Producing, Evaluating, and Upgrading Oils from Steam Liquefaction of

Chapter 9

Producing, Evaluating, and Upgrading Oils from Steam Liquefaction of Poplar Chips

Downloaded by EAST CAROLINA UNIV on October 30, 2016 | http://pubs.acs.org Publication Date: September 30, 1988 | doi: 10.1021/bk-1988-0376.ch009

D. G. B. Boocock, S. G. Allen, A. Chowdhury, and R. Fruchtl Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto M5S 1A4, Canada Chipped poplar wood can be liquefied using only high temperature/pressure steam (300-355°C). Approximately 60 per cent of the carbon feed appears in the o i l phase (acetone-soluble). The oils (= 45% mass yield) are just solid at room temperature but soften around 50°C (1.5% moisture). They typically contain 69% carbon, 5.8% hydrogen and about 25% oxygen. Both weight and number-average molecular weights, as measured by HPSEC, are less than 1000. The major fraction of the o i l is of lignin origin. Model compound studies show that the major o i l functionalities (phenolic OH and OCH ) can be hydrodeoxygenated using typical hydrodesulphurisation catalysts. 3

!

In the early 1970 s, i n response to the world o i l c r i s i s , studies on the d i r e c t thermal liquefaction of biomass were i n i t i a t e d . These studies could be c l a s s i f i e d into those on water-based processes and those on non-water-based processes. The focus i n this chapter i s on the water-based processes and, i n p a r t i c u l a r , on a process which uses no catalysts or reducing gases. I t should be noted that actual biomass used f o r commercial liquefaction would c e r t a i n l y contain s i g n i f i c a n t moisture, and water i s expected as a product of the l i q u e f a c t i o n . Despite t h i s , the water-based and non-water-based processes are s i g n i f i c a n t l y d i f f e r e n t . Two types of water-based processes were i n i t i a l l y studied. The f i r s t of these, based on Bergstrom's e a r l i e r work (1), employed sodium carbonate as a soluble catalyst and carbon monoxide as a reducing gas (2). The second technology, also based on e a r l i e r work, used n i c k e l metal catalysts and hydrogen (3). In a l l cases the substrate was powdered wood s l u r r i e d i n water. In both processes the role of the added chemicals was not clear. Sodium carbonate would i n i t i a l l y dissolve any phenolic and other a c i d i c products. However, i t s major role was assumed to be an aid f o r the addition of carbon monoxide v i a the formate ion. As we have shown, wood w i l l liquefy without the addition of sodium carbonate and carbon monoxide. Researchers at the University of Saskatchewan also noted that removal

c

0097-6156/88/0376-0092$06.00/0 1988 American Chemical Society

Soltes and Milne; Pyrolysis Oils from Biomass ACS Symposium Series; American Chemical Society: Washington, DC, 1988.


9. BOOCOCK ET AL.

OUs from Steam Liquefaction ofPoplar Chips

93

of carbon monoxide from their screw reactor system did not noticeably change the reaction, although they never went the extra step of eliminating the sodium carbonate (4)· We chose to study the nickel/hydrogen system, but here again i t was unclear what role the n i c k e l catalyst played with regard to a s o l i d substrate. I t was reasoned that gases or liquids must be formed before the catalyst could intervene. Since very l i t t l e l i q u i d was formed on slow heating i n the absence of n i c k e l and hydrogen, i t was also reasoned that the catalyst/hydrogen system must s t a b i l i s e the products. The early experiments were performed i n batch autoclaves, but subsequently we fed water s l u r r i e s of powdered wood and n i c k e l carbonate semicontinuously to a hydrogen-containing reactor (5). The feed was heated so fast that the n i c k e l carbonate decomposed to n i c k e l oxide instead of reducing to n i c k e l metal. In addition, when product was discharged from the reactor, both char and o i l — t h e l a t t e r i n 25 per cent y i e l d — w e r e present. We reasoned that the o i l y i e l d was i n i t i a l l y higher, but, because of prolonged residence times and the absence of the desired s t a b i l i s i n g system, some of the o i l had recondensed and charred. As a test of this theory, powdered wood was heated rapidly to 350°C, together with water only, i n small reactors heated by a f l u i d i s e d sand heater. O i l y i e l d s , up to 50 per cent by weight (based on dry wood) were obtained (6). U n t i l this time, a l l studies on water-based processes had been confined to powdered wood or sawdust, the general concern being that heat and mass transfer limitations i n wood pieces would prevent l i q u e f a c t i o n . We, therefore, studied the aqueous liquefaction of single poplar sticks (6.5 mm square cross section) i n the same small reactor and showed that complete l i q u e f a c t i o n occurred (no char) at 300°C and above (7). Although heating was f a i r l y rapid (2 minutes) for most experiments, heat-up times were extendable to at least 11 minutes without dramatic changes i n o i l y i e l d . By quenching the reaction at various temperatures and noting the change i n the physical state of the s t i c k s , as well as noting the change i n heat transfer c h a r a c t e r i s t i c s to the internal thermocouple, we were able to define the l i q u e f a c t i o n process at the macro l e v e l , as seen i n Figure 1. I n i t i a l l y steam enters the chip and swells i t . This results i n disruption of the matrix which then collapses to release condensed water. When too l i t t l e water i s present, the swelling phase removes free l i q u i d water from the reactor and results i n decreased heat transfer. When the matrix collapses, the reappearance of l i q u i d water increases the heat transfer. Chemical depolymerisation then results i n l i q u e f a c t i o n . The presence of l i q u i d water appears to retard the liquefaction s l i g h t l y . I t also s t a b i l i s e s the o i l and prevents charring. Samples quenched at temperatures of 290°C and 300°C do not t r u l y represent the swelling phase, since they have been dried (see Figure 1). Electron scanning microscope studies of the sticks have i d e n t i f i e d the i n i t i a l phases of the l i q u e f a c t i o n at the c e l l l e v e l (8). On the surface, the middle lamella i n i t i a l l y disrupts, but this i s quickly followed by fusion of the adjacent c e l l walls. The flowing matrix then engulfs the c e l l c a v i t i e s . Inside the s t i c k a d i f f e r e n t sequence of events occurs. Spherical structures appear, p a r t i c u l a r l y on vessel walls. These spherical structures, which result from


94

PYROLYSIS OILS FROM BIOMASS

softening matrix, eventually f i l l the i r r e g u l a r c a v i t i e s formed by the breakdown of c e l l walls, and gas and vapour bubbles can be seen within the flowing matrix. It i s assumed that this process continues, although this could not be observed, since the wood eventually had too l i t t l e mechanical strength for sample preparation, even using our special freezing techniques. More recently, we have investigated the liquefaction of softwoods. The l i q u e f a c t i o n process i s slower which i s probably due to decreased wood porosity. However, the spherical structures are much more prevalent during the l i q u e f a c t i o n (Figure 2) .


The Reactor On the basis of these encouraging r e s u l t s , i t was decided that a laboratory unit should be constructed for the purpose of studying the liquefaction of commercial-size (and larger) poplar chips (see Figure 3). A batch system was chosen employing a gravity feed and discharge system. The reactor i s designed to hold 100-120 g of poplar chips and as such has an internal volume of 700 mL. A single ingot of TP 316 stainless steel was used for machining the reactor which has an internal diameter of 1.5 inches and an external diameter of 3.0 inches. The length i s approximately 19 inches. At the top an Oteco hub i s threaded and seal welded to the reactor body. A one inch slim l i n e connector i s threaded to the base of the reactor. A t o t a l of 9 holes are d r i l l e d i n the reactor side to take i inch slim l i n e connectors. Three of these are for the steam i n l e t l i n e s . Three others are for 1/8 inch thermocouples, and one of the remaining three i s for the rupture disc l i n e . A vent line and pressure gauge occupy the f i n a l two holes. The reactor i s heated by two 6 feet long heaters joined i n p a r a l l e l . These are c o i l e d around the reactor and held close to i t by 4 longitudinal steel s t r i p s and eight c i r c l i p s . The maximum power drawn by the heaters i s 4 KW. The reactor i s insulated with ceramic brick which i s cut and f i t t e d to the contours of the external surface. The i n l e t valve i s a 1.5 inch b a l l type (Mogas Industries Ltd.) rated at 24.8 MPa (3600 psi) at 370°C. It i s joined to the reactor by a matching hub. The c o n t r o l l e r i s a i r operated and f a i l u r e of a i r pressure causes the valve to close. The valve i s insulated by three layers of 1.5 inch thick glass wool. A similar valve was planned for the outlet valve. However, because of cost considerations, a 0.5 inch b a l l valve (Crosby) rated at 18.4 MPa (2665 psi) at 370°C i s currently being used. This valve operates well, but we are reluctant to operate i t close to i t s design r a t i n g . The cooling lock has an internal column of 300 mL and i s machined from stainless s t e e l . An outer jacket allows for a variety of coolants. The steam generator i s a 2-L autoclave. A MiltonRoyal high pressure pump allows for the continuous addition of water to the hot generator, i r necessary. The product c o l l e c t o r i f a pyrex vessel approximately 8 inches i n diameter and 5 inches deep. Separated gas passes to a brine displacement vessel for volume measurement. A number of e l e c t r i c a l and mechanical overrides plus barriers are designed to prevent any injury to the operator as a result of accidental discharge from the reactor or steam vessel. In addition, the control panel i s located i n a separate room adjacent to the reactor.



9. BOOCOCKETAL.

original

250

OUs from Steam Liquefaction of Poplar Chips

270

278

290

300

310

319

330

95

350

Figure 1· Liquefaction of poplar sticks i n water (dried samples).

Figure 2. Scanning electron microscope picture of i n t e r i o r of black spruce s t i c k (surface heated to 309°C i n water).



96


Reactor Operation In operation, the reactor i s brought to temperature with the i n l e t valve open. A slow nitrogen flush through the i n l e t lines i s also employed. The feed i s contained i n a c y l i n d r i c a l steel mesh basket which i s manually dropped through the i n l e t valve. The purpose of the basket i s three-fold. F i r s t , i t allows rapid charging of the feed to the reactor. Secondly, i t prevents direct contact between the chips and the reactor wall; and t h i r d l y , i t retains any unconverted wood i n those experiments when large-sized feed and low reaction times and temperatures are used. After addition of the feed, the i n l e t b a l l valve i s immediately closed and the steam l i n e opened. Mechanical and e l e c t r i c a l overrides preclude accidental ejection of steam through the i n l e t valve. Ordinarily, steam i n j e c t i o n time i s short (3-7 s ) , since the r e l a t i v e low temperature at the bottom of the reactor encourages flushing of the steam vessel contents. After closure of the steam valve, a further two minutes i s allowed for r e action before the outlet valve i s opened and the products are allowed to enter the lock. After a suitable cooling period (up to 1 minute) the products are discharged to the product c o l l e c t o r . The basket i s retrieved through the i n l e t valve after venting the reactor. Results Figure 4 shows t y p i c a l temperature p r o f i l e s immediately after closure of the steam l i n e s . The temperature p r o f i l e highlights the major drawback of the reactor. This i s the r e l a t i v e l y low temperature at the base of the reactor due to (a) the low rating and low thermal mass of the outlet valve; and (b) the thermal conduction tô the cooling lock. Opening of the steam valve thus causes flashing of steam to this cooler area. I n i t i a l l y the middle and upper thermocouples are hotter than the steam, but the upper thermocouple rapidly cools as steam enters. The steam cools on expansion, and thus has a temperature less than 340°C when entering the reactor. Although heating and subsequent cooling of the bottom thermocouple i s occasionally observed ( p a r t i c u l a r l y i f the temperature i s below 300°C), the most common observation i s that the base temperature varies very l i t t l e . After approximately two minutes the temperatures at the top and bottom of the reactor are almost equal. However, the temperature at the middle i s t y p i c a l l y 40-60°C higher after 120 s of reaction time. O i l j water and gas exit the cooling lock and enter the c o l l e c t i o n v e s s e l . In a l l , over 30 runs have been made, and of these 13 have been completely analysed i n terms of carbon balance, elemental analysis and q u a n t i f i c a t i o n of the various phases. For chipped poplar, the varied parameters have been steam temperature, i n j e c t i o n time, water l e v e l i n steam generator and t o t a l reaction time. The l i q u e f a c t i o n of non-standard poplar i n the form of dowels has also been studied. Overall mass balances can be obtained, but these are not meaningful, given the r e l a t i v e l y large amounts of water involved i n the reaction. More important aspects are the carbon balance and the percentage of feed carbon found i n the various product phases. These phases are the o i l , including a separately obtained acetone wash o i l



9.

BOOCOCKETAL.

OUs from Steam Liquefaction of Poplar Chips

97

from cleaning the reactor, the aqueous phase and the gas phase. In addition, a small amount of insoluble material (usually acetonesoluble) i s found i n the aqueous phase. Figure 5 shows a t y p i c a l carbon d i s t r i b u t i o n i n the product phases. The gas phase i s 93% carbon dioxide, the balance being mostly carbon monoxide. About 7 l i t r e s of gas i s obtained for each 100 g of chipped wood (7% moisture). Thus approximately 7 per cent of the wood carbon appears i n the gas phase. The volume of the aqueous phase i s d i f f i c u l t to control, because of the flashing of steam to the reactor. T y p i c a l l y the volume of aqueous phase i s close to 300 mL for 100 g of chipped poplar. Interestingly, when 160 g of poplar i n the form of a poplar dowel 23 inches long and 1 inch i n diameter is l i q u e f i e d , the volume of water collected i s the same. This suggests that there i s a trade o f f between steam,required to heat the wood, and the i n i t i a l void space. The o v e r a l l major finding i s that, for wood chips, there i s very l i t t l e v a r i a t i o n i n carbon d i s t r i b u t i o n amongst the various product phases for reaction times of 45-120 s and steam generator temperatures of 335-355°C. This i s only true i f the amount of feed i s held at 100 g, and the steam vessel i s r e l a t i v e l y f u l l of water. Products The Aqueous Phase. As mentioned previously, the amount of aqueous phase collected under t y p i c a l operating conditions appears to be independent of void space or mass of wood i n the reactor. In general, s l i g h t l y less than 300 mL i s usually collected for each run. The concentration of 'soluble carbon i n this water i s about 45 g/L when 100 g of wood i s used. This corresponds to 15 g of carbon or about 30 per cent of that i n the wood. HPLC analysis has been performed for simple carboxylic acids (9) and some sugars. A more detailed study of the aqueous phase i s available for the small reactor experiments i n which single poplar sticks are subjected to hot water and steam generated i n t e r n a l l y . The combined aqueous phases from s i n g l e - s t i c k experiments have been analysed for heartwood, sapwood, and bark feeds. The a n a l y t i c a l procedures involved extraction of the aqueous phases with r e l a t i v e l y large amounts of ether. However, this ether extract only contained about 10 per cent by weight of the o r i g i n a l feedstock. Since this would leave the carbon balance d e f i c i e n t , i t i s concluded that the ether did not capture a l l the aqueous phase. Rapid hydropyrolysis of isolated poplar l i g n i n i n the same reactor not only yielded twice the amount of o i l but also produced much less ether extract from the aqueous phase. It i s thus concluded that the majority of organic compounds i n the aqueous phase derive from the h o l o c e l l u l o s e s . Table 1 supports this showing the major products as acetic acid, other carboxylic acids and f u r f u r a l s . While hemicellulose accounds for most of the a c e t i c acid and f u r f u r a l , we have shown (10) that the c e l l u l o s t y i e l d s the other f u r f u r a l d e r i v a t i v e s . The phenols obviously derive from the l i g n i n . 1

The O i l . The o i l y i e l d i s t y p i c a l l y 40-45% on a dry mass basis, and the carbon content i s about 59% of that i n the o r i g i n a l feed. This y i e l d varies very l i t t l e using steam temperatures of 335-355°C and


PYROLYSIS OILS F R O M BIOMASS

MESH BASKET

0.5

Inch


BALL

VALVE

RECEIVER

F i g u r e 3.

380

Schematic d i a g r a m o f wood l i q u e f a c t i o n u n i t .

H

τ

0

1

10

20

1

30

1

40

1

50

1

1

1

60

70

80

Γ

90

100

110 120

Time, s

F i g u r e 4.

Temperature p r o f i l e s a t t o p and bottom o f r e a c t o r . ACETDNE SOLUBLE D I L

58.8·/

ACETDNE INSDLUBLES 30.27. AQUEDUS PHASE Figure 5.

Carbon d i s t r i b u t i o n i n p r o d u c t phases.


9. BOOCOCKETAL.


Table I.

99

Oils from Steam Liquefaction of Poplar Chips Compositional Analysis of Ether-Solubles from the Aqueous Phase*

Substance

Sapwood

Heartwood

Bark

Douglas F i r

Acetic acid Other a l i p h a t i c acids Furfural 5-Methylfurfural 5-Hydroxymethy1furfural Phenols (including guaiacol, syringol and a l k y l derivatives) L e v u l i n i c acid and lactones Unidentified

30 5 30 5 3

40 10 10 3 4

20 10 12 10 5

10 5 10 10 8

5

5

10

5

7 15

10 18

5 28

20 32

* figures represent wt.% of the aqueous phase ether-solubles

reaction times of 45-150 s. The o i l softens around 50°C. On average the carbon and hydrogen contents of the dry o i l are 69% and 5,75%, respectively. The oxygen content i s about 25% and the nitrogen i s also measurable (0.3%). As mentioned previously the o i l i s separated e a s i l y from the bulk aqueous phase by f i l t r a t i o n . The trapped water (up to 20%) can be separated by then melting the o i l , and this lowers the water content to less than 5%. These two steps could, of course, be combined by preventing rapid cooling of the o i l . However, i n the laboratory we find i t convenient to operate i n the two step mode, although i t would be r e l a t i v e l y easy to provide heating to the product c o l l e c t i o n vessel. It should be noted that fast (non-aqueous) pyrolysis o i l s contain considerable quantities of water which i s inseparable because of the high content of poplar compounds. A b o i l i n g point d i s t r i b u t i o n curve, together with thermal evaluation analysis data, suggests that 36% of the o i l d i s t i l l s below 405°C which i s at the upper end of the range for heavy gas o i l i n petroleum d i s t i l l a t i o n . However, proton and carbon-13 nmr spectra c l e a r l y show that the o i l i s not alkane i n character. A strong methoxyl signal i s evident at 4.2 ppm i n the proton spectrum (Figure 6) and 56 ppm i n the carbon spectrum (Figure 7) and t h i s and other evidence indicates that the o i l i s predominantly of l i g n i n o r i g i n . Many of the sharp peaks found i n the carbon-13 spectra of fast pyrolysis o i l s are absent, and there i s evidence of peak broadening in both proton and carbon spectra. Spectra of co-mixtures of steam pyrolysis o i l and fast pyrolysis o i l show the broadening i s r e a l and is not caused by metal content i n the steam pyrolysis o i l (11). Solvent-separation of the o i l s , using diethyl ether, chloroform and acetone, followed by analysis, support the proposed o r i g i n of the o i l . Ether e s s e n t i a l l y dissolves monomeric material, while chloroand acetone dissolve successively higher molecular weight materials. Phenols, guaiacols and syringols are found i n the ether-soluble f r a c t i o n along with a l i p h a t i c acids (C-5 and higher) and various benzoic acids, the l a t t e r being of l i g n i n o r i g i n . The carboxylic acid f r a c t i o n appears to be twice as prevalent as the phenolic f r a c t i o n . The neutral f r a c t i o n showed spectral (nmr, EIMS)



100

TO


9

$

7

6

F i g u r e 6.

5

4

3

2

ϊ

OPPM

P r o t o n nmr spectrum o f o i l .

Solvent

200

120

160 Figure

7.

80

40

Carbon-13 nmr spectrum o f o i l .


20

P

P

M

0

9. BOOCOCK ET AL.

Oils from Steam Liquefaction of PopforChips

101

c h a r a c t e r i s t i c of straight chain a l i p h a t i c protons and may contain extractives reported to be mono-, d i - , and t r i g l y c e r i d e s (10). Spectroscopic evidence shows that the chloroform soluble f r a c t i o n i s of both l i g n i n and c e l l u l o s e o r i g i n . In addition, the use of either l i g n i n or c e l l u l o s e as feedstock produce s i g n i f i c a n t amounts of this f r a c t i o n (ether-insoluble/chloroform-soluble). The r a t i o of syringyl and guaiacyl contributions to t h i s f r a c t i o n can be estimated from the C-13 nmr spectra. This r a t i o i s 4:1 f o r heartwood, sapwood and l i g n i n o i l s . A l l the thermally produced o i l s are thermally unstable and, at s u f f i c i e n t l y high temperatures, 'react to form a char. This i s p a r t i c u l a r l y true of the fast pyrolysis o i l s which appear to have undergone less chemical conversion and contain more reactive f u n c t i o n a l i t i e s . Special processes, such as entrained flow, may have to be used to upgrade a l l the o i l s . High performance size exclusion chromatography (HPSEC) has been used to determine apparent molecular weight d i s t r i b u t i o n s and also to determine the effects of aging (12). The number and weight average molecular weights are less than 1000 based on polystyrene standards. Work i s currently being done to quantify variations i n detector response with molecular weight and sample concentration. The o i l s were found to age slowly at room temperature but rapidly repolymerized when heated above 100°C.


1

Upgrading of O i l s . I t i s obvious from examination of the o i l that i t must be transformed i n at least two ways. One of these transforma tions involves further molecular-weight reduction. This has to be achieved whilst avoiding the natural tendency of the o i l to react condensatively at high temperature y i e l d i n g chars. Some kind of entrainment may be necessary to obtain t h i s desired cracking at high temperature. Our own studies have been confined to methods f o r removing and/or changing the phenolic OH group and methoxy groups i n the product. The objective of this study was to f i n d ways to control the transformations i n specified directions and, therefore, provide maximum v e r s a t i l i t y . Although we have used many aromatic substrates, anisole and phenol have been the major focus of our work and our discussion w i l l be limited to them. We decided there were d e f i n i t e advantages to be gained i f catalysts already used for hydrodesulphurisation of petroleum could be used for deoxygenation and rearrangement of f u n c t i o n a l i t i e s . We have, therefore, studied molybdenum oxide/cobalt oxide (15%/4%) and molybdenum oxide/nickel oxide (15%/3%) (both on γ-alumina) catalysts both i n sulphided and unsulphided forms. A batch reactor was used for the study and, therefore, much larger conversions were obtained than i n previously used flow reactors. Disadvantages of such a system include the heat-up time required for the reactor and the non-removal of product water from the system. It was concluded from the study of phenol that the highest benzene production occurred when the sulphided cobalt catalyst was used at r e l a t i v e l y low temperatures and hydrogen pressures (350°C, 1.4 MPa). Ratios of benzene to cyclohexane y i e l d s are 14:1 or higher. However, the reaction rate may be too low at this temperature, but r a i s i n g the temperature and pressure causes the r a t i o to f a l l (2:1 at 450°C and 2.8 MPa). High conversion can be obtained at 450°C using the



102


unsulphided cobalt catalyst. The aromatic to cycloalkane r a t i o i s 5:1, but half the cycloalkane i s methylcyclopentane. The l a t t e r product i s favoured by the higher temperatures. The unsulphided n i c k e l catalyst i s s l i g h t l y more active at 450°C, but the aromatic/ cycloalkane r a t i o i s not as good (3:1). I f cyclohexane i s the desired product, i t i s best to use the sulphided n i c k e l catalyst at a r e l a t i v e l y low temperature (350°C), when a cyclohexane/benzene r a t i o of 11:1 r e s u l t s . Anisole readily demethylates, even i n the absence of hydrogen. Even without the catalyst, some demethylation occurs. At 350°C, using the unsulphided cobalt catalyst, some of the lost methyl groups alkylate the aromatic ring to y i e l d methyl and dimethylphenols. However, only 68% of the methyl groups can be accounted for ( i n c l u ding 14% unconverted anisole). The same accounting holds at 450°C when deoxygenation yields toluent (21%), xylene (12%), and even some trimethylbenzenes (3%). The use of the sulphided cobalt catalyst results i n a greater loss of the methyl groups (77%). Methylation of the ring i s thus obviously favoured by the oxided form of the catalyst. The lost methyl groups are probably converted to methanol or dimethyl ether, but we have not analysed for these compounds. An interesting observation was that recycled (3 times) sulphided cobalt catalyst gave enhanced benzene yields (68% vs 49%). Deoxygenation was v i r t u a l l y complete, whereas fresh catalyst l e f t s i g n i f i c a n t unconverted phenol. Obviously the degree of sulphiding has not yet been f u l l y exploited. Thrice used catalyst gave a r a t i o of aromatics to cyclohexane of 6.5:1. We suspect that similar effects would be observed i f phenol were used as substrate. Acknowledgments We thank the Natural Sciences and Engineering Research Council of Canada, Imperial O i l and Energy Mines and Resources (Canada) f o r supporting various aspects of this research.

Literature Cited 1. 2. 3. 4. 5. 6. 7. 8. 9.

Bergstrom, H. O.; Cederiquist, Κ. N. U.S. Patent 2 177 557, 1937. U.S. Bureau of Mines Technical Report of Investigation #7560, Appell, H. R.; Friedman, S.; Fu, Y. C.; Wender, I.; Yavorsky, P. M., 1971. Boocock, D. G. B.; Mackay, D.; McPherson, M.; Nadeau, S. J.; Thurier, R. Can. J. Chem. Eng. 1979, 47, 98. Eager, R. L . ; Mathews, J . F.; Pepper, J. M.; Zohdi, H. Can. J . Chem. Eng. 1981, 59, 2191. Final Report of Contract File #24SU.23216-3-6143, Boocock, D. G. B.; Renewable Energy Division, Energy, Mines and Resources: Ottawa, Canada, 1984. Beckman, D.; Boocock, D. G. B. Can. J . Chem. Eng. 1983, 61, 80. Boocock, D. G. B.; Porretta, F. J . Wood Chem. and Technol. 1986, 6, 127. Boocock, D. G. B.; Kosiak, L. Can. J . Chem. Eng. 1988, 66, 121. Status Report, McKinley, J., Industrial Chemistry Division, B.C. Research: Vancouver, B.C., 1987.


9. BOOCOCK ET AL.

Oils from Steam Liquefaction of Poplar Chips

Boocock, D. G. B.; Kallury, R. K. M. R.; Tidwell, T. Anal. Chem. 1983, 55, 1689. 11. Final Report of Contract File #38 ST.23216-6-6346, Boocock, D. G. B.; Chowdhury, A. Z., Renewable Energy Division, Energy, Mines and Resources: Ottawa, Canada, 1988. 12. Allen, S. G. B.A.Sc. Thesis, University of Toronto, Toronto, 1987. Received May 19, 1988


10.


103

Producing, Evaluating, and Upgrading Oils from Steam Liquefaction of

Recommend Documents