Experimental set–Up for amine solvent analysis

3.10.1. Experimental procedure

a)

A determined volume of aqueous amine solution at the desired concentration is loaded into the glass reactor and then immersed into the well-insulated heating bath. The heating oil in the bath provides the heat for the experiment at the desired temperature (usually 40 °C for CO₂ absorption). The level of the oil in the bath should cover at least three–quarters of the glass reactor (if the level of the sample in the reactor is at the half–way mark).

b)

After the amine solution in the reactor has reached thermal equilibrium for absorption (40 °C) process, the simulated flue gas (CO₂, N₂, O₂, H₂O, SOx, NOx, fly ash) is injected into the reactor at a determined flow rate. The CO₂ composition and the compositions of other impurities are also predetermined and should be constant throughout the experimental run. However, if impurities like SOx, NOx and fly ash are not available to be mixed with the simulated flue gas, they can be dissolved at the right concentration into the aqueous amine solution before the start of the experiment. Fly ash can be directly added to the amine solution while nitric acid (HNO₃), sulphurous acid (H₂SO₃) can also be added to the amine solution to represent NOx and SOx respectively. In this case, the simulated flue gas will only contain (CO₂, N₂, O₂ and H₂O). Aqueous solutions of metals like Copper(II) chloride (CuCl₂), Iron(III) chloride (FeCl₃), Calcium Oxide (CaO), Aluminium Oxide (Al₂O₃) can be added (at the desired ppm concentration) to the amine solution to represent the metal content of the flue gas [242]. Iron content can also be due to corrosion in the CO₂ capture plant. These impurities should be well mixed with the aqueous amine solution prior to starting the absorption experiment. Other impurities can also be added to the amine solution to represent flue gas components.

c)

A condenser is connected at the top exit of the reactor to limit amine losses through vaporization, which is achieved by using a cooling liquid set maximum at 7 °C.

d)

Once the simulated flue gas is flowing through the amine solution, time is allowed for the solution to reach equilibrium with CO₂. Equilibrium is attained when the dry CO₂ composition exiting the condenser remains constant for at least 10 min.

e)

Downstream of the condenser is a desiccant to dry any moisture content in the exiting vapor stream while the CO₂ analyzer or gas chromatography with thermal conductivity detector (GC–TCD) measures the amount of CO₂ in the dry vapor phase. Any of these instruments us used to confirm equilibrium.

f)

The dry gas stream can also be connected to Fourier transform infrared spectrometer (FTIR) for online quantification of vaporized amine solvent etc.

g)

Digital thermocouple (±0.1 °C accuracy) is used to measure the temperature of the amine solution and also the temperature of the wall of the reactor. A magnetic bar in the reactor provides the stirring during the experiment to ensure adequate mixing of the reacting components. The stirring speed (rpm) should be constant during all experiments and should not induce splashing on the inner wall of the reactor.

h)

During the absorption experiment the second oil bath should be heated to the desired temperature for desorption which should be representative of the actual regeneration temperature (110–120 °C). Slightly lower temperatures (95–105 °C) can also be used depending on the purpose of the experimental study, most especially for low temperature desorption investigation.

i)

At the end of the absorption experiment (equilibrium is attained) amine sample is taken for further analysis. The amine sample once taken is cooled in a refrigerator (about 10 °C) to stop any further reaction. Alternatively, the sample bottle containing the amine sample can be quenched in running cold water to bring down the temperature of the sample quickly for analysis. The analysis will cover CO₂ loading, free amine concentration and possible degradation products (NH₃, nitrosamine etc.). Then the reactor is transferred into the bath which is already at 95–120 °C for desorption process. For CO₂ loading, the amine sample must be analyzed right away to avoid further error.

j)

Desorption can be allowed to run for at least same time duration as the absorption experiment to allow for a significant change in free amine concentration (degradation and amine vaporization) to occur and be easily quantified. At the end of the desorption experiment the reactor containing the amine solution is sent to a refrigerator (about 10 °C) to cool the amine solution and stop further degradation reaction. The sample temperature can also be quenched in running cold water as described previously. After then amine samples can be taken to analyse for CO₂ loading, free amine concentration and possible degradation products (NH₃, nitrosamine etc.).

k)

Isokinetic sampling method can be used to trap all carryover species (like free amine, NH₃ and other degradation products) in the vapor phase [225], [226], [227], [228], [229], [230]. This method can also be used to quantify amine losses through vaporization. Considering the initial amine concentration at the start of the experiment, amine losses through vaporization can then be applied towards determining amine losses through degradation. When amine losses through vaporization or degradation products are to be analyzed and quantified in the off gas, the isokinetic set–up should be installed upstream of the dryer (absorption section) to avoid trapping the major products in the vaporized amine solvent. A dryer is not required during the desorption experiment. Also in this case, the temperature of the cooling liquid can be increased to 10–20 °C.

l)

On the other hand, due to the very low temperature and high flow rate of the cooling liquid flowing through the condenser, vaporization losses (amine and H₂O) can be assumed to be negligible (for simplicity).

m)

When salts of representative metals are added to the amine solution before CO₂ absorption, then their disappearance in the liquid phase can also signify degradation.

Since the experimental set–up is a semi–batch process Eq. (49) which based on Fourier’s law (heat transfer by conduction) can be used to determine the heat input during the desorption process which can be further used to quantify the heat of regeneration (heat duty) when divided by the amount of produced CO₂ as shown in Eq. (30). Any correlation for determining heat input can also be used.

(49)Hin=kA(dT)/d

where; Hi_n is the heat transferred to the amine solution (J/s), k is the thermal conductivity of the reactor (W/m ^oC or J/s.m. ^oC), A is the area of heat transfer (m²), dT is the temperature difference between the outer surface of the reactor and the amine solution in the reactor (°C) while d is the thickness of the reactor (m).

This proposed experimental procedure and analysis should be carried out at the maximum desorption/regeneration temperature (120 °C), thereby highlighting the actual condition of amine regeneration. Most studies [75], [196], [199], [200], [201], [202], [216], [218] investigated amine degradation at desorption temperatures (up to 170 °C) higher than the actual temperature used in the industry (120 °C). Results from such experimental condition can misrepresent the degradation rate of an amine solution.

Fig. 25 depicts degradation rates of two hypothetical amine solvents at various temperatures. It shows that at desorption temperature of 120 °C the degradation rates of Amine A and Amine B was 20% and 30% respectively while at 170 °C the degradation rates of Amine A and Amine B became 60% and 90% respectively. If the results at 170 °C are to be considered then Amine B will be seen as not applicable for CO₂ capture whereas results at 120 °C show that Amine B is competitive to Amine A.

1. Download : Download high-res image (60KB)
2. Download : Download full-size image

Fig. 25. Degradation rate and desorption temperature of two hypothetical amine solvents.