Optimal absorber conditions and absorber productivity

https://doi.org/10.1016/j.seppur.2021.118959

”

Table 8 shows the operating conditions that minimize the specific equivalent work of the overall capture process, $ω$ , for the most representative advanced configurations aiming at improving the energetic performance of NH₃-based capture processes. The RecVC has been included in all combinations of advanced configurations since it has shown a better performance than recycling the NH₃ and CO₂-rich stream recovered in the solvent recovery section to the CO₂ absorber. Subsequently, the RecVC has been combined with different designs of LVC configurations and with the MPD. Then, the VIS is always implemented in combination with the RecVC, when it is expected to be most efficient since the additional electrical work required by the vacuum pump can be compensated by reboiler duty reductions in the CO₂ desorber. In Table 8, the thermodynamic performance of the VIS is shown at two different working pressures in combination with the MPD and the RecVC, which is the combination that minimizes the specific equivalent work when using the IS. Configuration C2 using the IS (shown in Fig. 7) has been used as reference configuration for the assessment of advanced configurations that aim at improving the energetic performance of NH₃-based processes. Along with the minimum $ω$ value achieved for each configuration and the corresponding conditions of the process parameters, Table 8 also includes the results for other performance indicators such as consumption of chemicals, reboiler duties and temperatures, and electrical demand. Moreover, Table 8 also contains the results of the heat integration performance corresponding to the rich/lean heat exchanger, the NH₃-rich/NH₃-lean heat exchanger and the rich purge/lean purge heat exchanger. Such heat integration network allows to obtain the reboiler duties indicated in Table 8 for the CO₂ desorber, $q_{reb, {CO}_{2}}$ , and for the IS (or VIS), $q_{reb,IS}$ . All liquid/liquid heat exchangers are simulated as counter-current heat exchangers in which the pinch point temperature is set to 3 $° C$ . In the case of condensing vapour/boiling liquid heat exchangers above ambient temperature, such as the reboilers of the distillation columns and the heat exchanger introduced with the MPD, the pinch point temperature is set instead to 10 $° C$ [33], [39]. In the rich/lean heat exchanger, the hot CO₂-lean stream transfers heat, $q_{HTX,rich∕lean}$ , to the cold CO₂-rich stream, which amounts from two to four times the reboiler duty of the CO₂ desorber, depending on the process configuration and operating conditions. The heat transferred in the NH₃-rich/NH₃-lean heat exchanger, $q_{HTX,FG−WW}$ , from the hot NH₃-lean stream to the cold NH₃-rich stream, and in the rich purge/lean purge heat exchanger, $q_{HTX,IS}$ , from the hot lean purge stream to the cold rich purge stream, are of the same order of magnitude of the reboiler duty of the IS (or the VIS). Therefore, the proper design and simulation of such process heat exchangers is of paramount importance to find feasible reboiler duties in the distillation columns, i.e. CO₂ desorber and IS (or VIS), thus to determine the optimal set of operating conditions that minimize the energy demand of the capture process.

The RecVC (Configuration C2E1) is capable of reducing by 2.3% the minimum specific equivalent work of the capture process in comparison to the reference case in which the NH₃ and CO₂-rich stream recovered at the top of the IS is recycled to the CO₂ absorber. Such improvement is mainly obtained due to the decrease in $q_{reb, {CO}_{2}}$ achieved by means of the heat provided by the compressed NH₃ and CO₂-rich stream, which outgrows the increase in the electrical work demand required to compress such vapour stream to the pressure of the CO₂ desorber. The compression of the vapour distillate from the IS acts as a so-called “heat pump”, which increases its heat quality to the point of making it suitable to provide heat in the reboiler of the CO₂ desorber, at the cost of increasing the mechanical work of the process required for compression.

Table 8. Process operating conditions and performance indicators of the advanced configurations of the ${CO}_{2}$ desorber and of the solvent recovery section that allow for optimal operation in terms of minimum specific equivalent work requirements. The optimal ${CO}_{2}$ absorber conditions and ${CO}_{2}$ absorber productivity correspond in all cases to those of the mid ${\hat{c}}_{{NH}_{3}}$ case shown in Table 3. All cases have been simulated considering the reference configuration of the FG post-conditioning section operating at the optimal set of operating conditions (“Opt”) of the FG-WW column indicated in Table 6; the corresponding dimension of the FG-WW column can be found in Table 5. The pinch point temperature in the rich/lean, in the ${NH}_{3}$ -rich/ ${NH}_{3}$ -lean and in the rich purge/lean purge heat exchangers has been set to 3 $° C$ , while it has been set to 10 $° C$ in the reboilers and in the heat exchanger introduced with the MPD.

Variable	Configuration
Empty Cell	C2	C2E1	C2E2a	C2E2b	C2E2c	C2E2d	C2E3	C3E3
Empty Cell	IS	RecVC	LVC(4)	LVC(3)	LVC(4,3)	LVC(4,3,2)	MPD	VIS
Empty Cell	Empty Cell	(+IS)	(+RecVC+IS)	(+RecVC+IS)	(+RecVC+IS)	(+RecVC+IS)	(+RecVC+IS)	(+MPD+RecVC)
${CO}_{2}$ desorber parameters
$P_{{CO}_{2} des}^{top}$ [bar]	19.5	19.5	17.0	7.0	9.5	7.0	32.3	32.3	32.3
$P_{{CO}_{2} des}^{bot}$ [bar]	20.0	20.0	17.5	7.5	10.0	7.5	16.4	16.4	16.4
$f_{cr}$ [-]	0.0500	0.0500	0.0400	0.0600	0.0450	0.0425	0.0400	0.0400	0.0400
Solvent recovery section parameters
$N_{IS}$ [-]	9	13	13	13	13	13	13	13	13
$s_{{NH}_{3} - rich}^{IS}$ [-]	2	3	3	3	3	3	3	3	3
$F_{RecVC}$ [ $kg t_{{CO}_{2} captured}^{- 1}$ ]	0.0	62.3	63.4	63.4	64.2	65.2	61.4	64.0	71.8
$P_{IS}$ [bar]	1.05	1.05	1.05	1.05	1.05	1.05	1.05	0.35	0.10
Heat integration performance
$q_{HTX,rich∕lean}$ [ ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ ]	8.33	8.42	6.75	4.60	4.92	3.81	8.04	8.04	8.04
$Δ T_{c,rich∕lean}$ [K ]	4.1	4.3	3.5	4.6	3.7	3.2	3.5	3.5	3.6
$Δ T_{h,rich∕lean}$ [K ]	11.2	11.2	4.1	3.0	3.0	3.0	11.2	11.2	11.2
$x_{V,hot {CO}_{2} - rich}$ [mole frac.]	0.013	0.013	0.002	0.000	0.000	0.000	0.013	0.013	0.013
$q_{HTX,FG−WW}$ [ ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ ]	0.12	0.12	0.12	0.12	0.12	0.12	0.12	0.09	0.06
$Δ T_{c,FG−WW}$ [K ]	3.0	3.0	3.0	3.0	3.0	3.0	3.0	3.0	3.0
$Δ T_{h,FG−WW}$ [K ]	7.9	8.5	8.4	7.9	8.2	8.0	8.4	3.5	3.3
$x_{V,hot {NH}_{3} - rich}$ [mole frac.]	0.005	0.006	0.006	0.005	0.006	0.006	0.006	0.000	0.000
$q_{HTX,IS}$ [ ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ ]	0.02	0.02	0.02	0.02	0.02	0.02	0.02	0.01	N/A
$Δ T_{c,IS}$ [K ]	39.2	41.9	39.2	40.9	38.9	38.2	45.4	22.7	N/A
$Δ T_{h,IS}$ [K ]	3.0	3.0	3.0	3.0	3.0	3.0	3.0	3.0	N/A
$x_{V,hotpurge}$ [mole frac.]	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	N/A
Consumption of chemicals
$F_{H_{2} O}^{in}$ [ $kg t_{{CO}_{2} captured}^{- 1}$ ]	9.14	7.30	12.41	14.03	17.28	22.46	2.86	2.96	3.03
$F_{chem, {NH}_{3}}^{in}$ [ $kg t_{{CO}_{2} captured}^{- 1}$ ]	1.46	1.44	1.47	1.46	1.49	1.60	1.43	1.43	1.41
Energetic performance
$q_{reb, {CO}_{2}}$ [ ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ ]	2.19	2.03	1.94	1.96	1.91	1.79	1.86	1.84	1.81
$T_{reb, {CO}_{2}}$ [ $° C$ ]	143.9	143.9	134.7	117.7	121.3	115.4	139.1	139.1	139.1
$q_{reb,IS}$ [ ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ ]	0.17	0.15	0.15	0.15	0.15	0.15	0.14	0.14	0.16
$T_{reb,IS}$ [ $° C$ ]	100.1	100.1	100.1	100.1	100.1	100.1	100.1	72.3	45.7
$e$ [ ${MJ}_{el} {kg}_{{CO}_{2} captured}^{- 1}$ ]	0.287	0.321	0.371	0.483	0.453	0.557	0.366	0.386	0.417
$ω$ [ $MJ {kg}_{{CO}_{2} captured}^{- 1}$ ]	1.042	1.018	1.009	1.070	1.038	1.091	0.994	0.998	1.013
$ω$ savings [%]	N/A	2.3	3.2	−2.7	0.4	−4.7	4.6	4.2	2.8

When also introducing the LVC with expansion of the CO₂-lean stream to the inlet pressure to the fourth (last) stage of the multi-stage compressor introduced with the RecVC (Configuration C2E2a), the minimum specific equivalent work can be further decreased up to 3.2% with respect to the reference configuration, i.e. Configuration C2. While the greater vapour flowrate to be recycled to the CO₂ desorber increases the electrical work demand, this is outgrown as for Configuration C2E1, with RecVC but without the LVC, by the decrease of the reboiler duty required for solvent regeneration in the CO₂ desorber. In the case of Configuration C2E2a, such additional reboiler duty decrease is not only achieved by the additional flowrate of hot vapour recycled to the bottom of the CO₂ desorber, but also by the additional equilibrium stage provided by the CO₂-lean stream expansion, which allows to further separate CO₂ from NH₃ (and water) thus allowing for higher CO₂ loadings in the liquid exiting the reboiler of the CO₂ desorber, thus decreasing the requirements for the thermal separation. The expansion of the liquid exiting the reboiler of the CO₂ desorber decreases the temperature of the hot CO₂-lean stream, which lowers the amount of heat exchanged in the rich/lean heat exchanger and thus leads to lower temperature profiles along the CO₂ desorber, including lower temperature levels in the reboiler. The lower temperature levels along the column contributes to a better control of the NH₃ slip in the CO₂ stream exiting the top of the CO₂ desorber, as it can be deduced from the lower cold-rich bypass split fraction required with respect to Configuration C2E1 even at lower pressures of the CO₂ desorber. On the contrary, the lower optimal $P_{{CO}_{2} des}$ value obtained for Configuration C2E2a has the opposite effect, which overall leads to higher NH₃ concentrations in the CO₂ stream exiting the top of the CO₂ desorber thus to greater water make-up flowrates as a consequence of the greater water make-up flowrate required in the CO₂-WW section. Such increase in $F_{H_{2} O}^{in}$ when introducing the LVC, which is minor if compared with the process water requirements of the advanced configurations of the FG post-conditioning section using the SNA and the ANA, is allowed because it has a negligible effect on the reboiler duty of the IS. In fact, $q_{reb,IS}$ is mainly governed by variations in the greater flowrates of the NH₃-rich stream and of the purge stream, and not by small variations of water flowrate from the CO₂-WW section, as it can be inferred from the results shown in Fig. 11(a). The trends described for Configuration C2E2a including the LVC(4) can be extended for advanced configurations in which the CO₂-lean stream exiting the bottom of the CO₂ desorber is expanded to lower pressures either in one stage, as for Configuration C2E2b with the LVC(3), or in two or three stages, as for Configuration C2E2c with the LVC(4,3) and Configuration C2E2d with the LVC(4,3,2), respectively. Nevertheless, none of the LVC configurations with expansion to lower pressures decreases the specific equivalent work of the overall capture process with respect to Configuration C2E2a. Although expanding the CO₂-lean stream to lower pressures further decreases $q_{reb, {CO}_{2}}$ , its effect on the overall specific equivalent work is compensated or outgrown by the increase in the electrical work required to compress increasing flowrates of vapour to the pressure of the CO₂ desorber; while the thermal energy provided to the bottom of the CO₂ desorber increases, lower expansion pressures also decreases the distribution coefficient of CO₂ with respect to NH₃ in the recycled vapour stream, which approaches 1 at atmospheric pressure, thus worsening the separation of CO₂ from NH₃ (and H₂O) in the expansion stages of the LVC.

As far as the MPD is concerned, Configuration C2E3 decreases the specific equivalent work of the capture process by 4.6% with respect to the reference Configuration C2, which doubles the exergy savings achieved when only using the IS with the RecVC and that outperforms the energetic performance of Configuration C2E2a with the LVC(4). As for the RecVC and for the LVC, the improvement in the energetic performance of the capture process is triggered by a decrease of the reboiler duty of the CO₂ desorber, which reaches values as low as 1.86 ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ at conditions that minimize the specific equivalent work of Configuration C2E3. Similarly, such decreases in the $q_{reb, {CO}_{2}}$ values are obtained at the cost of increasing the electrical work required for the compression of the vapour exiting the top of the lower section of the CO₂ desorber to the higher pressure of the upper section of the column. The thermal energy gained by such compressed vapour stream is used to heat up the CO₂-rich stream before it enters the CO₂ desorber and to provide thermal energy for further CO₂ stripping in the upper section of the CO₂ desorber. At minimal specific equivalent work, the compressed vapour of the MPD adds 0.20 ${MJ}_{th} {kg}_{{CO}_{2} captured}^{- 1}$ to the CO₂-rich stream and still enters the upper section of the CO₂ desorber at 137.9 $° C$ . In addition, the partial condensation of the compressed vapour stream before entering the bottom of the CO₂ desorber upper section also allows for an additional equilibrium stage for the separation of CO₂ from NH₃ (and H₂O). The higher pressure in the upper section of the CO₂ desorber facilitates the removal of NH₃ from the CO₂ stream and allows to decouple the reboiler duty minimization from the control of the NH₃ slip in the exiting CO₂ stream. Therefore, the optimal $f_{cr}$ value in this case maximizes the heat exchanged in the rich/lean heat exchanger without increasing the NH₃ slip in the CO₂ stream thus the water make-up flowrate to the CO₂-WW section and the reboiler duty of the IS. These facts are confirmed by the results of Configuration C2E3 with the MPD shown in Table 8, where: (i) $q_{HTX,rich∕lean}$ reaches similar values to those obtained for configurations C2 and C2E1 at lower CO₂ desorber reboiler pressures thus for lower temperatures of the hot CO₂-lean stream; (ii) the optimal value of $f_{cr}$ that minimizes the overall specific equivalent work is lower than for the same configuration without the MPD, i.e. Configuration C2E1, even when $P_{{CO}_{2} des}^{bot}$ is decreased from 20.0 to 16.4 bar—the optimal $f_{cr}$ value is the same as for Configuration C2E2a with the LVC(4) but Configuration C2E3 also operates at lower $P_{{CO}_{2} des}^{bot}$ ; and, (iii) $F_{H_{2} O}^{in}$ is decreased with respect to Configuration C2E1 without the MPD, resulting from a better control of NH₃ emissions in the upper section of the CO₂ desorber. As a consequence, the MPD achieves lower $q_{reb, {CO}_{2}}$ values than the LVC(4) at optimal process operating conditions that minimize the specific equivalent work of the full capture process. Nevertheless, when comparing the results obtained with configurations C2E2a and C2E3 shown in Table 8, one can notice that the superior energetic performance of the MPD in comparison with the LVC(4) derives also from a lower increase of the electrical work demand; in the case of the MPD, the higher pressure at which the upper section of the CO₂ desorber operates allows to decrease the CO₂ compression work.

Regarding the VIS, it mainly affects the performance of the solvent recovery section, where the reboiler temperature of the stripping column decreases from 100.1 to 72.3 and to 45.7 $° C$ when decreasing the pressure from 1.05 to 0.35 and to 0.10 bar, respectively, without affecting significantly the reboiler duty values. The optimal process operating conditions that minimize the specific equivalent work of the capture process and the remaining performance indicators are neither affected significantly. The exception is the rich purge/lean purge heat exchanger, which cannot be implemented for the lower $P_{IS}$ value due to the fact that the rich purge stream has higher temperature than the NH₃-lean stream exiting the bottom of the VIS. Although low temperature excess (waste) heat might be available in the CO₂ point source for integration with the capture plant, the specific equivalent work computations provided in Table 8 only consider thermodynamic criteria and do not account for application scenarios. Therefore, operating the IS at vacuum conditions does not allow to decrease the specific equivalent work of the capture process since the increase in the electrical work demand driven by the vacuum pump, whose efficiency decreases for lower pressures, is not compensated by the decrease in the specific equivalent work associated with the reboiler duty of the VIS.

Something common to all advanced configurations presented in Table 8 is that they do not modify significantly the flowrate of aqueous NH₃ solution make-up, $F_{chem, {NH}_{3}}^{in}$ , with respect to the reference Configuration C2. Therefore, the advanced configurations developed here to improve the energetic performance of the capture process do not affect the performance of the FG post-conditioning section and could also be combined equally with any of the advanced configurations of the FG post-conditioning section developed in this work.

At this point, it is worth making a clarification about the simulation of the heat exchangers in Aspen Plus. As aforementioned, the proper simulation and design of such heat exchangers is of paramount importance to obtain optimal, at the same time feasible, reboiler duties thus energetic performances of the capture process. As far as the energetic performance of the capture process is concerned, the most critical heat exchanger is the rich/lean heat exchanger, which strongly affects the CO₂ desorber reboiler duty, the main energy consumer of the capture process. When the RSS is in place, the flowrate of the CO₂-rich stream flowing through the rich/lean heat exchanger, i.e. the cold stream, might be smaller than the flowrate of the CO₂-lean stream, i.e. the hot stream. As a consequence, the heat capacity of the former stream might be smaller than the heat capacity of the latter, so that the pinch point temperature might be reached at the hot side of the heat exchanger, instead of at the cold side as in the case of a configuration without RSS. Nevertheless, the higher CO₂ concentration in the CO₂-rich stream may lead to partial vaporization before leaving the rich/lean heat exchanger when approaching the temperature of the hot CO₂-lean stream, as confirmed by the results of molar vapour fraction of the hot CO₂-rich stream exiting the rich/lean heat exchanger, $x_{V,hot {CO}_{2} - rich}$ , given in Table 8. Consequently, the heat transferred to the CO₂-rich stream towards the exit of the rich/lean heat exchanger is used as latent heat and not as sensible heat to increase the temperature of the fluid. Therefore, the pinch point temperature in the rich/lean heat exchanger when using the RSS might be reached internally, instead of being reached at the cold or at the hot side as the values shown in Table 8 for $Δ T_{c,rich∕lean}$ and $Δ T_{h,rich∕lean}$ , respectively, confirm. Therefore, the rich/lean heat exchanger has been simulated in Aspen Plus, Version 8.6, by means of a “MHeatX” exchanger block divided in zones, which allows to identify the internal pinch point temperature. On the contrary, setting the pinch point at the cold side of the rich/lean heat exchanger when the RSS has been implemented leads to infeasible heat exchanger designs and CO₂ desober reboiler duties, thus to misleading optimal process operating conditions and energetic performances of the capture process, as in the case of Jiang et al. [20]. Design considerations for other critical heat exchangers where the pinch point temperature has been set to 3 $° C$ , i.e. in the NH₃-rich/NH₃-lean heat exchanger and in the rich purge/lean purge heat exchanger of the solvent recovery section, have been introduced in Section 4.2.1 when describing the advanced Configuration C2 using the IS. The heat integration performance results given in Table 8 for such heat exchangers confirm the avoidance of temperature crossovers or pinch point temperatures below 3 $° C$ . As far as the NH₃-rich/NH₃-lean heat exchanger is concerned, it is similarly simulated by means of a “MHeatX” exchanger block in Aspen Plus with zones division in order to be able to predict the internal pinch point temperature, if needed. Nevertheless, the heat integration performance results given in Table 8 show that the pinch point temperature in the NH₃-rich/NH₃-rich heat exchanger is always reached at the cold side, i.e. $Δ T_{c,FG−WW}$ . Since the goal of the NH₃-rich/NH₃-rich heat exchanger is not only to minimize the reboiler duty in the IS (or VIS), but also the chilling demand of the NH₃-lean stream before entering the FG-WW column, purging a fraction of the hot NH₃-lean stream before entering the NH₃-rich/NH₃-lean heat exchanger aims at equalling the heat capacities of the cold and the hot stream such that the temperature difference between the hot stream and the cold stream remains constant along the counter-current heat exchanger. Nevertheless, the hot side temperature difference increases due to the vaporization of the NH₃-rich stream when approaching the temperature of the hot NH₃-lean stream, as shown by the results of $Δ T_{h,FG−WW}$ and $x_{V,hot {NH}_{3} - rich}$ given in Table 8. Regarding the rich purge/lean purge heat exchanger in the solvent recovery section, it has been simulated in Aspen Plus by means of a “HeatX” exchanger block, setting the pinch point temperature at the hot side, $Δ T_{h,IS}$ , since the heat capacity of the rich purge stream is always smaller than that of the lean purge stream and the former stream is always at pressures equal or above 7 bar that avoid its partial vaporization within the heat exchanger thus a temperature crossover.

“

Cookie	Duration	Description
cookielawinfo-checkbox-analytics	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Analytics".
cookielawinfo-checkbox-functional	11 months	The cookie is set by GDPR cookie consent to record the user consent for the cookies in the category "Functional".
cookielawinfo-checkbox-necessary	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookies is used to store the user consent for the cookies in the category "Necessary".
cookielawinfo-checkbox-others	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Other.
cookielawinfo-checkbox-performance	11 months	This cookie is set by GDPR Cookie Consent plugin. The cookie is used to store the user consent for the cookies in the category "Performance".
viewed_cookie_policy	11 months	The cookie is set by the GDPR Cookie Consent plugin and is used to store whether or not user has consented to the use of cookies. It does not store any personal data.

Chunfei Wu

Follow:

Optimal absorber conditions and absorber productivity

Leave a Comment Cancel Reply

Useful Links

Contact Us

Phone: +44 (0)28 9097 5573

Email: c.wu@qub.ac.uk