New or revised pages:
(latest at top)
20 November 2009
Site history
|
|
Much has been published in the popular press,
in recent years, about hydrogen being the fuel of the future. Nothing is
less certain.
In reality, hydrogen can never be considered as
a fuel, in the same way as oil or natural gas. It is a means of storing
energy because it has to be manufactured using an energy-intensive process.
It is physically impossible to obtain more energy out of the manufactured
hydrogen than was put in to manufacture it. In reality, the holistic EROEI
(Energy Returned Over Energy Invested) is typically less than 0.8. It is
therefore evident that hydrogen should not be used when the original energy
invested could be used to better advantage. For example, there are two
principal ways of driving a car from an electricity supply: either one can
generate hydrogen and use a fuel cell to regenerate electric power to drive
a motor or one can use a rechargeable battery to drive an identical motor.
The former will have an electrical efficiency (kWh in to kWh out) of,
typically 45%: the latter will have an efficiency of better than 80%. So,
a priori, there is a better case for using rechargeable batteries. This
simplistic statement does, however, have many provisos, such as autonomy,
weight, lifetime and volume issues which could modify the equation.
The notion of using hydrogen is based on the
premiss that, when burnt or used in a fuel cell, the exhaust gases consist
only of water vapour. Even this is true only for fuel cells. If used as a
combustible gas, then polluting nitrogen oxides are also produced in small
quantities. However, one has to have a more holistic approach.
Proposals have been made to use hydrogen for:
-
mobile fuel cells in cars
-
static fuel cells for electricity
production (as a "fill-in" for when variable renewable generation is not
producing sufficient electricity)
-
mixed with natural gas as a fuel to reduce
its "greenhouse gas" potential.
This idea presupposes an important new
infrastructure for hydrogen production and distribution. Can such an
infrastructure be implemented in Cyprus?
There are two major ways of producing
hydrogen.
The cheapest, easiest and most common way is
chemically from hydrocarbons. Let us assume, for simplicity, that our
starting point is natural gas, methane, CH4. This is done in two
phases. The first is reacting it with water vapour over a catalyst:
CH4 + H2O > CO + 3H2
The second phase is to react the highly toxic
carbon monoxide with more water:
CO + H2O > CO2 + H2
Thus, we get four molecules of hydrogen out of
each molecule of methane. The problem lies in that we also get just as much
carbon dioxide, the main "greenhouse gas", as if we simply burnt the
methane, in the first place, but without the benefit of the thermal energy
from burning the carbon. Rather the opposite, because the reaction takes
place at several hundred degree Celsius and the methane needs to be heated.
Analogous reactions can take place with nearly every hydrocarbon, even coal,
but most other hydrocarbons have a much higher carbon:hydrogen ratio, so
would produce disproportionally more carbon dioxide than hydrogen.
This is produced by passing a low voltage DC
electric current through slightly acidified fresh water. Molecular hydrogen
bubbles off the cathode, while half the volume of molecular oxygen bubbles
off the anode.
2H2O > 2H2 + O2
This is a very exact physical phenomenon and a
current of 1 ampere flowing through the water for 1 second will produce
precisely 0.00001045 grams of hydrogen. The voltage required to do this
depends on the conductivity of the water and the configuration of the
electrodes and would normally be between 2 and 4 volts. To take a
hypothetical example of hydrogen required to fuel a car, a tankful would
typically be, say, 30 kg of highly compressed hydrogen to give the car an
autonomy of 500 km. This would require 2.87 billion A.s to produce. At about
3 V, this would require 2.4 MWh. Let's imagine that, one day, there will be
200,000 hydrogen-burning (e.g., fuel cell) cars in Cyprus, each averaging 40
km/day, it would require an extra power station of 1.6 GW to provide the
required amount of hydrogen; this is almost twice the current peak
electrical capacity for all industrial and domestic requirements. By
coincidence, 1.6 GW is the output power of a standard European pressurised
water reactor nuclear power station.
The overall efficiency of high-pressure
catalytic electrolysers is about 75%, on condition they work 24/7, but it
drops very considerably if they are not used continuously at full output.
This makes them unsuitable to be run uniquely from variable renewable
sources. Combined with the efficiency of the generating plant and that of
the fuel cell (currently about 45-50%) cars, their overall energetic
efficiency from the primary energy source to the cars' wheels is not likely
to be better than with ordinary internal combustion engines.
Furthermore, to produce this quantity of
hydrogen, it would require an annual consumption of over 1.5 million tonnes
of fresh water - in a country already strapped for sufficient water in
periods of low rainfall. By extra desalination in a small plant, say, with a
capacity of about 1/8 that of the existing Dhekelia plant, this quantity
would be feasible. However, the power required to desalinate this volume of
water would be considerable and would add to the energetic and economic
costs of the produced hydrogen.
Where would the energy for all this come from?
We can see in the essay on
Renewables that such a requirement would be impossible to supply from
wind or solar panels in the context of this island. Yet many persons are
banking on the advent of the fuel-cell car (see the essay on
Cars) within a 1 - 2 decade time frame.
As we can see in the essay on
Electricity, this leaves us with the choice of using fossil fuels, with
the consequent inability to meet Cyprus' commitments to the Kyoto Protocol
(see the essay on Climate change) and
the EU or nuclear fission, which would probably be unacceptable to many
people on the island.
If hydrogen were to be a popular source of
energy, whether for fixed or mobile uses, it would have to be distributed.
It would seem likely that a pipeline may be considered. Experience with
natural gas pipelines has revealed that there are always considerable
losses, often in the range of 1 - 2 per cent per 100 km, averaged with
pumping stations. Natural gas has a molecule which is 8 times more massive
than that of hydrogen, which would filter through the smallest leak much
more readily than would methane, especially as it would be necessary to pump
hydrogen through at a higher pressure (or to have an expensive and
potentially leaky pump at each filling station).
An alternative method would be by pressurised
insulated tankers, transporting liquid hydrogen at -253°C. A spherical tank
could be dropped off at a filling station. As the contents boiled, the
pressure would rise. The problem with this solution is that the outtake of
hydrogen would have to match the amount of heat being inputted, to keep the
pressure safe. This would involve a quite sophisticated installation, to
prevent leakage to the atmosphere. The advantage is that liquid hydrogen
would be less likely to produce an explosion than an equivalent weight of
compressed gaseous hydrogen. The density of liquid hydrogen is just under 71
g/l, so the tanks would have to be very large.
Hydrogen, no matter how produced, is a costly
fuel. The following estimates, just for the production, were published by
the International Energy Agency, a pro-hydrogen lobby, in the mid-1990s. The
prices would have to be adjusted upwards to cater for the increased cost of
energy, but the proportions remain. To put it in perspective, one litre of
refined petrol at that time had a production cost of about USD 0.10 and
produced 0.035 GJ or about USD 2.9/GJ.
|
Hydrogen produced from:
|
Cost range in USD/gigajoule
|
|
coal/gas/oil
|
1-5
|
|
natural gas minus CO2*
|
8-10
|
|
coal minus CO2*
|
10-13
|
|
biomass
|
12-18
|
|
nuclear power
|
15-20
|
|
onshore wind**
|
15-25
|
|
offshore wind**
|
20-30
|
|
solar cells**
|
25-50
|
|
* The IEA does not specify how the
carbon dioxide will be sequestered. In my opinion, this will not
be a practicable proposition (see the essay on
Sequestration)
|
|
** In ideal climates for the
technology
|
The cost of producing electricity from fossil
fuels, for comparison, is typically in the USD/GJ 6-12 region and for
nuclear generation USD/GJ 10-12 (including externalised costs such as
insurance and decommissioning).
No figures have been published on the cost of
hydrogen distribution, but it is estimated that this may approach the
doubling of the cost to the consumer.
If hydrogen is mixed with air in any proportion
between 4 and 74 per cent in a confined space, ignition will produce an
explosion. In the open, the risk is lower but not inexistent, because the
low density of the gas will cause it to dissipate upwards very rapidly. A
sudden leak in the open, such as a bursting hydrogen tank, where there is a
source of ignition, would usually produce a massive fireball which would
consume everything within a certain radius, but travel upwards, away from
the earth, within seconds. An indoor leak could be very violent.
As a matter of comparison, most violent
accidental gas explosions causing severe damage to buildings are caused by
natural gas or bottled LPG (butane or propane). The explosive limits for
methane are 5 to 15 per cent, and for propane and butane 2 to 9 per cent, in
round figures. It can therefore be seen that the danger of ignition with
hydrogen covers a much wider range of concentrations.
However, large quantities of hydrogen are
produced and used by the chemical industries, for example in the production
of ammonia, in almost complete safety, so the technology is available, at
least within the confines of a chemical plant. I am personally very
concerned about its use in general public applications, such as in cars. A
leak, especially in a confined space such as the body of a passenger car or
in a closed building, could possibly have disastrous effects, even if the
tanks were somehow made safe from rupture in the event of a road accident.
Similarly, I don't think I would wish to live within a few hundred metres of
a filling station dispensing hydrogen! (See the page
Hydrogen myths for further discussion).
As far as is known, widespread use of hydrogen
as a combustible, is not likely to have any important negative environmental
effects. The quantity of water vapour produced would be small compared with
the natural loading and it would be quickly dissipated. In large cities, the
relative humidity may rise slightly during periods of heavy use (e.g., rush
hour traffic). Natural atmospheric mixing would soon dissipate this. It
should not be forgotten that petrol and diesel traffic today also produce
large quantities of water vapour, so the difference would not be great.
If all cars throughout the world "burnt"
hydrogen, there could be a slight rise in global humidities, which would
result in slightly more cloud formation. This may affect precipitation
patterns in some places and even increase the earth's albedo, causing a
minute global cooling. However, major climate changes are not foreseen.
There is one cause for concern that has been
raised. Because hydrogen is light, any leakage will travel to tropopausal
levels quite rapidly, before any natural oxidation can occur.
Transtropopausal mixing mechanisms between the troposphere and the
stratosphere are quite violent and hydrogen entering the stratosphere will
cause ozone depletion (4H2 + 2O3 > 4H2O + O2).
This is not a catalytic chain reaction as is caused by, for example, the
halogens in CFCs and halons, but it is expected that the sheer volume of
hydrogen potentially reaching the stratosphere would cause significant
depletion and the resultant increase of stratospheric water vapour could
also contribute to global climate change. These phenomena have not yet been
fully modelled by atmospheric scientists, so there is no scientific
assessment of the potential effects of a global change to a hydrogen-based
economy, available at this time. Many scientists have expressed concern,
though.
Hydrogen may seem to offer a good solution in
the long term, as a combustible fuel or for fuel cells, but there remain
many technical and economic problems to be resolved before it can be
considered as a viable substitute for fossil fuels. In the Cyprus context
(and this also applies to many other countries), the most
environmentally-friendly way of generating hydrogen would be via
electrolysis with power coming from a nuclear fission plant, which would
also provide the energy for desalinating sea water for the electrolysis.
|