Carbon Resources Conversion 1 (2018) 183–190

Contents lists available at ScienceDirect

CHINESE ROOTS
Carbon Resources Conversion
GLOBAL IMPACT journal homepage: http://www.keaipublishing.com/en/journals/
carbon-resources-conversion

Greenhouses for CO2 sequestration from atmosphere T

a b b,c a,b,⁎
Jie Bao , Wei-Hua Lu , Jigang Zhao , Xiaotao T. Bi
a
College of Chemical Engineering, Fuzhou University, Fuzhou, China
b
Clean Energy Research Centre and Department of Chemical and Biological Engineering, University of British Columbia, Vancouver, Canada
c
School of Chemical Engineering, East China University of Science and Technology, Shanghai, China

HIGHLIGHTS

• AEnriched
novel approach to provide CO source to greenhouse is proposed.
• Potential CO
2
from Temperature Swing Adsorption for greenhouses.
• Potential tocostremove
2
more than 500 million tonnes CO from air annually.
•
2
of $15–$30/T-CO if effective adsorbents are developed.
2

ARTICLE INFO ABSTRACT

Keywords: Escalating threat of global warming and the steady growth in world population require the development of
Carbon enrichment for plant stimulation transformative greenhouse gas control technologies and food production systems of high energy efficiency, small
Greenhouses environmental footprint and low cost. To control the global temperature rise below 2 °C by 2050, global
Temperature swing adsorption greenhouse gas emissions need to be cut by more than 80%. At the same time, our land needs to be utilized more
Sequestration of atmospheric CO2
efficiently and productively in order to produce enough food to feed projected 9 billion people with less
available land area for food production in 2050. We propose to develop a modern urban vertical farming system,
i.e. greenhouses equipped with a Carbon Enrichment for Plant Stimulation (CEPS) system, to enhance land use
efficiency and thus increase food productivity and, at the same time, to sequestrate CO2 from ambient air. The
deployment of such a CEPS system will have a potential to remove more than 500 million tonnes CO2 from air
annually, and increase the current food productivity by more than 15 times than the open field operation. The
deployment of the CEPS technology will also promote locally produced food, benefiting urban economical de­
velopment and job creation.

1. Introduction temperatures associated with sorbent regeneration and subsequently

high energy penalty are needed for separating CO2 from air at a low
The newly released IPCC report revealed that the global greenhouse concentration of 400 ppm, and then recovering captured CO2 from
gas emission rate has continued its increasing trend, almost doubled solvents or adsorbents into nearly pure or highly concentrated CO2
over the past decade, and the carbon dioxide (CO2) concentration in the stream for secure disposal. Sequestrating CO2 from atmosphere has an
atmosphere has reached 400 ppm, leading us toward potential cata­ estimated cost of $35–300/T-CO2 with an energy penalty of 30–400 kJ/
strophe if no rapid actions are taken to control the global temperature mol-CO2 [3–7], which is more expensive than capturing CO2 directly
rise below 2 °C by 2050. To avoid the worst impacts of climate change, from point sources at a flue gas CO2 concentration in the order of
the report proposed an energy revolution to replace all fossil fuels by ∼10% or replacing fossil fuels with renewable fuels (< $100/T-CO2).
renewable energy sources including solar, wind and bioenergy. This, New technologies thus need to be developed to sequestrate CO2 from air
however, will require significant political and commercial changes. at low energy penalty and low cost.
Besides reducing carbon emissions from all human-made sources, CO2 Nowadays, direct air capture (DAC) has shifting researchers’ interest
can also be sequestrated from the atmosphere by photosynthesis of from off-site capture to on-site capture, which is capturing CO2 directly
plants or by “artificial trees” (e.g. absorption or adsorption) [1,2]. In from the atmosphere. The concept was first proposed by Lackner in
the air carbon capture and storage (CCS) process, high operating 1999 [8], and this technology is developing rapidly in the past decade.

⁎
Corresponding author at: Department of Chemical and Biological Engineering, 2360 East Mall, Vancouver V6T 1Z3, Canada.
E-mail address: tony.bi@ubc.ca (X.T. Bi).

https://doi.org/10.1016/j.crcon.2018.08.002
Received 26 February 2018; Received in revised form 4 August 2018; Accepted 11 August 2018
Available online 13 August 2018
2588-9133/ © 2020 The Authors. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

Zeman & Lackner [9], Keith [10], and Mazzotti [11] have proposed the field [24]. In an effort to increase the food productivity with limited
various pathways for DAC by aqueous solutions, which has a significant fertile land areas, an indoor controlled environment could provide the
drawback as a result of the multi-stage and energy intensive sorbent best growing conditions to plants. The switch of food production from
regeneration [12]. The DAC technologies have been investigated with a open field to indoor greenhouses, “urban agriculture” or “vertical
focus on adsorption, very little on desorption performance. Amine- farming”, represents a paradigm shift worldwide in order to produce
based nanofibrillated cellulose [13] was found to have a CO2 adsorption enough food for an anticipated global population of 9 billion in 2050
capacity of up to 0.695 mol/kg for direct air capture. Jones and Ka­ with food consumption almost doubled from now. At the same time, the
wajiri [5] proposed a five-step temperature vacuum swing adsorption greenhouses can be potentially used for sequestrating CO2 from am­
process for DAC using MOF adsorbents. An amine hybrid resorcinol- bient air when properly designed and operated.
formaldehyde/silica composite aerogel was synthesised by Shen and As direct air capture is gaining more and more attention, together
Fan [14] as an adsorbent for DAC, with an adsorption capacity of 1.80 with the fact that CO2 concentration has a positive effect on photo­
and 2.57 mol/kg-CO2 achieved under dry and humid conditions re­ synthesis, some researchers have proposed CO2 enrichment for plant
spectively. Elfving [15] studied the performance of polymeric ad­ growth using the captured CO2 from the air. Brilman [25] has proposed
sorbent in both dry and wet conditions by temperature swing adsorp­ a concept of ambient CO2 capture for microalgae cultivation using
tion, temperature-vacuum swing adsorption and pressure swing supported amine sorbent, which shows a cyclic capacity of 2 mol/kg.
adsorption, obtaining a CO2 adsorption capacity up to 1.06 mol/kg. Wang [26] also developed an integrated system combined direct air
Besides the well-known Class 1, 2, and 3 adsorbent for CO2 adsorption, capture and greenhouse agriculture, with a CO2 adsorption capacity of
Wilfong and Gray [16] developed a new category adsorbents which is 0.83 mol/kg reported. Commercially, Climeworks in Switzerland [27]
hybrid Class1/Class 2 supported amine sorbent for CO2 capture. The has piloted a system to capture atmospheric CO2 with an adsorbent. Air
adsorbent was synthesised by immobilizing polyamine/aminosilane is pumped through the unit and then chemically bound to the sorbent.
mixture on silica. Although CO2 adsorption performances haven’t been Once the filter is saturated with CO2, it is heated to around 100 °C to
evaluated in details, the mechanisms were well revealed, providing a release high-purity CO2, which is then collected and then transported, if
promising pathway to develop new category adsorbent for CO2 capture. necessary, to the greenhouses for plant growth stimulation. To facilitate
Carbon dioxide is the most important carbon source for plant the direct CO2 enrichment for greenhouses, we propose to enrich CO2
growth via photosynthesis. Each year, about 1–2 metric tonnes of CO2 is concentration in ambient air from 400 ppm to 1000 ppm by tempera­
sequestrated by trees in each 1000 m2 natural forest [17]. In an indoor ture swing adsorption. Different from the conventional carbon capture
environment like the greenhouse with sufficient lighting and water and process which captures CO2 from exhaust flue gases or from the at­
nutrient supplies, plant growth rate is limited by the CO2 concentration mosphere via chemical absorption/adsorption followed by desorption
in the atmosphere. An increase of CO2 concentration from 400 to at higher temperatures to produce pure CO2 and regenerate sorbents,
1000 ppm can stimulate the plant growth rate, leading to a yield in­ physical adsorption/desorption at ambient temperatures will be
crease of 21–61% for flowers and vegetables [18–20]. CO2 enrichment adopted for CO2 enrichment in the proposed Carbon Enrichment for Plant
has been widely practised in the greenhouse industry to increase plant Stimulation (CEPS) process to avoid energy intensive and expensive
productivity and to shorten the growth time. To elevate the CO2 level in sorbent regeneration at high temperatures.
the greenhouse air to the optimal level of ∼1000 ppm, CO2 is injected Following the same argument as House et al. [3], the theoretical
into the greenhouse at a rate of 10–20 kg/1000 m2/h in commercial minimal work required to separate CO2 in the air (400 ppm) into an
greenhouses [21], corresponding to 80–160 metric tonnes per year enriched product stream at 1000 ppm CO2 and a dilute stream of
which is more than 40 times higher than sequestrated by natural forest 200 ppm CO2 was calculated to be only 1.2 kJ/mole-CO2, 90% less than
[17] and 20% as much as sequestrated by microalgae in open ponds the CCS process which separates 400 ppm CO2 air into a 99% pure CO2
[22]. CO2 can be supplied from several sources including dedicated stream and a 200 ppm dilute air stream (17.8 kJ/mol-CO2) at ambient
natural gas or propane burners or liquefied CO2. In the summer when pressure and temperature (25 °C). At 15% thermodynamics second law
heating demand is low while CO2 demand is high as the plants grow efficiency with an enrichment factor of 2.5 (=1000 ppm/400 ppm), the
robustly, significant amount of natural gas is burned for the purpose of energy penalty for capturing and delivering one mole CO2 to the
CO2 enrichment only. greenhouse will be about 8.5 kJ/mol, one order less than for air CCS
Due to the increasing demand on locally produced food and vege­ process (30–400 kJ/mol).
tables, the urban agriculture industry has been expanding rapidly, with The proposed CEPS process is illustrated in Fig. 1. CO2-enriched air
the total greenhouse area in Canada alone increased from 2180 acres in will be directly injected into the greenhouse to stimulate plant growth,
1991 to 5670 acres in 2011 according to Statistics Canada. Globally, the replacing CO2 from burning fossil fuels. By producing CO2-enriched
total greenhouse area for vegetable production has reached 1 million ambient air at each greenhouse, costs associated with compression,
acres in 2012, and is expected to double in next decade [23]. Based on transportation and handling, distribution and storage will also be
an average CO2 demand of 15 kg/1000 m2/h and a total greenhouse avoided. Successfully developed and implemented, each greenhouse
area of 1 million acres, the total annual CO2 demand is estimated to be will be transformed into a distributed CO2 sequestration station. This
540 million tonnes, representing about 6% current annual global GHG paper for the first time focuses on examining the technical and eco­
emissions (9 billion tonnes CO2-eq.) from fossil fuel uses. If the addi­ nomical feasibility of the proposed CEPS system.
tional CO2 required by greenhouses in the world can be supplied di­
rectly from surrounding atmosphere, instead of from fossil-fuel sources, 2. Technical and economic feasibility analyses
all greenhouses in the world will become local “CO2 sequesters”, re­
moving up to 6% annual global CO2 emissions from fossil fuels sources 2.1 Materials and methods
from the atmosphere for enhanced food production.
For the growth of tomatoes, the productivity per square meters of The technical feasibility was evaluated experimentally using zeolite
land area can be more than doubled in the greenhouse than in the open 13X, a widely used commercial adsorbent for CO2 separation. It has
field. Because of the year-round operation, the annual productivity can been pointed out [28–35] that most of the previous researches have
be 15 times higher than the field plantation [24]. This implies that focused on the adsorption of CO2 by zeolite 13X for gases of relatively
every square meter’s land of indoor greenhouses can sequestrate 15 high CO2 concentrations, but not for gases of ultra-low CO2, the am­
times CO2 from the atmosphere than from the open field, and more than bient air for instance. Furthermore, little work has been done on des­
50 times higher than the natural forest. Water usage in terms of kg of orption of adsorbed CO2 from zeolite 13X, and none for CO2 enrichment
fruit per liter of water in the greenhouse is also only 20% of that used in at low concentrations. To prove the technical feasibility of the proposed

184
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

rate ranged from 0.2 to 1 L/min. The input gas flowed into the inner
column packed with adsorbent where adsorption and desorption took
place, with the outlet CO2 concentration (Cout) being measured by an in-
situ CO2 analyzer (Quantek Instrument, Model 906). Adsorbents were
first loaded into the inner column. The gas mixture was cooled down to
desired Tad via passing through the water bath before flowing into the
inner column. CO2 in the gas mixture would be gradually adsorbed by
the adsorbent, with the CO2 concentration in the gas mixture exiting the
bottom of the column being monitored by a CO2 analyzer continuously.
When adsorption was completed, the unit was switched to desorption
mode by quickly increasing the temperature of the column via
switching the water passing through the jacket from the low tempera­
ture water bath to the high temperature water bath. With the same gas
stream at a higher temperature, the adsorbed CO2 over the adsorption
period would be released from adsorbent into the passing gas mixture.
As a result, the gas stream exiting the column would contain more CO2
than the inlet, leading to a CO2-enriched gas stream for use in the
greenhouse.

2.2 Results and discussions

2.2.1 Adsorption and desorption characteristics

Fig. 3 shows typical adsorption and desorption curves for the 13X
Fig. 1. Illustration of (a) CCS (Carbon Capture and Storage) and (b) CEPS sample purchased from Shanghai Molecular Sieve Co. At the beginning
(Carbon Enrichment for Plant Stimulation) processes.
of adsorption, CO2 was adsorbed by the adsorbent as indicated by the
near zero CO2 concentration in the effluent stream. After this initial
process, experiments were conducted in a packed bed column using adsorption period, the adsorbent started to become saturated layer by
three types commercial 13X adsorbents by temperature swing adsorp­ layer, accompanied by a gradual increase in CO2 concentration in the
tion (TSA) at low temperatures, with the process diagram shown in outlet gas until a complete saturation was reached when the con­
Fig. 2. The test column is made of quartz glass, 6.5 mm in inner dia­ centration approached the inlet CO2 concentration. After the adsorbent
meter and 0.3 m long, surrounded by a water jacket to maintain the was completely saturated by CO2, the operation was changed to the
column at desired adsorption (Tad) and desorption (Tde) temperatures desorption mode by switching the hot water from the higher tem­
by two water baths. Before the adsorption experiments, all 13X samples perature water bath into the jacket surrounding the test column. As the
were dried in an oven at 110 °C for 12 h to remove the moisture. A CO2/ temperature increased abruptly, the adsorbed CO2 started to be des­
N2 gas cylinder with 1000 ppm CO2 and a cylinder of pure N2 were used orbed from the adsorbent. The concentration reached a peak and then
to prepare gas streams containing 1000 ppm and 400 ppm CO2. gradually dropped to the initial base level. Fig. 3 shows that desorption
A 13X zeolite sample was purchased from Shanghai Molecular Sieve is much fast than adsorption for zeolite 13X, facilitating fast re­
Co. and tested extensively. The particles have a mean diameter of 3 mm, generation of sorbent. The area underneath the adsorption curve (part
a BET surface area of 321 m2/g, a pore volume of 0.12 cm3/g and a pore A) would give the total amount of CO2 adsorbed, and the area in part B
size of 14.9 nm. Before the adsorption experiments, all 13X samples would provide the amount of CO2 desorbed.
were baked in an oven at 110 °C for 12 h to remove the moisture. A To characterize the CO2 enrichment performance, a moving average
CO2/N2 gas cylinder with 1000 ppm CO2 and a cylinder of pure N2 were concentration at time t after desorption begins is defined as
used to prepare gas streams containing 1000 ppm and 400 ppm CO2. t
5–30 g 13X samples were loaded into the column, and the gas flow C (avg ) =
0
Cout dt / t (1)

Fig. 2. Experimental setup.

185
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

Fig. 3. Typical adsorption and desorption curves using 1000 ppm CO2 balanced with N2 gas mixture (20 g of 13X, adsorption flow rate of 1 L/min, desorption flow
rate of 0.4 L/min, adsorption at 10 °C and desorption at 30 °C).

C(avg) represents the average concentration of enriched air over a proposed CEPS process is further evaluated.
desorption period of t, which can be directly compared to desired CO2
concentration in the greenhouse. The ratio of the average concentration
of enriched air over its initial inlet value is defined as the enrichment 2.2.4. Cost analysis
factor (f) The capital cost of a CEPS process is composed of the purchase of
equipment, adsorbents, auxiliary equipment, building and installation.
f= C (avg )/ Cin (2) The operating cost is mainly related to heat consumed over desorption
and electricity for air compression. Because of the low temperature
operation in physical adsorption, the whole unit including the ad­
2.2.2. Adsorption capacity, desorption capacity, and enrichment factor sorbent and the column is assumed to last 15 years. The annual interest
Fig. 4(a) shows typical moving average CO2 concentrations as a rate is set at 10%. The 13X zeolite adsorbent tested in the current study
function of desorption time at different desorption flow rates. The is about $2000/T (metric tonne). The selected sorbent loading for
moving average concentration increased with time, and reached a peak economics analysis is 15 tonnes, at a total cost of $30,000. The size of
before decreased. The enrichment increased with decreasing the deso­ the TSA fixed bed column to host 15 tonnes sorbent was 3 m in diameter
rption air flow rate and reached an optimal at 0.4 L/min, which was and 4.5 m tall. With galvanized carbon steel as the construction mate­
selected as the suitable air flow rate for economical feasibility analysis. rial, the column was estimated to cost $40,000 following the standard
The enrichment factor was evaluated as a function of adsorption engineering cost-estimation procedures for chemical vessels [36]. The
temperature and desorption temperature. As shown in Fig. 4(b), a installation cost, maintenance cost and the cost for building and aux­
maximum enrichment factor of about 3 was achieved at 400 ppm inlet iliary equipment (e.g. gas blower and heat exchanger) were assumed to
CO2 at a swing temperature of 20 °C, confirming that CO2 in the air can be 5%, 3% and 10%, respectively, of the cost of TSA column [36]. The
be enriched from its ambient value of 400 ppm to a maximum of estimated total annualized capital cost is about $10,000.
1200 ppm at ambient conditions. Furthermore, as shown in Fig. 4(c), The flow rates for adsorption and desorption are selected based on
the maximum enrichment factor increased with decreasing the ad­ the same space velocity (gas flow rate divided by catalyst volume) in
sorption temperature and increasing the swing temperature window. the current study, which has been commonly practiced in the sizing of
The results suggested that one could operate the unit at a low adsorp­ chemical reactors. For 1 L/min for adsorption and 0.4 L/min for deso­
tion temperature and a high desorption temperature to improve CO2 rption at a sorbent loading of 20 g in this study, a space velocity of 50 L/
enrichment. The adsorption capacity of three tested 13X sorbent sam­ min/kg-sorbent and 20 L/min/kg-sorbent were used for the adsorption
ples ranged from 0.035 mol/kg for the Shanghai Molecular Sieve Inc. and desorption, respectively. The pressure loss for air passing through
sample, as shown in Fig. 4(d), to 0.09 mol/kg for the Sigma-Aldrich the packed bed of sorbent particles is estimated by the well-known
zeolite 13X sample in the powder form (mean diameter of 2.34 µm, bulk Ergun equation, the most widely used equation for packed bed design.
density of 0.32 g/ml and pore diameter of 10 Å), not shown here, at an The electricity consumption is estimated by assuming an electrical
adsorption temperature of 20 °C. efficiency of 80% for the air blower. Adsorption is assumed to take
place with air at ambient temperature. The air temperature needs to be
2.2.3. Moisture effects raised to the level desired for desorption (e.g. 20 °C above the ambient
A humidification unit, as shown in Fig. 5, is added to the previous temperature). Since all greenhouses require heat to maintain a certain
process to evaluate moisture effect on the adsorption and desorption temperature suitable for plant growth, and the greenhouse industry has
capacity of 13X. As shown in Fig. 6, the adsorption capacity was found been exploring geothermal and solar energy for greenhouse heating, it
to decrease only slightly as humid gas mixture was used, which is likely is thus reasonable to assume that hot water produced from geothermal
due to the periodic temperature swing which prevents the permanent or solar thermal panels will be used to heat air for desorption. Since the
blockage of adsorption sites by water vapor. Those preliminary test hot CO2-enriched air is directly injected into the greenhouse, it will not
results using zeolite 13X clearly demonstrated the technical feasibility cause any energy loss. Therefore, the cost associated with air heating
of enriching CO2 in ambient air via temperature swing adsorption. for desorption is excluded in the current analysis. The annualized op­
Based on those technical data, the economical feasibility of the erating cost, including maintenance cost and cost of electricity, can be

186
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

Fig. 4. (a) Moving average CO2 concentrations with an adsorbent loading of 20 g. The adsorption temperature was 20 °C at a flow rate of 1 L/min and CO2
concentration ∼1000 ppm, and desorption took place at 50 °C with flow rates ranging from 0.2 to 1.0 L/min. (b) Enrichment factor as a function of desorption time at
swing temperature ΔT = 20 °C and inlet CO2 concentration of 400 ppm. (c) Maximum enrichment factor as a function of adsorption temperature at different swing
temperatures and two inlet CO2 concentrations. (d) Adsorption capacity as a function of adsorption temperature and swing temperature at an inlet CO2 concentration
of 1000 ppm.

Fig. 5. Process diagram of CO2 adsorption and desorption unit with humidification unit.

187
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

Fig. 6. Adsorption capacity of 13X sorbent particles as a function of inlet air

Fig. 7. Sensitivity of CO2 unit cost to adsorption capacity and adsorbent price.
relative humidity and CO2 concentration of 400 ppm.

studies on amine-based sorbents have been focused on its high ad­

obtained using the following equation:
sorption capacity, little on the desorption rate and desirable desorption
ccap = PV (1 + i)m i/[(1 + i) m 1] (3) temperature. For CEPS application, it is desirable to develop a sorbent
with a high adsorption capacity and fast desorption rate at low tem­
where PV represents the present value, i is effective annual interest rate, peratures.
and m is the life time in number of years. For the 15 tonne unit, the Table 1 compares the cost of CO2 from different sources for green­
annualized operating cost is estimated to be about $500, which is much houses. Since both heat and CO2 are supplied to the greenhouse from
lower than the annualized capital cost and is mainly associated with natural gas combustion, the total fuel cost is split into heat and CO2
electricity consumption for air compression. equally in view that natural gas is burned to provide CO2 in the summer
Our test results showed that it took about 3 h (2 h for adsorption and and day time, and for heating only during the night. Table 1 shows that
1 h for desorption) to complete one TSA cycle. Since there is no need for natural gas is a cheap source for greenhouse CO2 supply compared to
CO2 enrichment during the night when there is no light for plant liquid CO2. CO2 sequestrated from ambient air using CCS technologies
photosynthesis, the TSA unit is assumed to be operated only 12 h a day, reported in the open literature costs $130-$220. The CO2 unit price
or 4 cycles per day. The total amount of CO2 captured from air each from the proposed CEPS technology is around $120/T based on the
year is thus estimated based on the adsorption capacity of sorbent at the Sigma-Aldrich 13X tested in this study. However, the unit cost could be
selected operating conditions. The unit cost for supplying one tonne $70/T-CO2 based on 13X used by Wang et al. [28]. This cost is very
CO2 to the greenhouse is obtained by dividing the sum of annualized close to the natural gas option, and can be even lower ($40/T-CO2) if a
capital and operating costs by the total amount of CO2 captured. For the carbon capture credit of $30 (the carbon tax currently exercised in the
zeolite 13X purchased from Sigma-Aldrich, the adsorption capacity at province of British Columbia) is applied. Furthermore, we see a great
20 °C is about 0.09 mol/kg. The total amount of CO2 supplied to the potential for drastically reducing the CO2 unit cost to a level of $15-
greenhouse is estimated to be 85 tonnes per year, and the CO2 unit price $30/T-CO2 using the emerging amine-supported solids sorbents under
is $120/T-CO2. development.
In the proposed TSA process, the unit cost for enriched CO2 is in­ The greenhouse gas emission reduction potential of the CEPS
versely proportional to the CO2 adsorption capacity of the adsorbent. technology for the urban greenhouse industry consists of two compo­
To lower the unit CO2 capture cost, sorbent with a high adsorption nents: net CO2 sequestrated from the air by plants via photosynthesis
capacity and a low price ($2000/T for zeolite used in the current and displaced CO2 from natural gas combustion source. For each tonne
analysis) should be pursued. The sensitivity of CO2 unit cost to the CO2 captured from the air, one CO2 from natural gas combustion source
adsorption capacity and sorbent cost is plotted in Fig. 7. will be displaced. A fraction of this one tonne CO2 injected into the
As shown in Fig. 7, the CO2 unit cost is $120/T-CO2 based on the greenhouse will be also fixed by plants, depending on the plant growth
best adsorption/desorption performance data obtained from several rate and the residence time of the air in the greenhouse. Overall, more
samples tested so far in our lab. Wang et al. [28] reported an adsorption than one tonne net CO2 reduction will be achieved for each one CO2
capacity of 0.178 mol/kg at 60 °C for their tested zeolite 13X, which, if captured by adsorption column. To quantify the net CO2 reduction, a
used in the current design with the same operating temperatures and full life cycle analysis will be conducted in the future.
gas space velocity, would result a CO2 unit cost of $70/T-CO2 at a same
sorbent price of $2000/T.
It is also noted that much higher adsorption capacities in the range 3. Conclusions
of 1–2 mol-CO2/kg-sorbent have been reported in the literature using
advanced amine-supported adsorbents such as (tri-)amine-functiona­ We have proposed and demonstrated the potential of the Carbon
lized silica [37], hyperbranched aminosilica [38] and fumed silica Enrichment for Plant Stimulation (CEPS) technology for CO2 seques­
impregnated with polyethylenimine [39]. Again, with the assumption tration. In the future, the CEPS technology needs to be further devel­
of similar operating temperatures and space velocity as 13X sorbent oped in the following three areas: (1) Development of low-cost ad­
used in the current study but with an adsorption capacity of 1 mol/kg, vanced sorbents with a high adsorption capacity and fast regeneration
the CO2 unit cost is predicted to range from $15/T-CO2 at a sorbent rate at low temperatures; (2) Integration, demonstration and evaluation
price of $2000/T to $60/T at a sorbent price of $20,000/T, 10 times the in pilot and commercial greenhouses using different heat sources in­
price of 13X zeolite used in current study. It should be noted that most cluding geothermal and solar thermal; (3) A comprehensive assessment

188
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

Table 1
Estimated cost of CO2 supply for greenhouses from different sources.
Sources Processes Cost, $/T-CO2

Natural gas Natural gas burners, natural gas price of $8/GJ, 50%–50% cost split for heat and CO2 supplies 68

Liquid CO2 Food grade liquefied CO2 (Ontario Ministry of Agriculture and Food) 100–200

Air CCS NaOH scrubbing, causticization with lime, calcinations for regeneration [2] 136
Aerosol-type carbonator using Ca(OH)2,solar calciner & conventional slaker [40] 162–200
Spray tower with NaOH [41] ∼130
NaOH solution, causticization with lime, calcination-optimized APS Benchmark system [42] 309
Capture by anionic exchange resin, moisture swing desorption (steam flush) [1] 220
Capture by ammonium-based anion exchange resin, moisture swing adsorption (fixed bed)[4] 35
Temperature vacuum swing adsorption using MOF adsorbent [5] 75–140

Air CEPS 13X zeolite (Sigma-Aldrich), $2000/T, Tad = 10 °C, Tde = 30 °C, Qad = 50 L/min/kg, Qde = 20 L/min/kg, qad = 0.09 mol/kg 120
13X zeolite, $2000/T, qad = 0.178 mol/kg [28] 70
Amine-supported sorbents, $20,000/T, qad = 1 mol/kg 60
Amine-supported sorbents, $2000/T, qad = 1 mol/kg 15

of the CEPS integrated greenhouse system to quantify net greenhouse [12] J.V. Veselovskaya, et al., Direct CO2 capture from ambient air using K2CO3/Al2O3
gas reduction potential, financial benefits and other environmental and composite sorbent, Int. J. Greenhouse Gas Control 17 (2013) 332–340.
[13] C. Gebald, J.A. Wurzbacher, P. Tingaut, T. Zimmermann, A. Steinfeld, Amine-based
social benefits on local, regional and global scales. nanofibrillated cellulose As adsorbent for CO2 capture from air, Environ. Sci.
Technol. 45 (2011) 9101–9108.
Acknowledgements [14] Y. Kong, X. Shen, S. Cui, M. Fan, Facile synthesis of an amine hybrid aerogel with
high adsorption efficiency and regenerability for air capture via a solvothermal-
assisted sol–gel process and supercritical drying, Green Chem. 17 (2015)
Supports from Fuzhou University through a MinJiang Scholar pro­ 3436–3445.
gram and Natural Science and Engineering Research Council (NSERC) [15] J. Elfving, C. Bajamundi, J. Kauppinen, T. Sainio, Modelling of equilibrium working
capacity of PSA, TSA and TVSA processes for CO2 adsorption under direct air
of Canada through a discovery program are gratefully acknowledged.
capture conditions, J. CO₂ Utilization 22 (2017) 270–277.
[16] W.C. Wilfong, B.W. Kail, C.W. Jones, C. Pacheco, M.L. Gray, Spectroscopic in­
Author contributions vestigation of the mechanisms responsible for the superior stability of hybrid class
1/Class 2 CO2 sorbents: a new class 4 category, ACS Appl. Mater. Interfaces 8
(2016) 12780–12791.
J.B., W.-H.L and J.G.Z. conducted experiments and data analysis, [17] N. Sampson, D. Hair, Forests and Global Change, Vol. 2 Forest Management
and X.T.B. initiated and guided this work and prepared the manuscript. Opportunities for Mitigation of Carbon Emissions, Washington, 1996.
[18] Z.S. Chalabi, A. Birob, B.J. Baileya, D.P. Aikmanc, K.E. Cockshullc, Optimal control
strategies for carbon dioxide enrichment in greenhouse tomato crops part 1: using
Author information pure carbon dioxide, Biosystems Engineering. 83 (2002) 127.
[19] D.H. Willits, M.M. Peet, Predicting yield responses to different greenhouse CO2
The authors declare no competing financial interests. enrichment schemes – cucumbers and tomatoes, Agric. Forest Meteorol. 44 (1989)
275–293.
Correspondence and requests for materials should be addressed to [20] B. Tisserat, S.F. Vaughn, M.A. Berhow, Ultra high CO2 levels enhances cuphea
X.T.Bi. growth and morphogenesis, Indust. Crops Products 27 (2008) 133–135.
[21] S.D. Zhang, X.T. Bi, R. Clift, A life cycle assessment of integrated dairy farm-
greenhouse systems in British Columbia, Biore. Technol. 150 (2013) 496–505.
References [22] R. Sayre, Microalgae: the potential for carbon capture, Bioscience 60 (2010)
722–727.
[1] K. Lackner, Capture of carbon dioxide from ambient air, Eur. Phys. J. Spec. Top. 176 [23] G.W. Hickman, International Greenhouse Vegetable Production – Statistics (2014
(2009) 93–106. Edition), Cuesta Roble Greenhouse Vegetable Consulting. http://cuestaroble.com/
[2] D.W. Keith, M. Ha-Duong, J.K. Stolaroff, Climate strategy with CO2 capture from statistics.aspx, 2014.
the air, Climate Change 74 (2005) 17–45. [24] G. Smith, Practical Hydroponics and Greenhouses: Field vs. Glasshouse Tomatoes.
[3] K.Z. House, A.C. Baclig, M. Ranjan, E.A. van Nierop, J. Wilcox, H.J. Herzog, http://hydroponics.com.au/free-articles/issue-94-field-vs-glasshouse-tomatoes,
Economic and energetic analysis of capturing CO2 from ambient air, PNAS 108 2007.
(2011) 20428–20433. [25] W. Brilman, L.G. Alba, R. Veneman, Capturing atmospheric CO2 using supported
[4] C. Hou, Y. Wu, Y. Jiao, J. Huang, T. Wang, M. Fang, H. Zhou, Integrated direct air amine sorbents for microalgae cultivation, Biomass Bioenergy 53 (2013) 39–47.
capture and CO2 utilization of gas fertilizer based on moisture swing adsorption, J [26] T. Wang, J. Huang, X. He, J. Wu, M. Fang, J. Cheng, CO2 fertilization system in­
Zhejiang Univ-Sci A (Appl Phys & Eng). 18 (2017) 819–830. tegrated with a low-cost direct air capture technology, Energy Procedia. 63 (2014)
[5] A. Sinha, L.A. Darunte, C.W. Jones, M.J. Realff, Y. Kawajiri, Systems design and 6842–6851.
economic analysis of direct air capture of CO2 through temperature vacuum swing [27] Climeworks Switzerland, http://www.climeworks.com/, 2017.
adsorption using MIL-101(Cr)-PEI-800 and mmen-Mg2(dobpdc) MOF adsorbents, [28] Z.M. Wang, T. Arai, M. Kumagai, Adsorption separation of low concentrations of
Ind. Eng. Chem. Res. 56 (2017) 750–764. CO2 and NO2 by synthetic zeolites, Energy & Fuels. 12 (1998) 1055–1060.
[6] E.S. Sanz-Pérez, C.R. Murdock, S.A. Didas, C.W. Jones, Direct capture of CO2 from [29] S.U. Rege, R.T. Yang, M.A. Buzanowski, Sorbents for air prepurification in air se­
Ambient air, Chem. Rev. 116 (2016) 11840–11876. paration, Chem. Eng. Sci. 55 (2000) 4827–4838.
[7] K.S. Lackner, The thermodynamics of direct air capture of carbon dioxide, Energy [30] R.V. Siriwardane, M.S. Shen, E.P. Fisher, Adsorption of CO2 on zeolites at moderate
50 (2013) 38–46. temperatures, Energy Fuels 19 (2005) 1153–1159.
[8] K.S. Lackner, H. Ziock, P. Grimes, Carbon Dioxide Extraction from Air: Is It an [31] G. Li, et al., Capture of CO2 from high humidity flue gas by vacuum swing ad­
Option? Proceedings of the 24th Annual Technical Conference on Coal Utilization & sorption with zeolite 13X, Adsorption-J. Int. Adsorption Soc. 14 (2008) 415–422.
Fuel Systems, Clearwater, FL, 1999 pp. 885-896. [32] F.S. Su, C.Y. Lu, CO2 capture from gas stream by zeolite 13X using a dual-column
[9] F.S. Zeman, K.S. Lackner, Capturing carbon dioxide directly from the atmosphere, temperature/vacuum swing adsorption, Energy Environ. Sci. 5 (2012) 9021–9027.
World Res. Rev. 16 (2004) 157–172. [33] S. Sircar, W.C Kratz, Removal of water and carbon dioxide from air. US Patent
[10] D.W. Keith, M. Ha-Duong, J.K. Stolaroff, Climate strategy with CO2 capture from 4249915, 1981.
the air, Climatic Change 74 (2006) 17–45. [34] R. Kumar, Removal of water and carbon dioxide from atmospheric air. US Patent
[11] R. Baciocchi, G. Storti, M. Mazzotti, Process design and energy requirements for the 4711645, 1987.
capture of carbon dioxide from air, Chem. Eng. Process. 45 (2006) 1047–1058. [35] R. Jain, Pre-puri“cation of air for separation. US Patent 5232474, 1993.

189
J. Bao, et al. Carbon Resources Conversion 1 (2018) 183–190

[36] G.D. Ulrich, Chemical Engineering Process Design and Economics: A Practical (2011) 2420–2427.
Guide, second ed., Process Publishing, 2004. [40] V. Nikulshina, D. Hirsch, M. Mazzotti, A. Steinfeld, CO2 capture from air and co-
[37] Y. Belmabkhout, R. Serna-Guerrero, A. Sayari, Amine-bearing mesoporous silica for production of H2 via the Ca(OH)2-CaCO3 cycle using concentrated solar power –
CO2 removal from dry and humid air, Chem. Eng. Sci. 65 (2010) 3695–3698. thermodynamic analysis, Energy 31 (2006) 1715–1725.
[38] A. Goeppert, M. Czaun, R.B. May, G.K.S. Prakash, G.A. Olah, S.R. Narayanan, [41] J.K. Stolaroff, D.W. Keith, G.V. Lowry, Carbon dioxide capture from atmospheric air
Carbon dioxide capture from the air using a polyamine based regenerable solid using sodium hydroxide spray, Environ. Sci. Technol. 42 (2008) 2728–2735.
adsorbent, J. Am. Chem. Soc. 133 (2011) 20164–20167. [42] F. Zeman, Reducing the cost of Ca-based direct air capture of CO2, Environ. Sci.
[39] S. Choi, J.H. Drese, P.M. Eisenberger, C.W. Jones, Application of amine-tethered Technol. 48 (2014) 11730–11735.
solid sorbents for direct CO2 capture from the ambient air, Environ. Sci. Technol. 45

190