Treating irregularity

The perennial complaint about the major renewable energy sources (and the lament of their users and advocates) is that they are irregular.  If the sun shone all the time, you wouldn't need batteries; if the wind blew all the time, you wouldn't need backup generators; if it rained the same amount every day or even every month, you probably wouldn't need hydro dams and could make do with weirs and long penstocks.  The grid manager can't schedule a one of them.

Unless all supply-demand matching can be handled by DSM, any system which depends on alternative energy is going need storage of some type.  The devil is in the details, as ever.  Availability of energy varies on scales of hours, days and even by the seasons; a collection system which is adequate on the average is going to be deficient at some times and produce large surpluses at others.  There are loads which are well-matched to certain types of production (air conditioning vs. solar) but we aren't lucky enough to have all loads dovetail with renewable sources.  Some kind of buffer is needed to split the difference.

Buffers on the scale of seconds to hours are available, in the works or proposed.  Capacitors work well to buffer variations on the scale of milliseconds to seconds; flywheels are well-suited to smoothing differences lasting seconds to minutes, vanadium redox "flow batteries" have been suggested for handling time scales of seconds to hours, and compressed-air storage is being proposed as part of a wind-storage scheme.  But only compressed air is capable of dealing with daily variations at reasonable cost, and none of the above deal well with variations lasting weeks to months.

Fortunately, a suitable medium seems to be available.  Two, actually.

Storage media

The classic stockpiles of energy are stacked firewood and heaps of coal.  Oil is invisible by comparison; tank farms represent huge reserves, but they look the same whether full or empty and have no power as images.  Yet not all classic energy is meant for the fireplace; food laid by is energy too, whether it is dried or salted meat, jars of food canned or pickled, roots in the root cellar, packed granaries, full silos or haylofts stuffed with animal fodder.  Lately, fodder storage has expanded to square or round bales of hay kept either in or out of doors.

Fodder is biomass suited for ruminants.  For combustion it needn't even be edible, just dry.

The favored energy crops these days are coppiced willow or poplar and various perennial grasses.  The wet tropics produce good crops of sugar cane, but temperate zones are better suited to the likes of switchgrass or Miscanthus.  The latter is reported to produce 25-45 metric tons per hectare per year¹ with gusts up to 60.  The temperate-zone grasses are particularly attractive because they die back to the rhizomes during the winter, and the stems left above ground become dry and easy to harvest.  The standing crop represents stored energy.  Once cut and baled, it can be stored on scales of days to months.

Baled grass is not an ideal storage medium.  Its flaws include:

1. It requires proper and precise drying to attain its best characteristics.
2. It rots, loses energy and may spontaneously combust if allowed to get or remain damp.
3. Its general bulk limits how far it can be transported and how much can be stored on or near the field without impairing the next season's growth.

It would be best to convert the raw biomass to another medium:  something waterproof, compact, more easily transported and stable over the duration of a year or years.

This medium exists.  It is one of the oldest fuels known to man:  charcoal.

Charcoal is the solid product of the slow anaerobic pyrolysis or partial combustion of most forms of biomass.  It is so stable that it is used for radiocarbon dating of prehistoric human activities.  Aside from its friability, it is as storable as coal and can be handled in similar ways.  It contains little or no sulfur or heavy metals, burns much more cleanly than coal, and is usable as a substitute for coal or coke in many processes.

As I've noted before, the conversion of biomass to charcoal loses roughly half of its stored energy as heat and combustible gas.  The presumed 1.3 billion (dry) tons of available waste biomass has a total energy of 20.8 exajoules (19.7 quads) at 16 GJ/ton.  This is over 3 times as much energy as the natural gas used for electricity.  At an average productivity of 35 tons/ha for Miscanthus, each additional million hectares devoted to biomass production would yield another 560 PJ (0.53 quads).  Of the ~32 million ha devoted to maize alone, perhaps 10 million could be converted to Miscanthus as part of a price-support program.  This might yield in the neighborhood of 5.3 quads of fuel, for a total of 24 quads of biomass.  Other conversions might add to this.

Managing the energy supply

Once you've got the biomass, the question becomes how to manage it for the greatest benefit.  Some of the options are flexible, but are linked to cycles which are not terribly efficient.  For instance, the production of bio-oil by flash pyrolysis yields about 70% of the mass and energy of the input as oil; the rest becomes char or gas.  The bio-oil is suitable for boiler or perhaps gas-turbine fuel, but the net efficiency is low (if fed to a 35%-efficient steam plant, 24.5%; if fed to a 40%-efficient gas turbine, 28%).  On top of this, it does not store well.  It appears likely that charcoal will soon be convertible to electricity at 80% efficiency using direct-carbon fuel cells; maximizing the production of charcoal appears to be a good strategy.

Processing ~1.6 billion tons of biomass through a carbonization step which turns 28% of it into charcoal would yield ~450 million tons of charcoal.  For the purpose of this analysis, I'm going to assume that the charcoal is not typically used to produce electric power for the grid.  Some of it goes to mobile DCFC's aboard vehicles, replacing petroleum fuels in internal-combustion engines; an average demand of 183 GW would require approximately 220 million metric tons/year of charcoal for vehicle fuel.  The remainder could go for chemical production, metallurgy, strategic stockpiles or just to be sequestered.  For whatever reason, it's the last (renewable) choice for grid power.

If the carbonization process loses about 52% of the energy in the raw biomass, it seems prudent to use it well and wisely.  Charcoal stores well, so the carbonization could be scheduled to produce gas when needed.  If the heat and gas can be converted to electricity at efficiencies typical of simple-cycle gas-turbine powerplants (say, 40%), the recoverable electricity from the waste biomass would be 4.33 EJ (1200 billion kWH) with another 116 petajoules (32.4 billion kWh) from each million hectares converted to biomass crops.  This is an average power of 137 GW from the base amount of biomass and a further 3.7 GW from each million hectares converted; for the case of 10 million ha converted, the average available electric power from carbonization would be 174 GW.

The average US electric consumption in the year 2004 was roughly 450 GW.  An average power of 174 GW is over 38% of the total.  Could this fill the gaps in other renewable supplies?  The answer appears to be "it might, depending".  I have neither the time nor the data to do a comprehensive analysis of the adequacy or lack thereof, but just to try to get a feel for it I'll guesstimate this by geographic area.

The following includes a certain amount of hand-waving.  You have been warned.

South and southwest

In the south and southwest, air conditioning accounts for a very large fraction of the total electric consumption.  The peak demand coincides with sunny conditions, so it seems that solar might well suffice for half of total electric requirements there.  The use of ice-storage systems would allow a day's production to carry cooling demand overnight, eliminating the need to generate or store electricity for the purpose.  Solar could also be used to meet other power requirements during daylight.  How much is that?  As an educated guess, half seems reasonable.  Adding 38% from carbonization leaves only 12% to be met by other sources.  Nuclear already accounts for 20% of total US electric production, so that seems to be taken care of.

Midwest

The midwest is a fairly windy place.  The windiest sections are also among the least populated, but HVDC transmission may go some ways toward bringing supply and demand together.  The total amount of wind power potentially available is enormous; it is large even compared to total 2004 US consumption of 3953.4 billion kWh.  Per this list, the wind power from the top 20 states could meet this fraction of total US electric demand:

Rank	State	Billion kWh/yr	Fraction of 2004 US electric consumption
1.	North Dakota	1,210	31%
2.	Texas	1,190	30%
3.	Kansas	1,070	27%
4.	South Dakota	1,030	26%
5.	Montana	1,020	26%
6.	Nebraska	868	22%
7.	Wyoming	747	19%
8.	Oklahoma	725	18%
9.	Minnesota	657	17%
10.	Iowa	551	14%
11.	Colorado	481	12%
12.	New Mexico	435	11%
13.	Idaho	73	1.8%
14.	Michigan	65	1.6%
15.	New York	62	1.6%
16.	Illinois	61	1.5%
17.	California	59	1.5%
18.	Wisconsin	58	1.5%
19.	Maine	56	1.4%
20.	Missouri	52	1.3%
TOTAL		10470	265%

Of the top 10 states, I would put only Texas, Montana and Wyoming outside the Midwest; the remaining 7 have a potential of 6,111 billion kWh/year, or 155% of 2004 US consumption.  There's clearly enough energy there to do most anything we want.

Unfortunately, the availability does not coincide with the times we want to do it.  Wind has a capacity factor of about 30-40%; it may be feasible to use wind to meet more than that much demand using storage of ice and hot water, but it's not obvious to me without information I don't have right now (if demand is phased opposite to supply it could be less).  Still, if wind can handle 40% of demand and biomass carbonization can take care of 38%, the remaining 22% is very close to what's currently met by nuclear.

California and Pacific northwest

California, Oregon and Washington are problems.  California has 10% of the US population, but it ranks #17 for wind-energy resources and its potential is only 1.5% of US consumption; outside the sunny southern part of the state its renewable resources don't look good.  Oregon and Washington aren't even on the list.  Idaho's population of 1.4 million (~0.45% of US population) might be able to export much of the 1.8% of US consumption that its windpower represents, but that would easily be consumed by Washington alone.

Together, Washington and Oregon have approximately 7.8 million people, roughly 2.6% of the nation's population.  They are also major exporters of forestry products, and would be equally major sources of the biomass needed to power this scheme.  They might be able to capture a fair amount of their needs with hydro projects.  2.6% of the US's electric consumption is about 103 TWH/yr; if 38% of this comes from carbonization gas, 30% from imported wind and 20% from nuclear, the remaining 12% could come from a combination of nuclear and hydro with some extra available for export.  The exports might meet the remaining demand of northern California, or they might not.

The oceans might fill that gap.  The Pacific has considerable wave energy potential, and the wind between 5 and 50 miles off the east and west coasts might produce as much as 900 GW.  Unfortunately, the technologies for tapping this power are just being developed.

East

The East is in somewhat of a pickle.  On-shore wind resources are not great, they are far from the windier parts of the country (HVDC transmission might help, if opposition to transmission lines can be overcome), conditions are often cloudy compared to the less-humid West, and the dense coastal populations don't have anything like the per-capita biomass resources of the corn belt or PNW rain forests.  Only Maine hits the top 20 list for wind resources, at #19.

There are a lot of people to be served there.  The combined populations of New York and Florida alone are about as much as the west coast states; adding in MA, NJ, MD, VA, NC, SC, and GA brings the total to 86.3 million, or almost 29% of the total population.  Wind and wave energy from the continental shelf might also address this deficiency, but we don't even have a significant test program in place for such devices, let alone a plan for installing sufficient capacity.

If the East has to get 30% of its electricity from something outside the set of (nuclear, solar, wind, carbonization) it might be a reason to tap that 230 million tons/year of carbon I set aside above.  If the east accounts for 29% of total US electric consumption (~1150 billion kWH out of 3953.4), direct carbon fuel cells could produce 30% of that using about 47 million metric tons of carbon².  This is about 20% of the possible set-aside, and appears well within reason.

Net effects

A system based on biomass cycles lasting months and carbon storable on a scale of years addresses the irregularity of wind, solar and hydro.  The quantities available appear more than sufficient to fill in missing average supply, and it appears likely that it can be scheduled to cover immediate shortfalls.  A more detailed analysis will be needed to determine the specifics.

The next thing to notice is that it appears feasible to use renewable carbon burned in DCFC's to replace all vehicle fuel³, and wind, solar and carbonization gas to replace all coal and gas burned in electric powerplants.  Getting the job done completely probably requires off-shore wind/wave energy harvesting plus solar energy systems on a scale never seen before, but all of these technologies are under development and even being installed on a pilot scale.  About 4.8 billion barrels/year of petroleum products, 6-some quads of natural gas and a billion tons of coal would be surplus.

The impact of petroleum reductions alone would be huge.  The US balance of payments deficit for some years ran roughly equal to the value of petroleum imports.  Eliminating the demand from motor vehicles would cut a fraction roughly equal to US imports; other petroleum needs might be replaced by materials derived from the biomass processing.  The replacement of coal mining by grass cutting would eliminate many dangerous underground jobs and also many open pit mines and spoils heaps.  The mining jobs would be replaced by industrial and service jobs building, maintaining and operating the wind, solar and biomass systems.  The almost inevitable result is that the dollar would rise, and so would wages.  This translates to higher standards of living for millions.

Wind, solar and biomass would replace about 28 quads of petroleum, 6 quads of natural gas and perhaps 25 quads of coal.  Some of these would be replaced by wind, solar or biomass-derived charcoal.  As much as 230 million tons/year of charcoal would be kept back from mobile use.  This charcoal could be devoted to uses (e.g. soil amendment to retain nutrients) which sequester carbon, or used in stationary industrial processes which allow the carbon to be captured and stored.  Properly executed, this would allow the system to be strongly carbon negative.

The following would be largely or wholly eliminated:

• noise, VOC, HC, NOx and particulate emissions from vehicles
• sulfur, mercury, NOx and ash emissions from powerplants
• Net carbon emissions from the economy (some 230 megatonnes/year might be sequestered)
• Petroleum imports and the US trade deficit.
• Farm price supports for food products.

Conclusion:  to a first approximation, this is doable.  If it's doable, it's highly desirable.

Footnotes:

[1] Personal communication, Emily Heaton, assistant to Professor Steven Long, UIUC.  (back)
[2] Calculated as follows:  93960 cal/mol heat of combustion, 12 g/mol = 7380 cal/gm, 4.184 J/cal produces 32761 J/g or 9100 kWh/tonne.  80% efficiency produces 7280 kWh/tonne, so the production of 345 billion kWh would require ~47.4 million tonnes of carbon. (back)
[3] Plug-in hybrids could replace a great deal of the vehicular charcoal consumption with electricity consumed directly; the charcoal would become available for industrial use or grid power. (back)

¶ 1/09/2006 01:44:00 AM