**4. CCS benefits**

#### **4.1. Private benefits and public benefits from employed CCS technologies**

The private benefits of carbon capture and storage include the proven ability of injecting CO<sup>2</sup> underground into geologic crude oil and natural gas deposits to enhance extraction of oil and gas from these deposits. These benefits can be quantified by multiplying the price of the additional crude oil or natural gas extracted as a result of CO2 injection by the going market price of oil or gas. As this chapter is currently being written in early 2016, the real prices of crude oil and natural gas resources received by oil and gas producers are at record lows worldwide. These relatively low prices have a negative impact on the private benefits of CO<sup>2</sup> injection projects for enhancing oil and gas projects.

Thus, a c ritical component of whether or not such projects will be economically feasible to oil and gas companies is the expected price path of future oil and gas prices. Such price paths are difficult to estimate empirically [2, 8, 11]. However, based on economic theory and Hotelling's rule in particular, we expect theoretically that the market price of any exhaustible, non-renewable natural resource, including crude oil and natural gas, to follow an upward-sloping price path in the long run as the resource becomes scarcer.

³For consistency, in the chapter the units of MtC and US \$/tC are being used to describe economic values for marginal costs and benefits on average, assuming MC ≈ AC in all long-run CCS operations. We will note where the units of MtCO<sup>2</sup> and US \$/CO2 are applied as alternative measures. All are equivalent: (1) US \$27.3/tCO2 (= US \$100/tC) [26]; (2) US \$10/ tCO2 is approximately equivalent to US \$37/tC [15].

⁴Because the atomic weights of carbon are 12 atomic mass units and carbon dioxide is 44 atomic mass units, a ratio factor of 3.67, or 44/12, is used, meaning one ton of carbon equals 3.67 tons of carbon dioxide, which can also approximately equal 1 tC = 3.7 tCO2 (as computed from (US \$37/tC)÷(US \$10/tCO2 ), or 3.66 tCO2 = (US \$100/tC) ÷ (US \$27.3/tCO2 )). However, only the factor of 3.67 is applied for computation of all estimates in this chapter.

⁵These costs can be easily and quickly observed in Anderson and Newell (2004) (please see Table 3 in Ref. [8]).

<sup>6</sup> The later measures may be slightly higher after having been adjusting for inflation over time. Assuming a gas price of US \$3 per million Btu (MBtu), which was the average price over the past decade, transport and storage costs of \$37/tC stored were reported in [8]. Moreover, one can apply the following formulas to see how adjusted/expected benefits and costs are affected by inflation rates over time, that is, adjusted benefits in current-year = dollars in base-year × (CPICurrent-year/ CPIBase-year), and adjusted costs in current-year = dollars in base-year × (PPICurrent-year/PPIBase-year), where CPI is the consumer price index, and PPI is the producer price index.

From economic theory, we can also predict that increasing market prices of exhaustible, nonrenewable energy resources such as crude oil and natural gas will eventually lead to the substitution of these relatively high-cost energy sources by relatively cheaper energy sources. For example, sometime in the future it may be economically feasible and desirable to shift completely over to some "backstop technology" for producing energy including solar and wind power and the "holy grail" of virtually unlimited energy production—nuclear fusion ([5], Chapter 3).

In addition to the private benefits to oil and gas companies of CCS projects that enhance oil and gas production, private benefits of CCS projects as a whole also include private benefits of global warming mitigation such as reduced health costs to individuals, reduced damages to agricultural crops, and reduced damages to human-built structures in flood-prone areas. These private benefits can be quantified using private health-care expenditures, the market value of agricultural crops, and the costs of replacing or repairing human-built structures [16].

There are also many public benefits of CCS projects associated with global warming mitigation. These public benefits include economic values associated with protecting fish and wildlife habitat (e.g., Polar Bear habitat in Artic regions) and human cultures (e.g., Indigenous, Native or First-Peoples in Artic regions). Non-market economic valuation techniques including contingent valuation and choice experiments can be used to quantify these types of nonmarket benefits ([5], Chapter 13).

The total benefits of carbon capture and storage (*<sup>T</sup> Bccs* )can be expressed in the equation form as

$$TB\_{\rm css} = \left(TP \, B\_{\rm cs}^{\ast}\right) + \left(TS \, B\_{\rm cs}^{\ast}\right) \tag{3}$$

where,

dollar per ton carbon dioxide (\$/tCO2

246 Recent Advances in Carbon Capture and Storage

between US \$225/tC and \$315/tC (or US \$61/CO2

storage, which range from US \$5/tC to \$10/tC per 100 km [8]<sup>5</sup>

tional crude oil or natural gas extracted as a result of CO2

projects for enhancing oil and gas projects.

the long run as the resource becomes scarcer.

is approximately equivalent to US \$37/tC [15].

price index, and PPI is the producer price index.

stored at the storage site, which range from US \$20/tC to \$55/tC [14].6

**4.1. Private benefits and public benefits from employed CCS technologies**

dioxide.4

**4. CCS benefits**

tCO2

6

equal 1 tC = 3.7 tCO2

),3

According to recent literature, an estimated avoided total cost of CCS per unit (MCccs) is

in MCccs can arise in the near future because of continuously technological improvements in CCS [8]. To give a breadth of findings, estimates of marginal cost avoided can be shared in three cost components: (1) marginal costs of carbon captured at the point of source, which range from US \$200/tC to \$250/tC [8]; (2) marginal costs of captured carbon transportation to

The private benefits of carbon capture and storage include the proven ability of injecting CO<sup>2</sup> underground into geologic crude oil and natural gas deposits to enhance extraction of oil and gas from these deposits. These benefits can be quantified by multiplying the price of the addi-

of oil or gas. As this chapter is currently being written in early 2016, the real prices of crude oil and natural gas resources received by oil and gas producers are at record lows worldwide.

Thus, a c ritical component of whether or not such projects will be economically feasible to oil and gas companies is the expected price path of future oil and gas prices. Such price paths are difficult to estimate empirically [2, 8, 11]. However, based on economic theory and Hotelling's rule in particular, we expect theoretically that the market price of any exhaustible, non-renewable natural resource, including crude oil and natural gas, to follow an upward-sloping price path in

³For consistency, in the chapter the units of MtC and US \$/tC are being used to describe economic values for marginal costs and benefits on average, assuming MC ≈ AC in all long-run CCS operations. We will note where the units of MtCO<sup>2</sup>

⁴Because the atomic weights of carbon are 12 atomic mass units and carbon dioxide is 44 atomic mass units, a ratio factor of 3.67, or 44/12, is used, meaning one ton of carbon equals 3.67 tons of carbon dioxide, which can also approximately

The later measures may be slightly higher after having been adjusting for inflation over time. Assuming a gas price of US \$3 per million Btu (MBtu), which was the average price over the past decade, transport and storage costs of \$37/tC stored were reported in [8]. Moreover, one can apply the following formulas to see how adjusted/expected benefits and costs are affected by inflation rates over time, that is, adjusted benefits in current-year = dollars in base-year × (CPICurrent-year/ CPIBase-year), and adjusted costs in current-year = dollars in base-year × (PPICurrent-year/PPIBase-year), where CPI is the consumer

⁵These costs can be easily and quickly observed in Anderson and Newell (2004) (please see Table 3 in Ref. [8]).

), or 3.66 tCO2

and US \$/CO2 are applied as alternative measures. All are equivalent: (1) US \$27.3/tCO2

(as computed from (US \$37/tC)÷(US \$10/tCO2

However, only the factor of 3.67 is applied for computation of all estimates in this chapter.

These relatively low prices have a negative impact on the private benefits of CO<sup>2</sup>

and \$86/tCO2

where one ton of carbon equals 3.67 tons of carbon

), but a considerable reduction

; and (3) marginal costs of carbon

injection by the going market price

(= US \$100/tC) [26]; (2) US \$10/

= (US \$100/tC) ÷ (US \$27.3/tCO2

injection

)).

*TP B*ccs *<sup>s</sup>* is the total private benefits of carbon captured and stored

*TS B*ccs *<sup>s</sup>* is the total social benefits of carbon captured and stored.

For economic efficiency purposes, we must also measure the marginal benefits of humanengineered CCS. The short-run marginal costs (MBCCS) of human-engineered CCS are defined as

$$MB\_{\rm ccs} = \frac{\partial TB\_{\rm ccs}}{\partial Q\_{\rm c0}} \tag{4}$$

where *Q*CO2 is the quantity of CO2 captured and stored.

#### **4.2. Measures of total marginal benefits of CSS**

In Section 4.1, we describe that the total benefits of carbon capture and storage technologies are received by both public and private entities. For economic efficiency analyses, we use the total marginal benefits of human-engineered CCS (MBccs) given by the private marginal benefits of carbon captured and stored (PMBccs) plus the social marginal benefits (SMBs) of carbon captured and stored (SMBccs). In the following sections, we discuss quantitative estimates of the private marginal benefits of CCS (PMBccs) and the social marginal benefits of CCS (SMBccs).

#### *4.2.1. Private marginal benefits and carbon capture and utilization*

A Canadian Pembina Institute Publication [17] reported post-CO2 capture diverging into two pathways—carbon sequestrations (CCS) (already discussed so far) and carbon capture and utilization (CCU) (discuss in this subsection). CCU applications fall under two main approaches: the conversion approach and nonconversion approach.7 Since the twenty-first century, technological advances have made various CCU applications under these two main approaches more practical and profitable [18–20].

CCU conversion approach applications range from mineralization (e.g. varied utilized forms of carbonate applications), biological transformation (e.g. algae cultivation applications), and chemical transformation (e.g. liquid fuel applications including methanol, polymer/chemical feedstock, and urea yield boosting<sup>8</sup> ). CCU non-conversion approach applications are generally aimed for the purposes of desalination and enhanced techniques including enhanced oil recovery (EOR), enhanced geothermal systems, and enhanced coalbed methane [17].

Thus, the economics of CCU technologies then lies in potential net benefits received from reutilizations of the captured CO2 . Within 5–10 years, CCU conversion approaches including mineralization (considered as permanent-based performance) and biological and chemical innovations (considered as non-permanent-based performance) have been estimated to be utilized in a range of 5 to more than 300 MtCO2 per year [17]. Within the same time frame, CCU nonconversion approaches will yield, in both permanent and non-permanent potential performance, an estimated 5–300 MtCO2 in enhanced techniques and between 30 and 300 MtCO2 in desalination [17].9

According to Refs. [21, 22], it is estimated that each year about 80%, or 9 million metric tons (MtC) of captured CO2 used by commercial industry, are in EOR operations. The net marginal benefits (PMBccs) of stored carbon to EOR and enhanced coal-bed methane recovery operations have been estimated in the range of US \$15/tC to \$30/tC [23].10 There certainly exist

⁷CCU is also called carbon capture and reuse or carbon capture and recycling (CCR) [17].

⁸Urea, also known as carbamide, is an organic compound with the chemical formula CO(NH2)2 and one of the most common forms of solid nitrogen fertilizer. Urea is produced by the reaction between ammonia and CO2. See ([24], Appendix B).

⁹Permanent and non-permanent potential performances are referred to permanent and non-permanent storage. According to the Global CCS institute, reuse technologies that permanently store CO2 are considered to be an alternative form of CCS and referred to as "alternative CCS." EOR, ECBM, EGS, carbonate mineralization, concrete curing, bauxite residue carbonation, and potentially algae cultivation (depending on the end product) are considered to be alternative forms of CCS. See ([24] Part I: Section 3.2).

¹⁰In recent work [25], it was estimated that EOR storage of CO2 could generate net benefits as high as \$335/tC stored, or cost as much as \$270/tC stored. In a base-case calculation, EOR generates average net benefits of about \$45/tC stored [8].

additional net benefits from other applications described above, but empirical estimates of these benefits are not yet available.

#### *4.2.2. Social marginal benefits*

benefits of carbon captured and stored (PMBccs) plus the social marginal benefits (SMBs) of carbon captured and stored (SMBccs). In the following sections, we discuss quantitative estimates of the private marginal benefits of CCS (PMBccs) and the social marginal benefits of

two pathways—carbon sequestrations (CCS) (already discussed so far) and carbon capture and utilization (CCU) (discuss in this subsection). CCU applications fall under two main

century, technological advances have made various CCU applications under these two main

CCU conversion approach applications range from mineralization (e.g. varied utilized forms of carbonate applications), biological transformation (e.g. algae cultivation applications), and chemical transformation (e.g. liquid fuel applications including methanol,

applications are generally aimed for the purposes of desalination and enhanced techniques including enhanced oil recovery (EOR), enhanced geothermal systems, and enhanced coal-

Thus, the economics of CCU technologies then lies in potential net benefits received from

mineralization (considered as permanent-based performance) and biological and chemical innovations (considered as non-permanent-based performance) have been estimated to be

CCU nonconversion approaches will yield, in both permanent and non-permanent potential

According to Refs. [21, 22], it is estimated that each year about 80%, or 9 million metric tons

ginal benefits (PMBccs) of stored carbon to EOR and enhanced coal-bed methane recovery operations have been estimated in the range of US \$15/tC to \$30/tC [23].10 There certainly exist

⁸Urea, also known as carbamide, is an organic compound with the chemical formula CO(NH2)2 and one of the most common forms of solid nitrogen fertilizer. Urea is produced by the reaction between ammonia and CO2. See ([24],

⁹Permanent and non-permanent potential performances are referred to permanent and non-permanent storage. According to the Global CCS institute, reuse technologies that permanently store CO2 are considered to be an alternative form of CCS and referred to as "alternative CCS." EOR, ECBM, EGS, carbonate mineralization, concrete curing, bauxite residue carbonation, and potentially algae cultivation (depending on the end product) are considered to be alternative

¹⁰In recent work [25], it was estimated that EOR storage of CO2 could generate net benefits as high as \$335/tC stored, or cost as much as \$270/tC stored. In a base-case calculation, EOR generates average net benefits of about \$45/tC stored [8].

⁷CCU is also called carbon capture and reuse or carbon capture and recycling (CCR) [17].

capture diverging into

Since the twenty-first

). CCU non-conversion approach

. Within 5–10 years, CCU conversion approaches including

used by commercial industry, are in EOR operations. The net mar-

per year [17]. Within the same time frame,

in enhanced techniques and between 30 and 300

*4.2.1. Private marginal benefits and carbon capture and utilization*

approaches more practical and profitable [18–20].

utilized in a range of 5 to more than 300 MtCO2

performance, an estimated 5–300 MtCO2

in desalination [17].9

forms of CCS. See ([24] Part I: Section 3.2).

(MtC) of captured CO2

polymer/chemical feedstock, and urea yield boosting<sup>8</sup>

A Canadian Pembina Institute Publication [17] reported post-CO2

approaches: the conversion approach and nonconversion approach.7

CCS (SMBccs).

248 Recent Advances in Carbon Capture and Storage

bed methane [17].

MtCO2

Appendix B).

reutilizations of the captured CO2

Unlike quantifying direct total private benefits, an attempt to measure public or social benefits can be quite a challenging task for researchers since there involve concepts of types of costs and nonmarket valuations of public goods and services provided to the population.

Before it is attempted to explain how social marginal benefits arrive in this subsection of the chapter, there are three concepts needed to explain since we simply use the reported range (not derived explicitly) of the estimates for SMB from various sources. First of all, we simply define that private costs are the costs that individual decision makers are facing given actual established market prices. Second, social costs are the private costs plus the costs of economic externalities on society. These social costs are the prices derived from market prices, where opportunity costs are taken into account. Finally, social cost of carbon (SCC) is the discounted monetized sum of the annual net losses from impacts caused by an additional unit of carbon emitted presently and is measured in US \$/tC or US \$/tCO<sup>2</sup> ([26] Chapter 3, p. 135).

According to the economic theory, at an economically efficient mitigation level the marginal social benefits of carbon reduction (SMB) are equal to the social costs of carbon, where SCC is defined as avoided total damages for an additional ton of carbon abated ([26], Chapter 3, p. 233). Thus, using estimates of SCC ([26], Chapter 20) and the assumption that SCC = SMB at an economically efficient carbon price, we can infer estimates of SMBCCS currently in the range US \$14/ tC to \$350/tC (or US \$4/tCO2 to \$95/tCO2 ).11 By assuming a 2.4% per year increase in emissions, the estimated range for SMBCCS in the year 2030 is between US \$29/tC and \$694/tC (or US \$8/ tCO2 to 189/tCO2 ).12 By adding private marginal benefits (PMBCCS) from the previous section to social marginal benefits (SMBccs) from the current section, we estimate total marginal benefits of CCS (MBccs) to fall in a range of US \$29/tC–\$380/tC currently to US \$49/tC–\$735/tC in 2030.13

## **5. Optimal CCS provision**

#### **5.1. Concept of economically efficient level of CSS size**

According to economic efficiency, the optimal level of carbon capture and storage is where the marginal benefits and marginal costs of CO<sup>2</sup> captured and stored are equal. In **Figure 1**, we show the marginal benefit curve for CCS (MBCCS), and the marginal cost curve for CCS (MCCCS). The marginal benefit curve is downward sloping because, following the law of diminishing

¹¹Median and 95th percentile estimates reported in [27].

¹²The estimated social cost of carbon reported by [28] including uncertainty, equity weighting, and risk aversion is \$44 per ton of carbon (or \$12 per ton CO2) in 2005 US\$. Second, including uncertainty increases the expected value of the SCC by approximately 8%. Finally, equity weighting generally tends to reduce the SCC.

¹³For consistency, we assume there is also a 2.4% per year increase in the PMBCCS reported in [23]. Thus, for 2030 the estimated range for PMBCCS is between US \$20/tC and \$41/tC.

**Figure 1.** Economically efficient level of CCS.

returns, each additional unit of CO2 captured and stored provides less private and social benefits. The marginal cost curve is upward sloping because both the private and social costs of CCS go up with each additional unit of CO2 captured and stored. The upward-sloping nature of the marginal cost curve indicates that it would be very expensive (and likely cost prohibitive) to capture and store 100% of all CO2 found in emissions from a point source such as a coal-fired power plant or industrial factory.

The economically efficient level of CCS (Q\*) is shown graphically in **Figure 1** where the marginal benefit curve and marginal cost curve for CCS cross; at this point,

$$\mathbf{M} \, \mathbf{C}\_{\text{CCS}} = \frac{\partial T \, \mathbf{C}\_{\text{CC}}}{\partial Q\_{\text{CO}\_{i}}} = M \, B\_{\text{CCS}} = \frac{\partial T B\_{\text{CC}}}{\partial Q\_{\text{CO}\_{i}}} \tag{5}$$

If all private and social benefits and costs of CCS could be "internalized" into economic markets, transactions between buyers and sellers could lead automatically to an economically efficient level of CCS, given certain conditions (e.g., perfect competition). It is notoriously difficult, however, to "internalize" all social benefits and costs because of the public good (or "bad") characteristics of these benefits and costs such as nonexclusiveness and nonrivalry. Thus, achieving an economically efficient level of CCS would most likely require some degree of government intervention into markets such as economic incentives (e.g., taxes and subsidies) and/or direct regulation ([5], Chapter 10).

#### **5.2. Estimates of CSS optimal level**

As previously described in this chapter, under the condition where marginal benefits and marginal costs of CO2 captured and stored are equal, there exists a relationship between the optimal carbon price and the optimal level of carbon capture and storage. For a given carbon price range of US \$146–\$257/tC (or US \$40–\$70/tCO2 ), the optimal level of CO2 captured and stored is in the estimated range of 0–8MtC (or 0–29.48MtCO2 ) per year [29, 30].

#### **6. Summary and conclusions**

From a public policy perspective, since the general public also benefits from carbon dioxide being captured, stored, and prevented from entering the atmosphere, there is economic justification for public policies targeted at providing economic incentives for private companies to invest in CCS technology, such as direct subsidies or tax breaks. Whether or not CCS technology will prove to be one of the "tools" in the global warming, mitigation "tool box" in the long run is yet to be seen.

In addition to the Petra Nova project in the United States, private companies in Canada, Germany, and China are investing in large-scale CCS projects, with mixed economic feasibility results from a private firm perspective. Scaling-up from the private firm level to the society level where public benefits from global warming mitigation are taken into account, the private and public economic benefits of CCS projects seem likely to outweigh the private costs. Thus, public polices, which help private companies to defray the high costs of large-scale CCS projects, may be justified from an overall benefit-cost analysis perspective.

#### **Author details**

returns, each additional unit of CO2

**Figure 1.** Economically efficient level of CCS.

250 Recent Advances in Carbon Capture and Storage

to capture and store 100% of all CO2

power plant or industrial factory.

CCS go up with each additional unit of CO2

*<sup>M</sup> <sup>C</sup>*CCS <sup>=</sup> <sup>∂</sup>*<sup>T</sup> <sup>C</sup>* \_\_\_\_\_\_ CCS

dies) and/or direct regulation ([5], Chapter 10).

captured and stored provides less private and social ben-

found in emissions from a point source such as a coal-fired

∂*Q*CO2

captured and stored. The upward-sloping nature

(5)

efits. The marginal cost curve is upward sloping because both the private and social costs of

of the marginal cost curve indicates that it would be very expensive (and likely cost prohibitive)

The economically efficient level of CCS (Q\*) is shown graphically in **Figure 1** where the mar-

If all private and social benefits and costs of CCS could be "internalized" into economic markets, transactions between buyers and sellers could lead automatically to an economically efficient level of CCS, given certain conditions (e.g., perfect competition). It is notoriously difficult, however, to "internalize" all social benefits and costs because of the public good (or "bad") characteristics of these benefits and costs such as nonexclusiveness and nonrivalry. Thus, achieving an economically efficient level of CCS would most likely require some degree of government intervention into markets such as economic incentives (e.g., taxes and subsi-

<sup>=</sup> *<sup>M</sup> <sup>B</sup>*CCS <sup>=</sup> <sup>∂</sup>*<sup>T</sup> <sup>B</sup>* \_\_\_\_\_\_ CCS

ginal benefit curve and marginal cost curve for CCS cross; at this point,

∂*Q*CO2

John C. Bergstrom\* and Dyna Ty

\*Address all correspondence to: jberg@uga.edu

University of Georgia, Athens, GA, USA

#### **References**


[19] Kenyon D. The technologies behind carbon utilization. Calgary, Canada: The Pembina Institute; 2015.

[4] Pires J.C.M., Martins F.G., Alvim-Ferraz M.C.M., and Semoes M. Recent developments on carbon capture and storage: an overview. Chemical Engineering Research and

[5] Bergstrom J.C. and Randall A. Resource economics: an economic approach to natural resource and environmental policy, 4th edition. Cheltenham, UK and Northampton,

[6] Nichani K., Bapat S., and Kumari A. A study of combating climate change with carbon capture and sequestration. International Journal of Engineering and Technical Research.

[7] Jenkins J. Financing mega-scale energy projects: a case study of the petra nova carbon capture project. University of Chicago, Chicago, IL, USA: Paulson Institute; 2015.

[8] Anderson S. and Newell R. Prospects for carbon capture and storage. Annual Review of

[9] Herzog H.J. Scaling up carbon dioxide capture and storage: from megatons to gigatons.

[10] Azar C., Lindgren K., Larson E., and Mollersten K. Carbon capture and storage from fossil fuels and biomass—costs and potential role in stabilizing the atmosphere. Climate

[11] Rubin E.S., Chen C., and Rao A.B. Cost and performance of fossil fuel power plants with

for mitigating global climate change: a white paper. Cambridge, MA: Massachusetts

[15] David J. and Herzog H. J. The cost of carbon capture. Presented Paper, Fifth International Conference on Greenhouse Gas Control Technologies; 13–16 August 2000; Cairns, Australia.

[16] Baker E., Chon H., and Keisler J. Carbon capture and storage: combining economic analysis with expert elicitations to inform climate policy. Climate Change.

[17] Kenyon D. and Jeyakumar B. Carbon capture and utilization. Calgary, Canada: The Pembina Institute; March 2015. [Cited September 2016]. Available from https://www.

[18] International Energy Agency (IEA). Tracking clean energy progress 2015: energy technology perspectives 2015. Excerpt IEA Input to the Clean Energy Ministerial, IEA

sequestration. Science. 2003;300(13):1677–1678.

capture, reuse, and storage technologies

capture and storage. Energy Policy. 2007;35:4444–4454.

[13] Rhodes C.J. Carbon capture and storage. Science Progress. 2012;95(4):473–483.

Design. 2011;89:1446–1460.

252 Recent Advances in Carbon Capture and Storage

MA, USA: Edward Elgar; 2016.

Environmental Resources. 2004;29:109–142.

Energy Economics. 2011;33(4):597–604.

[14] Herzog H.J., Drake E., and Adams E. CO<sup>2</sup>

Institute of Technology Energy Laboratory; 1997.

pembina.org/pub/carbon-capture-and-utilization

2014;2(9):269–272.

Change. 2006;74:47–79.

[12] Lackner K.S. A guide to CO<sup>2</sup>

2009;96:379–408.

Publications; 2015.

CO2

