T-2854
STEAM PYROLYSIS OF A HYTORTED EASTERN
SHALE OIL FOR
CHEMICAL INTERMEDIATE PRODUCTION
By
Steven Paul Sherwood
ARTHUR LARES LIBRARY COLORADO SCHOOL ot MINES GOLDEN, COLORADO 80401 ProQuest Number: 10781147
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a com plete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest
ProQuest 10781147
Published by ProQuest LLC(2018). Copyright of the Dissertation is held by the Author.
All rights reserved. This work is protected against unauthorized copying under Title 17, United States C ode Microform Edition © ProQuest LLC.
ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 T-2354
A thesis submitted to the Faculty and the Board of Trustees
of the Colorado School of Mines in partial fulfillment of the
requirements for the degree of Master of Engineering (Chemical
and Petroleum-Refining Engineer),
Golden, Colorado
Date ///&Q/&*/_ -A Signed:
Steven Paul Sherwood
Approved! ^jr
Dr. Victor Yesavage Thesis Advisor
Golden, Colorado
I ate
DjV A.J. Kiqnay L-epatment Head Chemiacal and Petroleum Refining Engineering T-2854
ABSTRACT
Eastern shale oil from a Hytort oil shale retort was pyrolyzed with steam at atmospheric pressure in a bench scale tubular reactor packed with ceramic balls. The reaction variables and ranges studied were: temperature from 1395°F to 1632°F, residence time from 0.445 seconds to
1.42 seconds, and a steam to oil mass ratio from 0.449 to
0.931.
The maximum yield of olefin product as a weight percent of feed were:
Ethylene 12.54
Propylene 5.14
Total Olefin 20.70
The results of these experiments were correlated with a pyrolysis severity factor which combines the effects of three reaction parameters into a single quantity. Increased severity resulted in increased production of hydrogen and carbon monoxide. Ethylene and olefin yield were found to go through a maximum at a temperature of 1478°F, residence time of 0.530 seconds, and steam to oil mass ratio of 0.449.
Olefin production was much less then other shale oils investigated at Colorado School of Mines. Therfore this oil would not seem to be an appropriate feed for the petro chemical industry.
iii T-2854
TABLE OF CONTENTS
Fage
ABSTRACT...... iii
LIST OF FIGURES...... vi
LIST OF T A B L E S ...... vii
ACKNOWLEDGEMENT ...... viii
INTRODUCTION...... 1
EXPERIMENTAL BASIS...... 6
Chemistry of Pyrolysis ...... 6
Previous Investigations...... 6
Institute of Gas Technology ...... 7
Laramie Energy TechnologyCenter ...... 7
Colorado School of Mines...... 8
DuPont...... 12
Stone and W e bster...... 15
REACTION VARIABLES...... 18
Feed Composition...... 18
Reaction Temperature ...... 18
Residence T i m e ...... 19
Steam to Oil Mass Ratio...... 20
Pyrolysis Severity Factor...... 21
FEEDSTOCK CHARACTERISTICS ...... 22
iv T-2854
TABLE OF CONTENT (cont.) Page
EXPERIMENTAL APPARATUS...... 24
Feed System...... 24.
Reactor System ...... 26
Cooling and Sampling System...... 31
SAFETY FEATURES ...... 33
EXPERIMENTAL PROCEDURE...... 34-
RESULTS AND DISCUSSION...... 37
Effect of Pyrolysis Condition on Yields. . . 37
Effect of Reaction Temperature...... 37
Effect of Residence Time...... 38
Effect of Steam to Oil Mass Ratio . . . 44
Effect of Severity Factor ...... 46
Optimum Ethylene and Olefin Yields ...... 50
Overall Mass Balance...... 52
Elemental Mass Balance ...... 52
Reproducibility of Results ...... 54
Comparison with Previous Studies ...... 57
CONCLUSION...... 59
RECOMMENDATIONS...... 61
REFERENCES CITED...... 62
APPENDICIES...... 67
A. Computer Program Listing and Notation . 67
A1. Sample Calculations and Equations . . . 76
B. Computer Program EOSMP.FOR Data Output. 80
C. Simulation Program Data Output...... 101 v T-2854-
LIST OF FIGURES
Figure Page
1. Experimemtal Apparatus for Pyrolysis System. . . 25
2. Reactor Dimensions ...... 27
3. Reactor S y s t e m...... 29
4.. Thermocouple Location...... 30
5. Effect of Temperature (Wt.? of Feed Vs. Temp.) . 4-0
6. Effect of Temperature (SCF of Feed Vs. Temp.). . 4-1
7. Effect of Residence T i m e ...... 4-3
8. Effect of Griswolds's Severity Factor...... 4-7
9. Effect-of Fritzler’s Severity Fa c t o r ...... 4-8
vi T-2854-
LIST OF TABLES
Table Page
1. Maximum Light Olefin Yields of Tosco II Crude Shale O i l ...... :9
2. Maximum Light Olefin Yields of Whole Shale Oil. . 9
3. Maximum Light Olefin Yields of Distillates. . . . '9
4.. Maximum Light Olefin Yields of Hydrogenated Vacuum Distillates...... 13
5. Maximum Light Olefin Yields of Hydrogenated Shale Oils...... 13
6. Steam Pyrolysis of Paraho Shale O i l ...... 14-
7. Steam Pyrolysis Yeilds (Ruderhausen and Thompson) 16
8. Steam Pyrolysis of Parahoe Shale Oil and Petroleum Gas oil (Korsi, Halle, and Virk)...... 17
9. Physical Properties of Eastern Shale Oil. .... 23
10. Medium Residence Time Results ...... 39
11. Effect of Residence Time on Product Yield .... 4-2
12. Effect of Steam to Oil Mass Ratio on Product Yield 4-5
13. Results of Steam Pyrolysis of Eastern Shale Oil . 4-9
14-. Maximum Olefin Y i e l d s ...... 51
15. Overall Mass Balance...... 53
16. Elemental Mass Balance...... 55
17. Reproducibility of Pyrolysis Run...... 56
18. Comparison with Computer Model...... 58 ACKNOWLEDGMENT
I would like to thand Dr. lesavage and Dr. Dickson for their help with this thesis. I would also like to thank Mike Robinson and Doug Nishamoto for helping with the pyrolysis system.
viii T-2854 1
INTRODUCTION
Recent fluctuation in crude oil supply and cost have stimulated an interest to utilize the vast reserves of oil shale found in the United States as an alternative fuel source. These cyclic economics are a deterrent to some shale oil projects (1), but Union Oil appears to be determined to produce shale oil from Colorado. In order to produce a synthetic fuel from shale oil, process schemes have been developed which involve coking, hydrostabiliza tion, hydrodenitrogenation, reforming and cracking (2,3).
Foreign and domestic shale oils were analyzed by The
Bureau of Mines in Laramie (4). The results indicate that shale oils differ greatly in such properties as sulfur and hydrocarbon composition. The sulfur containing species are evenly distributed throughout the boiling ranges of the oils, while the nitrogen compound concentration in the higher boiling fractions is much higher then the lower boiling fractions (4). Shale oil naphtha is found to contain about 10 percent nitrogen compounds, light distllates (200-310?C) contain about 20 percent, heavy distilaltes (310-430°C) contain about 40 percent and residuum (430°C+) will contain about 85 percent nitrogen species (5,6). Nitrogen compounds found include pyrroles, pyridines, and quinoline homologs, all of which have the T-2854 2
chemical properties of a weak base (7). Typically shale oil will have a nitrogen content of about 2 weight percent
(8,9). Cady et al. (7) report that NTU shale oil consists of 39 percent hydrocarbon compounds and 61 percent nonhydrocarbon species. Of the nonhydrocarbons, 59 percent are nitrogen containing compounds, 10 percent are sulfur containing compounds and 31 percent are oxygen containing compounds.
The basic nitrogen compounds found in shale oil act as poisons to modern acidic type catalysts (2). Since most of the processes in a refinery are catalytic, it is necessary to remove the nitrogen before processing (10). This could be done by hydrodenitrogenation (2), of hydrofining (9).
Cheveron research reports that it would cost between $6.50 and $9.70 (1978 prices) a barrel to refine shale oil. Most of this cost is associated with the initial step of reducing the nitrogern content of the oil (11). Any alternative use for shale oil that does not involve prerefining may be both practical and economical.
An alternative use for shale oil may be as a feedstock for the production of petrochemical intermediates such as ethylene, propylene, butadiene, benzene, toluene and xylene.
Steam pyrolysis of hydrocarbons is the most extensively used method of production of these intermediates (12,13). Since pyrolysis is a noncatalytic process, it does not require expensive prerefining to reduce the nitrogen content of a T-2854 3
shale oil. Pyrolysis to form pertrochemical intermediates may therefore be an attractive alternative use for shale oil.
Until recently natural gas and refinery gas were the major feedstocks for the petrochemical industry in the
United States. In 1970, 85 percent of the national petrochemical feedstock came from these two sources (14).
Light hydrocarbon feedstock produce high yields of ethylene and propylene, but the availability of additional natural gas is not keeping up with light hydrocarbon demand. In the
1980’s light hydrocarbons will produce less then half of the ethylene made in the United States (14), and in the 1990's
70 percent of the ethylene produced will come from liquid feedstock (13).
The petrochemical industry has been involved in looking for an alternative feedstock that will supplement the lagging supply of natural gas. The continuity of a feedstock supply is almost more important then feedstock price (15). Chambers and Potter (16) report that when designing a new pyrolysis plant, maximization of ethylene yield has been the primary objective of current units, however rising demand is increasing the value of other light olefins. Europe and Japan have been using naphtha as their primary feedstock due to the lack of natural gas resources and light hydrocarbon availability (14,17). Because of the large demand for naptha by refineries in the United States, T-2854 4
the petroleum industry has been looking at gas oil as a primary feedstock (14). Current olefin plants crack feeds ranging from ethane to gas oil. Most of these plants were designed to utilize only a single feed. New process plants will have to be flexible enough to handle a larger range of feedstocks (14,18).
The amounts of paraffins, olefins, naphthenes and aromatics (PONA) in an oil can be a good indicator as to how well the oil will perform as a pyrolysis feedstock. It is believed that normal paraffins make the best feedstock for ethylene production, as they are easiest to crack, followed by branched paraffins, olefins, naphthenes and aromatics.
Naphthenes are less easy to crack then paraffins as they must first be opened (19). At low temperatures naphthenes tend to dehydrogenate to form aromatic compounds. Benzene rings pass through the furnace unchanged, while aromatics with sidechains may be hydrodealkylated to form olefins and benzene (20).
A research program has been initiated at The Colorado
School of Mines to study the feasability of using shale oil produced from several different retoring processes as feedstock for steam pyrolysis to produce petrochemical intermediates. Previous studies have looked at shale oil that had been produced from Eocene shale found in the western United States. In this portion of the program an eastern shale oil is pyrolyzed. The shale oil was produced 2854 5
from Devonian shale by means of the IGT Hytort retorting process. Netzel and Miknis (46) analyzed eastern and western shale oils by NMR. The results of this study are given below.
EASTERN WESTERN SHALE OIL SHALE OIL
MOLE % TOTAL AROMATIC 57.0 - 62.9 26.5 - 28.1
MOLE % TOTAL ALKANES 33.9-39.1 66.7-67.1
The objective of this study is to find reaction conditions that will produce maximum ethylene and light olefin yields. These results will then be compared with those obtained from previous studies at The Colorado School of Mines on western shale oil. T-2854 6
EXPERIMENTAL BASIS
Pyrolysis is a highly endothermic process which involves reactions such as thermal rearrangement, decomp- ositon and polymerization (22). Pyrolysis reactions are reported to follow a free radical mechanism (23). The initialization step is a homolysis of a carbon-carbon bond.
R -| CH2-CH2R2 R-| CHg • +R2CH2 •
Further scissions occur at the carbon-carbon bond that is in the beta position, to give ethylene and a smaller radical.
RCH2-CH2-CH2* -> RCH2 * + H2C=CH2
The repetition of this reaction leads to the formation of large amounts of ethylene.
Previous Investigations
Shale oil has been investigated as a potential fuel or petrochemical feedstock since the early 1950's. Early pyrolysis tests were performed on shale oil by The Institue of Gas Technology (24), and at The Laramie Energy Technology
Center of The Department of Energy (25,26). The Colorado
School of Mines has initiated a research program involving steam pyrolysis of various shale oils. Several studies by
Dupont (27,28) and Stone and Webster (29) have looked at pyrolysis of crude shale oil and shale distillates. T-2854 7
Institute of Gas Technology
The studies made by The Institute offf Gas Tchnology involve steam pyrolysis off crude shale oil in a vertical tube reactor. The oil shale was produced from Green River
Formation shale. Experiments were performed at temperatures ranging from 1400°F to 1550°F with reaction times over the
1.0 to 4.5 seconds range and partial pressure off product gases ranging from 0.64 to 0.76 atmosphre. Hydrogen and methane yields increased with reaction time. This was accompanied by lower yields of ethane and heavier Paraffins.
Optimum ethylene was found at a temperature of 1470 f and a reaction time of 2.5 seconds. A pyrolysis severity factor was developed to combine the effects of temperature and reaction time into a single quantity
Laramie Energy Technology Center
A high temperature retort was constructed to crack vapors released from shale during the retorting process
(25). The goal was to produce a naphtha high in aromatics.
Shale was crushed into fine particles to ensure efficient heat transfer .
In another series of experiments (26), shale oil from a conventional retort was cracked for a comparison study with high temperature retorts. These latter series of T-2854 8
experiments are more relevent to the investigations at The
Colorado School of Mines. The reaction times ranged from
2.0 to 3.0 seconds, and temperatures ranged from 950°F to
1200°F. The study reported that high temperature retorting enhanced naphtha production.
Colorado School of Mines
A research program was initiated at The Colorado School of Mines to determine the suitability of shale oil as a feedstock for the petrochemical industry. Griswold (30) pyrolyzed shale oil from the Tosco II retorting process in a tubular reactor packed with ceramic balls. The reaction variables and ranges were: temperature 1300°F to 1600°F, residence time 0.4 to 1.3 seconds, steam to oil mass ratio
0.4 to 1.2. The results were correlated with a modified pyrolysis severity factor which combines the effects of these three variables into a single quantity. The maximum yields of ethylene, propylene, and light olefins are given in Table 1.
Next Fritzler (31) pyrolyzed crude shale oil from the
Tosco II process and an NTU simulated in-situ oil shale retort. The reaction variable ranges in this study were: temperature from 1348^F to 1637°F, residence time from .3 to
1.3 seconds, and steam to oil mass ratio from .4 to 1.6. A modified residence time was developed for a nonisothermal T-2861 9
TABLE 1 MAXIMUM LIGHT OLEFIN YIELDS (WT % OF FEED) OF TOSCO II CRUDE SHALE OIL (30)
TOSCO II CRUDE SHALE OIL Ethylene 21.2 Propylene 10.8 Total Light Olefins 39.2
TABLE 2 MAXIMUM LIGHT OLEFIN YIELDS (WT % OF FEED) OF WHOLE SHALE OIL (31 )
TOSCO II WHOLE SIM. IN-SITU WHOLE SHALE OIL SHALE OIL Ethylene 19.86 22.11 Propylene 8.16 11.37 Total Light Olefins 27.98 38.71
TABLE 3 MAXIMUM LIGHT OLEFIN YIELDS (WT % OF FEED) FROM DISTILLATES (32)
TOSCO II SIM. IN-SITU VACUUM DISTILLATE VACUUM DISTILLATE Ethylene 28.8 22.62 Propylene 11.1 12.3 Total Light Olefins 4.0.5 17.2 T-2854 10
reactor temperature profile. Maximum yields of ethylene, propylene and light hydrocarbons are given in Table 2.
Smith (32) pyrolyzed the vacuum distillates of shale oil obtained from the NTU simulated in-situ and TOSCO II retorting processes. Reaction variables and ranges were: temperature 1397°F to 1671°F, residence time 0.388 to 1.126 seconds, and steam to oil mass ratio 0.261 to 1.124.
Increased severity resulted in increased yield of solid products, volumetric gas yields, and amounts of ethylene, methane and hydrogen. The results of the distillates were greatly improved over Tosco II and NTU whole shale oils.
Results of maximum yields of ethylene, propylene and light hydrocarbons are given in Table 3.
Kavianian (33) studied the liquids produced from steam pyrolysis of shale oil. The formation of aromatics and distribution of nitrogen and sulfur in liquid products from whole and vacuum distillate oils were studied. The results indicate that the carbon to hydrogen ratio of the products were much higher then that of the feedstock. The concentra tion of nitrogen and sulfur were higher in the liquid products then the feed shale oil.
In a related study, Angelos (34) pyrolyzed several different shale oils at various levels of hydrogenation at optimum ethylene production conditions. The liquid products were analyzed for benzene, toluene, and xylene (BTX). The severity of hydrogenation effected the maximum BTX T-2854 11
formation. The greatest amount of BTX was from shale oil
containing between 0.3 and 0.4 weight percent nitrogen.
The effect of hydrogenation severity on pyrolysis
product yields was a source of other studies. Vacuum
distillates of Tosco II and simulated in-situ shale oils were hydrogenated at three diferent levels: mild, moderate, and severe. These studies were performed by A.A.Ballut,
H.R. Kavianian, and C.F. Griswold. Ballut (35) investigated
the mildly hydrogenated oils. The results showed that the
olefin yields were higher then those produced from the crude
or vacuum distillate oils. Kavianian (36) pyrolyzed moderately hydrogenated oils. These results indicated that
these oils could make an excellent feedstock for pyrolysis.
A simulation model was also developed for steam pyrolysis of
liquid hydrocarbons. The validity of this model with respect
to eastern shale oils is examined in this study. Griswold
(37) pyrolyzed severely hydrogenated oils. This study
indicated that these oils should make an excellent feed. An investigation into gasification was also conducted
by studying the role of steam reforming in pyrolysis reactors. Severely hydrogenated oils have a tendency to undergo steam reforming reactions during high temperature conditions. The results of these experiments are given in
Table 4. T-2854 12
A high temperature (1602 F to 1775 F) reactor with low residence time capability was used by El-Kabule (38). Paraho and Tosco II oil at various levels of hydrogenation severity were used in this study. Steam to oil mass ratio remained at about 1.0 for all of the pyrolysis runs. Results of maximum ethylene, propylene and light hydrocarbon yields are given in Table 5. The severely hydrogenated oils underwent steam reforming reactions at high temperatures. Increased severity resulted in decrease of ethylene, propylene,
1,3-butadiene, and olefin yields.
Dupont
Dupont initiated programs to evaluate new sources of feedstocks for petrochemicals which will be available in this century and early in the next. Several experiments on shale oil have been preformed by Glidden and King (27,28).
Pyrolysis was done on various cuts of lightly hydrotreated
Paraho shale oil. Results of some of these experiments are given in Table 6 . The amount of olefins produced by the heavy gas oil fraction is increased dramatically by hydro- teating. The ethylene yield is increased by hydrotreating of the light gas oil fraction but the amount of propylene and heavy products is decreased (27,28). In another study
(46,47) the potential of Paraho shale oil as a petrochemical feedstock was investigated. The processing scheme used in T-2854 13
TABLE 4 MAXIMUM LIGHT OLEFIN YIELDS (WT % OF FEED) OF HYDROGENATED VACUUM DISTILLATES (35,36,37)
TOSCO II SIMULATED IN-SITU HYDROGENATED SEVERITY HYDROGENATED SEVERITY LOW MED. HIGH LOW MED. HIGH Ethylene 28.1 28.7 24.5 29.5 34.0 35.7 Propylene 12.9 12.7 14.7 31 .1 14.7 15.4 Total Olefin 44.9 46.3 45.7 45.2 52.5 55.4
TABLE 5 MAXIMUM LIGHT OLEFIN YIELDS (WT % OF FEED) OF HYDROGENATED SHALE OILS (38) (HIGH TEMPERATURE PYROLYSIS)
PARAHO TOSCO II HYDROGENATED SEVERITY HYDROGENATED SEVERITY LOW MED. HIGH LOW MED. HIGH Ethylene 26.7 28.6 14.6 23.0 30.1 22.5 Propylene 9.65 13.4 4.45 9.6 12.6 8.1 Total Olefin 42.2 51 .7 21 .9 38.8 52.2 35.7 T-2854- U
TABLE 6 STEAM PYROLYSIS YIELDS (WT % OF FEED) OF PARAHO SHALE OILS (27,28)
VIRGIN HYDROGENATED LGO HGO NAPHTHA LGO HGO
TEMP. F U 11 1371 14-76 1111 1371
REACTION TIME 0.32 0.32 0.36 0.34- 0.32 (SEC)
RATIO 1.0 1.0 0.75 1.0 1 .0
YIELDS (WT%)
Ethylene 15.6 10.6 21.1 16.1 14.7
Propylene 11.6 7.5 11.2 11 .3 10.6
Butenes 4.7 2.0 3.6 4.1 3.0
Butadiene 3*3 1.5 3.9 2.9 2.3
Ethane 3 .6 3.7 3.8 3.2 4.0
Propane 0.6 0.5 0.5 0.1 0.5
Butane 0.1 0.1 0.1 0.1 0.1
Conversion to C,minus 4-9.8 36.1 57.8 16.9 16.0 (Wt J6) T-2854 15
this study involved distillation, hydrogenation, reforming, and pyrolysis of various oil fractions (whole, naphtha, light gas oil, and medium gas oil). Results of these experiments are given in Table 7.
Stone and Webster
A study into the pyrolysis of alternative feedstocks was made by Korosi, et al.(29) for The Stone and Webster
Engineering Corporation. The first stage of the investi gation compared results obtained from pyrolysis of the vacuum gas oil fraction of Paraho shale oil with results obtained from an Arabian vacuum gas oil. The effects of hydrogenation of the oils was also investigated. Two samples of Occidental shale oil were obtained, one was hydrogenated, the other one not. The two oils were pyrolyzed and the results compared with the results of a mideast gas oil. Hydrogenated Paraho kerosene gave the maximum ethylene yields, 23.6 weight percent. Results of this study are given in Table 8 . T-2854 16
ii n O O oo
-P £
r — CO 00
s co < a w o Eh M CO EH • l—l o nO cv cv -si- ^— c-- Q PP •• • • • • a CO cv cv cv T— x— -si- o o O
pq i-P PQ o o O n 00 ITN ICS cn *=i o o n O o X— n O O n EH C- 00 CO CO 00 C" a |x| to -Nf P i Eh EH •» EH P i ^ N a K Q EH a PP Q < i o O p p a /—V o Eh pq 1-- 1 •'-- N x o pp M Eh < ! S o c a :-■•■■■ *sj m PP hP < CO < pq o T— o oo ON 00 CN2 O 0 0 I • «• 0 • s PQ > -s t ^— o • s t -< t ON t ) I—I x— cv •*— c r\ w S I-q o CO P i 525 ^—. Eh «aj o pq M O o o T— o CV cv P h PQ O CV 1 ••••• <«j > -s t T_ sO c \2 -s t 0 0 P i r— -s t I - q < M O PQ pq pq CO Eh O >H o pq M Ph Q P i Q PC CO pq Eh M pq M pq W O O CO pH P h U a H o *=i t-q P=h PQ I s o s £>- o T— o » r\ o O O S *A C\2 1 • * ••• p i < P i i-P -s t T— T“ C f\ u~\ feS. i-q r — T“ -S t Oh t-P f=> Eh M Ph S t s O -— ■ pq W O e-t t-Q I s O t > OQ O NO O CO «*} < • CV •••• • t c P i o ■st O V— -S t O -s t CO -=£j -r~ -s t > O t n O «aj i s o T - CV 0 ON (T\ NO t n p i < *i • CXD • •••• <*! P i S -s t O o U~\ o o -\ £ > EH ^— P h -r_ ■ r_ -s t o < Ph p c O < PQ P i pq < Eh pq P-. < S S M pq co n O r — NO -s t O O l 0 0 ••• o P i ITN o Reaction Variables Feed Composition- Pyrolysis yields are greatly dependent on the composition of the feed. The ease of conversion of various molecular species are summarized by Zdonik (14) and Vanderkooi (39). Normal paraffins are easiest to convert to ethylene, followed by isoparaffins, naphthenes, olefins and aromatics. Paraffins will react to form light olefins as the major products. These light olefins can inturn polymerize and condense to form benzene, toluene, or form heavier olefins (19). Naphthenes can dehydrogenate to form aromatics (20). Generally yields of / olefinic and aromatic products decrease as the average molecular weight of the feedstock increases. Properties of the oil used in this study is given in a later section. Reaction Temperature-The total volume of gaseous products increase rapidly with increasing reaction temperature (24). Pyrolysis at low pressure, high temperatures and short residence time gives maximum ethylene yield (16). Product stability is effected by high temperatures. As the temperature increases the yields of hydrogen, methane, ethylene and aromatics will increase, propanes, butanes and pentanes will first increase then decrease ; h-exanes and nonaromatics will decrease (39). Yields of propylene and butene will decrease with increasing temperatures while the fraction of butadiene ARTHUR LAKES ~ COLORADO SCHOOL of MINES GOLDEN, COLORADO 80401 T-2854 19 within the butene yields will increase (12). The Institute of Gas Technology (24) conducted pyrolysis experiments between 1400° and 1500°F. Dupont Chemical (27,28) operated between 1382° and 1562°F. The maximum reactor temperatures studied at Colorado School of Mines are 1300° to 1630 F(30,31), 1400° to 1670°F (42), and 1600° to 1775 ^ (38). In this study the highest tempera tures in the reactor will range from 1395° to 1632°F. Residence Time-Low residence times are essential for maximum production of light olefins (16). As pyrolysis begins in the reactor, decomposition reactions take place, forming valuable olefins. As the reaction time increases the concentration of reactive olefins increase and polymer ization reactions (coking) begins to occur (40). These polymerization reactions can be reduced by decreasing the time that the reactive intermediates are in the pyrolysis furnace (reduce the residence time of the reactants) . Several investigations have been involved with looking at the effects of residence time on product composition. Residence time ranges of some of these experiments are 1.0 to 4.5 seconds (29), 0.5 to 1.5 seconds (39), 0.27 to 1.34 seconds (31), 0.04-0.1 seconds (35), and 0.01 to 0.1 (38). The high temperatures that are required to reach the lower residence times will break down metals used in pyrolysis T-2854 20 units. In this study residence times range from 0.445 to 1.423 seconds. Steam to Oil Mass Ratio-Zdonik and Green (41) report that the functions of steam in pyrolysis of hydrocarbons are: lowering of the hydrocarbon partial pressure resulting in better selectivity for olefins, steam has an oxidizing effect on surface metal of the reactor diminishing catalytic effects of iron and nickel which whould promote coke forming reactions. The steam is also used as an oil preheat. Chambers and Potters (16,42) report that decreasing the steam to oil mass ratio decreases the olefin yield and increases the fouling rate inside the reactor. As steam .is expensive , the steam to oil mass ratio should be kept to a minimum consistent with good furnace operation. Previous studies have been made with steam to oil mass ratio of 0.5 to 1.0 (42) and 0.2 to 0.8 (39) for petroleum feedstocks. Ranges for shale oil are 0.4 to 1.2 (14) and 0.4 to 1.6 (31). It has been reported (37) that steam to pil mass ratio has less of an effect on product yield then either residence time of reactor temperature. The steams to oil mass ratio in the study range from 0.449 to 0.931. T-2854 21 Pyrolysis Severity Factor The Institute of Gas Technology (24) showed that temperature (°F), T, and residence time (seconds), can be combined into a single variable called the Severity Factor,S .06 S=T0 This quantity was developed for correlating pyrolysis results . Griswold (30) modified this severity factor to include x the steam to oil mass ratio,Ratio;S * .06 .05 S =T9- Ratio xx Fritzler (31) introduced another severity factor,S which took into comsideration the nonisothermal temperature profile of the reactor. A residence time is calculated using the formula proposed by Davis and Farrell (43),O; _.06 .05 S =T-Q- Ratio In this study a comparison is made between the severity factor proposed by Griswold (30) and the modified severity factor of Fritzler’s (31). T-2854 22 FEEDSTOCK CHARACTERISTICS Until recently, interest in oil shale development in the United States has concentrated on the Eocene Shale found in Colorado, Utah and Wyoming. However, Devonian Oil Shale found in the eastern United States has become an interesting prospect due to the Hytort (hydroretort) process developed by the Institute of Gas Technology (44). This retorting process uses controlled heating of the oil shale in a hydrogen atmosphere under elevated pressures (500 psig)(45). The hydrogenation process shifts the composition of the oil shale towards a higher paraffinic concentration by a combination of ring saturation and bond breaking. In a conventional retort organic carbon conversion for eastern oil shale is only about 55 percent. The Hytort process boosts the conversion to as high as 90 percent (45). The oil used in this study was obtained from the Institute of Gas Technology. It is an eastern shale oil that was retorted by the Hytort process (sample number 82-ppc-5). Properties of this oil are given in Table 9. T-2854- 23 TABLE 9 PHYSICAL PROPERTIES OFi1' EASTERN SHALE OIL (HYTORT) SAMPLE § 82-PPC-5 ELEMENTAL ANALYSIS (wt. *) C-86.38 I:1?lot C/H-8.11 N- 1.95 ASH- 0.00 SPECIFIC GRAVITY (60°F): .9264. VISCOSITY (1OQ°F); 5.2 cSt loTJR POINT: -4-5°F AVERAGE MOLECULAR WEIGHT: 200-210 grams/mole WEIGHT % AROMATICS: 55-60% DISTILLATION FRACTIONS Degree F Volume ! 14-7 5 236 10 283 20 365 30 4-4-5 10 513 50 570 60 619 70 663 80 700 90 736 95 760 END 96% Recovery 3% Residual T-2854 24 EXPERIMENTAL APPARATUS The experimental apparatus used in this study is essentially the same as the system used in previous pyrolysis studies at The Colorado School of Mines. The system, pictured in Figure 1, can be divided into three sections: 1. Feed Section 2. Reactor Section 3. Cooling and Sampling Section Feed Section Deionized water for steam generation is fed from a 250 ml buret by a Lapp pump to a 19.5 inch section of 1 inch O.D. schedule-40, 316 stainless steel pipe which is packed with 3/8 inch Coors ceramic balls. The balls are held in place by two pieces of wire mesh. The pipe is positioned inside a single zone Lindberg furnace with a controller that has an operating range of 200°to 1200°C. The steam flows from the furnace through an 18 inch section of 1/4 inch O.D. 316 stainless steel tubing that is wrapped with Briskheat tape and insulated to decrease heat loss from the steam. The Briskheat tape temperature is adjusted by a 7 1/2 amp. rheostat. T-2854 25 snasNaaNoo nasNaaNOO aasNaaNOO i W O CO aovNHna o tHK s m o z aaHHi eu 3A1VA ix a a v s aovNHna 3n o z aiowis O < E-i W 3 «t Oh w < X .Eh Y 110 31VHS I! 1 I 1 • 1 ! I H3IVM . 11 , 11— iaaxvw HONano T-2854 26 Shale oil is pumped from a 250 ml buret by a Lapp pump through 1/4 inch O.D. 316 stainless steel tubing. This tubing is wrapped with Briskheat tape to preheat the oil. The heating tape temperature is adjusted by a 7 1/2 amp. rheostat. The shale oil is then pumped through 1/4 inch O.D. 316 stainless steel tubing that is wrapped with insulation. Shale oil and steam are mixed at a "tee” junction directly above the reactor inlet. The temperatures of the oil and steam are monitored by type K 1/8 inch O.D. 316 stainless steel sheathed ground thermocouples wired to an Omega 1695-automatic scanner and an Omega 200 digital temperature indicater. Reactor Section The reactor is a 14 inch length of 2 1/2 inch O.D., 2 in I.D. 316 stainless steel tubing. The tubing is threaded 1 inch on each end to accomodate machined 316 stainless steel plugs (Figure 2). The top plug has two drilled and tapped holes. The center hole has a 5/16 inch diameter hole that is tapped with 1/4 inch NPT thread to accomodate 1/4 inch pipe to 1/4 inch Swagelok thermocouple fittings. The feed hole is 3/8 inches in diameter and tapped with 1/4 inch NPT threads. Details of the construction can be seen in Figure 2. T-2854- Tapped i" NPT 3/8" Hole CM FIGURE 2 REACTOR 27 T-2854 28 The bottom plug has one concentric hole that has a 3/4 inch diameter tapped with 3/4 inch NPT threads to accomodate a 3/4 pipe with 1 inch Swagelok fittings. The reactor and plugs were machined to specification given in Figure 2. The reactor is packed with 705 3/8 inch Coors ceramic balls (Alumina grinding media in Figure 3). The ceramic balls are used to reduce the volume of the reactor, cause turbulence in the flow, and increase the heat transfer to the reactant gases. The ceramic balls are held in place by two pieces of nichrome wire screen. A three-zone Lindberg furnace is used to heat the reactor. The temperature of the furnace is controlled by a Lindberg controlling unit that can operate between 200° to 1200°C. The control unit responds to the platinum II thermocouples which are located in the middle of each of the three zones within the furnace. The temperature inside the reactor is monitored by 5 1/16 inch k type 316 stainless steel sheathed thermocouples. Thermocouple locations within the reactor are pictured in Figure 4. Thermocouples are wired to an Omega 1695-auto matic scanner and an Omega 200 digital temperature indicater. A quench tube is attached to the bottom of the reactor. The hot process gases flow down through this tube and are cooled by a quench deionized water stream to a temperature between 500 °F to 700 OF. The quench water is pumped by a T-2854 29 Feed Center Therocouple Well Top Plu Nicromium Wire Alumina Grinding Media Coors Ceramic Balls Nicromium Wire Bottom Plu; Quench Water Inlet Quench Thermocouple FIGURE 3 ASSEMBLED REACTOR T-2854- 30 -#1 :0.0" .#2:1 .0 " -#3:3 1/4-” ■U-5 3/4-” .#5:8.0" #6 :12.0" -______#10:25 1/2" FIGURE ^ THERMOCOUPLE LOCATION T-2854 31 quench tube 3 1/2 inches from the bottom of the reactor (Figure 1). Cooling and Sampling Section After quenching the reaction product mixture flows into a 2000 ml. surge flask where a liquid-gas separation occurs. The gas passes through a 3/8 inch O.D. 316 stainless steel tube where a pressure gauge is located. The gas is then cooled by an 18 inch single pipe condenser. The gas then passes through 1/4 inch O.D. 316 staainless steel tubing to a 1000 ml. flask that is cooled in an ice-brine water bath. After condensation of liquids the gas flows through a two way Whitey valve (VI). This valve is used to vent product gas to the atmosphere if an unexceptable pressure increase should occur within the system, or send gas through the remaining part of the system. In the latter case the gas flows through a 16 1/2 inch long, 1/2 inch O.D. single pipe condenser. Product gas is then cooled in a 500 ml. flask which is submerged in a refrigeration bath. The bath is filled with a 50-50 mixture of ethylene glycol and water. This mixture is also used to cool condensers. The gas then goes through two 24 inch, 1/4 inch O.D. single pipe condensers arranged in a vertical array, and then through another submerged 500 ml. flask. Gas then passes through a 1/4 inch O.D. 316 stainless steel tube into a demisting flask which is packed T-2854 32 with Pyrex glass wool to trap entrained liquids. The gas is directed towards two valves (V2 & V3) arranged in series. These valves are used to send gas to the Rockwell Dry Gas Meter or to sampling bombs. Two bombs are taken during a run, the first at twenty minutes and is taken in a 300 ml. bomb. The second at thirty minutes and is in a 1000 ml. bomb. After metering, the gas is bubbled through a diethanolamine solution to remove H S before venting to the atmosphere. Gas Meter Calibration The gas meter is calibrated before each run by the procedure discribed by El-Kabule (38). Three calibration attempts are made and the average correction factor is determined. T-1854 33 Safety Features -The pressure in the reactor, condensing and steam feed systems are monitered by two gauges, one is placed on the steam feed line, the other is on the ’’tee" directly after the reactor system. A 25 psig safety valve is connected to the first gauge line and is vented to the atmosphere. -The pyrolysis unit is enclosed in a safety cage to prevent operators from explosions. The cage is fitted with a fan that is vented to the outside to reduce accumulation of product gas within the cage in the event of a leak in the system. -An M-S-A series 1-501 combustable gas detector is used to analyze the laboratory atmosphere for flamable gasses. If harmful gases are detected an alarm is activated. -A fire extinguisher is mounted next to the operator’s console. -Gas masks are worn during each pyrolysis run. -Emergency shutdown procedures are posted on the operator’s console. T-2854 34 EXPERIMENTAL PROCEDURE The involved experimental procedure which follows requires the efforts of at least two people. Prior to and during each run the following procedure was followed. 1. Calibrate the Rockwell Dry Gas Meter to obtain the correction factor to be used for the run. The weighed-water method is used as described by El-Kabule (38). 2 .Weigh out all components of reactor and condensing sections that had been previously cleaned with acetone and air. Weigh out one 2 liter flask, one 1 liter flask and two 500 ml. flasks. 3.Count out 705 ceramic balls and wash them with acetone Weigh the dry balls and place them into the reactor. Assemble the system as shown in Figure 3. 4.Set controller at desired reaction temperature on the three zone Lindberg furnace. In order to obtain an acceptable profile within the reactor,the top zone must be set about 75 degrees higher then the middle zone. The middle zone should be set at the desired reaction temper ature. Turn on the reactor furnace and let it warm up for four hours or until a steady temperature profile is obtained within the reactor. 5. Run program ESTIM.SIT (App A) to get estimation for oil and water flow rates for a desired residence time, temperature and steam to oil mass ratio. T-2854 35 6 . While the reactor is heating up set the oil and water pumps to desired flow rate. Fill burets with oil and water. 7.Turn on gas to gas chromatograph and turn on G.C. using method described by El-Kabule (21), which is posted above the G.C.. 8 . At 1/2 hour before reactor reaches temperature turn on the single-zone furnace, oil preheat and steam superheat. 9. When reactor reaches temperature, turn on the condenser pump, and let run for ten minutes. Record temperature profile, buret readings and attach tared flasks to the system. 10. Run steam through the system for fifteen minutes to get desired flow rate on water pump. 11. Record gas meter reading, turn VI to condenser position, open V2, turn V3 to meter position,simultaneously turn on oil and quench pumps (steam pump is still running from step 10, take reading on steam buret when turning on other pumps), start stop watch. 12. Temperature profiles are taken every ten minutes, while gas meter and quench readings are taken every five minutes. Steam water and oil buret readings are taken every minute to insure desired flow rate. 13. Gas samples are taken at twenty and thirty minutes from the start of the run. Both sample bombs are filled with helium before a sample of product gas is taken. The twenty minute sample is taken in a 300 ml bomb and the 30 T-2554 36 minute sample is taken in a 1000 ml. bomb. 14. After 30 minute bomb is taken, turn off the quench pump, oil pump and record level on buret. Leave steam water pump on for 15 minutes. 15. Turn off three zone furnace, steam furnace, preheats and condensing pumps. Open VI to vent position. 16. Read gas meter then flush it with air for 12 hours. 17. Run gas chromatograph on bomb samples by using the instructions provided by El-Kabule (21). The small bomb is used to establish attenuation factors for the 30 minute bomb . 18. After reactor system is cooled, weigh out the reactor and all of the condensers and flasks. 19. Run program E0SMP.F0R (appendix A) to get component gas yields from the run. T-2854 37 RESULTS AND DISCUSSION All pyrolysis runs were made with an eastern shale oil that had been produced by means of the Hytort retorting process. The oil was obtained from The Insitute of Gas Technology, which also provided the physical properties of the oil shown in Table 9. Nine pyrolysis runs were made, eight of these runs were used to obtain conditions for maximum ethylene productions. Conditions were changed by changing first the temperature then residence time and steam to oil mass ratio. The final run was used to establish the reproducibility of a given pyrolysis run. A summary of reaction conditions, weight percent of feed conversion to gaseous products , liquid, gas and solid product yields are given in Table 13 and Table 15. Print outs of EOSMP.FOR are given in Appendix B Effect of Reaction Temperature The first series of pyrolysis runs (4,5,6 ,7 and 9) were used to evaluate the effect of temperature on product gas yields. Maximum temperature inside the reactor was changed on each run while the residence time remained between 0.716 to 0.932 seconds (modified reaction times between 0.479 and 0.656 seconds) and a steam to oil mass ratio from 0.783 to 0.91. Maximum reactor temperatures used in this part of the T-2854 38 study were: 1395,1529,1632,1573 and 1527 degrees Fahrenheit. Results of these experiments are given in Table 10 and graphed in Figure 5 and Figure 6 . These results indicate that the weight percent of feed that is converted to gaseous products is maximized at 1573 F. Maximum ethylene yield is obtained at 1529°F, while maximum propylene is obtained at 1395°F. Yields of hydrogen, carbon monoxide and carbon dioxide increase with increasing temperature, while butene yields decrease. Effect of Residence Time A series of experimental runs (5,8 and 10) were made to investigate the effect of residence time on gaseous product yield. The runs were made at approximately 1520°F, this is the temperature that gave maximum ethylene yield in the previous series of experiments. The residence times looked at were: 0.445, 0.790 and 1.423 seconds, this corresponds to a modified residence time of 0.262, 0.530 and 0.970 seconds. The modified residence time proposed by Fritzler (31) was used to compensate for nonisothermal reactor conditions. Results of this series of experiments are given in Table 11 and graphed in Figure 7. Total gas production dropped off at longer residence times while the production of hydrogen, carbon monoxide and carbon dioxide increased. Ethylene yield remained constant T-2854 39 TABLE 10) MEDIUM RESIDENCE TIME RESULTS RUN § 4 5 6 7 9 12 TEMP.°F 1395 1529 1632 1573 1518 1527 RETIME°N 0,855 0,790 °-716 0 •932 0.768 0.789 MODIFIED REACTION 0.599 0.530 0.4-80 0.656 0.471 0.501 TIME STEAM TO OIL MASS 0.783 0.803 0.909 0.826 0.753 0.789 RATIO TOTAL GAS YIELD 28.3 31.6 28.0 33.6 31.1 31 .4 (WT^FEED) TOTAL GAS YIELD 5.18 6.86 7.88 8.50 6.55 6.95 (SCF/#FEED) . MAJOR PRODUCT YIELD (WTfoFEED) (SCF/#FEED) 9.61 12.3 8.81 11.7 11 .6 12.1 °2H4 1 .30 1 .66 1.19 1 .58 1 .57 1 .63 3.18 0.83 2.10 3.97 3.17 °3H6 5.65 0.51 0.29 0.08 0.19 0.36 0.29 TOTAL 1 .45 0.37 0.07 0.23 0.53 0.35 0.10 0.02 0.00. 0.01 0.04 0.02 °4H8 ,s 1 .07 0.91 0.63 1 .26 0.99 °4H6 0.89 0.06 0.08 0.07 0.04 0.09 0.07 TOTAL 17.6 16.9 9.97 14.9 17.3 16.6 OLEFIN 1 .97 2.05 1 .29 1 .72 2.06 2.01 T-2854- Q_ Total Olefin Q_ I 3501450 1 450 11550 550 1650 TEMPERATURE ( F) FIGURE 5 EFFECT OF TEMPERATURE ON GAS YIELDS T-2854 SCF/tt OF FEED Q 1 IUE EFC O TMEAUE N A YIELDS GAS ON TEMPERATURE OF EFFECT 6 FIGURE 350 TEMPERATURE ( TEMPERATURE F) 1 450 1550 oa Olefin Total oa Gas Total 1650 41 T-2854 42 TABLE 11 EFFECT OF RESIDENCE TIME ON PRODUCT YIELD RUN § 8 10 TEMP.°F 1529 1 520 1535 REACTION 0.790 0.445 1 .42 TIME MODIFIED REACTION 0.530 0.262 0.970 TIME STEAM TO OIL MASS 0.803 0.931 0.810 RATIO TOTAL GAS YIELD 31 .6 32.0 26.7 (WTJFEED) TOTAL GAS YIELD 6.86 6.30 6.70 (SCF/#FEED) MAJOR PRODUCT YIELD (WTJGFEED) (SCF /#FEED) 12.3 12.3 9.21 C2H 4 1 .66 1.66 1 .25 3.18 4.56 0.28 C3H6 0.29 0.41 0.03 TOTAL o’: 37 0.67 0.23 Ws 0.02 0.04 0.02 TOTAL 16.9 19.1 10.3 OLEFIN 2.05 2.22 1 .34 ARTHUR LAKES LIBRARY COLORADO SCHOOL of MINES GOLDEN, COLORADO 80401 4-3 T-2854- Total Gas Total Olefin Q_ d_ 0 . 00 0.^0 0.90 1 .20 1 .60 RESIDENCE TIME (SECONDS) FIGURE 7 EFFECT OF RESIDENCE TIME ON GAS YIELDS T-2854 44 at 12.3 weight percent of feed (1.66 SCF/pound of feed) at lower residence times but was reduced to 9.21 weight percent of feed (1.25 SCF/ pound of feed) at the longest residence time. Propylene yield initially increased from 3.18 to 4.56 weight percent of feed with increasing residence time (0.445 to 0.790 seconds). At longer residence times, 1.42 seconds, the propylene yield droped off to 0.28 weight percent of feed. Maximum butenes and total olefin yeilds were obtained at the intermediate residence time (Table 11). Effect of Steam to Oil Mass Ratio A pyrolysis run was made to determine the effect of reducing the steam to oil mass ratio on the product gas yield. It was determined from the previous series of experiments the maximum light olefin yields were obtain ed in run number eight (Temperature: 1520°F, Residence Time: 0.445 seconds, Modified Residence Time:0.262 seconds and a Steam to Oil Mass Ratio:0.931). In this run, number 11, the maximum temperature was 1478°F. Although attempts were made to increase this temperature, we were unable to reach the desired temperature of 1520°F.. This could be due to the smaller amount of steam available to preheat the oil. The increased amount of oil in the feed mixture would also make the endothermic effects of the pyrolysis reactions more pronounced. Residence time was 0.530 seconds (modified T-2854 TABLE 12 EFFECT OF STEAM TO OIL MASS RATIO ON PRODUCT YIELD RUN § 8 11 TEMP.°F 1520 1478 REACTION 0*445 0.530 TIME(sec) MODIFIED REACTION 0.262 0.415 TIME(sec) STEAM/OIL 0.931 0.449 MASS RATIO GRAMS OF COKE r 2-6 .9.8 TOTAL GAS YIELD 31.97 33.97 (WTfoFEED) TOTAL GAS YIELD 6.30 6 .46 (SCF/#FEED) MAJOR PRODUCT YIELD (WT$FEED) (SCF/#FEED) ETHYLENE 12.29 12.54 1 .66 1 .70 PROPYLENE 4.56 5.14 0.41 0.46 TOTAL BUTENES 0.67 1 .06 0.04 0.08 TOTAL 19.1 20.7 OLEFIN 2.22 2.37 T-2854 46 residence time of 0.415 seconds) and the steam to oil mass ratio was 0.449. The decrease in steam to oil mass ratio showed a slight increase in ethylene, total light olefin and total gas yields. Results of this experiment are given in Table 12. The reactor and packing material showed a significant increase in coke formation and the quench tube was almost plugged with coke and liquid products. Effect of Severity Factor The severity factor is a variable that combines reaction parameters (temperature, residence time and steam to oil mass ratio) into a single quantity. Two severity factors are used in this study, the first one was proposed by Griswold (30) and is refered to as the "OLD SEVERITY FACTOR" in Figure 8 . The second severity factor was pro posed by Fritzler (31) and uses a modified residence time to compensate for nonisothermal reactor conditions. Fritzler’s severity factor is graphed in Figure 9. Fitzler's severity factors were substantially reduced when compared with factors evaluated by Griswold’s method, as is shown in Table 13, they also appear to form a smoother curve when ploted against gas yield, Figure 8 and Figure 9. Ethylene yield varies from 9.21 to 12.5 weight percent of feed. Maximum ethylene yield is observed at a Fritzler’s T-2854- FIGURE UT % OF FEED 1 .00 10.00 20.00 3 0 .0 0 0 0 3 8 EFFECT OF GRISWOLD'S SEVERITY FACTOR ONYIELDS SEVERITYFACTOR OFGRISWOLD'S EFFECT L -SEVERITYOLD FACTOR 1 400 oa GasTotal oa OlefinTotal 1600 T-2854- 48 X' 1— !— !— !— [— !— !— !— I— r T T 1 440 00 0 1 5 0 01 0 SEVERITY FACTOR FIGURE 9 EFFECT OF FRITZLER'S SEVERITY FACTOR ON YIELDS T-2854- 49 TABLE 13 RESULTS OF STEAM PYROLYSIS OF EASTERN SHALE OIL Run § J±__ __5__ _6 _ _7_ _8_ _9 _ _LO 12 CONDITIONS Reactor Temp 1395 1529 1632 1573 1520 1518 1535 U 7 8 1527 Res. Time (s) 0.86 0.79 0.72 0.93 0.45 0.77 1.42 0.53 0.79 Mod. Res. 0.60 0.53 0.48 0.66 0.26 0.47 0.97 0.4-2 0.50 Time (s) Steam/ Oil Mass 0.78 0.83 0.91 0.83 0.93 0.75 0.81 0.45 0.79 Ratio Severity Factor 1365 1491 1592 1551 1442 1473 1551 1367 1488 (Griswold) Severity Factor 1336 1455 1554 1519 1397 1430 1516 1347 1448 (Fritzler) MAJOR PRODUCT YIELDS (wt. % of feed) 6 .A. . 5 7 8 9 10 11 12 Ethylene 9.61 12.3 8..81 11 .7 12.3 11 . 6 9. 21 12.5 12.1 Propylene 5.70 3.18 0,► 83 2. 10 4.56 3. 97 0.28 5.14 3. 17 Total 1 .45 0.37 0..07 0. 23 0.67 0. 53 0. 23 1 .06 0. 35 cJ s Total 17.6 10 Olefin 16.9 9.■ 97 14 .9 19.1 17 .3 .3 20.7 16 . 6 T-2854 50 severity of 1347. Optimum Ethylene and Olefin Yields Maximum olefin and ethylene yields were obtained in run number 11 (Temperature:1478°F,Reaction Time: .530 seconds, Modified Reaction Time: .415 seconds, and a Steam to Oil Mass Ratio: .449). A large amount of coke and fouling was found in the reactor and quench systems. Ethylene yield was 12.54 weight percent of feed (1.7 SCF/pound of feed), propylene was 5.14 weight percent of feed(.46 SCF/pound of feed) and total olefin yield was 20.66 weight percent of feed (2.37 SCF/ pound of oil). Run number 8 also showed high light olefin yields with much less coking then run number 11. Conditions for run number 8 were; temperature: 1520°F, residence time: .445 seconds, modified residence time: .262 seconds, steam to oil mass ratio:.930. Ethylene yield was 12.29 weight percent of feed (1.66 SCF/pound of feed), propylene yield was 4.56 weight percent of feed (.41 SCF/pound of feed) and total olefin yield was 19.12 weight percent of feed (2.22 SCF/pound of feed) Results of runs 11 and 8 are given in Table 14. Although conditions number 11 showed maximum olefin and ethylene yield, conditions of run number 8 may be more suitable for industrial porposes as much less coke is formed T-2854- 51 TABLE U MAXIMUM OLEFIN YIELDS (WT. % OF FEED) RUN # 1_1 8 REACTION CONDITIONS TEMP. °F U 7 8 1520 REACTION n ™ 0.4-45 TIME (SEC) MODIFIED REACTION 0.4-15 0.262 TIME (SEC) STEAM/OIL 0.931 MASS RATIO 0.449 SEVERITY FACTOR 1367 1442 (GRISWOLD) SEVERITY FACTOR 134-7 1397 (FRITZLER) GRAMS OF COKE 9.8 2.6 GRAMS OF COKE 4-7.5 22.8 PER KG OF OIL FEED PRODUCT YIELD (WT. % OF FEED) ETHYLENE 12.54 12.29 PROPYLENE 5.14- 4.56 BUTENES 6.20 6.13 BUTADIENE 1.92 1 .60 TOTAL OLEFINS 20.70 19.10 ETHANE 1.81 1 .30 PROPANE 0.18 0.11 TOTAL GAS 33-95 31 .97 T-2854 52 inside the reactor per kilogram of feed (Table 14). Overall Mass Balance An overall mass balance was preformed on all experimental runs. This was done to help establish the validity of the experimental results. The total weight of material into the system was calculated from the weight of steam water, oil and quench water that was introduced into the system. The weight of material out of the system was determined from the amount of product gas and the weight of liquids and solids found in tared flasks, condensers and reactor. The results of the mass balance is given in Table 15. These results indicate that there was less then four percent error in the overall mass balance in a given run. Elemental Mass Balance An analytical analysis for carbon, hydrogen, nitrogen and sulfur was performed on the organic liquid product of run number 5. Results of this analysis is given in Table 16. Assumptions made to perform the balance were: 1. All solid products found inside reactor were assumed to consist solely of carbon. 2. Composition of all liquid organic products were assumed to be the same, no matter if the substance was T-2854- TABLE 15 OVERALL MASS BALANCE i— i p£ C5 M E o CO O M H E a £3 E <3 O s s CO Pi < E H I— I O E <£j 0H O O T*25eT £3 HH £3 O h h h h h £3 =8= ’ O EC E EH s O £3 O M s o (-3 O £3 CO M ffi CO CO M Q M i-4 3 O i-i CO O O Q NO NO NO O- £> ON in m CV o as ON T— 02 T o co co 02 00 ON -'4- in O- N-st ON NON ON -00 e- - * • • • • • • • ONO NO NO NO NO NO in \— 00 T— c- ON co * V o T NO 00 02 02 -4 nin in o- -st CO -4- -4- -4" — • • • • • • NO in o 02 00 T— co 02 CO CO o T— o 02 O02 CO 00 O' -4 in 00 CO co 02 00 CO ON CO T ON — • • • • • • • NO NO NO NO o- 02 co 02 o o in 02 T— ON ON ON in -4- -4- -Sf in ON CO '4- oo o \— » • • • • • • • • • ONO NO NO n ON in 02 ON o in 00 o o co in 00 e- 2 CO ^2- o ON in V— 02 4 e' -4- E> CO in 02 C- o NON ON f— -4- 02 Q • • • • • • • • • • • • • NO NO NO co CO CO in en CO o T— Y— 02 T— in co -4- -Sf -4- ON in 02 • • • 0 0 2 2 • 53 T-2854 54 found in a flask or a condenser. 3. Aqueous layer in separation flasks did not contain any product. Results of the elemental mass balance is given in Table 16. Carbon balance showed a 2 percent decrease in carbon out to carbon in ratio. The hydrogen balance showed a 9 percent decrease in the same ratio. Sulfur balance showed the largest difference in the balance. This could be caused by sulfur compounds in the aqueous layer of the separation flasks. Reproducibility of Results Run number 12 was made to evaluate the reproducibility of the results obtained in run number 5. The results of these two runs are given in Table 17. Exact reproducibility of water and oil flow rates and reactor temperature profile was difficult to obtain. Even with these difficulties less then 2 percent difference was found in ethylene and total olefin yield. Less then 1 percent difference was observed in total gas yield. T-2854 55 TABLE 16 ELEMENTAL MASS BALANCE MASS IN LIQUID MASS GAS MASS ELEMEMT (grams) CARBON 68.41 45.67 21 .44 HYDROGEN 8.44 2.85 4.87 SULFUR 0.80 0.47 0.05 NITROGEN 1 .54 1 .37 — C +/C. = 0.981 out m H ./H. out m = 0.915 s ./S. = 0.665 out m N +/N. = 0.890 out m ELEMENTAL ANALYSIS WEIGHT % OF WEIGHT % OF ELEMENT OIL FEED LIQUID PRODUCT CARBON 86.38 88.27 HYDROGEN 10.66 5.51 SULFUR 1 .01 0.91 NITROGEN 1.95 2.65 T-2854- 56 TABLE 17 REPRODUCIBLITY OF PYROLYSIS RUN RUN ff 5 1_2 °F 1529 1527 REACTION 0.790 TIME(sec) 0.789 MODIFIED REACTION 0.530 0.501 TIME(sec) STEAM/OIL MASS RATIO 0.803 0. 794- SEVERITY 14-56 14-4-8 FACTOR PRODUCT YIELDS(WT. % OF FEED) ETHYLENE 12.29 12.07 PROPYLENE 3.18 3.17 BUTANES 0.37 0.35 BUTADIENE 1.07 0.99 TOTAL OLEFIN 16.9 16.6 ETHANE 1.07 1 .21 PROPANE 0.06 0.06 HYDROGEN 1.06 1.13 TOTAL GAS 31 .62 31 -4.3 T-2854 57 Comparison with Previous Studies Results of this study were compared with results of previous studies at The Colorado School of Mines. Kavianian (36) proposed a simulation model that was designed to predict pyrolysis results from western shale oil. This program was used to see if it could also predict yields of from an eastern shale oil. The physical properties of the oil and reaction conditions were substituted into the program. Printout of the simulation program is given in Appendix C. Results are given in Table 18. These results show that the simulation program may be able to predict ethylene yield fairly well, but it over estimates yields of other light olefins. Predicted yields of hydrogen and methane were lower then those actually obtained. Ethylene and light olefin yields obtained in this study, Table 12, are much less then those obtained from previous studies at The Colorado School of Mines, Table 1 through Table 5. This may be due to the larger amount of aromatics, 55 to 60 weight percent of feed, found in this eastern shale oil. T-2854 58 TABLE 13 COMPARISON WITH COMPUTER MODEL COMPU' RUN jj 5 m o d e : TEMP.°F 1529 1529 REACTION 0.790 0.790 TIME(sec) STEAM/OIL 0. 804 MASS RATIO 0.803 PRODUCT YIELDS (WT. % OF FEED) HYDROGEN 1 .06 0.29 METHANE 9.87 4.09 ETHANE 1 .07 1.18 ETHYLENE 12.56 12.29 PROPANE 0.30 0.06 PROPYLENE 3.18 8.41 BUTENES 0.37 3.86 BUTADIENE 1 .07 2.03 TOTAL GAS 31 .62 32.72 CONVERSION T-2854 59 CONCLUSION In this series of experiments an eastern shale oil was pyrolyzed to obtain maximum ethylene and light olefin yields. A vertical tubular reactor, packed with ceramic balls was used as a pyrolysis furnace. Maximum temperature inside the reactor ranged from 1395°F to 1632°F, reaction times ranged from 0.46 to 1.42 seconds and steam to oil mass ratio ranged from 0.449 to 0.931. Maximum olefin yield was obtained at a temperature of 1478°F, residence time of 0.530 seconds and a steam to oil mass ratio of 0.449. Yields at these conditions were: ethylene;12.54 weight percent of feed (1.7 SCF/pound of feed),propylene;5.14 weight percent of feed (0.46 SCF/ pound of feed) and total light olefins; 20.7 weight percent of feed (2.37 SCF/ pound of feed). These results are much less then those obtained by previous studies at The Colorado School of Mines, Table 1 through Table 5. Therefore, although shale oil may be a feasible feedstock for pyrolysis units, western shale oil is much more suited for light olefin production then the eastern shale oil used in this study. The low yields of light olefins is consistent with the composition of the oil. The eastern Hytoted oil contains approximately 60 mole percent aromatics and are unreacted in the pyrolysis furnace. Only 39 percent of the oil is made of alkanes, which are easiest to crack during pyrolysis. The western T-2854 shale oils contain about 67 percent alkanes. Although olefins yields in this study were less then previous studies at The Colorado School of Mines, further studies should be made into eastern shal oil. A larger range of residence times and temperatures should be used to see if better yields can be obtained. A BTX cpmposition should be obtained on the liquid feed and all liquid products to investigate the effect of reaction conditions on aromatic content of product liquids. T-2854 61 RECOMMENDATIONS The study into pyrolysis of eastern shale oil should be continued with emphasis on the following areas. 1. Look at a larger range of residence times with each temperature selected. 2. Increase the temperature studied with ceramic reactor. 3. Do an elemental mass balance on all runs. 4. Look at the the BTX composition of the liquid products . T-28554 62 REFERENCES CITED 1 . Wall, H.D., "Shale and Tar Sands are Going," Hydrocarbon Processing, V. 61, No. 6 , 1982, p. 93-96. 2 . Montgomery, D.P., 1968, "Refining of Pyrolytic Shale Oil," I&EC Product Research and Development, V. 7, No. 4, p. 274-282. 3. Lovell, P. F., M. G. Fryback, H. E. Reif, and J. P. Schwedock,"Maximize Shale Oil Gasoline," Hydrocarbon Processing, V. 60, No. 5, p.125-130. 4. Dinneen, G. U., J. S. Bull, H. M. Thorne, "Composition of Crude Shale Oil," I&EC, V. 44, No. 11, 1952, p. 2632-2635. 5. Cook, G. E., "Nitrogen Compounds in Colorado Shale Oils," ACS Div. of Petroleum Chemistry Preprints, V. 10,No. 2, 1965, p. C35-38. 6 . Poulson, R. E., H. B. Jenson, G. L. Cook, "Nitrogen Bases in Shale Oil Light Distillates," ACS Div. of Petroleum Chemistry Preprints, V. 16, No. 1, 1971 p. A49-A55. 7. Cady, W.E., and H. S. Cook, "Composition of Shale Oil," I&EC, V. 44, No. 11, 1952, p. 2636-2641. 8 . Sladek, T. A., "Recent Trends in Oil Shale," Mineral Industries Bull., V. 18, No. 2, 1975, p. 6 . 9. Sullivan, R. F., "Catalytic Hydroprocessing of Shale Oil Distillate Fuels," ACS Div. of Petroleum Chemistry Preprints, V. 22, No. 3, 1977, p. 998. 10. Atwood, M. T., Raw Shale Oil Inspection, The Oil Shale Corporation, Golden, Colorado, 1975. 11. Sullivan, R. F., B. E. Strangeland, H. A. Frumikin, Paper presented at API Refining Department, 43rd Midyear Meeting, May 10, 1978, Toronto, Ontario. 12. Goldstein, R. F., A. L. Waddams, The petroleum chemicals industries: London, E. & F. N. Spon. Ltd., 1967, p. 97. T-2854 63 13. Yesavage, V. F., C. F. Griswold, and P. F. Dickson, "Current Status of the Production of Chemicals from Oil Shale," ACS Div. of Petroleum Chemistry Preprints, V. 25, No. 3, 1980, p. 608-614. 14. Zdonik, S. B., E. J. Bassler, and L. P. Hallee, "How Feedstocks Affect Ethylene," Hydrocarbon Processing V. 53, No. 2, 1974, p. 73-81. 15 . Stork, K, M. A. Abrahams, and A. Rhoe, "More Petrochemi cals from Crude," Hydrocarbon Processing, V. 53. No. 11, 1974, p. 157-166. 16. Chambers, L. E. and W. S. Potter, "design of Ethylene Furnaces: Parti, Maximum Ethylene," Hydrocarbon Processing, V. 53, No.l, 1974, p. 121-126. 17. Kamptner, H. K., "Theory and Technique of the high- Temperature Pyrolysis of Petroleum Hydrocarbons and Their Influence on Some Manufacturing Processes," ACS Div. of Petroleum Chemistry Preprints, V. 13, No. 4, 1968, p. A23-A35. 18. Maples, and P. Adler, "Options for Olefin Feedstocks," Hydrocarbon Processing, V. 57, No. 7, 1978, p. 171-175. 19. Sakai, T., K. Soma, Y. Sasaki, H. Tominaga, and T. Kunugi, "Secondary Reactions of Olefins in Pyrolysis of Petroleum Hydrocarbons," ACS.Div of Petroleum Chemistry Preprints, V. 14, No. 4 1969, p. D40-D53. 20. Hatch, L. F., S. Matar, "From Hydrocarbons to Petrochemicals: Part 8 ; Production of Olefins," Hydrocarbon Processing, V. 47, No. 1, 1978 p. 135-139. 21 . El-Kabule, A. S., personal correspondence. 22. Hurd, C. D,, Pyrolysis in McGraw-Hill Encyclopedia of Science and Technology: New York, McGraw-Hill Book Co., V. 11, p 122-125. 23. Sakai, T., and D. Nohara, "a Kenetic Study of the Formationof Aromatics during Pyrolysis of Petroleum Hydrocarbon," Industrial and Laboratory Pyrolysis, ACS Symposium Series, No. 32, 1976, p. 152. T-2854 64 24. Shultz, E. B., J. J. Guyer, and H. R. Linden, "Pyrolysis of Crude Shale Oil,” I&EC, V. 47, No. 12, 1955, p. 2479-2482. 25. Brantly, F. E., R. J. Cox, H. W. Sohrs, W. I. Barnet, and W. I. Murphy, "High Temperature Shale Oil," I&EC, V. 47, No. 12, 1955, p. 2479-2482. 26. Sohns, H. W., E. D. Jukkola, and W. I. R. Murphy, "Development and Operation of an Entrained Solids, Oil Shale Retort," U. S. Bureau of Mines, RI 5522, 1959. 27. King, C. F., and H. J. Glidden, "Light Olefin from Shale Oil Distillates," ACS Div. of Petroleum Chemistry Preprints, V. 25, No. 3, 1980, p. 155. 28. Glidden, H. J., and C. F. King, "Light Olefins from Shale Oil," CEP, V. 16, No. 12, 1980, p. 47-51. 29. Korosi, A., L. P. Halla, and P. S. Virk, "Steam Cracking of Colorado Shale Oil and Athabasca Bitumen Feedstocks for Petrochemicals," ACS Div of Petroleum Chemistry Preprints, V. 25, No. 3, 1980, p . 621. 30. Griswold, C. F., "Steam Pyrolysis of Tosco II Shale Oil for Production of Chemical Intermediates," Master’s Thesis, 1976, Colorado School of Mines, Golden, Colorado. 31 . Fritzler, E. A., "Steam Pyrolysis of Tosco II and Simulated In-Situ Shale Oil for Chemical Inter mediates Production," Master’s Thesis, 1977, Colorado School Of Mines, Golden, Colorado. 32. Smith, P. D., "Stam Pyrolysis of Tosco II and Simulated In-Situ Sale Oil Vacuum Distillates for Production of Petrochemical Intermediates," Master’s Thesis, 1977, Colorado School of Mines, Golden, Colorado. 33. Kavianian, H. R., "Aromatic Formation and Distribution of Sulfur and Nitrogen in the Liqui Products of Steam Pyrolysis of Petrochemical Intermediates," Master's Thesis, 1978, golden, Colorado. 34. Angelos, C. P., "Production of Commercial Monoaromatics by Steam Pyrolysis of Hydrogenated Paraho Shale Oils," Master's Thesis, 1981, Colorado School of Mines, Golden, Colorado. T-2854 65 35. Ballut, A. A., "An Experimental and Theoretical Study of Steam Pyrolysis of Shale Oil Feedstocks for Petrochemical Intermediates," Ph.D Thesis, 1979, Colorado School of Mines, Golden, Colorado. 36. Kavianian, H. R., "A Simulation Model for the Steam Pyrolysis of Shale Oil for the Production of Chemical Intermediates," Ph.D Thesis, 1979, Colorado School of Mines, Golden, Colorado. 37 . Griswold, C. f., "Utilization of Shale Oil Feedstocks for Petrochemical Production," Ph.D Thesis, Colorado School of Mines, Golden, Colorado. 38. El-Kabule, A. S., "Millisecond Steam Pyrolysis of Shale Oil for Petrochemical Intermediates Production," Master's Thesis, Colorado School of Mines, Golden, Colorado. 39. Vanderkooi, W. N., "Pyrolysis Experimentation and Coputation," ACS Div. of Petroleum Chemistry Preprints, V. 19, No. 1, 1974, p. 125-139. 40. Nelson, W. L., "Petroleum Refining Engineering: McGraw-Hill Book Company, New York, p. 647. 41. Zdonik, S. B., and E. J. Green, "Function of Dilution Steam in Cracking," Oil and Gas Journal, May 27, 1968, p. 103-108. 42. Chambers, L. E., and W. S. Potter, "Design of Ethylene Furnaces: Part 2, Maximum Olefins," Hydrocarbon Processing, V. 53, No. 3, p. 95-100. 43. Davis, H. G. , and T. J. Farwell, "Relative and absolute Rates of Decomposition of Light Paraffins under Practical Operation Conditions," I&EC Process Design and Development, V. 12, no.2, p. 171-184. 44. Nowacki, P., Oil Shale Data Handbook: New Jersey, Noyes Data Corporation, 1981, p. 128-134 45. Feldkirchner, H.I., "The Hytort Process," Institute of Gas Technology Symposium Papers,"Synthetic Fuels from Oil Shale," 1979, p. 525-547. T-2854 66 46 Netzel, D.P., F.P. Miknis, "N.M.R. Study of US Eastern and Western Shale Oils, Produced by Pyrolysis and Hydropyrolysis," Fuels, v. 61, No. 11, 1982, p. 1101-1109. T-2854- 67 APPENDIX A COMPUTER PROGRAM LISTINGS AND NOTATION FOR EOSMP.FOR AND ESTIM.SIT T-2854- 68 COMPUTER PROGRAM NOTATION ARRAYS TT(5) Reactor thermocouple temperatures DT(5) Reactor volume increment/thermocouple V0L(16) Individual gas component volume RES(16) Gas chromatograph response factor COUNT(16) Component chromatograph peak area X(16) Mole % of components XX(16) Mole % of components (normalized) WM(16) Component molecular weight AT(16) Standard attenuations used for chromatograph FW(16) Weight % of gas F (16) Weight % of gas (normalized) FGIN0(16) Weight % conversion of oil VARIABLES TRMAX Maximum temperature ( F) inside reactor WOIL Weight of oil VH20 Volume of steam water OMW Oil molecular weight SP Specific gravity of oil VREAC Volume of reactor PRESS Atmospheric pressure TIME Run time (minutes) OILMOL Moles of oil OILVAP Oil vapor volume (ml) VSTEAM Volume of steam (ml) TOTVOL Total volum of gas feed TAU Reaction time (seconds) RATIO Steam to oil mass ratio TAUEQ Modified residence time (seconds) T-2854 69 PSF Severity factor using TAU PSF1 Severity factor using TAUEQ TIT,LE Run number POFG Pounds of gas produced CAL Peak area of calibration gas (methane) GOFG Grams of gas produced GTOO Gas to oil feed weight ratio FOFLO Liquid product to oil feed weight ratio FOFC Coke product to oil feed weight ratio FGINO Fraction of gas in oil T-2854- 7Q ■s o x •A j ("A » a • H s :» « a aCU— X Cb Sb JJVO 3 • x =«*»-4 CO as u « cuxxa x a u II >->cox 3 e-> as Cb Z X Cb u u cn «< J x X X u a. X M —. s Cb a * QvO eo x U —t x x >• z t IZ w A U H a HA • e-OS t-vO Cb *A u —a. vO u z —«a x 3 * x Cb ze-«a«-'t9 u X Uld m w o x xcn x «*u vun Q U VS J U M —AO XX cob.vo«-ie-ia. CO a CO HA X fiA SQ h w m > X CO z in X OX X » was-or 6- u AC CB X SB X u n n z v*i X X x ao • Cb- X ti-ei ’SuHfNraiScs*-• r-ifXCO'T. Ir u IfND i <»-. ts«s s * -1 CN a • K 3» ao « Z Cb VO s » *-i in II ^ «. V Cb X 3B VO X Cb vo z v a: ov 4-1 p * 3 N VO V5 • X -H N * % •v X WK^{J V « z fr-vOCb CN vO t-> z h o in N % 3V-C3 <-4 *. a x v n • Cb u vocn s • v 'N J io in cu b » <”WO SOS® a h vo^6- ta ® ® m pi w r w >=. r;. »s. rsi ® ® &< ® ® p. Kj *r "*»ao c ^oo' wn OV4»**-b-1 4** • • • 4 X •* X l O C O — 1 • c o m m w W 1 X X a x m i n »tn z «j » 4 u i - a -X o O ' t n t o 4 > • % xea Cb Z V T O C N f O X 'O 3 CNf-4-x x 3 4-4 W V ) 4 0 C b -X w % p» 4COC9 z 4 + x 3 *«• x i n • 3 Ob X Cb 4-X 400 us X X a 4rX0DCN « 4 X 4*4 S5 004»*«a % -X vo 4M inr» • • (X w V • • P M V O z W\OCJ V X*T3* t- 3 f-4-XCb V I - X X 4i US X Z x ^ O CD4a*0'Ov X ^ 4 s x u vocnaava x n o 44 O X X CD • • • • ♦ 4 4 o ^ w o t • 4 w * 4 r 4** • X n o jus N30N*tn CN c a X <-x-_| NtoJ *C— S S !■*! <1 < a osier.« !s»?s>»-'fN«'s>«a5»®U7:as»m'^®ir>p'Ssseastas>uip«*ss'«ia4sa sNUi^ * X* SV « »9JX X aemm c* >Ow • O » /«* * coao Cb qj ts> cs sj o if) k> intsi 6> & d E3 ca «b c» ess a &: es ea ea sa sc> os d aa © ® ea sj esi r-p't^t^r* (J Cfa o Ifa N s 0 J (-1 a a . W (Ml » N 5» (fa b] 1 <-* £ o (fa 1 tO 3 (fa I —4 co 044 (4# 1 o- N /Oft, 1 X 3 vO f- 1 X £*« -40 38 1 <* x W OW 0 1 —* 3 0 —40 X 1 <*s to sen >11 CO * — KG;eac«<»Q>e>Pie:e&QBe:«sdeBC£CBBBt£S(^escuc=Cic;e9c;b:SiEiS(BGca^cae3isiG»OQA atasasa s» casta Barca's ■a *azie»«na'sii5«a'sna,3r'a;a;ae»3iis*s;si'sea's!a's«fa>a-3»''54a?s(a>si ^r4n,Tir5vor'Coa'5Ei-4rMfo^,invor»ooO'»s»HrM«n,5rirvoc~ ARTHUR Lft*£s UBivnAi -STcSa- si. T-2854 75 U ra.-trMm^rtn'O -‘Z z- m j *■— 1» I» i> Ii > N > «•% U3f*CMlOO VSSHXHN* ^ 2 3 X 3 3 3 3 ^ » <^3S55ZI«4 » «*>33B3B32E33l3356b 6->U3Ub:CL]b:Ub]Ub]h}U'!UUUUUwww>^A^A^^w«^wv^ww "••••*••••»••••* » c IC H iC iC it it it it »C 't It i[ iC it-*C ’< Z m txctatreQSQSQ;ca£Kiz;Q;iZ:QsaiCco;ir-*Q sb.iLU.iaLCbb.b.b.tbtbit.tbbCbibaaoooaaaooaoaoootds rOMl II II 11 ‘I II Ii—1 II II 1| II II II II Ifiifi if I iCiifi if! ifi III ifi iff I'lifiifi ifi ifi ,r../>r-.] ShHlNWIfl'O cse-osearatss.ore?a eats co a APPENDIX A1 SAMPLE CALCULATIONS AND EQUATIONS T-2854 77 SAMPLE CALCULATIONS 1. Reactor Volume Calculations a .ball volume by volumetric displacement 18.7 ml. H20/10 balls = .4675 ml/ball 27.7 ml. H20/60 balls = .4671 ml/ball 32.5 ml. E^O/70 balls = .4643 ml/ball average = .464 b. packing volume (705 balls)(.464 ml/ball) = 327.12 ml. - = 19.95 inches c. thermocouple volume (3•14/4)(1/4)^(12 inches) = .589 inches^ d. reactor volume (3.14/4)(2 inches)^(12 inches)= 37.70 inches^ e. quench tube reactor volume (2 inches)(3/4 inches)^(1/4)(3.14) o + (2 inches)(1/2)(1/2)(1/4)(3.14) = 1.276 inches f. feed tube volume (3 inches ) (3/8)^ (1/4) (3 • 1 4) = *331 inches*^ g. total reactor volume 37.70 + 1.28 + .33 - .59 -19.95 = 18.767 inches3 = 307.54- cm3 2. Approximate interval defined per thermocouple a. void volume per inch of reactor (37.70 - .589 - 1 9. 95)inches^/1 2 inches= 1 .438.-^— m reac. 2854- 78 adjusted residence time= 6 o (-E/R(incremental T R ))(residence time)(void frac.) 2______(-E/R(maximum reactor temperature) steam to hydrocarbon mass ratio _ mass of H20 in mass of oil in product composition 1. chromatograph component area= (peack area)(attenuation)(response factor) 2. 100$ methane calibration= (peak area) (1024.) 3. product mole fraction= chromatograph component area 100$ methane calibration 4. normalized product gas mole fraction= product mole fraction ^ product mole fraction 1 5. molecular weight of product gas= 1 6 (molecular weight)(normal mole fraction) 1 6. product gas mass at metered P and T= (P/RT)(volume of product gas produced) molecular weight of product gas 7. gas weight $ of feed= (product gas mass)(100) mass of oil in T-2854 79 8. solid weight % of feed = (solid mass)(100) mass of oil in 9. liquid weight % of product gas = 100-gas$-solid$ 10. component weight % of product gas = (normalized mole percent)(molecular weight) product gas molecular weight 11. component wt % of feed = (gas wt% of feed)(component wt % of product gas) 12. volume of product gas per pound of feed at standard conditions (RT/P)(component wt % of feed) component molecular weight 13. pyrolysis severity factor = (reaction temp °F)(residence time)*^ (ratio)*^ T-2854 80 APPENDIX B COMPUTER PROGRAM EOSMP.FOR DATA OUTPUT T-2854 RUNt'! = 4SSH7'5>i3 TR= 1855.£**= .7b29TJWI = ,8551PSF= 1365.15PSF1= 1 3 3 6 .25T A»JE(i= .5986 COUJ in-: rJ o z ■oz o = o =: too /O C > c/iO cn 2 Ul C£ j - CU 2 U3 o <#> O fr-« 9P Cb a S *■4 3 3 rsi *-4 X 6- J o : U: r* TT CN CO K CM rN ro Ci- <5’ CM ro ro O' H oo o> o> vo Cii o G G •sO S co es» G ra -o •a • • • • • • • 13394 • • • • • • • rl u> CO ro CN CM ' O G in K. H C* CN 00 V G in in * o u< o jcnocco i > com o c •>• >a u £ u. >*a. n m O cn m o c j ** U u «x CO OK9ZZZ_J«*C9CO£- b c a - E O C O C Z 3 1 3 - £ O C 9 C * « J _ Z Z Z 9 K CO 1 6 o O Cb c a l lib e- oo 1 3 3 OP z n c v o o a ' m p w o v N C ' O n r v N f i w n C 09 CO CO 00 V h. «< e- z u CO s v o i a x - v s s e a S M S S M O S H O N i • • • • • • • • • • • • • • • • ! • t • • • • • • • • • • # • « « • ! pioixcLacixscciz pioixcLacixscciz z=HME->UHUtocoH o c o t U H U > - E M H = Qi—i Cju ZIb» —«e-«OCfaZ X-4tfaCfaX
ra e;< O' in G NCO CN G <7* G 00 G in OD <3* TO G OG VO G in G • • • • • j 0 *"4 fN n cn H G c. - c G H IN ro G G G g G G G G G G G "a