The effect of vapour residence time on organic liquid yield is relatively well understood although the interaction of temperature and residence time is less understood. Studies by Diebold et al., Liden et al and Knight et al. have attempted to interlink both primary liquids formation and secondary cracking, but one essential component which is neglected is the variation of water yield with temperature and residence time. It is believed that at temperatures below 400*C, secondary condensation reactions occur and the average molecular weight of the liquid product decreases. Boroson et al. have demonstrated that the average molecular weight decreases with the degree of secondary reaction, i. e. increasing residence time and temperature (10).

For chemicals, it is considered necessary to "freeze” the process at the appropriate time-temperature point in the envelope to maximise yield. In one case this has led to a commercial reactor design where vapour residence times down to 90 ms are claimed.

Fuels have less specific process requirements and most work has focused on maximising liquid yield rather than product quality. The window for fuel production requires more R&D to better understand the processes and match the product quality requirement to process parameters. There is no definition of product quality in terms of physical or chemical properties or composition, and this area will need to be addressed as more applications are tested and alternative supplies of bio­fuel-oil become available.

Long vapour residence times and high temperatures (> 500*C) cause secondary cracking of primary products reducing yields of specific products and organic liquids. Lower temperatures (< 400 eC) lead to condensation reactions and the subsequent formation of lower molecular weight liquids which can also react.

The objective of the process was charcoal production from which a by-product oil was derived by condensation of the vapours in the offgases. A continuous, 50 kg/h demonstration plant process was available in Switzerland and several commercial plants have been sold up to 2 t/h capacity. The oil from one of these units in Spain was successfully utilised in a local hospital boiler. This also supplied oil for testing across Europe at the same time that the Alten plant was operating and served to both encourage interest in direct production of liquid fuels as well as to cause concern about the quality of the oil and the problems in handling and use from high viscosity and water separation. The product was a secondary oil from the long vapour residence times and high formation temperatures and thus had a high viscosity and low water tolerance leading to phase separation in some samples.

The following technologies were proposed:

• Pyrolysis

• Gasification

• Liquefaction

• Combustion

• Upgrading

• Chemicals production

For each technology the following data should be provided:

maximum 50 words

for cross referencing to DATABASE and PRODUCTS Can be established to link to a FEEDSTOCK file

kg/h feed d. a.f. basis kg/h feed d. a.f. basis kg/h feed d. a.f. basis

for cross referencing to DATABASE and PRODUCTS for cross referencing to DATABASE and PRODUCTS

A detailed review of the science and technology of fast pyrolysis and related processes has been carried out covering the basic scientific principles of fast pyrolysis for producing liquids; their application to working processes with identification of the main problems; a description of the more commercially and technically advanced of these processes; and opportunities for chemicals recovery and upgrading of vapours and liquids to more valuable chemicals and transport fuels. The contribution that has been made to the development of a Biomass Information System is described and copies of specimen data are included. Finally the conclusions summarise the current status and opportunities for each of the areas described.

3.6.1 General properties

The crude pyrolysis liquid is dark brown and with low viscosity which approximates to biomass in elemental composition, and is composed of a very complex mixture of oxygenated hydrocarbons with an appreciable proportion of water from both the original moisture and reaction product. Solid char (39, 40, 41124, 125, 126) and dissolved alkali metals from ash (124) may also be present.

The complexity arises from the degradation of lignin, cellulose, hemicellulose and any other organics in the feed material, giving a broad spectrum of phenolic and many other classes of compounds that result from uncontrolled degradation as described above. The liquid from fast or flash pyrolysis has significantly different physical and chemical properties compared to the liquid from slow pyrolysis processes which is more like a tar.

The primary liquid product is readily combustible and can be used directly, for example in boilers and kilns, or it may be subjected to further chemical processing to give a higher quality fuel or chemical product. The characteristics of the liquid and descriptions of the technologies for upgrading are summarised in Table 3.4 below.

Changes occur in storage due to moisture content changes which affects viscosity, oxygen absorption or reaction, polymerisation and other innate chemical inter­reactions. The effect manifests as changes, usually increases, in viscosity, and the extent and rate of change depends on the mode of production, particularly residence or contact time, feedstock, and conditions of storage. These effects have not yet been fully evaluated, but are being investigated.

5.15.1 Summary

Union Fenosa undertook the construction of a substantial pilot plant in 1989 for the production of pyrolysis liquids for fuels. After a thorough review of available technologies at that time, a license was agreed for the Waterloo Flash Pyrolysis Process (WFPP) developed by the University of Waterloo in Canada (q. v.).

Construction started in 1990 and commissioning started around October 1992. By mid 1993 the unit was operating reasonably satisfactorily at 160 kg/h dry wood throughput, although some problems with liquid collection had to be resolved. Work is continuing on developing the process and considering scale-up.

5.15.2 Description

The flowsheet is shown in Figure 5.13 (6, 69). After storage the eucalyptus feed is ground then dried with propane. A cyclone removes fines prior to venting the combustion products to atmosphere. The dried feed is stored in a closed hopper prior to metering to the fluid bed pyrolyser. The fluidising gas is recycle gas supplemented with nitrogen as required. The gas is heated with a propane fired heat exchanger.

The product char is separated in a cyclone and stored in a closed container. The vapours are cooled in two water cooled heat exchangers in series, then passed through a demister and a final cooler to remove as much oil as possible. All liquid product streams are combined.

5.15.3 Products

Detailed mass and energy balances and product analyses are not available at this time, but reported char yields of around 20% wt on dry feed (60) are closer to 15% (71). Oil yields in excess of 55% wt. have been indicated (60) although it seems likely that the oil yield is nearer 65-70% wt. . At the end of the first phase of the work in summer 1993, Fenosa admitted that it had not been possible to produce pyrolysis oil of the right quality. The early operational difficulties in producing a consistent quality oil have now been mostly solved and the plant is producing a consistent relatively viscous liquid with a water content of 15 wt%.

This modelling approach allows for the formation of intermediate components and their subsequent conversion to final products, as shown in equation M.5.

biomass —> product component j M.5

where the rate of formation of component j with a yield Vj at a given time t is given by equation M.6:

koj exp (- Ep (Vj* — V) M.6

eft RT

where:

Vj* ultimate attainable yield of product j, i. e, the yield at high temperature and long residence times (77).

The constants k0j, Ej and Vj* cannot be predicted beforehand and must be

estimated from experimental data, a problem that increases as the number of reactions postulated increases. The model provides a simple scheme that can be used to predict product yields, This method has been used by Krieger et al. (77).

Some models have taken into account the competitive nature of some of the pyrolysis reactions which have been postulated to account for the variations in product yield. This is done by Bradbury et al. (44) for cellulose as shown in Figure 2.7.

Figure 2.7 Proposed Pyrolysis Model for Pure Cellulose (44)

Bradbury et al. (45) used this approach for their kinetic model as shown above. This model was based on the pyrolysis of pure cellulose. Theoretical and experimental results for weight loss agreed to within ± 5%. As a consequence this mode) has been used to account for char yields in the models of large particle pyrolysis (77, 78, 79, 80).

Koufopanos et al. (41,42) and Nunn et al. (76) proposed that the biomass pyrolysis rate could be related to the individual pyrolysis rate of the biomass components i. e.

biomass = a (cellulose) + b {hemicellulose) + c (lignin)

where:

• bracketed terms () represent the fractions of the biomass components not transformed into gases or volatiles

• a, b, c are the weight fractions of the corresponding biomass components in the virgin biomass.

The reaction scheme of the individual components then followed a similar reaction scheme as shown in Figure 8. In both cases, theoretical and experimental results for the weight loss agreed to within ± 10%. However, for the model of Koufopanos et al. (41,42) there was no indication that it could be used to predict product yields. Nunn et a!. (76) found that, in general, the calculated values fitted laboratory data within ± 7% for temperatures up to about 950-1000°C. Similar approaches have been adopted by Simmons (65), Salazar (68), Samolada (69) and Varhegyi (55).

Fast pyrolysis of biomass was “invented” in North America around 1979 with early work being performed by NREL and the University of Waterloo. A number of commercial and demonstration plants for flash pyrolysis are operating in North America at a scale of up to 1000 kg/h. Ensyn (Canada) are marketing commercial flash pyrolysis plants of up to 4 t/h throughput which are offered with a performance guarantee and a number of conditional sales have been concluded (9). Interchem were building a second generation 1360 kg/h demonstration plant in Kansas based on the NREL vortex ablative pyrolysis process (1°) but this is now understood to be shelved. There are other smaller scale activities at research and development scale which are listed in Table 4.1 and described later.

The pyrolysis liquids can be used for production of higher value commodities as chemicals rather than fuels. Most are based on physical extraction processes, of which some are commercial and proprietary (32) and some are under development and subject to patent applications and again data is not available (17, 18, 19, 20, 21,22, 30,31).

b) National Renewable Energy Laboratory, Golden, CO, USA

Phenols are one of the largest group of chemicals in flash pyrolysis liquids and chemical recovery has focused on these compounds. Polyphenols, for example, are potentially valuable chemicals for substitution in phenol-formaldehyde resins for wood processing such as in plywood manufacture (30, 31, 48). Anisole has been derived from phenol extracted from pyrolysis oil (25), in which both proprietary and in-house manufactured catalysts modified with Mn and Cd were used with up to 72% conversion of phenol.

Levoglucosan is a major intermediate in the thermal degradation process. Yields of up to 20% on a dry feed basis have been reported when the feed is pre-treated by acid washing to reduce the alkali metals which catalyse sugar decomposition (47, 48, 49). The extraction and recovery of the levoglucosan at high purity has recently been patented by BC Research (19). High yields of hydroxyacetaldehyde have similarly been reported by the University of Waterloo, some through additions of simple (unspecified) catalysts to the biomass prior to pyrolysis (47, 48, 49).

This has long been a major difficulty for researchers. The pyrolysis vapours have similar properties to cigarette smoke and capture by almost all collection devices is very inefficient. The product vapours are not true vapours but rather a mist or fume and are typically present in an inert gas at relatively low concentrations which increases cooling and condensation problems. They can be characterised as a combination of true vapours, micron sized droplets and polar molecules bonded with water vapour molecules. This contributes to the collection problem as the aerosols need to be impinged onto a surface to permit collection, even after cooling to below the dew point temperature.

Electrostatic precipitators are effective but can create problems from the polar nature of the product and arcing of the liquids as they flow, causing the electrostatic precipitator to short out. Larger scale processing usually employs some type of quenching or contact with cooled liquid product which is effective. Careful design is needed to avoid blockage from differential condensation of heavy ends. The rate of cooling appears to be important. Slow cooling leads to preferential collection of the lignin derived components which is a viscous liquid which can lead to blockage of heat exchange equipment and liquid fractionation. Very rapid cooling of the product has been suggested to be effective as occurs typically in a direct contact quench. Transfer lines from the reactor through the cyclone(s) to the liquid collection system should be maintained at > 400’C to minimise liquid deposition and collection.

At present, there are no recognised design methods and most work has been empirical and specific to the characteristics of the feedstock being processed. Commercial liquids recovery processes are usually proprietary and may be specific to individual feedstocks and reactor configurations/.