Evaluating the Performance of Micro-Encapsulated CO2 Sorbents during CO2 Absorption and Regeneration

[NNadir]

The paper I'll discuss in this text is this one: Evaluating the Performance of Micro-Encapsulated CO2 Sorbents during CO2 Absorption and Regeneration Cycling (Joshuah K. Stolaroff et al Environ. Sci. Technol., 2019, 53 (5), pp 2926–2936)

One of the co-authors of this paper is Dr. Joan Brennecke, of the University of Notre Dame. Over a year back, I posted a link to one of her lectures at Princeton University on the subject of phase change ionic liquids for the capture of carbon dioxide:

On the Solubility of Carbon Dioxide in Ionic Liquids.

More recently, I discussed a class of "ionic liquids," these are salts that are liquids at room temperature, most often composed of organic cations and anions, although sometimes one of the ions is inorganic in nature - a common example being the hexafluorophosphonium anion, that was being utilized for the recovery of precious metals from spent automotive catalytic converters:

Ionic Liquid Based Separation of Precious Metals from Spent Automotive Catalytic Converters.

It is a remarkable reflection on the flexibility of ionic liquids that the cation used to recover palladium and rhodium from automotive catalysts is similar to that utilized to capture carbon dioxide. The carbon dioxide capture phase change ionic liquids are designated IL NDIL0309, and IL NDIL0230. The cations in these species are tetralkyl phosphonium ions, as they are in the catalytic metal capturing, although the anions differ, in the later case being chloride ion, and in the former, imide type anions of aromatic species.

There is also a discussion of CO2-BOLs (CO2 Binding Organic Liquids) and a cyclic organic tetramine known as "cyclen" (1,4,7,10-tetraazacyclododecane).

The idea in the paper is to encapsulate these agents to improve their performance.

From the introduction to the paper:

Global carbon dioxide emissions are projected to continue increasing in the near term, with coal-consuming countries such as China and India contributing to the expected growth over the next several years.(1) The rate of increase in anthropogenic CO2 emissions more than doubled in the period from 2000–2014, to 2.5–2.7% per annum, relative to the 1.1% per annum increase in the 1990–1999 period.(2,3) The concentration of CO2 in the atmosphere now exceeds 400 ppm, the highest it has been in 670 000 years.(4) The increasing global anthropogenic CO2 emissions present a challenge to meet the international target of <2 °C increase in global temperature relative to preindustrial levels. Thus, investigating carbon capture, storage, and utilization strategies is critical to mitigating CO2 emissions and maintaining <2 °C global temperature increase.

Globally, the combustion of fossil fuels accounts for 75% of anthropogenic CO2 emissions.(2) In 2016, 35% of the U.S. energy-related CO2 emissions came from the electric power sector fed by fossil fuels.(5) One strategy for CO2 emission mitigation is postcombustion capture of CO2 from point-sources such as power plants. The conventional approach to CO2 capture from flue gas is amine-scrubbing using aqueous solutions of amines such as monoethanolamine (MEA) or diethanolamine (DEA), which is already used in a number of plants but suffers from high energy demand, mostly to regenerate the amine solvent, as well as degradation of the amine solvent and the formation corrosive degradation products.(6,7) These challenges limit the economic feasibility of postcombustion CO2 capture from powers plants. Therefore, developing novel CO2 sorbents and processes that reduce the energy requirement and/or increase the CO2 absorption rate is a vital area of research to meet CO2 emissions and global temperature goals...

Nothing I say in this post, said with the deepest respect for Dr. Brennecke, should be construed as indicating that I find carbon capture from the "flue gas" of dangerous fossil fuel power plants acceptable. This said, I do regard capture of carbon dioxide from oxyfuel closed combustion of biomass as to be one option to remove carbon dioxide from the atmosphere. Thus the basic technology attracts me, with the expectation that it can be utilized for something else besides putting lipstick on the fossil fuel pig.

I don't believe in "flues" exposing combustion products to the air we breath, but believe strongly that closed combustion is technologically feasible, particularly in light of developments in materials science. This is only possible if we can quantitatively capture carbon dioxide and chemically modify it in useful ways with the effective removal of the gas from our current modern favorite waste dump, the planetary atmosphere.

Anyway, another excerpt from the introduction:

Micro-Encapsulated CO2 Sorbents (MECS) are a recently developed CO2 capture technology that may be an effective strategy for enabling the use and recyclability of advanced solvents, but they have not been rigorously evaluated for practical use within an absorption/regeneration process until now. MECS consist of a CO2-absorbing solvent or slurry encased in spherical, CO2-permeable polymer shells.(15,16) We have demonstrated this technology with carbonates, CO2BOLs, and ionic liquids to achieve increased surface area, resulting in an enhancement of CO2 absorption rates by an order of magnitude relative to a thin film of the solvent.(16) CO2 promoters such as carbonic anhydrase,(17) sarcosine,(18) or zinc(II) cyclen complexes(19,20) may be combined with carbonate solutions in MECS to further increase the rate of CO2 absorption. We have previously explored the addition of zinc(II) cyclen, a mimic of the enzyme carbonic anhydrase that catalyzes CO2 absorption into aqueous solution, to carbonate MECS to enhance the rate of CO2 absorption and found that the capsules reached CO2 saturation 2–3 times faster than carbonate MECS without cyclen.(15) Here, we rigorously demonstrate successful encapsulation, compare absorption rates and capacities of CO2 at low partial pressure, and evaluate stability of six advanced liquid CO2 solvents within polymer shells: Na2CO3 solution (uncatalyzed, with sarcosine, with cyclen); ionic liquids (task specific ionic liquid NDIL0230, and phase-changing ionic liquid NDIL0309), and a CO2BOL (Koechanol). We build upon our previous encapsulation tests to characterize MECS performance over a wider range of temperatures, CO2 loadings, and absorption/regeneration cycles.

The preparation of MECS is described here:

Encapsulated liquid sorbents for carbon dioxide capture (Jennifer Lewis et al, Nature Communications volume 6, Article number: 6124 (2015))

The experiments herein did not involve actual flue gases, but rather carbon dioxide in the presence of water. The apparatus utilized is shown here:



The caption:

Figure 1. Process diagram of the pressure drop apparatus used to measure CO2 absorption rates as a function of time for all six MECS types.

Pressure drop was followed using the following equation to determine the mass uptake over an hour in this apparatus:



This is obviously derived from the "ideal gas law." Dr. Brennecke is the editor of the Journal of Chemical and Engineering Data where she has set a high (and appropriate) standard for precision and accuracy in data. The ideal gas law is very imprecise, but this can be forgiven here, since probably the paper is intended for qualitative analysis.

Many higher accuracy equations of state for gases exist, either for generalized gases with fitting parameters such as acentric factors for individual gases, the Peng-Robinson equation for example.

Carbon dioxide itself has a very highly precise equation of state known as the Span-Wagner equation:

A New Equation of State for Carbon Dioxide Covering the Fluid Region from the Triple‐Point Temperature to 1100 K at Pressures up to 800 MPa (Roland Span and Wolfgang Wagner Journal of Physical and Chemical Reference Data 25, 1509 (1996); https://doi.org/10.1063/1.555991)

Although modern computers can certainly address the use of cubic equations like Peng Robinson and more complex formulations like Span-Wagner, again, it's probably overkill here.

Using these microencapsulating systems, these carbon dioxide capture agents were evaluated through a number of capture and release cycles. The release, of course, requires heat, one can imagine circumstances where this heat might come from very high temperature devices in a kind of "combined cycle" approach during cool down phases, reducing exergy losses to entropy.

The regenerating device is shown here:




The caption:

Figure 2. Process diagram of regeneration in a tube furnace (A), where CO2 was stripped from all six MECS types during preliminary testing. (B) Process diagram of the regeneration apparatus used to strip CO2 using a dry stream of N2 at ∼90 °C during the 10-cycle experiments (MFC = mass flow controller, TC = temperature controller, TI = temperature indicator, CO2 = CO2 sensor, RH = relative humidity sensor, and MFI = mass flow indicator).

Upon the conclusion of these experiments, they were repeated using "simulated coal flue gas"

...We have previously measured the gas permeability of the MECS shell materials to nitrogen and CO2 separately, shown in Table 1, but had not investigated the effect of a mixed gas stream on the gas absorption rate of the MECS until now. The selectivity of a membrane between a pair of gases is given by the following:

(3)

in which PA is the permeability of the more permeable gas, and PB is the permeability of the less permeable gas.(22) In the case of both types of silicone shells used in the MECS, the selectivity for pure CO2 over pure N2 is ∼11, which is on the low end compared to other materials reported in literature.(23,24)Previous research suggests that the presence of nitrogen can reduce CO2 permeability in silicones that have similar molecular structure to the MECS shell materials,(25) which would be detrimental to the rate of CO2 absorption from coal flue gas in which nitrogen makes up ∼75% of the gas stream. In fact, Scholes et al. reported that the CO2/N2 selectivity in PDMS decreased from ∼11 with pure gases to ∼3–4 with a blend of 10% CO2, 90% N2.(26) This indicates that the CO2 absorption rate may be limited by mass transfer of CO2 to the solvent since the selectivity for CO2 is decreased in the presence of N2.


Permeability is related to molar flux, J, by the following:

(4)

where d is the thickness of the membrane, or in this case the shell thickness, and Δp is the pressure difference across the membrane. Combining eq 4 with eq 3, the ratio of fluxes is proportional to the partial pressure gradient across the shell, given by eq 5:

(5)

In order to test whether the CO2 absorption rate was affected by the presence of air, CO2 absorption tests were performed in the pressure drop chamber with 0.1 bar of injected CO2 in both cases and an additional 0.9 bar of ambient atmosphere (79% N2) injected in the case with air. The resulting total gas absorption for these two cases are compared in Figure 3 over the first 10 min. It should be noted that in the case of the experiment with air, the gas absorption includes both air (predominantly N2) and CO2 across the shell material...

The authors expected some variation with air but didn't see it:



The caption:

Figure 3. (A) Gas absorption versus time for uncatalyzed Na2CO3 MECS exposed to 0.1 bar CO2 with and without 0.9 bar air in the pressure drop apparatus at 25 °C. The loading was practically the same in pure CO2 as in CO2 with air. (B) Model calculation of CO2 mole ratio (including CO2(aq) and HCO3– ions) of CO2 and N2 in MECS (shell and core) in 0.1 bar CO2 and 0.9 bar N2 condition as a function of CO2 loading capacity. The ratio quickly approaches 100%, meaning N2 contribution to the flux and loading may be assumed to be negligible.

There is some discussion of the reasons for this.

Some of the CO2-BOLs did not stay encapsulated. This was the case for a rather interesting material known as koechanol (1-((1,3-dimethylimidazolidin-2-ylidene)amin)propan-2-ol) the synthesis of which is described here:

Low viscosity alkanolguanidine and alkanolamidine liquids for CO2 capture (David J. Heldebrant et al, RSC Adv., 2013, 3, 566-572)

Here's a photograph of the leaking koecheanol along with the effect of the leaking on cycling the microcapsules through capture and regeneration cycles:



The caption:

Figure 4. Microscope images of Koechanol CO2BOL MECS after exposure to desiccant for 4 days before cycling, with visible droplets on the shell surface. The mass of Koechanol MECS decreases dramatically with each absorption/regeneration cycle (right).

The MECS were tested with some test solutions, "promoters" without the ionic liquids or CO2-BOLs.




The caption:

Figure 5. Comparison of CO2 absorption vs time for the three types of Na2CO3 MECS: uncatalyzed MECS, MECS with sarcosine promoter, and MECS with cyclen promoter. All of these MECS contained 17 wt % Na2CO3 and were soaked in 17 wt % Na2CO3 solution prior to testing. All data were collected in the pressure drop at room temperature (25 °C). (A) CO2 absorption rate normalized by MECS mass and initial pressure, (B) CO2 loading normalized by MECS mass.

Sarcosine, which is the N-methylated analogue of the natural (and simplest) amino acid glycine, leaked from the capsules and was not pursued.

The comparison for the sodium carbonate and cyclen systems was then made with the ionic liquids:



The caption:

Figure 6. Comparison of CO2 absorption rate over the first 10 min by MECS with (A) Na2CO3-cyclen, (B) NDIL0309, and (C) NDIL0230, and at 25, 40, and 60 °C. (D) The bar graph displays CO2 loading after 30 min (left axis), and the markers represent the percent of stoichiometric capacity (right axis) of each solvent.

Not only do the ionic liquids capture more carbon dioxide per unit mass (stoichiometric capacity), but they capture it faster:




The caption:

Figure 7. Comparison of CO2 loading vs time for three MECS (NDIL0230, NDIL0309 and Na2CO3-cyclen) during Cycle 0 at 25 °C and 0.1 bar CO2. CO2 loading is presented in terms of mol/kg solvent (left) and as a percentage of total stoichiometric capacity (right).

They hold up well through multiple cycles, a very, very, very, very important issue for these types of systems.



The caption:

Figure 8. Comparison of CO2 absorbed by three MECS types (NDIL0230, NDIL0309 and Na2CO3 w/cyclen) across 10 cycles. CO2 loading capacities are the cumulative CO2 absorbed (per kg solvent and initial pressure) after the MECS have been exposed to CO2 in the pressure drop apparatus for 30 min. (A) CO2 loading is presented in terms of mol/kg. (B) The mass of the MECS is shown over 10 cycles (bars, left axis) and CO2 loading is shown as a percentage of total stoichiometric capacity (lines, right axis).

Some microscope photographs of the microcapsules, before and after:




The caption:

Figure 9. Microscope images of the three final candidate capsules taken before (top) and after (bottom) ten absorption/desorption cycles: (A) Na2CO3 before cycling; (B) NDIL0309 before cycling; (C) NDIL0230 before cycling; (D) Na2CO3 after ten cycles; (E) NDIL0309 after ten cycles; and (F) NDIL0230 after ten cycles. The scale bar for each image is 500 μm.

The authors conclude:

Overall, across the characteristics tested here—absorption rate, capacity, and cyclic stability—the ionic liquid MECS appear to be a potential competitor to aqueous amines. The major challenge to these MECS are common to solid sorbents: the need for a process configuration that is similar in capital cost and energy efficiency to aqueous solvent systems. Further work is needed on material fabrication and testing at larger scale to establish the viability of IL MECS.

While I do not believe the paper under discussion here is open sourced, there are two excellent government reports related to her DOE grants that are. They are here:

Hybrid Encapsulated Ionic Liquids for Post-Combustion Carbon Dioxide (CO 2) Capture 1337563

Hybrid Encapsulated Ionic Liquids for Post-Combustion Carbon Dioxide (CO 2) Capture 1406897

Again, I am against coal combustion, petroleum combustion and gas combustion, but I believe that biomass combustion and/or supercritical water or supercritical carbon dioxide oxidation does represent a path forward for capturing carbon dioxide from the atmosphere, and should not be overlooked, even if the combustion of biomass, as practiced now, is responsible for slightly less than half of the seven million air pollution deaths now taking place while many of us whine about Fukushima.

This kind of technology is potentially of huge importance.

I wish you a very pleasant Sunday evening.