Results from A1 - DEF

Concerning Action A1, the activities for DEF were subdivided as it follows:

• Sub-action A1.1_DEF Selection of alkaline compounds through a literature review

• Sub-action A1.2_DEF: Simulation of absorption with alkaline compounds

• Sub-action A1.3_DEF: Preliminary test runs of absorption pilot plant.

• Sub-action A1.4_DEF: Technical report preparation.

Sub-action A1.1_DEF Selection of alkaline compounds

Among the existing methods for CO2 removal from gaseous streams (Olajire, 2010) (Zhao et al., 2010) (IEA; 2007) (IPCC, 2005), absorption with alkali compounds was preliminary identified as the most suitable one, in order to realize a regenerative process, aimed to the final capture of the CO2 in solid form, through the carbonation of solid residues.

The aim of the alkali absorption process is to produce a load solution in which CO2 is transformed in the form of carbonate ions (CO3--) or bicarbonate ions (HCO3-) (Kohl and Nielsen, 1997), so that such ions can react with available calcium in the solid residues selected for regeneration, in order to form mainly calcium carbonate (CaCO3), regenerating the original compound of the absorbing solution.

The selected alkali compounds for the reaction were: Potassium Hydroxide (KOH), Sodium Hydroxide (NaOH) and Sodium Carbonate (Na2C O3) (Kohl and Nielsen, 1997) (Corti et al., 2001) (Corti, 2004).

Sodium Hydroxide and Potassium Hydroxide are occasionally used to remove the last traces of carbon dioxide from hydrogen, or other gases, where the bulk of the carbon dioxide has been removed by a more economical, but less efficient regenerative process (Kohl and Nielsen, 1997). Caustic scrubbing is also used to remove CO2 from small volumes of air where CO2-free air is required.

The CO2 absorption using aqueous solution of KOH was applied by Altair Chimica S.p.A. in their industrial plant in Saline di Volterra (Pisa, Italy). The plant produces several chemical products, and potassium carbonate (K2CO3) is one of them. In the conventional process, potassium carbonate is produced by reaction between potash (KOH) and carbon dioxide (CO2). An alternative production module has also been implemented, producing K2CO3 by using the CO2 present in combustion exhaust gases as reactant (along with KOH). The CO2 capture efficiency ranged from 83% to 97% (Carnevale et al., 2007). According to this process, the feasibility of the absorption process with KOH, applied to remove CO2 contained in the landfill gas was demonstrated in the frame of the Project LIFE05 ENV/IT/000874. (Lombardi et al. 2007; Carpentieri et al., 2008; Lombardi et al., 2008 a; Lombardi et al. 2008 b; Lombardi and Corti 2009).

When an aqueous solution of Sodium Carbonate (Na2CO3) is used, the following reaction applies:

Na2C O3 + CO2 + H2O = 2 Na+ + 2 HCO3-

2 H2O <--> H3O+ + OH-

CO2 + 2 H2O <--> H3O+ + HCO3-

HCO3- + H2O <--> H3O+ + CO3--

NA2CO3 --> 2 NA+ + CO3--

When an aqueous solution of Sodium or Potassium Hydroxide (NaOH or KOH) is used, the following reactions apply:

2 H2O <--> H3O+ + OH-

CO2 + 2 H2O <--> H3O+ + HCO3-

HCO3- + H2O <--> H3O+ + CO3--

NAOH --> NA+ + OH- KOH

KOH --> K+ + OH-

A KOH/NaOH excess leads to an increased production of potassium/sodium carbonate

2 KOH + CO2 --> K2CO3 + H2O

or

2 NaOH + CO2 --> Na2CO3 + H2O

while a CO2 excess leads to the following reactions:

K2CO3 + CO2 --> 2 KHCO3

or

Na2CO3 + CO2 --> 2 NaHCO3

Sub-action A1.2_DEF Simulation of absorption with alkaline compounds

The simulation of the absorption step of the CO2 contained in the landfill gas by means of aqueous solution of alkaline compounds, was carried out using a commercial chemical and thermodynamic simulation engine.

Preliminarly it was necessary to clearly identify which are the targets of CO2 removal efficiency that are to be reached by the process, that, in the specific case of landfill gas/biogas gas upgrading are concerned with pre-defined requirement in terms of quality of the exiting gas

Up-grading targets

There is no international technical standard for biogas quality before its injection into the grid.

Some countries have developed national standards and procedures for biogas injection (Persson et al., 2006). For example, some countries like Sweden, Switzerland, Germany and France have standards for injecting biogas into the natural gas grid. The standards have been set to avoid contamination of the gas grid or end use.

Quite often demands on the Wobbe Index (have been set to avoid influence on gas measurements and end use. In the standards there are limits on certain components for instance sulphur, oxygen, particles and water dew point, whereas a minimum methane purity of around 96% is required. These demands are in most cases possible to achieve with existing upgrading processes. As a matter of fact the existing national standards are quite equivalent among themselves, a part from slight differences.

In this work it was chosen to follow the German National Standards, in particular concerning the value of the Wobbe Index (Higher Wobbe Index HWI=46,1-56,5 MJ/Nm3, that corresponds to % vol. of CH4 >97,5%; Lower Wobbe Index LWI=37,8-46,8 MJ/Nm3, that corresponds to % vol. of CH4 between 87-96,5%) and the relative density (RD=0,55-0,75).

As a matter of fact, the French regulation considers more stringent limits for the lower bounds of the HWI and LWI ranges, being those values equal respectively to 48,24 and 42,48. This fact will be taken into consideration in the discussion of the results.

Single stage up-grading process simulation

The preliminary schematic for the simulation of the absorption process is quite simple and based, as a preliminary approach, on a single stage absorption column, followed by the upgraded gas cooling down to ambient temperature (25 °C), in order to remove the water eventually evaporated during the absorption process. As a matter of fact, the selected absorption reactions are characterized by being exothermal, especially in the case of KOH and NaOH, with a subsequent increase in the solution/gas temperature at the exit from the column. For this reason, in a hypothetical industrial scale process, it is necessary to cool down the upgraded gas, to condense the evaporated water, in order to obtain an acceptable volumetric fraction of methane.

From the simulation results it was evident how the aqueous solution of Na2CO3 are very poorly reactive with respect to KOH and NaOH. For this reason, this compound will be neglected in the following and the attention will be focused only on KOH and NaOH.

Two and three stages up-grading process simulation

As it is better illustrated in the paragraph devoted to the preliminary results obtained by UNIROMA, a limiting factor for the alkali concentration in the absorbing solution exists. As a matter of fact at increasing concentration the yield of regeneration decreases for the combination of two negative effects: decrease in the chemical regeneration and increase in the solution losses in the solid separation process after regeneration. UNIROMA concluded that a maximum of 4 eq./liter of carbonates in the solution to be regenerated is acceptable. Hence, a maximum of 4 eq./liter of alkali in the absorbing solution is acceptable, as well.

For this reason the possibility of working in the absorption process with lower alkali concentration was investigated. Of course, in order to keep high CO2 removal efficiency, high CH4 volumetric fraction and to respect the assumed limits for the Wobbe Index and the relative density – being the dimension of the pilot column appropriate for a maximum solution flow rate of 60 l/h - it is necessary to start to work with more than one absorption stage.

On an hypothetical industrial scale, the possibilities are two: increase the solution flow rate with respect to the gas flow rate, with an appropriate sizing of the column; keep similar solution/gas flow rate ratio and repeat the absorption process more than one time (more than one stage).

In the following the possibility of working with two or three absorption stages is investigated, relatively to the use of KOH and NaOH.

In particular the considered constraints are related to the maximum acceptable concentration of KOH and NaOH in the solution, that will be about 2-3 eq./liter, to keep within the range indicated by the preliminary tests executed by UNIROMA. In particular, the additional simulations were carried out with the aim of finding the concentration conditions which allows reaching the target values of Wobbe Index and relative density, when a two or three stage configuration is applied.

The solution mass flow rate assumed for the simulation is 60 l/h entering the two or three columns. The gas entering the first absorption stage is at the same conditions assumed before (flow rate of 20 Nm3/h, 40 °C and 1,03 bar; composition 50% vol. CH4 and 50% vol. CO2).

From the simulation results, in the case of using KOH, it is possible to comply with the required values of Wobbe Index and relative density, if the KOH concentration in the first, second and third stage is 14% in mass, and the solution mass flow rate in the first, second and third stage is 60 l/h.

In the case of using NaOH, it is possible to comply with the required values of Wobbe Index and relative density, if the KOH concentration in the first, second and third stage are respectively 11, 11 and 12 % in mass, and the solution mass flow rate in the first, second and third stage is 60 l/h.