---

Department of power generation

New concept of CO2 removal technologies in power generation, combined with fossil fuel recovery and long term CO2 sequestration

Ph. Mathieu, R. Dubuisson, S. Houyou, R. Nihart

University of Liège

Department of Mechanical Engineering

Rue E. Solvay, 21; 4000 Liège; Belgium

Phone: +32-4-366 92 68; Fax: +32-4-252 54 39; e-mail:pmathieu@ulg.ac.be

http://www.ulg.ac.be/genienuc/

1. ABSTRACT

The so-called MATIANT cycle, presented in the previous ASME conference, belongs to the family of CO2/O2 cycles using CO2 as theworking fluid and O2 as the fuel oxidiser. The CO2 generated in the combustion process is not released in the atmosphere and is removed through a simple purge, in liquid or supercritical state at highpressure (up to 300 bar). The cycle efficiency ranges from 40 to 50%.

In this paper, the mentioned cycle has been transformed in a CO2 regenerative Ericsson-like cycle and therefore is named E-MATIANT. The removed CO2 can still be available at a pressure higher than the critical one (73 bar). When optimising the cycle, the calculated optimum pressure will be around 60 bar; this makes the technical issues easier to deal with than when using a supercritical fluid, namely the material strains and corrosion behaviour. Asensitivity analysis is performed with respect to the CO2 delivery pressure in order to evaluate the performance changes. The fuel flexibility is an important asset of the newly designed cycle: mixtures of CO and H2 produced either in gasification or steam reforming processes can indeed be burnt in the combustion chamber. In a future work, the combination of a solid oxide fuel cell (SOFC) and this cycle both fed by a CO and H2 mixture will be considered as an option for the improvement of the global efficiency.

If not fixed in a chemical or biological system, the delivered CO2 can be used in industry or in the enhancement of fossil fuels recovery from their deposits, with a marginal compression consumption work. In this paper, CO2 injection is used to enhance methane recovery from coal seams by some 20 to 30%, in comparison with waterpumping. The depleted seam can afterwards be used as the host site for long run CO2 sequestration.

As a conclusion, the combination of quasi-zero emission powerplants with CO2 geological storage and enhanced fuel recoveryprovides a CO2 flow, otherwise considered as a waste or a by-product, with an exergetic and a commercial added value. This makes this option a serious alternative to other CO2 control technologies.

2. INTRODUCTION

According to the Kyoto protocol, the 15 countries of Europe have committed themselves to reduce their CO2 emissions by an average of 5.2% with respect to the 1990 level by 2012. On its side, Belgium has to reduce its emissions by 7.5% by 2012.

In order to achieve that target, several techniques aiming at CO2 emissions reduction have to be implemented:

- increase of the efficiency of energy and fuel use in powerplants. Promotion of cogeneration and co-firing of fossil fuels and biomass;

- more intensive use of low C/H content fuels (shift from coal tonatural gas);

- increase of the share of CO2 neutral technologies : renewables(windmills, photovoltaic, hydrostation, biomass) or nuclearenergy.

However should the target of the coming summits be more and more lowered and the cut of CO2 emissions be more and more larger,then all these measures, even when applied altogether and at their maximum potential complying with the public acceptance, could not be able to meet the requirements. Then the removal of CO2 from fluegases becomes unavoidable and if the CO2 emissions reduction has still to be decreased, new types of power plants based on new concepts could come on the stage and be alternatives or complementthe other options available in the fuel basket of a country.

With quasi-zero emission power plants, a country will be able to go on burning fossil fuels, avoiding social and economic disruptions, while having a total control on the CO2 releases in the atmosphere. In addition, it can sell "CO2 Emission Permit" to more polluting countries.

In this paper, we will present a combination of the E-MATIANT cycle with a geological CO2 storage option. The main asset of the E-MATIANT cycle is the removal of nearly 100% of CO2 coming from the combustion, almost pure and at a high pressure. This gives to the system the possibility to be combined with enhanced oil recovery from oil wells or with methane recovery from coal beds (see Fig. 1). After its reuse the CO2 can then be sequestrated in stable geological sites for the long run.

The aim of the research is to achieve a cycle efficiency of atleast 45% on natural gas. A CO2 compressor from 1 bar to 40 bar or higher in liquid and supercritical state and cooled expanders at high temperature are required. A prototype needs to be built, without cooled expanders, to demonstrate the technical feasibility of the total control of removed CO2 flow [Mathieu et al (1998), Mathieu and Nihart (1999)].

Issues linked to CO2 chemical behaviour especially in supercritical state (corrosion, dissolution in other fluids, reactions with materials and behaviour as a solvent) are to betackled and introduced in the modelling.

3. PERFORMANCES OF THE E-MATIANT CYCLE

The main distinctive features of this cycle are:

1) CO2 is used as the working fluid and replaces the nitrogenof air, notably as the thermal ballast for the control of the flame temperature. The excess CO2 generated in the combustion process is thus extracted out of CO2 itself through a simple valve. This avoidsthe need for CO2 removal from the flue gases (scrubbers, membranes, cryogenic systems) and the associated high penalties on efficiency (typically 10% pts), on capital and O&M costs.

2) O2 is used as the fuel oxidiser in a CO2 atmosphere; this requires hence an air separation unit, and its electricity consumption as well as its cost have to be taken into account in the global balance.

3) The removed excess CO2 is in liquid or supercritical state at a high pressure, ready to be reused and/or sequestred.

4) Water extraction from the staged compressor is modelled.

5) Efficiency penalty due to the cooling of the blades is taken into account (Mathieu at al, 1999)

Fig. 1 Synergy between CO2 Power Plant and Methane Recovery from Coal Seams

Fig. 2 The E-Matiant Cycle on the T-s Diagram

Roughly speaking, Fig. 2 shows that it is a CO2 Ericsson-likecycle with 2 quasi- isothermal processes (the intercooled compression and the staged expansion with reheat) and 2 isobaric processes (in the heat exchangers and in the combustion chamber). The cycle operates in the largest possible range of temperatures compatible with the materials and cooling techniques currently available, namely 30 to 40°C at the lower part where the intercoolers are and1300°C on the upper part, corresponding to the present turbine inlet temperature (TIT) of the most advanced gas turbines. The cycle contains a 4 stages compressor designed such that the compression does not penetrate into the CO2 saturation line. This is one of theassets since this allows avoiding the implementation of a condenserand consequently a leakage of CO2 when it is vented for thenon-condensable gases extraction.

The cycle is regenerative and the sensible heat of the exhaustgases may be recovered either internally, then we have the CO2 gas cycle shown on Fig. 2 or externally and then the design is a combinedcycle, but this option is not investigated here.

This requires a CO2 turbine, which is not currently available. The E-MATIANT cycle has so far only been modelled for natural gasfuel but could be integrated to an O2-blown coal gasifier with a cleanup system or to a methane steam reformer in order to use synthesis gas in future.

The E-MATIANT cycle is also particularly well adapted to fuelslike a mixture of H2 and CO produced in a gasifier, or even to amixture of H2 and CO2 when in addition a shift reaction is used to convert CO in CO2. Here the separation of CO2 and H2 may be avoided and the fuel is now H2 with an extra ballast of CO2. The combustion product with O2 is simply water, which is extracted from the cyclefor its major part and the extra CO2, which is removed at the desired pressure for an application or for its sequestration.

This cycle is also well adapted to carbon-rich fuels. This is again an asset, especially for the coal users, since there commendation of moving from coal to natural gas is no longervalid.

Fig. 3 shows the performance diagrams (efficiency versus specific work) using the operating parameters mentioned in table 1 for the E-MATIANT cycle without reheat and that with and without taking the electricity consumption of the ASU into account. In bothcases the cycle efficiency shows a maximum with respect to the uppercycle pressure, a flat one at some 90-100 bar without ASU and as harper one at about 65 bar with the ASU. The ASU consumption is observed to be very penalising since the optimal efficiency is around 60% without and 47% with the ASU. This means an efficiency penalty of some 13%pts, thus significantly higher than the penalty brought about in a CO2 MEA scrubber installed in the flue gas.

Upper cycle pressure (P1) 40-110 bar Pinch-point at therecuperator outlet 20°C

Lower cycle pressure (P2) 1 bar Maximum inlet temperature in therecuperator 700°C

Pressure drop in the combustion chamber 3 %psu Expander inlettemperature 1300°C

Isentropic effectiveness of the 3 expanders 0,87 Cycle lowtemperature 30°C

Isentropic effectiveness of the O2 compressors 0,75 Isentropiceffectiveness of the fuel compressor 0,75

Isentropic effectiveness of the intercooled CO2 compressor : 0,85for the 3 first stages - 0,8 for the last one

Table 1 Operating Parameters of E-MATIANT Cycle

Fig. 3 Performances diagrams (efficiency versus specific work) of the E-MATIANT cycle with and without the air separation unit consumption, the upper cycle pressure (P1) being the parameter.

The occurrence of a maximum can be explained as follows. For pressures below the optimum (95 and 65 bar respectively), the expansion work of course decreases as the pressure ratio is smaller and since the temperature of the hot gases is limited at 700°Cat the regenerator inlet (680°C for cold gases as the pinch is20°C,see table 1)), a small amount of heat at high temperature is lost since the exhaust temperature of the gases is higher than 700°C. For pressures beyond the optimum, the expansion workincreases but the carbon dioxide being less preheated before entering the combustion chamber, more fuel is needed so that more oxygen is needed. The increase of both the oxygen mass flow rate and its pressure gives rise to an increase of the oxygen compressor consumption. However, this consumption is surpassed by the increase of the expansion work, but due to the increase of the fuel consumption the cycle efficiency decreases.

Fig. 4 shows that the efficiency penalty, caused by the increase of the ASU consumption when it delivers oxygen at a higher pressure and mass flow, increases with the pressure ratio. It ranges from about 11,5 %pts at 40 bar up to about 14 %pts at 110 bar.

Fig. 4 Efficiency penalty of the E-MATIANT cycle due to the air separation unit consumption

In (Mathieu et al., 1999), the fraction of carbon dioxide leaving the cycle dissolved in the extracted water and the fraction of water flowing out of the separator in the main CO2 flow and towards the compressor are calculated. The model takes into account the two-phase CO2/H2O equilibrium. The extraction of water isperformed both before the first compressor and at the outlets of the intercoolers. This location of the extraction points is favourable:indeed the dew point is higher because of the higher operating pressure of the separators.

When there is no reheat in the cycle, 56% of the water generated in the combustion chamber is extracted before the first compressor, 41% at the outlets of the intercoolers and the remaining 3% go on flowing with CO2 in the gaseous phase. The carbon dioxide balance shows that 442 g CO2/kWhe are extracted from the cycle at apurity of 99.25% weight and 6 g CO2/kWhe are lost in the extracted liquid phase; this represents 1.5% of the removed carbon dioxideflow.

4. INFLUENCE OF THE REHEAT

The impact of a reheat on the performances of the E-MATIANT cycle for the optimal pressure ratio of 60 are shown on fig 5, without and with an ASU. The 2 curves show that both the efficiency and the specific work pass through a maximum with respect to the reheat pressure. Typically the reheat increases the specific power of the cycle, but we will show that it does not necessarily improve the efficiency.

Fig. 5 shows that the efficiency decreases when the reheat pressure decreases (here from 54 down to 6 bar). Indeed, for low reheat pressures and for a given reheat temperature (1300°C), the temperature at the outlet of the second expander is quite high. As the temperature of the hot gases at the inlet of the recuperatoris limited at 700°C for technical feasibility and considering a preheating temperature of the CO2 flow at 680°C (the pinch is indeed 20 °C), this means that a significant part of the heat at high temperature is lost at the outlet of the second expander, and that is why the efficiency decreases when the reheat pressure goes down.

When there is a reheat, the specific work increases with the flow in the second expander, but this does not offset the increase of fuel consumption. Hence, the efficiency is lower when using areheat.

It is also observed that the efficiency penalty due to the air separation unit consumption ranges from about 10.5 %pts at 6 bar upto 12.5%pts at 54 bar. The situation here is quite different from the case without reheat; here the highest penalty occurs when the ASU has the lowest consumption but then its relative value to the expansion work is the biggest.

Fig. 5 Performances diagrams of the E-MATIANT cycle (withreheat) with and without the air separation unit consumption, there heat pressure (P3) being the parameter (P1 = 60bar).

5. ENHANCED RECOVERY OF COAL BED METHANE

Once the CO2 is available under pressure, there are two main options to be considered : CO2 storage or CO2 utilisation. Carbon dioxide can be stored by injection in the ocean or underground, for instance in depleted oil and gas reservoirs or in deep saline aquifers. The retention time is expected to be sufficiently long tomitigate the greenhouse effect. The storage potential of these options is tremendous, particularly in the oceans (IEA Report, 1994). However, in this case, attention must be paid to the fact that the CO2 is expected to be released again to atmosphere after a fewhundred years.

Even if CO2 storage shows high promises of mitigating the greenhouse effect, due to the high amounts involved and to the long retention times, it offers no possibility to generate any economic added value with the delivered CO2. This reason explains why it is also important to consider CO2 utilisation, though only a few quantity of recovered CO2 can be dealt with utilisation.

Carbon dioxide can be used in the chemical industry as feedstock to generate some molecular compounds (plastic, carbonates,…), or to be converted into an alternative fuel such as methanol, by the following equations :

CO2 + 3 H2 ® CH3OH + H2O

CO2 + 3 CH4 + 2 H2O ® 4 CH3OH

These equations are not very interesting strictly from energy utilisation point of view, because the production of hydrogen consumes more energy than the combustion of methanol can deliver. A lot of research is still needed to increase the catalysts efficiency (Herzog et al, 1997).

Carbon dioxide can also be fixed by biological systems such as micro-alguae production or reforestration, but also in the enhancement of fossil fuels recovery from their deposits, i.e. enhancement of oil recovery from oil reservoirs and enhancement of methane recovery from unminable coal seams. These two utilisation options involve higher amounts of CO2 than any other one. The technology of enhanced oil recovery is already developed in the United States, and several deposits are under exploitation with this technique. The carbon dioxide and the oil are tending to achieve a single-phase flow, but the CO2 flows out with oil and must be separated and then recycled to the injection wells for long-termstorage. In the case of coal bed methane recovery, the CO2 replaces the methane adsorbed on the coal particles surfaces and is stored atonce, which avoids any energy consuming separation process, as is the case in enhanced oil recovery.

This new way of producing methane from coalbeds emerged from the United States a few decades ago. Indeed coal seams are filled withgas, commonly named firedamp, and even in unminable coal seams it is feasible to recover this gas: in 1996 coalbed methane production in the United States amounted to 5 % of the total US natural gas production [Stevens and Riemer (1999)].

The commonly used method of coalbed methane recovery is based on depressurisation of the coal seam. Indeed, 90% of the gas is adsorbed on the coal particle surface and this adsorption phenomenon strongly depends on the pressure as shown in Fig 6. The drop of the total seam pressure was produced until recently by pumping water outof the coal seam. Fig 6 shows that a decrease of the pressure by anorder of magnitude (10 to 1 bar) in a test cell at the temperature ofthe host site leads to a desorption of the initially present methaneby around 50%. Consequently the pressure effect on coalbed methane recovery is very high.

Water recovery at production wells occurs before that of methane. According to the observations reported on Fig 7 the maximum production of water is observed after about three years. While waterproduction decreases methane recovery increases and reaches its maximum after five or six years. A plateau is observed during a few years and afterwards the methane recovery rate decreases constantly. The lifetime of the exploitation site is around thirty years[Mostade (1999)]. A new way to enhance methane recovery from unminable coal seams consists in producing methane desorption from coal surface by injection of pressurised CO2 in the coal seam. CO2 causes a significant drop of partial pressure of methane because of its dilution in another gas. Moreover it contributes to sweep more efficiently the water out of the deposit. Indeed, it has been observed that water production at the outlet of the wells increases nearly as soon as CO2 is injected. About 70% of the methane in place (instead of 50% with water pumping) can be recovered using this technique.

Fig. 6 Adsorption Isotherm (reservoir temperature) of Methane on Coal Particles

Fig. 7 Estimation of Production Flows per km² when Using Water Pumping

Experimental studies have showed that CO2 has more affinity with coal surface than methane does. The amount of CO2 adsorbed on coal surface is two to three times higher than the corresponding amount of methane at their critical temperatures.

Unlike EOR, no significant increase in CO2 concentration isobserved at production wells outlet [Mostade (1999)]. There is no need to process the outcoming flow of methane whereas in case of EOR, CO2 and oil have to be separated and CO2 has to be reinjected. From that point of view, ERCBM has a major advantage incomparison to EOR. Moreover as coal affinity for methane is muchlower than for CO2, it can be considered as a semi-natural way toimplement underground CO2 sequestration.

Some authors, and more particularly Stevens (1998), have indicated that, considering some reservoir screening criteria, theworld-wide potential for CO2 storage in place of the methane in coal seams is of 60 Gt CO2 at a cost under $50/t CO2, and 150 Gt CO2 at a cost under $120/t CO2. Beyond this amount, the cost of the storage of an additional ton of CO2 is shown to increase tremendously.

6. CONCLUSIONS

A simplification of the MATIANT cycle (suppression of the high pressure section, Mathieu and Nihart 1999) is not penalising the efficiency, which is around 47 % minus around 2.5% when the cooling of the blades is included in the modelling. Thanks to this new structure, capital costs should be lower and prototype easier tobuild.

The lost of CO2 entrained in the extracted water flow (1.5% ofthe extracted CO2 flow) is higher than in a previous MATIANT cycle because the water extraction at the outlet of the intercoolers is here taken into account.

As the E-MATIANT cycle provides a pressurised CO2 source for ERCBM, the synergy between the two applications is obvious from the technico-economic point of view. Indeed, methane is recovered and around two or three times that amount of CO2 may be sequestred forthe long run, without the need for additional equipment.

An illustration of the synergy between zero-emission E-MATIANTcycle for electricity production and ERCBM as shown in figure 1 is described in short as TRIPLE win option :

- On top of heat and power generation with good performance, the CO2/O2 cycle delivers a flow of liquid or supercritical CO2 underelevated pressure;

- the valorisation of this CO2 flow, in particular through its use for the enhancement of methane production from coal beds; this increases the natural gas resources;

- the sequestration of CO2 in the depleted reservoir.

The model presented here is used to optimise the E-MATIANT cycle design such that it meets the requirements of themethane recovery from a given coal seam.

In practice, an E-MATIANT based power plant could be installed above coal seams with a possible tuning of its power output in orderto follow the production capacity of the coal seams. On the otherhand, such a location is recommended, as it is much easier to transport electricity in high voltage wires than to transport chemical substances like CO2 and CH4 on long distances.

7. NOMENCLATURE

ASU : Air Separation Unit

CBM : Coal Bed Methane

EOR : Enhanced Oil Recovery

ERCBM : Enhanced Recovery of Coal Bed Methane

GT : Gas Turbine

T : Temperature (°C)

TIT : Turbine Inlet Temperature (°C)

P1 : Upper cycle pressure (bar)

P2 : Lower cycle pressure (bar)

P3 : Reheat pressure (bar)

psu : Combustion chamber inlet pressure (bar)

8. REFERENCES

Herzog, H., Drake, E. and Adams, E. (1997). CO2

Capture, Reuse and Storage Technologies for

Mitigating Global Climate Change. White Paper, DOE

Order DE-AF22-96PC01257.

IEA Greenhouse Gas R&D Programme (1994). Carbon

Dioxide Disposal from Power Station. ISBN 1898373

07 8, Cheltenham, UK

Mathieu, P. (1998). Presentation of an innovative zero-

emission cycle for mitigating the global climate change,

Int. J. of Applied Thermodynamics, Vol. 1, 1998

Mathieu, P. and Nihart R. (1999), Zero - Emission

MATIANT Cycle, ASME Journal of Engineering for

Gas Turbines and Power. January 1999

Mathieu P. and Nihart R. (1999). Sensitivity Analysis of

the MATIANT Cycle. Energy Conversion &

Management. 1-14.

Mathieu P., Dubuisson R., Houyou S., Nihart R. (1999) Combinationof Quasi Zero Emission Power Cycles and CO2 Sequestration. Fifth International Conference on Technologies and Combustion for a Clean Environment, Lisbon July 1999

Mathieu P., Iantovski E., Nihart R (1998) The zero

Emission MATIANT cycle : technical issues of a novel

Technology. Second International Workshop on Zero

Emission Power Plants, Liège Jan 1998

Mostade M. (1999). Coalbed Methane Potential of the

Southern Coal Basin of Belgium, paper submitted to

the 1999 International Coalbed Methane Symposium,

Tuscaloosa, USA

Stevens S., Riermer P. (1998). CO2 Sequestration in Deep

Coal Seams : Pilot Results and Worldwide Potential,

Presented at The Fourth International Conference on

Greenhouse Gases Control Technologies, Interlaken,

Switzerland

Toward a world based on near zero CO2 emission energy systems

Philippe MATHIEU

University of Liège - Department of Mechanical Engineering-B52/3

Chemin des Chevreuils,1 - 4000 Liège 1 - Belgium

Tél. : +32-4-3669268 - Fax : +32-4-3669563

E-mail: pmathieu@ulg.ac.be - Website: http://www.ulg.ac.be/genienuc

ABSTRACT

In this paper, near zero CO2 emission power cycles (ZEP) using an oxyfuel with total recycling of the flue gas, the so-called MATIANT cycles, are presented in various configurations: regenerative Brayton-like CO2/O2 gas cycles, steam injected CO2/O2 gas cycles, CO2/O2 combined cycle, CO2/O2 gas cycles with integrated solid oxide fuel cell. Performance, costs and technical issues of these CO2/O2 cycles are discussed and compared to other ZEP cycles, like H2O/O2 cycles (Water cycle and Graz cycle) and to other technologies with CO2 sequestration. Except for the AZEP cycle, the efficiencies of the ZEP cycles are similar and range from around 45 to 55%. The cost of the kWh generated by a MATIANT cycle is 50% to 100% higher than that of a standard Combined Cycle. The total cost (i.e. generation + external costs) of the ZEP generation technology with storage of CO2 compares quite well with that of other technologies such as windmills and Combined Cycles with capture and storage of CO2.

Nomenclature

ASU : Air Separation Unit

CC : Combined Cycle

NG : Natural Gas ( a mixture of 84% methane and 16% CO2 : LHV = 42MJ/kg)

w : specific work (kJ/kg)

h : efficiency (%)

CoE : Cost of Electricity (c €/kWh)

GT : Gas Turbine

NGCC : Natural Gas fired Combined Cycle

PC/SC : Pulverized Coal/ Super Critical

ZEPP : Zero Emission Power Plant

1. INTRODUCTION

In the framework of a sustainable development, mankind has to manage a transition period from the present situation towards a world using carbon-free hydrogen and electricity as its energy carriers. Meanwhile one of the best ways to reduce drastically the greenhouse gases emissions is to produce electricity and heat through near zero emission conversion systems (like renewable energies, nuclear energy) and through the use of fossil fuels with carbon capture and sequestration. The Kyoto protocol is a very first modest step towards a stabilisation of the concentrations of the greenhouse gases in the atmosphere. In order to comply with the global climate change target as defined by the UNFCC, it is urgent (the sooner, the better) to implement from now on a strategy of drastic reduction of the emissions of the 6 greenhouse gases by 50% or more in the long run (30 years). Three types of technology are available for doing so in fossil fuels based power generation :

· the flue gas decarbonisation (80-90% CO2 retention rate)

· the fuel decarbonisation (85-95% CO2 retention rate)

· the use of ZEP generation (nearly 100% retention rate).

All three options of course involve the sequestration of CO2 once removed.

In this paper, the objective is to present ZEP cycles and compare them with technologies using CO2 sequestration as well as with renewable energies from the points of view of performance, costs and environmental impacts. Here we will focus more particularly on the so-called MATIANT cycles [21 to 27] in their various configurations, namely :

1) the E-MATIANT cycle : it is a regenerative CO2/O2 Ericsson-like gas cycle with cryogenic ASU or with a membrane for oxygen production (AZEP cycle)

2) the SI- MATIANT cycle : it is a steam injected E-MATIANT cycle

3) the CC- MATIANT cycle : it is a Combined Cycle

4) the FC- MATIANT cycle : it is a hybrid cycle with the integration of a high temperature fuel cell into a E-MATIANT cycle.

For completeness, we remind shortly that other near zero emission cycles exist and use oxyfuels and H2O as working fluid. The so-called “Water cycle” [30] is a Rankine-type oxy-fuel power cycle using O2 as the fuel oxidiser and steam as the working fluid. The “Graz cycle” [17-18] is like a regenerated steam injected oxy-fuel GT. All of them have comparable performance ( 48-55%) and environmental impact but differ in their complexity, in their need for new developments (especially the design of GTs operating at high temperatures : 1200°C and above) and in the end in their feasibility and their costs [26].

2. RATIONALE FOR NEAR ZERO EMISSION POWER GENERATION OR ZEP CYCLES

The design of the ZEP cycles has to meet the following objectives :

- nearly no releases in the atmosphere of CO2 but also of other pollutants (NOx, SO2)

- generation of electricity in a cost effective way (high efficiency and cost of electricity competitive with other advanced technologies)

- flexibility with respect to the fuel (natural gas, coal, biomass)

- technical feasibility, the system having to look as much as possible like existing technologies (turbomachines, heat exchangers, combustion chambers)

The basic idea is to use pure oxygen as the fuel oxidiser and replace the nitrogen of air by either CO2 or H2O. Since the combustion products are mainly CO2 and H2O, nearly all the CO2 generated in the combustion may be removed either by extraction from CO2 itself or by separation from H2O.

In the MATIANT cycles [20-25], CO2 itself is used as the working fluid instead of air so that the excess CO2 produced in the combustion process may be removed from the cycle, i.e from CO2 , through a simple valve. On top of that, if nearly pure O2 (+Ar) as fuel oxidiser is used, the flue gas is highly enriched in CO2. The separation of CO2 and H2O is then easy and there is no longer need for a very penalising scrubber separating CO2 from N2 in the flue gas.

Typically when using CO2 chemical scrubbing, the penalties are – 10/13 % pts on efficiency, -20/25 % on power output and + 50 to 70 % on cost of electricity, depending on the fuel and type of plant [3; 7;8 ; 12; 13; 14; 16; 25; 29; 31].

As shown in table 1, the considered technologies have different retention rates of CO2, different NOx emissions and different efficiency penalties compared to a reference cycle.

With the above-mentioned CO2 capture technologies, the recovery of CO2 ranges from 80% to 95% whereas the NOX emissions are similar to those in fossil fuels-fired power plants. Subsequent reduction of NOX requires the implementation of further processes such as a Selective Catalytic Reactor. In addition, CO2 removal infrastructure (and a syngas manufacturing plant for the fuel decarbonisation) is required at 1 bar for amine based chemical absorption and under pressure for physical absorption. This increases significantly the total plant capital cost. Combustion in O2/CO2 atmospheres enables nearly 100% CO2 recovery as well as a reduction of NOX emissions below 1 ppm. However, this concept calls for expensive and highly electricity-consuming oxygen supplies, as well as the development of new (and costly) turbomachinery equipment, like a gas turbine using CO2 as the working fluid. With commercial cryogenic air separation techniques for the production of pure oxygen (e.g. cryogenic separation or pressure swing absorption (PSA) the electricity consumption is 250 to 300 kWh/ton oxygen produced. This is very penalising on performance (loss of at least 20% of the net power output and 10%points of the thermal efficiency) and on cost (increase of CoE by some 50%).

An apparently less penalising way to produce O2 would be the use of membranes of the ITM type (Ion Transport Membrane) operating at high temperatures (700- 900°C). This option (the AZEP cycle) developed by Alstom and Norsk Hydro is presented later.

3. MODELLING OF THE MATIANT CYCLES

· 3.1 the E-MATIANT cycle

The so-called MATIANT cycles [20-24] (contraction of Mathieu and Iantovski) are of the O2/CO2 type. In this novel concept, the trick is to replace the nitrogen of air by CO2 itself which is now the working fluid and the thermal ballast for the control of the flame temperature, and use O2 as the fuel oxidiser. This requires hence an ASU leading to performance and cost penalties. Fig. 1 shows that the E-MATIANT cycle uses an oxyfuel (mixture of a fossil fuel and of O2 produced in an air separation unit (ASU)) and CO2 as the working fluid and thermal moderator, with total recycling of CO2. The combustion products are mainly H2O and CO2. A water cooler/separator is used for the water removal before the compressor. The non-extracted water is removed at the intercoolers' outlets in the staged compressor. The excess CO2 (produced in the combustion process) is nearly totally removed in liquid or supercritical state at high pressure, (above 73 bar and at 30°C) from the working fluid, CO2 itself, through a simple valve at the compressor’s outlet. It is ready to be transported , normally between 15-30°C and 100-150 bar as a supercritical fluid with the viscosity of a gas and the density of a liquid. It may be reused (for example, for enhanced recovery of oil from oil wells or of methane from coal beds), and finally for permanent disposal. If necessary, an additional compressor is used to bring about the removed excess CO2 flow from the combustion chamber pressure up to the pressure required for a possible use of CO2 (like enhanced recovery of oil or methane) or for its long term sequestration.

It is designed as a regenerative Ericsson-like cycle with 2 nearly isothermal processes (a staged compressor with intercoolers and a staged expansion with a reheat) and 2 nearly isobaric processes (in the regenerator, in the 2 combustion chambers, and in the cooling of the flue gas down to the compressor inlet temperature). Fig. 2 shows how efficiency and power output change when the reheat pressure increases from 5 to 36 bar .Under the conditions given in table 1 (the turbine inlet temperature (TIT) is 1300°C, the upper and lower cycle pressures are 60 and 1 bar, respectively), the net efficiency increases with the reheat pressure and ranges from around 40% when the optimal reheat pressure, corresponding to the highest specific work, is about 10-12 bar up to a maximum of 47-48% when this pressure rises up to 36-40 bar. The specific work decreases then by 17%, from 575 kJ/kg down to around 475. The highest efficiencies are achieved for the highest reheat pressures but then the mechanical stresses are also the highest. Given that efficiency and specific work vary in an opposite way when the reheat pressure increases above 10 bar and that the efficiency curve starts to flatten above 25 bar, a reheat pressure of 25 bar appears as a good compromise with respect to the global performance (h = 45%, w = 515kJ/kg). The penalties due to the ASU's consumption and to the cooling of the hot parts like vanes and blades of the turbines and the walls of the combustion chambers (2.5% points) are taken into account. Due to its electricity consumption, i.e. 0.25kWh/kg O2 at 1bar, 15°C and a 99.5% purity (0.5% Argon), the ASU makes the efficiency fall by some 13% points. That is the price to pay in order to avoid the use of CO2 chemical scrubbers and a CO2 compressor aimed at bringing the extracted CO2 up to a supercritical state (80-110 bar). No miracle however, what has been saved downward is repaid upward.

The total recycled CO2 throughput is 1.022kg/s or 4.9 kgCO2 /kWhe .

The Advanced Zero Emission Power cycle (AZEP) is a variant of the E-MATIANT cycle and is being developed under the leadership of Alstom Power and Norsk Hydro [4-5]. Instead of a cryogenic ASU, a mixed conducting membrane MCM integrated in the combustion chamber is used here for O2 production. This is supposed to be cheaper than an ASU and less penalising on global efficiency (the promoters claim a loss of efficiency as low as 2 %pts). The basic process in MCM is O2 ion diffusion which is all the more high since the temperature is high, above 700°C and which generates highly pure O2. The fuel is introduced in the flow enriched in O2 (permeate side of the MCM). The products CO2 and H2O are expanded and separated in a cooler. CO2 is then compressed up to 100 bar. Due to the rather low temperature of the air open cycle (700-900°C), the net efficiency in a CC configuration calculated with our modelling is lower than 40%. The performance is significantly limited by the operating temperature of the membrane. The situation would be better if the membrane was NOT integrated in the combustion chamber but instead was used in an external system Technical developments are still needed like the design of very high temperature membranes with ceramic materials, of CO2 high temperature turbines with CO2 (or steam) cooling, of combustion under pressure in pure O2 in a CO2 atmosphere.

· 3.2 SI-MATIANT cycle or Steam-Injected CO2/O2 gas cycle

When the extracted water (point 9, fig 1) is recycled and injected as steam in the low pressure combustion chamber (5-6 on Fig 1), the net efficiency climbs up to 50%. This is called the SI-MATIANT cycle. The steam is produced in a heat recovery boiler fed with water extracted from the cycle. The amount of recycled water is determined by the choking condition at the expander’s inlet. As far as the cooling of the hot parts is concerned (vanes, blades, walls), CO2 is used as the coolant and is either bled from different stages of the compressor or the excess CO2 extracted from the cycle can be partly or totally re-injected in the cooling circuit. Here steam could be used as the coolant in a external circuit like in the GE gas turbines of the H-class.

· 3.3 CC-MATIANT cycle

Another option is the Combined Cycle (called the CC-MATIANT cycle) consisting of a O2/CO2 Brayton-like cycle without and with reheat and of a Rankine cycle comprising a heat recovery steam generator with 2 pressure levels and operating at subcritical pressures (under 221 bar). Here the compression is nearly isentropic instead of nearly isothermal. Due to the high temperature at the GT expander's exhaust, a regenerator is used to cool the gas (down to the acceptable temperature at the heat recovery boiler inlet (700°C) in a CC configuration) and to preheat the recycled CO2 flow from the compressor outlet. This heat could also be given up to a high temperature fuel cell like in the FC-MATIANT below. Under 700°C, the sensible heat of the exhaust gas is transferred to the steam cycle. Using a heat recovery boiler with various configurations, with two optimised pressure levels up to 3 pressure levels with reheat in supercritical steam conditions, the net cycle efficiency is in the range 49% -55%. This is the efficiency of a standard NGCC. Like in the E-MATIANT cycle, the amount of CO2 escaping from the cycle inside the water flow leaving the separator is about 0.5% of the removed CO2 flow. The additional cost of electricity compared to that of a combined cycle CC (55% efficiency) without CO2 removal amounts to 50-100% higher than that of a CC kWhe The cost of the electricity generated by a E-MATIANT plant would be similar to that of a current combined cycle plant equipped with a CO2 scrubber.

· 3.4 FC-MATIANT CYCLE or FUEL CELL–MATIANT CYCLE

A solid oxide fuel cell (SOFC: 55-60 % efficiency operating at 900-1000°C) may be integrated at the hot gas exhaust (points 7-8). The heat is used for fuel, oxygen and water preheating up to the fuel cell (or its reformer) temperature (800-900°C). This is the FC-MATIANT [27] shown on Fig 3. The 2 technologies fit quite well since this increases both the power output and the global efficiency .The best integration provides a gain in efficiency around 3-4% pts and of 15-20% in power output Unfortunately the cost is presently prohibitive. Considering the performance increase in the integrated system, the investment is not cost-effective at present. In addition, a lot of developments on both the ZEP cycle and the SOFC have still to be made before the technology arrives at the commercial stage.

Figure 3: Lay-out of the FC-MATIANT cycle

4. COST OF ELECTRICITY

Calculations show that, in the structure of the total cost (generating cost + environmental cost), the share of transportation (transportation by pipeline: 1-3 €/t of CO2 per 100 km) and sequestration (storage: 1-3 €/t of CO2 ) is much smaller (5 to 25 %) than the cost of capture (capture and compression to 110 bar: 30-50€/t for a 500 MWe gas- or coal-fired plant at current fuel prices 2.5 and 1.5/GJ respectively). These costs compare favourably with other options such as the widespread use of renewable energy sources and are expected to fall as the technology matures. As an example, figure 4 shows that the total costs of the kWh generated by natural gas and coal fired power plants with CO2 capture and sequestration (6 and 6-7.5 c€/kWh resp.) and is of the order of magnitude of the total cost of windmills (on the coast 4.5c€ and off-shore 6c€/kWh).

5. TECHNICAL ISSUES

The technical issues are linked to the composition of the working fluid turbomachines and to the operation on CO2. Developments are required on cooled CO2 expanders at 1300°C and higher temperatures combustion in pure oxygen in a CO2 atmosphere under pressure for water cycles, steam turbines at 1300°C and higher and compressors and condensers of mixtures of CO2 and H2O at low pressures for AZEP cycles, membranes at high temperatures. All the ZEP cycles ask for advances in materials and cooling techniques. They look technically feasible; however their complexity and hence their cost are not the same and so are the R&D they require. The easiest to build early and cheap will be the winner. Will it come from developments of steam or of gas turbines? This is one of the main questions.

6. CONCLUSIONS

Benefits:

- ZEP cycles are particularly well-suited to use in power generation and, in combination with other CO2 abatement techniques, could enable deep reductions to be achieved with least impact on the global economy. They will enable a country to continue to burn efficiently and cleanly fossil fuels, avoiding social and economic disruptions.

- thanks to its modular and compact structure, the ZEP technology will be well suited for both small (10 - 50 MW ) and large (400-500 MW) scale power plants, either offshore or onshore;

- the efficiency of the considered ZEP CO2/O2 cycles ranges from 45 to 55% Consequently the efficiencies of both CO2/O2 and H2O/O2 ZEP cycles are similar. They differ in their complexity, costs and needs for development of new components;

- the E- and CC-MATIANT cycles release a small amount of CO2 flowing out of the cycle dissolved in the extracted water, typically 5-6 g CO2/kWh. This is 60-70 times less than a current CC (55%; 350 g CO2/kWh) and 6 –7 times less than this CC with CO2 sequestration. The extracted excess CO2 is nearly pure ,hardly contaminated with NOx, and may be directly injected into a storage reservoir;

- although matching 2 technologies such as ZEP cycles and SOFC, the FC-MATIANT cycle comprises components under development and is still too expensive.

Ancillary benefits:

- no use of chemicals, no wastes, simple exhaust gas treatment, easy separation of CO2 and water;

- when the captured CO2 is used in enhanced oil or methane recovery schemes, the income produced helps to offset the sequestration costs.

Challenges are:

- the increase of the CoE;

- the development of new components;

- the separation of O2 from air. Some 10% pts efficiency being lost here due to the electricity consumption of ASU based on cryogenic distillation, R&D efforts have to be made on O2 separation using membranes;

- unlike CO2 capture technologies, ZEP generation as well as long term storage are not proven technologies yet.

As a conclusion of our analysis, the E- and CC- MATIANT cycles, only designed for CO2 emission mitigation , appear to be attractive alternatives to other technologies with CO2 sequestration, especially when combined with enhanced fossil fuels recovery. Even if the balance advantages/drawbacks is not cost effective yet, political and social incentives, such as the Kyoto protocol, a higher value of the environment (taxes), as well as a change of economy such as the use of electricity and carbon-free hydrogen as energy carriers for centralised, distributed and mobile energy users should make the ZEP technologies commercially available to combat climate change in the long run (30 years).

BIBLIOGRAPHY