Intelligent Machines

A Practical Road to Lightweight Cars

Sure, we could make the leap to a plastic “supercar” but who could afford to buy it? The auto industry can get just as far, and achieve lower costs, by taking one step at a time.

The automobile is the defining technological artifact of the twentieth century. Its familiarity, however, belies its complexity. It is no mean feat to design a car that is fast and powerful yet comfortable and safe-and still affordable. Factor in a few more constraints-durability, ease of repair, enough room for a few kids and the family dog, and an ample power supply for the electric windows, air-conditioning, CD player, and heated seats-and the challenge becomes clear. Precisely because the automobile has become an integral part of our lives, consumer expectations establish a set of formidable and often conflicting design objectives.

Over the last 25 years, automakers have faced growing pressure to incorporate environmental objectives into their designs as well. In particular, consumers and the federal government have pushed for improvements in fuel economy as a way to conserve oil and control pollution. The automobile industry has responded: the gas mileage of the average new car rose from 14.2 to 28.2 miles per gallon between 1974 and 1995.

Now public pressure to improve fuel economy is again rising, in part because of concern over the prospect of global climate change. (Automobiles account for about one-quarter of carbon dioxide emissions, a major contributor to the greenhouse effect.) The key to improving a vehicle’s fuel economy is weight reduction: the smaller a vehicle is, the less power it requires to accelerate and the less energy to maintain a fixed speed. Traditionally, the automotive industry has reduced weight primarily by downsizing, a strategy that has succeeded in cutting the weight of a typical car from 3,500 pounds to 2,500 pounds over the past 20 years. Today, that strategy has reached its limits. Substantial improvements will be possible only through a new approach: making the automobile body out of lightweight materials instead of basic carbon steel.

This story is part of our January 1997 Issue
See the rest of the issue

Although the body accounts for only about one-third of the weight of an automobile, reducing the weight of the body is the sine qua non of the lightweight, fuel-efficient automobile. A car with a lighter body can use a lighter engine, a less massive suspension, and a less elaborate structure. These secondary weight savings can roughly double the benefits: for every 10 pounds saved by reducing the weight of the body, another 10 pounds can be saved by downsizing other parts of the car.

What’s more, many new technologies designed to improve fuel economy are feasible only for cars that are substantially lighter than today’s. Automobile engines, for instance, must balance the goals of efficiency (energy per distance traveled) and power (the force needed to accelerate the car). High-efficiency internal combustion engines, electric engines, or hybrid engines that combine the two are all far less powerful than conventional engines and will achieve a comparable level of performance only with a much lighter vehicle. Reducing the mass of the body is essential to creating a synergy between light weight and new engine technologies.

In 1993, a highly influential paper by energy analyst Amory Lovins of the Rocky Mountain Institute suggested that major automakers (or anyone else with the gumption) could use existing materials and technologies to produce an ultra-lightweight, highly fuel-efficient vehicle. The “supercar” he envisioned would incorporate lightweight plastics, computerized controls, and a hybrid powerplant-a power system that would combine a traditional heat engine and an electric motor, like a modern locomotive. It would weigh roughly 1,000 pounds and achieve well over 150 miles per gallon-yet it would retain the safety and convenience features of today’s automobile.

Lovins pointed out, correctly, that the materials and technologies that would make a supercar possible are fundamentally incompatible with the design, manufacturing, and organizational processes around which the automobile industry is structured. He therefore argued that only a revolution in the industry would lead to a supercar; efforts to improve fuel economy and performance through the incremental adoption of new materials and technologies would cost too much and yield too little.

The supercar concept attracted a great deal of attention among environmentalists, auto industry leaders, and policymakers and even helped inspire an unusual alliance-though its goals fall somewhat short of Lovins’s. In 1994, U.S. auto companies and the federal government joined forces to launch the Program for a New Generation of Vehicles, an aggressive research and development project whose goal is to produce a car that meets a fuel-economy standard three times higher than today’s 27.5 miles per gallon and that offers the performance and convenience of a conventional car-for the same price. By combining the resources of the national laboratories and the major U.S. automakers, PNGV researchers hope to develop a prototype vehicle within 10 years and to mass produce and market it within 20.

The question is not whether an ultra-lightweight vehicle offering revolutionary improvements in fuel economy can be built. Automakers already know that it can. The question is whether such a car can be made affordable, and what kinds of changes in the automobile industry will be necessary to bring us closer to that goal. In particular, automakers and supercar proponents are debating the costs and benefits of two classes of materials that could serve as lightweight substitutes for steel in vehicle bodies: aluminum, which can be adopted with only incremental change in the industry’s design and manufacturing processes; and plastics, which cannot.

Aluminum’s Pluses and Minuses

A light metal 45 percent as dense as conventional steel, aluminum has been used as a major structural material in the aerospace industry for many years. Although it is expensive-aluminum sheet sells for about $1.50 per pound, compared with about 30 cents per pound for steel sheet-researchers in the automobile industry have begun to investigate the possibility of substituting aluminum for steel in vehicle bodies.

One of the main advantages of switching to aluminum, compared with other lightweight materials, is that it can be formed using many of the techniques already applied in making automobiles out of steel. Thus the industry could continue to use much of its existing equipment. And designing for aluminum is not drastically different from designing for steel-an important advantage in an industry where engineers are reluctant to experiment with relatively untried materials.

Of course, the fact that automobile bodies are not largely aluminum today suggests that the material also has disadvantages. Besides being more expensive than steel, aluminum is only about one-third as stiff-a crucial limitation in automobile body design. Stiffness can be increased somewhat by changing the geometry of the design (curved shapes are stiffer than flat ones), but this is problematic in an industry where shape and style are important sales concepts. An easier solution is to make flat aluminum body panels-fenders, hoods, and doors-thicker than steel panels to ensure that they perform equally well. This imposes higher material costs, however, and offsets the weight advantage to some extent.

Another problem is the high electrical conductivity of aluminum, which makes spot welding difficult. Spot welding is the standard method for assembling steel automobile bodies. The two parts being joined are clamped between two electrodes and electrical current is applied, thereby heating the two parts at the point of contact, leading to diffusion bonding. (The metal does not actually melt, since this would reduce the material’s performance and lead to corrosion and part failure.)

Because aluminum conducts heat better than steel, it takes a lot more electricity and larger electrodes to make the metal hot enough to bond. And because the electrodes stay in contact with the aluminum longer while the current is being applied, aluminum atoms are more likely to diffuse into the electrode, shortening its useful life. Aluminum vehicles will probably therefore rely on alternative assembly techniques, including seam welding (in which a strip of molten metal is applied more or less like glue), adhesives, and mechanical fasteners.

Unibody versus Space Frame

The challenge facing the automobile industry is how to design an aluminum automobile so as to capture the advantages of the material and minimize the disadvantages. There are two competing possibilities: a unibody, short for “unitized body,” the design used for steel automobiles; or a space-frame design, essentially a large truss structure covered with a thin skin.

In a unibody, the vehicle’s body panels are joined together to form a shell structure. This makes efficient use of the high stiffness of the body panels. Although aluminum is not as stiff as steel, if the panels are made thick enough and appropriate joining techniques are used, the unibody design will work well with this material.

However, the unibody design poses two related problems. First, it is relatively difficult (and therefore expensive) to make complex surfaces, such as cutouts or elaborate curves, from relatively stiff metal body panels. If designers attempt to circumvent this problem by using materials that are easier to form, the second problem arises: because the unibody derives most of its structural performance from the way its parts are attached, those parts must be composed of materials that can easily be joined. Without an inexpensive way to fasten two dissimilar materials to one another, the unibody design essentially requires the automaker to manufacture cars using a single class of materials.

In response to these objections, designers are exploring the space frame. In this design, the vehicle structure is composed, in effect, of a lattice of metal rails, similar to a bridge truss. The vehicle does not rely on body panels for structural performance and in fact can be driven without any panels attached. This design does not work well for steel, in part because complex steel rails are not that much easier to make than complex steel body panels. Today the consensus among automakers is that the unibody is the most efficient way to make a mass-market vehicle out of steel.

However, the space frame is gaining renewed attention from designers working with alternative materials, especially aluminum. It is easier to make complex rails out of aluminum than steel because, unlike steel, aluminum can be extruded-formed into complex tubular shapes-in a process similar to pasta-making. These extruded, hollow rails can be far stiffer than solid bars of equivalent weight. Extrusion is easily adapted to mass production; it is already used on a large scale to manufacture construction shapes such as window frames and pipes. Several designs for aluminum space-frame vehicles have been developed, each using differing combinations of extrusions, castings, and sheet metal. While the jury is still out, with the right combination of materials the space frame may someday challenge the unibody in mainstream automobile production.

Is Aluminum Affordable?

An aluminum vehicle based on either design would bring us closer to the goal of building a lightweight car at a relatively moderate increase in cost. A typical steel unibody weighs just under 600 pounds, while an all-aluminum unibody weighs about 325 pounds and various aluminum space-frame designs would weigh between 285 and 385 pounds. Thus either design could cut the weight of the body nearly in half; a lighter engine, suspension, transmission, and so forth could double the number of pounds saved. (Of course, weight may be added in other areas to compensate for the deficiencies of the new design-for instance, a lightweight car cannot rely on its structural components to protect passengers in the event of a crash and so will need to employ additional systems, like airbags, which add some weight.)

Just how much fuel savings are generated by lightweighting the body alone? Reducing the weight of the vehicle by 300 pounds can increase fuel economy by as much as 15 percent. This would increase the gas mileage of a typical mid-sized car, such as the Ford Taurus, from about 22 to about 25 miles per gallon, and reduce carbon dioxide (CO2) emissions from about 410 grams of CO2 per mile driven to about 355 grams per mile. Secondary weight savings would double the improvement in fuel economy and the reduction in emissions. More dramatic improvements in fuel economy would result in proportional decreases in CO2 emissions, but these would require much more drastic measures than mere lightweighting: more efficient engine technologies, for instance, and probably less room and fewer conveniences than the American consumer typically expects.

A lightweight aluminum car based on either of these designs is likely to be somewhat more expensive than today’s steel car when produced in large volumes, according to cost analyses by members of the Materials Systems Laboratory at MIT. At very low production volumes (less than 20,000 vehicles per year), aluminum space frames are actually cheaper than a steel unibody: the least expensive space-frame design would cost about $4,500, compared with $5,800 for a steel unibody and $7,200 for an aluminum unibody.

However, these production volumes are much too low for mass-market vehicles. Popular models such as the Ford Taurus are produced in volumes of 300,000 to 500,000. Even niche vehicles-luxury cars like the Lincoln Continental-have production runs between 40,000 and 80,000. To be considered affordable, a lightweight vehicle must be able to be manufactured inexpensively in large quantities.

At production volumes of about 100,000, the steel unibody is the cheapest design, at an estimated unit cost of $2,500. Aluminum space frames are a bit more expensive-the cheapest design costs about $2,800-while the aluminum unibody costs about $3,600. For more typical production runs of 300,000, the cost of the steel unibody declines to an estimated $1,400, and the aluminum unibody becomes cheaper than the aluminum space frame ($2,000 compared with $2,400).

The changing cost profiles for the three designs result from differences in their manufacturing processes. Metal stamping-the process by which both steel and aluminum unibodies are made-is better able to capture economies of scale than extrusion. As a result, the unit costs of both kinds of unibodies decline as they are produced in greater quantity; the cost differential between them is largely explained by the difference in the cost of the raw material.

The space frame follows a different pattern. Because the capital costs of extrusion are far lower than those of steel stamping, space frames are less expensive than unibodies at low production volumes. But extruded parts require finishing and heat treating, which are time consuming. Furthermore, the rate at which extruded parts can be formed is far slower than the rate at which stamped parts can be made. As a result, unit costs do not decline as dramatically when production volumes increase. Higher production volumes ultimately shift the economics in favor of the unibody.

Given that a vehicle with an aluminum body is going to cost $300 to $1,100 more than a vehicle with a steel body, will increases in fuel economy make up for the increased cost over the lifetime of the vehicle? The answer depends on a variety of factors: the total weight (and cost) of the vehicle, the efficiency of its engine, and the price of fuel. However, the increase in fuel economy attributable to the aluminum body alone would pay for itself only if the price of gasoline were to rise. If the price of gasoline remains between $1.20 and $1.50 per gallon, the money saved on gas would not be enough to make up for the higher cost: the life cycle cost of an aluminum unibody produced in volumes of 300,000 would remain about $300 more than that of a steel unibody. But if the price of gasoline rose to $2.30 per gallon, the owner of the aluminum-based car would break even over the vehicle’s lifetime. It is reasonable to think that under these circumstances, consumers might be willing to pay the higher up-front cost of an aluminum-based car.

The Appeal of Plastics

Advocates of the revolutionary approach, however, stress the advantages of plastics as a more radical lightweight alternative to steel. Plastics are more than twice as light as aluminum and can be formed into a much wider variety of shapes. Moreover, the equipment used to manufacture plastics costs much less than the heavy stamping equipment required to make metal parts. These qualities have attracted automakers’ interest since the 1960s.

Today the industry has incorporated plastics in a variety of uses; they form the interior components of most cars, for example, as well as bumper covers and fenders. Manufacturers and designers have also used polymeric composites-plastics reinforced with either glass or carbon fibers-in the bodies of race cars and some commercially produced vehicles. In the 1980s, as automakers looked for new ways to reduce vehicle mass, many in the industry began to investigate the use of polymeric composites to substitute for steel in automobile bodies.

Like aluminum, composite materials have their disadvantages. For one thing, they are more expensive than other automotive materials. The plastic resin mixture costs between $1 and $10 per pound and glass fiber prices start around $1 per pound. Glass fiber polymeric composites are price competitive with aluminum or steel only when used in small quantities or in complex shapes that are prohibitively expensive to form from metal.

In addition, ordinary plastics are between one-thirtieth and one-sixtieth as stiff as steel, while reinforced plastics are about one-fifteenth as stiff as steel. The traditional uses of plastics in automobile interiors capture the advantages of light weight and ease of formation without requiring a high degree of stiffness. Unibodies, however, have to be stiff to perform effectively. Structural panels composed of reinforced plastics must therefore be much thicker than their metal counterparts, offsetting the reduced weight and raising costs even further.

Carbon fiber composites have drawn the industry’s interest as an alternative to glass fiber composites because they are stiffer. Panels composed of these materials can be made thinner-and thus lighter-than their glass-reinforced counterparts. However, carbon fiber composites are prohibitively expensive: carbon fiber prices start at $20 per pound and rise dramatically with increases in fiber strength and stiffness.

Polymer-based unibodies are also difficult to manufacture. Although bodies made of reinforced composites would require only one-third as many parts as conventional metal bodies, these parts would have to be made to fit together exactly-something that is beyond the state of assembly art today. Since plastic resin and carbon fibers contract at different rates as they cool, the parts are bound to warp and shrink slightly in ways that vary unpredictably from piece to piece. That’s not unusual-steel changes shape as it cools, too-but materials like steel can be bent and twisted into shape. For instance, assembly-line workers use wooden mallets and two-by-fours to make sure steel car doors hang properly and seal when closed. Reinforced plastic components cannot be deformed in this fashion-plastic will break sooner than bend-so there is no easy way to compensate for slight imperfections in the way parts fit.

Finally, producing an affordable vehicle requires large-scale production, with volumes of at least 30,000 units per year and possibly an order of magnitude higher. While nonstructural plastic components can easily be manufactured on this scale, processing technologies for reinforced plastics are better suited to lot sizes of hundreds or thousands rather than hundreds of thousands. The cheapest way to shift to mass production of polymeric materials would be to speed up the process, making many more parts with the same equipment. But the processes involved in manufacturing and shaping reinforced polymer-based materials are not particularly amenable to this kind of straightforward scale-up.

The critical problem is that processing these kinds of plastics is inherently slow. The parts are formed by preparing a mixture of ingredients and waiting for them to cool or react chemically. For parts the size of automobile body panels, this process can take a minute or more. By comparison, steel parts can be stamped in less than 10 seconds. It is hard to find ways to increase the rate of chemical reactions or the rate of heat transfer-if plastic cools too rapidly it becomes brittle, and if chemical reactions are sped up they become difficult to control.

To make a large number of plastic parts, then, automakers would need to buy multiple machines and set up parallel production lines-steps that would more than offset the capital advantage of plastic production and increase administrative overhead. While parallel production lines may sound feasible in theory, they are very difficult to coordinate in practice. As a result, automakers have tended to avoid processes that require more than two parallel production lines.


How much weight could a plastic unibody save, and at what cost? The most radical polymer system is the Ultralite, a “concept car” based on carbon fiber composites that was developed by GM researchers given a mandate to obtain the highest possible gas mileage. The car, which was built by hand, incorporated a variety of weight- and fuel-saving technologies. Although the car was capable of getting more than 100 miles per gallon, it cannot be considered a prototype for a mass-market vehicle: it did not contain the space or safety features most consumers would consider essential and was never road- or crash-tested. Nevertheless, at 308 pounds, it represents the lightest auto body yet built of polymeric materials.

Although the Ultralite weighs about the same as an aluminum space frame, it would cost significantly more to produce in large volumes. At production volumes of 100,000, for instance, each Ultralite-style unibody would cost about $6,400. This estimate is based on the assumption that carbon fiber prices will remain at about $20 per pound. Proponents of polymeric materials have argued that the price of carbon fibers will decline as demand rises. But even if the price of carbon fibers fell to $5 per pound-a trend we do not foresee, since the production of carbon fibers is not necessarily amenable to economies of scale-the plastic unibody would still cost $3,500, compared with $2,500 for a steel unibody and $2,800 for an aluminum space frame at comparable production volumes. Moreover, at higher production volumes, the price of a steel or aluminum unibody will fall considerably, while the price of a polymer-intensive unibody will fall much less, making it an even less economically sound choice.

It is unlikely that the increase in fuel economy attributable to the body alone would make up for the higher cost of a polymer-based body. At prices of $1.20 to $1.50 per gallon of gasoline, the Ultralite body would still cost some $4,500 more than either a steel or an aluminum unibody over its life cycle. In fact, carbon fiber-reinforced polymer-intensive bodies would still cost about $4,000 more than steel bodies even if gasoline prices rose to $4.00 per gallon, as is the case in Europe.

What Manufacturers Are Doing Now

Given the state of manufacturing art, the automobile industry has been taking an incremental approach to the use of new materials, gradually adopting new applications of aluminum, polymers, and advanced steels. For example, Ford is working closely with several aluminum companies on a project called Concept 2000 to produce 20 to 40 all-aluminum Taurus sedans, which the company is now testing and evaluating. The vehicle, which uses a unibody design, is only a few hundred pounds lighter than its steel counterpart, largely because the project engineers did not change the powertrain or suspension or redesign the vehicle to achieve other secondary weight savings. The project was intended only as a test of the manufacturability of an all-aluminum car, with the goal of identifying the changes in forming technology that would be needed to produce it. It is not yet clear whether Ford regards the experiment as successful.

Alcoa and Audi have collaborated on the Audi A8, a luxury sedan based on an aluminum space frame that is being produced at low volumes and marketed in Europe. Much of the weight savings gained by the use of aluminum are canceled out by accoutrements intended to boost the car’s appeal in a high-end market. The vehicle does, however, demonstrate the viability of a design that utilizes aluminum extrusions and castings as well as the wrought sheet used in the panels.

The automobile industry is also attempting to develop production techniques to put plastics on mass-produced vehicles (notably GM’s Saturn car lines), but even here the plastic components are not critical structural elements of the vehicle. All Saturns, for instance, use plastic body panels to cover a steel space frame. Because they have no structural role, the panels are made not of reinforced composites but of ordinary plastics, which can be produced in quantities of hundreds of thousands. The choice of material is governed less by weight considerations than by cosmetics: plastic panels give the vehicle its distinctive shape and resist dents and scratches. In fact, the weight saving achieved by the use of plastic panels is at least partly offset by the need to use more steel in structural components to maintain the expected level of performance.

Automakers have found that, with an aggressive effort, they can substitute polymers for steel in a handful of major nontraditional applications, such as roofs, hoods, floor pans, and engine cradles, but many are also discovering that the costs are too high and the weight savings unimpressive. GM has also experimented with glass fiber composites on the body panels of its APV vans for a number of years but recently concluded that the material is just too expensive. The company plans to return to using steel.

While they continue to experiment with glass fiber-reinforced polymers in niche-market vehicles-a well-established platform for innovation-automakers appear to have decided that these materials are not useful in applications with production volumes over 80,000, because at these volumes the benefits do not justify the costs. Moreover, it appears that the industry is already using plastics in most of the applications that are best suited to the material’s strengths. Further substitutions of plastics for steel will be much harder to accomplish, because these are the uses that capitalize specifically on the properties of metals.

Another material that may play a role in incremental change is high-strength steel. The thickness of steel parts used in automobiles is usually determined by the degree of stiffness they require, but in about 20 percent of applications the important property is strength. For instance, a beam in every car door protects passengers in the event of a crash. New high-strength steel alloys are two to three times as strong as conventional carbon steel, so a beam made of the new material could weigh one-half to one-third as much as the beam used in car doors today. A number of steel companies based in different countries have hired Porsche Engineering Services to come up with a body design incorporating all the potential applications of lightweight steel. They estimate that the body could weigh 10 to 20 percent less than a conventional steel unibody, at a cost up to 15 percent higher.

The Program for a New Generation of Vehicles, meanwhile, is investigating the potential uses of advanced steels, plastics, and aluminum, as well as such exotic-and expensive-substances as magnesium and titanium. At this early stage, researchers are trying to identify the technologies that could make up the platform for an affordable advanced vehicle. They appear to be focusing their efforts on the concept of a hybrid diesel-electric engine, for instance, and on aluminum as the dominant material for structural applications (although the vehicle will undoubtedly incorporate a variety of advanced materials for other uses.) Whether or not the program ultimately succeeds in developing a vehicle that is affordable-and there are rumblings that insiders believe it won’t-the effort will give the auto industry valuable experience with new materials and technologies.

Concentrating on What We Can Do

Whatever strategy the industry adopts, a vehicle made of lightweight materials is clearly going to cost more than today’s conventional car. The fuel economy of these vehicles is also going to depend upon a lot more than the shift to lightweight materials; significant gains will require changes in consumers’ expectations. Given our assumptions about how roomy a car should be, how swiftly it should accelerate, how fast it should go, and how comfortable it should be to ride in, it is difficult to make a car much lighter than, say, the all-aluminum Taurus that will still be a vehicle most of today’s consumers want to buy.

Nevertheless, the specter of the supercar haunts the debate over carbon-dioxide-induced global warming and feeds public pressure for government to mandate more radical reforms. If we can make a better tennis racket out of Kevlar, the argument goes, why can’t we make a better automobile out of the same kind of material? One answer is: although consumers may be willing to pay three times as much for their advanced composite tennis rackets, they are unlikely to be willing (or able) to pay quite the same price premium for an advanced composite car.

A supercar like that envisioned by the Program for a New Generation of Vehicles-one that achieves 80 miles per gallon, maintains the same level of convenience, and costs the same as today’s car-is beyond our capabilities today and for the near future. Any two of these three objectives can be achieved today, but putting all three together will require major technological breakthroughs. It is thus impractical for the industry to jettison today’s automobile designs and technology to pursue this technological chimera.

Because we cannot mass-produce an affordable, ultra-lightweight polymer-based vehicle body, we should concentrate instead on what we can do. For instance, we can make an aluminum body that performs as well as the steel alternative but costs only marginally more. The incremental application of the broad spectrum of advanced materials technologies available today can yield real benefits in efficiency, utility, and performance without incurring insupportable costs. Although relatively unexciting and unglamorous, incremental strategies for vehicle weight reduction are the only credible approach for beginning the transition to an economical, fuel-efficient vehicle.

Tech Obsessive?
Become an Insider to get the story behind the story — and before anyone else.
Subscribe today

Uh oh–you've read all five of your free articles for this month.

Insider Premium

$179.95/yr US PRICE

More from Intelligent Machines

Artificial intelligence and robots are transforming how we work and live.

Want more award-winning journalism? Subscribe to Insider Plus.

  • Insider Plus {! !}*

    {! insider.display.menuOptionsLabel !}

    Everything included in Insider Basic, plus ad-free web experience, select discounts to partner offerings and MIT Technology Review events

    See details+

    What's Included

    Bimonthly home delivery and unlimited 24/7 access to MIT Technology Review’s website.

    The Download. Our daily newsletter of what's important in technology and innovation.

    Access to the Magazine archive. Over 24,000 articles going back to 1899 at your fingertips.

    Special Discounts to select partner offerings

    Discount to MIT Technology Review events

    Ad-free web experience

You've read of free articles this month.