An approach is proposed to enable direct air captured€02 to be produced and stored at food grade specifications without the need for the very costly small-scale C02 liquefaction to achieve the needed purity or needed on site storage capacity. This removes a major impediment to realizing the benefit of creating a distributed food grade C02 system to take advantage that direct air captured€02 can be produced anywhere, specifically at the site where it is being used.

Generally, a preferred method for obtaining high purity C02 comprises cyclically removing carbon dioxide from carbon dioxide-carrying ambient air by directing a flow of carbon dioxide-carrying air through a carbon dioxide capture contactor structure having pores, said contactor structure supporting sorbent which is capable of binding carbon dioxide, to remove carbon dioxide from the air by binding carbon dioxide to the sorbent, thereafter causing said carbon dioxide capture contactor structure to be exposed to regeneration conditions, whereby saturated steam of a temperature not greater than about 120 degrees C is directed at the carbon dioxide bound to the sorbent, thereby facilitating separation of the carbon dioxide from the sorbent and regenerating the sorbent, then withdrawing the carbon dioxide that has been unbound from the sorbent together with any remaining steam, and selectively thereafter repeatedly exposing said carbon dioxide capture structure to a flow of carbon dioxide-carrying ambient air, thereby enabling the regenerated sorbent to be again used to bind carbon dioxide, to remove carbon dioxide from the newer flow of carbon dioxide-carrying ambient air. This can be achieved at a reasonable total cost by utilizing a system for carrying out this invention including a location for removing carbon dioxide from the atmosphere, a€02 extraction system including a substrate having an amine sorbent on its surface which captures€02 from air flowing over its surface; and a second, regeneration and €02 collection location, where the captured€02 is removed from the amine sorbent and the removed carbon dioxide is withdrawn and isolated in a third location.

At the second location, the€02 containing sorbent is exposed to steam at a temperature of not greater than 120°C in a sealed chamber, and the removed€02 is passed into the third location in a substantially pure state. It has been found that wdien carrying out the initial contacting of the ambient air with a contactor structure that the material actually containing on its surfaces the sorbent material includes pores extending at least partially through the entire thickness of the material. The pores are formed such that the sorbent material is present on its surface or may be filling at least partially the full cross-section of the channel .

One process for obtaining a highly pure carbon dioxide from not only ambient air but ambient air mixed with a minor proportion of a carefully free processed flue gas to form a uniform high C02 content gas is the system described in published patent application number US 2011/0296872 In this case of course the preprocessing is critical in order to avoid the presence of undesirable or toxic impurities. The complete disclosure of this published application, commonly owned, is incorporated herein by reference as if fully set forth.

Another procedure for achieving pure carbon dioxide is disclosed in an air contact structure containing a substrate having the desired channels extending below its surface and a CQ2 sorbent present on the porous surfaces of the substrate. The C02 containing gas, whether ambi ent air or a mixture of ambient air and pre-cleaned exhaust gases are allowed to flow over the surface of the substrate and upon reaching the desired concentration of CQ2 on the sorbent, the contact structure is then caused to be in a sealed regeneration container from which any remaining trapped gases are exhausted and the total pressure reduced to below 0.4 barA and preferably to below 0.3 bar A and optimally to as low as between 0.1 and 0.2 barA In a most desirable situation, there are a pair of such cycling systems involving 2 regeneration chambers and at least 2 contact structures that are passed into the regeneration chambers upon completing the adsorption of C02,

In the regeneration chamber the substrate is exposed to a stream of process steam at temperatures of not greater than 120° and preferably at temperatures below 100°C. In all of these systems, the initial temperature of the C02 rich gas contacting the capture structure is preferably ambient temperature and optimally not greater than 25°C this allows for the desi red load temperature regeneration of the sorbent and capture of the C02.

By utilizing the paired regeneration chamber's the regeneration chamber which has completed the regeneration is filled with a combination of C02 and remaining steam at a pressure greater than the original vacuum pressure described above which can then be utilized to be passed into the 2nd regeneration chamber to preheat the material so as to increase the efficient use of energy by preferably after at least a major portion of the regenerated CQ2 is removed from the chamber by a combination of a pumping system and a and the flow of steam. The 2nd chamber is then exposed to the pumping system to reduce the total pressure therein to below 0.4 as explained above and the re-generating steam is then passed into the 2nd sealed chamber to continue the regeneration process. These procedures are all carefully and fully describe in the above recited published patent applications incorporated herein. In addition to the above procedures and as the additional invention of the present application the following steps are carried out.

It begins by adding a step of a C02 flush after pump down to a pressure of between 0 2 and 0.4 bar A, before regeneration by steam. It is in addition to our previously planned approach of adjusting the transition from exhausting the stream emerging during the beginning of regeneration to collecting the CQ2

(1) The pure collected C02 produced during regeneration does contain moisture even after passing through a condenser to knock out some of the water and may be contaminated by the steam (eg oxygen). The first step therefore is to dehydrate the C02 via established approaches most likely using molecular sieve technology, see http://www.ieaghg.org/docs/General_ ocs/Reports/2014-04.pdf. This approach also allows for having“getters” in the pores to remove any remaining impurities in the pure C02 gas stream. Once pure C02 is obtained, the system capacity is chosen, to first order, to deliver it directly to the user at the working pressure desired by the user. Due to the anticipated volatility of demand and depending upon the magnitude of the volatility, different storage options should be available in order to achieve the most economically desirable results. The solutions will vary from not operating when there is no demand, using the excess C02 to produce another product (eg cannisters of C02 for retail sale), high pressure ballast tanks to higher volatility solutions that involve storage in, for example, zeolites, covalent organic frameworks or activated carbon under modest pressures at room temperatures (around 35 bar) and to achieve storage densities of about 20% of liquid C02, see

http://Yaghi.berkeley.edu/pdfPublications/09MOFexception.pdf and

http : // pub s . acs . org/d oi/ab s/ 10.1021 /j a9015765. Which solution represents the best tradeoff in terms of cost and performance and to what pressure one goes to will need further study as will how to best integrate it with the dehydration step. This in principle produces to first order a scale independent cost of C02 that is low capital and low operating expense. We can follow the same approach currently used of having two storage tanks one that is being filled and the other that is being drained of CQ2 to delivery C02 . The deliver}- pressure will be considerably lower than the storage pressure and will determine how low one can empty the storage system. One will need of course a control system that delivers the C02 at the pressure and flow rate needed and can respond adequately to fluctuations in demand. All this seems eminently feasible for an approach that does not depend upon a change in state.

Production and storage of Food Grade Purity€02

In this approach one will have two pumping systems and lines, one that is used for the exhaust and one for collecting the CQ2. There should, preferably, be separate exit ports from the regeneration chamber. From the exhaust port one would pump from 1.0 bar A to 0.2 prior to the C02 flush step. From the C02 collection port, one will pump from 0.7 bar and also have a step to remove the water and possible remaining impurities as described above. For example, the C02 can be pressurized to slightly above the working pressure for direct use with a ballast tank and fro that pressurized working tank pump into a higher pressurized storage tank, storing the C02 in higher pressure ballast tanks or in the porous materials mentioned earlier This step can be carried out more slowly than the first step because it will only be required when the demand decreases below the amount delivered by the low pressure system. If demand increases above the amount delivered by the low pressure system, the other filled ballast or storage tank would be available to make up the additional amount required for the increased C02 demand. As a result, pumping to higher pressures becomes a second order process greatly reducing the amount and rate of pressuri zation and thus energy use, and thus cost.

If needed as described above getters can be provided in the bal last/storage tank as well to preserve and maintain purity.

Preferably, the system should include a line from the low working pressure collected C02 back to the steam injection manifold for injecting C02 into the box to assist in the removal of pure C02. Ail the C02 lines and tanks would probably need to be made from whatever is needed to keep the C02 pure and other than the regeneration box none of the C02 collecting lines will see anything but pure C02.

The Operating System Described

Step 1 loaded monoliths enters the regeneration box.

Step 2 box is pumped down to 0.2 barA.

Step 3 pure C02 from the collection tank is pushed through the steam injection system with the pump preserving 0.2 barA pressure.

Step 4 the box is then connected to the other regeneration box which is filled with C02 at 0 9- 1.0 barA where the condensed water is evaporated from the hot box into the box with 0.2 barA C02 and condenses on the cold monoliths preheating them.

Step 5 the steam is injected the CQ2 removed at 0.7 barA as before and switched to the collection tank from the exhaust line to achieve the desired purity. The exact details of the transition from C02 injection to steam injection will need to be empirically determined.

The following suggests that this approach is feasible and low cost, where the desired purity can be achieved . For the case of oxygen, to reach not more than lOppm 02, requires a further dilution of 4000. The ratio of the C02 collected per cycle to the oxygen remaining in the monolith at our design of 22 liters is about 150, requiring that we remove 97.5% of the remaining air. This can be accomplished via the three stages, flush with C02, heat transfer step flush, and if needed, adjustment of when to start collecting the C02, and possible use of an oxygen scavenger in the sealed chamber or in the storage medium. Step by step analysis;

1. Per 3.4 liter monolith assume an extra volume for the space on either end of monoliths making the volume per monolith of about 4.5 liters and then subtract the filled space of the monolith and structure which for a void fraction of .65 and a loaded wall with sorbent of about .6 gets you close for ease of calculation to about 4 liters per monolith of volume.

2. At 0.2 bar the amount of air in that volume is .8 liters and oxygen 0.16 liters. Which as noted above is a small fraction of the C02 collected in a cycle.

3. If one injects one volume worth of C02 (0.8 liters) pumping to keep 0.2 bar one

should remove the 0.2 bar of air since other studies at SRI showed good plug flow behaviour- One can adjust the amount of C02 injected up or down as needed for purity based on the analysis of what is produced by the steam injection step

4. To move 1 liter (to be conservative) of€02 through the roughly 196 monoliths per array for a 4000tpy plant requires we move about 196 liters per box of C02 at one bar. e want to do so in time T.

5. If the C0 ₂ has to travel a distance D from the injection point to the collection point at the exit from the box, one has to pump 196 liters per time T and the velocity of injection needs to be D/T.

6. If the width of the box is about 20 cm one can use 50 cm for the longest total distance (Depends upon spacing of exit ports and injection path) and let us say one wants to do this in 5 seconds one than has a velocity of 10 cm per second and a volume flow rate of 2056 liters per minute or about 70 cubic ft per minute or 112 cubic meters per hour. To get the actual volume needed to pump out the C02 and air at 0.2 bar, one needs to multiple by 5 to get the pump size needed (350 CFM) not accounting for efficiency (see below). To pump down the 4 liters to 0.2 bar in say 5 seconds is also about the same CFM so the same pump can be used for both steps that exhaust the pumped gases at 1 bar into the air. Note this exhaust will have some excess C02 should be exhausted such that it can possibly be pulled back into the system.

7. In the next step the cold box is connected to the hot box that has finished collecting C02. That box is tilled with€02 at 0.9bar. The important point is that the 0.9 bar pure C02 provides a second sweep opportunity. It is possible that this step alone might provide the needed purity since it is 2.8 liters’ worth of C02 but because it is coming across with water vapor that condenses the plug flow properties of this stage are more complicated. In any case one would start collecting the C02 being transferred at the point that produced the desired purity of C02. Between the two sweep steps it seems dear that the desired purity can be achieved, the only question being how much CO2 is“wasted” to achieve it. Flow much is actually wasted can in turn be reduced by the way the pump exhaust is distributed. For example, it could be distributed such that it was pulled in via the fan on the monolith array on either side of the regeneration boxes A secondary benefit of the first low temperature flush might be to remove trace impurities in the air and oxygen before they are exposed to water and higher temperature. This could improve lifetimes. The cost and effectiveness of the two sweeps and time to switch to collection needs to be empirically determined. Another embodiment is to use some of the CQ2 trapped in the hot box at the end of regeneration as the source of the C02 used in the sweep before heat transfer by briefly pumping into a storage tank used as the source of CQ2 in the first sweep.

8. For the regular collection of C02 of say 22 liters per monolith at 0 7 bar in 30 seconds is almost the identical CFM.

9. Clearly all this scales with the tpy and so a lOOOtpy plant would have monolith arrays with ¼ the volume and thus ¼ the CFM requirement and being smaller in size would reduce the velocity if one wanted to keep the 5 second time for doing it and thus further reduce the power requirement for the compressor.

The actual pressure of the first step for producing C02 for the working tank will be impacted by the dehydration step. Overall this process will need more detailed analysis to optimize it. The purpose of this analysis is to determine whether it is feasible to directly produce food grade C02 and the above analysis indicates an affirmative result, but that further evaluation will be useful at specific ambient conditions to achieve optimum effect.