Chapter 5 Appendices 1

Appendix 5A: Natural Gas Use in Industrial

Boilers
Figure 5A.1 Use of Natural Gas Boilers in the U.S. Manufacturing Sector
Industrial boilers consumed 2.1 Tcf of natural
gas in 2006, accounting for 36% of total natural
gas in manufacturing.
1
In this appendix, we
provide further detailed analysis of two potential
drivers affecting demand for natural gas in
boilers: modernization of the current natural
gas boiler eet with more efcient units, and
replacement of coal boilers with new natural
gas boilers.

Figure 5A.1 shows that use of industrial boilers
is concentrated in the energy-intensive industries,
with the four largest applications in chemicals
(39%); food processing (17%); paper (13%); and
petroleum and coal products (13%). There is
strong competition among boiler fuels in the
energy-intensive industries, which employ larger
boilers and have ready access to alternative fuel
supplies. The EPA inventory shows, for example,
382 boilers greater than 100 MMBtu/hr, of
which 68% were coal red. Natural gas is the
predominant boiler fuel in other manufacturing
industries, which typically employ smaller boilers
and do not have the same opportunities for use
of by-product and waste fuels. The EPA data
show 4,132 natural gas boilers, with an average
size of 71 MMBtu/hr.
2

For purposes of our analysis, we used a boiler
size of 100 MMBtu/hr as a benchmark.
A 100 MMBtu/hr boiler is at the large end of the
scale of natural gas boilers; over 95% of existing
boilers (all types and all fuels) are less than this
size, comprising about 60% of total fuel input
capacity (all fuels).
3
A boiler size of 100 MMBtu/hr
is comparable to many coal boilers, which
typically have a larger average size than natural
gas. As a sensitivity analysis, we also analyzed
smaller size boilers (50 MMBtu/hr).
All Other includes:
bovo..go cb.ccc |.cJucts (14)
oxt||o V|||s (16)
oxt||o |.cJuct V|||s (0)
App..o| (02)
\ccJ |.cJucts (12)
|.|rt|rg |o|.toJ Suppc.t (02)
||.st|cs |ubbo. |.cJucts (2)
|crnot.|||c V|ro..| |.cJucts (10)
|.|n..y Vot.|s (2)
|.b.|c.toJ Vot.| |.cJucts (1)
V.c||ro.y (08)
cnputo. ||oct.cr|c |.cJucts (09)
||oct.|c.| |ou|pnort, App||.rcos,
cnpcrorts (03)
..rspc.t.t|cr |ou|pnort (19)
|u.r|tu.o |o|.toJ |.cJucts (<01)
(Conventional and CHP/Cogeneration Boilers)
Food
17%
Paper
13%
Petroleum
& Coal
Products
13%
Chemicals
39%
All Other
18%
Source: EIA MECS
2 MIT STUDY ON THE FUTURE OF NATURAL GAS
Modernization of the Natural Gas
Industrial Boiler Fleet
Boilers have long service lives; about 85% of
U.S. boilers have been in operation for about
30 years, and almost half of all boilers are
nearly 50 years old, as illustrated in Figure 5A.2.
There are a variety of existing boiler types and
sizes. The two common types of boilers are
retube boilers, which are used primarily for
hot water applications, and watertube boilers,
which are for larger-scale, steam-generation
applications.
Most existing natural gas boilers are typically
non-condensing boilers, whose exhaust gases
retain signicant quantities of waste heat. These
boilers typically have energy efciencies in the
range of 65% to 70%.
4
The waste heat in the
exhaust gases consist of both the latent heat that
can be recovered from condensing the water
vapor into a liquid, as well as the sensible heat
contained in the hot temperatures of the
exhaust.
Since the mid-1980s, new natural gas boilers
have incorporated additional heat recovery
systems (i.e., condensing technology) to capture
the latent heat and a portion of the sensible
heat in the exhaust gases. In addition, use
of economizers allows for waste heat to be
recovered by pre-heating the boiler feedwater.
These improvements boost overall energy
efciency to the 80% to 85% level. In 2004, the
DOE set minimum energy efciency standards
for new natural gas boilers in the range of
77% to 82%, depending upon boiler size and
boiler technology.
5

Further technology advances have demon-
strated efciency levels in the range of 94%
to 95%. DOE cost shared RD&D with the
Gas Technology Institute (GTI) has led to the
commercialization of a Super-Efcient boiler
capable of achieving 94% efciency in retube
boilers.
6
The super boiler employs a multi-stage
combustion system to improve combustion
efciency and reduce NOx emissions. This
technology also incorporates a combination of
a Transport Membrane Condenser (TMC) and
compact humidifying air heater to extract more
of the sensible as well as latent heat content of
the exhaust gas.
7
In addition, DOE and GTI have
collaborated on RD&D on another technology,
the Ultramizer System.
8
The Ultramizer
consists of a TMC condenser combined with
Figure 5A.2 Age Distribution of U.S. Industrial Boilers (All Types and Fuels)
Source: Energy and Environmental Analysis, Inc.
47%
29%
10%
8%
7%
Pre-1962 19631972 19731982
Vintage
19831992 19932002
C
a
p
a
c
i
t
y

(
M
M
B
t
u
/
h
r
)
0
200,000
400,000
600,000
800,000
1,000,000
1,200,000
Chapter 5 Appendices 3
high-temperature and low-temperature waste
heat recovery systems that preheat the boiler
feedwater and provide additional make-up
water recovered from the exhaust gas. Boiler
efciencies of 95% are possible for both
retube and watertube boiler applications.
9

We compared the net present value costs,
pre-tax, of potential replacement with either a
high-efciency or super-high-efciency natural
gas boiler for an existing 100 MMBtu/hr
natural gas boiler. Our analysis employed
equipment capital costs and energy efciency
assumptions provided by the GTI,
10
combined
with the 2010 average natural gas price for
industrial delivery of $5.19 per mcf. The results
under these assumptions (Table 5A.1) show
that replacement of current natural gas boilers
with high-efciency models would, at a 15%
discount rate, yield a reduction of 8% in
annualized costs on a pre-tax basis. Replace-
ment with super-high-efciency boilers would
yield annualized savings of 20%. A sensitivity
analysis comparing 50 MMBtu/hr natural gas
boilers yielded similar results.
The payback periods for these boiler replace-
ments range from 1.8 to 3.6 years, based on
2010 actual industrial natural gas prices, and
assuming no increase in natural gas prices over
this period. Higher natural gas prices would
improve the results; lower natural gas prices
would reduce the projected annualized savings
and extend the payback period.
The addition to these equipment expenditures
of soft costs (management, supervision, etc.)
and expenses attending the change in particular
installations will reduce these returns somewhat.
Also, in particular instances the attractiveness
of boiler modernization will depend on other
factors such as the remaining book value of
existing boilers that a rm might write off; the
availability of investment capital; the return on
investment in boiler modernization relative to
other opportunities; and the availability of tax
incentives, such as accelerated depreciation or
investment tax credits. Considering all these
factors, however, it appears that replacement
will be cost-effective in many installations.
Existing
Natural Gas Boiler
Replacement
Natural Gas Boiler
Parameter Units Base Case High Ef ciency
(80%)
Super High
Ef ciency (94%)
Assumptions
Boiler Size MMBtu/hr 100 100 100
Boiler Capital Cost $ million 1.00 1.25
Natural Gas Fired Boiler
Ef ciency
% 70 80 94
2010 Industrial Natural Gas
Price
$/MMBtu 5.19 5.19 5.19
Economic Results
Net Present Value of Costs $ Million 22.9 21.0 18.3
Payback Period Years 3.6 1.8
Impact on CO
2
Emissions
CO
2
Emissions Tons CO
2
/yr 35,616 31,164 26,523
Reduction in CO
2
Emissions Tons CO
2
/yr 4,452 9,093
Table 5A.1 Cost Comparison of Natural Gas Boiler Modernization Options
Source: MIT
4 MIT STUDY ON THE FUTURE OF NATURAL GAS
Two scenarios can provide an indication of
the impact on natural gas consumption:
(1) a replacement of 50% of current natural gas
industrial boiler capacity with high-efciency
natural gas boilers would reduce demand for
natural gas by 129 Bcf annually, while (2) a
replacement of 50% of current natural gas
boiler capacity with super-high-efciency
natural gas boilers would reduce demand by
263 Bcf annually. The reduction in CO
2
emis-
sions ranges from about 4,500 to over 9,000
tons per year per boiler.

These results show that replacement of existing
industrial natural gas boilers with higher
efciency models could cost-effectively reduce
natural gas demand and reduce GHG emissions,
suggesting that the DOE should review the
current energy efciency standards for commer-
cial and industrial natural gas boilers and
assess the feasibility of setting a more stringent
standard.
Replacement of Existing Coal Industrial
Boilers with Efcient Natural Gas Boilers
A CO
2
emissions reduction requirement could
lead to a signicant level of replacement of
existing coal boilers to natural gas. Absent a
carbon constraint, a potential driver for fuel
switching of coal boilers to natural gas is the
establishment of National Emissions Standards
for Hazardous Air Pollutants (NESHAPS) for
boilers which could lead to a similar result.
Our analysis is based on the EPA February 23,
2011, emissions standards for mercury, metals,
dioxin, acid gases and other hazardous air
pollutants emitted from industrial boilers and
process heaters. On May 16, 2011, EPA Adminis-
trator Jackson stayed the implementation of the
standards to provide for additional review;
however, the February 23 standards, and the
June 2010 proposed standards, provide a general
benchmark for analysis of the trade-offs between
retrotting existing industrial coal boilers with
post-combustion controls and replacement of
existing industrial coal boilers with efcient, new
natural gas boilers.
The Clean Air Act requires that emissions
reduction standards for each hazardous air
pollutant be based upon the emissions reductions
that can be attained through installation of the
Maximum Achievable Control Technology
(MACT), which is dened as the level of perfor-
mance achieved by the top 12% performing
facilities within the subcategory of facilities
subject to the standards. The EPA dened 15
different subcategories of industrial boilers and
process heaters, setting standards for 11 of the 15
subcategories; existing natural gas boilers (as well as
boilers fueled with renery gas and certain other
types of clean gases) fall within the four subcat-
egories for which there are no specic emissions
standards. EPA estimates of the emissions
reductions achievable from application of the
proposed standards are shown in Table 5A.2.
Table 5A.2 Estimated National Emissions Reductions from the February 2011
MACT Standards for Industrial Boilers and Process Heaters
Hazardous Air Pollutant Annual Emissions Reductions (Tons/yr)
Hydrogen Chloride 30,000
Mercury 1.4
Non-mercury Metals 2,700
Particulate Matter 47,000
Sulfur Dioxide 440,000
Volatile Organic Compounds 7,000
Source: EPA
Chapter 5 Appendices 5
Three subcategories subject to new MACT
standards are coal boilers utilizing different
technologies stoker, uidized bed and
pulverized coal combustion. Achieving the
emission standards for coal boilers will require
the installation of wet scrubbers and fabric
lters. Installation of activated carbon injection
for control of mercury emissions also may be
required, although the EPA noted that it is
assuming that the units subject to the MACT
calculations were able to achieve the standards
for mercury emissions reductions through the
use of fabric lters only.
At that time of the initial proposed rules in
June 2010, the EPA also analyzed fuel switching
from coal to natural gas as a compliance
option,
11
but concluded that this measure was
uneconomical relative to the installation of
post-combustion control technology at coal
boilers. This EPA conclusion was heavily
inuenced by two assumptions that were
disadvantageous to natural gas: (1) the analysis
used the 2008 average natural gas price for
industrial delivery of $9.58 per mcf, which
represented a period of high natural gas prices
relative to today and anticipated in the future
(see Chapter 3); and (2) the analysis assumed
that boiler owners would retrot the burners
on existing coal boilers to burn gas rather than
replace the boilers entirely with new high-
efciency boilers designed for natural gas. The
EPA estimates that burner retrot reduced
boiler energy efciency by 5%.
We performed a similar analysis for a single
100 MMBtu/hr coal boiler, a relatively large
boiler that can be deployed in a number of
industry sectors. Four different options were
analyzed and compared to a base case. The four
options include: (1) retrot of post-combustion
controls (using EPA cost assumptions); (2) retrot
of natural gas burners within the existing coal
boiler (using EPA efciency assumptions); (3)
replacement of the existing coal-red boiler
with a high-efciency natural gas boiler; and
(4) replacement of the existing coal boiler with
one of the new, super-high-efciency natural
gas boiler technologies.
Table 5A.3 Cost Comparison of Industrial Boiler MACT Compliance Options
Existing Coal Boiler Replacement Natural Gas Boiler
Parameter Units Base Case Post-Combustion
Controls Retroft
Natural Gas
Burner Retroft
High Ef ciency
(80%)
Super High
Ef ciency (94%)
Assumptions
Boiler Size MMBtu/hr 100 100 100 100 100
Boiler Capital Cost $ Million 5.5 0.34 1.00 1.25
Coal-Fired Boiler Ef ciency % 65 65
Natural Gas Fired Boiler
Ef ciency
% 62 80 94
2010 Industrial Coal Price $/MMBtu 2.88 2.88
2010 Industrial Natural Gas
Price
$/MMBtu 5.19 5.19 5.19
Economic Results
Net Present Value of Costs $ Million 12.7 18.2 24.4 19.6 17.1
Impact on CO
2
Emissions
CO
2
Emissions Tons CO
2
/yr 63,840 63,840 37,339 28,938 24,628
Reduction in CO
2
Emissions Tons CO
2
/yr 0 0 26,501 34,902 39,212
Incremental Cost of CO
2

Reductions
$/Tons CO
2
34 5 -5
Source: MIT
6 MIT STUDY ON THE FUTURE OF NATURAL GAS
Table 5A.3 shows the results, including the net
present value at a 15% discount rate and effects
on CO
2
emissions. The capital cost shown in
the table is the cost of equipment, and for the
purposes of this comparison it is reasonable
to assume that the soft costs and other costs
attending the particular rm or installation are
roughly the same. The option of retrotting
natural gas burners in existing coal boilers
(i.e., the option analyzed by the EPA) was the
highest cost option on a net present value basis.
12

The reason is that there is a small energy-
efciency penalty from retrotting natural gas
burners, and thus total fuel costs are higher.
This nding is consistent with the EPA regula-
tory analysis. Replacement of the existing coal
boilers with high-efciency natural gas boilers
(i.e., 80% efciency) is slightly more expensive
than installing post-combustion controls, but
boiler replacement with a super-high-efciency
(i.e., 94% efciency) natural gas boiler is more
cost-effective. A sensitivity analysis comparing
smaller size boilers (50 MMBtu/hr) yielded
similar results.
This cost comparison is dependent upon two
assumptions: (1) the estimates of capital equip-
ment cost for retrotting post-combustion
controls for coal; and (2) the relative prices of
coal and natural gas. Our analysis uses the EPA
capital cost assumptions for installation of
post-combustion controls at coal boilers,
consisting of wet scrubbers and fabric lters,
but without activated carbon injection. For coal
boilers that may require additional controls
to achieve MACT limits for mercury emissions,
costs would increase substantially, making
the options for replacement with natural gas
boilers much more cost-effective. The compara-
tive results also are sensitive to natural gas prices.
The price differential between coal and natural
gas used in our analysis was $2.31/MMBtu,
based on actual average delivered prices in 2010.
A lower price differential (i.e., a smaller price
spread between natural gas and coal) would
make conversion to natural gas more attractive.
The potential impact of replacing industrial
coal boilers with new, high-efciency natural gas
boilers is signicant. The EIA MECS data show
that industrial coal boilers and process heaters
currently use 892 trillion Btu (0.9 quads) of
coal each year. Conversion of this capacity to
natural gas would increase demand for natural
gas by 0.87 Tcf per year. The actual rate of
market penetration would be dependent upon
individual facility analyses.
Replacement of existing coal boilers with new
efcient natural gas boilers in order to meet
NESHAPS requirements could reduce annual
CO
2
emissions by 52,000 to 57,000 tons per
year per boiler. Assuming that high-efciency
(i.e., 80%) natural gas boilers are installed, the
net present value cost is slightly higher than
installing post-combustion controls. If this
incremental cost is assigned solely to CO
2

reduction (ignoring the benets from further
reductions of other pollutants), the incremental
cost to achieve the CO
2
reductions is about
$5/ton.
Context 7
NOTES
1
U.S. Energy Information Administration, 2008 Manufacturing Energy Consumption Survey.
2
U.S. EPA Industrial/Commercial Boiler Survey.
3
Energy and Environmental Analysis, Inc., Characterization of the U.S. Industrial Boiler Population,
Report prepared for Oak Ridge National Laboratory, May 2005.
4
Energy Efciency in boilers is measured as AFUE or Average Fuel Use Efciency.
5
The DOE Energy Efciency Standards can be found at 10 CFR Part 431.
6
DOE Webcast: GTI Super Boiler Technology, by Dennis Chojnacki, Senior Engineer and Curt Bermel,
Business Development Manager R&D, Gas Technology Institute, November 20, 2008.
7
U.S. Department of Energy, Industrial Technologies Program, Super Boiler, June 2007.
8
See http://www.cannonboilerworks.com.
9
GTI licensed its patented Transport Membrane Condenser (TMC) to Cannon Boiler Works on
October 6, 2009. Cannon Boiler Works announced plans to release the Ultramizer technology to the
public around the end of 2010. See http://www.cannonboilerworks.com/ultramizer.html.
10
Ron Edelstein, Gas Technology Institute, personal communication.
11
The EPA analysis was summarized in an April 10, 2010 Memorandum from Graham Gibson, ERG,
to Jim Eddinger, U.S. EPA.
12
The analysis in Table 5A.3 is drawn from a variety of data, including the ERG Memorandum of April 10,
2010; the GTI data; the EPA regulatory analysis, which can be found at http://epa.gov/ttn/atw/boiler/
boilerpg.html; coal and natural gas prices can be found at http://www.eia.doe.gov. The net present value
estimates are based on a discount rate of 15%, representing a typical internal rate of return for analysis
of corporate capital investments.
Chapter 5 Appendices 7