Nguyen V. Dung

Yields and chemical characteristics of products from fluidized bed steam retorting of Condor and Stuart oil shales: effect of pyrolysis temperature

Abstract Condor and Stuart oil shales were pyrolysed with superheated steam at 450–600 °C in a 48 mm diameter semi-continuous fluidized-bed reactor. Temperature had a significant influence on the yields of products and on the ratio of light oils (C5 − 340 °C) to heavy oils (+ 340 °C). By contrast, the organic carbon conversions to char in spent shale were constant, and the elemental composition of oils changed only slightly over the whole temperature range. The yields of C1-C4 alkanes increased linearly with temperature. The relationships between temperature and the yields of other gases (C2-C4 alkenes, hydrogen and carbon oxides) were also linear, but with a transition from a low rate of increase to a much higher rate above a certain temperature. This transition temperature was 525 °C for Stuart and 550 °C for Condor oil shale. Oil yields, which appeared to be constant in the low temperature range, decreased rapidly above the transition temperature. Compared with experiments using nitrogen as the sweep gas, steam pyrolysis produced higher oil yields in the low temperature range. However, above the transition temperature steam was partly responsible for increases in hydrogen and carbon dioxide production at the expense of oil. This would appear to result from steam gasification of kerogen/bitumen/oil vapour catalysed by shale minerals. Low oil vapour concentration in the retort was also found to be essential to reduce oil coking losses within shale particles.

Continuous fluidized bed retorting of Condor and Stuart oil shales in a 150 mm diameter reactor

Abstract Pyrolysis of Condor and Stuart oil shales in a continuous fluidized bed reactor heated by through-wall thermal conduction gave oil yields close to modified Fischer Assay; however, when these shales were pyrolysed in the same reactor with an ash-to-shale recycle ratio of two, there was up to 28 % oil loss. The lost oils were mainly heavy fractions which adsorbed onto shale ash. The main reaction appeared to be coking. The high internal surface areas of the Condor and Stuart ashes are possibly a key factor in the reduction of oil yield. Options for reducing this loss are outlined.

Factors affecting product yields and oil quality during retorting of Stuart oil shale with recycled shale: a screening study

Abstract Seven process variables—retort temperature, solids recycle ratio, char content of recycle solids, recycle solids temperature, pretreatment of recycle solids with ammonia, solids residence time and steam concentration—were studied. The Plackett-Burman statistical experimental design, highly effective for detecting main effects of large number of variables with a minimum number of experiments, was used to screen these variables. Each variable was tested at low and high levels. The bench-scale integrated retorting-combustion oil shale (BIRCOS) facility at Lucas Heights was used to obtain product yields and compositions under realistic recycle shale conditions. The significance of the main effects of the variables on product yields and carbon conversions decreased in the order: retort temperature, solids residence time, steam concentration, char content, solids recycle ratio, recycle solids temperature, and ammonia treatment. Retort temperature was the only variable to affect the H C atomic and nitrogen content of the oil. The sulfur content of the oil was independent of all variables.

Pyrolysis of stuart oil shale in the presence of recycled shale

Abstract The pyrolysis of Stuart oil shale in the presence/absence of fully combusted shale (shale ash) in superheated steam or nitrogen was conducted in the range 450 to 600 °C in a 48 mm diameter semi-continuous fluidized-bed reactor. The detrimental effect of shale ash on oil yields in a recycled shale retorting system was confirmed. The coking of product oil vapour on shale ash led to oil losses that increased with increase in the ash-to-shale ratio. Reduction in surface area of the shale ash by high temperature treatment lowered oil coking losses. However, while pyrolysis in steam produced more oil than pyrolysis in nitrogen, the type of sweep gas (steam versus nitrogen) did not affect the rate of oil coking on shale ash. Certain types of clay minerals in the shale and changes in their structure during the course of pyrolysis/combustion were postulated as having a significant influence on the extent of the oil coking. Oil coking losses on shale ash, which were largely independent of temperature in the range 450 to 525 °C, increased significantly with temperature above 525 °C. Except for methane and hydrogen, yields of gaseous products and the atomic H C ratio and sulphur content of the oils were unchanged by the coking reactions.

Pyrolysis behaviour of Australian oil shales in a fluidized bed reactor and in a material balance modified Fischer assay retort

Abstract The pyrolysis behaviour of several Australian oil shales was determined using the material balance modified Fischer assay and a bench-scale fluidized bed pyrolysis reactor, with nitrogen or steam as the sweep gas. The assay oil yield, which ranged from 5.3 to 15.7 wt% of the dry shale, did not correlate well with the organic carbon contents of the shales. However, under both assay and fluidized bed pyrolysis conditions, a shale of high kerogen H C had high organic carbon conversions to oil. Compared with the Fischer assay, nitrogen pyrolysis gave 7 ± 4 wt% more oil for the shales studied, and steam pyrolysis gave 15 ± 4 wt% more oil for all shales except one, which showed a 35 wt% increase in oil yield. However, the oils from both nitrogen and steam pyrolysis had lower H C s, higher sulphur and nitrogen contents, and more high boiling point fractions than those from the Fischer assay. Nitrogen pyrolysis oils were of higher quality than those produced by steam pyrolysis. With steam as the sweep gas, much more hydrogen and hydrogen sulphide were produced for all shales; in most cases, there was also more carbon monoxide and less hydrocarbon gases.

Basis of reactor design for retorting Australian oil shales

Abstract Correct selection and design of the retort is of vital importance to the technical and commercial success of an oil shale conversion process. A set of performance characteristics for an efficient retort was therefore identified through a review of oil shale retorting technology and analysis of the processing properties of Australian feedstocks. Based on these characteristics and aided by computer modelling, a retort was designed. A description of the recently constructed integrated retorting/combustion facility and a comparison between its measured and predicted performance are presented.

A new concept for retorting oil shales

Abstract A new concept in which oil shale is rapidly heated and retorted in the absence of shale ash is proposed. The shale heater is a bank of horizontal tubes, where shale particles are conveyed in dilute phase, immersed in a fluidized bed of combusting spent shale. The process is illustrated by computer simulation which uses a simple mathematical model. The advantages of this new concept are: small total retort volume; no oil loss associated with the presence of shale ash; small oil loss due to thermal cracking; it is energy self-sufficient; and it has a simple oil condensation system.

Modelling of oil shale retorting block using ASPEN process simulator

Abstract An integrated process model for the retorting block of an oil shale conversion process is presented. The model integrates only the key process operations such as drying/preheating and pyrolysis of raw oil shale, combustion of spent shale and sensible heat recovery from combusted shale. The flowsheeting package ASPEN has been used to achieve mass and energy balances around the process. Performance data for individual process units were generated using detailed reactor models developed previously.

A mathematical model of dilute phase transport combustion of spent oil shale

Abstract A model for the combustion of a mixture of spent oil shale and shale ash for a range of particle sizes in a vertical dilute phase transport contactor is presented. The “grain” model for a sphere of unchanging size is used to describe the combustion of a single particle of spent shale. The rate of gas-solid heat transfer is assumed to be controlled by gas film forced convection. The aerodynamic properties, namely terminal and slip velocities of particles and choking conditions, are taken into account. Realistic conditions of inlet air and solids to the combustor are determined from the mass and energy balances for a hypothetical, 50,000 barrels per day (7.95 10 6 1/day), energy self-sufficient oil shale processing plant. The fraction of the largest particle size is shown to govern the aerodynamic characteristics and hence the design of the combustor. The sensitivity of the profile of average particle temperature to a number of shale-related properties is evaluated; for a typical material such as Condor brown spent shale, the ranking in descending order is as follows: heat of combustion, particle specific heat capacity, initial char content, particle terminal velocity, particle density, intrinsic combustion rate coefficient, pore effective diffusivity and choking velocity of the top size particle. The rate of combustion of the Condor spent shale/shale ash mixture as a whole in the contactor is controlled by the intrinsic kinetics.

Processing Stuart oil shale in an integrated retorting/combustion recycle system

Abstract The importance of oil yield to the economics of an oil shale conversion process, and the detrimental effect of recycling solids on oil yield from Australian tertiary shales, are well known. A set of retort characteristics aimed at producing high oil yields has therefore been identified and, aided by computer modelling, used in the design of the new CSIRO integrated retorting/combustion facility. The operating conditions to achieve oil yields close to the Fischer assay yield have been determined both experimentally and through computer simulations.