INTRODUCTION

Carbon dioxide emissions to the atmosphere have risen steadily since the beginning of the industrial revolution. At present worldwide combustion of fossil fuels emits about 22 Gt of carbon dioxide to the atmos here.”2 The measuredannual increase in atmospheric C02 is approximately13 Gt. The difference between total output, which includes some additional emissions from deforestation and other anthropogenic sources, and the observed increase in atmospheric C02 is absorbed into natural sinks like the ocean and the biosphere. The substantial absorption indicates that the current state of the atmosphere is far from a steady-state equilibrium. The level of atmospheric carbondioxide has risen by 30% from its pre-industrial value of 280 ppm to about 365 ppmtoday. Most of this rise (about 60 ppm) has occurred during the last 50 years.The size of readily accessible fossil fuel deposits is extremely large. Easily accessible,oil and gas may be limited, but oil shales, tar sands and coal de osits are virtuallyinexhaustible.Coal deposits alone are estimated at 10,000 Gt, which should becompared to a worldwide

Aanual consumption of 6 Gt of carbon.*Methane hydratedeposits have become of recent

interest and may dwarf all others carbon resources. BestCombustionof fossil fuels could

riveatmospheric carbon dioxide levels very much higher. The available 10,000 Gt of

arboncorrespond to 4,700 ppm of atmospheric C02.7 While the detailed effects of carbon

dioxide on climate and environment are still debated, it is undisputed that carbon dioxide

is a greenhouse gas that could cause climate change. Carbon dioxide affects the acidity of

the ocean, it is of physiological importance and thus can directly affect the ecological

balance of species. Hardly anybody would advocate doubling natural COZ levels, yet

current consumption patterns inexorable will lead to this result. To stabilize C02 at 600

pprn requires a drastic reduction in C02 emission. Ultimately, emissions would have to be

reduced to about 30% of those of 1990.2 For 10 billion people sharing into such a COZ

budget the per capita allowance comes to about 390 of that of the average US citizen

today.

In summary, it appears to be extremely difficult to stop the growth of fossil energy

demand, yet to stabilize C02 levels requires a drastic reduction in COZ emissions. The

only way out appears to be some means of collecting and subsequent disposing of the gas

after it has been generated. If proven feasible, extraction from air would provide a

powerful approach to the problem. It completely avoids a restructuring

infrastructure, it uses the atmosphere to transport the carbon dioxide from its source to the

disposal site and it would make it even possible to lower the atmospheric levels of carbon

dioxide, if this turns out to be necessary or desirable.

We have looked into the feasibility of extracting carbon dioxide from the air and here

we provide simple dimensional arguments that suggest that there are no fundamental

obstacles to this approach of carbon dioxide sequestration.

The energy cost of this effort

appeam to be tolerable, and the infrastructure cost may well turn out to be low compared

to some of the alternatives.

WHYCOLLECTCARBON DIOXIDEFROM AIR?

Collection of C02 from the air opens up new options and possibilities. It makes it possible to retain a transportation sector that is based on an extremely convenient energy source of hydrocarbons. It opens up for sequestration a multitude of dispersed carbon dioxide emitters which otherwise would require a potentially costly rebuilding of the infrastructure that relies on a carbon free energy form, e.g. electricity or hydrogen. Carbon dioxide disposal requires carbon dioxide collection. Typically, carbon dioxide collection is integrated with carbon dioxide emitters. Carbon dioxide extraction from air would allow to the integration of the collection process with the disposal or sequestration step. In effect, this happens in biomass production. Biomass generation is, however, a very inefficient approach because it is coupled with the reduction of the carbon which requires as much energy as was released in the combustion. Biomass generation requires much land and is costly. There are various approaches of C02 disposal. Disposal in the deep ocean, injection into underground reservoirs and the chemical stabilization as carbonate minerals all offer means of keeping carbon dioxide out of the atmosphere. All have in common that they work best in specific locations, which provide the appropriate conditions. For example, mineral sequestration requires magnesium minerals that are abundant but nevertheless concentrated in specific locations.s Underground disposal requires special circumstances that guarantee safe and stable formations able to accommodate large quantities of carbon dioxide. Ocean disposal, in order to last for a long time, is likely to be limited to special locations where the absence of deep ocean currents would guarantee a long residence time. Thus, in all cases one would either have to relocate the emission sources near the disposal site, or alternatively transport the carbon dioxide to the location. Large scale bulk transportat adds substantial costs to the process and in many cases is not practical. For example, collecting C02 on board of an automobile would greatly add to the weight and cost of the car, it also would require a new infrastructure for C02 collection which can handle a mass flow that is three times larger than that of the gasoline distribution system. Extraction from the air would overcome this obstacle, as it would allow for the collocation of extraction and disposal. The atmosphere is well mixed and the COZ level is roughly the same everywhere. Even the Southern Hemisphere lags only a couple of years behind the Northern Hemisphere in C02 concentrations. Mixing along a given latitude occurs in a matter of weeks. Using the atmosphere as a vehicle for transporting the C02 does not pose an environmental risk. It is the increase in C02 levels over decades that matters not the accumulation of a few months. C02 extraction from the atmosphere opens up disposal sites which otherwise would not be of practical interest. For example, large deposits of serpentine in Oman would otherwise be of little interest for C02 disposal. If collection of C02 from the air would prove feasible, these deposits could be used to chemically bind C02 as magnesium carbonate.Similarly, large and well-suitedaquifers in Albertag would become accessible to C02 emitted anywhere in the world.Ocean disposal in the mid-ocean would become accessible to this method as well. In this

ense, extraction from air opens technological options for C02 disposal.

.While stationary, large-scale emitters may find ways of

disposing of C02, without technologies that can extract C02 from air, the same is not

possible for the myriad dispersed and mobile sources of carbon dioxides. Thus, there will

be a strong pressure towards abandoning the use of fossil energy for example in the

transportationsector in residential households and in commercial buildings.

Biomassfuels, or non-carbon based energy carriers would be the only remaining options. None of

these options have been shown to be economically viable. All of them would demand a

drastic rebuilding of the entire infrastructure. Carbon dioxide extraction from air, would

allow the continued use of carbon based fuels for distributed energy production.

Unlikeother approaches to the problem it would integrate the disposal process not with the

emission but with the collection scheme. Extraction from air would completely eliminate

the need for an entirely redefined and reshaped energy infrastructure.

A move to ahydrogen economy may still be considered on its own merits but it would not be required

in order to mitigate greenhouse gas emissions. Again, extraction of carbon dioxide from

air opens new technological options.

In the more distant future; renewable energy sources may become competitive with

fossil fuels. Then the extraction of carbon dioxide from air opens up another interesting

technological option. Renewable energy sources could be applied to turn carbon dioxide

into hydrocarbons.To this end a variety of chemical pathways have been studied,

particularly in Japan. Most of them start out with hydrogen and carbon dioxide to form

methanol or other hydrocarbons.

All of them have in common that they require

substantial amounts of energy to “refill the carbon with energy.” In turn the carbon-based

fiel can be used anywhere, for example in the transportation sector, and the “empties” are

returned via the atmosphere to the carbon dioxide collection site. Whether the “empties”

are discarded in C02 disposal or refilled with renewable (or nuclear) energy is ultimately

an issue of cost. Given the availability of fossil hydrocarbons we expect that it will take a

long time before the alternative energy approach becomes economically more viable.

FEASIBILITY

Extraction of carbon dioxide from air is feasible.

Photosynthesizing

plants already

collect carbon dioxide from the air. Many air liquefaction

schemes start with the

extraction

of carbon dioxide from air, since solid C02 would interfere with the

liquefactionprocess. Both examples prove the feasibility of the process, but both

examples provide a poor gauge of the technological difficulties, as they solve a more

difficult problemBiological extraction is rate limited by access to sunlight rather than

C02. Industrial processes that need to generate C02 free air are much more demanding

than processes that only need to extract a substantial fraction of the total, but are not

driven to reach extremely low concentrations in the output stream.

Whether or not carbon dioxide extractionfrom air will become economically

competitive to other means of carbon dioxide mitigation will depend on essentially two

issues, the cost of the collection process and the energetic of the process. The cost of the

collection process ought to be small compared to the cost of generating energy. Indeed

the cost of energy from fossil fuel plus the cost of the collection and disposal process

must be less than the cost of alternative forms of energy.

Furthermore,

the energy

demand of the process needs to be so small that its own C02 emissions don’t overwhelm

the COZ collection.

In this section we will address these two issues and conclude that both requirements

are likely to be satisfied. We begin with a simple dimensional argument that shows that

although carbon dioxide is dilute in the atmosphere it is not so dilute as to make

extraction hopeless. We then look at absorbents that could collect C02 out of air in spite

of the low concentration of 365 ppm. Finally we try to obtain an order of magnitude

estimate of the cost of a possible implementation of a candidate process.

The C02 in air is commonly considered too dilute to justify its collection. Here we

present a different point of view that suggests the opposite. Consider a cubic meter of air.

It contains roughly 40 moles of air (at T = 300K) and 0.015 moles of C02. If we were to

remove this carbon dioxide, some energy producer elsewhere is allowed to inject an equal

amount of C02 back into the atmosphere.

The combustion of this amount of carbon

comes to 6 kJ of energy. As coal is effectively CH, the heat value per mole of carbon is

slightly higher, resulting in 7 kJ per cubic meter of air.

Thus the removal of C02 from one cubic meter of air can be viewed as an integral part

of producing 7 kJ of thermal energy from coal, or 10 kJ of energy from gasoline. The

same cubic meter of air moving along at a strong wind blowing at 10m/s (22.5

mileslhour)

contains 58 J of kinetic energy.

Extractingwind energy from air isconsidered economical and actually proves quite cheap at about 5@Wh. If we consider generating energy from fossil fuels and collecting an equivalent amount of C02 from the air to avoid a net increase in atmospheric C02 then processing one cubic meter of air for C02 is much more effective than extracting its kinetic energy for alternative energy. If measured against its heat of combustion, the C02 in air is much less dilute than the wind energy contained in the air.

To pursue this comparison even further, if the same wind blows through a system that

removes carbon dioxide, harnessing the air fIow through one square meter of cross

section can compensate for 70 kW of thermal energy. The same cross section would tap

into 580 W of raw wind energy. Note that the actual useful energy in both cases is less. A

square meter unit of solar energy collectors would produce maybe 50W which represent

a much higher energy quality and it represents about 25% of the typical solar flux in the

US.The equivalent output in biomass collection is about 3 W of potential heat of

combustion and amounts to about 1.5% of the solar flux. Thus again we find that in a

power comparison,carbon dioxide extraction stacks up very favorably against the

obvious competitors.

Page

‘

What are the means of collecting C02 from air? There is a multitude of separation

schemes that could be used to separate gas streams, but most are not practical for

extracting trace gases, and the concentration of carbon dioxide in the air is only 365 ppm

by volume. For example, refrigeration processes are in principle possible but in practice

they would be too inefficient.

In cooling and re-heating a mass of air that exceeds the

mass of collected C02 by a factor 1800, the slightest inefficiency would cause the

energetic of the process to become prohibitively expensive.

Cooling the air to form dry

ice removes 2.2 MJ of heat from the air for every mole of C02. This should be compared

to 400kJ/mole in the heat of combustion that generated the C02. The same argument

applies to membrane technologies that would drive air through a membrane. A pressure

drop of 1 bar would require on the order of 7 MJ/mole of C02.

On the other hand

building up a pressure gradient in the partial pressure of C02 across a membrane is

virtually impossible given the low partial pressure of C02 in the input stream. Thus in the

end one is limited to absorptive processes that find a way of binding the C02 to a

chemical or physical absorber.

One example of a chemical absorber is a solution of

Ca(OH)2 which readily remcves C02 from ambient air.

The low pressure gradient that needs to be maintained in order to keep the gas flowing

through an absorption system is comparable to the kinetic energy in the flow.

Thus

following the same calculations as before we find an energy expenditure on the order of

60 J/m3 of air or about 4kJ/mole of COZ which amounts to about 1% of the associated

heat of combustion. Quite likely, a practical implementation

would utilize the natural

convection of air to accomplish this task rather than provide external energy for the

process.

Most of the energy demand for an absorption process is in the recovery of the

absorbent. In order to bind rapidly and effectively the absorbent needs to have a

substantial binding energy with C02. In a subsequent step of separating the C02 from the

absorbent, this energy needs to be supplied from external sources. The minimum binding

energy is given by the free energy of mixing.

The speed of the reaction is in part

determined by the excess in the binding energy of the sorbent.

One can easily calculate from first principles the change in free energy that would be

incurred in extracting C02 from the air. The free energy of mixing is given by

RT log P/PO

Where P is the ambient C02 partial pressure and P. the desired pressure of the C02 in

output stream. R = 8.314 J/mole/K is the gas constant and T is the absolute temperature

measured in Kelvin.

The free energy required for separating C02 from air at ambient

temperature and for providing an output stream at 1 atm is therefore 20kJ/mole.

This is the theoretical minimum energy expenditure that does not depend on the

specific choice of the separating scheme. Any-practical implementation will require more

energy, possibly substantially more energy. The minimum energy expenditure is only 570

of the energy releaSed in the combustion of carbon. Thus compared to the energy gained

in the combustion process the penalty is quite small. One should keep in mind though

that inefficiencies in power generation and in the extraction could rapidly add up. We

have given above some extreme examples using refrigeration or membrane technologies.

6

Page 10

.

.

In practice, most effective absorbents will bind much more strongly then is required strictly by thermodynamicconsiderations.Ca(OH)2 is a point in case. The heat ofcalcination of CaCOs is 179 kJ/mole.How long needs a absorption column be so that it can absorb the C02 from air flowingthrough?The answer will depend on the efficiency of the absorber, but even the best absorber will be limited by the rate of diffusion of C02 through air. The diffusion coefficient of C02 through air at ambient temperatures is D = 1.39 XIO-5m2/s. The mass flux to a absorbing surface is given byZV=Dgrad A?If we make the simplifying and optimistic assumption that theat the absorptionsurface vanishes,then the mass flux to

approximately by

N,b, = D pc02 L

partial pressure of C02

a boundaryis given

L is the distance over which the diffusion occurs, i.e. the typical distance to the wall,

Thus let us consider the case of air flowing through a set of parallel tubes, 1 mm in

diameter. If the inside walls are coated with a liquid film that strongly absorbs C02, then

we find that the flux rate implies that most of the C02 will be removed after about 30 cm

of flow. Based on the more accurate empirical the length could be shorter by about a

factor of 3.6.10

The length of the tube depends on the flow velocity and is inversely

proportional to the flow speed.

The pressure drop on such a system would be comparable to the kinetic energy in the

air flow. Thus typical pressure drops are of the order of 50Pa. The same argument that

implies that an air flow can transfer its C02 content to the side walls, also implies that the

gas will transfer its momentum to the wall.

One can easily imagine a variety of geometries for contacting the air and extracting

C02. The numbers we have given here give typical mass transfer rates and similar rates

are obtained by using mass transfer rates as tabulated for various flow geometries in the

chemical engineering literature. For some geometries, e.g. thin fibers, the mass transfer

rates are substantially higher. Possible flow geometries include air filters, droplets of

liquid falling, or packed towers. A particularly simple geomet is a provided by a panel

exposed to the airflow. In this special case, it would make more sense to think of the

contact area rather than the cross section of the airflow as the characteristic variable that

will describe the phenomenon.

Again based on Perry’s Chemical Engineering Handbook

we estimate that small sheets of absorbing surface would collect C02 at a rate of about

4 x10-4 mole/(m2s). This rate again depends on the wind speed.

The dependence is apower law with a coefficient between 1/3 and 1/2. Again we assumed 10 nis as the speed of the air flow. Thus, a simple collecting surface would operate at a collection efficiency equivalent to 190 W/m2, which is much lower than the flow per unit area normal to the wind direction, but it is still better than a photovoltaic system. 7

Page 11

In a somewhat indirect fashion this comparison also suggests that the mechanics of

collecting C02 from air does not require too large an investment. In essence the argument

says that C02 extraction equipment from an air flow could on a per area basis be much

more expensive than either wind energy or solar energy units without having much

impact on the overall price of energy. Since the extraction unit effectively handles 80

times as much power as an equally sized windmill, 1000 times more power than a solar

collector and 20,000 times more than agriculture, the price for C02 extraction equipment