The Greenhouse-Gas Impact of Various Chiller Technologies

Editor's note: The following article is based on a presentation given by the author during HPAC Engineering's 2009 Engineering Green Buildings (EGB) Conference and Expo, held Sept. 24 and 25 in Nashville, Tenn., part of HVACR Week. HPAC Engineering is accepting proposals for sessions for EGB 2010, to be held Sept. 23 and 24 in Baltimore. For more information, contact Executive Editor Scott Arnold at 216-931-9980 or [email protected].

Through 2009, 663 college and university presidents had signed the American College & University Presidents' Climate Commitment (ACUPCC) (www.presidentsclimatecommitment.org), pledging "to eliminate net greenhouse-gas emissions from specified campus operations and ...promote ... research and educational efforts ... to equip society to re-stabilize the earth's climate." In part, the ACUPCC states the signatories' agreement to "initiate the development of a comprehensive plan to achieve climate neutrality as soon as possible."

Comparing Carbon Footprints

With such a daunting goal, discussion of the engineering fundamentals of the production of greenhouse gases is in order. This article will discuss the production of the most prevalent greenhouse gas, carbon dioxide (CO₂), by an energy-consuming system found on many college and university campuses: the chilled-water system. Specifically, it will focus on the central energy consumer of that system, the water chiller.

In the United States, more than 83 percent of utilized energy originates from a combustion process.¹ This includes fuels burned to create steam and hot water for heating and cooling and fuels burned to create steam to turn turbines/generators to produce electricity. Although this article will focus on the combustion of coal and natural gas, the logic applies for all combustible hydrocarbon fuels, including propane, No. 2 fuel oil, No. 6 fuel oil, and gasoline.

Determining emission factors is a site-specific task because energy sources vary by region, as do methods of generating and transporting electric energy to electric infrastructure. Many colleges and universities determine their emissions using Campus Carbon Calculator, available from Clean Air-Cool Planet (www.cleanair-coolplanet.org). Campus Carbon Calculator divides the nation into sub-regions and uses U.S. Environmental Protection Agency data to determine regional emissions. Additional supporting information can be obtained from the U.S. Energy Information Administration (www.eia.doe.gov). Utilizing Campus Carbon Calculator Version 6.4, CO₂-emission factors were determined for a chiller project in St. Louis:

• Coal combustion

  The average to be used for St. Louis is 207.9 lb of CO₂ per million British thermal units of coal burned, although the Wyoming Powder Valley coal largely burned by the local electric utility indicates a slightly higher value (see sidebar).
• Natural-gas combustion

  The average to be used for St. Louis is 116.3 lb of CO₂ per million British thermal units of natural gas burned, although the fuel furnished by the local gas utility indicates a slightly higher value (see sidebar).

• Electricity generation

  The average to be used for St. Louis is 1.844 lb of CO₂ per kilowatt-hour of electricity delivered. It is based on a mix of energy sources, including coal, natural gas, nuclear, hydroelectric, and wind. For reference, the use of coal alone would produce 2.19 lb of CO₂ per kilowatt-hour of electricity delivered, while the use of natural gas alone would produce 1.21 lb of CO₂ per kilowatt-hour of electricity delivered (see sidebar). Obviously, electricity in St. Louis is generated predominately through the combustion of coal.

The carbon footprints of eight types of chillers common to college and university campuses were determined. The chillers are:

• High-efficiency electric centrifugal ("Option 1" in Table 1).

• Standard-efficiency electric centrifugal ("Option 2").

• Single-stage steam absorption ("Option 3").

• Two-stage steam absorption ("Option 4").

• Gas-fired two-stage absorption ("Option 5").

• Natural-gas-engine/generator-powered electric without heat recovery ("Option 6").

• Natural-gas-engine/generator-powered electric with heat recovery ("Option 7").

• Electric centrifugal heat recovery ("Option 8").

Page 2 of 9

The design operating conditions were a chiller-module size of 600 tons of cooling capacity, an evaporator design based on 1,200 gpm of chilled water cooled from 54°F to 42°F, and a condenser design based on 85°F condenser water available at a flow rate capable of producing a 12°F rise in the condenser-water stream.

Only full-load conditions were compared. For each chiller, the following were considered:

• Input energy -- the electricity, steam, or natural gas used to power the chiller and auxiliary equipment (e.g., pumps, cooling tower) or the natural gas used to power an engine creating electricity to power the chiller and auxiliary equipment.

• Recovered heat -- heat recovered from the condenser-water stream or the engine/generator producing the input electricity.

Input energy and recovered heat were determined using manufacturer data. Cooling-tower-fan power is the catalogued power needed for heat rejection. For absorption chillers, input energy included catalogued power needs for solution, refrigerant, and vacuum pumps. For chilled-water and condenser-water pumps, chiller input-power contribution was calculated based on catalogued evaporator or condenser head loss and the following formula:

Pump power (kw) = (gpm x ΔH x 0.746) ÷ (3,960 x η_pump x η_motor)

where:

ΔH = evaporator or condenser head loss (feet) catalogued by chiller manufacturer

η_pump = pump efficiency (70 percent assumed)

η_motor = motor efficiency (90 to 93 percent assumed, depending on motor size)

The catalogued values for each of the chiller options are given in Table 1.

High-efficiency electric centrifugal chiller

Figure 1 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a high-efficiency electric centrifugal chiller. Power for the chiller and auxiliary equipment is on the left, with cooling output on the right and rejected heat above. (For clarity, chiller ton-hours produced is symbolized by an arrow leaving the process; more accurately, this is heat energy removed from the chilled-water system by the chiller.) This chiller is the yardstick by which the others will be compared.

Standard-efficiency electric centrifugal chiller

Figure 2 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a standard-efficiency electric centrifugal chiller. Six-percent-higher kilowatts per ton, 6-percent-higher emissions ¡ª no surprise here.

Page 3 of 9

Single-stage steam absorption chiller

Figure 3 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a single-stage steam absorption chiller. Natural gas fuels a boiler producing low-pressure steam at an assumed efficiency of 80 percent. This likely casts the system in the best-possible light, as Energy Star would assume an efficiency, including distribution losses, for a central plant more on the order of 69 percent.² Input energy includes natural gas as well as electricity for auxiliary equipment. CO₂ emissions are 132-percent higher than those of the high-efficiency electric centrifugal chiller.

Two-stage steam absorption chiller

Figure 4 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a two-stage steam absorption chiller. The basic parameters are the same as those of the single-stage steam absorption chiller, except high-pressure, 100-psig steam is utilized, increasing the chiller coefficient of performance from 0.71 to 1.35. Input energy includes natural gas as well as electricity for auxiliary equipment. CO₂ emissions are 29-percent higher than those of the high-efficiency electric centrifugal chiller.

Gas-fired two-stage absorption chiller

Figure 5 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a gas-fired two-stage absorption chiller. Notice the absence of a boiler because natural gas is utilized directly in the chiller. Input energy includes natural gas as well as electricity for auxiliary equipment. This proves to be the best of the absorption-chiller options, but CO₂ emissions still are 24-percent higher than those of the high-efficiency electric centrifugal chiller.

Natural-gas-engine/generator-powered electric chiller without heat recovery

Figure 6 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a natural-gas-engine/generator-powered electric chiller without engine heat recovery. In this case, a nominal 400-kw natural-gas-fired engine/generator set is used to provide electricity to power the high-efficiency electric centrifugal chiller, along with electric auxiliary equipment, of Option 1. CO₂ emissions are 26-percent below those of Option 1, but at the cost of a significant increase in system complexity.

Page 4 of 9

Natural-gas-engine/generator-powered electric chiller with heat recovery

Figure 7 shows the pounds of CO₂ resulting from the production of 1 ton-hr of cooling by a natural-gas-engine/generator-powered electric chiller with heat recovery. In this case, the engine/generator-powered electric centrifugal chiller of the previous option has been equipped with a full engine heat-recovery system, one that includes engine jacket water and engine exhaust. For a typical internal-combustion, natural-gas-fired engine of this size, the shaft energy produced would be about 32 percent of the input energy, leaving about 68 percent to be dissipated as heat. For typical heat-recovery applications, the dissipated heat can be recovered as a percentage of input energy largely as follows:

• Jacket heat recovery (180°F water), 30 percent.

• Exhaust heat recovery (steam or hot water), 18 percent.

• Lube-oil/aftercooler heat recovery (130°F water), 8 percent (not utilized in this case).

• Unrecoverable losses, 12 percent.

For this application, then, 48 percent of the heat was considered recovered from the input-energy stream through jacket water and exhaust heat. Additional auxiliary electricity was considered for the jacket-water and exhaust heat-recovery pumping systems.

The natural gas not burned because of useful heat recovery becomes a credit to the process, reducing emissions to 0.35 lb of CO₂ per ton-hour of cooling provided, or 70-percent below the emissions of the high-efficiency electric centrifugal chiller (Figure 8). If heat is not recovered, emissions will be similar to those of a natural-gas-engine/generator-powered electric chiller without engine heat recovery (0.86 lb of CO₂ per ton-hour). Every real heat-recovery system operates somewhere between those two extremes.

Unfortunately, a natural-gas-engine/generator-powered electric chiller with heat recovery adds a significant amount of first cost and operating complexity to a system.

Page 5 of 9

Electric centrifugal heat-recovery chiller

An electric centrifugal heat-recovery chiller, or water-to-water heat pump, as it sometimes is called, is capable of producing 42°F chilled water while also producing 155°F condenser water that can be recovered for space heating or reheat (Figure 9). Following the basic precepts of reverse Carnot-cycle efficiency, this machine requires significantly more electrical power than a standard high-efficiency chiller (1.370 kw per ton vs. 0.571 kw per ton).

Figure 10 shows the comparatively high emissions per ton-hour of cooling of an electric centrifugal heat-recovery chiller. If heat from the machine can be recovered for useful purposes, however, this can be a super-low-emissions option (Figure 11). Such a machine never should be operated in situations in which heat cannot be recovered; not only will the CO₂ emissions be the highest of any of the options discussed here, operating costs will be incredibly high.

Summary

Obviously, the next step is to develop annual numbers. However, on the basis of the full-load comparisons, several conclusions can be drawn:

Emissions comparisons are highly related to site location and the source of electrical generation.

Page 6 of 9

Facility characteristics favoring heat-recovery-chiller systems are:

• Year-round requirement for space heating or reheat, as in laboratories and hospitals.

• HVAC systems that accept recovered heat at heating water temperatures of 155°F or below.

• Relatively high natural-gas costs and relatively low electrical costs.

• Machines sized so that recovered heating matches minimum steady reheat load.

Facility characteristics favoring cogeneration options are similar, exceptions being:

• The availability of higher-temperature hot water or steam.

• The favoring of relatively high electrical costs and relatively low natural-gas costs.

References

1. EIA. (2009). Annual energy review. Retrieved from http://www.eia.doe.gov/emeu/aer/pdf/aer.pdf

2. EPA. (2009). Energy star performance ratings methodology for incorporating source energy use. Retrieved from http://www.energystar.gov/ia/business/evaluate_performance/site_source.pdf

Did you find this article useful? Send comments and suggestions to Executive Editor Scott Arnold at [email protected].

A principal of 8760 Engineering and a longtime member of HPAC Engineering's Editorial Advisory Board, Gerald J. Williams, PE, LEED AP, is an expert on chilled water, air systems, and system analysis. Since 1973, he has taught as an affiliate professor of mechanical engineering at Washington University in St. Louis, which awarded him its School of Engineering and Applied Science Alumni Achievement Award in 2009. From 1984 to 1995, he taught air-system design for energy and cost-effectiveness as part of the American Society of Heating, Refrigerating and Air-Conditioning Engineers' (ASHRAE's) Professional Development Seminar series. He is a past president of the St. Louis chapter of ASHRAE.

CARBON-DIOXIDE PRODUCTION IN COMBUSTION PROCESSES

In a combustion process, energy is liberated by the chemical reaction and combination of the combustible portions of a fuel -- carbon, hydrogen, and sulfur -- with oxygen contained in atmospheric air.

Page 7 of 9

COAL

As coal is burned, the amount of heat that is released and the chemical byproducts of combustion depend to a large extent on the constituents of the coal. In St. Louis, the local electric utility burns mostly a Wyoming Powder Valley coal that is surface-mined and referred to as subbituminous coal. Following is an ultimate analysis of this coal:

• Carbon, 50.6 percent by weight.

• Oxygen, 33.0 percent by weight.

• Hydrogen, 6.1 percent by weight.

• Nitrogen, 1.2 percent by weight.

• Sulfur, 1.1 percent by weight.

The higher heating value is 8,700 Btu per pound.

The combustion of coal substantially is a combination of three active chemical reactions:

• The combination of carbon (C) and oxygen (O₂) to form carbon dioxide (C + O₂ = CO₂).

• The combination of hydrogen (H₂) and oxygen to form water vapor (H₂ + 0.5 O₂ = H₂O).

• The combination of sulfur (S) and oxygen to form sulfur dioxide (S + O₂ = SO₂).

In each case, the number of atoms on the left side of the equation is equal to the number of atoms on the right. Also, the weight in pounds of a mole of a gas is the same as the numerical value of the molecular weight of the element. Thus, the approximate weight of a mole of hydrogen (H₂) is 2 lb, a mole of oxygen (O₂) 32 lb, and so on (see table). The moles of constituents on the two sides of the equation do not necessarily balance. However, according to the law of conservation of mass, the weight of all of the constituents on the left side of the equation must equal the weight of all of the constituents on the right (the products of combustion). Thus, "C + O₂ = CO₂" could be restated as:

1 mole C + 1 mole O₂ = 1 mole CO₂

or:

12 lb C + 32 lb O₂ = 44 lb CO₂

Carbon makes up 50.6 percent, or 0.506 lb, of a pound of Wyoming Powder Valley coal. If 1 mole of carbon (12 lb) is combined with 1 mole of oxygen (32 lb), 1 mole (44 lb) of carbon dioxide is produced. For every pound of carbon burned, then, 3.66 lb (44 divided by 12) of carbon dioxide is produced. So, for every pound of Wyoming Powder Valley coal burned, 1.85 lb (0.506 times 3.66) of carbon dioxide is produced. (Similar calculations can be performed to show the amounts of water vapor and sulfur dioxide produced per pound of coal burned.)

To determine the carbon dioxide produced per unit of energy consumed in the combustion of Wyoming Powder Valley coal, the heating value of the coal must be considered:

(1,000,000 Btu) x (8,700 Btu per lb coal) = 114.94 lb coal

The carbon dioxide produced in burning this amount of coal would be:

(114.94 lb coal) x (1.85 lb CO₂ per lb coal) = 212.6 lb CO₂ per MMBtu coal

Page 8 of 9

NATURAL GAS

The combustion of natural gas is a bit more complex. Although methane typically is the major constituent, natural gas often contains a number of other combustible organic materials, depending on the gas field from which it was obtained. In St. Louis, the local gas supplier delivers natural gas with the following major constituents:

• Methane, 94.8 percent by weight.

• Ethane, 2.5 percent by weight.

• Carbon dioxide, 1.4 percent by weight.

• Nitrogen, 0.6 percent by weight.

• Propane, 0.6 percent by weight.

• Trace hydrocarbons, 0.1 percent by weight.

The higher heating value is 1,032.2 Btu per cubic foot. The specific gravity is 0.5934.

As with coal, analysis of the combustion of natural gas begins with consideration of the reaction of carbon, hydrogen, and, if it is present, sulfur with oxygen:

• Methane combustion: CH₄ + 2 O₂ = CO₂ + 2 H₂O

• Ethane combustion: C₂H₆ + 3.5 O₂ = 2 CO₂ + 3 H₂O

• Propane combustion: C₃H₈ + 5 O₂ = 3 CO₂ + 4 H₂O

The carbon dioxide produced by each of the combustion reactions can be calculated for each pound of natural gas burned:

• Carbon dioxide produced from methane: 0.948 lb x (44 ÷ 16) = 2.607 lb

• Carbon dioxide produced from ethane: 0.025 lb x(88 ÷ 30) = 0.073 lb

• Carbon dioxide produced from propane: 0.006 lb x(132 ÷ 44) = 0.018 lb

• Carbon dioxide in original constituents: 0.014 lb

The total is 2.712 lb of carbon dioxide.

Page 9 of 9

At standard temperature and pressure, the specific volume of natural gas is:

1 ÷ (0.5934 x 0.075 lb per cu ft) = 22.469 cu ft per lb

Thus, 1 lb of natural gas has a heat content of:

1,032.2 Btu per cu ft x 22.469 cu ft per lb = 23,192.5 Btu per lb

One MMBtu of natural gas, then, would equal:

(1,000,000 Btu) ÷ (23,192.5 Btu per lb) = 43.117 lb natural gas

Because 1 lb of natural gas produces 2.723 lb of carbon dioxide, 43.117 lb of natural gas would produce 116.9 lb of carbon dioxide (43.117 times 2.712), or burning 1 MMBtu of natural gas would produce 116.9 lb of carbon dioxide.

For other fuel constituents and the typical chemical equations that apply, see Chapter 28 of the 2009 edition of ASHRAE Handbook ¡ª Fundamentals.

GENERATION OF ELECTRICITY

A large amount of the electricity produced today is generated in an enhanced Rankine steam cycle. Typically, coal or natural gas is fired in a boiler, producing high-pressure steam that is delivered to a steam turbine. In the turbine, energy is released, turning an electrical generator. The expanded steam condenses at a low temperature and is pumped back to the boiler, where the cycle is completed.

The performance of such plants often is measured by gross station heat rate:

Gross station heat rate (Btu per kwh) = (W_f x HR) ÷ (kwh generated)

where:

W_f = amount of fuel burned per hour in pounds or cubic feet

HR = heating value of fuel in British thermal units per pound or British thermal units per cubic foot

kwh = electric energy produced by generator

Such a calculation is a good measure of boiler-turbine efficiency, but does not address parasitic loads, such as feedwater and cooling-water pumping. To better account for the overall efficiency of a plant in delivering electrical energy to end users, net station heat rate often is utilized. Net station heat rate is based on station output minus the power consumed by plant electrical auxiliaries:

Net station heat rate (Btu per kwh delivered) = (W_f x HR) ÷ (kwh generated - kwh used in plant)

Note that net station heat rate always will be higher numerically than gross station heat rate.

For steam Rankine-cycle plants operated by the local utility in St. Louis, the average gross station heat rate is approximately 9,800 Btu per kilowatt-hour, while the average net station heat rate is in the range of 10,300 Btu per kilowatt-hour (or an equivalent overall thermal efficiency of 3,413 divided by heat rate, or 33 percent).

For a coal-burning Rankine-cycle plant, carbon-dioxide emissions per kilowatt-hour of electricity delivered to end users would be:

(212.6 lb CO₂ per 1 MMBtu coal) x (10,300 Btu per kwh) x (1 MMBtu coal ÷ 1,000,000 Btu) = 2.19 lb CO₂ per kwh

For a natural-gas-burning Rankine-cycle plant, carbon-dioxide emissions per kilowatt-hour of electricity delivered to end users would be:

(116.9 lb CO₂ per 1 MMBtu natural gas) x (10,300 Btu per kwh) x (1 MMBtu natural gas ÷ 1,000,000 Btu) = 1.20 lb CO₂ per kwh

In St. Louis, where the average regional electric emissions rate is 1.844 lb of carbon dioxide per kilowatt-hour, the majority of electricity is generated through the use of coal. (Electricity generated with wind energy, solar energy, hydroelectric energy, and nuclear energy has equivalent net emissions rates approaching 0.0 lb of CO₂ per kilowatt-hour of electricity generated.)