Are liquid-fuelled laptops close at hand?

Level: Padawan
Required knowledge: Electrical power, resistance, chemical reactions, valency

Every energy-yielding electrochemical reaction is characterized by a transfer of electrons from a negatively polarized substance called the anode to a positively polarized substance called the cathode (the polarization is reversed if the reaction is energy-consuming). This transfer occurs when a suitable medium is available between the two centres for the electrons to flow through, a medium called an electrolyte. Sometimes, the electrolyte itself may become polarized because of the electrons flowing through it, but the eventual purpose of the reaction remains the same: if the electrolyte gets electrically charged, it acts as either the anode or the cathode depending on the way it's been polarized.

[caption id="attachment_22673" align="aligncenter" width="300" caption="Simple setup of an electrolytic cell"][/caption]

When a substance loses electrons, it said to have been oxidized and acquires a positive charge. When a substance gains electrons, it is reduced and acquires a negative charge. By these definitions, H⁺ hydrogen ions are strong oxidizing agents because they strips the atoms in their vicinity off their electrons and become reduced themselves (to H). Thus, an electrochemical reaction is often a redox (reduction-oxidation) reaction because the anode loses electrons to the cathode. Such reactions are handy because they can be, and are, used to power rechargeable batteries: the transfer of electrons between the electrodes generates an electric current that can be drawn out and wired to small machines... like laptops.

In a laptop, chipsets continuously perform hundreds of thousands of mathematical operations per minute in their tasks as the bridges between the microprocessor and other devices. During operation, they generate copious quantities of heat, enough to warrant a strong fan that has to keep running to prevent the heat from increasing the electrical resistance in wires (resistance increases with temperature if the wires are made of metal). If the fan stops, the system will either shut down or be damaged. The problems with the fan are that it is a big sink of electrical power coming in from the battery, and it needs cleaning and/replacement once every few years.

[caption id="attachment_22675" align="aligncenter" width="320" caption="Such quantities of dust accrued inside a cooling fan trap heat and prevent it from being expelled"][/caption]

Because of the needs of the ongoing tech. revolution, the implication is that more efficient systems have to be designed that work on lesser amounts of power, are longer-lived and energy-efficient. A fan, obviously, is none of these things. That's where an electrochemical cell can come in, according to a research led by IBM's Bruno Michel at the tech. giant's Zurich Research Laboratory, Switzerland. Michel's proposal is to dig in small channels within the chipsets through which a small quantity of an electrolyte containing vanadium ions is pumped in. Next, because the surface of the chips will be coated with a suitable catalyst, vanadium will become reduced while the catalyst will become positively polarized - in the process generating a small current that can be harnessed, just like an electrochemical cell.

After the reaction has occurred, the electrolyte - in this case, also the cathode - can be pumped out, making it a plausible coolant as well that carries the heat from the chipsets out of the system and into a nearby heat-sink. When the electrolyte is pumped back in, it can be electrolyzed using an external power source to replenish the vanadium ions before it comes in contact with the catalyst. The principle advantages of using such a cycle - encapsulated as a vanadium redox battery (VRB) - are that such cells have low discharge rates (making them suitable for long-term storage of electrical energy) a low response time, and large storage capacities (although this is applicable only for large installations because the volume of electrolyte required is very high, then).

[caption id="attachment_22677" align="aligncenter" width="457" caption="This VRB cell-stack can store 43.2 gigajoules of energy. The costs of setting up and maintaining a VRB are justified only for high-energy applications."][/caption]

The corresponding excerpt from an article that appeared in The Economist in its Technology Quarterly edition (Q1), 2012:

Dr Michel’s proposal is to use the ... system to power the chips, too. Instead of a single network of channels running through a stacked chip of this sort, he suggests there should be two. Each would carry a fluid doped with vanadium ions, but those ions would be in different oxidative states in the different channel systems—at least, they would be when they entered the chip. At the heart of the device, however, the channels would be lined with a catalyst that reacted with the electrolyte and also acted as an electrode.

When the fluid was pumped around the chip a piece of chemistry called a redox reaction would take place, as one sort of vanadium ion gave up electrons to the other sort. For that to happen, the electrons would have to pass between the channel systems via the wiring of the chip. This would produce a current to power the chip. The electrolytes would then be pumped out of the chip, carrying heat away with them. Once cooled, they would be reinvigorated back to their original ionic states by an external electric current, and pumped back into the chip.

However, there are many issues associated with the VRB that are not immediately evident. First, because of the quantity of heat being generated by the chipsets, the amount of electrolyte required will be quite high. Second, because a large quantity of electrolyte will be pumped in and out of the system, the frequency of electrochemical reactions occurring will be higher, which will in turn quickly deplete the catalyst, necessitating continuous and troublesome replenishment. Third, because a VRB has a low energy-to-volume ratio, even high volumes of the electrolyte will produce (only) slightly higher "volumes" of energy, giving it a low thermodynamic efficiency as a coolant. In fact, the energy density of a VRB is about 25 Wh/kg, whereas those of lithium-ion batteries are 80-200 Wh/kg (although researchers from the Fraunhofer Institute for Chemical Technology, near Karlsruhe, Germany, announced a significant upgrade in 2009 which is yet to find deployment).

[caption id="attachment_22678" align="alignleft" width="214" caption="The four vanadium ions"][/caption]

The fourth and most important reason has to do with the four oxidation states of vanadium ions: V²⁺, V³⁺, VO₂⁺ and VO²⁺. All four states are extremely acidic (because they are strong oxidizing agents) and will therefore require special plumbing within the laptop lest they etch into important circuitry and destroy them.

I'm only pointing out these things because the The Economist article ends on a very hopeful note, that if some other electrical wiring problems (not discussed here) are fixed, we could soon have "a new type of liquid-fuelled computer". I'd think the day is really far off when, despite the demands plaguing their design and production, lithium-ion batteries and cooling fans can be replaced by liquid coolants and flowing sources of power. At the same time, while this might seem like a step in the right direction if only for its environmental implications because VRBs consume less net power than a fan would, the production of sulphuric acid would shoot up. The strong acid is required for the manufacture of vanadium-ion-impregnated electrolytes.