Content uploaded by Antonio Garcia-Olivares
Author content
All content in this area was uploaded by Antonio Garcia-Olivares on Dec 05, 2015
Content may be subject to copyright.
Article
Substitutability of Electricity and Renewable
Materials for Fossil Fuels in a Post-Carbon Economy
Antonio García-Olivares
Received: 7 September 2015; Accepted: 17 November 2015; Published: 25 November 2015
Academic Editor: Robert Lundmark
Spanish National Research Council (CSIC), Institute of Marine Sciences, Ps. Maritim de la Barceloneta 37-49,
Barcelona 08003, Spain; agolivares@icm.csic.es; Tel.: +34-932309500
Abstract: A feasible way to avoid the risk of energy decline and combat climate change is to
build a 100% renewable global energy mix. However, a globally electrified economy cannot grow
much above 12 electric terawatts without putting pressure on the limits of finite mineral reserves.
Here we analyze whether 12 TW of electricity and 1 TW of biomass (final) power will be able to
fuel a future post-carbon economy that can provide similar services to those of a contemporary
economy. Contrarily to some pessimistic expectations, this analysis shows that the principle
economic processes can be replaced with sustainable alternatives based on electricity, charcoal,
biogas and hydrogen. Furthermore, those services that cannot be replaced are not as crucial so
as to cause a return to a pre-industrial society. Even so, land transport and aviation are at the limit
of what is sustainable, outdoor work should be reorganized, metal primary production should be
based on hydrogen reduction when possible, mineral production should be increasingly based on
recycling, the petrochemical industry should shrink to a size of 40%–43% of the 2012 petrochemical
sector, i.e., a size similar to that the sector had in 1985–1986, and agriculture may require organic
farming methods to be sustainable.
Keywords: post-carbon economy; 100% renewable mix; substitutability; electrification;
sustainability; sustainable industry; steady state economy
1. Introduction
It is not clear what level of energy consumption would allow a future 100% renewable energy
(RE) mix. Some studies suggest that a RE mix would allow the onset of a new growth cycle based on
green technologies until a final stationary state were reached with an input of energy larger or similar
to the present one [1,2]. However, they also conclude that if the investment in electrification of the
economy is not enough, the future stationary RE production could be under the present level. This
scenario is close to the scenario of “energy decline” defended by [3]. Holmgren, like other members of
the Permaculture Movement, assumes that the peak of fossil fuels will involve a reduction of energy
input to the economy, because the substitution of renewables for fossil fuels will not be sufficient
to maintain the enormous diversity of economic activities that were fueled by oil, coal and gas.
He suggests that a lower energy density of RE will, over time, force a ruralisation of settlements
and the economy, with less consumption of energy and resources, a progressive decline in human
populations, and abandonment of high technology. He also outlines four potential ways in which
our global society could respond and adapt to the fossil fuel peak and climate change (Figure 1):
(i) techno-explosion; (ii) techno-stability; (iii) controlled energy descent; or (iv) collapse.
The techno-explosion scenario (i) would be related to the discovery of new energy sources that
would allow a rising consumption of energy despite the fossil fuel decline; techno-stability (ii) would
involve a deployment of RE sufficient to sustain stationary consumption of resources, population and
Energies 2015,8, 13308–13343; doi:10.3390/en81212371 www.mdpi.com/journal/energies
Energies 2015,8, 13308–13343
economic activity, with new electric industrial processes able to maintain if not improve the quality
of services currently available.
Energies2015,88,page–page
2
populationandeconomicactivity,withnewelectricindustrialprocessesabletomaintainifnot
improvethequalityofservicescurrentlyavailable.
Figure1.Fourscenariosofenergyconsumption:techno‐explosion,techno‐stability,controlled
energydescent(“permaculture”),andcollapse,basedonHolmgren[3].
Theenergydescentscenario(iii)wouldinvolvereductionofenergyconsumptionand
economicactivity,adoptionof“lowtechnology”systemsandpermaculturefarmtechniques,anda
finalstationaryenergyconsumptionsomethingoverthepre‐industriallevel;finally,thecollapse
scenario(iv)wouldbeanuncontrolledbreakdownofeconomicandsocialsystemsclosetothe
Duncan[4]projection.Itwouldinvolveamajor“die‐off”ofhumanpopulationandalossofthe
knowledgeandinfrastructurenecessaryforindustrialcivilization.
Weconsiderimplausiblethetechno‐explosionscenariobecausenewtechnologiesandenergy
systemstakeabout50yearstodiffusethroughouttheeconomy[5].Thus,thenewenergysystems
thatwillreplacefossilfuelsinfiftyyearsaremostprobablythosethatarecurrentlybeingtested,i.e.,
renewables.However,REdeploymentisverydependentuponasetofmaterialswithlimited
reserves,especiallycopper[1,6].Therefore,itishighlyimprobablethata100%REeconomymay
continuethecustomaryexponentialgrowthofenergysupply,asisdiscussedinSection2.
Weassumethatoursocietyhasenoughmeansandprudencetoavoidthecollapsescenarioat
anyprice.Aswasdiscussedby[6],REtechnologiesarepresentlyavailablewhichallowthesupply
ofabout12TWofelectricitygloballywithoutshrinkingpresentreservesofcopper,lithiumand
nickel.Itisimprobablethat,havingaglobalenergysourceavailable,knowledgeandtechnology
willnotbedevelopedtouseitefficiently.Wedonotbelieveinthecornucopianmythaccordingto
whichtherewouldalwaysbeenoughenergyresourcesifonlythestockofknowledgeincreased
quicklyenough,but,inlinewithanargumentusedbyGreer[7]wedobelievethat,providedanew
energysourcehasbeenfound,humaningenuitywillcreatetechnologiesabletoexploitit,especially
ifthealternativeiscollapse.
Thetworemainingscenarios(iiandiii)arebothplausibleandwethinkthattheevolutionof
oureconomywillbecloserto(ii)orto(iii)dependingonthedegreeofsubstitutabilityofelectricity
forfossilfuelsinthepresenteconomysectors,andtheextenttowhichirreplaceableprocessesare
crucialfortheworkingoftherestoftheeconomy.
Inthisstudywetakeasareferenceafuturestationaryeconomybasedona100%renewable
mix,andweevaluatewhichpresenteconomicprocessesshouldbereplacedbyothersbasedon
electricityandrenewablefeedstocks,theconcreteformthatthenewprocessesshouldadoptin
termsofmaterialsandenergyflows,andthedegreeofsubstitutabilityofthemaineconomic
sectors.Theobjectiveisenvisageifapost‐carbonsocietywouldbecompatiblewithanindustrial
economysimilartothepresentone,andwhatwouldbethemainchangesthatthedifferent
economicsectorswouldsufferinsuchfuturesociety.
Somestudieshavequantifiedthemonetary[8]andenergeticcosts([9],submitted)that
infrastructureswouldrequireina100%REeconomy.Hereweassumeascenariowherethishuge
Figure 1. Four scenarios of energy consumption: techno-explosion, techno-stability, controlled energy
descent (“permaculture”), and collapse, based on Holmgren [3].
The energy descent scenario (iii) would involve reduction of energy consumption and economic
activity, adoption of “low technology” systems and permaculture farm techniques, and a final
stationary energy consumption something over the pre-industrial level; finally, the collapse scenario
(iv) would be an uncontrolled breakdown of economic and social systems close to the Duncan [4]
projection. It would involve a major “die-off” of human population and a loss of the knowledge and
infrastructure necessary for industrial civilization.
We consider implausible the techno-explosion scenario because new technologies and energy
systems take about 50 years to diffuse throughout the economy [5]. Thus, the new energy systems
that will replace fossil fuels in fifty years are most probably those that are currently being tested,
i.e., renewables. However, RE deployment is very dependent upon a set of materials with limited
reserves, especially copper [1,6]. Therefore, it is highly improbable that a 100% RE economy may
continue the customary exponential growth of energy supply, as is discussed in Section 2.
We assume that our society has enough means and prudence to avoid the collapse scenario at
any price. As was discussed by [6], RE technologies are presently available which allow the supply of
about 12 TW of electricity globally without shrinking present reserves of copper, lithium and nickel.
It is improbable that, having a global energy source available, knowledge and technology will not be
developed to use it efficiently. We do not believe in the cornucopian myth according to which there
would always be enough energy resources if only the stock of knowledge increased quickly enough,
but, in line with an argument used by Greer [7] we do believe that, provided a new energy source has
been found, human ingenuity will create technologies able to exploit it, especially if the alternative
is collapse.
The two remaining scenarios (ii and iii) are both plausible and we think that the evolution of our
economy will be closer to (ii) or to (iii) depending on the degree of substitutability of electricity for
fossil fuels in the present economy sectors, and the extent to which irreplaceable processes are crucial
for the working of the rest of the economy.
In this study we take as a reference a future stationary economy based on a 100% renewable
mix, and we evaluate which present economic processes should be replaced by others based on
electricity and renewable feedstocks, the concrete form that the new processes should adopt in terms
of materials and energy flows, and the degree of substitutability of the main economic sectors. The
objective is envisage if a post-carbon society would be compatible with an industrial economy similar
13309
Energies 2015,8, 13308–13343
to the present one, and what would be the main changes that the different economic sectors would
suffer in such future society.
Some studies have quantified the monetary [8] and energetic costs ([9], submitted) that
infrastructures would require in a 100% RE economy. Here we assume a scenario where this huge
investment has already taken place, and we focus on the main features of the resulting post-carbon
sustainable economy.
The manuscript is organized as follows: Section 2discusses the level of power that a 100% RE mix
would be able to supply. Sections 3–6discuss the substitutability of the current fuel-based economic
processes by electricity in a future stationary post-carbon society. The economic sectors analyzed
are, respectively: agriculture, forestry and fisheries (Section 3), transportation (Section 4), commercial
and residential sectors (Section 5), and industry (Section 6). Special care is taken in analyzing the
main material and energy inputs of the industrial sector and their future substitutes, and the technical
details of this discussion are shown in the Appendix. Section 7studies the inputs of biomass, charcoal
and biogas that such an RE mix would require and the limits that these materials would impose on
the economy. Finally, Section 8summarizes the main conclusions resulting from the analysis.
2. Energy Supply in a 100% RE Economy
In previous works [1,6] we have argued that the fossil fuels peak could take place around 2030.
In that context, a feasible way to avoid the risk of energy decline and to combat climate change is to
build a worldwide 100% renewable energy mix. However, a globally electrified economy cannot grow
much above 12 year-average electric terawatts (TWe) without putting pressure on the limits of copper
reserves. A way to achieve RE up to 12 TWe would involve deployment of floating turbines over 10%
of the continental shelves to depths of 225 m, land turbines over 5% of non-frozen continental areas,
and installation of concentrating solar power farms over 5% of the areas of high insolation (deserts).
New photovoltaic (PV) silicon panels do not use silver metallization or other scarce materials and
could contribute up to 1 TW of decentralized residential power. Hydroelectricity has a potential
of 1 TW but a fraction of this would have to be sacrificed for energy storage purposes. Hydro,
concentrating solar power, wave energy and grid integration at continental scales may be sufficient
to fit supply to demand while avoiding intermittency [6,10].
In 2005 1.47 TW of primary biomass was used for energy purposes, resulting in 1 TW of final
energy after discounting losses in transport and processing [11]. This consumption has not changed
significantly over the last few decades and we can assume that it will remain the same in a future
post-carbon society. Therefore, the total renewable power (including biomass) consumed by end
users may be 13 TW, and the contribution of biomass to the renewable mix would be 8%.
An RE mix such as the one proposed above would have an Energy Return on Energy Investment
(EROEI) of about 15, 35% lower than the estimated EROEI of the present energy mix. That should be
sufficient to sustain an industrialized economy in the second half of this century provided substitution
of electricity for fossil fuels is feasible and intelligently made [10].
Could technological progress enable this cap of about 12 TW to be surpassed, and hence make
indefinite exponential growth of energy production possible? The incorporation of fusion energy to
the electric grid is expected at some date close to the end of this century [12]; however a fusion-based
energy system has the same dependence that renewables have on copper-based devices, and thus the
electrified economy that it makes possible is as copper-consuming as the one based on renewables.
An energy production mix based on renewables and fusion could be scaled exponentially only if a
major substitution of copper for aluminum (which is abundant), graphene, and high temperature
superconductors (HTS) were successfully implemented in electricity generators, engines and wires.
Most of the consumption of copper in a post-carbon economy will come from windings in generators
and electric motors [1]. Squirrel-cage motors frequently use aluminum instead of copper for the
conductive bars, however they are built for low and medium power (a few kW) and are outside
the range required by a power generator (MW). Thus, the feasibility of this substitution is very
13310
Energies 2015,8, 13308–13343
uncertain [2] and, also, even if new aluminum-based generators and motors were developed in the
future, technological innovations take 40–50 years to expand throughout the economy [5]. Thus, the
scenario of a ceiling of 12 TW for a future 100% RE mix at the second half of this century seems more
realistic and prudent than assuming major breakthroughs in the expected technological development.
World primary energy production was 496 EJ in 2005, which is equivalent to 15.7 TW of
mean annual power. Primary energy includes the energy embodied in energy repositories such
as oil (167.4 EJ), natural gas (99 EJ), coal (122.2 EJ), nuclear power (28.5 EJ), biomass (46.3 EJ),
hydropower (30.1 EJ) and new renewable energies captured by solar panels, wind turbines, etc.
(2.3 EJ). Due to conversion and distribution losses the final energy that fueled the global economy
was 330 EJ, or equivalently, 10,460 GW of fuels and electricity. The expected error value is a few
percent due to uncertainties in the statistical processing of the original data. The share of this power
consumed by end-use sector in 2005 was: 27.7% transportation, 26.8% industrial, 9.5% commercial,
24.6% residential, 9.2% feedstocks, and 2.3% agriculture, forestry and fisheries ([11], Table 1.2 and
Figure 1.5).
In what follows we analyze whether 12 TW of electricity and 1 TW of biomass (final) power
will be able to fuel a future post-carbon economy that can sustain similar services to those of the
2005 economy, and what changes should be introduced into the main processes to maintain, as far as
possible, the current economic services.
3. Agriculture, Forestry and Fisheries
Energy input to agriculture was 0.24 TW in 2005 (Table 5) (from [11], Table 1.2). We will assume
that 8% of world biomass input of energy into the economy is also representative of the biomass
consumption of agriculture. The share of fossil fuels in agriculture is highly variable between
countries. In the USA 60% of energy input was fuel (diesel and gasoline), 5% gas and liquid
petroleum, and 35% electricity in 2013 [13]. We will take this share as representative of all developed
countries. However, western agriculture is an extreme case of high use of machinery and irrigation
and low use of labor, while developing countries are closer to the opposite extreme. Direct energy
inputs of a typical rainfed arable crop farm are probably close to this second extreme and, according
to Sims ([14], see Figure 2.19) consist of 20% liquid fuels (diesel), 0% electricity and 80% feedstocks.
We will take this latter share as representative of the fossil fuels used for agriculture in developing
countries. According to Pimentel [15] roughly the same energy is used in developed and developing
countries in agricultural production.
We assume that oil is used mainly for transportation, which will be electrified in the future.
Tillage sometimes requires high power tractors that, in the future, could be powered by fuel cells.
However, for other generic farm work, if a grid connection point is installed on the farm, a fleet
of smaller electric tractors would have no problem doing the open field work, since the battery
recharge could be as frequent as needed, and some electric tractors would be working while others
are recharging. For small farmers, the minimal traction equipment required would consist of one
tractor and two rechargeable batteries. Of course, in such a future system, full connection of farms
to the electric grid will become necessary. If that connection was not available or the farmer could
not afford to own a tractor, tillage would have to be based on human labor, as is the case currently
with small farmers in developing countries. We will assume that electric battery vehicles will replace
only 23% of the total oil consuming machines of the sector, and 77% of it will be replaced with fuel
cells tractors. The former figure corresponds to the “miscellaneous” category in the diesel input of a
typical arable farm (see Figure 2.20 in [14]).
We assume that gas is used for heating, which will be replaced by electric resistance heating
that is 97% efficient [16]; and that biomass will be used in the future with the same efficiency, about
22% for heat production in rural stoves [17]. We assume that gas is used mainly for water and space
heating with the typical efficiency of a condensing boiler (about 90%) [18]). Oil is assumed to be used
for transport and tilling purposes with the efficiency of diesel motors (40%) (Table 1).
13311
Energies 2015,8, 13308–13343
Given the above efficiencies, the 2005 power used in this sector (240 GW) will become 188 GW
in a post-carbon economy. The expression used to make the calculation is the following:
pr“pfr0.5 p0.92 ˆ0.60 p0.23 edi{eba `0.77 edi{efcq ` 0.08 est {est `
p0.92 ˆ0.05qecb{eer `0.92 ˆ0.35q ` 0.5 p0.77 edi{efc `0.23 edi{eba qs
where pris the mean annual power demanded by a renewable agricultural sector; pfis the demand of
the sector in 2005; and edi, eba , efc, est, ecb, eer are the efficiencies of diesel engines, battery motors, fuel
cell motors, rural stoves, condensing boilers, and electric resistance, respectively. Similar expressions
are used for each economic sector hereafter.
Appendix A7 of the Appendix analyzes the feasibility of producing ammonia and other
agricultural feedstocks in a post-carbon society. While it seems feasible to produce ammonia
renewably from biogas and hydrogen, renewable sources of phosphorous and potassium are not
available to the present-day agriculture and this is a major problem that must be solved in a future
post-carbon economy.
In the long term (from a few decades to a century) the world will run out of potassium
and phosphorus, so the only available way to sustain productive agriculture will be to resort
to organic farming [19]. A move to fully organic farming will require good knowledge of soil
ecosystems, and must be fine-tuned to the local climate. Also, it could reduce the output of grains
by 20%–30% in the short term. However, in the long term, organic soils hold up in quality and
even improve, and they resist erosion better than standard farmed soils. Therefore, in a few decades
organic soils may achieve productivities close to those of artificially fertilized soils. Some models
indicate that organic agriculture could produce enough food for the current world population:
2640–4380 kcal/person/day [20–22]. In addition, in the majority of cases studied, organic systems
are more profitable than non-organic ones due to lower input costs [23]. Use of plant, animal and
human wastes for production of compost and natural fertilizers will allow recycling of phosphorous
and other nutrients. This is the only available solution to the future decline of phosphorous mining
that has been predicted for 2040–2050 [24,25].
Organic farming as a solution to the decline of fertilizer availability is, however, not compatible
with an always rising population, since available arable land per capita has decreased from 0.5 to
0.24 ha between 1961 and 2005 [26]. There is not much room for additional decrease of arable
land per capita, because the supply of the principle cereals is saturating at their biological limits:
7–8 tonnes/ha for wheat and 7 tonnes/ha for rice, despite the increasing input of fertilizers [27].
Depletion of groundwater basins must also be curbed, which requires putting a stop to population
growth. About one third of the Earth's largest groundwater basins are being rapidly depleted
by human consumption, which means that significant segments of the population are consuming
groundwater without knowing when it might run out [28].
Renewable exploitation of forests will also be crucial in a post-carbon economy, not only for
reasons of climate and biodiversity, but also because of their importance in renewable production of
wood biomass and charcoal (see Section 7).
Regarding fisheries, fishing fleets consisted of about 4 million vessels in 2002, 1/3 decked and
2/3 undecked, with 65% of undecked boats not using mechanical propulsion systems [29]. This is
equivalent to 1.32 million vessels with large motors and 0.23 million vessels with light motors;
0.005 and 4 ˆ10´4times the number of large and light motors being used for the transportation
sector, respectively. In principle, all these combustion engines could be replaced by fuel cell motors
in a future post-carbon economy, as discussed in Section 4.
4. Transportation
Hydrogen has certainly been proposed as an energy source that is similar to oil and natural
gas, and that could be used for transportation. However, present electrolytic systems require around
13312
Energies 2015,8, 13308–13343
60 kWh to produce 1 kg of hydrogen [30], which implies an energy efficiency of 65% if we take HHV
of hydrogen as output. This implies that hydrogen produced and consumed on-site has 1.53 times
more electricity embedded than its own HHV content. If losses along the hydrogen conversion chain,
i.e., containing, liquefaction, transport and handling are also taken into account, the result is that the
production of hydrogen for consumption by a jet turbine or fuel cell requires 1.8 times its HHV energy
content in the form of electricity [31] or, equivalently, 2.1 times its LHV content that is the metric that
is being used in Table 4.
In addition, electrical motors are more efficient than fuel cell motors (Table 1) and, for both
reasons, a fuel cell vehicle requires 3.6 times more integrated electricity consumption than an electric
vehicle [31]. Also, the hydrogen produced is five times more expensive than the direct use of
renewable electricity [32]. Thus, the direct use of electricity by motors is a cheaper and more efficient
way to produce movement, and is the most promising option in future ground transportation [1].
The exception would be aircraft and other forms of transport that are not able to receive energy from
the electric grid, as well as vehicles with specific requirements for both autonomy and power, such as
ambulances, fire engines and police cars.
Efficient land transport should ideally be based on electric trains for freight and passengers
between cities, and electric vehicles (EV) for short-distance transport between cities and
villages [1,33]. However, we will assume, pessimistically, that future land transport will instead be
based on the electrification of the present vehicle fleet.
About 600 million small vehicles, 205 million commercial (heavy) vehicles and 215 million
motorcycles were circulating in 2005 if we linearly extrapolate the trend observed by [34] for the
period 2000–2003. The number of commercial vehicles that will use fuel cells is very dependent of the
future weight given to trains for long distance freight. We will assume that its number will be only
10% of the number of commercial vehicles, because with this percentage 99% of Pt reserves (and 22%
of Pt and Palladium reserves) would have to be used in the fuel cell electrodes. Palladium is more
abundant than Pt and it has been reported to be a possible substitute of Pt for fuel cells, although not
with identical performance [35]. Lithium-ion batteries have the largest energy density and, for this
reason, are the most used in current electric cars. Taking typical battery capacities and power for these
three classes of vehicles (Table 1) and the density of metal used in their respective motors (Table 2),
the quantity of lithium (Li) that such a fleet would require would be 7.8 Mt. Alternatively, nickel
Na-NiCl2(Zebra) batteries are technically feasible. If these batteries were used to renew the world
fleet, 65 Mt of nickel would be used [1]. These figures amount to 58% and 80% of present reserves of
Li and Ni, respectively [36]. If 50% of Li batteries and 50% of Ni batteries were used, 29% and 40%
of the present reserves of Li and Ni would be used, respectively. Electrification of vehicles will be
necessary, but given that reserves cannot be indefinitely expanded [2], the number of vehicles that a
future post-carbon society could sustain is roughly the number we have currently. A larger number
would endanger the availability of Li and Ni for other economic demands.
Table 1. Classes of electric vehicles (column 1), global number in 2005 (column 2), typical peak power
of its battery (column 3) and battery capacity (column 5). The parameters used are discussed in [1].
Kind of EV Number in 2011 (millions) Power (KW) Battery Capacity (KWh)
Light 600 60 22.4
Heavy 205 179 67
Motorcycle 215 3.6 1.2
13313
Energies 2015,8, 13308–13343
Table 2. Values used for the estimation of metals required by the transport system. Density refers to
the mass of metal used per unit of power or per unit of energy stored in engine, battery or fuel-cell.
The parameters used were discussed in [1].
Metal Density Reserves (106t)
Copper 0.73 kg/kW 680
Lithium 0.3 kg/kWh 13.5
Nickel 2.5 kg/kWh 81
Platinum 0.004 kg/kW 0.015 (0.07) a
aReserves of Pt group metals (Pt, Pa, Rh, Ru, Ir, Os).
Marine shipping consists of about 87,500 ships, not including naval and fishing boats [37]. Naval
ships are only a few thousand globally [38], therefore we can take 100,000 as an upper estimate of the
number of non-fishing ships. Assuming that this kind of vessel uses four heavy engines, the total
number of marine engines would be 1.72 ˆ106heavy and 0.23 ˆ106light engines. We will use
60 kW as a typical power for a light engine, and 179 kW as the power of a heavy engine [1].
To complete the number of light and heavy motors that would probably require fuel cells, we
need to add the world number of heavy farm tractors, the number of ambulances, and the number of
police cars and fire vehicles. Farm tractors were estimated to number about 28,570,900 in 2005 [39].
We will assume that 10% of these will be high power tractors that will need fuel cells. The number of
ambulances per capita was about 1 per thirty thousand people in Turkey [40] and 1 per 4350 people
in Australia ([41] Table 9A.39), figures that we will consider to be representative of an average
developing country and a mean developed country, respectively. We will assume that the number
of police cars is the same as the number of ambulances. Fire engines will be considered to be far
fewer in number. Under these assumptions, and considering that developing countries included 81%
of world population in 2005, we obtain a total number of 1,200,000 light engines and 4,577,000 heavy
engines, which must be considered as orders of magnitude, and not precise estimates. Assuming that
0.2 kg of Pt or its substitute, palladium, is needed for a fuel cell of 50 kW [42], the mass of Pt or Pa
necessary for such a fleet of fuel cell engines is 18,250 t. This is 28% of present reserves of platinum
and palladium (66,000 t) ([36], platinum). Thus, fuel cell deployment could be sufficient in principle
to meet the current requirements of special vehicles and 10% of commercial vehicles, even though not
very scalable, except if palladium and new materials were developed to fully substitute for platinum
in the fuel cell catalyzers.
Assuming that fuel cell vehicles will mainly be ambulance, police and 10% of commercial
vehicles, and taking into account the energy required for producing electrolytic hydrogen, and the
efficiencies given in Table 3, the power required for road transport would be 893 GW.
We assume that half of world train transport has already been electrified and that the other half
uses diesel locomotives. Assuming that most of the energy consumption of train transport is for
locomotive traction, and using the motor efficiencies shown in Table 1, we estimate that the power
required for 100% electrified railroad transport is 52 GW. The energy demand of the transport system
could be substantially reduced if current ground transportation based on cars were replaced with a
substantial increase in rail transport. As an example, a typical intercity train transports eight times
more seated passengers per MW than a car (204 people/MW vs. 25 people/MW) [10]. Therefore, if
well organized, it has the potential to reduce the energy consumption of future road transport by a
factor of eight.
Regarding marine transport, we assume that most of the energy consumed by the sector goes
to vessels and mechanical motors. Under this assumption, we estimate that to produce the service
obtained in 2005 with 285 GW, a post-carbon economy would require 830 GW. This factor of three
increase in energy demand derives from the need to produce hydrogen for fueling the marine fuel
cell engines. However, 10%–35% of this energy could be saved if wind systems were used in tandem
with the motors of the vessel, such as the SkySails kite system [43]. In kite propulsion systems,
13314
Energies 2015,8, 13308–13343
high-altitude wind energy is directly converted into traction without intermediate conversion to
electricity, and with the low cost and maintenance that is typical of low-technology systems.
About 89% of the transport system could possibly be electrified in a post-carbon society, since
world air transport uses only 11% of the energy demanded by the transport system ([11], Table 1.2). To
sustain the same volume of air traffic as in 2005, air transport would need 3% (11% of 27%) of 10.4 TW
from fuel, or 0.32 TW from hydrogen (H2) and biofuels. If electrolysis were used to supply H2, about
696 GW of renewable electricity would be required. This result has been included in Table 5.
As we will see in Sections 6and 7in a post-carbon economy demand for woody and non-woody
biomass will be superior to the available renewable biomass production if we want to maintain the
economic levels of industrialized economies such as those of 2005 or 2012. For this reason, hydrogen
would be a more convenient fuel than biofuels in the long term. On the other hand, biofuel-oriented
crops may compete with food crops for arable land and should be avoided wherever possible.
However, as we will see in Section 7, about 98 GW of biogas could be saved if future agriculture
becomes fully organic and ammonia is no longer industrially produced. A similar situation would
result if ammonia were produced directly from hydrogen and air and not from biogas. This would
make available a fraction of agricultural wastes able to produce around 98 GW of biofuels. Provided
that the present structure of the transport system is not modified, this would amount to 11% of the
future aviation demand.
5. Commercial and Residential Sectors
Energy consumption in this combined sector was between 3.6 and 3.8 TW in 2005. The fractions
of biomass, electricity and central heat, gas, coal, oil and new renewables used by this sector have
been calculated in Grubler et al. ([11], Figure 1.5).
The services required by this sector include food processing, hygiene (water heating), thermal
comfort (air conditioning), illumination, mechanical work, and communication. All these services
could, in principle, be supplied by electricity. However, we will assume that current level of use of
biomass for heating and cooking will remain the same in a future post-carbon economy (about 1 TW).
About 1 TW could be produced by residential PV power if 12.5% of world populated areas were
covered with PV panels [6]. This would meet 39% of residential needs, with the remaining 61%
obtained from the grid.
Table 3. Current efficiencies of different motors and heating devices.
Device Efficiency Reference
Gasoline motor 0.25–0.30 [44]
Diesel motor 0.40 [45]
Battery-powered motor 0.80 [46]
Overhead line electric motor 0.95 [46]
Fuel cell motor 0.50 [47]
Coal braziers 0.97
Coal cooking stoves 0.22 [17]
Electric resistance (air heating) 0.97 [16]
Electric resistance (cooking) 0.74 [48]
Heat pump 3.0 [49]
Condensing boiler 0.90 [50]
Gas hob 0.40 [48]
We assume that oil consumed by this sector goes entirely to internal combustion engines, which
will be replaced by electric motors; that coal is used 50% in coal braziers and 50% in cooking stoves,
where the former will be replaced by electric resistance and the latter will be replaced 50% by electric
resistance and 50% by heat pumps. We assume that natural gas is used 50% for cooking and 50% for
air heating, where the former will be replaced by electric resistance and the latter will be replaced
13315
Energies 2015,8, 13308–13343
50% by condensing boilers and 50% by heat pumps. Table 3shows the current efficiencies of different
motors and heating devices.
Under these assumptions, 1 TW of biomass and 1.93 TW of electricity may be sufficient to supply
the same services that the 2005 economy supplied to these sectors with 3.8 TW. The first lines of
Table 5summarize the energy consumed in 2005 by the transport, commercial and residential sectors,
agriculture, forestry and fishery sector, and the energy that would be probably consumed by these
sectors in a post-carbon society.
6. Industry
In 2005 the total energy used by industry was 88.2 EJ of fossil fuels and electricity and about
25 EJ of feedstocks„ which is equivalent to 3.59 TW of mean annual power ([11], Figure 1.5). Table 5
(column 2) shows the share of the industrial use of energy that year.
The industrial sector is strongly dependent on electricity, coal, gas and oil, in this order (see
Table 4, based on [51], Table 8.22). However, oil is especially important in the petrochemical and
construction sectors. Future decline of many feedstocks coming from oil and coal will force us to
substitute these feedstocks for others coming from renewable sources, or to abandon the process.
The degree of substitutability of fossil fuels in the different industrial sub-sectors and their
energy cost has been analyzed in detail in the Appendix. Expressions similar to Equation (A1) have
been used for each industry to estimate the energy use in a post-carbon economy. The parameters
of Table 3have been used for the efficiency of different devices, and the efficiency of the central
heating and electro-chemical processes of the specific industry, as well as other processes such as
steam generation, have been estimated from the Sankey (energy flow) diagrams produced by the
US-DOE for the specific industry [52], when available. The final result of this analysis is shown in
Table 5, rows 10 to 24. As can be seen in that table, a post-carbon industry would use less energy
than the present industry in all the sectors except iron and steel and non-ferrous metals. The energy
saving would be specially high in nonmetallic minerals production due to the high heat loss of
conventional kilns, which could be minimized by using electric and microwave assisted heating
(see Appendix A6). Many industries, such as transportation equipment, machinery, textiles and
others are partially electrified in their central processes and their full electrification should not be
a major problem.
The work done in open terrain for wood extraction, construction and mining will have to be
fully electrified. Indeed, mobile electric construction machines will have more limitations than fuel
powered ones and therefore open field work will have to be planned in a different way. For instance,
it will require the installation of temporary grid connections, fleets of (electric) vehicles with lower
autonomy and power but in larger number than at present, and different design approaches for
the extraction and transport of heavy loads. If we examine a map of the distribution of electricity
sub-stations in Spain, which supply power to villages and factories, we can observe that even in
less densely populated regions (La Mancha) no point is further than 40 km from a sub-station, and
that villages with an electrical supply can be found at a maximum of about 20 km [53]. Thus, any
mining or construction project outside of urban areas will have to include the building of a power
connection (of up to 40 km in Spain, maybe longer in some developing countries) to the nearest
power sub-station. Once the connection is installed, power shovels could be connected to the grid
as do actually many dragline excavators and giant power shovels and bucket wheel excavators used
presently in mining. Transport of moderate loads could be made by a fleet of electric vehicles, since
the battery recharge could be as frequent as needed, and many electric vehicles would be working
while others are recharging. Finally, transport of heavy loads on complex orography could use fuel
cell powered vehicles.
13316
Energies 2015,8, 13308–13343
Table 4. Energy input (GW) for the global industrial sector by industry and energy type in 2005 (based on [51], Table 8.22). Energy from fossil fuel feedstocks is
excluded. A zero is displayed when the figure is under 0.05 GW.
Industry Coal and
Derivatives
Crude, NGL,
Feedstocks
Petroleum
Products
Natural
Gas Geothermal Solar, Wind,
Other
Combustible Renewables
and Wastes Electricity Heat Total
Iron and steel 250.8 0 20.1 77.8 0 0 8.5 103.3 15.7 476.3
Chemical and
petrochemical 58.1 0.1 80.9 150.4 0 0 3.0 113.2 45.9 451.6
Non-ferrous metals 15.4 0 10.3 19.2 0 0 0.2 67.5 2.7 115.3
Non-metallic minerals 181.3 0 48.2 66.7 0 0 6.7 42.7 3.4 349.0
Transport equipment 4.8 0 4.0 13.4 0 0 0 18.7 4.3 45.2
Machinery 13.2 0 15.0 27.9 0 0 0.1 66.3 6.1 128.6
Mining and quarring 9.6 0 18.1 12.8 0 0 0 26.2 3.3 70.0
Food and tobacco 25.2 0.1 34.4 43.4 0 0 35.1 40.7 11.3 190.2
Paper, pulp and printing 25.2 0 19.1 33.9 0.2 0 65.6 53.9 6.7 204.5
Wood and wood products 2.8 0 4.5 3.8 0 0 13.1 11.1 6.8 42.0
Construction 6.8 0 26.6 4.9 0 0 0.2 6.8 1.6 46.9
Textile and leather 14.3 0 11.6 11.6 0 0 0.3 25.1 7.6 70.5
Non-specified industry 75.0 4.9 133.7 108.0 0.2 0.2 105.4 130.4 30.6 588.3
Total (GW) 682.4 5.2 426.4 573.8 0.4 0.2 238.2 706.0 145.9 2778.4
13317
Energies 2015,8, 13308–13343
Iron and many non-ferrous metals require carbon as a reducing agent in their production, which
could be obtained from renewable charcoal in a post-carbon economy. However, demand for charcoal
by the petrochemical industry will saturate in a future post-carbon economy (see Appendix A8) and
also its future renewable production will probably not be sufficient to supply the large quantities
of coke needed for iron reduction (see Table 6). Fortunately, sponge iron produced by the direct
reduction process is a proven technology that is becoming increasingly popular due to the shortage
of high-quality coking coals [54,55]. In the process of producing sponge iron, hydrogen is frequently
employed as the reductive gas, through the following aggregate reaction:
3Fe2O3`9H2Ñ6Fe `9H2O
and this process could be the best suited to a future post-carbon economy. Then, crucible steel may
be produced by diffusion of charcoal, a process that is known in the industry as “carburization” [56].
Other metals that could be primarily produced by direct reduction with hydrogen are copper, tin
and nickel (see Appendix A5). Some metals and ferro-alloys require coal as reducing agent, however,
as we will see in Section 7, present levels of ferro-alloy production could be supplied through the
future use of renewable charcoal or methane. Secondary production of the main metals (recycling)
already involves the use of electric arc furnaces which can be, in principle, fully electric.
Table 5. Energy consumed by the transport, commercial and residential sectors, agriculture, forestry
and fishery and industrial sectors, in 2005 (column 2) and in a post-carbon economy (column 3).
Industrial energy use in 2005 is based on Figure 8.1 in [51]. The energy required for future hydrogen
production in some sectors is included in the corresponding box of column 3. An important fraction
of energy consumed by the chemical and petrochemical sectors (759 GW) is embedded in feedstocks,
as well as 34 GW of coal for chemicals, graphite and steel production, which have been implicitly
accounted in “Chemicals and Petrochemicals” (26 GW), or explicitly included in “Non-ferrous metals”
(1 GW) and “Iron and steel” (4.8 GW).
Energy End Use Final Energy in 2005 (GW) Final Energy in Post-Carbon Economy (GW)
Road 2100 893
Rail 73 52
Shipping 285 830
Air 330 696
Pipelines 90 0
Total transport 2900 2471
Residential and commercial 3800 2954
Agriculture, forestry, fishery 240 188
Wood and wood products 42 35.8
Transport equipment 45.2 39.2
Machinery 128.6 110.7
Construction 46.9 27.1
Mining and quarrying 72 14
Textiles and leather 72 64.1
Renewable feedstocks - >541 to >596 *
Non-ferrous metals 115 + 1 115 + 1 + 1.6 + 0.2
Food and tobacco 190 162.6
Paper, pulp and print 216 173.2
Non-metallic minerals 350 236
Iron and steel 476 + 4.8 720 + 4.8
Chemicals and petrochemicals 1057 628 to 785 **
Others 588.3 424.5
Total Industry 3405 3292 to 3504
Total 10,343 8912 to 9124
* This power is embedded in the raw materials used for producing charcoal and biogas and it does not add
to the electrical power produced by a post-carbon economy. The range depends on the use or not of best
available technologies (BAT) for charcoal production; ** The first (second) value corresponds to 15% (the same
as in 2005) production of ammonia and 49% (45%) of HVCs production of 2005, respectively.
13318
Energies 2015,8, 13308–13343
The chemical and petrochemical sector is the most energy demanding of the industries,
1057–1075 GW in 2005, including the embedded feedstock ([51,57], (Table 5)). About 50% of
this energy, 544 GW, was used to produce high value chemicals (HVC), mainly olefins (ethylene,
propylene and butylene) and aromatics (benzene, toluene and xylenes). An important fraction was
also used for ammonia production (188 GW), and for methanol production (37 GW). From these
primary chemicals, a large number of secondary chemicals and end-products are produced, by
processing and polymerization, which consumes 4% of the sector energy (44 GW).
The HVC production is currently based on naphtha, one of the main derivatives of petroleum,
which may be replaced by charcoal in the coal to olefins process (see Appendix A8 for details).
However, for obtaining the level of HVCs production of 2004 we would precise 3.7–4.6 times the
renewable potential of charcoal production estimated in Section 7. This suggests that maintaining
the current level of petrochemical production may be unfeasible in a post-carbon economy, and
that we will have to reduce our present consumption. Appendix A8 of the Appendix details the
parameters used in the estimation of energy consumed by the chemical and petrochemical sectors,
which is showed in Table 5. Some biological substitutes for the principle petrochemical products are
also suggested.
7. Biomass, Charcoal and Biogas Production in a Post-Carbon Economy
Charcoal and biogas will have to be produced in larger quantities in a post-carbon economy
than they are at present, in order to compensate for the decline of fossil coal, oil and gas. That
increased production should be obtained from renewable biomass production. Governments will
have to force this to happen with laws forbidding plant cultivation for biofuels and biogas production,
and nonrenewable logging. If they do not enforce this, it is probable that production of biological
feedstocks will compete with agriculture, as is presently happening in the USA, Indonesia, Africa and
other regions [58,59]. Clearing of tropical rainforests for rapeseed and palm oil production is currently
destroying biodiversity and releasing 17 to 420 times more CO2than the greenhouse reduction that
these biofuels would provide by displacing fossil fuels [58].
The post-carbon stationary economy that we suggested in the Introduction would avoid these
problems since the exploitation of wood and agricultural waste (needed for producing charcoal
and biogas) would be stationary and sustainable, and would use agricultural crops to produce
biogas. If we focus exclusively on the sustainable exploitation of temperate and tropical forests,
up to 3 t/ha/year of woody biomass could be extracted from these forests whilst maintaining
stationarity [60]. The carbon content of tropical wood is 47.3% on average [61] and similar in
temperate wood [62]. Forests occupied 4000 million Ha in 2012 [63], assuming that 10% of this
extent were sustainably exploited for wood extraction and that 5 t of wood are needed to produce
1 t of charcoal [64], 240 million tons per year of charcoal could be produced, almost 5 times the
present production.
An alternative estimate can be made from the potential of renewable biomass, which is about
103.8 EJ/year, 40% of it coming from woody biomass [65]. World biomass consumption for cooking,
heating, and industrial wood-related activities was 46.3 EJ/year in 2005, 75% of it in developing
countries and 25% in developed countries. Assuming that these end-uses are inflexible, and taking
into account that 40% of biomass used comes from wood, 18.5 EJ/year of wood consumption must
be reserved for these end-uses and 23 EJ/year of wood would be available for other industrial uses,
which is equivalent to 1.2 ˆ109t/year of dry wood if we take 19,000 kJ/kg as the heat content of dry
wood. Assuming again 5 tons of wood for 1 t of charcoal produced, 1.2 ˆ109t/year of wood could
be sufficient to produce 240 ˆ106t/year of charcoal, the same result as in our previous estimation.
Mass yields from a Casamance kiln and a well-managed traditional mound kiln are somewhat
higher than the world average, about 25% [64], therefore the previous potential could be increased to
300 ˆ106t/year if future governments encouraged the use of best available technologies. Finally, this
13319
Energies 2015,8, 13308–13343
figure could be increased by an additional 6% if governments of North America and Europe would
force the replacement of wood fuel with electric heating (see Table 3 of [65]).
Methane obtained from biogas could be an alternative source of a reducing agent for ferro-alloys
and other metals. In the future, energy crops will probably compete with food production for soils
and so should be avoided, however biogas could be obtained from crop residues and urban wastes.
Current global production of methane from biogas is only 20–26 ˆ106t/year, but global potential
has been estimated at approximately 900 ˆ109m3[66] or, equivalently, 600 ˆ106t/year. If only
the European Union, USA and China were able to make the investments needed to develop the new
biogas infrastructure, the potential for CH4production would be approximately 215 ˆ106t/year.
As a consequence, present levels of tin, lead, zinc and ferro-alloy production could be supplied
through the use of charcoal or methane by a small expansion of the present level of woody biomass
and biogas production. This is apparently feasible in both cases.
Charcoal and natural gas production do not involve rock grinding or underground digging and
ventilation, both of which are very energy consuming. Therefore, the energy needed to supply
the reducing carbon from charcoal should not be larger than that currently needed to extract it
from mining.
HVCs production is the most demanding sector regarding charcoal and natural gas and the route
from methane to olefins via methanol and dimethyl-ether is the most efficient (see Appendix A8).
After discounting 65.6 ˆ106t of natural gas needed for ammonia production, 149.4 ˆ106t of CH4
remain, which would allow production of 92.6 ˆ106t of HVCs using that process. If we need to
produce 287 ˆ106t of HVCs (Appendix A7), 194 ˆ106t should be produced from charcoal. Given
that 4.1 t of carbon are needed to produce 1 t of HVCs (Appendix A8) and assuming 75% carbon
content in charcoal [67], we find that 1063 ˆ106t of charcoal would be required.
Table 6summarizes the new production of charcoal, hydrogen and biogas required for the
different sectors if a post-carbon society had to reach the same production level as in 2005. The global
production of charcoal and biogas that is probably obtainable in a renewable way is also indicated.
High quality industrial charcoal has 75% carbon content [67], and a factor 1.33 has been used to
translate carbon required for charcoal used in each sector. All the energy requirements of the marine
and aviation sectors in 2005 are assumed to go to propulsion, in the form of hydrogen, and the same
efficiency to produce movement is assumed for the energy of hydrogen and kerosene in aviation jets.
13320
Energies 2015,8, 13308–13343
Table 6. New production of charcoal, hydrogen and biogas required for the different sectors if a
post-carbon society had to reach the same production levels as in 2005. The second figure in the third
column represents the electrical power required to produce the hydrogen.
End-Use Sector Charcoal
(103t/year)/(GW)
Hydrogen
(103t/year)/(GW)
Biogas
(103t/year)/(GW)
Vehicles fuel cells * 17,000/138
Marine fuel cells 60,048/411
Aviation 85,902/588
Iron and steel 64,000/440
Copper reduction 600/4
Tin reduction 40 or 5/0.03
Nickel reduction 81.8/0.6
Lead reduction 710
Zinc reduction 1600
Ferro-alloys 30,100
Graphite 1334
Ammonia 24,706/169 or 65,634
High value chemicals 1,063,230 149,366
Total 1,097,010 252,343/1,751 215,000
Global potential 240,000/221 to 300,000/276 Not applicable 215,000/320
Percent of global potential 366 to 457% Not applicable 100%
* We suppose that 10% of commercial (heavy) vehicles use fuel cells.
As can be seen in this table, charcoal needed for a future post-carbon economy is about
4 times above the global potential. This implies that, if the charcoal demand of metal industries is
respected, only 45%–49% of the HVCs produced in 2004 could be produced in such a future economy,
92.6 ˆ106t from biogas and 37.7–48.7 ˆ106t from charcoal.
The routes from carbon and natural gas to HVCs are also more energetically expensive: 45 GJ per
ton of HVC produced and 29 GJ per ton of HVC produced, respectively, to be compared with 18 GJ per
ton of HVCs produced with the conventional naphtha route (embedded energy not included). If we
take into account these factors, the quantities of available charcoal and biogas and the quantities of
charcoal and gas to be used for HVC production (Table 6), and compare with the energy used in 2004
to produce HVCs ([51], Table 8.5) we find that 544 GW would be required, to be added to the energy
cost of producing ammonia, ethanol and all the petrochemical products. We will assume that, in a
post-carbon economy, ammonia production will initially be the same as in 2005 (177 GW); but that it
will decrease to 15% of this figure following the implementation of a future organic agriculture. 15%
of ammonia is expected to continue to be necessary for textile fiber processing, nitric acid production,
refrigeration for bulk food storage, water purification, antimicrobial agents, rubber production, metal
plating, and other uses [68,69]. We will assume also that other petrochemical activities different from
HVC and ammonia production will diminish in the same proportion (45%) as the decrease of HVCs.
With these assumptions, the total consumption of a future petrochemical sector amounts to 785 GW,
decreasing to 628 GW when full organic agriculture is implemented.
8. Discussion and Conclusions
A future post-carbon economy is the only sustainable solution to the present challenges of energy
security and rising environmental impacts. Such a future economy should be based on a mix of
decentralized and centralized renewable sources. Some studies have warned of the necessity and
difficulty of the transition to such a society, and others argue that such a transition is incompatible
with the maintenance of an industrial society.
The present analysis shows that a future post-carbon society seems capable of sustaining an
industrial developed economy, if the investments needed to implement the appropriate process
13321
Energies 2015,8, 13308–13343
substitutions are done. Actually, a post-carbon economy based on direct use of renewable electricity
is able to provide similar services to those of the 2005 economy.
However, global electrification would put the finite supplies of copper, nickel, lithium and
platinum under rising stress. Therefore, a post-carbon economy will have to adapt, sooner rather
than later, to a stationary consumption of energy and materials. Further growth in the throughput
of that future economy will be dependent on improvements in efficiencies of charcoal and biogas
production, as well as on dematerialization of processes.
About 10.3–10.4 TW of final energy was required in 2005 for all the economic end uses. As can be
seen in Table 5, the conclusion of our analysis is that a 100% RE economy trying to provide the same
services as the economy of 2005 would require initially about 9.1 TW, and 8.9 TW after BAT in charcoal
production and implementation of full organic agriculture. Thus, a 100% renewable economy would
require 87% of the energy consumed in 2005 to provide similar services, except for the petrochemical
sector, which would be similar to the one of 1985. No savings due to efficiency improvement have
been assumed, however demand-side energy-conservation measures could allow savings of about
5%–15% of energy demand [31], which would amount to 0.5–1.3 TW in a post-carbon economy.
Once we have obtained these general figures, a comparison with the recent or near
future economy becomes easier. As an example, total final consumption of energy in 2012 was
8979 Mtoe [70], which is equivalent to 11.9 TW. If a post-carbon economy would require 87% of
that energy, 10.4 TW would supply similar services to those supplied by that recent economy. Thus,
12 TW of renewable electricity and feedstocks and 1 TW of biomass (that is the maximum supply
compatible with a prudent use of the mineral reserves) would be more than sufficient to supply
similar services than the present economy, except for the petrochemical sector. On the other hand,
economic growth of developing countries is expected to increase the world energy consumption in
the next decades, which has been estimated in 460 to 520 EJ/year in 2030 [71]. This range is equivalent
to 14.6–16.5 TW, or 12.7–14.4 TW if the services were supplied for a post-carbon economy. If we
accept that a sustainable level of final energy supply is not much above 12 TW of electricity and new
feedstocks and 1 TW of traditional biomass, the conclusion is that in 2030 the economy may be on the
limit of what is sustainable.
On the other hand, world's population continues increasing and, in a future globalized society,
industrial economy will have to grow accordingly to world's population in order not to get a per
capita decline. The medium variant of World Population Prospects of the UN [72] estimates that
world population will increase from 6520 million in 2005 to 8500 million in 2030, 9725 million in
2050 and 11200 million in 2100. In order to maintain the same amount of energy per capita than in
2005 to this higher number of people, energy production would have to be 13.4 TW, 15.4 TW and
17.7 TW in 2030, 2050 and 2100, respectively or, in RE equivalents, 11.7 TW, 13.4 TW and 15.4 TW.
Thus, even maintaining a constant energy use per person, expected population growth may take us
to unsustainable levels of energy production after 2050. Under present growth expectations, if some
technological alternatives to copper do not appear on time, RE might probably not be enough to
maintain an industrial society as we presently understand it in the long term, and all these over-shoot
scenarios mentioned above would imply a forced de-growth in a future post-carbon economy.
Over the scale of several decades to one century we cannot discard technological innovations
that enable replacement of copper and lead to a possibly larger energy supply, and even unexpected
techno-explosions. However, even in that case, we cannot know if such technological surprises
will permit sustained exponential energy growth, linear growth or only logistic growth in the long
term. Therefore, in a post-carbon economy we should be prepared to work under all the range of
possibilities, from stationary energy supply to exponential growth. Such an economy would adapt
its growth to the availability of new energy sources and feedstocks, and not the opposite, and in this
context any improvement in economic efficiency would allow the energy available (e.g., 13 TW) to
satisfy a larger service demand. This will probably require structural transformations of capitalism
13322
Energies 2015,8, 13308–13343
such as we know it [73]; it is high time to stop considering any change in capitalism as the end of
the world.
Provided that future economy is able to adapt to a stationary energy supply and that population
can be limited to less than 9700 million, the above analysis shows that the main economic processes
can, in principle, be replaced by sustainable alternatives based on electricity, charcoal, biogas and
hydrogen. And contrarily to the most pessimistic expectations, those services that cannot be replaced
are not as crucial as to cause a return to a pre-industrial society.
RE are physical energies and in many economic sectors, such as transport, people demand
essentially physical services as well. Conversion of RE into chemical energy and then back again
into physical energy wastes an important fraction of the initial RE produced, and should be avoided
through direct use of electricity in transport [31]. This implies avoiding the use of hydrogen in
transport except where connection to the grid or battery use are impossible, e.g., in aviation, shipping,
and specialist (emergency) vehicles.
Even so, if land transport at present levels were based on a fleet of electric cars, trucks and
motorcycles, and 50% of batteries were made of lithium (Li) and 50% of nickel (Ni), 29% and 40% of
the present reserves of Li and Ni would be used, respectively, as well as 22% of platinum-palladium
reserves. The limited reserves of these metals constraints the number of trucks that could be
powered by fuel cells to a fraction not much above 10%. Given that reserves cannot be indefinitely
expanded [2], the viability of the present level of land freight transport may depend on future
development of high capacity batteries and higher use of rail freight. Electrification of the entire
global fleet is a feasible option, but the number of vehicles that a future post-carbon society may
sustain is roughly the number we have currently. A larger number would endanger the availability
of Li and Ni for other economic sectors. Given that demand for vehicles will probably increase in
the future due to demographic and economic growth, especially in developing countries, the present
land transport might not be fully sustained. However, a reorganization with a larger emphasis on rail
transport, would alleviate the increased demand on these metals and thereby probably maintaining
sufficient mobility for people and freight.
The energy demand of transport in a future post-carbon economy will be very similar to the
present one. On one side, aviation and shipping will notably increase energy use due to the
necessity to produce electrolytic hydrogen to be consumed in jet engines and fuel cells, respectively.
But this enhanced demand will be compensated for savings coming from road transport, since electric
motors have greater efficiency than combustion ones. Actually, future aviation would have energy
consumption larger than trains and close to that of road transport, if present transport structure
remained in a post-carbon economy. Given that aviation is less energy efficient than vehicles and
trains for the transport of people and goods, the use of jet aircraft may diminish in importance in the
future. Thus, present levels of air transport may not be fully sustained. A similar mechanism could
enhance the use of rail transport in a substitution of cars and trucks.
Fishing and mercantile activities frequently require long periods of navigation, which make
battery-based motors inappropriate. Fuel cells may be the best solution in these circumstances. They
are a relatively inefficient way to use electricity, however the size of the fishing and merchant fleets is
small in comparison with that of land vehicles, and they would not imply an insurmountable problem
in terms of energy demand.
Open field work in farming, mining and construction sometimes requires high power tractors
that should also be supplied by fuel cell vehicles, while other generic farm work could be done using
many smaller electric tractors which would recharge their batteries in the grid. Thus, full connection
of farms to the electric grid would become necessary in the future economy. For similar reasons, any
project involving open field construction will have to plan for the building of a connection to the grid.
This reorganization of open field work does not necessarily create an insurmountable problem.
Mining in a post-carbon economy seems sustainable in the short and medium term, but is
unsustainable in the long term due to the decline of ore grades [74]. Thus, a fully sustainable
13323
Energies 2015,8, 13308–13343
post-carbon economy should increasingly base its mineral production on recycling. The form that
this transition may take is a crucial subject that would require discussion in a separate paper.
The future industrial sector will initially demand more energy than the present one to deliver
the same services (see Table 5), due to the increased use of charcoal, biogas and hydrogen. However,
if we subtract the energy embodied in future feedstocks (see Table 6), we find that the electricity
needed for industry is 2751–2908 GW, which is smaller than the 2005 industrial energy demand
(3403 GW). A post-carbon industry would use similar or less energy than the present industry in
all the sectors except iron and steel and non-ferrous metals. The energy savings would be especially
high in nonmetallic minerals production due to the severe heat loss associated to flue gases in high
temperature furnaces, which could be minimized by using electric heating.
To make these estimations, the efficiency of electricity for machinery, vehicles and heating has
been assumed to be equal as reported in current systems. When Sankey (energy flow) diagrams
are available and fuels are the main energy input for heating in an industry, the present efficiency
of industrial heating has been calculated from these Sankey diagrams. When this method is not
applicable, the reported efficiency of current furnaces is used. A similar procedure has been used to
estimate the efficiency of steam production in different industries. Actually, the different industrial
and economic sectors are complex and the precise form and efficiency that their processes will have
in a post-carbon economy remain uncertain. For these reasons, the energy demand calculations made
here cannot be considered as precise projections but as indicative estimates that should be improved
through continuous research.
To compensate for the withdrawal of coal, oil and gas, charcoal and biogas will have to be
produced in larger quantities than currently. Most of this demand will originate from the production
of High Value Chemicals (HVC) required by the petrochemical industry.
To achieve economic activity similar to that of 2005, 240–300 million tonnes of charcoal per
year should be produced from woody biomass. This will require governments to encourage people
to abandon the use of wood for heating and cooking purposes, at least in developed countries,
where electric alternatives will exist. 215 million tonnes of biogas will also be required for ammonia
production, at least until organic agriculture is fully developed. This quantity can be obtained
through the current production by developed countries and China alone, given that many developing
countries may need their biogas for rural self-consumption.
Use of direct iron reduction with hydrogen in steel production will alleviate an important
fraction of the future carbon (and therefore charcoal) demand (Table 6). Even so, the estimated
potential of renewable charcoal production is insufficient to maintain the services that the world
economy produced in 2005. If 100% of the biogas renewable potential were used to produce 100%
of ammonia and 32% of the HVCs consumed in 2005, and all the available charcoal is used for the
coal to olefins process (after setting aside the demands of the metal sectors), 130 to 141 ˆ106t of
HVCs could be produced. This amounts to 45%–49% of HVC production of 2004.
Olefin production was 220 ˆ106t in 2012 [75], 10.9% higher than in 2004 ([51], Table 8.5). If
HVC production had the same relative increase, 318 Mt were produced in 2012. If the calculations
of Table 6are made with this value, only 40%–43% of HVCs produced in 2012 can be produced in
a post-carbon economy. This range could be increased by a few percent if ammonia were produced
from hydrogen (Appendix, Equation (A6)) and not from biogas. In any case, the present growth in
HVC demand takes us progressively away from what is sustainable from a post-carbon perspective.
Given the limitations of the renewable production of natural gas and charcoal, systematic
utilization of these products as fuel should be legally discouraged in the future, since electricity is
a more efficient energy supply for end uses, and these products are critical for supplying feedstocks
to the petrochemical industry.
The petrochemical industry cannot be fully maintained at its present size in a post-carbon
economy and, to a first approach, should shrink to a size of 40%–43% of that of the 2012 petrochemical
sector, which is the size that the sector had in 1985–1986 [76]. Fortunately, many of the products
13324
Energies 2015,8, 13308–13343
currently provided by the petrochemical industry can be replaced with similar products based on
natural feedstocks although their supply rate must adapt to their natural availability. Substitution
of natural feedstocks such as those mentioned in Table A2 for HVC consumption would allow an
increase of the size of the petrochemical sector above the mentioned 40%–43% factor. In particular,
there is a global potential of non-woody biomass that is not fully utilized. Starch and non-woody
biomass are rich in xylenes and glucose, sugars required for the production of polylactic acid (PLA).
Thus, production of polystyrene from PLA would alleviate the demand for charcoal, saving a fraction
of it that could be used for olefin production.
Polyurethanes, polyesters, sealants, many pharmaceutical drugs and some dyes cannot be
replaced by natural substitutes and should be produced from HVC and, ultimately, from charcoal.
However, given the insufficient availability of renewable charcoal, these products would be produced
at a lower rate than at present; up to 2.5 times lower, making them scarce in a future post-carbon
society. The services that these products offer include cushioning foams, foams that do not melt
when heated, and some synthetic fibers with special qualities.
More critical can be some pharmaceutical drugs whose lack would affect quality of life. Scarcity
of these drugs may worsen the treatment of some specific health conditions, although it will probably
not affect life expectancy. Indeed, the main factor influencing life expectancy is the incidence of
infectious diseases, which are mainly related to poor basic hygiene and poverty [77] and not to
specific drugs.
While electricity, charcoal and biogas may be the base of a sustainable transport and industry,
they will be necessary but not sufficient for future sustainable farming. Agriculture is not sustainable
at present given its dependence on fossil fuels and minerals such as phosphorous and potassium.
In a post-carbon economy, organic farming may be the only sustainable solution able to (almost)
fully recycle these essential nutrients. A detailed discussion of future sustainable farming would
require another paper. However, sustainable farming will be one of the key elements of a future
post-carbon society.
Acknowledgments:
This work has been partially supported through the project “Vade Retro” (CTM2014-56987-P)
of the Research and Development Spanish program. I thank two anonymous reviewers for their detailed
comments which helped improve the article.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix. Energy Demand of Industry in a Post-Carbon Economy
A1. Transportation Equipment and Machinery
We have taken the energy share of liquid oils, gas, coal, renewables, wastes, electricity and heat
consumed by the transportation equipment sector from Table 4. Given the absence of global statistics
we have taken the energy and carbon analysis of the US-DOE for this sector as representative of
the efficiency of use of fuels and electricity by this industry in the world [52]. Finally, we have
calculated the energy cost of a possible substitution of fuel inputs with electricity. The expression
used is the following:
pr“pfpf1esp{eer `f2ecb{eer `fo3 ein{eel `fo5 ein {eba `fg4 0.5 pecb{eer `ecb{ehp q ` fh`feq(A1)
where pris the mean annual power demanded by a post-carbon transportation equipment sector;
pfis the demand of the sector in 2005; f1, f2, are, respectively, the fractions of fuel inputs to steam
generation and process heating, relative to the total energy input to the sector; fo3, is the fraction
of fuel (assumed to be oil derivative) used for machine drive, fo5, is the fraction of fuel (assumed
to be oil derivative) used for onsite transportation; fg4 is the fraction of fuel (assumed to be natural
gas) used for facility HVAC (heating, ventilation and air conditioning); fhand feare the fractions
of heat and electricity consumed by the sector, respectively; and esp, ein, eel, eba , ecb, ehp, ecb , are
13325
Energies 2015,8, 13308–13343
the efficiencies of steam production, internal combustion engine, motor connected to the electric
grid, battery motors, condensing boilers, heat pump, and industry heating process (conventional
boiler), respectively. For onsite transportation we assume a future substitution of electric vehicles
for internal combustion vehicles, with an efficiency ein intermediate between those of gasoline and
diesel vehicles (0.30) (Table 3). The efficiency of electric resistance for heating and steam production,
er, is assumed to be 0.97 (see Table 3and [78]), the efficiency of conventional heating in the industrial
process, ecb, has been assumed to be 0.83, that is a typical value for a conventional gas boiler (see [79],
p. 18), given that the use of condensing boilers in industry is not yet extended. Steam production
efficiency, esp (0.81), as well as the fractions f1, f2, fo3, fg4, fo5, fhand fewere estimated from the Sankey
diagrams of the US industry ([52], “Transportation equipment”). We assume that heat imported from
other industries will have the same efficiency of production than at present, which is probably a
conservative assumption. Under these assumptions, the present consumption of the sector, 45.2 GW,
would become 39.2 GW (Table 5).
Similar expressions have been used to estimate the power required in other post-carbon
industrial sectors. A similar analysis can be made for the machinery sector, by using Table 4and [80].
The present consumption by that sector, 128.6 GW, would become 110.7 GW (Table 5).
A2. Wood Products, Mining, Construction
The timber sector consumed 42 GW in 2005. Fractions of fossil fuels, biomass and electricity
used by this sector are taken from Table 4. An expression similar to Equation (A1) is used to make
the energy demand estimation. Coal, biomass and gas are assumed to be used for onsite generation
of steam, which will be replaced by electric resistance; oil for onsite transportation, which will be
replaced by electric locomotion; gas for process heating and HVAC, which will be replaced with
electric resistance and heat pumps, respectively. HVAC and gas heating are assumed to be 83%
efficient, and the efficiency of onsite steam generation (72%) is estimated from [81]). Assuming an
efficiency of 0.3 for the present conversion of oil energy to mechanical work (Table 3), and a future
substitution of electric locomotion for combustion locomotion, we obtain a demand of 36 GW for a
post-carbon economy.
In a post-carbon economy the mining and quarrying sectors would decrease or eliminate their
activity related to energy producing materials (fossil fuels) which currently contribute most of
their added value. It is difficult to estimate the mean energy consumption of energy producing
mining because of its great variability. However, the consumption of goods and services related
to energy-producing mining is 79% of the one made by the entire mining and quarrying sector ([82],
Table 2.4). Assuming that this fraction is a good estimate of the fraction of energy consumed by
energy producing mining, all this fraction of energy would be saved in a future mining and quarrying
sector, which would comprise only 0.4% of industry’s energy usage (14 GW) instead of the present
2% (72 GW).
The fractions of oil, coal, gas, electricity and biomass consumed by the construction sector in
2005 are taken from Table 4. No Sankey diagram is available from the US DOE for this sector, and the
following expression is used for the calculation of future energy demand:
pr“pfppfc`fbqesp{eer `foein{eba `fg0.5 pecb{eer `ecb {ehpq ` fh`feq(A2)
where pris the mean annual power demanded by a post-carbon construction sector; pfis the demand
of the sector in 2005; fc, fb, fo, fg, fhand feare the fractions of coal, biomass, oil, gas, heat and electricity
consumed by the sector, respectively; and esp, ein, eba, ehp, ecb, are the efficiencies of steam production,
internal combustion engine, battery motors, heat pump, and industry heating process (conventional
boiler), respectively. Given the lack of information on the distribution of the different fuels between
the main industry processes, the expression assumes that all the coal and biomass energy is used
for steam production and conventional boilers (83% efficient), the totality of oil is used in internal
13326
Energies 2015,8, 13308–13343
combustion engines, and the totality of gas is used in conventional boilers and process heating (83%
efficient) and will be replaced with 50% of electric resistance power and 50% of power based on heat
pumps.
Under these assumptions, the energy consumption of 2005 (46.9 GW) would become 27.1 GW.
A similar expression to Equation (A2) is used for other sectors where no Sankey diagram is available.
A3. Textiles, Paper Pulp, Food and Tobacco
The textile chain consists of the production or harvest of man-made and natural fibres;
conversion to yarns; weaving, knitting, or other mechanical processes; printing, coating, washing and
drying to produce end products such as clothes, carpets, knitwear, etc. [83]. The energy consumed
in the chain is mainly related to mechanical processes done with electric engines. The electricity
consumption rate in the total consumed energy for individual textile production stages is 93% for
spinning, 85% for weaving, 43% for wet processing, and 65% for clothing manufacture [84]. The main
constraints that this industry may suffer in a post-carbon society relates to shortage of feedstocks
that are based on oil, such as oil-derived polymers, oil-derived organic detergents and oil-derived
dyes. However, as will be commented on in Appendix A8, many of these petrochemical products
have substitutes.
Thus, energy use for textile and leather production can, in principle, be completely electrified.
The expression used to estimate it is similar to Equation (A1).
Steam production efficiency, esp (0.81), as well as the fractions f1, f2, fo3, fg4, fo5, fhand fewere
estimated from the Sankey diagrams of the US industry ([52], “textiles”). Under these assumptions,
the 72 GW consumed in 2005 in a future post-carbon economy would become 64 GW.
Paper pulp is made by grinding and cooking wood chips in an aqueous solution of chemicals
(mainly lime and chlorine dioxide) and by recycling fibers. Paper manufacture involves pressing and
heating the pulp, drying and coating, and packaging the sheets. None of these processes are required
to use fossil fuels. No Sankey diagram is available from the US DOE for this sector, and an expression
similar to Equation (A2) is used for the calculation of future energy demand.
Under these assumptions and using Table 4, the 216 GW consumed in 2005 would become
173 GW in a future post-carbon economy. A similar analysis can be made for the food and tobacco
processing, which involve essentially human work and mechanical processes that can be electrified.
From 190 GW consumed in 2005, a future post-carbon economy would require 163 GW.
A4. Iron and Steel
Iron and steel is produced using basic oxygen coal blast furnaces (70%) and recycling of scrap in
electric arc furnaces (29%) [85]. In the first process, 0.6 t of coke are needed to produce 1 t of steel; in
the second process no coal is needed. Thus, the effective coke input is 0.42 t of coke per tonne of steel
produced. The production of raw steel was 1 ˆ109t in 2005 ([36], steel), which required 420 million
tonnes of coke. A proven alternative is direct reduction from hydrogen.
The production of iron was 7.47 ˆ108t/year in 2005, and using an efficiency of hydrogen
electrolysis of 65% [30], we find that about 275 GW power would be consumed in the hydrogen
synthesis needed for iron production. On the other hand, direct reduction processes use about 11 GJ
per metric ton of iron produced ([86], Section 10.1). Thus, about 261 GW would be needed for global
iron reduction.
The sponge iron produced in this way can be used to produce crucible steel by diffusion
of charcoal, a process that is known in the industry as “carburization” [56], leading to typical
concentrations of around 0.5% carbon in steel. Thus, to produce 1.65 ˆ109t/year of steel (the
production of steel in 2014) [36] we would need 8.3 ˆ106t/year of carbon, a supply that would
consume 23% of the present charcoal production. The 0.4 GJ/t of energy needed for carburization is
that required to maintain a constant melting furnace temperature [87], and it would amount to 13.8
GW to cover the production of steel in 2005. In total, 550 GW would be needed for iron production and
13327
Energies 2015,8, 13308–13343
subsequent carburization to produce steel. The figure would probably be slightly larger if pumping
and other small electrical consumptions were considered in the estimation. In the US “iron and steel”
sector of 2010 ([88], “Iron and Steel”), the fraction of energy used in non-process (HVAC, lighting,
facility support, onsite transportation and other non-processes), process cooling and refrigeration,
machine driving and other process uses was 31% in relation to the main process energy use (process
heating and electro-chemical). If we assume the same fraction to be approximately valid for a direct
reduction completely-electrified steel production, 31% of 550 GW should be added to our estimate.
The final result is 720 GW required for iron and steel production in a post-carbon economy, 51% more
energy power than at the present.
A5. Non-Ferrous Metals
The non-ferrous metals are mainly copper, aluminum, lead and tin, zinc and cadmium, precious
metals, ferro-alloys, nickel and cobalt and carbon and graphite electrodes. A large proportion
of energy consumption in metal production is associated with milling operations, responsible for
approximately 40% albeit with great variations, followed by dewatering, ventilation and transfer of
materials after grinding. However, these processes are essentially mechanical and can be electrified.
Smelting and refining consumes about 50% of the energy, with large variations depending on the
metal. In this process, metal reduction is a necessary step that frequently uses fossil coal. We will
analyze if that coal can be replaced by other materials or be obtained renewably.
Generating primary aluminum is extremely energy intensive, however, essentially it involves
calcination of high alumina containing bauxite and electrolysis of alumina ([74], pp. 184–185). The
latter process is electricity based and the former can be done with an electric furnace. Thus, future
aluminum production is expected to use a similar amount of power as that used at present.
Copper production involves chalcopyrite grinding, floatation, filtering, roasting, smelting and
electro-refining ([74], pp. 187–188). Roasting and smelting consist of strongly heating chalcopyrite
with silicon dioxide and oxygen-enriched air in a furnace. Part of the process is auto-thermal due to
the high exergy content of chalcopyrite, and does not necessarily require fossil fuels. The other steps
can also be electricity-based. Methane is frequently used as a reducing agent, through the reaction:
4CuOpsq ` CH4pgq Ñ 4Cupsq ` 2H2Oplq ` CO2pgq(A3)
However a much more efficient reducing agent is hydrogen [89] through the reaction:
CuOpsq ` H2pgq Ñ Cupsq ` H2Oplq(A4)
which is also environmentally clean. In a post-carbon economy hydrogen reduction would be a good
choice for production of primary copper given the relative scarcity of methane (see Section 7), and the
probable decreasing costs of a future hydrogen economy. Production of copper was 14.9 ˆ106t/year
in 2005 ([36], copper). If hydrogen were used, copper reduction would require 0.6 ˆ106t/year of
hydrogen for reaction Equation (A4). Taking into account the efficiency of hydrogen electrolysis [30],
about 3.2 GW would be consumed using the electrolytic hydrogen production process. This energy
expenditure is almost two times the energy content of the stoichiometric methane required in
Equation (A3), which amounts to 1.7 GW for the 2005 copper production. All things being equal,
changing from Equations (A3) to Equation (A4) adds 1.6 GW to energy demand, as explicitly included
in Table 5.
Hydrogen requires cooling, compression and confining, expenditure that is hard to quantify but
also avoid energy costs that are related to process Equation (A3), such as carbon capture energy for
avoiding CO2emissions. Hereafter, we will assume that, aside from the energy differences discussed
above, other energy costs related to hydrogen reduction of non-ferrous metals are similar to those of
conventional coke or methane reduction.
13328
Energies 2015,8, 13308–13343
Extraction of tin involves concentration from cassiterite (SnO2) mixed with other minerals, fusion
reduction, and electric furnace treatment. Fusion reduction involves the reduction of SnO2using
carbon as the reducing agent:
SnO2`CÑSn `CO2
This process normally uses coal. Given that a fraction of carbon is directly oxidized, about 2 t
of CO2is generated to produce 1 t of tin [90]. To produce 296,000 t/year, which was the world tin
production in 2014 ([36], “Tin”), we need 161,455 t/year of carbon.
An alternative is direct reduction with hydrogen:
SnO2`2H2ÑSn `2H2O
which is favorable at temperatures between 823 and 1023 K [91]. The power needed to electrolytically
produce the hydrogen would be 0.03 GW.
Nickel production involves mining, floatation, drying up, and flash furnace heating. Even
though coke is normally used as a reducing agent, the reduction can be done with hydrogen, for
similar reasons to those discussed above for iron, through the following reaction:
NiO `H2ÑNi `H2O
The 2005 production of Ni was 1.5 ˆ106t/year [36]. The hydrogen necessary to reduce
that nickel amounts to 51,107 t, which could be electrolytically produced by using 0.35 GW in a
post-carbon economy. Given that 0.165 GW of carbon can be saved, the net cost increase for using
hydrogen reduction will be 0.18 GW, demand that has been explicitly included in Table 5.
Precious metals (silver, gold and platinum) can be obtained as byproducts from the processing
of anode slimes from copper production, leach residues and crude metal from zinc and lead
production [74]. Their quantities are marginal, therefore their reduction reactions do not require any
significant mass of charcoal in comparison with the estimates made above.
Lead production involves mining of galena (PbS), concentration via crushing, grinding,
floatation and sintering, extraction of metal via smelting, and refining. During the smelting process
(in a vertical blast furnace) the galena is roasted (reaction with O2) to remove the sulfur [91]:
2PbS `3O2Ñ2PbO `2SO2
The formed lead oxide is reduced by coke to metallic form:
2PbO `CÑ2Pb `CO2
With some help of the parallel reaction:
2PbO `PbS Ñ3Pb `SO2
The coke rate for reduction in lead primary production is typically 130 kg of coke per tonne of
lead produced [92].
Secondary production of lead from recycled scrap amounts to 50% of world production [93].
An Ausmelt/ISASMELT furnace uses 5000 t of coke and coal in order to produce 125,000 t of
secondary lead bullion, and a QSL plant uses 15,000 t of coal to produce 135,000 t of lead bullion ([94],
Tables 5.5 and 5.6). Assuming that 50% of present systems for secondary production use the former
system, while the other 50% use the latter with its associated coal and coke efficiencies, we estimate
that 591,500 t of coke-equivalent are needed to obtain the current production of lead (5.46 ˆ106t/year
according to [36], “lead”). Using the relative contents of carbon in charcoal (0.75) and coke (0.90), this
is equivalent to 709,800 t of charcoal.
13329
Energies 2015,8, 13308–13343
This reducing coal is essential for the process and cannot be replaced with electricity. The others
steps in the chain of production could, in principle, be electrified.
Early blast furnaces used charcoal as fuel because coke was not well known and not available
in sufficient quantities [95], and carbon as a reduction agent could be supplied again from renewable
charcoal in a future post-carbon economy. Another feasible alternative would be hydrogen reduction,
in a process similar to the one commented above for iron.
Zinc production involves mining of sphalerite (ZnS), concentration via crushing, grinding,
floatation, sintering, roasting, leaching and electro-winning. Again, the reduction of the sphalerite
with coal is the step that cannot be replaced with electricity. The smelting itself can be done with an
electric arc [96], or by using the heat given off by the combustion of a fraction of the carbon supplied.
The reduction reaction is:
2ZnO `CÑ2Zn `CO2
To obtain the current production of zinc (13.3 ˆ106t/year according to [36], “zinc”) we would
need 1.2 ˆ106t/year of carbon which, in principle, could be supplied from renewable charcoal
production (see Section 7).
Regarding ferro-alloys, ferro-chrome, along with ferro-nickel, are the major alloys in
the production of stainless steel. Silicon metal, ferro-silicon, ferro-manganese, ferro-nickel
and silicon-manganese are used as additives and alloys in important industrial products.
Ferro-vanadium, ferro-molybdenum, ferro-tungsten, ferro-titanium, ferro-boron, ferro-niobium, and
other alloys are also produced in small quantities.
Ferro-niobium reduction occurs as an alumino-thermic process, ferro-molybdenum reduction
most commonly uses a silico-thermic process and ferro-titanium is usually reduced through a
metallothermic process. Thus their production does not involve coal as reducing agent [94]. Table A1
shows the annual production of the main ferro-alloys in 2005, their consumption of carbon per ton
produced, and the carbon needed per year for their production.
Table A1. Annual production of the main ferro-alloys in 2005, their consumption of carbon per ton
produced, and the carbon needed per year for their production. Sources: [94], Tables 8.5 to 8.10; [36],
year 2005.
Ferro-Alloy Production (Mt/year) Reducing Carbon (kg/t) Carbon (Mt/year)
Ferro-chrome 6.6 550 3.6
Ferro-silicon 5.4 1150 6.2
Silicon metal 0.7 1300 0.9
Ferro-manganese 4.6 500 2.3
Silicon-manganese 6.9 550 3.8
Ferro-nickel 1.1 4000 4.4
Others (Fe-Bo,Fe-Ti,Fe-Va . . . ) 2.3 1.4
Total ferro-alloys 27.6 22.6
We have assumed that half the alloys grouped in the “others” category will need carbon as
reducing agent at a rate of 600 kg/t. The total flow of carbon required to produce the components
listed in Table A1 is then almost 23 ˆ106t/year, a substantial part of the annual charcoal currently
produced. Given that some furnaces use a fraction of carbon for heating, perhaps 10% of this figure
could be saved if electric arc furnaces were the only technology used. An alternative source of
reducing agent for ferro-alloys could be methane obtained from biogas.
Carbon electrodes are usually manufactured from graphite, however fabrication of a carbon
electrode is possible by using activated carbon [97], which can easily be obtained from charcoal.
Graphite electrodes are used for melting scrap iron and steel (and sometimes directly-reduced iron)
in electric arc furnaces, which comprise the vast majority of furnaces for recycling steel scrap. World
13330
Energies 2015,8, 13308–13343
production of synthetic primary graphite was 1.5 Mt/year in 2011 after growing 5% per year since
2001 [98], therefore synthetic primary graphite production was about 1 Mt/year in 2005.
In a post-carbon economy, graphite could be produced following these steps: wood Ñcharcoal
Ñcoke Ñbaking Ñgraphitizing. Charcoal production does not require energy inputs except for
transportation and milling. Charcoal to coke process (“coking”) is exothermic due to the production
of coke oven gas [99]. We will assume 1.33 t of charcoal per metric ton of graphite produced if charcoal
has 75% carbon content [67]. Energy input for baking is up to 11 GJ/t and graphitizing requires
9–20 GJ/t [94].
In a post-carbon economy, the 2005 level of synthetic primary graphite production could be
maintained by using 3.6 ˆ106t/year of biomass, which is 0.3% of the sustainable biomass potential
(Section 7). The energy consumed in that process would be about 1 GW, as explicitly included in
Table 5.
A6. Non-Metallic Minerals
The industry for non-metallic mineral products is made up of the cement, ceramics, glass
and lime sectors. These are all traditional, well-established manufacturing sectors characterized by
the transformation of naturally occurring minerals such as limestone, silica, and clays, through an
energy-intensive process [100]. World production of cement was 2540 million tonnes in 2006 [101].
Energy use for cement production is about 3.3 GJ/t in India, which has the most efficient cement
industry in the world [51].
The main constituents of the raw materials required for cement production are calcium oxide
(CaO, coming from lime), silicon dioxide (SiO2), aluminum oxide (Al2O3) and iron oxide (Fe2O3).
The process consists of the crushing of raw minerals, calcination, clinkering and final milling [101].
Calcination can be forced by electric heating, and clinkering is a complex set of chemical reactions at
1400–1500 ˝C where belite ((CaO)2SiO2) is formed from the raw materials and alite ((CaO)3SiO2) is
formed from belite and calcium oxide through the following reaction:
pCaOq2SiO2`CaO Ø pCaOq3SiO2
Fossil fuels are normally used to heat the furnace, however clinkering is also possible to do
in an electric furnace provided that the lime is melted down from the surface of the bath before it
comes into contact with the carbon electrode of the arc furnace, so avoiding the production of calcium
carbide [102]. Crushing, milling and calcination can be also done by electrical means, hence cement
production can, in principle, be electrified. The same is true for lime production, which involves
similar steps as cement but without the clinkering process.
It is difficult to know what efficiency will have future electric industrial furnaces because
presently the practical totality of furnaces is partial or totally fuel-fired. However, most of the heat loss
of an industrial furnace comes from the warm flue gases released by chimneys. In a gas based furnace
with a typical oxygen excess of 5%, 36% of heat is lost for furnace temperatures of 800 ˝C (80% for
temperatures of 1600 ˝C) [103]. Flue gases are therefore under the low efficiency of high-temperature
furnaces such as cement kilns.
A future 100% electric furnace could highly improve efficiency by avoiding the release of hot
gases. Internal recirculation of hot gas, which is useful to homogenize temperature, can be made
with electric fans which make possible a minimum gas release outside the furnace. Home electric
furnaces use presently that principle and are able to reach high annual fuel utilization efficiencies,
from 95% to almost 100% [104]. The main difference of these furnaces and future industrial electric
furnaces will be the larger size of the latter. But a larger size increases the volume to surface ratio of
the furnace, which decreases skin losses and tends to increase its efficiency. Thus, in our calculations
we have used an efficiency of 97% for future electric industrial furnaces, a figure that is equal to that
reported for air heating with electrical resistance (Table 3).
13331
Energies 2015,8, 13308–13343
In ceramics production, raw materials are mixed and cast, pressed or extruded into shape. Water
is used for a thorough mixing and shaping. This water is evaporated in dryers and the products are
either placed by hand in the kiln or placed onto carriages that are transferred through continuously
operated kilns. In most cases, the kilns are heated using natural gas, but liquefied petroleum gas, fuel
oil, coal, petroleum coke, biogas/biomass or electricity are also used [105]. All these steps could be
done with electricity, in principle.
In glass production, silica sand, process cullet, or post-consumer cullet with a set of intermediate
and modifying materials (such as soda ash (Na2CO3), dolomite (CaCO3.MgCO3) and other inorganic
salts) and colouring/decolouring agents (such as iron oxide (Fe2O3), carbon or pyrite) are mixed with
a fluxing agent, normally sodium oxide, heated in a furnace to 1350–1500 ˝C and melted. Essentially,
the silica from the sand combines with the sodium oxide and with other batch materials to form
silicates [106]. Heating in an electric furnace is one of the techniques commonly used in this industry,
which could be completely electrified, in principle.
The energy consumed in the production of non-metallic mineral products was 350 GW in 2005
(Table 5), of which 51.9% was for coal, 13.8% petroleum products, 19.1% natural gas, 1.9% biomass,
12.2% electricity and 1% heat (Table 4). We will assume that 0.4% of power is for heating, ventilation
and air conditioning (HVAC) [107] by gas and will be converted by heat pumps, that 1% is for onsite
transportation using oil and being converted to electrical transport, that electricity uses (machine
power, lighting and HVAC) stays the same, and that the remaining fuels are used for clinkering
and other energy processes, which will be electrified. The efficiency of coal, gas and oil heating in