures indicated in the ASTMstandards.

fractions – the commonly accepted analytical protocolinvolves the preparation of four classes of compoundscalled: saturates, aromatics, resins and asphaltenes(SARA analysis). Each class can then be analysed toidentify its individual components or to evaluate itsmost important molecular characteristics, dependingon the complexity of the mixture and/or therequirements.

Compared with traditional crudes, the quantity ofsaturated hydrocarbons contained in heavy crude oilsand bitumens is considerably lower (Fig. 2). From aqualitative point of view, there are major differencesdue to a lower concentration of n-paraffins in favour ofiso-paraffins and naphthenes with a high degree ofcondensation that very often contain sulphur in theirstructure.

The non-aliphatic component is made up ofaromatic and heteroaromatic hydrocarbons withdiffering degrees of condensation, differentalkyl-substitution and functionalization. Even in thiscase, for the same distillation cut, the hydrocarbonstructures display a higher degree of condensationcompared with what is seen in conventional crudes.

Another characteristic of heavy crudes andbitumens is the fact that they contain significantamounts of heteroatoms (especially sulphur andnitrogen), as well as heavy metals, such as nickel andvanadium in particular.

By far, the most common heteroatom is sulphurwhose concentration can reach values of 6-8% byweight. Sulphur is distributed in increasingpercentages in the products with the highest boilingtemperature and is present predominantly asthiophenic sulphur in condensed structures (benzo-,

dibenzo- and naphthobenzo- thiophene), but also asaliphatic sulphur in sulphide and disulphide typefunctional groups. These functionalities are often usedto create links between hydrocarbon clusters.

Nitrogen, which is contained at levels of 0.5-1% byweight, tends to concentrate in the heaviest fractionson the distillation curve. This heteroatom is found inboth basic type (predominantly primary aliphatic andaromatic amines and pyridines) and neutral type (inthe form of indoles, carbazoles, imides, as well asporphyrin nitrogen) functional groups.

Oxygen is present in the crude, in heavy oils andbitumens in small amounts, about 1.0-1.5 wt% or lessand, since it ends up by preference in hydroxylic typegroups (phenols, alcohols and carboxylic acids), itconcentrates in the most polarized components of thecrude such as resins and asphaltenes. Naphthenic acidscertainly make up the most researched class ofoxygenated compounds, above all because of theircorrosive properties. More rarely, the oxygen canappear in the form of ethers or cycloethers, or coupledwith other heteroatoms to form sulphoxides andamides.

As far as metals are concerned, nickel andvanadium are by far the most abundant elements (up toseveral hundred ppm), even though some oils cancontain significant quantities of sodium, iron andmolybdenum. These metals are contained inoil-soluble metallo-organic structures and areconcentrated in the heaviest parts of the oil, so muchso that they are found in abundance in asphaltenes. Aconsistent part of the metallo-organic component isthe porphyrinic type, but many other structures whichare often difficult to identify broaden the case studiesof the metal-containing compounds present inasphaltenes.

Bitumens obtained by extraction from oil sandsbitumen can contain inorganic material, typically clayand sand which are dispersed within the oily matrix ingranules with linear dimensions in the order ofmicrons (silt). The amount of the inorganic materialdepends on the extraction technology and on theprocesses used to separate the organic phase from thesand; this content is from 0.5 to 1.1% by weight if thebitumen is produced using traditional miningprocesses, whereas it can drop to values of between500 and 1,000 ppm where more modern productiontechnologies are used, such as SAGD technology(Steam Assisted Gravity Drainage) in particular.

Chemical and chemical-physic characteristicsof asphaltenes

The vacuum residues, i.e. the quantitatively greaterfraction of heavy crudes and bitumens, mainly consistof aromatic hydrocarbons condensed to varying



100%aromatics

conventionalcrude oils Athabasca, Canada

Cold Lake, Canada

Lloydminster, CanadaCherokee, USA

100%saturates

100% resins �asphaltenes

Fig. 2. Distribution of the principal hydrocarbon typesmaking up heavy crudes and bitumens.

degrees and distributed across a wide range ofmolecular weights and polarities. They constitute acompositional continuum even though, as mentionedearlier, they are generally divided into aromatics,resins and asphaltenes. The differentiation of thesefractions cannot, however, disregard the operatingparameters used to carry out the separation.Deasphalting is a typical example which is largelyinfluenced by the type of precipitating agent used aswell as the operating conditions (Speight, 2004). Thedirect consequence is that to be able to carry out acomparison between different samples, it is essentialto standardize the analysis conditions, specifying theprecipitating agent used (usually n-pentane orn-heptane) and supplying details of the method used.Conceptually, the separation of asphaltenes can becompared to the fractionating through distillation ofthe volatile component of the crude oil. In both cases,the constituents of the mixture are divided by defininga ‘cut point’, such as the boiling temperature(distillation) or the solubility parameter of theprecipitating agent (deasphalting). The solventstrength of the hydrocarbons, or better still, its abilityto act as a solvent or anti-solvent in precipitatingasphaltenes, can be correlated to the solubilityparameter d defined by Joel Hildebrand and RobertScott as:

DHV�RTd ��11112�

124

2

V

where DHV is the molar enthalpy of vaporization ofthe hydrocarbon concerned, V is its molar volume, andR and T are the universal constant of gases and theabsolute temperature respectively. The solubilityparameter may also be estimated by using theequation:

d2�AV�

124

3g

where A is a constant, g is the surface tension and V isthe molar volume of the hydrocarbon concerned(Barton, 1991).

A very effective bi-dimensional diagramillustrating the molecular characteristics of asphalteneswas suggested by Robert Long (Fig. 3), in which thepolarity and molecular weight of the hydrocarbonspresent in the oil constitute the primary parameterswhich, depending on the solvent used, determine theprecipitation of specific fractions (Long, 1979). Inaddition, this diagram indicates that asphaltenes mustbe considered to be a class of compounds which, froma chemical point of view, can be very broad anddiverse.

As well as the method used during the separationphase, the chemical characteristics of the asphaltenesdepend on the crude oil origin. As shown in Table 4,

asphaltenes from different sources vary considerablyin their aromaticity, degree and nature of alkylsubstitution, as well as the heteroatom content(Cimino et al., 1995).

As far as their molecular weight is concerned, theissue is complicated even further because of thetendency of the asphaltenes to form aggregates, tosuch an extent that the molecular weightmeasurements made by means of Vapour PressureOsmometry (VPO) or Size Exclusion Chromatography(SEC) are significantly affected by the polarity of thesolvent used. For this reason, the average molecularweight values quoted in literature, where you find avariability ranging from 103-104 units of atomic mass,have a (relative) significance if considered within a setof measurements taken, but can be given littlecredence as far as the actual molecular dimensions ofthe sample are concerned.



polarity

C5 asphaltene

crude oil

C7 asphaltene

mol

ecul

ar w

eigh

tFig. 3. Representation of the molecular characteristics of asphaltenes (Long, 1979).

Table 4. Compositional variability of C7 asphaltenesprecipitated from crudes and bitumens

Yield of oil based asphaltenes (% by weight) Up to 30

H/C ratio 0.8-1.4

Sulphur (weight %) 0.5-10.0

Nitrogen (weight %) 0.6-2.6

Oxygen (weight %) 0.3-4.8

Aromaticity factor 0.45-0.70

n (average number of atoms of C per alkylicsubstituent)

4-7

The chemical-structural analysis of asphaltenes canbe carried out effectively through techniques ofnuclear magnetic resonance (1H- and 13C-NMR).Using these techniques, various average molecularparameters may be determined, such as aromaticityfactor, degree of alkyl substitution, average chainlength, etc. These are extremely useful for giving anidea of the hydrocarbon skeleton, as well as thechemical functionality of the molecules present in thesample. The results of these analyses tend to show theasphaltenes as macromolecules made up ofpolycondensed aromatic clusters, substituted invarious ways with alkylic chains which can be fairlylong (>C10), and linked together by saturated andhydrocarbon chains and heteroatoms. The degree ofcondensation of the aromatic units can be high, butnormally it does not exceed a figure of 5-6 rings(Speight, 1980).

As with the other classes of compounds, theasphaltene component of heavy crudes and bitumens isalso significantly different from that of light crudes. Inaddition to the high heteroatom and metal content,asphaltenes have significantly higher molecularweights, due to the high concentration of sulphurwhich favours the formation of sulphide anddisulphide bridges among the aromatic clusters, acharacteristic which makes these structures veryreactive in relation to thermal cracking andhydrogenating reactions (see above).

From the molecular point of view, the asphaltenescontained in bitumens can be represented effectively

by ‘archipelago’ models, that is, structures made up ofislands of little groups of condensed rings, linkedtogether by aliphatic chains and sulphide bridges, asdepicted in Fig. 4 (Sheremata et al., 2004).

The resins are compounds that fall betweenasphaltenes and hydrocarbon components (saturatedand aromatic); they consist of polar molecules similarto those of asphaltenes, but contain longer lateralaliphatic chains and smaller aromatic rings.

As already stated, because crude oil is acontinuum, it is the separation procedure thatdetermines the difference between asphaltenes andresins; resins could be considered to be asphalteneswith a low molecular weight, just as asphaltenes couldbe seen to be resins with a high molecular weight.

From the 1950s onwards, thermo-dynamic modelshave also been proposed in which the asphaltenesappear as colloidal particles dispersed in the oil due tothe action of the resins which surround them.Displaying a comparatively greater polarity than therest of the oil, the resins are adsorbed onto the surfacesof the colloidal particles. In accordance with thesemodels, the asphaltenes are stabilized or ‘peptized’ bythe resins; if a change in conditions of temperature,pressure or composition leads to the desorption of theresins from the colloidal particle surfaces, there is aseparation (precipitation) of the asphaltenes (Murgichet al., 1996).

This model, which describes asphaltenes aslyophobic colloids, is being increasingly discarded infavour of a description of asphaltenes as lyophilic



CH3

CH3

CH3

CH3

OH

OOH3C

H3C

H3C

H3C

S

S

N

N

S

S

O

A B

O

O

O

S

S

S

S

S

S

SNH

HN

HN

Fig. 4. Average molecular structures representing asphaltene molecules from different sources: A, asphaltenes from traditionalcrudes; B, asphaltenes from Canadian bitumen (Sheremata et al., 2004).

colloids, solvated by the surrounding medium. In thismodel, the separation phase of the asphaltenes islinked to a decrease of the medium’s solvent strength,and the resins cease to play the key role in the system(Cimino et al., 1995).

Nature and chemical characteristics of oil shalesOil shales are sedimentary rocks, generally

silicates and carbonates, containing significantamounts of insoluble organic material which canbe recovered through pyrolytic distillation (aprocess which is better known as retorting). In therocks which cover a site of potential commercialinterest, the amount of organic material must begreater than 10 gal/t (45 l/t), even though as anorm, in the richest formations, this value isbetween 30 and 40 gal/t (for Athabasca bitumen itis around 22 gal/t). Oil shale deposits can extendfor hundreds of square kilometres with thicknesseswhich can reach 700 m, so that the quantity ofrecoverable oil per unit of surface area is greaterthan the Canadian oil sands by an order ofmagnitude. The productivity of the biggest deposits(for example, Colorado oil shale) can, in fact,reach figures of up to 0.73 bbl of oil per tonne ofmaterial extracted (Bunger et al., 2004).

The organic component of oil shale is made up ofcomplex hydrocarbon molecules comparable tokerogen (Fig. 5), from which petroleum originates,containing significant quantities of oxygen (5-6% byweight) and, to a lesser extent, sulphur and nitrogen.The hydrogen content of the kerogen itself is

significantly greater than that of coal, with an H/Cratio of 1.5-1.6 compared with values of 0.8-0.9 forbituminous coals. In a similar way to coal, this organicmaterial is usually divided into groups of maceralsaccording to their optical (reflection of light) andmorphological characteristics, echoing the nature ofthe biological material which created them. Maceralscan be grouped into three primary types calledtelalginite, lamalginite and bituminite which, in turn,can be further divided into subgroups.

2.5.3 Chemistry of the conversionand upgrading processes

The purpose of converting and upgrading petroleumresidues, heavy crudes and bitumens is to transform asubstrate consisting of high molecular weighthydrocarbons, which are viscous and rich in toxicelements and metals, into products which are lighterand more fluid (synthetic crude oil), and arecomparable to traditional crudes; or better still, intodistillates that can be further upgraded to yieldgasoline and diesel for motor vehicles. Thistransformation can be achieved directly throughthermal or hydrocracking processes, or indirectlythrough transformation of the feedstock (a quantity ofmaterial for feeding the reactor) into syngas (i.e. amixture of CO and H2) through gasification (see Vol.II, Chapter 7.3), and the subsequent production ofparaffin through Fischer-Tropsch synthesis (seeChapter 2.6).



O

O

OO

O

O

O

OH

OH

OH

OH

HO

HO

HS

HO

HO

OH

OH

OHO

O

O

O

O

O

OH

OHOH

OH

Cl

O OOH

O

O

O

NH

OHO

O

OO

S

HO O

O

S

O

S

Fig. 5. Representation of the molecular structure of kerogen (Lille et al., 2003).

Processes for conversion into distillatesThe processes for direct conversion of heavy

feedstocks into distillates are particularly complex andinvolve the reduction of the molecular weight of thefeedstock constituents through reactions which breakthe bonds of the hydrocarbon molecules (cracking)and increase the H/C ratio; the latter can be achievedthrough the removal of carbon (C-rejection process) orthe addition of hydrogen (H-addition process).

The C-rejection processes are thermal processesthrough which the heavy hydrocarbons in thefeedstock are disproportionated, generating distillateswith a high H/C ratio and releasing a highly aromaticresidue (tar or coke). This is a radicalic type processand involves the homolytic breaking of the C�C andthe C�heteroatom bonds, followed by b-scissionreactions through which, as the reaction progresses,increasingly lighter hydrocarbon fragments areproduced, generating distillates and gas. However, thearomatic radicals produced by dealkylation (radical p)tend to react among themselves, giving rise to highlycondensed polynuclear structures which become everless soluble in the reaction mixture and lead to theformation of mesophase and, hence, coke above acertain level. The residue’s tendency to form coke islinked to the degree of polycondensation of the heavyaromatic structures and is quantified by the CCR value(Conradson Carbon Residue), which is measuredaccording to ASTM (American Society for Testing andMaterials) methodology D 189.

The principal types of reaction that operate in theseprocesses are, therefore, the dealkylation of aromaticstructures, the dehydrogenation of naphthenes andcondensation. All these reactions are favoured by thetemperature which is generally in excess of 450°C.

From a kinetic point of view, at least as far asvisbreaking is concerned, the production of crackingproducts follows an apparent first order kinetics withactivation energy values of around 230 kJ/mol, whichindicates that the reaction speed doubles for eachincrement in temperature of 14-15°C.

In general, thermal processes are not very selectivetowards the production of distillates since, as the

severity of the process increases, the gas yieldincreases, and problems are encountered in relation tothe stability of the reaction products (see below). Thequality of the distillates is poor, because thermalcracking alone is not capable of removing, in anysignificant way, the heteroatoms present in the heavyfeedstocks. Moreover, naphtha and gasoil are rich inolefins and dienes and, hence, must be stabilizedthrough hydrotreatment.

In the H-addition processes, conversion of theheavy feedstocks and the distillates is achievedthrough the combined reaction effects of cracking andcatalytic hydrogenation of the reactive fragments. Inthis way, it is possible to control the propagation of theradicalic reactions more effectively, above all, withregard to the condensation processes of the aromaticsand, hence, to reduce the problem of coke formation(Fig. 6). Moreover, depending on the reactionconditions and the type of catalyst used, it is possibleto add hydrogen to the products, saturating thearomatic structures and facilitating the elimination ofthe heteroatoms. For this reason, the quality of thedistillates and the conversion residue obtained fromthe hydrocracking processes is definitely better thanthat of distillates obtainable through thermalprocesses.

With regard to the thermodynamics of the process,the aromatic structures’ balance of hydrogenationreactions is facilitated by a high hydrogen partialpressure, whereas it is hindered by an increase intemperature. Therefore, the requirement to operate attemperatures in excess of 380°C to promote thermalcracking makes it necessary to push the hydrogenpartial pressure up to levels exceeding 100-120 bar.

The ideal catalyst for the upgrading of heavyfeedstocks must facilitate the process of hydrogenaddition to the products generated in the thermalcracking phase, minimizing the amount of cokeproduced. Moreover, it must permit the removal of thepoisons present in the feedstock throughhydrogenation of the substrate, or rather it must favourthe processes of desulphurization(HDS, Hydrodesulphurization), denitrogenation



.S

RD

R1 CH2.

CH2

S

R1

S

R1

SH

R1H2/

catalyst

mesophase coke

distillates

Fig. 6. Simplifiedrepresentation of the process of hydroconversionof heavy feedstocks.

(HDN, Hydrodenitrogenation), demetallization(HDM, Hydrodemetallization) and reduction of carbonresidue from the products (HDCCR, Hydro ConradsonCarbon Residue Removal).

The most active catalytic species for thesereactions are certain heavy metal sulphides such asMo, Ni, Co, W and Rh in particular, often used incombinations (Ni/Mo, Co/Mo and Ni/W) anddeposited on appropriate porous supports (preferablyalumina) or mixed with the feedstock in the form ofpowder (catalysis in slurry phase).

The main problem that is encountered when usingsupported catalysts to process particularly heavyfeedstocks is that of limiting the deactivation of thecatalyst due to the depositing of both the metals andcoke. However, it should be made clear that thedeposited coke can be removed through regenerationof the catalyst, whereas the activity loss due to themetals is permanent and, hence, the catalyst has to bereplaced.

The deactivation due to deposits of coke leads to aloss of activity because the active sites of the catalystare covered by carbonaceous material, primarily fromasphaltenes; in fact, this loss increases with the levelof conversion of the feedstock into distillates, that is,when conditions favour the emergence of stabilityproblems. The deactivation by coke can be offset by anincrease in the hydrogen partial pressure.

As regards the metals, the deactivation comesabout through the obstruction of the porous structureand the covering of the active sites by the metalscontained in the metallo-porphyrinic structures whichare destroyed during the reaction generating thecorresponding sulphides.

However, the physical characteristics of the supportand, in particular, its porosity are as fundamentallyimportant as those of the active phase in determiningthe catalyst’s behaviour. Its high viscosity and thepresence of high molecular weight compounds(asphaltenes and metallo-organic compounds), whichare characteristic of heavy feedstocks, make access tothe substrate within the catalyst particles difficult; thediffusion process within the porous structure can be aproblem and can represent the slow stage of thereaction. If the catalyst does not have a sufficientlyporous structure and the diffusion of the molecules inthe pores is impeded, most of the metals are depositedon the external surfaces, causing obstruction of thepores and, therefore, impeding the reaction.Consequently, in the case of heavy feedstocks,macroporous materials are used, often as ‘sacrificialbeds’, on which most of the demetallization reactionsare made to take place in order to go ahead with theconversion reactions and upgrading of the feedstockusing ad hoc catalytic beds.

The catalysts used in the slurry processes are oftenclosely associated with carbon material (coke) whichis either produced during the reaction or is purposelyadded. Compared with supported catalysts, on whichconventional hydrocracking technologies are based,these materials are not very sensitive to the presenceof poisons, since they do not display the classicproblems resulting from the depositing of coke andmetals onto the pores of the support. The use ofdispersed catalysts based on metal sulphides fromgroups V, VI and VIII (in particular Fe, Mo and V) forupgrading residues, heavy crudes, bitumens and coalis well-known and has been fully described inscientific literature for more than thirty years. The firstsignificant works published on this subject makereference to Clyde Aldridge and Roby Bearden(Aldridge and Bearden, 1978), and describe the use ofMo introduced in the form of oil-soluble precursors.Subsequently, numerous variations were tried andproposed in regard to both the use of various types ofprecursors and the ex situ synthesis of the catalyst inorder to improve its specific activity.

The most active dispersed catalysts, however, arestill those based on molybdenum and obtained throughthe decomposition of oil-soluble precursors such asnaphthenates, oxalates, xanthates, dithiocarbamates orother metallo-organic derivatives such as Molyvan A(N,N-dibutyldithiocarbamate of oxothiomolybdenum)which are supplied at the hydrocracking/hydrotreatingstage together with the feedstock (Delbianco et al.,1995). The in situ decomposition of these precursorsin the presence of hydrogen and sulphur generates an



10 nm

Fig. 7. Structure of micro-crystalline molybdenite obtainedthrough TEM microscopy (Transmission ElectronMicroscopy; Panariti et al., 2000).

extremely fine powder made up of nano-sized layersof molybdenum sulphide (molybdenite, MoS2) with alow degree of aggregation (nanocluster) and highlydispersed within the feedstock. The active catalyticphase is, therefore, the molybdenite, a well-knownlayered hexagonal structure with molybdenumbetween two layers of sulphur. The adjacent layers arelinked together by the weak dispersion forces (Van derWaals type) in action between the respective sulphuratoms. For this reason, the molybdenic structure canbe easily peeled (delaminated) until the elementaryfoils are obtained which display a very low degree ofstacking, guaranteeing a high dispersion in the oilymatrix (Fig. 7). On average, the radius is in the rangeof 2-4 nm. Microscopic analysis reveals that themolybdenite crystals tend to aggregate into particleswith a length in the order of a micron which appear inthe shape of irregular clusters that have an averagediameter of 0.5-2 mm (Panariti et al., 2000).

Its morphological characteristics and the absenceof porous supports make molybdenite particularlywell-equipped to operate effectively as ahydrogenation catalyst in very severe conditionsbecause of the presence of high concentrations ofpoisons, especially heavy metals. The catalytic activityof molybdenite in a hydrogenating environment seemsto be due to both the formation of sulphur vacancieson the profiles of the nanoclusters due to the effect ofthe interaction of the hydrogen with MoS2, and theformation of –SH groups which evolve into H2S.

The reactivity of heavy feedstocks tohydrocracking is strongly influenced by the nature ofthe substrate which can display differing degrees ofreactivity, depending on the average molecularstructure and on the concentration and nature of theheteroatoms present.

The most reactive substrates are characterizedby the fact that they have average molecularstructures containing relatively small aromaticsclusters, linked together by alkylic chains orC�heteroatom bonds with low bonding energies.As regards the hetero-atoms, the elimination of the

sulphur comes about through the formation of H2Sand may or may not be difficult, depending on thenature of the sulphuretted compound that containsit. In general, the reactivity of the sulphurettedspecies follows a decreasing scale: aromatic S(condensed thiophenes�thiophenes)�naphthenicS�paraffinic S (thio-ethers and disulphides).

The second heteroatom usually present in heavyfeedstocks is nitrogen, which is removed in the formof NH3 through hydrogenation of the heteroaromaticstructures that contain it. The elimination of nitrogenis more difficult than that of sulphur since the energyof the C�N bond is greater than that of the C�S bond (360 kJ/mol against 320 kJ/mol for amineand alkylic sulphide, respectively).

As far as metals are concerned, the removal of Niand V (demetallization reactions) is performed bypassing the porphyrinic structures in which they arecontained through a hydrogenation process. This leadsto the formation of pyrrhotitic type sulphides, i.e.Ni1�xS, V1�xS (with x�0.1), which are released in thereaction mixture or trapped in the porous structure ofthe supported catalysts used in the process.

The main reactions involved in the residuehydrocracking processes are exothermic. The heatgenerated depends on the nature of the feedstock beingtreated and the degree of conversion, as well as thelevel of upgrading achieved by the process. Thethermal tonality of the different reactions, which occurin the upgrading processes, can be estimated based onthe hydrogen consumption within a determined rangeof values, as illustrated in Table 5 (Tominaga andTomaki, 1997).

Stability of petroleum residuesFrom a general point of view, stability expresses the

capacity of a petroleum residue to tolerate dilution withfluxants (cutter stocks) which are predominantlyparaffinic in nature, without giving rise to theprecipitation of asphaltenes. The stability is assessed bydetermining the P-value, that is, the value of P derivedfrom the equation: P�1�Xmin, where Xmin is the valueof the sample’s maximum dilution in cetane (n-C16H34)where there is no precipitation of asphaltenes,expressed as ml of cetane per gram of the sample.

In primary distillation residues (straight run), thestability is an intrinsic quality of the product anddepends on the compositional characteristics of theasphaltenes compared to those of the non asphaltenichydrocarbon phase (maltenes). The conversionprocesses, both thermal and catalytic, modify thechemical nature of these two pseudo-compounds,causing a progressive reduction in stability as theseverity of the treatment grows (linked to the reaction’stime-temperature combination). The reasons for this



Table 5. Estimate of reaction heat for the main reactionsinvolved in the hydrocracking of heavy feedstocks

Type of reaction kJ/(mole H2 consumed)

Cracking and ring opening 20-45

Aromatic saturations 55-70

Olefin saturations 115-125

Hydrodesulphurization (HDS) 55-75

Hydrodenitrogenation (HDN) 60-85

reduction in stability are due to the fact that during thecourse of the reaction, the asphaltenes are dealkylatedand become increasingly aromatic and, hence,increasingly less soluble in the maltene phase whichtends to become more paraffinic for the same reason.Above a certain conversion level, the asphaltenesprecipitate (hence a separation of the liquid-liquidphase is seen), triggering processes in whichmesophase is formed and, hence, coke.

This phenomenon can be observed in all residueconversion processes, both thermal and hydrocracking,so much so that the stability of the conversionproducts actually determines the maximum conversionlevel obtainable for a given feedstock. For this reason,it is not possible to push the conversion beyond20-30% in visbreaking plants or beyond 40-50% infixed bed hydrocracking plants.

2.5.4 Carbon rejectiontechnologies

Thermal processes: visbreaking and coking

VisbreakingVisbreaking is a very simple technology for the

treatment of petroleum residues, which is verywidespread globally. The process involves heating thefeedstock for just a few minutes to temperatures inexcess of 450-460°C and at low pressure. In theseconditions, the high molecular weight hydrocarbonstructures that make up the heavy feedstock undergo apartial thermal cracking process, resulting in theproduction of a limited quantity of distillates, generallyless than 30% by weight, and of a residue with reducedviscosity. This residue has to be fluxed until the requiredviscosity is obtained to produce a fuel oil, using aquantity of diluent that is less than that of the feed itself(see Vol. II, Chapter 5.2). The limit on the severity of theprocess is, in fact, linked to the stability of the residue ofwhich the P-value must be greater than 1.1-1.2.

The main aim of using visbreaking to upgradeheavy crudes and bitumens is to make these productsmore fluid, facilitating their transportation via pipelinewithout the need to use diluents (naphtha).

Obviously, given the characteristics of the process,thermal treatment does not reduce the levels ofpollutants, and for this reason, traditional visbreakingis of little interest for use in the field of extra heavycrudes and bitumens. To overcome these limitations, atleast in part, the Institut Français du Pétrole (IFP) hasproposed variants of the process that call for the use ofhydrogenating atmospheres (hydrovisbreaking) and,where necessary, metallic additives capable ofpromoting hydrogenation reactions (catalytic

hydrovisbreaking) which are called Tervahl H andTervahl C, respectively. The improvement inperformance in terms of HDS activity is, however,limited to values of around 20% compared withclassical visbreaking.

Another solution, proposed by PDVSA-Intevep andpatented in conjunction with Foster-Wheeler/UOP, is aprocess called Aquaconversion. In this instance, thevisbreaking operation is carried out in the presence ofwater and a catalyst capable of promoting the partialbreakdown of the water to produce hydrogen in situ,which is used for a partial upgrading of the feedstock,while the oxygen is used to produce CO2. The processwas developed to pilot plant level and, later, on ademonstration scale using a suitably modified existing18,000 bbl/d visbreaking unit. In the reaction conditionssuggested by PDVSA, the cracking reaction can bemanaged at a level of severity above that of classicvisbreaking, making it possible to increase the conversionyields for the same level of stability of the residue andimprove the quality of the product as well (Table 6).

Then there are thermal cracking processes thatoperate at greater severity than visbreaking that aim tofurther increase the degree of conversion into distillatesand produce a residue capable of being pumped.Because the residue is unstable, it cannot be utilized asfuel oil but has to be burned directly in fluid bedboilers or gasified. Such is the case of the processcalled Deep Thermal Conversion developed by Shell.

Other solutions, and in particular, the Eurekaprocess perfected by Chiyoda Corporation and theHSC process (High-conversion Soaker Cracking)developed by Toyo Engineering Corporation, operatein fields of severity which fall between visbreaking



Table 6. Aquaconversion process performance.Feedstock treated: extra-heavy crude from the Orinoco

Belt (6.5°API)

Visbreaking Aquaconversion

Temperature (°C) Base Base � 5

Conversion (weight %)

Naphtha 2.9 7.5

Distillates 500°C�* 28.2 36.6

Upgrading feedstock

API gravity ofatmospheric residue

3.7 5.4

P-value 1.2 1.2

* This is a way of indicating the fraction of distillates that boilsbelow 500°C.

and coking. In both cases, use is made of steam tocontrol coke formation. The Eureka process, inparticular, is very much like delayed coking in that itoperates with two reactors alternately. The thermalcracking reaction is carried out in the presence of arecycled oil and high temperature steam to favourstripping of the distillates. The cracking residue is afluid pitch at high temperature and, as such, can easilybe recovered from the bottom of the reactor vessel andthen cooled and pelletized (Table 7).

CokingDelayed coking is the technology most often used

nowadays for upgrading heavy crudes and bitumens.The process of coking (see Vol. II, Chapter 5.1)involves heating the feedstock in a furnace and thensending it to reactors known as coking drums,operating at high temperatures (around 500°C) and forextended reaction times. This process promotesthermal cracking of the hydrocarbon structures so as tostimulate the production of gas and distillates frompart of the component with a higher H/C ratio and toleave a carbon residue (coke) in which most of themetals are concentrated (over 90%), and an amount ofsulphur and nitrogen (about 30 and 70%, respectively).

Application of the coking process to heavyfeedstocks becomes relatively simple from a technicalpoint of view, but it involves the production of hugequantities of coke which is a highly polluting materialthat can be utilized as fuel in power generatinginstallations or as feedstock for the production ofhydrogen in gasification plants. The yield of coke can,in fact, be directly correlated to the tendency to formcarbon residues (CCR) in accordance with theformula: coke (% in weight) = 1.6 · CCR.

Table 8 lists the results of treatment by coking of thevacuum residues from three typical extra heavy crudes.

Contrary to what happens with delayed coking, theprocesses of Fluid Coking and Flexicoking (anextension of Fluid Coking) use the coke generated bythermal cracking as a reaction medium and as a carrierof heat. In Flexicoking, the coke is used as a reagent inan integrated gasification reactor where it is gasifiedwith air. In this way, production of pet-coke, theproduct of classic coking plants which is of leastvalue, is eliminated.

Both processes were developed by Exxon in the1950s. The first Fluid Coking plant was set up in 1954at the Exxon headquarters in Billings and, since then, atotal of 18 units have been constructed. The processtakes place in two fluid bed reactors, connected toeach other so as to permit circulation of the coke. Inthe first reactor vessel, the feedstock is converted bymeans of thermal cracking into gas and distillates at atemperature of 510-560°C in the presence of particlesof carbonaceous material on which the coke producedby the reaction is deposited. The solid is then removedfrom the bottom of the first reactor vessel and sent tothe second one, where it is partially burnt to recoverheat needed for the process, while the excess amountis discharged. Alternatively, the residual coke can begasified with air and steam at a temperature of820-900°C in a third reactor vessel (Flexicoking) toproduce a fuel gas with a low heating value



Table 7. Comparison of the conversion yields ofdelayed coking and the Eureka processes. Feedstock

treated: vacuum residue with 5.9°API

Delayed coking Eureka

Conversion (weight %)

Hydrocarbon gases C1-C4 10.4 5.3

Distillates C5-350°C * 39.3 33.6

Vacuum gasoil 16.3 28.4

Coke 34.0 0.0

Pitch 0.0 32.7

* This is a way of indicating the fraction of distillates that goes frompentanes to hydrocarbons that boil at 350°C.

Table 8. Product yields and qualities from delayed coking

Feedstock Zuata Cold Lake Maya

TBP cut 510°C� * 565°C� * 565°C� *

API gravity 2.4 0.4 0.5

Sulphur (weight %) 4.4 6.2 5.8

Product yields (weight %)

Hydrocarbon gases C1-C4 7.5 8.3 8.2

Naphtha 10.0 11.0 11.4

Atmospheric gasoil 23.6 20.3 21.1

Vacuum gasoil 26.1 23.9 25.6

Coke 32.8 36.5 33.4

Distillate characteristics

API gravity 30.3 29.5 28.2


Coke characteristics


Ni � V (ppm) 1,976 1,018 2,296

* This is a way of indicating the fraction of distillates that boilsabove the temperature indicated.

(4.3-5.3 MJ/Nm3) called flexigas, and made up ofCO, H2, N2, CO2 and heterogas (that is, H2S, NH3, etc.).

Besides the advantage of eliminating the cokeproduced, the Flexicoking solution allows for bettercontrol of the cracking process, contributing to ahigher yield of liquids (Table 9).

Other technologies have been developed on thesame principle as Fluid Coking that differ in thesolutions adopted with regards to the reactor or thetype of materials used for transmitting the heat. Forexample, the LR Coker (Lurgi Ruhrgas) uses aconversion reactor inside which a type of rotatingscrew helps to obtain optimum contact between thefeedstock and the heat carrier, allowing the system tooperate like a plug-flow reactor. There is one versionof this technology that has been specifically adaptedfor treating extra heavy feedstocks with a CCR contentof up to 70% (Satcon process).

Finally, a mention should be made of the processknown as Rapid Thermal Processing (RTP), currently

in the development phase in a 1,000 bbl/ddemonstration installation at the Canadian companyEnsyn Group, specifically for the processing of heavyhydrocarbon materials from bitumens to ligneousbiomasses. The process involves a rapid heating of thefeedstock using hot sand at the relatively lowtemperature of 500°C in a transported bed reactor ontowhich the coke is deposited. In this way, it is possibleto process bitumens from oil sands bitumen, producingstable liquids with conversion yields of up to 80%.

Extraction processes: solvent deasphaltingThe fact that most of the metals present in crudes are

concentrated in asphaltenes means that by using SolventDeasphaltating (SDA), it is possible to recoversignificant amounts of partially demetallized anddeasphalted oil (DAO) from the residues which can beprocessed economically in FCC or hydrocracking units.The main advantages of SDA are low investment andoperational costs, while its principal limitation is that toobtain DAO with a low content of pollutants (inparticular, sulphur, nitrogen, metals and CCR), the yieldmust be limited. As a result, significant quantities ofby-products are generated (asphaltenes) that can be usedas components of low quality fuels or as sources ofcarbon for the production of syngas and, therefore,hydrogen in suitable gasification units. The yield and thepollutant content in DAO are, in fact, directly connected,as shown in Fig. 8; therefore, the quality of the productfalls as the yield increases (see also Vol. II, Chapter 7.1).

The application of the SDA process in the extraheavy crudes and bitumens upgrading sector has led tothe development of ad hoc technologies in which the



0

25

50

75

100

0 20 40 60 80 100

cont

amin

ants

dis

trib

utio

n (%

)

DAO yield (weight %)

sulphur

nitrogen

metals

Fig. 8. Comparison of yield vs contaminants content of DAOobtained via deasphaltating.

Table 9. Comparison of product yields and qualitieswhen processing an Arabian Heavy vacuum residue

(1.8°API and 6.0 weight % of sulphur)via coking processes

Delayed coking Flexicoking



Naphtha 13.4 10.8

Atmospheric gasoil 17.9 15.9


Net Coke 40.4 2.7

Flexigas (expressed as fuel oilequivalent)

32.0


API gravity 29.6 23.4

Sulphur (weight %) 3.6 4.1

Coke characteristics

Sulphur (weight %) 6.4

Metals (ppm) 698

Composition of the gases (volume %, dry basis)

N2 53CH4 2H2 15CO 20CO2 10

classic extraction unit is often combined with otherprocesses. This is the case with the OrCrude process,developed specifically for the upgrading of Canadianbitumens by ORMAT Industries. The process involves:a) the distillation of bitumen (atmospheric andvacuum); b) the deasphaltating treatment of theresidue during which the asphaltenes are removed; c)the thermal cracking of the DAO;d ) the recycling of the cracking products into thefeedstock to the atmospheric distillation, so as torecover the distillates and separate the asphaltenecomponent produced by the thermal treatment. In thesame way as with coking, the OrCrude processproduces distillates and a heavy residue (asphaltenes).They are used as feedstock for the gasification plantthat produces syngas to generate energy and the steamneeded for extraction of the bitumens via SAGD(Steam Assisted Gravity Drainage), as well as thehydrogen needed for further upgrading of theproducts. OPTI Canada and Nexen Petroleum aremoving in this direction; a joint venture between thesetwo companies is developing a project (the Long LakeProject) for the recovery and treatment of 70,000 bbl/dof bitumen in the province of Alberta, in Canada(Fig. 9). Applying the process to this type of feedstockwill enable a yield of 60,000 bbl/d of synthetic crudeoil at 22°API which may be further treated indesulphurization/hydrocracking units to produce asweet synthetic crude oil with a gravity of 39°API,while the remaining 3,100 t/d of asphaltenes willbecome the feedstock for the gasification unit.

Catalytic processes: catalytic crackingThe technology of catalytic cracking, conceived in

1936 with the first industrial fixed bed installation, ischaracterized by the fact that it uses acidic catalystsbased on zeolite to assist the reactions for cracking

heavy crudes, capable of producing distillates andnaphtha, in particular. In the configuration, which ismost widely used and known as Fluid CatalyticCracking (FCC), the catalyst is mixed with thefeedstock, then circulated between the riser, the reactorvessel and the regenerator.

Initially, the feed consisted of gasoil, butsubsequently, and particularly with the evolution ofcatalysts, it has been possible to feed FCCs with awide range of hydrocarbons, from naphtha toatmospheric residue.

In general, the size of the hydrocarbon moleculesthat make up the feedstock to an FCC plant is notcompatible with the pores of the zeolitic crystal. Firstof all, these molecules have to be reduced in size by apreliminary cracking that takes place on the externalsurface (matrix) of the zeolite. This matrix can alsohave other functions, such as removal of the metalspresent in the feedstock.

The feedstock enters the riser where it comes intocontact with the regenerated catalyst. This contactleads to the partial vaporization of the feedstock; theoil/catalyst mixture then flows upwards along the riser.The cracking reactions take place almost completelyalong the riser; being globally endothermic, they causea reduction of the temperature. The residence times arebetween 1 and 4 seconds, during which most of thecracking reactions takes place.

In some cases, the feedstock is pre-treated toreduce the content of metals and asphaltenes by meansof extraction with solvents, deasphaltating withpropane, or treatment with hydrogen.

A high content of asphaltenes and/or aromatics inthe feedstock favours the formation and deposit ofcoke on the catalyst, thus reducing its activity.

Over the years, the process has undergonesignificant improvements, as regards both the flow



OrCrudeprocess

SAGD

hydrocracking

premium synthetic crude

hydrogen

steam

asphaltene

bitumen

sour synthetic crude

powergasificationprocess

Athabascaoil sands

Fig. 9. OPTICanada/NexenPetroleum’s Long Lake Project for the exploitation of Canadian bitumens(Zuideveld and de Graaf, 2003).

sheet of the process and the catalysts, making itpossible to improve the selectivity towards theproduction of naphtha to be used in the gasoline pooland increasing its flexibility as far as feedstocks withhigh contents of pollutants are concerned. At atechnical level, improvements have been made in thesystem for distributing the feed in the riser and liftingthe catalyst, in the internal arrangements of thereactor-regenerator system, and in the addition of anexternal heat exchanger, through which part of thecatalyst circulates to remove part of the heat producedby the combustion of the coke deposited on the spentcatalyst.

There have also been new developments in theconfiguration of catalytic cracking, for example, in theone perfected by UOP in conjunction with BAR-COIndustries, which features a very low contact time(MSCC MilliSecond Catalytic Cracking).

The reduced contact time minimizes the formationof gas and coke; this solution makes it possible to feedthe plant with feedstocks (residues) with a highercontent of carbon residue. New catalysts have alsobeen formulated which are more resistant to poisons(nitrogen and metals such as nickel and vanadium).

To maintain a satisfactory level of catalyst activitycirculating within the plant, provision is made for agreater quantity of fresh catalyst to be fed in with acorresponding withdrawal of spent catalyst. The freshcatalyst make-up will also compensate for the loss ofcatalyst fines which are not trapped by the cyclones.Moreover, additives (passivators) have been developedto reduce the deactivation effect of the catalyst causedby the presence of metals. The second improvementmade by UOP goes under the name of X DESIGN. Avessel is added between the reactor and the regeneratorin which the spent and the regenerated catalyst aremixed at a temperature below that of the regeneratedcatalyst. The overall result is a reduction in the thermalreactions in the riser.

To sum up, the improvements enable the catalyticcracking process to be used for the conversion ofrelatively heavy feedstocks, even though fairly strongconstraints persist regarding the feedstock’s content ofmetals and carbon residues that limit its use for thetreatment of non-conventional oils.

2.5.5 Hydrogen additiontechnologies: hydrocracking

Technologies with supported catalystsAs mentioned, the application of hydrocracking

technologies to the upgrading of non-conventional oilscan be strongly affected by the presence of metals andcarbon residue in the feedstock to be treated. These

poisons can actually cause a rapid deactivation of thecatalyst and an increased pressure drop in the reactioncircuit with a subsequent reduction in the cycle length.

To overcome these difficulties, appropriatecatalytic systems can be used which make it possibleto minimize the problems associated with the highcontent of contaminants in the feedstock and/or tomodify the technology, so as to prevent theaccumulation of coke and/or to enable the replacementof part of the catalyst during operation of the plant.

The hydroconversion plants that use supportedcatalysts can be divided into two categories, based onthe technology used: those with fixed bed reactors andthose with expanded (or ebullated) bed reactors.

The plant layout, where fixed bed reactors areused, generally consists of three or more reactors inseries and a fractionating section where the reactoreffluents are separated by atmospheric and vacuumdistillation. The liquid feedstock must be filteredbefore being mixed with the hydrogen and sent to thereactors in order to eliminate the particles of solidspresent and prevent them from being deposited ontothe catalytic beds. The flow through the reactors is ofthe downward type (from the top towards the bottom).

The liquid feedstock and the hydrogen are heatedup in two separate furnaces until they have reached therequired reaction temperature.

One particular innovation developed by Chevronfor fixed bed technology consists of adding anOnstream Catalyst Replacement system (OCR) whichmakes it possible to replace part of the catalyst evenwhile the plant is kept in operation. This configurationconsists of the addition of a reactor upstream of thenormal reaction train that has an upward flow (fromthe bottom to the top) and special internals that allowthe removal of the spent catalyst and the addition offresh catalyst from the top of the reactor. Handling ofthe catalyst is made possible by a special system.

Its extreme sensitivity to high concentrations ofpoisons makes fixed bed technology particularlysuitable for treating the atmospheric residues fromconventional crudes, but not suitable for feedstocksproduced from non-conventional oils.

In a similar way to fixed bed hydrocracking, theflow scheme with ebullated bed reactors is alsogenerally made up of two or three reactors in serieswith a fractionating section where the reactor effluentsare separated by atmospheric and vacuum distillation.

Also in this arrangement, the liquid feedstock andthe hydrogen are heated up separately to the reactiontemperature, and the flow through the reactors is of theupward type. The liquid flow, which will allow thecatalyst to be expanded (ebullated), is ensured by apump that re-circulates part of the liquid collected inthe upper portion of the reactor. The recycle pump can



be installed directly on the bottom head of the reactoror even outside the reactor.

Part of the catalyst contained in the reactors isreplaced on a daily basis in order to keep the activityconstant.

There are two different technologies available onthe market: the first, that keeps its original nameH-Oil, was originally developed by HRI and thenpurchased by Axens, while the second, calledLC-Fining, was developed by ABB Lummus Global.Subsequently, ABB Lummus Global made a strategicalliance with ChevronTexaco to develop and marketthe technology.

Unlike fixed bed technology, ebullated bedtechnology is suitable for treating feedstocks with ahigh content of contaminants and, as such, is used toprocess vacuum residues even from particularly heavyfeedstocks. Furthermore, this solution displays a highdegree of flexibility for feedstocks coming fromdifferent crudes, provides almost constant yields andproduct quality, and has high operational flexibility.Table 10 shows product yields and qualities obtainableby processing vacuum residues from a Canadian

feedstock and from a mixture of Mexican crudes witha more than 60% Maya content.

Technologies with catalysis in dispersed phase(slurry)

These hydrotreatment technologies usenon-supported hydrogenation catalysts finelydispersed in the substrate to be hydrogenated. Thesecatalysts consist of transition metals sulphide andtherefore have no acidic actions, as a result of which,the conversion (cracking) remains purely thermal.

The development of hydrogenation technologieswith dispersed catalysts in the field of upgradingheavy hydrocarbon feedstocks can be traced back toFriedrich Bergius who, in the 1930s, developedprocesses for the hydrogenation of heavy crudes andthe direct liquefaction of coal in which the catalystsconsisted of a mixture of Fe, Al and Ti oxides (seeChapter 2.4). Subsequently, many other catalysts basedon Fe, Mo and, in general, metals of the VIII grouphave been proposed and used, even thoughmicrocrystalline molybdenite generated fromoil-soluble precursors remains by far the most efficientcatalyst. As mentioned earlier, these catalysts are notvery sensitive to poisons and, therefore, do not presentproblems of deactivation through the deposition ofcoke and metals on the pores of the support. Thatmakes them of particular interest when processingfeedstocks which have very high concentrations ofmetals, sulphur, nitrogen and asphaltenes.

The main limitation of dispersed catalysts isundoubtedly the difficulty of recovering them from theunconverted residue. Consequently, the cheapestcatalysts have almost always been preferred(particularly those based on Fe), or solutions that usemore active materials, but in low concentrations, toavoid the separation phase and recovery from theunconverted product. In both cases, the level ofupgrading of the products is medium-low. Thisproblem has strongly influenced the development at anindustrial level of all the technological solutions basedon the use of catalysts in dispersed phase, which areclassified according to the nature of the catalyst, theconfiguration of the reactor, the method of recoveryand recycling of the catalyst, etc. Some of thesesolutions have reached the point of demonstration inpilot or pre-industrial plants, but until now, no realplants have been set up at an industrial level (SFAPacific Inc., 2003).

VEBA CombiCracking process (VCC)This process has resulted from the German

industry’s decades of experience in the field ofhigh pressure hydrogenation of heavyhydrocarbons feedstock and coal. The heart of the



Table 10. Product yields and qualities from ebullated bed

Feedstock AthabascaMexican

blend

TBP cut 540°C+ * 538°C+ *

API gravity 1.9 1.5




Naphtha 5.6 5.6



Vacuum residue 43.2 43.2


API gravity 29.1 29.3


Vacuum residue characteristics

API gravity 5.9 6.0


Metals (ppm) 290 393

* This is a way of indicating the fraction of distillates that boilsabove the temperature indicated.

process is a slurry reactor where cracking andhydrogenation reactions take place in liquid phase(LPH, Liquid Phase Hydrogenation) at hightemperature (450-490°C) and pressure (over 250bar), and in the presence of a catalyst/additivebased on iron/coal that limits the formation of cokeand favours metal removal. This unit is integratedwith a fixed bed reactor operating in gas phase(GPH, Gas Phase Hydrogenation) for furtherhydrogenation of the products.

The VCC technology was developed to the point ofa demonstration plant of 4,000 bbl/d at Ruhr Oel’sBottrop refinery. This unit operated for several years(1988-1993) processing different feedstocks (residues,heavy crudes, coal and plastic residues) and was thendismantled in February 2000. At the moment, nofurther developments of the technology areanticipated, also because of issues related to theincorporation of VEBA Oel by BP.

HDH and HDHPlus processesThe HDH process (Hydrocracking-Distillation-

Hydrotreating) was developed by Intevep-PDVSA inthe 1980s and its scheme follows a similar pattern tothe VEBA CombiCracking process. Compared withthe latter, HDH works at relatively less severeconditions, that is, at a temperature of 420-480°C, 130bar pressure, space velocity (WHSV, Weight HourlySpace Velocity) of 0.4-0.8 h-1, and using an Fesulphate as the catalyst.

The performances reported indicate maximumconversion into distillates of around 95% broken downas follows: naphtha 21%, atmospheric gas oil (AGO)

43% and vacuum gas oil (VGO) 36%. The distilledproducts, however, have to be reprocessed, whereas theresidue is incinerated to recover the catalyst.

The process, initially developed by Intevep usingtwo sizes of pilot plant of 0.3 and 3 bbl/d, was laterdeveloped on a larger scale (150 bbl/d plant) using theVEBA Oel facilities in the Bottrop refinery where,from 1986 to 1988, several trial campaigns wereconducted for over 7,000 hours using Morichal crude.

In the most recent version, called HDHPlus, thecatalyst also contains Mo incorporated on coke used asa support, so as to improve the degree of dispersionand to make it easier to separate it from the heavyresidue (pitch) after the reaction.

Later, in the period 1992-1997, a survey wascarried out to set up an industrial plant of 15,000 bbl/dcapacity at the Venezuelan Cardon refinery to processthe heavy crude from the Orinoco Belt.

Canmet processThe process was initially studied by NCUT (the

National Center for Upgrading Technology) in the1970s and was later developed by Petro-Canada andSNC-Lavalin. The hydrocracking reactor is an emptyvessel in which the feedstock is processed in thepresence of an Fe sulphide based catalyst deposited onparticles of coal. Because the desulphuration activityof Fe is very low, Mo can be added at a level of tens ofppm in the form of naphthenate.

The reaction products are fractionated and sent tothe hydrotreatment unit (Unifining and Unicracking),while the unconverted residue (5-10% of thefeedstock) can be burnt or gasified.

The performances recorded in the treatment ofCanadian bitumen for different degrees of conversionunder the typical reaction conditions of thehydrocracking unit, that is temperature of 400-490°C,pressure up to 140 bar and LHSV ( Liquid HourlySpace Velocity) in the range of 0.5-2 h�1, are shown inTable 11.

In 1986, a demonstration unit with a capacity of5,000 bbl/d was set up at the Montreal refinery whichis still working.

Microcat-RC processThe process, originally called M-Coke, was

developed by Exxon during the 1980s and 1990s up tothe present configuration which consists of:• A hydrocracking stage in the presence of

molybdenum based micronic catalysts supportedon coal particles in dispersed phase (slurry); thereaction is carried out in a reactor that operateswithin a temperature range of 440-470°C, at ahydrogen pressure that can reach 170 bar, withLHSV in a range from 0.5 to 2 h�1.



Table 11. Canmet process performance. Feedstocktreated: Cold Lake vacuum residue

Conversion 525°C� * (weight %) 84.2 93.5



Naphtha 15.8 19.8



Upgrading feedstock

% HDS 62 70

% HDN 31 41

Hydrogen consumption (weight %) 1.6 2.5

* This is a way of indicating the fraction of distillates that boilsabove 525°C.

• The separation of the reaction products throughflash and vacuum distillation to obtain a distillatewith a high yield of the fraction that boils below560°C and to produce a residue in which thecatalyst and all the metals contained in thefeedstock are concentrated (this residue can beburnt or gasified).

• The hydrotreatment of the distilled fraction in afixed bed reactor for the finishing of the productsthat are subsequently fractionated into naphtha,atmospheric gasoil and vacuum gasoil.This process can reach conversions of up to 95%

with a typical product distribution of naphtha 10-15%,AGO 45-55% and VGO 30-40%. It was tested at a pilotplant level of 8 bbl/d at the beginning of the 1990s, butthere have been no further developments of thetechnology even though ExxonMobil remains active inthe field, at least judging by the patents being filed.

(HC)3 processThe process called (HC)3 (High

Conversion/Hydrocracking/Homogeneous Catalyst)was developed at the end of the 1980s by the AlbertaOil Sands Technology & Research Authority(AOSTRA) and the Alberta Research Council (ARC).

The upgrading reaction is carried out in thepresence of a catalyst defined as ‘colloidal’, formed insitu starting with oil-soluble precursors such as ironpentacarbonyl or molybdenum 2-ethylhexanoate and, ifnecessary, by an aromatic diluent to prevent asphalteneprecipitation. The conversion product is fractionatedunder vacuum and the residue can be recycled.

The operating conditions of the hydrocracking unitcall for a reference temperature of around 450°C and apressure of 140 bar. Also for this process, conversion

levels (of the fraction that boils above 500°C) into gas,light distillates and heavy distillates can be obtainedup to 95%.

The process was developed at a pilot plant level of1 bbl/d (Fig. 10). During 2002, a patent agreement wasreached between ARC and Hydrocarbon Research, acompany controlled by Headwaters, to develop andmarket the (HC)3 technology.

Eni Slurry Technology (EST) ProcessThe Eni Slurry Technology (EST) process was

developed recently by Snamprogetti andEniTecnologie, both companies of the Eni Group(Montanari et al., 2003). Unlike the technologiesavailable today, EST operates in such a way as topermit the almost complete conversion of heavypetroleum feedstocks into distillates and avoid theproduction of fuel oil and coke.



feed

reac

tor

hot s

epar

ator

vacu

umto

wer

mix

ed f

eed

hydr

otre

ater

�370°C

�370°C

cata

lyst

prec

urso

r

make-upH2

recyclegas

recyclegas

recyclegas

wash waterlean amine

fuel gas

SCO

sour water

H2S rich amine

recycle

residue

conditioning

Fig. 10. Flow sheet for (HC)3 technology(Lott and Lee, 2002).

reac

tor

frac

tion

ator

SD

A

purgeasphaltene recycle and catalyst

DAOdistillates

feedstock

hydrogen

Fig. 11. Flow sheet for EST technology (Montanari et al., 2003).

The heart of the process consists of a hydrotreating(HT) reactor in which the heavy crude feedstockundergoes a hydrogenation treatment in relatively mildconditions (410-420°C and 160 bar), limiting theconversion per pass to distillates, but ensuring asatisfactory margin of stability for the unconvertedresidue. The hydrotreatment is carried out in thepresence of several thousand ppm of a molybdenumbased catalyst finely dispersed in the liquid mass so asto promote the upgrading reactions (metal removal,desulphuration, denitrogenation and reduction of thecarbon residue). The hydrotreated products leaving theHT unit are sent to a fractionating section to recoverthe distillates. The unconverted residue from thebottom of the fractionating column is then sent to asolvent deasphaltating section (SDA) to recover thedeasphalted and demetallized oil (DAO), while theasphaltene stream that contains all of the catalyst isrecycled to the HT section to be reprocessed togetherwith additional fresh feedstock (Fig. 11). After anumber of recyclings, a stationary steady-statecondition is reached that makes it possible to obtainalmost total conversion levels, overcoming thetraditional limitations of the classical conversionprocesses, i.e. the loss of stability of the reactionproducts and, consequently, the deposition of coke.

The technical validity of the EST process wasdemonstrated through an experimental activityconducted at a pilot plant level of 0.3 bbl/d capacity.

The process proved extremely flexible in treatingdifferent types of heavy feedstocks, such as vacuumresidues from conventional crudes (Ural and ArabianHeavy), heavy crudes (Maya) and extra heavy crudes(from Orinoco), as well as bitumens from oil sands(Athabasca). In all cases, the technical validity of theprocess was confirmed, especially regarding thecatalyst life, the recyclability of the asphaltenes andthe minimum amount of purge stream needed toprevent the accumulation in the plant of the metalscontained in the feedstock. The EST process enablesan almost complete conversion of the feedstock(�95%) and guarantees an excellent level ofupgrading of the products (Table 12). The technologyis in an advanced phase of development in acommercial demonstration plant of 1,200 bbl/d insidethe battery limits of an Eni refinery.

Other processesIn addition to the processes discussed above, there

are other industrial initiatives that broaden the range oftechnologies potentially available, even though at themoment, they are in the early stages of development.

Among these, there is the Hydrogen TransferCracking process, developed by Toyo EngineeringCorporation (Japan) in the early 1990s. It is based onthe use of iron catalysts on active coals that act so as toadsorb and desorb the radicals produced by thecracking, thus preventing the formation of coke. The



Table 12. EST process performances

Feedstock Zuata Maya Athabasca Bitumen

TBP cut 530°C� * 500°C� * 300°C� *

API gravity 2.5 1.5 7.9



Hydrocarbon gases C1-C4 15.1 9.9 12.9

Naphtha 14.0 3.9 4.1

Atmospheric gasoil 39.1 26.9 39.1

Vacuum gasoil 23.3 34.9 32.1

DAO 8.5 24.4 11.8

Products Upgrading

% HDS 86 84 83

% HDN 59 52 47

% HDM �99 �99 �99

% removal of CCR 98 96 95

* This is a way of indicating the fraction of distillates that boils above the temperature indicated.

process operates at relatively low pressures (70-100bar) but, nevertheless, makes it possible to obtain goodlevels of upgrading even when processing very heavyfeedstocks, such as vacuum residues from ArabianHeavy and Maya or the extra heavy Cerro Negro. Theperformances reported indicate maximum conversionsof between 85 and 90%, with good levels of upgrading(HDM and HDS around 95 and 80%, respectively).Toyo recently reached an agreement with the InstitutoMexicano del Petróleo (a company affiliated toPEMEX) to build a demonstration unit at Tula.

Another process proposed by Nikko Consultingand Engineering Corporation is the Succeed Processthat operates in two stages: hydrocracking of thefeedstock with catalysts in slurry phase using adispersed catalyst based on transition metals to obtaina conversion of 60-65%; coking of the part that hasnot reacted so as to reach a total conversion of 85%.The Japanese company is planning to build ademonstration unit of 5,000 bbl/d.

The CASH process (Chevron’s Activated SlurryHydrocracking), proposed by ChevronTexaco, is theresult of a long period of research activity carried outin the 1980s and recently restarted. CASH is based onthe use of a dispersed catalyst based on molybdenumin the presence of nickel as a promoter. This catalytic

formulation seems particularly suitable for facilitatingnitrogen removal. The process also foresees recyclingof the catalyst from the bottom of the distillationcolumn. The operating conditions reported by Chevroncover a very wide range of conditions: temperature400-480°C, pressure 95-130 bar and concentration ofthe catalyst which can go up to 10,000 ppm from 500ppm. Development of the CASH process is currentlyat the stage of a small pilot plant (0.1 bbl/d).

Finally, the technology called GNO-V, proposed byGenoil, is also in the development stage at a pilot plantlevel. The hydrogenation process operates withcatalysts in slurry phase at 400°C and 130 bar, andmakes it possible to upgrade Canadian bitumen,raising the specific gravity from 7 to 28°API, withyields in volume above 100% and reductions ofsulphur from 5 to 0.2%. In recent years, Genoil hasfinalized agreements with Syneco Energy (a Canadiancompany owning reserves of non-conventional oilsand coal) and ConocoPhillips to carry outdemonstration tests on a 6 bbl/d pilot unit, theconstruction of which was started in January 2003 atTwo Hills (Alberta, Canada).

2.5.6 Technologies for theexploitation of oil shale

The first utilization of oil shale goes back toseventeenth century, when, in Sweden, this rock wasroasted to extract sulphates for use as dyestuff. Thefirst exploitation for energy usage, however, is dated atabout the middle of the Nineteenth century with thestarting up of mines and the first processing plants inPennsylvania which were developed in the followingcentury (Fig. 12).

In recent decades, the exploitation of oil shale hasbeen concentrated in just a few countries, reaching apeak of 45 million t/a worked in 1980; assuming a yieldin oil of 10%, that equates to an annual production ofabout 4 million t. In the two following decades, thisfigure fell significantly, to the extent that in 2002 theworldwide production of oil from shale (thereforecalled shale oil) was less than 600,000 t. The country inwhich, by far, the greatest consumption of oil shale wasrecorded is Estonia, where even today, a good part oftheir energy needs are met from this source; in thiscase, however, the oily mineral is used almostexclusively as fuel in fossil-fuelled power stations.

The traditional technologies for the exploitation ofoil shale involve three principal phases: the recoveryof the mineral in mines; its thermal treatment for theproduction of gas and crude oil; hydrotreatment(upgrading) of the liquids up to the production ofdistillates for the fuel market.



Fig. 12. Shale oil retorting furnace with a capacity of 200-330 t/d built in the United States in the 1940s (USBM).

Recovery of the oil from the rock in which it iscontained passes through a treatment of pyrolysiscarried out at high temperature for which differentsolutions have been developed over the years (Johnsonet al., 2004).

The process, known as retorting, starts at 200°C,finishing at higher temperatures (500-600°C), and iscarried out in reactor vessels called retorts.

The main objective in the design of a retort is toheat at low cost by using as little energy as possible.There are different types of retorts distinguished bythe way the heat is applied to the crushed oil shale:retorts heated indirectly by gas; retorts heated directlyby gas by internal and external combustion; retortsheated directly by solids (Matar, 1982).

In the first type of retort, the heating comes fromoutside and the heat is applied through a wall,generally of cast iron. Because of its limited capacityand the low thermal efficiency of the method, this typeof retort has not been developed further. The secondtype uses hot gases that pass directly through the oilshale, usually in a vertical furnace (Union B Processdeveloped by Unocal). The gases can be generatedthrough the combustion of a portion of thehydrocarbons produced within the retort (internalcombustion) or by partial gasification with air andsteam at 800-850°C of the carbon residue that remainson the mineral after the pyrolysis treatment (SGRtechnology, Stream Gas Recirculation). This is a lowcost method, but its yield in hydrocarbons iscorrespondingly low. In the third type of retort,preheated solid carriers, for example ceramic balls 0.5inch (1,27 cm) in diameter, are mixed with the oilshale and provide the heat necessary for the pyrolysis.The first process carried out using this solution,known as TOSCO (The Oil Shale Corporation) II, wasdeveloped at the end of the 1960s by a consortium ofseveral companies. It uses a rotating reactor operatingat 500°C that is fed continuously with the mineral, andinvolves a rather complex system of circulation of thecarrier, but offers advantages in terms of the oil yield.

A conceptually similar technology was laterdeveloped by UMA Engineering (Alberta TaciukProcess, ATP), and is currently considered thereference technology for this type of application. Inthis process, the mineral is first pyrolyzed in a rotatingoven in order to produce oil (500°C) and is then sentto a second combustion furnace where the organicmaterial which has not reacted is burnt (750°C),before being partially returned into circulation so as torecover the heat needed for the retorting.

The gas produced during the retorting consistsmainly of carbon monoxide and carbon dioxide, as aresult of the release of most of the oxygen present inthe initial kerogen, hydrogen, hydrogen sulphide,

methane and heavier hydrocarbons, and can be usedfor the production of hydrogen needed for theupgrading processes or as fuel gas.

The mineral residue, generally waste material, has tobe transported back to the mine or dumped in a rubbishtip. In some cases, depending on the nature of thefeedstock and the method of pyrolysis, the mineralresidue may still contain a considerable quantity oforganic material and can, therefore, be burned to supplyheat for the process. Similar to what takes place incoking furnaces, the pyrolysis reaction, in fact, producesa sort of disproportionation of the hydrocarbon matrix,producing a liquid enriched with hydrogen and a carbonresidue (coke) that remains in the mineral.

The product generated from the retorting of oilshale is a dark, viscous liquid with a high content ofheteroatoms and unsaturated compounds producedduring the pyrolysis. Compared with heavy crudes orbitumen, the liquids from oil shale display a highconcentration of hydrocarbon structures with a highH/C ratio, owing to the fact that the initial kerogencontains a high percentage of paraffin structures.

The distribution of the distillates in shale oils islinked to the retorting technology used but, in general,it is similar to that of the products from the coking ofpetroleum residues.

As regards the heteroatom content, shale oils arecharacterized by the fact that, in addition to sulphur,they contain significant quantities of oxygen andnitrogen. The pyrolysis liquids can include several tensof ppm of metals and, amongst these, iron inparticular, but also traces of nickel, vanadium andarsenic; the last element can create seriouscontamination problems for the hydrogenationcatalysts. In fact, the upgrading of shale oils toproduce fuels has its roots in the classic technologiesdeveloped in the field of petroleum refining and, inparticular, hydrotreatments.

An alternative solution to convert the oil shale intoliquids is known as in situ conversion. This type oftechnology was initially put forward in the USA aroundthe 1960s by several mining companies, among them inparticular, Occidental Oil Shale. The idea was to injectair and steam into the deposit, causing partialcombustion of the organic material (in situ combustion)and, therefore, the heating up of the formation, reachingtemperatures of 700-800°C (Braun et al., 1984). In thisway, the pyrolysis process is carried out directly in thedeposit, from which the conversion products can thenbe recovered using appropriate production wells. Withthe aim of optimizing the technology, several solutionshave been proposed that relate, above all, to the depositexploitation methods, i.e. the layout of the injector andproducer wells, but the different trials have alwaysexperienced serious difficulties in controlling the



combustion front and, hence, in guaranteeing areasonable level of productivity of the oil. A differentapproach is currently being evaluated that provides forthe boring of a series of vertical wells into the oil shaledeposit, which are then heated either electrically, or bymeans of superheated steam, or using the hot gasesproduced by partial combustion of the organic material.In this way, the formation is brought up to a temperaturein the order of some hundreds of degrees Celsius andkept in this condition for several years. The hightemperature accelerates the natural process ofdegradation of the kerogen into oil (at about 300°C, thedegradation kinetics of kerogen increase significantly;Lewan et al., 1979) that can then be extracted liketraditional crude from a production well. Theconversion yields are lower than that which could beobtained with traditional retorting (about 20% in weightcompared to the organic material), but the quality of theoil produced is decidedly superior, as shown in Table 13and, above all, the environmental problems linked torecovery of the mineral and disposal of the processingresidues are significantly reduced.

2.5.7 Main development projects

At present, there are numerous projects (started or atan advanced stage of development) relative to the

industrial exploitation of non-conventional oils.As far as extra heavy crudes are concerned, in

Venezuela four projects for exploitation of the crudesfrom the Orinoco Belt have been carried out for a totalof 634,000 bbl/d, with production of 573,000 bbl/d ofsynthetic crude with API gravities varying from 16 to32. Negotiations are taking place with the Venezuelanstate oil company (PDVSA) for an increase inproduction to about 2 million barrels a day by2010-2012. Table 14 shows the most significant datarelating to these projects.

In Canada, in the province of Alberta, about 1.1million bbl/d of oil sands are treated annually toproduce, in part, synthetic crudes and, in part, dilutedbitumen with a diluent (consisting generally ofnaphtha) in order to reduce its viscosity to enable it tobe pumpable to allow transportation. An updated listof the main industrial projects is shown in Table 15.

Finally, as far as oil shale is concerned, at aninternational level, there are several research anddevelopment initiatives aimed at experimentingsolutions for its exploitation, particularly in Canada,Australia, China, Russia and Israel, but applications atan industrial or demonstration level are still very limited.

Amongst these, a project worthy of note is the onebeing run by two Australian companies, SouthernPacific Petroleum NL and Central Pacific MineralsNL who, in 1995, signed an agreement with the



Table 13. Comparison of product yields and qualities when processing American oil shaleby means of different processes

Traditional retorting TOSCO process Union Oil process Shell ICP process

Oil characteristics

API gravity 19.8 21.2 18.6 38

H/C 1.63 1.64 1.70 1.84

Sulphur (weight %) 0.7 0.9 0.9 0.5

Nitrogen (weight %) 2.1 1.9 2.0 1.0

Oxygen (weight %) 1.6 0.8 0.9 0.5

Bromine number* 33 49 n/a n/a

Ni � V (ppm) 10 9 5 2

Fe (ppm) 108 100 55 9

Distillate distribution (volume %) **

C5-230°C 11 23 5 45

230-343°C 25 21 25 39

343-554°C 54 56 70 16

554°C� 10 – – –

* The bromine number is a parameter that quantifies the olefin content in a hydrocarbon cut.** The distillates are divided into groups starting with pentanes through to hydrocarbons which boil at the temperatures indicated.



Table 14. Projects for the exploitation of extra heavy Venezuelan crudes

Petrozuata Cerro Negro Sincor Hamaca

Operator Petrozuata ExxonMobil Sincor Petrolera Ameriven

Percentage participationConocoPhillips 50.1PDVSA 49.9

ExxonMobil 41.7PDVSA 41.7BP 16.6

Total 47PDVSA 38Statoil 15

ConocoPhillips 40PDVSA 30ChevronTexaco 30

Reserves (gbbl) 1.4 1.4 2.3 2.2

Total CAPEX (billions of dollars) 3.9 2.0 4.6 4.5

Upgrader CAPEX (billions of dollars)

1.5 0.7 2.5 2.0

Production of heavy oilinto upgrader (bbl/d)

120,000 120,000 204,000 190,000

Syncrude productionout of upgrader (bbl/d)

108,000 105,000 180,000 180,000

Syncrude density (°API) 20-26 16 32 26

Upgrader technologyDelayed coking

Naphtha hydrotreatingDelayed coking

Naphtha hydrotreating

Delayed cokingHydrocrackingHydrotreating

Delayed cokingHydrocrackingHydrotreating

Table 15. Main Canadian projects for the industrial exploitation of non-conventional oils

Company Project Capacity (bbl/d)Cost

(millions of dollars)Status Start-up

Shell Canada, Chevron Canadaand Western Oil Sands

Muskeg River 155,000 5,700 Operative 2003

Syncrude Canada Aurora 140,000 5,000 Under construction 2005

Suncor Energy Inc. Firebag 105,000 2,100 Under construction 2007

Petro-Canada Meadow Creek 80,000 800 Proposed 2007

Nexen Petroleum & OPTICanada

Long Lake 30,000 2,600 Proposed 2007

Imperial Oil Nabiye 30,000 1,000 Proposed 2006

TrueNorth Energy & UTS Energy

Fort Hills 95,000 3,500 On hold ?

Japan Canada Oil SandsLtd.

Hangingstone 35,000 250Preliminary proposal

filed2007

Husky Energy Inc. Tucker Lake 30,000 350 Proposed 2006

EnCana Corp. Christina Lake 70,000 900 Pilot 2007

Devon Canada Corp. Jackfish 35,000 400 Preparing proposal 2007

Syneco Energy Northern Lights 80,000 3,500Preliminary proposal

filed2007

ConocoPhillips Canada,TotalFina & Devon Energy

Surmont 100,000 1,000 Proposed 2006

Canadian Natural ResourcesLtd.

Horizon 232,000 8,000 Proposed 2011

Deer Creek Energy & Enerplus Resources Fund

Joslyn Creek 40,000 450 Preparing proposal 2008

Black Rock Ventures Hilda Lake 20,000 260 Proposed 2005

Canadian company, Suncor Energy, aimed atexploiting the Australian deposit known as Stuart. Thefirst phase of this project was completed in 1999 withthe creation of an industrial complex for the treatmentof 6,000 t/d of mineral, from which 4,500 bbl/d of alight oil (42°API) with a sulphur content of 0.4weight % and a nitrogen content of 1 weight % isproduced. The processing technology used forextraction of the oil is the previously mentionedAlberta Taciuk Processor (ATP).

In Estonia, there are currently three retortinginstallations in operation that produce 8,000 bbl/d ofliquid. Looking to the future, the private chemicalcompany, Viru Keemia Grupp (VKG), has plans forthe setting up of a 4 million bbl/a plant for distillates(naphtha and diesel) whose profitability would beincreased by the possibility of extracting oxygenatedderivatives for the chemical industry (in particular,phenols and cresols) from the pyrolysis oil.

Finally, it should be remembered that in Brazil, oneof the biggest retorting plants is already in operationusing gas combustion (Petrosix), through which 3,870bbl of liquid are produced daily.

As far as the in situ conversion process isconcerned, in the last fifteen years, several studies andtests have been carried out in the field to optimize thetechnology. Particularly worthy of note is the work thatis being carried out by Shell on the development of thetechnology referred to as ICP (In situ ConversionProcess), which is in an advanced experimentationphase in the Piceance Creek basin in Colorado, while anew project has been announced in Jilin province inNorth East China.

References

ALDRIDGE C.L., BEARDEN R. JR. (1978) US Patent 4066530 toExxon Research Engineering.

Altgelt K.H., Boduszynski M.M. (1993) Composition andanalysis of heavy petroleum fractions, New York, MarcelDekker.

Barton A.F.M. (1991) CRC handbook of solubility parametersand other cohesion parameters, Boca Raton (FL), CRC.

Braun R.L. et al. (1984) Results of mathematical modellingof oil shale retorting in an aboveground, internal combustionretort, «Fuel Processing Technology», 9, 125-138.

Bunger J.W. et al. (2004) Is oil shale American’s answer topeak-oil challenge?, «Oil & Gas Journal», August, 16-24.

Cimino R. et al. (1995) Solubility and phase behaviour ofasphaltenes in hydrocarbon media, in: Sheu E.Y., MullinsO.V. (edited by) Asphaltenes. Fundamentals andapplications, New York, Plenum Press.

Delbianco A. et al. (1995) Upgrading heavy oil using slurryprocessing, «CHEMTECH», November, 35-43.

Dyni J.R. (2004) Oil shale. World energy council, in: Surveyof energy resources 2004, Amsterdam, Elsevier.

IEA (International Energy Agency) (2004) World energy outlook2004, Paris, Organization for Economic Cooperation andDevelopment/IEA.

Johnson H.R. et al. (2004) Strategic significance of America’soil shale resource, 2v.: Volume II: Oil shale resources,technology and economics, US DOE Office of navalpetroleum and oil shale reserves.

Lewan M.D. et al. (1979) Generation of oil-like pyrolysatesfrom organic rich shales, «Science», 203, 897-899.

Lille Ü. et al. (2003) Molecular model of Estonian kukersitekerogen evaluated by 13C MAS NMR spectra, «Fuel», 82,799-804.

Long R.B. (1979) The concept of asphaltenes, in: Proceedingsof the 178th meeting of the American Chemical Society,Washington (D.C.), 10-11 September, 891-900.

Lott R., Lee L.K. (2002) Upgrading of heavy crude oils andresidues with (HC)3TM hydrocracking technologies,Lawrenceville (NJ), Hydrocarbon Technology Inc.

Matar S. (1982) Synfuels. Hydrocarbon of the future, Tulsa(OK), PennWell.

Montanari R. et al. (2003) Production of ULS fuels fromheavy feedstocks, «Petroleum Technology Quarterly»,Winter, 61-67.

Murgich J. et al. (1996) Molecular recognition and molecularmechanics of micelles of some asphaltenes and resins,«Energy & Fuels», 10, 68-76.

Panariti N. et al. (2000) Petroleum residue upgrading withdispersed catalysts. I: Catalyst activity and selectivity,«Applied Catalysis A. General», 204, 203-213.

Perrodon A. et al. (1998) The world’s non-conventional oil& gas, «Petroleum Economist», March.

SFA Pacific Inc. (2003) Upgrading heavy crude oils andresidues to transportation fuels. Technology, economics,and outlook. Phase 7, Multisponsored reports, Petroleumprocessing program, SFA Pacific Inc., 4, 39-48.

Sheremata J.M. et al. (2004) Quantitative molecularrepresentation and sequential optimization of Athabascaasphaltenes, «Energy and Fuels», 18, 1377-1384.

Speight J.G. (1980) The chemistry and technology of petroleum,New York, Marcel Dekker.

Speight J.G. (2004) Petroleum asphaltenes. I: Asphaltenes,resins, and the structure of petroleum, «Oil & Gas Scienceand Technology», 59, 467-477.

Tominaga H., Tomaki M. (1997) Chemical reactor and reactordesign, New York, John Wiley.

Zuideveld P., de Graaf J. (2003) Overview of Shell globalsolutions’ worldwide gasification developments, in:Proceedings of the Gasification technology conference,San Francisco (CA), 12-15 October.

Alberto DelbiancoEniTecnologie

San Donato Milanese, Milano, Italy

Romolo MontanariSnamprogetti

San Donato Milanese, Milano, Italy



Emerging Technologies for the Conversion of Residues

Documents

volume iii

alternative fossil resources

toyo engineering corporation

hydrocarbon gases c1

fixed bed reactors

veba combicracking process

retorts heated directly

oil sands bitumen