6. Cost estimates
The above mass and energy balances provided the basis
for estimating capital, coal, and non-fuel operating and
maintenance costs for each process configuration. Capital
costs (± 30 % accuracy, expressed in 2002 US$) were estimated
for commercially-mature (‘‘Nth plant’’) systems by
major equipment area for manufacture and construction at
a United States location primarily based on reference
overnight installed costs from one of two sources [Moore,
2003; Kreutz et al., 2003] (Table 8).
The reference capital costs were scaled (using scaling
exponents in Table 8) to the capacities corresponding to
the plants described in Tables 5 and 7 for methanol and
DME, respectively. Since fuel production is of primary
interest in the present analysis, to facilitate cost comparisons
among different process configurations, we further
developed our capital cost estimates for process configurations
producing the same amount of liquid fuel in all
cases. We used the fuel output of the recycle case as the
common output level. The plant capacities in Tables 5 and
7 for the OT cases have been scaled up accordingly. The
resulting plant sizes and installed capital cost estimates
are shown in Tables 9 and 10 for methanol and DME
respectively. Note 1 in these tables provides some additional
details on our cost estimating methodology.
The above capital costs for methanol and DME would
be lower if estimated on the basis of Chinese manufacture
and construction, rather than on US conditions. For components
of ICL processes that in the near future could be
manufactured in China (essentially all components exceptthe gas turbine), as well as for engineering and construction
labor, significant cost reductions are likely to be
achievable relative to the estimates in Tables 9 and 10.
For example, a pre-feasibility study for construction of an
830,000 t/d facility that would make DME from coal in
Ningxia Province, China, indicates that installed capital
costs based on US Gulf Coast construction should be multiplied
by 0.75 to estimate capital costs for China [Lucas,
2002]. A study of IGCC costs indicates a 0.65 China ‘‘location’’
multiplier on total US overnight installed capital
costs for a plant built in China instead of in the USA
[Stoll and Todd, 1996]. For gasification projects, Shell
China typically applies a location factor of 0.60 to European
plant costs, on the basis of its own detailed evaluations
of local Chinese manufacturing and construction
costs [Wang, 2002].
6.1. Methanol
For plants located in the US, levelized factory-gate production
costs for methanol are shown in Table 11. The
coal input price for this analysis is $ 1/GJLHV ($ 23.5/t).
The methanol costs include credits for revenue from exported
electricity, assuming the sale price of the electricity
is equal to the cost for generating electricity from a new
IGCC plant (with no CO2 capture), for which emissions
per kWh of local pollutants such as SO2 and NOx would
be somewhat higher than emissions from the methanolelectricity
co-production facility (see Table 11, Note 5).
For a range of electricity prices, Figure 6 shows themethanol cost.
Without considering carbon capture, the least costly
methanol ($ 8.6/GJ) is produced in the RC case. To produce
methanol at this cost in the OT case requires an
electricity sale price of $ 0.045/kWh, which is slightly
higher than the baseline value used in Table 11 (Figure
6). The cost of methanol is insensitive to electricity price
in the RC case (since almost no exportable electricity is
produced).
When CO2 is captured at the plant and injected below
ground 100 km from the plant for long-term storage, the
cost of methanol increases relative to the lowest-cost case
with CO2 vented. At the reference electricity sale price,
the costs of methanol in the OT and RC cases with CO2
capture and storage are $ 11.3/GJ and $ 9.4/GJ, respectively.
When CO2 and H2S are co-captured and stored
jointly below ground, the cost of methanol is considerably
lower than when CO2 alone is captured and stored. In
fact, the methanol cost is only slightly higher than the
cost for the same plant configurations with CO2 vented.
This surprising result is due primarily to the considerably
simplified sulfur capture system (including elimination ofthe Claus and SCOT units) when H2S and CO2 are cocaptured
for underground storage.
The methanol costs shown in Table 11 can be compared
to the value of methanol as an octane-enhancing gasoline
additive and to the cost of methanol as a neat fuel (100 %
methanol) used in vehicles replacing petroleum-gasoline
vehicles.
When used as a gasoline additive, methanol will boost
fuel octane without significantly affecting engine efficiency.
Thus, the value of methanol as an additive can be
estimated from the difference in value between a regulargrade
gasoline and a mid-grade gasoline. Consider the
case of Beijing on April 11, 2003 (when the world crude
oil price was about $ 23/bbl = $ 169/t; 1 bbl or barrel =
about 136 kg). The average refinery-gate price (including
taxes) for regular-grade gasoline (90 octane) from 36 large
refineries in China on that date was 3172 RMB/t (or $
0.287/l) [Changchun, 2003]. The mid-grade (93 octane)
gasoline wholesale price that same day was 3408 RMB/t
($ 0.308/l). Since methanol has a blending octane of 120
[Wyman et al., 1993], a 10 % volume addition of methanol
to 90-octane gasoline would increase the octane of
the blended fuel to 93, and the added methanol would
have a value of $ 0.50/l, or $ 28/GJLHV
[5]. While the value
of methanol in this application would be far above the
estimated production cost, the market potential of this application
is limited by fuel blending limits.
Much larger potential markets are for methanol as a
neat fuel or as M85 (85 % MeOH, 15 % gasoline). The
value of methanol in a neat application will depend on
the efficiency of the methanol engine relative to a conventional
gasoline engine. An optimized spark-ignited
neat-methanol engine will be more energy-efficient than
a conventional gasoline spark-ignited engine primarily because
of higher compression ratios that can be achieved
with methanol. Wyman et al. [1993] indicate that a 20 %
engine efficiency advantage can be expected. Conservatively,
we consider a 15 % gain here, which gives methanol
production costs of $ 0.24--0.25/l gasoline-equivalent
for the cases with CO2 venting and $ 0.25--0.32/l gasoline-equivalent
for process configurations with CO2 capture
(Table 11). These values can be compared with
average Chinese refinery-gate gasoline prices mentioned
above ($ 0.29-0.31/l). As another comparison, the average
sale price of petroleum gasoline by US refiners to resellers[6]
in April 2003 was $ 0.26/l [EIA, 2003]. These comparisons
suggest that the methanol costs shown in Table
11 (even for some cases with CO2 capture and storage)
would be competitive with gasoline made from crude oil
when the world oil price is $ 23/bbl or more, particularly
considering that capital costs for methanol plants built in
China would be lower than those estimated in this analysis
(for reasons discussed at the start of Section 6).
6.2. Dimethyl ether
For plants located in the US, levelized factory-gate production
costs for DME are shown in Table 12 for the same
input parameter values as discussed above for methanol,
including electricity sold for $ 0.429/kWh. For other electricity
prices, Figure 7 shows DME costs as a function of
electricity sale price.
Without considering carbon capture cases, DME produced
in the OT case ($ 8.7/GJ) is less costly than in the
RC case ($ 9.5/GJ). The ‘‘break-even’’ electricity price for
the once-through versus recycle configurations is $
0.040/kWh, which is below the assumed reference electricity
sale price.
When CO2 is captured and stored, the calculated costs
of DME in the OT and RC cases are $ 10.0/GJ and $
10.4/GJ, respectively, using the reference electricity price.
When CO2 and H2S are co-captured/co-stored, the DME
cost in the OT case is 21 % lower than with capture of
CO2 alone, and the DME cost in the RC case is 8 % lower
than with CO2 capture alone.
Moreover, in the RC case with co-capture/co-storage,there is essentially no cost penalty for carbon capture relative
to the RC configuration with CO2 venting. In the OT
case with co-capture/co-storage, there is actually a cost
advantage compared to the OT case with CO2 vented. This
is unlike the case for methanol production (Table 11),
where the methanol cost in the OT case with CO2 co-capture/co-storage
is higher than for the OT case with CO2
vented. The advantageous relative costs for DME arise
from the fact that less CO2 is captured and stored in the
OT case for DME than for methanol (7.7 kgC/GJDME and
13.1 kgC/GJMeOH).
The above DME costs are also expressed in Table 12
on LPG-equivalent, diesel-equivalent, and gasolineequivalent
bases.
The LPG-equivalent costs, ranging from $ 399/t to $
437/t when CO2 is vented and from $ 380/t to $ 478/t
when CO2 is captured and stored, can be compared with
costs of LPG in China. Typical ex-refinery wholesale
prices of LPG at Chinese refineries were 2980-3370
RMB/t in April 2003 [Cui, 2003], or an average of US$
383/t. Wholesale prices of imported LPG at coastal terminals
in China were in this same range. Since the LPGequivalent
DME costs in Table 12 are based on a USA
plant location, it is likely that a plant located in Chinawould produce DME at costs that would be very competitive
with wholesale LPG prices.
The above LPG and DME costs can also be compared
with estimated health-related costs of indoor air pollution
arising from cooking with solid fuels, which the World
Health Organization ranks as the eighth most serious contributor
to the total burden of disease in ‘‘low-mortality
developing countries’’ (which includes China) [WHO,
2002]. The World Bank has estimated that the annual
health cost of indoor air pollution in rural China was between
$ 3.7 and $ 10.6 billion in 1995 [World Ba