b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

Available at www.sciencedirect.com

http://www.elsevier.com/locate/biombioe

Review

Wood would burn

Jim Swithenbank a, Qun Chen a,*, Xiaohui Zhang a, Vida Shari a, Mohamed Pourkashanian b
a b

SUWIC, Department of Chemical and Biological Engineering, University of Shefeld, Mappin Street, Shefeld S1 3JD, UK School of Process, Environmental and Materials Engineering, Leeds University, Clarendon Road, University of Leeds, Leeds LS2 9JT, UK

article info
Article history: Received 23 September 2009 Received in revised form 16 December 2010 Accepted 21 December 2010 Available online 12 January 2011 Keywords: High pressure oxy-fuel combustion Carbon dioxide recovery and sequestration Condensing boiler Wood combustion Power generation Low grade heat recovery

absract
In view of the world-wide problem of energy sustainability and greenhouse gas production (carbon dioxide), it is timely to review the issues involved in generating heat and power from all fuels and especially new (to the UK) solid fuels, including high moisture fuels such as wood, SRF, oil shale, tar sands and brown coal, which will become major international fuels as oil and gas become depleted. The combustion properties of some of these materials are signicantly different from traditional coal, oil and gas fuels, however the technology proposed herein is also applicable to these conventional fuels. This paper presents some innovative combustion system options and the associated technical factors that must be considered for their implementation. For clarity of understanding, the novel concepts will be largely presented in terms of a currently developing solid fuel market; biomass wood chips. One of the most important characteristics of many solid fuels to be used in the future (including oil shale and brown coal) is their high moisture content of up to 60%. This could be removed by utilising low grade waste heat that is widely available in industry to dry the fuel and thus reduce transport costs. Burning such dried wood for power generation also increases the energy available from combustion and thus acts as a thermal transformer by upgrading the low grade heat to heat available at combustion temperatures. The alternative approach presented here is to recover the latent heat by condensing the extrinsic moisture and the water formed during combustion. For atmospheric combustion, the temperature of the condensed combustion products is below the dew point at about 55e65  C and is only suitable for recovery in an efcient district heating system. However, in order to generate power from the latent heat, the condensation temperature must be increased to the level where the heat can be used in the thermodynamic power cycle. This can be achieved by increasing the combustion pressure to above 80 bar, resulting in the recovered latent heat being available at more than 200  C. It can then be used to increase the cycle efciency by about 15% by pre-heating the boiler water and/or combustion air etc. A further advantage is that the high pressure of the combustion gases also reduces the superheater tube stress since it can balance the steam pressure. The key advantage of this high pressure ue gas is that it is above the pressure at which carbon dioxide condenses to a liquid or supercritical gas at atmospheric temperature. Thus when used with oxy-fuel combustion, the carbon dioxide ue gas from which the moisture has been condensed can be cooled to atmospheric temperature and the supercritical CO2 can

* Corresponding author. Tel.: 44 (0) 114 222 7563; fax: 44 (0) 114 222 7501. E-mail address: q.chen@shefeld.ac.uk (Q. Chen). 0961-9534/$ e see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2010.12.026

1000

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

be fed directly into the pipes leading to the sequestration site. An important consideration of these strategies is to ensure that non-condensable gases in the exhaust, including oxygen and nitrogen, do not adversely affect the condensation processes. When oxy-fuel combustion is used, the ame temperature must be moderated by a cool diluent. Recycled carbon dioxide is often proposed for this duty. However, since the latent heat is recovered, the moisture or even additional water can full this role. This latter option may be advantageous since it is more efcient to pump wood chip fuel in water into the high pressure zone rather than feed solid wood particles. Surplus water can be simply drained and the wet wood chips are a good fuel when the latent heat of the moisture in the fuel gases is recovered into the power cycle. Bearing in mind that it is much more efcient to pump a liquid to high pressure than to compress the same material as a gas, indicates that cryogenic oxygen is a suitable material to use for an efcient power station that generates energy from biomass (or other fuels such as coal etc). Finally, combustion of the hydrogen from the wateregas reaction with oxygen allows the steam temperature in the turbine to be increased to the gas-turbine engine range of 1000e1400  C and hence the biomass and/or fossil fuel cycle efciency can be well over 60%! 2010 Elsevier Ltd. All rights reserved.

1.

Introduction

Large quantities of wood are burned world-wide for cooking and heating using combustion systems varying from three stones to support a cauldron to large systems such as a 400 MW uidised bed boiler burning wood chips. Recent plans include the construction of wood-red boilers of 700 MWe output and the scale of equipment generating power using wood biomass is clearly increasing. In 2008, the global trade of wood chips was estimated to have reached 29 million tones [1]. Although no wood-red energy plants in the UK are currently dependent on imported biomass in the form of chips or wood pellets, power plants are starting to ship wood from Brazil, Canada, Scandinavia and South Korea to meet their growing demand for timber. Meanwhile, it is being planned to develop large wood-red energy plants that will use imported chips, pellets or both. If all these planned wood-red plants are built, the import quantity could rapidly rise to about 27 million tonnes per annum [1]. It has been estimated that the UK annually generates approximately 5 million tones of waste wood. In 2008, around 40% of the waste wood was recycled and reused [2]. The majority of the recycled waste wood was used to produce chipboard and breboard for building and furniture manufacture. Statistics

shows, however, this fraction was reduced by 5% in 2009 whereas the proportion for recycled wood used as biomass energy increased by 33% [2]. It is timely to review the issues involved in generating power from wood fuel since the properties of this material are signicantly different from traditional coal, oil and gas fuels. The aim of this paper is to discuss some of the options with an innovative open mind, rather than to carry out a detailed comparative analysis of the many technical factors that must be considered before actual plant construction.

2.

Wood fuel characteristics

Conventional fossil fuels consist almost exclusively of carbon, hydrogen and inert ash. On the other hand, wood is largely composed of cellulose, hemi-cellulose and lignin, materials that include about 30% oxygen within their structure. As shown in Table 1, the effect of this oxygen is to reduce the caloric value of the wood to about one half that of conventional fossil fuels. During combustion this oxygen is incorporated into the water and carbon dioxide in the combustion product gases. Furthermore, raw wood can contain more than 50% moisture; hence the combustion products generally contain a much

Table 1 e Properties of coal, wood and RDF samples [3,4]. Coal

Proximate analysis (%, dry) Fixed Carbon Volatiles Ash C H O N S 55.8 33.9 10.3 84.2 5.6 5.5 1.3 3.5 31.75

White Oak
17.2 81.3 1.5 50.2 5.5 43.8 0.4

Pine bark
33.9 54.7 0.4 56.5 5.6 37.9

RDF
9.8 67.6 18.9 61.2 8.2 26.6 1.3 0.2 22.30

Ultimate analysis (%, dry ash free )

HHV (MJ/kg)

19.42

21.78

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

1001

higher fraction of moisture than would be obtained from conventional fuels. The effect of the moisture on wood fuel can be very dramatic. First of all, when raw wood fuel is transported, whether by road, rail or sea, it is apparent that more than 50% of the cargo is often water. This obviously has an impact on the cost, fuel efciency and emissions of transportation. Analysis of the potential production of wood in the UK shows that it would not be feasible to produce enough wood fuel to satisfy our total power needs. The import of sustainable wood supplies by sea from Africa or South America etc. will probably be carried out in huge wood tankers whose fuel consumption to deliver the load is a tiny fraction of the energy of the cargo [5]. The impact of the water on combustion is also dramatic. This can be seen from its effect on the caloric value (CV) of the fuel since every 10% increase in the moisture content reduces the CV of the fuel by about 2 MJ/kg. Thus the lower CV of the raw fuel is about 8 MJ/kg whereas that of dry wood fuel is about 18 MJ/kg. It is not necessary to dry wood completely to obtain an advantage, as can be seen from the following argument. Simply reducing the moisture from 50% to 30% would increase the CV from 8 MJ/kg to 12 MJ/kg. This represents a 50% increase in caloric value for a modest amount of drying. Clearly, determination of the optimum level would require a detailed economic analysis for any specic case. Dry wood fuel is thus a value added product and drying could provide an energy efcient means of using the huge amount of low grade heat that is currently discharged to the atmosphere by industry. Not only could this result in reducing transport costs, but also it is thermodynamically remarkable since although the low grade heat is unsuitable for generating power, the drier wood regenerates the energy at ame temperatures, when it can be used to generate power. Thus drying can upgrade heat and this process can be regarded as part of a thermal transformer. The option of using solid recovered fuel (SRF) as an alternative to wood could be considered since it consists of more than 50% biomass [6], and huge quantities are now becoming available from material recycling facilities (MRF) in Europe. Its morphology depends on the process used for production and may vary from damp ock to rigid cylindrical pellets. Many of the points discussed below are applicable to such fuels. The recovery of the latent heat due to the high moisture level in oil shale and brown coal is similarly important, as also is the recovery of latent heat from the ue gas of natural gas due to the large proportion of hydrogen in methane (CH4 3O2 0 CO2 2H2O).

temperature is about 55e65  C and it follows that the temperature of the water returned from a central heating system (or from the hot water heating system), must be about 30  C, i.e. well below this temperature. This requires an underoor heating system of a high surface area of the radiators in the building. Similar considerations apply to district heating schemes if the latent heat of the moisture is to be recovered from the ue gases. Most district heating schemes in the UK presently use delivery and return water temperatures of about 120  C and 70  C respectively, although some Scandinavian district heating schemes do utilise the latent heat [7]. However, if wood is to be used for a combined heat and power (CHP) or district heating scheme, then there will be a signicantly larger amount of latent heat that could be recovered by condensing the ue gases.

4.

High pressure combustion systems

3.

Condensing boilers

The means of recovering energy from moisture in the ue gases is to use a condensing boiler. These are used widely in the UK on natural gas red domestic water heating systems but they are rarely used in industry. They recover the latent heat of the moisture in addition to the sensible heat and hence exploit the higher CV, thus increasing the efciency of the boiler by about 15%. The condensation of moisture from the ue gases requires that the gases be cooled below their dew point. In the case of natural gas combustion products at atmospheric pressure, this

Bearing in mind the fact that energy in the form of electricity is more highly valued than low grade heat, it is worth considering whether recovery of the latent heat of the ue gas moisture from wood ring is practicable in power generation systems. Thermodynamic considerations show that heat available at less that 50  C is not signicantly useful for power generation, either within the steam cycle or for pre-heating boiler water or combustion air. However, if the combustion takes place at high pressure, Fig. 1 shows that the dew point increases to more than 200  C and the latent heat is then available to increase the efciency of the electrical power production by about 10%. A coal red system utilising such a system has been investigated analytically by CANMET [8]. For raw wood, oil shale and brown coal with 60% moisture content burning with oxygen, the moisture content of the ue gases is about 20e35%. The high pressure combustion system is well matched to fuels with such high moisture content since the latent heat is recovered at a useful temperature. Furthermore, the ow of moisture through the combustion system serves to moderate the combustion temperature and thus reduces the need for carbon dioxide ue gas recycle that is required with coal red oxy-fuel systems. It must also be noted that the recycled gases are signicantly reactive compared to the nitrogen present with an air-fuel combustion system. The high ue gas pressure also reduces the superheater tube stress when it is matched to the steam pressure. If it exceeds the steam pressure then the inversion of the superheater tube load from tension to compression could allow novel materials such as silicon carbide/nitride to be used. The thermal conductivity of these refractory materials is four times higher than stainless steel and their temperature capability is much higher. This potential extension to the proposed system would allow steam temperatures to be increased resulting in a substantial enhancement in power generation efciency.

5.

Heat recovery by the condensation of water

As mentioned above, the condensation of water from the high pressure ue gases recovers heat at more than 200  C. This is

1002

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

250

200

Flue gas dew point (C)

150 35 % H2O 30% H2O 25% H2O 20% H2O 15% H2O 10% H2O 5% H2O 0 10 20 30 40 50 60 70 80 90 100

100

50

Pressure (bar)
Fig. 1 e Flue gas dew point temperature as a function of pressure and moisture content.

Pressure (bar)

at a sufciently high temperature that it can be used within the power generation system and thus increase the plant efciency. The question of where it should be returned to the power cycle has been considered in the case of coal ring [9]. Possible locations are: heating the oxygen used for combustion, or heating the boiler feed water. It was found that the latter was more benecial in the case of coal ring. Although Fig. 1 shows the ue gas temperature as a function of pressure, this gure assumes that the gas consists of pure carbon dioxide and water vapour. In practice, other contaminants will be present. These are due to contaminants in the oxygen, particularly nitrogen, and contaminants in the fuel, thus leading to possible ue gas contaminants including sulphur dioxide, HCl and oxides of nitrogen. The actual ue gas condensation temperature must therefore take the effect of these contaminants into account. Fortunately, ue gas scrubbing technology is well developed and scrubbing should precede the condensation. Also, as with domestic condensing boilers, the heat transfer tubes in the condensation section must be made of plastic uorocarbon to avoid corrosion problems. If required, very low levels of moisture in the sequestered carbon dioxide can be achieved by polishing it with silica gel. The water remaining in the sequestered carbon dioxide is an important parameter from its potential contribution to pipe corrosion problems, especially if sulphur oxides are present. However wood is relatively free from sulphur and this problem will be less severe with biomass combustion. If the carbon dioxide is to be marketed at low pressure, then even if it is saturated with water at high pressure, expansion to low pressure will cause its humidity to drop remarkably.

this curve applies for pure gas and for a real system and account must be taken of the actual composition of the mixture. This issue is discussed in the literature [9] where it is shown that gas impurities can have a profound effect of the phase equilibrium of the carbon dioxide rich stream. One main conclusion from this observation is that high quality oxygen (0.5% N2) should be used for the combustion process. Also, products of incomplete combustion such as CO and contaminants such as sulphur compounds must be removed at an appropriate point. The desire to operate with a high ue gas condensation temperature and an output of supercritical liquid carbon dioxide both require high pressure combustion at above 80 bar. Detailed design studies are needed to optimise this pressure for the fuel combustion; however, 100 bar can be regarded as an approximate value at this stage of the discussion.

7.

Fuel feeding system

The technique used to feed a solid fuel into a high pressure environment poses some interesting problems. If a lock hopper system is used, then the chamber between the two valves will ll with gas as the solids move into the high pressure environment.

80 70 60 50 40 30 20 10

Critical Point Liquid CO2 Gas CO2 Triple Point

-80 -60 -40 -20
o

Solid

6.

Carbon dioxide recovery

The main advantage of the pressurised combustion system is that carbon dioxide can be recovered for transmission and sequestration. For pure carbon dioxide, the pressure temperature relation for its liquefaction is shown in Fig. 2, where the liquid region lies above the curve. Again, it must be noted that

0 -100

20

40

Temperature ( C)
Fig. 2 e Phase diagram for pure carbon dioxide.

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

1003

The hopper full of gas will then release to the low pressure region and must be discharged, or it must be pumped back up to the high pressure level before being returned. This loss can be minimised by using a displacement piston. Compressing solid fuels to high pressure requires very little energy since they are largely incompressible but the space between particles represents a problem. This problem is usually solved with coal red gasiers by using slurry made from the fuel in water. For a combustor operating at high pressure with latent heat recovery, the energy required to evaporate the water added to make wood fuel slurry will be largely recovered in the condenser. The effect of the amount of water that must be added the wood to make pumpable material and the size of the wood particles will obviously have an effect on the combustion as discussed below. Slurries that contain particles can interfere with the operation of valves in the pump. A valveless system can be designed whereby the particles are fed through the pressure gradient produced in water spinning in a zero liquid ow centrifugal turbine conguration. Such a system would allow the particles to be transferred to the high pressure system with a smaller proportion of water. Nevertheless, the optimum proportion of moisture in the high pressure combustor still remains to be determined. The problem of feeding solid recovered fuel (SRF) as an alternative to wood depends on the physical nature of the material. Flock can be fed as a water slurry as above, whilst the energy to feed semi-solid SRF may be minimised by extruding a cylindrical worm of material directly into the high pressure environment.

8. High pressure combustion of solid fuel (wood)

There are several potential combustor congurations that can be considered for the high pressure combustion of wood. These

are listed in Table 2, together with their advantages and disadvantages. Bearing in mind the requirement for high pressure means that the whole combustion equipment must be enclosed in a pressure vessel, it is apparent that the chosen system should be simple to maintain and operate. This suggests that a moving grate or rotary kiln would be unsuitable. A shaft combustor could be used for small scale but would not be suitable for damp wood chips. Entrained wet sawdust would also be generally undesirable. A circulating uidised bed is quite exible but tends to be subject to high erosion. The bubbling uidised bed is regularly used for slurries and for wood chips and at rst sight would be the preferred technology for the proposed application. The temperature of the bed must be controlled by in-bed heat loading, fuel-oxygen ratio, ue gas recycle and/or water. A typical bed temperature would be in the range of 800e900  C. The use of added water to limit the combustion temperature is interesting since it could partially or completely replace the recycle of the carbon dioxide ue gas, but the total ow through the heating section of the boiler would be similar. The overall heat balance for the cycle is almost unchanged since the heat required to evaporate the water would be recovered in the condenser. However, the combustion chemistry would be signicantly changed as the ue gases would contain a greater proportion of hydrogen. If the oxygen were reduced below stoichiometric, the heat output would decrease, and obviously the system would approach the role of a gasier. Heat transfer in the high pressure system differs signicantly from that in conventional boilers since both radiation and convection contributions will increase with pressure. The radiation increases modestly due to the greater concentration of the emitting species which are largely water and carbon dioxide. Convection heat transfer increases more rapidly with pressure than radiation and account must be taken of this in the design of the heat transfer surfaces in the boiler. The

Table 2 e Potential combustor congurations for high pressure wood combustion [10,11]. Advantages
Bubbling uidised bed No moving parts in the chamber; High exibility for moisture content and fuel kind No moving parts in the chamber; High exibility for moisture content and kind of biomass fuel; Homogeneous combustion conditions in the furnace; High specic heat transfer capacity High NOx reduction by efcient air staging; Good load control and fast alteration of load possible Low investment costs for small-scale plant; Low operating cost; Low dust load in the ue gas No moving parts in the hot zone; Relatively low investment and operating costs; Most versatile and high fuel exibility

Disadvantages
High investment and operating costs; Possible bed agglomeration for high alkali biomass fuels; High dust load in the ue gas High investment and operating costs; High alkali biomass fuels; High dust load in the ue gas; High sensitivity for ash slagging

Circulating uidised bed

Entrained pulverised combustion Moving grate

Limited particle size; Extra start-up burner is necessary Usually no mixing of fuel; High excess oxygen decreases efciency; Less homogeneous combustion conditions Limited mixing of fuel in the chamber; Fairly high particulate and hydrocarbon concentrations in the ue gas at the kiln exit

Rotary kiln

1004

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

Mass concentration (mg/m3)

result is that smaller heat transfer surface area will be required in regions such as the superheater. Bearing in mind the fact that the system could be designed so that the steam and ue gas pressures are almost equal means that there is little, or no, stress on the heat exchanger elements. Furthermore, the high densities of the uids results in liquid like heat transfer coefcients. Thus a very compact heat exchanger such as a plate design unit could be used. In this case the volume of a 100 MW heat exchanger would be about the size of a large ofce desk!

30 25 20 15 10 5 0 Wood firing plant (literature) Retrofitted Boiler (measured)

Fig. 3 e Mass concentration of particles from a wood boiler and retrotted coal boiler.

9.

Oxygen production

The improvement of about 15% in the power yield from the proposed high pressure combustion system comes at the expense of high pressure oxygen input. Bear in mind the fact that pumping gases to high pressure requires an enormous power input, whilst pumping the same ow as an incompressible liquid requires very little power, implies that the oxygen is required as a liquid. Technology for the cryogenic production of oxygen is well established, and is appropriate for this application. However, cryogenic air separation is an energy-intensive process. The specic energy for producing O2 with purity of 95e99.5 vol% is about 300 kW h/tonne. If cryogenic production of oxygen can be achieved as a by-product of process industries, the costs could be lower. The above discussion shows that the quality of the oxygen is also important to the process and the economic choice of oxygen containing a few percent of nitrogen, or a fraction of one percent of nitrogen will be a key factor in the optimisation of a high pressure, wood red, power generation system.

atmospheric equivalents, but the need to contain the very high pressure will undoubtedly pose special problems. It is also expected that the condensation of water stage in the proposed system will be effective in removing sub-micron y ash particles. It is particularly noteworthy that there will be little (or no) atmospheric emissions from the plant and the power industry will no longer require chimneys! Due to the high density of the sequestered carbon dioxide, the exhaust pipe from a 500 MW plant will be about 300 mm diameter.

11. 10. Emissions

Issues of scale

With wood combustion, the amount of bottom ash is likely to be small, and there is considerable experience in ash removal when operating a uidised bed system at atmospheric pressure. Reversing the feed system discussed above for fuel implies that the ash should be removed from the high pressure zone suspended in water. The main gaseous contaminants of the ue gases are likely to be O2, N2 CO, NOx, SOx, HCl, y ash, organics and other minor species. The oxygen arises from the need to burn out the fuel, whilst the nitrogen is the main impurity in the feed oxygen. The emission of ne particulates from atmospheric pressure wood combustion has given particular concern [12] and data from the literature [13] together with data from our measurements on a boiler converted from coal ring are shown in Fig. 3. Other contaminants will be at the ppm level. Flue gas emissions from the combustion of contaminated wood and C&D (construction and demolition) wood are currently regulated by Waste Incineration Directive (WID). The WID is a European law aiming to control emissions from the incineration and co-incineration of waste. However, it is anticipated that most gas cleaning equipment such as scrubbers, activated charcoal and bag lters will be tted in the high pressure system as discussed below. Due to the high density of the gases, these units will be signicantly smaller than their

The proposed pressure in the plant is about half that in a conventional high pressure gas cylinder and this clearly leads to critical features of the plant design. It is likely that parts of the equipment with a thermal load, such as the uidised bed, will be placed within a separate pressure vessel. It follows that the pressure vessel construction material must be chosen for its strength rather than its thermal properties and it will be lined with insulation. The economics of most process plants follow the rule that their cost increases at the 0.6 power of their scale whereas the throughput increases linearly with scale. This leads to the well known economy of scale, and the result is often a very large power plant. However, high pressure systems require a large amount of metal and many parts such as valves need an amount of material that scales as the cube of a linear dimension, whilst the valve port area only increases as the square. This property can result in the specic material cost per unit of throughput increasing with scale. The overall result is likely to be that there is an optimum size for the nal power plant. It should be noted, however, costs and market issues for such power plants rely heavily on system safety, experience with the technologies, as well as policy constrains. These are well beyond the discussion scope of this paper. The detailed economic characteristics of the system will be evaluated at a later stage of the research programme.

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

1005

12.

Ultra-high efciency prospects

The potential to increase steam temperature and hence cycle efciency [14] by using new superheater materials that exploit the inverted pressure load on the heat transfer tubes was presented above. Nevertheless, the temperatures are then still well below the established technology temperatures used with cooled turbine blades on aircraft gas turbines. This gives us scope to dramatically increase the efciency as follows: The intrinsic and extrinsic water present in the combustion chamber will lead to the production of substantial quantities of hydrogen, principally through the wateregas reaction. A number of studies are currently investigating the separation of this H2 fuel from the carbon dioxide by various types of membranes that operate at high temperatures. For several years, we have been producing ultra-superheated steam at temperatures of 1500  Ce2000  C by burning hydrogen rich fuel in a steam-oxygen mixture. Thus, utilising this established technique to burn the hydrogen with some oxygen allows the steam temperature in the turbine to be increased to the established gas-turbine engine range of 1000e1400  C (Figs. 4 and 5) and hence the biomass and/or coal cycle efciency for electrical power generation can be increased to well over 60%! The fact that cold nitrogen and/or oxygen from the cryogenic air separator is used to cool the condenser following the steam turbine also contributes to this high efciency. Nevertheless, a staged development of the power cycle is envisaged. Phase 1 would be the lower performance system shown in Fig. 6. Here the uid bed is operated slightly substoichiometrically to yield a gas containing a little CO and hydrogen. The uid bed is followed by gas cleaning with a cyclone/ceramic lter to remove dust from the uidised bed. The CO, H2 and H/C gases are then burned to raise the temperature and destroy any heavy hydrocarbons in the form of vapourised tar before the rst heat exchanger. The main attraction of this Phase 1 design, which recovers latent heat, is that it delivers CO2 sequestration without incurring the huge energy penalties for atmospheric combustion associated with high CO2 compression and amine scrubber uid recovery. These penalties typically amount to about 30% power loss.

Fig. 5 e Experimental high temperature steam concept.

Fig. 4 e An experimental high pressure, high temperature (up to 1400  C) SiC tube heat exchanger.

Details of the Phase 1 process would take into account the actual fuel being used. To develop the heat exchanger and turbine, natural gas would be an appropriate clean fuel. In this case the Rankine cycle top/bottom temperature difference could exceed that of a combined (GTCC) cycle with the advantage that the energy input from the fuel corresponds to its higher heating value, and high pressure CO2 ready for sequestration is produced. The trade-off between oxygen use and atmospheric CO2 removal is expected to lie in favour of the former. Changing the fuel to biomass, brown coal or oil shale reveals the large advantage of the high pressure combustion system since there then a large benet from the latent heat recovery that allows the higher heating value of the (intrinsic and extrinsic moisture containing) fuel to be exploited. In order to achieve ultra high efciency, the Phase 3 scheme shown in Fig. 7 is required. This scheme achieves very high efciency by using a higher turbine entry temperature, and obviously requires the adaptation of cooled blade technology, already used in gas turbines, to the steam turbine. Again, the details of the actual plant will vary depending on the fuel to be used. For example, wood fuels have low ash, low sulphur and low chlorine, but generally have high alkali content. The alkali can give corrosion problems with conventional superheater tube materials. To mitigate this problem, the rst step is to remove the solid y ash immediately after the uidised bed, before the temperatures are raised above the ash fusion point. The second step is to take account of the low mechanical stress on the superheater heat transfer material and use corrosion resistant refractory materials such as silicon carbide and coatings. Many other fuels contain sulphur and/or chlorine. There is considerable experience with these pollutants and the location of their removal in a phase 3 system depends on the scrubbing technology used. Most existing scrubbers use reagents containing calcium carbonate, lime or sodium bicarbonate, often incorporating activated carbon to remove hydrocarbons and heavy metals. An interesting new technique employs liquid tin as the reagent [15]. This system operates at temperatures above 250  C and removes sulphur as tin sulphide dissolved in the tin. The sulphur is then removed from the tin by reaction with zinc to precipitate zinc sulphide.

1006

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

Fig. 6 e Phase 1 stage of development of the Hi-h power generation system.

Fig. 7 e Phase 3, nal stage of the Hi-h power generation system with CO2 sequestration for fuels varying from coal, oil, gas, biomass and SRF, to the oil shale and brown coal fuels that will be used increasingly as gas and oil are depleted.

b i o m a s s a n d b i o e n e r g y 3 5 ( 2 0 1 1 ) 9 9 9 e1 0 0 7

1007

13.

Conclusions
[2]

 Decarbonise the atmosphere with advanced biomass and fossil fuel burning technology.  There is a high level of intrinsic and extrinsic moisture in biomass, SRF, shale oil and brown coal combustion products. Either dry the fuels using low grade heat to recover the extrinsic moisture energy, or condense the moisture in a high pressure combustion system to recover the latent heat of water in the ue gas and hence exploit the higher caloric value of the fuel.  Burning and condensing at high pressure not only recovers the latent heat at >200  C for energy generation but also allows condensation of CO2 for direct delivery of supercritical CO2 to the sequestration site.  Pump liquids or fuels in water (e.g. wood chips as slurry) to compress efciently to high pressure, and then drain excess water to deliver a damp fuel.  Inverted heat transfer tube stress (compression) or balanced steam/ue gas pressures allows use advanced materials and heat exchanger designs.  Oxy-fuel Combustion: Control ame temperature with water to form H2 by the wateregas reaction or utilise recycled CO2.  Pump cryogenic oxygen as liquid to high pressure.  Use cold nitrogen and/or oxygen from the air separation unit to cool the steam condenser to w2  C to maximise cycle temperature difference and thus increase efciency.  Raise power generation efciency for fossil fuels (coal, oil, gas, brown coal, oil shale) and SRF/biomass to >60% by oxyhydrogen superheating steam to w1400  C at the turbine entry.

[3] [4]

[5] [6]

[7]

[8]

[9]

[10] [11] [12]

[13]

references

[14]

[1] ConFor (Confederation of Forest Industries). Wood bre availability and demand in Britain 2007 to 2025, http://www.

[15]

confor.org.uk/Upload/Documents/37_WoodFibreAvailability DemandReportnal.pdf [accessed November 2010]. Wood Recyclers Association. Waste wood to markets statistics 2009, http://www.woodrecyclers.org/documents/ stats2009.pdf [accessed November 2010]. Gaur S, Reed T. Thermal data for natural and synthetic fuels. New York: Marcel Dekker; 1998. Hernandez-Atonal FD, Ryu C, Shari VN, Swithenbank J. Combustion of refuse-derived fuel in a uidised bed. Chem Eng Sci 2007;62:627e35. Drax Ltd, Private Communication. European Committee for Standardization. CEN/TR 14980: 2004 e solid recovered fuels: Report on relative difference between biodegradable and biogenic fractions of SRF; 2005. Paappanen T, Leinonen A. Fuel peat industry in EU, country reports e Finland, Ireland, Sweden, Estonia, Latvia, Lithuania; 2005. Canmet energy, carbon capture and storageehigh pressure oxy-fuel (HiPrOx), http://canmetenergy-canmetenergie. nrcan-rncan.gc.ca/eng/clean_fossil_fuels/carbon_capture_ storage/hiprox.html [accessed 19.01.09]. Zheng L, Pomalis R, Clements B. Technical and economic feasibility study of a pressurised oxy-fuel approach to carbon capture, http://www.thermoenergy.com/userles/File/TIPS_ Feasibility_Study_Mar_2007.pdf [accessed 19.01.09]. van Loo S, Koppejan J. The handbook of biomass combustion and co-ring. London: Earthscan; 2008. Niessen WR. Combustion and incineration processes. New York: Marcel Dekker; 2002. AEAa. Technical guidance: Screening assessment for biomass boiler. Report to the Department of Environment, Food and Rural Affairs; 2008. ED48673005/R2655. Ehrlich C, Noll G, Kalkoff WD, Baumbach G, Dreiseidler A. PM10, PM2.5 and PM1.0 e emissions from industrial plants e results from measurement programmes in Germany. Atmos Environ 2007;41:6236e54. Swithenbank J, Boysan F, Liu F. Keynote lecture (Shefeld SUWIC HIC 534). In: 10th intl heat transfer conference, Brighton; August 1994. Ryu C, Ismail MHS, Sharif VN, Swithenbank J. Ind Eng Chem Res 2007;46:9015e21.