The present invention relates to a process and a plant for

hydrothermal Synthetic Natural Gas (SNG) production from carbon- containing substrates like, for example, biomass and biomass waste, and industrial residues.

Hydrothermal gasification of carbon-containing substrates in supercritical water is one of the rare pathways that potentially allows for a complete conversion of wet carbon-containing

substrates into Synthetic Natural Gas (SNG) . A flowsheet for the production and separation of the crude gasification product is given in the European Patent Application EP 1 931 753 Al . The flowsheets are based on a liquid-vapour (LV) at high (i.e. at or slightly below gasification) pressure or only a LV at high and a gas separation step at low (i.e. at or slightly above gas grid or ambient) pressure. This prior art reference therefore discloses only a single gas separation step. Further, no energy (i.e. power) recovery form the crude product at high pressure is considered and both the liquid and vapour phase are expanded in valves. The heat requirements of the process are satisfied by combustion of part of the final product (i.e. grid-quality SNG) . This application presents process design configurations for the conversion of carbon-containing substrates into SNG and discusses some key aspects of the adaption of the process integration for different feedstocks to reach optimal process design and

performances. It thereby demonstrates that the process design and performance is influenced by available technology, catalyst

deactivation issues, process scale and the characteristics of the processed substrate.

Conventional biomass conversion technologies for the production of fuel and power require relatively dry and clean feedstock and thus suffer from increasing competition for a relatively scarce

resource. Hydrothermal gasification of biomass in supercritical water is a promising process alternative to produce synthetic natural gas (SNG) since it relaxes this requirement and grants access to a large range of carbon-containing, low quality

feedstocks such as wet lignocellulosic biomass and biomass wastes that are difficult to valorise by other means and thus relatively cheap .

Several authors have reviewed process fundamentals, chemistry and catalysis for hydrothermal gasification while mainly focussing on the pathway to hydrogen. The production of methane has been

experimentally demonstrated in a batch reactor from wood substrate, and technology and process development for a continuous process is under way. In this regard, an overall process model has been developed and has provided a first investigation of the process design and life cycle assessment for the hydrothermal production of SNG from wood and manure.

Most of these studies discuss general process principles, present laboratory and pilot units or focus on detailed experimental investigations. To our knowledge, no-one has yet presented a process design model that quantitatively takes energy integration and recovery into account. At the time of the general process principle developments, only limited insight into some process steps was yet available. Energy integration has been performed on a scenario basis without optimisation, and the synergies between the reaction and separation subsystems through process integration have been disregarded. For this reason, the conceptual process design of hydrothermal gasification for the co-generation of SNG and power from wet lignocellulosic biomass and biomass wastes has been systematically investigated. The known model has been

improved with both more general and detailed technology models that are reconciled and validated with data from experimental investigations. A general superstructure for integrated product separation, power recovery and heat supply for the process has been developed, and multi-objective optimisation has been applied to explore the design alternatives for selected candidate substrates. This has allowed for generating the innovative process configurations that are part of this application.

This application summarises some key aspects of the process integration and the resulting process configurations, and

discusses the influence of the feedstock on the optimal process design and its performances.

The conversion of carbon-containing substrates into methane and carbon dioxide is based on the conceptual overall net reaction, which can be written for a typical composition of lignocellulosic matter :

CHi.35 Oo.63 + 0.3475 H ₂ 0 → 0.51125 CH ₄ + 0.48875 C0 ₂ , (1) Aii _r ° = -10.5 kJmol ^_1 CHi.350o.63

Technically, the conversion requires a heterogeneous catalyst and is thus impossible to perform directly with the solid feed since the big macromolecules cannot access the active sites. The most envisaged conventional route is thus to first decompose the solid feedstock by gasification and then catalytically synthesise the obtained H ₂ /CO-rich producer gas into CH ₄ and CO ₂ . The conversion of (1) therefore splits up in an endothermal gasification step above 800°C and an exothermal synthesis step at 350-450°C at which CH ₄ is thermodynamically favoured. This limits the product yield since a considerable part of the energy content of the feed is required to form intermediate H ₂ /CO and is then converted into excess heat in its highly exothermal methanation. The hydrothermal route omits the endothermal step at high

temperature and targets a direct conversion at 300-500°C into CH ₄ , C0 ₂ and minor amounts of H ₂ and CO (EP 1 931 753 Al), or at 400- 800°C into CH ₄ , H ₂ , CO and C0 ₂ (DE 102 59 928 Al), both references herewith incorporated by reference. Instead of forming an

intermediate gas, the biomass is hydrolysed and gasified in a supercritical aqueous environment at around 250 to 500 bar, which allows for an efficient contact with the catalyst. The thoroughly fluid processing thereby requires a carbon-containing feed in form of a pumpable slurry with a total solids content in the range of 5-80%wt (preferably 15-40%wt) depending on the type of substrate. Although this makes the process suitable for wet biomass since the heat requirement up to the gasification temperature is reduced by high pressure and drying is not required, the design must take care of the high amount of water that accompanies the reacting species throughout the process. As this represents the major share of the heat transfer requirements, the overall performance gets sensitive to the energy integration of the plant.

Compared to the conversion at moderate pressure, high pressure thereby considerably reduces the heat demand up to the

gasification temperature. This makes the process suitable for wet biomass and other, preferably wet carbon-containing substrates, although the design must take care of the high amount of water that accompanies the reacting species throughout the process.

Depending on the humidity and type of substrate that is processed, the first step in the block flow diagram of Fig. 1 is to

mechanically dry or grind and dilute the feed. The slurry is then compressed to typically 300 bar and heated close to pseudo- critical conditions, during which hydrolysis and other reactions occurs. When passing the pseudo-critical point, inorganics present in the feedstock will precipitate as salts and risk to plug the equipment and deactivate the catalyst if they are not efficiently removed. To do so, the high pressure slurry is injected through a dip-tube into a heated vessel, in which the salts precipitate at supercritical conditions. The nutrient content of these salts may be recovered and used as fertiliser. The hydrolysate then passes through a fixed bed of suitable catalyst (e.g. nickel- or

ruthenium-based, broad variety of metal and/or metal oxide

catalysts is disclosed in the prior art) , which converts, at ideal conditions, more than 99.9% of the organic matter into CH ₄ , CO ₂ , some residual ¾ and only traces of CO. The catalyst may be a honeycomb- or plate type catalyst or dynamic bed catalyst, wherein catalyst support material are selected from a group consisting of titania, zirconia, ytterbia, silica, zeolithes and mixtures thereof and catalytic active compounds are selected from a group consisting of transition metals and their respective oxides, precious metals, rare earth metals and their respective oxides. The parameter of the reaction are for example disclosed in EP 1 931 753 Al and DE 102 59 928 Al as well the document cited in the search reports to these applications which are herewith

incorporated by reference. In order to inject the produced SNG into the natural gas grid, the methane must be separated from water, carbon dioxide and possibly hydrogen and carbon monoxide. For the typical lignocellulosic feedstock of Eq. 1 diluted to 20%wt total solids, the crude product from gasification contains approximately 84 mol% of ¾0 and 8 mol% of each CH ₄ and CO ₂ in a near- or supercritical mixture at 350-450 °C and around 300 bar. As indicated in Fig. 1, the process furthermore requires additional heat for reaching the conditions of the salt separator and/or gasification. It is therefore the aim of the present invention to provide a system and a process for hydrothermal SNG production from wet carbon-containing substrates having a design of the product separation that not only considers the grid quality specifications for SNG, but also the recovery of the exergy potential of the crude gas and the supply of required heat for the plant.

With respect to the process this aim is achieved according to the present invention by each of the processes given in claims 1, 3 and 5.

With respect to the system this aim is achieved according to the present invention by each of the systems given in claims 7, 9 and 11. The present invention comprises several process alternatives for separating the crude product while recovering energy in the form of electricity. The following solutions according to the present invention provide:

First solution (cf. Figure 7)

The first solution provides a process and a system for

hydrothermal SNG production from carbon-containing substrate, comprising the steps of:

a) generating a pumpable slurry having a predetermined solids content as calculated in dry mass, such as 5 to 80 weight%; b) pressuring and heating the slurry above 50 bar and 200°C; c) gasifying the hydrolysed slurry, from which precipitated

solids are optionally separated before and/or after the gasification,

d) performing LV separation and gas separation at high pressure to generate a liquid stream and a vapour stream;

e) for the liquid stream, liquid expanding with or without power recovery or evaporating and turbine expanding followed by LV separation; and

f) valve expanding or reheating and turbine expanding of the

vapour stream followed by combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power of the so-generated SNG gas and/or supply of the so- generated SNG gas to a gas grid. A preferred embodiment of this first solution can specify the process that prior to the supply of the so-generated SNG gas to the gas grid performing at least one further gas separation step followed by combustion and expansion in a turbine, and/or

electrochemical conversion to heat and/or power of the separated depleted gas and/or supply of the so-generated further purified SNG gas to a gas grid.

Second solution (cf. Figure 8)

This solution provides a process and a system for hydrothermal SNG production from carbon-containing substrate, comprising the steps of: a) generating a pumpable slurry having a predetermined solids content as calculated in dry mass, such as 5 to 80 weigth%; b) pressuring and heating the slurry above 50 bar and 200°C; c) gasifying the hydrolysed slurry, from which precipitated

solids are optionally separated before and/or after the gasification,

d) performing LV separation at high pressure to generate a

liquid stream and a vapour stream;

e) for the liquid stream, liquid expanding with or without power recovery or evaporating and turbine expanding followed by LV separation, and

f) valve expanding or reheating and turbine expanding of the

vapour stream followed by combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power, and/or gas separation of the so-generated raw SNG gas followed by combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power, and/or supply of the so-generated SNG gas to a gas grid. Again, a preferred example of the second solution can specify that prior to the supply of the so-generated SNG gas to the gas grid performing at least one further gas separation step followed by combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power of the separated depleted gas and/or supply of the so-generated further purified SNG gas to a gas grid.

Third solution (cf. Figure 9)

The third solution provides a process and a system for

hydrothermal SNG production from carbon-containing substrate, comprising the steps of:

a) generating a pumpable slurry having a predetermined solids content as calculated in dry mass, such as 5 to 80 weight%; b) pressuring and heating the slurry above 50 bar and 200°C; c) gasifying the hydrolysed slurry, from which precipitated

solids are optionally separated before and/or after the gasification, d) valve expanding or reheating and turbine expanding of the gasified crude product;

e) performing LV separation at lower pressure to generate a

liquid stream and a vapour stream;

f) waste water treating of the liquid stream, and

g) for the vapour stream, combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power, and/or gas separation of the so-generated raw SNG gas

followed by combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power of the separated depleted gas and/or supply of the so-generated SNG gas to a gas grid.

Again, a preferred example of the third solution can specify that prior to the supply of the so-generated SNG gas to the gas grid performing at least one further gas separation step followed by combustion and expansion in a turbine, and/or electrochemical conversion to heat and/or power of the separated depleted gas and/or supply of the so-generated further purified SNG gas to a gas grid.

The present invention consists in several process alternatives for separating the crude product while recovering energy in the form of electricity. These options increase the process efficiency, decrease the capital cost, provide a more versatile process design and increase the robustness of the process performance with respect to measures that may be necessary to prevent catalyst deactivation .

Preferred embodiment of the present invention are hereinafter described with reference to the following drawing which depict in:

Figure 1 a block diagram of the hydrothermal SNG production

process; Figure 2 a schematic superstructure including all flowsheet alternatives for combined crude product separation and expansion (crude product is the product after

gasification has taken place) ;

Figure 3 an illustration of the minimum energy requirements under product expansion without power recovery (a) and under power recovery by reheating the entire crude product (b) ; Figure 4 an illustration of the optimal thermodynamic and thermo- economic trade-off at 20 MW _t h, biomas s , wherein in (a) the maximal partial efficiencies, in (b) a full

consideration of the catalyst cost, and in (c) without catalyst deactivation is shown, left image without and right image always with power recovery from the high pressure vapour phase;

Figure 5 optimal process configurations at 20 MW _t h, biomas s depending on the substrate feedstock;

Figure 6 separation flowsheets according to EP 1 931 753 Al ;

Figure 7 separation flowsheets with LV and gas separation at high pressure ;

Figure 8 separation flowsheets with LV at high pressure and gas separation at low pressure;

Figure 9 separation flowsheets without separation at high

pressure;

Figure 10 detailed flowsheets of the alternative valve or turbine expansion route in Figures 7 to 9; and Figure 11 detailed flowsheets on the possible energy technology

involved into the processes according to Figures 7 to 9; Table 1 properties of exemplary candidate biomass feedstocks; and

Table 2 energy balances with and without supplying additional heat from other sources than the combustion part of the gasification product.

Following the developed methodology for the conceptual design of thermochemical production of fuels from biomass, a decomposed modelling approach has been applied to identify the inventive flowsheets. The material conversion in the process units and their energy requirements are computed in energy-flow models, which are assembled in a process superstructure of all relevant

technological options. The material flows defined by this

superstructure (Fig. 2) act as constraints in the energy- integration model that is formulated as a mixed integer linear programming problem in which the heat exchanger network is

represented by the heat cascade. Considering waste and

intermediate product streams as fuel to supply the required heat, the combined SNG and power production is optimised with respect to operating cost. For the so-determined flowsheet, design heuristics and pilot plant data have been used for rating and costing the equipment required to meet the thermodynamic design target.

While no major technology alternatives for feed pretreatment, hydrolysis, salt separation and gasification exist, several distinct strategies for the separation and expansion of the crude product are conceivable and might influence the process

performance markedly. The crude product from gasification contains more than 80% ¾0, approximately equal amounts of CH ₄ and CO ₂ , and some marginal ¾ and CO. Due to the supercritical conditions, its upgrade and expansion to grid conditions potentially allows for recovering mechanical energy, which however competes with the supply of thermal energy required for the process steps. Another important aspect of the separation system design is the quality of the depleted stream, which may be used to supply the required heat and thus relax the need for a higher methane recovery in the separation. The given boundary conditions thereby suggest different strategies for combined product separation and expansion that are outlined in the general superstructure of Fig. 2. Apart from gas separation at the gas grid pressure by, for example, absorption with a dedicated liquid solvent, adsorption to a solid by pressure or temperature swing adsorption or any other gas separation technology, possibly followed by a second separation step (e.g. a membrane stage to remove residual hydrogen), the better solubility of CO ₂ in water when compared with CH ₄ may become technically relevant at the prevailing process pressure. A trade- off between selectivity and good absolute solubility might thereby occur with respect to pressure. In any case, the separation is best at low temperature, and additional water is required for absorbing the bulk CO ₂ to reach gas grid quality. In order to recover mechanical energy from the crude product at high pressure, the separated vapour phase - or the entire

supercritical bulk, if no high pressure separation is applied - may be expanded through turbines (see Path 1) . It might thereby be advantageous or even necessary to preheat the stream above the gasification outlet temperature, which increases the thermal efficiency of the recovery and prevents an expansion too far into the two-phase region. Compared to an isenthalpic expansion through valves, this causes less heat to be available from the crude product stream since energy is withdrawn at high temperature. For the liquid phase obtained from the separation at high pressure, the available exergy can be recovered by liquid expanders (see Path 2) . As an alternative, the liquid phase could also be

reheated and expanded into the vapour domain, which would allow for extracting more mechanical energy from the available potential, but also requires a considerable amount of heat to be supplied (see Path 3) .

If the product is not upgraded to grid quality at high pressure, the liquid vapour and gas separation need to be carried out after the expansion of the crude product and similar technology as in the conventional route applies. For the complete gas separation at grid pressure, a column for physical absorption in a liquid solvent (such as Selexol, Rectisol, etc) or any other technology seems appropriate (see Path 4) . The combination of both high pressure and grid pressure separation is also conceivable. In order to reduce the amount of required additional water and thus pump power, the gas could only be pre-separated at high pressure and upgraded to grid-quality after expansion. For this purpose, one or several membrane stages or any other gas separation

technology around gas grid pressure could be used (also Path 4) . Fig. 3 shows the minimum energy requirements (MER) of the

principal flowsheeting options for wood at the default operating conditions. The composite curves that identify the contributions of the process sections highlight that the layout of the product separation and expansion section determines the pinch point and influences the energy demand markedly. If no power recovery from the crude product is performed (Fig. 3a), the process pinch is situated at the salt separator where 186 kW MW ^_1 _b i _omass are required at 440°C. Below, the specific and latent heat of the crude product is sufficient for preheating and hydrolysis of the feed, and an excess of about 150 kW MW ^_1 _b i _omass can be recovered between 250 and 400°C. Limited power recovery by liquid expansion of the high pressure condensate and/or expansion of the incondensable mixture with previous reheating to the process pinch does not change the MER and only marginally influences the amount of excess heat.

If no separation at high pressure is applied and the crude product including the bulk water vapour is expanded in a turbine, the energy withdrawn as mechanical work is not available anymore at the gasification outlet temperature. As a consequence, the pinch point shifts to the turbine outlet temperature and results in an increased heat requirement but at a lower temperature (Fig. 3b) . Reheating the crude might thereby be required to avoid

condensation in the final turbine stages and enhances the

thermodynamic conversion efficiency, which leads not only to an increased power output but also to an increased heat demand. If the condensable phase from separation at high pressure is evaporated, reheated and expanded to atmospheric pressure, the characteristics of the process integration change drastically. For such a configuration, the pinch point would shift to the

saturation temperature of the mixture at atmospheric pressure and the MER increases to 64-68% of the raw material's heating value. This would require to burn a large part of the produced gas and to turn the generation of electrical power to the plant's main purpose .

The process concept of hydrothermal gasification principally addresses the conversion of wet biomass and biomass waste with considerably different properties. For this reason, the influence of the substrate properties on the process design and performance is discussed here for some representative examples.

Tab. 1 provides the relevant properties of a selection of

exemplary feedstocks for hydrothermal gasification. Among the potential substrates, manure and sewage sludge are abundant biomass wastes with a large potential. Coffee grounds and lignin slurry represent typical energetically exploitable by-products. While the former is a residue from the food industry, large amounts of biomass are retrieved as slurries with high lignin content in the pulp and paper industry or in a future production of fuel ethanol from lignocellulosic biomass. In case of the latter, excess heat from the SNG production might thereby also satisfy the requirement for biomass pretreatment and ethanol distillation, and favourable effects might emerge from process integration. Finally, microalgae are considered as a

photosynthetically efficient energy crop that is cultivable in photo-bioreactors on marginal land, from which a reduced

environmental impact is expected.

Compared to wood, all these substrates offer a higher hydrogen fraction and thus an increased theoretical methane yield from the dry, ash-free substance according to (1) . Except coffee grounds and lignin slurry, they yet suffer from a higher ash content which reduces the effective biomass content if diluted to the same dry solids content. Among the substrates, manure suffers from a

particularly low solids content on an as-received basis and is the only substrate for which water purification by reverse osmosis is considered necessary.

The process design for hydrothermal conversion is particularly flexible with respect to the co-production of fuel and power. In order to explore this particular trade-off, the co-generation potential is addressed in a first optimisation step that considers the partial SNG and electric efficiencies defined by the ratios of the SNG and electricity yields to the biomass input, respectively, as objectives. In a second step, thermo-economic optimisations of the process design are carried out with and without considering the catalyst cost to investigate the importance of catalyst

deactivation in the design. The chemical efficiency, defined as the equivalent SNG yield if electricity is substituted by the amount of gas consumed for its generation in a combined cycle at an exergy efficiency of 55%, is thereby used as an aggregated thermodynamic objective. As economic objective, the specific investment cost per installed capacity is used, including also the total catalyst cost over the entire plant lifetime if catalyst deactivation is considered. Fig. 4 provides the Pareto fronts of the overall best

configurations for all substrates in the different optimisation steps. The maximum partial efficiencies in Figure 4(a) assess a nearly equal co-generation potential for coffee grounds and lignin slurry, which are performing slightly better than wood. Microalgae, manure and sewage sludge are consecutively worse. In comparison with Tab. 1, this order mainly follows the ash content of the substrates. With an equal total solids content of 20%, the net dilution of the reactive biomass in water almost doubles in the worst case of sewage sludge and has a fatal impact on process efficiency since the amount of water to be entrained is doubled as well. In addition to the maximum combined efficiency situated close to the maximum SNG yield, power recovery from the high pressure vapour phase allows for a high marginal efficiency in substituting the SNG by electrical power generation. This leads to a second peak with respect to the combined efficiency at net SNG yields below roughly 50%, which is particularly beneficial for low quality substrates like sewage sludge.

The efficiency considerations have a big impact on the thermo- economic performance of the conversion. Compared to coffee grounds and lignin slurry which are dominating the common Pareto domain of Figures 4 (b) and 4 (c) , the conversion of wood is slightly less efficient and more expensive due to the higher CO ₂ share in the crude product that requires more effort for CO ₂ separation. It is thus competing with microalgae whose conversion is disfavoured by a slightly higher ash content. The waste substrates are clearly worst. While sewage sludge is seriously penalised by its low thermodynamic performance, manure suffers from high investment cost for dewatering and especially wastewater treatment by reverse osmosis . Fig. 5 illustrates the evolution of the process configuration on the thermo-economic Pareto fronts and clearly highlights that the optimal choice depends not only on the availability of energy recovery technology, catalyst deactivation and plant scale, but also on substrate properties. Although the use of a single

separation technology is more efficient, its combination with, in this example, a polymeric membrane is less costly since the purification requirement is relaxed. The flowsheets with

absorption of CO ₂ in water thereby require less investment than Selexol, but are disfavoured at higher efficiency. An exception is observed if power recovery from high pressure vapour is excluded and catalyst cost can be disregarded, for which water absorption is the unconditionally best technology for all substrates. If catalyst cost is considered and power recovery feasible,

superheating and expansion of the bulk crude product emerges as an interesting alternative since its efficiency is less sensible to the design constraints imposed to avoid excessive deactivation. For the economically best configurations, an almost neutral power balance at high SNG yield seems best if catalyst deactivation does not need to be considered. Otherwise, a yield distribution in which up to 10% of the biomass input is converted into power is more advantageous. The assessed break-even costs for coffee waste, lignin slurry and microalgae are similar or higher to those of wood, which results in considerably higher plant profitability if lower substrate prices apply. Although manure conversion suffers from high investment cost, such plants might yet be profitable since also low compensations for the feedstock can be expected. The conversion of sewage sludge increases the energy efficiency of wastewater treatment, but economical benefits should principally emerge from avoiding another type of waste treatment.

As schematised in Figures 7 to 9, the inventive flowsheets

introduce multiple gas separation steps and propose internal energy recovery from the crude gasification product at high pressure. In particular, after cooling of the crude product to reach subcritical conditions and liquid-vapour phase separation (U11-U13) , the use of liquid expanders (U22) is introduced to recover mechanical energy from the liquid phase at high pressure, which is not possible with valves (U21) in prior art representing Figure 6. Alternatively, the liquid phase might be evaporated, superheated and expanded in the vapour phase (also U22), which increases the power output to the expense of net SNG production. Similarly, the vapour phase at high pressure might be optionally reheated and expanded in a turbine (U32) to recover mechanical energy and convert it into electricity. If reheating in unit U32 is omitted, cooling energy below ambient temperature can be supplied for refrigeration from the expanded stream (s) . These different configurations of unit U32 are detailed in Figure 10 for the process path with a single (a) and multiple (b) gas separation steps .

Contrary to the prior art flowsheets shown in Figure 6, these flowsheets consider depleted and intermediate product streams instead of the final product to satisfy the heat requirements of the process. Furthermore, they introduce the use of energy technologies that convert the calorific value of (gaseous) fuels into useful heat and/or power (U92), such as electrochemical conversions (e.g. fuel cells) and/or standard and/or advanced combustion units, i.e. the use of a catalyst that allows for a complete combustion at low temperatures or a partial oxidation gas turbine (POX-GT) for power co-generation, which is far more

efficient in the present application than a standard gas turbine. These different configurations of unit U92 are detailed in Figure 11 for a first path with the conversion to heat and power by standard or catalytic combustion (a) and for a second path with the conversion to heat and power by a partial oxidation gas

turbine POX-GT (b) and for a third path with the conversion to heat and power by electrochemical conversion. Figure 7 illustrates the flowsheet alternatives with LV and gas separation at high pressure (U13) . Compared to the prior art solution according to EP 1 931 753 Al shown in Fig. 6(a), Figure 7(a) introduces power recovery technology for both the liquid (U22) and the vapour (U32) phase. The depleted gas recovered from the liquid phase is further explicitly intended for conversion into useful heat and/or power (U92) . In addition, Figure 7(b) suggests an additional gas separation step (U52) by for example a membrane stage in order to relax the separation requirement at high

pressure. All depleted streams are considered for conversion to heat and/or power (standard, catalytic and/or POX-GT combustion, and/or electrochemical conversion (e.g. fuel cells), U92) and the heat requirement may be balanced by combustion of intermediate products that is withdrawn upstream the final separation step. Figure 8 summarizes the inventive flowsheet alternatives with LV separation at high pressure (U12) and gas separation at low

pressure (U53 in Figure 8 (a) and U52 and U53 in Figure 8 (b) ) .

Compared to the prior art solution according to EP 1 931 753 Al shown in Figure 6 (b) , Figure 8 (a) introduces mechanical energy recovery technology for both the liquid (U22)and the vapour (U32) phase, explicitly intends the depleted streams for conversion to heat and/or power (standard, catalytic and/or POX-GT combustion, and/or electrochemical conversion (e.g. fuel cells), U92) and balances the heat requirement with intermediate product. Figure 8(b) adds a second gas separation step (e.g. membrane stage, U52) after the bulk separation (e.g. an absorption column with a technical solvent, U53) at low pressure.

Figure 9 summarizes the inventive flowsheet alternatives without gas separation at high pressure. Compared to all other

alternatives, the entire crude product resulting from the

gasification is thereby (optionally) reheated and expanded (U32) without prior LV separation which allows for considerably

increasing the power co-generation or heat for refrigeration. LV separation (U42) and - as already outlined with Figure 8 - gas separation is done at low pressure in one or several steps of different technologies (e.g. absorption column with a technical solvent (U53) followed by a membrane stage (U52)) . Again, the depleted streams are explicitly intended for conversion to heat and/or power (standard, catalytic and/or POX-GT combustion, and/or electrochemical conversion (e.g. fuel cells), U92) and the heat requirement is balanced with intermediate product.

As an alternative to balancing the heat requirements of the processes by combusting part of the gasification product or SNG, the required heat may be supplied by an external source such as solar collectors similar to the ones used in solar thermal power plants, or from a thermal power plant or any industrial process. From the pinch analysis presented in Figure 3, the only condition for such an external heat input to be useful is that the heat is supplied above the pinch point, i.e. above 650-700K if the product is expanded without power recovery (Figure 3a) , or above 500 to 550K if the entire crude product is reheated and expanded in a turbine (Figure 3b) . If this is the case, less or no gas needs to be withdrawn from the product streams and combusted. Hence, the total SNG yield from the substrate is increased. Table 2 compares the overall energy balances without and with supplying additional heat from other sources than the combustion of part of the

gasification product. These values are calculated with the developed process model for the cases with and without LV separation. The data demonstrate that in any case, the use of additional heat from an external source is beneficial since it increases the net production of SNG without generating thermal losses. For this reason, all the external heat is converted into SNG and electricity and the total marginal efficiency (defined as the increase of the SNG and electricity production divided by the amount of additional heat supply) exceeds 100%. Further

considering that the limiting factor in the production is the availability of biomass, external heat supply thereby allows for increasing the yield of fuel and power from biomass from 58-62% to 85-94%.

Therefore, the feature of the supply of external heat (e.g. from solar collectors) to the process could be claimed as a depending claim to all the attached claims. If necessary, the heat exchange can be specified. If no power recovery from the vapour phase of the crude product is considered (i.e. Unit 32 configured as shown in Figure 10b), the heat must be supplied to the salt separator. If power recovery from the vapour phase of the crude product is considered (i.e. Unit 32 configured as shown in Figure 10a), the heat must be supplied to the salt separator and the heat exchanger upstream the turbine. The heat supply might be direct (a) or indirect (b) :

a) the salt separator is (somehow) placed in the focal point of solar heat collectors

b) the heat is transferred via a fluid (e.g. thermal oil, gas) that circulates between the salt separator and the external heat source

This application therefore summarises the systematic analysis of the process design alternatives for hydrothermal production of SNG from wet biomass and biomass waste. Based on a general

superstructure for combined product separation and internal energy recovery from the supercritical conditions, the possibilities for an efficient co-generation of SNG and power have been explored. Even with conservative hypothesis on practical design limitations, a sound process integration and energy recovery allows for an energetically and economically viable process design. Thermo- economic optimisations have revealed that the hydrothermal

conversion should thereby be regarded as a polygeneration system in which SNG and electricity yields are to a large extent on a par.

It is demonstrated that the process design and performance is not only influenced by available technology and catalyst deactivation, but also the characteristics of the processed substrate. Wet but energetically valuable industrial by-products with a high hydrogen and low ash content such as lignin slurries or coffee grounds have been identified as a particularly well suited feedstock that allow for greater efficiencies than wood. Biomass wastes with high ash content such as manure and digested sewage sludge are less

advantageous since their effective biomass content is severely reduced if processing is limited to slurries containing no more than about 20% total solids. Nevertheless, from the perspective of waste treatment with disposal as principal objective, also

marginal profit from a complete energy recovery from wastes might yet be valuable.

Nomenclature section

LV: liquid-vapour

MER: Minimum energy requirements NGCC : Natural gas combined cycle SNG: Synthetic natural gas

E: Electricity, MW

M: Mass flow, kg s ^"1

H: Heat, MW

LHV: Lower heating value, MJ kg ^_1 _daf EXV: Exergy value, MJ kg ^_1 _daf ε : Energy efficiency, %

n: Exergy efficiency, %

Φ: Biomass humidity, wt%

Subscripts

daf: dry, ash free