Energy and transportation(∗)

— Transportation takes a considerable and increasing fraction of the energy use worldwide, and more than half the oil consumption. By far the largest part is used by cars powered by internal combustion engines. The advantage of using internal combustion engines is that the energy density of liquid fuels is extremely high. The disadvantage is that gasoline and diesel engines have a poor performance: 20 to 25% only. How does this compare with electric cars? What are the alternative transportation systems and their efficiencies anyway? In this lecture we analyse the efficiency of various transport systems, using elementary physics principles. We will look at cars, buses, trains and TGVs, ships and aircraft. In addition, the efficiency of human powered vehicles will be considered. New and promising developments in the field of Intelligent Transportation Systems, like Cooperative Adaptive Cruise Control, are also discussed.


Transportation
Transportation has gradually become the leading item on many personal energy bills.This is not only because gasoline is heavily taxed: Also in terms of energy, the average family car may be the chief private energy user in many developed countries [1].For example, if a family drives an average of 20000 km annually, it consumes ca.1400 litres of gasoline.In terms of energy use, this is about the same as the 1600 m 3 of natural gas consumed annually by the average Dutch household.
So let us look at how we travel.In most developed countries, the average resident uses the car as his or her primary mode of transport.In the Netherlands, for example, the average person travels a total of 32 km per day, of which 25 km is by car, 3.6 km by public transport and 2.5 by bicycle (data 2007).In rapidly developing countries like China, the use of cars may still be moderate, but automobile sales are soaring.
It is, therefore, a good idea to look first at the automobile.How efficient is it?Can we improve it?If so, how?

-Cars
In most countries, the largest proportion of car travel is on horizontal roads.This means that it does not involve much climbing and descending, and, since the two compensate one another to a large extent anyway, let us focus on flat roads.First, we obviously need energy when we accelerate.The energy needed to reach speed v is 1/2 mv 2 .This means that accelerating to 100 km/h takes 4 times as much energy as accelerating to 50 km/h.This is good to keep in mind whenever we notice the traffic light turning yellow in the distance.If we proceed carefully we may be able to avoid braking and having to accelerate again.Anticipation is the keyword here.
How much fuel does an extra full stop take?If we do the calculations we discover that, roughly speaking, accelerating from 0 to 100 km/h takes as much fuel as driving 1 km on a highway.Incidentally, for a train this would be about 15 km.This is because a train is relatively heavy and is pretty economical once in motion.The lesson is that trains perform well provided that they do not stop too frequently, unless they recover the braking energy.
Since gasoline (or diesel, or whatever energy resource) is consumed primarily on the highway, let us look at this situation and find out what exactly determines energy use.When we drive at constant speed, our car has to cope with a certain amount of resistance.In fact, the engine has to push the car with exactly the same force that is needed to overcome that resistance.And it is precisely the resistance that determines the fuel consumption, as we will see later.
What is that resistance we have to cope with?There are two forms of resistance, and we will discuss them separately.
1. Rolling resistance.-The ball bearings do not really play much of a role when it comes to the rolling resistance of a car.Rolling resistance is almost entirely caused by the deformation of the tires on the road.That may seem odd: Is rubber not elastic, and does it not bounce back nicely into its original form after it is compressed?
That is indeed true, but the process of compression and expansion does take some energy.This is because the force necessary to compress rubber is greater than the "pushback force" that the tire delivers when it expands (see Box 1).The difference between the two -the net energy -is converted into heat.You can feel how warm the tires get after you drive on a highway for a while!The resulting resistance from all this -the rolling resistance, or rolling friction -is proportional to the weight of the car.It is, in fact, a fixed percentage of that weight, given a certain type of tires at a certain tire pressure.That percentage is called C r , the "rolling resistance coefficient".
What is the value of this C r ?Could it be around 1 or 0.1?These high values can not be right, and here is why.Suppose our car is parked on a slope, and we release the brakes.How much of a slope do we need for the car to start rolling?It turns out that a slope of about 1% is enough.What is the force that pushes the car forward on such a slope?Let us look at the weight of the car (which points in the vertical direction) on this sloping road.What we need is its component parallel to the road.We find that this component is 1% of the car's weight in this case.That is the force which pushes the car forward.So, if that is enough to overcome its rolling resistance, the value of C r must be ca.0.01.And sure enough, it is -at least if the tire pressure is right.It is only logical that C r increases as the tire pressure decreases, because the whole story boils down to the deformation of the tire.It turns out that the rolling resistance increases by about 5% if the pressure is decreased by 10% [2].
The value of C r varies for different tires and road surfaces.In this respect concrete performs best.It does not give way at all and consequently has a somewhat lower resistance than asphalt, which is just a bit softer.
For the average car, a reasonably precise value for C r is 0.010.But, for modern tires, it is already a bit lower at 0.009.Future figures of 0.006 seem plausible.
It is important to note that speed does not play a role here.The rolling resistance is, for all practical purposes, independent of speed.

Air resistance ("drag"
). -The air resistance, or drag, behaves differently.It goes up with the square of the speed.We could guess that from experience: If we stick our hand out of the window at 100 km/h, the force we feel is far larger than at 50 km/h.It is 4 times as large, to be precise.
Furthermore, the air resistance is proportional to the density of the air.This means that the air resistance increases with increasing atmospheric pressure.It also means that, in the heat of the summer, the air resistance is lower than in the winter, given a certain atmospheric pressure.Analogously, also the thinner air at higher altitudes reduces drag.This also holds for bicycles and speed skaters: breaking a speed record is easier at higher altitudes.
Another factor that determines the drag is the size of the car, or -more preciselythe frontal area.Low and narrow cars have less air resistance than high and wide ones.
Finally, the air resistance is proportional to the famous drag coefficient C d , which is determined by how streamlined the car is.While the value of C d can be as high as 1 for clumsy objects that are not streamlined at all, it can go down to below 0.3 for some late-model cars.
Why are rolling resistance and air resistance so important?It is because they have a direct influence on fuel consumption.Together, they comprise the total resistance that a car undergoes when it is driven along a horizontal road.This is equal to the force that the engine must continuously produce to maintain a constant speed.The interesting point is that this force is equivalent to the work done -or the energy used -per unit distance.And this can be directly translated into an automobile's energy consumption (see Box 2).The rolling resistance F r is written as with C r being the rolling resistance coefficient, m the mass and g the acceleration due to gravity.This resistance is mainly due to the fact that the force needed to compress rubber is larger than the force which is delivered by the rubber as it tries to bounce back to its original form (in terms of physics: the closed integral F • dS is not equal to zero).The energy "lost" in this process is released as heat.
The air resistance or "drag" F d is proportional to the square of the speed v.We may derive directly from Bernoulli's law (p+1/2ρv 2 = constant, where p is the pressure and ρ the mass density of air) that the pressure right in front of a flat plate, where v = 0, must be equal to 1/2ρv 2 .Consequently, the force on such a plate is A × 1/2ρv 2 , where A is the area of the plate.However, most objects have some level of streamlining, which reduces that force, or resistance.This is expressed in the "drag coefficient" C d .So, the air resistance on an object with a frontal area of A becomes (2) The value of C d is ca.0.25 for late-model cars.The lower limit for the ideally streamlined vehicle is 0.05.Although the value of C d is not quite independent of speed, we may consider it as a constant for the relevant speed range.

Family car
Fig. 1. -The resistance that a car experiences increases dramatically at higher speeds, due to the rapidly increasing air resistance or drag (blue curve).This increases fuel consumption at higher speeds.The figure corresponds to a car with a mass of 1000 kg, a rolling resistance coefficient of Cr = 0.01, a drag coefficient C d = 0.4 and a frontal area A = 2 m 2 .
To assess the influence of the two types of resistance on the fuel consumption separately, we need to know their relative magnitude.The two are shown for a mid-sized model car in fig. 1.We notice that, when we drive through residential areas at lower speeds (below 50 km/h), the rolling resistance is dominant.It is clearly an advantage in this situation to have a light car (especially since it uses less energy during the frequent stop-and-go cycles).
Things change on the highway, however.Here air resistance is dominant.Now it is more important to have a small car in terms of its frontal area: low and narrow.Moreover, streamlining is essential here.So, driving with your windows open or carrying stuff on top of the car is not particularly efficient: it can actually increase fuel consumption dramatically.Since the air resistance increases with the speed squared, driving fast does consume more fuel.
Fortunately, things are not as bad as suggested by the steeply rising curve.This is because the efficiency of an engine increases at higher speeds, since it has to work harder (the efficiency is zero when the engine is idling!).This effect partly compensates for the steep increase in air resistance.But driving fast, in general, remains inefficient.Therefore, some sort of speed limit may be a good idea after all, if we consider energy use.An experimental illustration of how fuel consumption is dependent on speed is highlighted in fig. 2. With decreasing speed and while driving in fifth gear, an efficiency of 29 km/l of gasoline was achieved, compared to 20 km/l at 90 km/h.
Obviously, the results will vary for different cars.But the trend remains the same: it pays off to watch our speed.Moreover, we must aim for high engine efficiency.In this case, this was achieved by using fifth gear throughout, even at 40 km/h.

1
. 3. Room for improvement.-The ordinary gasoline-or diesel-powered car can be substantially improved if we are willing to sacrifice a bit of comfort.A striking example is the experimental XL1 car developed by VW.It is a low and well-streamlined two-seater having a drag coefficient C d = 0.19, a mass m = 795 kg, a height of 1.5 m and a small 800 cc, 35 kW diesel engine with an efficiency of 0.3.All this results in a fuel consumption of 2 L per 100 km.If driven as a hybrid (see below) with an additional 20 kW electric motor, it achieves a fuel consumption of about 1 L per 100 km.This result is reflected in its name: the experimental 1 liter car.Despite this remarkable efficiency it reaches a top speed of 160 km/h.

1
. 4. The hybrid car .-It does not sound particularly logical that in order to increase efficiency we equip a car with two engines: an ordinary internal combustion engine and an electric motor.Yet this is indeed the case with a hybrid car.It derives its energy exclusively from gasoline or diesel, so all it does is use this energy more efficiently than an ordinary car.How does it achieve this?
The secret is that its electric motor -in combination with a relatively large battery -makes driving in the city, with its frequent stop-and-go cycles, much more efficient than an ordinary car.At a traffic light, the motor simply turns off; it does not idle like the engine of an ordinary car.Furthermore, it recovers its braking energy by converting it into electricity that recharges the battery.In an ordinary car, that energy is lost in the form of heat in the brakes.So, all of a sudden, driving in the city has become much more efficient.
On the highway, this advantage disappears.Moreover, the battery has a relatively short range and the car must revert to running on gasoline or diesel, just like a "normal" car.This means that, for long-distance travel, a "normal" car with the same amount of streamlining is just as efficient.
Since the electric motor may be used in conjunction with a normal engine for a short time, to up a hill or overtake another car, it does not hurt to have a smaller internal combustion engine than you would normally find in such a car.This can further enhance engine efficiency, since the efficiency increases when the engine "has to work harder".
In conclusion, the advantage of a hybrid is mainly evident in the city.So, it may be a good idea for taxis to switch to hybrid cars.Not only would that make driving more economical, it would also reduce air pollution in the city.

1
. 5. The plug-in hybrid and the all-electric car .-The hybrid car discussed above still gets all of its energy from fossil fuel.We can now go a step further and really drive electrically.For long-distance travel this is not yet a practical option, since current battery performances keep us from having enough energy on board (see sect. 5 .

2).
An intermediate option is the plug-in hybrid.This enables us to drive on electricity drawn, for example, from the power outlet in our own homes.But if the batteries are running low, we can still cover a reasonable distance by switching to gasoline.The charging can be done at selected times when electricity consumption is low (and cheap), for example at night.Alternatively, if solar electricity is used, the batteries can be charged in the middle of the day during periods of maximum sunshine.
The all-electric car is currently an option for short-distance travel only.Its range (typically 150 km) may be sufficient for most commuter traffic.If and when battery development improves, cars may drive electrically for distances comparable to those commonly covered by current combustion-engine cars.
Obviously, the question arises whether the electric car is more energy-efficient than the traditional car running on fossil fuels.To compare the two, we must compare the efficiency of an internal combustion car engine (20% on average) with electricity generation (note that the efficiency of an electric motor is close to 100%).If we use a power plant fuelled by oil or natural gas, a net efficiency of 35% is a reasonable figure.So, if losses in the battery-charging process and in the electric motor are small, the electric car would prevail.The advantage is even larger when we consider the lighter weight of the electric motor compared to the internal-combustion engine, plus the fact that an electric car is more efficient in city traffic with its many stop-and-go cycles, as pointed out above.However, energy losses in the charge/discharge cycles can largely offset these advantages.
Electric cars may also be powered by electricity from fuel cells, which are addressed in sect.5 . 1.

2
. 1. Buses.-Buses and cars have in common that they both use rubber tires.Therefore, the rolling resistance coefficients should be similar, but the rolling resistance itself is for a bus much larger because of its weight.For a fair comparison, we should take the rolling resistance per seat.If we do that, we discover that cars and buses are quite comparable (see table I).In other words, if we assume that both have a 100% passenger load factor, cars and buses should have almost the same energy use rates per person at low speeds.However, things change at higher speeds, when air resistance, or drag, is more prevalent.It is granted that the bus has a larger frontal area and less streamlining than most cars, but this is compensated for by the large number of passengers a bus can transport.The result is that a bus has only half the drag per passenger compared to a car.So, at higher speeds the bus would beat a car by a factor of 2, as seen in table I.
Experimental data for buses.-An average bus for 50 passengers consumes about 1 litre of diesel every 3.5 km.This amounts to 175 passenger-kilometres per litre (the number of passenger-kilometres is the number of passengers multiplied by the number of km travelled).By contrast, a typical diesel car with 4 people on board and travelling at the maximum bus speed limit may reach 80 passenger-kilometres per litre.This, indeed, means that the car is only half as efficient as a long-distance bus.

2
. 2. Trains.-Trains are a lot like buses if we consider total air resistance.However, trains carry a much larger number of passengers than buses do.So, if we consider the air resistance per seat, the train gains a large advantage (see table I).It beats buses by a factor of 2 and cars by a factor of 10.Another big difference is revealed in the rolling resistance.Trains have steel wheels and run on steel rails.Therefore, the rolling resistance coefficient is much smaller.It may not be quite as low as one would expect from the stiffness of steel.This is caused by non-ideal wheel-rail contact that causes some extra friction, and/or by poor track maintenance.Reasonable figures for C r range from 0.0015 to 0.0020.This is still much lower than figures for rubber tires on roads.This advantage is partially offset by the relatively large weight of a train, but the rolling resistance of trains per seat remains about 3 times lower (see table I).
So, trains beat buses and cars especially at speeds at which air resistance is greater than rolling resistance.For trains, the two resistances are equal at around 70 to 100 km/h, depending on the type of train.Remember that for cars this happens at a somewhat lower speed, ca.50 to 70 km/h, again depending on the type of vehicle.
Stopping and accelerating has not yet been taken into account.Trains are relatively heavy, so accelerating is rather costly.Since cruising is relatively efficient, a train can cruise for quite a distance and use the same energy as needed for accelerating to cruising speed.If we work it out for the train used in table I, which consists of four double-decker cars, we find that this distance is ca.15 km.For a car, this figure is only ca. 1 km.The conclusion here is that trains are efficient especially at longer distances.The more they stop the less efficient they become, unless the braking energy is recovered (a technique that is being used more and more frequently).
During the winter, trains have to be heated, just like cars.But, unlike cars powered by internal combustion engines, they do not have the waste heat to recycle.Instead, they use electric heating, which can substantially reduce a train's energy efficiency.
Experimental data for trains.-A double-decker train consisting of 4 cars (372 seats in total, 254 tons) turns out to have an energy efficiency of 158 passenger-km per litre of gasoline, if converted to electricity.A 100% load factor and an average of 14 km between stops has been assumed.This is about three times as efficient as a full car, which reaches about 56 km/litre, assuming four occupants and an average of 14 km per litre.
The advantage of trains increases at longer distances.For 50 km between stops, the energy efficiency increases to 260 passenger-km per litre.Again, it is assumed that the braking energy is simply lost.

2
. 3. High-speed trains.-TGV (train à grande vitesse), High-speed trains, MagLev (magnetic levitation) all move at very high speeds through air at ambient pressure, so they are bound to be inefficient.The reason is that the air resistance increases with the speed squared, just like for cars.So, twice as fast means four times the energy.Compared to the air resistance, the rolling resistance becomes negligible at speeds above 300 km/h approximately.Therefore, magnetic-levitation trains (which have no rolling resistance) offer barely any improvement in terms of energy efficiency.
So, one would not expect high-speed rail travel to be very efficient.Despite its excellent streamlining (low C d ) we should expect that a TGV, travelling more than twice as fast as a normal long-distance train, will use about four times the energy.
Experimental data for TGV .-In practice, the TGV uses roughly five times as much energy as an ordinary long-distance train if this is expressed in terms of gasoline consumed (as we did with the trains above).It achieves about 50 passenger-km per litre.This confirms our calculations above, where we expected it to be roughly a factor of 4 worse LNES 2014 than a regular long-distance train.One aspect that may improve TGV performance in practice is the relatively high occupancy rate.

-Aircraft
The fuel consumption of planes per passenger-km depends on a number of variables.Is the plane full?What type of plane is it?How close are the seats to one another?Is the plane carrying cargo in addition to passengers?Is it a long-distance flight?
The last question has an interesting answer.Obviously, short flights are not very efficient, because of the relatively large amounts of fuel used for take-off and landing.So, longer flights should be more efficient.But wait, very long flights require extra heavy loads of fuel.Indeed, flights above, say, 5000 km become less efficient, since there are fewer passengers on board because of the weight of the extra fuel.We can estimate this by using a rule of thumb: an aircraft at cruising altitude consumes almost 10 cm 3 of fuel per second per seat, irrespective of the type of aircraft.For a flight of 5000 km that takes 6 hours, we can expect fuel consumption to be 6 × 3600 × 0.01 litres per passenger.This is 216 litres, or the equivalent of the weight of 2 passengers!To estimate fuel consumption, let us again use the rule of thumb that the fuel consumption for each passenger is almost 10 cm 3 per second.This number was fairly accurate for many different types of planes during the 1980s, but it has been gradually decreasing somewhat as planes are becoming more efficient.Even so, it is convenient to keep this in mind as a rule of thumb.
It follows that flying for one hour consumes ca.36 litres of kerosene per passenger.Since we cover almost 1000 km during an hour, the consumption should be about 3.6 litre per 100 km, or 28 km per litre per passenger.This is equivalent to the fuel efficiency of a car consuming 14 km per litre with 2 occupants.
The conclusion is that a full plane is (per km and per passenger) roughly as efficient as a half-full car.Since both planes and cars are becoming increasingly efficient, this rule of thumb may hold for many years to come.
There are some factors that complicate matter, however.Take cargo.Most planes carry a lot more cargo than just passengers' luggage.This makes it difficult to calculate efficiency with any precision (see also experimental data for aircraft below).
It may come as a surprise that planes are relatively efficient, given the fact that the air resistance increases steeply with speed.Why is that?We should bear in mind, first, that the cruising altitude for planes is at (or even above) 10 km.The air density is only one-quarter of the air at sea level.Furthermore, planes have excellent streamlining.Last but not least, their dead weight is relatively small.If we compare the total mass of passengers plus luggage plus cargo with an empty aircraft, we find a ratio of ca. 1 to 10.This is particularly favourable if compared with a passenger ship (see table II).
Experimental data for aircraft.-A long-distance flight by a Boeing 777-200, carrying 260 passengers and 12000 kg of cargo, consumes about 68000 kg or 85000 litres of kerosene over a distance of 9000 km.This is 28 passenger-km per litre, which corresponds to the

07001-p.10
Table II.-Energy efficiency in terms of number of passenger-km per litre of fuel, assuming a passenger load factor of 100% and long-distance travel.For electric vehicles, the electricity is assumed to be generated in an oil-fuelled power plant, and for the bicycle the energy equivalent of food is expressed in litres of oil.above figures.But we should also consider the cargo.If we assume the above 12000 kg of cargo to be the equivalent of 120 passengers including their luggage, all of a sudden we have 380 people on board.This would give us 40 passenger-km per litre instead of 28.

3
. 1. Comparison.-How do various means of transportation compare, if we consider energy use?Table II quotes some approximate values.The poor performance of the passenger ship is striking, especially in view of its low speed.The reason is not only that it has to move through water, the density of which is a factor of 10 3 greater than air at sea level.More importantly, the dead weight of a passenger ship -or a cruise ship -is enormous.Unlike other means of transportation, these ships have been designed to spend a lot of time at sea.Therefore, they have dining rooms, shops, ballrooms, even swimming pools and the like.In fact, these ships are floating villages.For example, take the relatively modern Queen Elisabeth 2. This ship is 345 metres long, has a mass of 70000 tons and carries 1900 passengers.If we include their baggage the passengers may weigh in at ca. 200 tons.This is only about 0.3% of the ship's total weight.The passengers on an aircraft comprise ca.10% of its total weight, as mentioned above.
Care should be taken not to generalise the poor performance of passenger ships to ships in general.The load carried by cargo-boats and oil tankers is usually very large compared to their own weight.Therefore, their energy efficiency is very good.

-The bicycle
Mankind managed to improve its transport efficiency by a factor of 4 by inventing the bicycle.Walking is already fairly efficient compared to flying and other modes of transport, including those in the animal world.But cycling beats every mode of transport, if one measures the amount of energy per kg of displaced mass over a fixed distance [3].

07001-p.11
Just how much energy does it take to ride a bicycle?There is an easy way to calculate this.The power that it takes to ride a bike is quite comparable to climbing stairs at a pace of one step/s, which requires ca. 100 W. Now imagine the following experiment in which we ride a bicycle for 24 hours continuously.The above 100 W of mechanical energy means that total energy use for cycling is 400 W, given a muscle efficiency of 25%.A continuous energy use of 400 W corresponds to one litre of oil per day.So, if we cycled for one day, we would use the equivalent of 1 litre of oil for a 24-hour bike ride.In those 24 hours, we cover a distance of 24 × 20 km, approximately 500 km.
Conclusion: A bicyclist consumes roughly "1 litre per 500 km".That is enormously efficient.But there is a catch.We do not cycle on oil or gasoline, but on bread, milk, steaks, and cheese.All of this food takes a lot more energy to produce and deliver to our home than its own energy content represents.Take milk, for example.The cows have to eat, they have to be milked (electrically), and the milk must be cooled down, transported, heated to be pasteurized, cooled again, and transported again.If you add it up, one litre of milk may cost half a litre of oil before it gets to your kitchen table.Much the same can be said regarding the rest of the food as well.So, while bicycling is wonderful and very healthy, if we want to save energy we might be better off buying a light motorcycle.

4
. 1. Rolling resistance and drag for bicycles.-Let us consider cycling on a horizontal road, so we do not need to fight gravity.The bike itself offers very little mechanical resistance, provided it is in good condition.The ball bearings cause a loss of about 1% only, a well-oiled chain about 1.5%, the gears (derailleur ) 5% at most [4].Once we reach cruising speed, the real forces that we have to compensate for are rolling resistance and air resistance.It is the same as for a car (see fig. 1).But with a bicycle, we are obviously working with different figures.
The rolling resistance coefficient C r is very comparable to the value we saw earlier for a car, again depending on the tire pressure.Typical figures for standard tires are between C r = 0.006 and 0.01.For racing bikes a typical value is 0.003.
The air resistance or drag coefficient (the "C d -value") is determined by the streamlining.For a bicyclist sitting upright C d is about 1.For a hunched-over racer it is about 0.88 (in this case, the frontal area is also reduced, leading to a double gain).But for a cigar-shaped super bike, the HPV (Human Powered Vehicle), it can be as low as 0.1.
The behaviour of the two resistances for an ordinary city bicycle is analogous to the behaviour shown for a car in fig. 1, except that, for a bicycle, the rolling resistance dominates for speeds up to about 15 km/h.At higher speeds, the air resistance is dominant.For reaching high speeds, the streamlining is therefore essential, and setting a speed record essentially means reducing air resistance.Super bikes (HPVs) can reach speeds over 100 km/h, which an ordinary racing cyclist cannot even think of.
As noticed for the case of a car, the total resistance, or force, is a direct measure of energy use per kilometre.At a speed of 20 km/h, the total resistance for such an ordinary city bike (C r = 0.006; m = 90 kg; C d = 1.1;A = 0.50 m 2 ) is found to be 16 newtons.Mechanical energy use therefore is 16 kJ/km.With a muscle efficiency of 25%, EPJ Web of Conferences 07001-p.12Table III.-Properties of molecular hydrogen (H2).( a ) This is similar to natural gas.The "higher" heating value includes the heat of condensation of the water vapour; the "lower" value does not.

Heat of combustion (higher
this means a total energy use of 64 kJ/km.Since the energy content of oil or gasoline is 40 MJ/kg, this corresponds to 1.6 gram per km.So our earlier rough estimate of 1 litre per 500 km, i.e., 2 grams per km was fairly accurate after all.
And how much fat would we lose on our bicycle if we do not eat anything extra?If we take the above case and we realise that the energy content of oil and fat are quite close, we conclude that we burn 2 grams of fat by cycling 1 km.So, if we want to slim down by 1 kg on your bicycle, we find that this takes a distance of some 500 km.

-Alternative energy carriers
After the "fossil-fuel era" alternative energy carriers will have to be used.The most promising candidates are likely to be hydrogen and advanced batteries.Let us have a closer look at both, and include "super capacitors" which have enjoyed a remarkable development lately.

5
. 1. Hydrogen as an energy carrier .-The relevant properties of hydrogen are given in table III.Once electricity is available (for example, by using solar PV panels or wind turbines), hydrogen can be produced by electrolysis.The efficiency of this process is around 80% for modern equipment.In the future, H 2 may be also efficiently produced directly from solar energy by "Photo-electrolysis" or "Biophotolysis".
Hydrogen's efficiency as a fuel depends on how it is used.
-Hydrogen may be used simply as a fuel for internal combustion engines.However, we must realise that the efficiency will not exceed 20 to 30%, as it is limited by the Carnot efficiency.If we want to use hydrogen to produce mechanical energy, we can do much better: -Hydrogen can also be used in a fuel cell, which directly converts H 2 and O 2 into electricity.Remember that an electric motor has an efficiency close to 100%, in contrast to an internal combustion engine.By going directly from fuel to electricity we avoid the inefficient step of converting heat to mechanical energy.An efficiency of 60% in low.Alcohol may be one of the only alternatives.But alcohol production itself requires energy.
This may add an extra incentive for conserving oil.It is too handy a fuel to simply burn like scrap wood.All alternatives are much less convenient for transportation purposes.

5
. 2. Electricity storage for transportation.-The hybrid and the electric cars were both discussed in sect. 1.The basic problems with electric cars are twofold: the batteries are still too heavy and bulky, reducing the range to some 150 km for most cars, and "refuelling" still takes too long.
If one compares the refuelling rates of petrol vs. electrical energy, one comes up with some amazing figures.Charging a car battery at a standard 230 V electric outlet at home may be done at a maximum rate of about 3.5 kW, if we want to prevent the 16 A fuse from blowing.Just how slow this transfer rate is becomes clear if we compare it to petrol.Refuelling a normal car at a standard service station takes about 100 seconds for 50 litres.The transfer rate is thus 0.5 litres per second or about 18 MJ/s, or 18 MW!In other words, this is 5000 times faster.
So, the future electric car may need services such as battery exchanges rather than battery recharging.This reminds us of the old days of the stagecoach.The difference is that in those days horses were exchanged, not batteries.
Batteries and supercapacitors.-Storing electrical energy is not easy.Almost all of the current cars still rely on traditional lead batteries, but these do not have a lot of energy.An average 12 volt battery has a charge of 40 Ah.Combining the two figures yields 480 Wh, or 0.480 kWh of energy.This is the equivalent of a wine glass full of petrol.Taken per kg, this battery contains about 0.03 kWh/kg of energy only.
The nickel-metal-hydride battery does a much better job (see table V).It is relatively cheap, safe and contains 0.06 kWh/kg of energy, which is twice as good as the lead battery.It is currently used in many hybrid cars.
Lithium-ion batteries in mobile phones and laptops perform even better, achieving ca.0.14 kWh/kg.However, they are relatively expensive, complex and present some safety issues if used in cars.Moreover, the Li-ion battery is not particularly durable as a result of various chemical processes.It looses performance more quickly at higher temperatures and when fully charged.Optimum storage conditions imply lower temperatures with the battery loaded to 50% of capacity.Useful information is found at: www.batteryuniversity.com.
Batteries have problems delivering large amounts of energy quickly, i.e., if they must produce a high power.This is because the chemical processes that form the basis of the storage mechanism react slowly.True electricity storage occurs in capacitors, which do not display these problems.They can be charged and discharged extremely quickly.But common capacitors as found in electric appliances are totally useless in storing large amounts of energy.Modern "supercaps" have a much larger capacity.They are extremely well suited for storing the braking energy of a car or a trolley, and subsequently EPJ Web of Conferences 07001-p.16re-delivering it for quick acceleration.They are also called "boostcaps" since they give brief boosts to racing cars.Another advantage is that they can be charged and discharged millions of times without performance loss, in contrast to batteries with a much more limited number of charge/discharge cycles.One drawback, however, is that, while they can deliver large amounts of power, they cannot store large amounts of energy.Moreover, they are still expensive.

-Intelligent transport systems
Modern micro-electronics and on-board sensors make it possible to achieve appreciable improvements in car efficiency by using inter-vehicle communication.The key here is reducing air resistance by driving in "the slip stream" of the preceding vehicle.In fact, by achieving automated short-distance vehicle following in a safe way, improvements in three fields are obtained.First, improved energy efficiency by reduced drag.Second, an increase in road capacity (or "throughput"), and third, an increase in safety.An example of this approach is Cooperative Adapted Cruise Control, which uses a combination of wireless communication and on-board sensors including radar [5,6].
The first requirement is that the system allows a car to accurately follow the preceding vehicle at a fixed short distance.To this end, one-or two-vehicle look-ahead communication is used.Secondly, the system must provide "string stability": fluctuations which may arise when the preceding vehicle slows down must be damped.This is in contrast to what often happens on busy roads when a car suddenly brakes: the following driver has the tendency to brake more abruptly, and so on, which sometimes results in a virtual traffic jam.
Finally, in order to meet safety requirements, the system must also be capable of coping with flaws in the communication; in other words: it must provide "graceful degradation" if data transmission would fail.
A prototype platoon of six vehicles appeared to yield string-stable behaviour at an inter-vehicle time gap of 0.7 s.A time gap down to 0.3 s seems feasible [5].These figures show that such systems are also capable of dramatically increasing road capacity.This is easily seen when comparing these time gaps with the "two-seconds rule" used in some countries.This "two-seconds rule" yields a road capacity of 1800 vehicles per hour in the asymptotic limit where the car length can be ignored.In contrast, a time gap of 0.7 s beats this value by almost a factor of 3.

Box 1 .
-Rolling resistance and Air resistance.
Fig. 2. -Fuel efficiency in terms of number of kilometres per litre for speeds between 40 and 90 km/h in fifth gear (Toyota Yaris, courtesy of Herbert Blankesteijn).

Table I .
-The approximate values for the quantities that determine the energy use of cars, buses and trains.The figures for rolling resistance and air resistance are calculated per seat.These numbers thus provide a direct comparison, if we assume an equal passenger load factor (see also table II).

Table V .
-Approximate energy densities of batteries and supercapacitors, in kWh per kg.