Thursday, April 7, 2011

Time to Think About Time


Time to Think About Time

Beware of stopping to think about time and inspecting its relevance in philosophy, logic, science, history, administration, religion, or politics. On the one hand it is a confusing field, of vital importance in all sorts of things on all sorts of scales; on the other it is a snare for anyone who wishes to make sense of it in unsuitable connections or on invalid or irrelevant assumptions.

Commonly the most pathological nonsense emerges when obsessive personalities try to apply abstract metaphysical concepts to practical problems; they effortlessly progress from nonsense to nonsense without so much as coming within sight of either the sublime or the mundane on the way. It takes an uncommonly strong stomach (and commonly an uncommonly weak head) to read some of the history and the debates concerning the spiritual significance of religious calendars for example; compulsive zealots spend their lives blinding their bodies (their minds went blind shortly after they first took up their scriptures) poring over the texts and smugly yielding non-sequiturs from tawdry metaphysics.  As Shakespeare had Julius Caesar say:
“Such men as he be never at heart's ease
           Whiles they behold a greater than themselves,
           And therefore are they very dangerous.”

Perhaps even more appositely, ‘twas Kurt Vonnegut who said:
"Beware of the man who works hard to learn something, learns it, and finds himself no wiser than before," Bokonon tells us. "He is full of murderous resentment of people who are ignorant without having come by their ignorance the hard way."
Just questioning the views of such a man, let alone contradicting him, is to open yourself to poisonous vilification, and sometimes even to take your life in your hands. And please do not think for one moment, either that the debates have been settled in any rational way, or that you can safely select any particular sect as being immune to such nonsense. The more blindly devout, the greater the nonsense and the greater the danger, often the actual physical danger, of entering into debate on the matter. People have died in debates ostensibly over lunar calendars and year lengths, and whether elephants can jump, and what time of the day which day started or which year ended, and which thus denoted day is the right day to let your slave cheat or eat the stranger within your gates. And many of those who died did not even know there was a debate, let alone what it was about. 

It really is not funny, when you get into it; it is a sadness to realise the intellectual and ethical depths of rationalisation and rage to which dedicated minds, sometimes quite good minds, can permit their preconceptions and bigotry to sink them.

At the other extreme, in science and technology people work on scales of measurement of time beyond the imagination of the unassisted minds of the religious. I have yet to encounter among the spite-driven text-quibblers, any sign of true comprehension of deep time, such as we see it in palaeontology, astronomy or cosmology. "From everlasting to everlasting" is an empty phrase coined by peoples whose temporal horizons were measured in lifetimes, or occasionally, vaguely, in millennia. As for the religious significance of Planck time, the less said the better.

The interesting thing is that to the layman in scientific matters, such time intervals seem nonsensical; and why? Simply because they cannot conceive them. And yet to those silly scientists working with meaningless time intervals, milliseconds are as meaningful as seconds, and femtoseconds as meaningful as microseconds or petaseconds.

Somewhere in there, there should be a lesson, I cannot help thinking...

However, to move from the obscenely ridiculous to the sublimely ridiculous, let us contemplate a few hardy perennials in contemporary politics and administration.

Times and Lines

These matters were not always ridiculous, in fact all of them are surprisingly recent concepts and through most of our intellectual history most of them never emerged at all. For one thing, they depend on comparatively recent technological and scientific advances.  In earlier centuries they simply did not arise at all as day-to-day concerns, and at most rarely as abstract intellectual concerns — angel on pin stuff.  Nowadays however, we not only have the technology to measure time with greater precision than our everyday activities demand, we also have the ability to contact each other pretty well anywhere on the planet at a few seconds notice. In an age when it took a good sundial and a competent reader of sundials to tell the time with a probable error much smaller than an hour, and when it would take special, expensive arrangements to get a message say, from London to Bristol in less than two days, or to New York in less than two months, the very concepts we discuss here would hardly have meant anything.

Time Zones

Let's begin with the concept of time zones. Now, one could hardly get more down to earth and obvious than that, could one? Time zones are not only a practical consideration, but practically inescapable, right?  The sun does after all move round the Earth (or is it the other way round?) in twenty four hours, right? If we accept one hour as a time zone, that gives us 15 degrees per zone, right?

The concept certainly escaped most people in days when pocket watches did not exist, and even after the marvellous creations of John Harrison and Thomas Tompion, it would have been very rare for people in different towns, or even in neighbouring villages out of earshot of each other, to synchronise their clocks. Almost throughout history people have been without precise time measurement, and even when reasonably accurate and reliable timepieces were no longer objects of wonder, practically every village kept its own time. And time everywhere amounted roughly to local solar time. To the best of my knowledge hardly anyone ever imagined time zones. Susan Glaspell and George Cram Cook referred to such ideas in a light, satirical one-act playlet called "Tickless Time". It opened in 1918, and as such was well behind the time, technologically speaking, but it was good-humouredly thought provoking.

The main impetus behind the synchronisation of timekeeping in various villages and towns around the world, was not the requirement for everyday local schedules, which managed very nicely for most purposes, thank you; it was the introduction and growing significance of electronic communication (which in those days amounted to telegraph!) and high-speed transport. Railway schedules became more than just matters of neatness and prestige; they became matters of  money – and life and death – in fact probably thousands of people have died because of timing errors on railways.  

But affordable watches became more accurate and soon his pocket watch was as characteristic of the train guard as his uniform. Not only was it possible for different towns to keep tolerably synchronised, but for different towns to keep their own times, though indeed, largely different times. Each town understandably preferred to fit its own time to its own activities.  This largely meant that towns kept solar time according to the local noon, something that they each could check on independently, and needed no telegraph for.

Around the world, bit by bit, that largely came to an end, though the ending took a long time. From the first faltering suggestions in the mid-19th century, it was some 70 years before the implementation of time zones became effectively universal.  Even a rotten idea, like chopping the planet into 15 degree slices, may be slow to catch on.

The problem that time zones addressed, consciously or not, was coordination of times, both between dates and daytimes, and between neighbour and neighbour. Time zones have precious little to do with scheduling local daily activities. Some countries did their best to upset the apple cart; They asserted the personality and independence of their savants by defiantly and daringly adopting half-hour offsets, and a few even used quarter-hour offsets — there is no accounting for... mental confusion, if I might put it so delicately.  The problem that fractional hours aggravated was that people confused the time shown on their clocks as local time with “real time” or something like that. The time to get up was say, six in the morning, and the time for shops to open might be say, eight. Not surprisingly,  this led to great inefficiency, because six in the morning might be dark at one extreme of the country, but broad daylight at the other, not to mention seasonal changes. If there were inconveniently placed mountain ranges in the country,  that could add more than an hour to local discrepancies.  

A hint at the right way to do things was demonstrated by China. After the Communist revolution in 1949 the entire country adopted Beijing time. Note that, partly because of its closeness to the North Pole, China spans no less than five time zones; possibly more if one counts the islands it lays claim to.

Forcing five time zones into one may sound terrible, but enforcing that single time zone was one of the most constructively and benignly effective acts of that government. Even politicians can achieve great things if they get confused enough! And did China collapse in the confusion of having just one time zone? Hardly! In areas in which Beijing time was inconvenient, people simply adjusted their working hours accordingly. If the clock said that it was 10 AM when day dawned, then people in such a town got up about 10 AM, perhaps four hours after Beijing. To them breakfast at 10 did not seem strange or inconvenient, and why should it? Seven in the morning is a mark on the clock, not the definitive time for breakfast. If everything else in China had worked so simply and constructively, we would all be speaking Chinese by now.

All the tools had been in place to avert the time zone problem since the mid-1880s after the International Meridian Conference of 1884 had established the Prime Meridian at Greenwich and formulated and proposed the adoption of the Universal Day.  If we had all dropped the idea of time zones and used UTC 24-hour time, as the Chinese now use Beijing time, there would have been enormous economies in administration and reductions in confusion world wide. No AM/PM confusion, no doubts as to the actual time anything happened anywhere on the planet.  

Daylight Silly Time

Now, if it had been the case that the time-zone system had obviated all problems dealing with everyday timekeeping, I should have had nothing more to say, but in fact we have no end of niggles with our time zones, sometimes dangerous and expensive niggles.  These niggles are so widely unsatisfactory that they have given rise to another idiocy of unbelievable ingenuity: Daylight Saving Time!

Could you believe it? Some genius observes the horrifying reality that in some places the sun does not rise at the same time every day... This means that on some days we use our natural light less optimally than on other days...
Horrors!
Something Must Be DONE!
But what?

Illuminate the whole planet? Tempting no doubt, but...

Rise and shine at different times during the year? Let shops open an hour later in winter than in summer? Ridiculous! Impossible! Get a bigger ‘ammer!

Oh well, then let us rise at the same time every day all year round. That should do it! To solve the insolation problem,  all we do is change time instead of changing times. Reset our watches to  a fictitious time two to four times a year! What could be more logical?  If the farmers don’t like it, tell them to bully their cows into compliance.

Resetting clocks would save everyone from, say four times a year, deciding to set the alarm and open the shops perhaps half an hour earlier or later, or maybe twice a year in zones nearer to the equator, such as Brighton.   Nowadays we could have our clocks automatically programmed to reset the alarm without danger of  confusion or misunderstanding, or clashes between neighbours in neighbouring time zones or neighbouring hemispheres.

What’s my line? International Dates?

Now, another idiocy is less official than Daylight Saving Time, though even more abject: The International Date Line! Actually it has no official standing at all, though unofficially it tacitly is widely recognised.  The International Date Line, much like time zones,  is the result of confusing days with daytimes.  Like most products of confusion it is hard to make sense of the line in hindsight.  The idea of the International Date Line always was ridiculous, but by now it has become unbelievable! Every country has the right to declare that the line does not pass through its territory, and most recently (1995)  Kiribati shifted the line in the most complex pattern so far. If you were to take that shift seriously, it would be possible to change your date, not just by travelling east or west, but by crossing the equator! Or if you prefer something less exciting, you can instead cross the date line at least three times in succession on the same great circle course. Maybe more if you are sneaky. You could do it in a couple of hours if you travelled by air.

The Bermuda Triangle and Sargasso Sea have nothing on the International Date Line, if you ask me!

Now why do we need an International Date Line at all? If we used a Universal Day, the very same form of universal day proposed in 1884, we would not need any International Date Line! Anyone wanting to know what day it is, and what time of that day it might be, could consult a watch; wherever we were on Earth, whether North or South Pole, equator or ocean, any latitude whatsoever, including 180 degrees near the Bering Strait, or 0 degrees near Greenwich, whether up in the air or down a mine, winter or summer, midnight or noon; that watch would tell us the date and the time.  And to within a fraction of a second, that time would be UTC time, the same as Greenwich. What could be simpler? (Did I hear someone say: “The brain of a politician?” Go and stand in a corner; if politicians had brains, we would not have the problem!)

Just a minute, you might reasonably object, that would be fine if I happen to live in Greenwich; I even would not mind resetting my alarm with the seasons, but everywhere else, every day as measured by the sun, I would have two dates as measured by the calendar.

This is true up to a point, but it is neither an unfamiliar type of problem, nor at all a serious problem in practice. Consider: even as things stand in our current ridiculous system, for practically the whole of every day, there are two different dates in various places around the world. Take off in your private jet plane and follow the line of midnight around the planet from hour to hour (and notice that the line becomes pretty confusing in some places, given our current time zones!) You then will find that you have one date to the east your time zone, and a different date to the west. If that is not confusing, what is? And on the same planet at the same time, as measured by your portable atomic clock, whether it is midnight or not, what is happening to the date at the International Date Line?

Having embraced a system like that without objection, who are we to sneer at a more rational system? In particular, a simple convention would adequately solve the two-dates-to-the-day problem almost without thought in all longitudes far away from Greenwich. Stupidity is optional; we have no need to confuse dates with daytimes.

We simply accept some convention (there are several alternatives according to taste) such as that at any point in any context where the contrary had not been stated, the local daylight would be taken as spanning the date that applied at the time of dawn. So for example, if I hire a labourer on a temporary basis, then he gets one day's pay for work done starting at sunrise, even if we happen to be in Hawaii, where the problem would be nearly at its worst. In much of Europe and East-Coast America the problem would hardly be noticeable. Under circumstances where that expedient does not suit, one simply mentions UTC when specifying the date as appropriate, much as people in the United States mention the state when speaking of a city; “Birmingham, Alabama”, “Memphis, Tennessee, and so on. That is all very simple; in comparison the complications of our current time zoning system are positively labyrinthine.

In writing the foregoing I did some homework and you might find it instructive to consult both Wikipedia and “International Conference Held at Washington for the Purpose of Fixing a Prime Meridian and a Universal Day. October, 1884.” That document is available on the WWW at Project Gutenberg. I wasn't there at the time of the conference. Perhaps it is just as well. They had some clever people there, but the voting majority would not have listened to me. I notice that at least a couple of delegates liked the “Standard Day” idea, so it cannot just be me!


Wednesday, January 5, 2011

Stop Mucking With Geothermal



Stop Mucking With Geothermal

Table of contents

Before reading on, please note:
The Theme
Geothermal Schemes in General
The "Power Bubble" Concept
Power Bubble Bootstrapping, or "Why Bother?"
Power Bubble Applications
Drilling Wet or Drilling Dry?
Power Bubble Politics
Technical alternatives to Power Bubbles
Objections and Concerns
Adverse Effects of Salt Injection or Disposal
Blowouts and Pressure Excursions
Bubble wandering
Drilling in Hot Rock
Impossible Pressures
Impossible Temperatures
Bursts and Seismic Effects
Shaft Collapse
Quakes and Geological Flaws
So...
What to aim for
What to do

Before reading on, please note:
This is not a document cast in stone, but a non-technical draft proposal.  Anyone with doubts, queries, suggestions, objections, or corrections should please feel welcome to contact me.
If you do so, please let me know whether your correspondence is in confidence, or you would like to be credited with any changes or additions that you might inspire. 
Of course there also would be no hard feelings if you preferred to write an independent document of your own without consulting me. 

We know very little, and yet it is astonishing that we know so much,
       and still more astonishing  that so little knowledge
               can give us so much power.
Bertrand Russell
This essay remarks briefly on geothermal power as a range of resources that in most forms are intrinsically so transient, so feeble, and on so small a scale, as to be mainly of local interest.  It then deals at greater length with certain strategies of exploitation of particular classes of geothermal energy resources that diverge from that pattern.  In their implications and objectives they differ relevantly from any geothermal schemes usually contemplated.  They are the only ones that bid fair to solve global energy problems and radically transform international energy politics and economics.  They offer prospects on a scale exceeding that of the organic fossil fuel industries.  
The implications are of critical importance in determining the nature and scope of the long term future of human energy usage on the planet.  If the proposals are successfully implemented, our current usage patterns of organic fossil fuel consumption will be obsolete.  The international political consequences will be enormous.  Successful implementation obviously must change the face of industry in many ways.  For one thing we could belatedly begin to use our carbonaceous fossil materials predominantly as chemical feedstocks instead of simply burning them.  Burning fossil carbonaceous material (and oxygen with it) is about as sensible as burning money.  For certain currencies it may be a good deal less sensible yet.  
I realise that some readers will be puzzled to see that I apparently do not know that fossil petrochemical feedstocks have been used for other purposes than fuel for over a century.  I hasten to reassure them that I am thoroughly aware of it.  What they may not realise is what a tiny proportion of the material (a few percent at most; in many countries a fraction of a percent) is used as anything but fuel.  We burn a staggering proportion of our birthright for far less than a mess of pottage, but a great mess all the same.  
Existing or projected geothermal schemes are of several types.  You may read some accessible articles on the subject in Wikipedia, and of course thousands of items of assorted levels of reliability and rationality may be found through various search engines.  Books on such subjects also are no novelty.  I do not discuss such material in much detail here.  From the point of view of this article most of the various conventional geothermal options differ mainly in their duration, scale, and intensity of production of energy.  Some, such as the use of heat pumps to extract heat from warm ground, are only of local interest and many produce energy of marginal quality at a marginal profit.  Some even work at a loss, either thermodynamic or monetary, and those are justified only by special circumstances or special needs, rather than by primary energy production as such.  Some amount to heat storage or heat conditioning rather than heat production.  
Other installations running on low-grade energy certainly are prime power producers, but at marginally useful temperatures, barely above the boiling point of water. Some actually work at still lower temperatures.  Those are of little interest here.  Another class, such as "hot dry rock" sources, also are of little interest here.  They do produce heat at attractive temperatures, but their scope of production is local and their duration is modest.  More valuable sources of heat, such as in Iceland, variously produce hot water, steam and so on.  
I mean no disrespect to local schemes of modest scope and function.  They often meet needs of great importance, but such is not the theme of this discussion.  What I have in mind is energy production of global significance for centuries or millennia at least.  The scale of energy production that I urge is to be of major national or international importance, for periods of many millennia at least.  
The objective is to establish an energy regime at least as significant as those based on all fossil organic fuels combined, and eventually to replace them for the foreseeable future.  
For really high efficiency and versatility one needs high temperatures and indefinite resources.  Consumers like Iceland can put up with low efficiency because they have plentiful supplies and modest consumer requirements.  That is very well in its place, but it is not the subject at issue.  For a sense of perspective, consider that the current global power crisis is doing a lot of harm in Iceland as well.  
It will occur to the reader that there is increasing interest in hot magma geothermal sources.  They do meet the  need for high temperatures (several hundred K, sometimes over 1000K), but not even these meet the requirements of this discussion.  Some such schemes have great implications, of which not all are primarily concerned with energy production, but they still are not particularly relevant to the subject of this essay.  They do not all lend themselves to indefinite production, or indefinite expansion.  
In particular they do not achieve what I shall call "bootstrapping".  By this I refer to the ability to use part of the energy production directly to increase the yield of a scheme indefinitely. 
The point of this essay is to introduce a concept of energy production that could attain all those objectives.   It must do so on a basis of technology that is feasible, if in some ways ambitious.  It could achieve certain other objectives as well, but that too is beyond the scope of the essay. 
As it happens, the source of energy I propose is in fact geothermal, but that does not imply that it simply is an elaboration of traditional geothermal energy schemes.  To argue that would make about as much sense as to regard the internal combustion engine as an elaboration of coal burning on open fires.  
Readers might occasionally wonder whether I have considered concepts such as various connotations of the term "hubris".  If they do not, then to that extent at least I have failed in my objectives. 

The general approach of most geothermal schemes is to drill down to some hot material and extract the heat for use.  Typically extraction takes the form of pumping up subterranean liquids or vapours, or, increasingly often, injecting fluids and extracting the heat from them.  
Note that in these contexts I use the term "fluid" to denote substances that can flow appropriately, whether liquid or gaseous.  
Typically it sooner or later becomes necessary to drill extra holes so that cold fluids pumped down one hole can be extracted hot from other holes.  Such multiple hole designs may or may not be appropriate to the proposals under consideration in this discussion.  
So far this description of geothermal schemes is largely consistent with the proposal below.  Certain principles are pretty well common to any geothermal work, and some of the common ones follow. 
Readers no doubt accept that:
·       Ground temperatures increase more or less progressively as one digs down.  At a sufficient depth the rocks become hot, even plastic or actually molten (magma).  
·       The depth at which the temperatures become high enough to melt rock into magma is variable, though in the absence of actual volcanoes, it is usually at some tens of km.  
·       Sometimes magma rises and forms isolated pockets without any prominent connection to deeper sources of molten material.  Sometimes there still is a connection to the source, but narrow enough to be trivial.  In such cases the amount of heat one might extract from such a pocket is practically limited to what already is inside the pocket.  Such limited deposits may be rewarding, but they are not the subject addressed in this essay.  
·       Within manageable limits the higher the temperature of a heat source, the the greater the efficiency with which one could exploit energy from that source, and the greater the versatility with which one might use it.  For instance, pressurised air at temperatures approaching 1000C can be sold either for generating electricity directly, for metallurgical purposes, or for generation of hydrogen, and so on.  Alternatively air that is too hot for other purposes can be diluted with cooler fluids as part of the heat extraction.  In contrast, fluids at 100C can hardly be used for more than district heating and inefficient generation of electricity.  
·       Depending on their location and nature, magma temperatures typically are in the range of about 700C to 1200C.  Rock might be plastic within similar temperature ranges, depending on the pressure and their constitution.  Factors that affect the rate of energy transfer include the temperature of the rock,  the area exposed, and the thermal conductivity of the rock.  By flow in general, and convection in particular, fluid rock can transfer heat over significant distances (metres or more) dramatically faster than solid rock can transfer heat by conduction.  
·       The behaviour of rock in very hot regions under high pressures is strongly hydrostatic.  It exerts powerful effects of buoyancy and convection etc.  Where the materials are cooler this remains true, but there the hydrostatic effects are not pronounced enough for us to take it into account in most connections.  For example small-scale convection in cooler materials is negligible.  

 ·    Even where rock is effectively solid, it is elastic to a degree that can show dramatically in the curvature of masonry pillars under stress for example.  (See the brilliant book by J. E. Gordon: "Structures, or Why Things Don't Fall Down".)  The combined effects often are of great importance.  For example, where rock is placed under stress by overlying ice or other overburden, it sinks, eventually to isostatic levels.  When such overburden is removed the rock then slowly rebounds.  Finland for example still is rebounding from its most recent period of glaciation.  Conversely, where underground fluids thrust up under pressure, they cause the rocks above them to bulge.  Usually this is hardly perceptible, but sometimes they bulge dramatically before they burst out as volcanic eruptions.  Such bulges are among the most significant intimations of immanent eruptions. 

·       The elasticity of hot rock has important implications for deep engineering techniques that depend on the displacement of rock by pressure.  If one digs down to thoroughly plastic or marginally molten rock, one can apply hydraulic pressure to form a bubble or a fracture, depending on the nature, pressure, and temperature of the rock.  Even rock that one would hardly imagine to be capable of distortion can deform plastically under such conditions.  For example, if surrounded by sufficient pressure, apparently solid marble can be deformed amazingly without cracking, almost like putty.  
·       Conversely, as a rule the higher the temperature of any rock that does not decompose under the ambient conditions, the more easily it flows.  This is a familiar effect to anyone with experience in glass blowing, though of course, it is dangerous to extrapolate too simplistically from experience of a homogeneous material like glass, to heterogeneous, partly crystalline, materials such as molten rock mixtures that might fractionate and react chemically. 

Suppose that we have drilled down to thoroughly plastic rock.  We apply internal pressure by injecting gas or fluids such as water (which of course will quickly become supercritical).  Possibly we even might use liquids such as molten salts and similar compounds and fill the shaft with them to counteract back-pressure.  If we pump in fluids at sufficient pressure, a bubble results.  The shape and behaviour of the bubble depend on the texture and buoyancy of the rock and the injected fluid.  There should be little obvious distortion on the surface, because the depth of the bubble is several km at least, so the surface bulge would be spread over a wide area.  
The inner surface temperature of such a bubble would be roughly 700C to 1700C, depending on the nature of the surrounding rock, the depth, and so on.  Large surfaces at such high temperatures can be used as a source of enormous amounts of power for long periods of time.  For example, energy may be extracted thermoelectrically or by pumping working fluids suitable for driving electric power generators through the bubble.  So in this essay I refer to them as "power bubbles".  
As the bubble grows, especially if it is partly or even completely filled with dense molten salt or possibly molten alkali, or with a puddle of iron, the tendency to expand downwards  will compete with the tendency to expand upwards.  However, the heat flow generally will be from below,  so that especially if the molten salts act as a flux, that would increase the tendency for the bubble to expand downwards, because hotter rock deforms more easily than cooler.  

The implications will be familiar to anyone with experience of glass blowing.


To increase such downward tendencies in the expansion of the bubble, the sides of the growing bubble could be cooled to increase their viscosity.  The most convenient means of doing this might be by spraying them with (relatively) cool incoming working fluids or powders.  
The eventual effect would be to produce a bubble in the plastic rock.  If all worked perfectly, it would progressively expand to force its way through the yielding material downwards into deeper, hotter, softer rock.  Ideally this should continue until eventually its lower surface reached completely liquid rock.  However, such depths probably would only be practical in limited circumstances, such as near a mid-ocean ridge or a hot spot.  
The pressures required for bubble expansion at depths of tens of kilometres would be in the range of mega-pascals.   Currently such pressures are non-standard in industry, but certain approaches could change that.  
Supercritical fluids such as steam could be bled off from the bubble to yield heat and high pressure, much as existing very hot sources do, only more so than most.  They could be used to drive generators and machinery, supply heat, and extract industrially valuable dissolved substances. 
The best form in which to supply fluids to the bubble would depend on the requirements and the processes to be serviced.  Water, either fresh or brine, would be an obvious candidate.  Residual salt from the brine should act as a flux in contact with the softest, hottest rock.  Other candidate fluids include CO2, air, and molten salts from other sources.  Sodium chloride, which is about as cheap and plentiful a salt as any, has some quite interesting properties in such applications.  It becomes especially cheap when it is available in the form of raw seawater or of highly salty ground water that renders soil unusable until it can be removed.
Whether to use hot material direct from the bubble, or extract energy via heat exchangers, would depend on the practical factors associated with any particular installation.  
A crucial attribute of the bubble scheme is that once a bubble is well established and yielding power at a sufficient rate, it can be expanded continuously simply by using part of its power yield to continue indefinite inflation of the bubble.  On the venerable principle of assuming a spherical cow, we safely and conservatively may assume a spherical bubble, because, short of various disastrous outcomes, a sphere is the least productive shape for extracting heat.  
In practice a highly spherical bubble is vanishingly unlikely of course, because hydrostatic factors would tend to flatten the bubble drastically, if nothing else did.  
Two examples of disastrous shapes include:
·       nodes that pinch off or extrude sideways to such an extent as to be effectively excluded from the heat extraction flow, and
·       parts of the bubble floating up to well above the nozzle of the drill hole. 
Such effects would be wasteful at best, and possibly destructive.  Continuous control of the hole shape might be necessary.  Selective heat extraction and directing the injection of working fluid so as to cool areas where high viscosity is desirable, should be a promising and economical approach.  Elsewhere I also describe how to encourage continuous downward growth with added fuel.   
At all events, assume a spherical hole with a volume of perhaps 1 cubic km, giving a wall surface of about 5 km2.  Assume a wall temperature of 1000C.  This is a ridiculous oversimplification of course, because in real life not only would the wall be all sorts of shapes, but the wall temperature would vary considerably over its surface.  However, any deviation from a true spherical shape would improve the surface to volume ratio.  The only concern would be to prevent the resulting cavity from developing deep bulges that would frustrate good heat exchange.  
Consider in contrast another academic example.  If the 1 cubic km sphere were flattened to a disk with an average thickness say, of 10 m  it would have a surface area of some 200 km2, about forty times better than the sphere.  Without taking that figure too seriously, it still is plain that less compact proportions are more attractive than spherical bubbles. 

It is certain that many people who read as far as this will be objecting that the whole scheme, apart from being a technologically inane pipe dream, is just a dusting-off of the weary, limited, and ineffectual theme of geothermal power.  Without digressing to point at Iceland or other geothermal success stories, I point out that just because a given system relies on power from underground, does not imply that it is identical to other technologies that draw power from the same direction. 
By way of helpful illustration, a monkey might sneer at a proposal for an Atlas rocket.  "Firstly," he might say, "we already know about going up.  We monkeys have been going up trees for millions of years.  This rocket thing of yours obviously is just the same as climbing trees, but noisier, more expensive and plainly impossible anyway.  And besides, even if it worked, what is the point?  Who wants to go to the moon in the first place?  It will never pay.  Rather gather the low-hanging fruit!" 
To such a reasonable monkey it is difficult to point out that not only would climbing a tree not get anyone into orbit, but getting into orbit enables us to do many more marvellous and worthwhile things than going to the moon.  The profits from either weather observation or communications alone, however unromantic, could pay handsomely for the whole of our space investments so far.  
Similarly, power bubbles may resemble existing geothermal technology in no more than a sense in which rocketry might be seen as analogous to tree climbing.  The most important difference is that not one of the "conventional" geothermal technologies takes the bootstrapping concept beyond trivial levels.  Consider hot rock heat extraction for example.  If all goes well (which understandably it sometimes doesn't) we drill down to hot rock, generally creating at least two shafts.  We then interconnect the shafts by cracking the intervening body of hot rock.  As a rule one uses hydraulic pressure to do so.  Then, for as long as circumstances permit the cracks to stay open, one pumps the working fluid down one hole and up the other.  
This may sound a little unpractical, but it really does work, sometimes quite profitably.  What is more, it has a very interesting implication.  A non-obvious opportunity arises from the situation once the hole is established.  As the rock gives up its heat, why should one not apply more pressure from time to time, either to extend the period of viable yield, or to increase the rate of yield?  
An excellent idea; so excellent indeed, that this does get done in practice.  It is a sort of low-grade bootstrapping.  Get your first hole going at some acceptable level of performance, then use part of your new asset to increase its own scale.  
Now, hot rock, and every other form of geothermal energy extraction, is limited to obtaining such energy as is available from the immediately surrounding hot material.  Such energy may well be not only profitable, but plentiful, and more strength to the industry, say I.  However, the accessible energy still is limited.  In most sites one may estimate the expiry date with a reasonable degree of accuracy. 
 
In contrast consider the idealised power bubble.  Yes, it is dug into very, very hot material with very, very large amounts of energy available in forms that offer far wider ranges of applications than conventional geothermal.  But those are matters of detail rather than the essential contrasts.  Yes, its own power can be used to bootstrap its size.  But that is the end of the resemblance to other technologies that lend themselves to bootstrapping.  (Most geothermal technologies do not lend themselves to anything of the type at all!)  
When one can blow bubbles into what amounts to soft, flowing rock and grow the bubbles to huge size, important qualitative differences emerge.  Firstly, one can aim for a scale of production that can serve whole countries.  As opposed to mere profit, this has implications for political power.  (Ask the OPEC leaders!)  
From the point of view of the investor the initial investment into a working installation goes on increasing its yield indefinitely by balancing the power offtake constructively.  Simply by controlling the rate and choice of materials injected, together with the rate of power extraction, one keeps the bubble growing and exposing more hot material.  All it takes to complete the process is competence in managing the shape and state of the bubble.  
Growing the bubbles sounds good, but what about the supply of new heat?  Not even power bubbles violate thermodynamics.  
True.  However, when the bubble has penetrated far enough into rock that is fluid enough to exhibit significant convection on time scales relevant to the process, simply cooling the bottom of the bubble by extracting power will cause it to sink into hotter rock beneath, exposing new hot surfaces indefinitely.  
In fact, it suggests horizons of power extraction at depths and temperatures that currently would seem nonsensical.  
Infinite power?  Certainly not.  Eternal power?  Of course not.  Free power?  Definitely not — not even nearly free for some time, and who at this time can guess what the running cost may be?   
Do remember old forecasts of "power too cheap to meter".  This turned out to be a hopelessly over-optimistic idea.  And yet nuclear technology did prove to be clean and economical.  All the same, clean as such power has turned out to be, bubble power promises to be even less polluting.  In fact, any "pollution" it produces might well be worth extracting as ore or raw materials.  The calcium, magnesium, aluminium, silicon, and trace elements extracted from material pumped out, could provide your camera lenses, window glass, electronics, metals, and building bricks.  
Though neither infinite nor eternal, bootstrapped power production could serve the planet for longer into the future than our species has been clearly human in the past.  If by the time that bubble power begins to tail off we humans have not found our way off the planet, maybe we do not deserve to.  
It all is quite enough to make bootstrapped power worth serious thought and serious effort. 

There are more applications and implications to bubble power than appear at first sight.  This follows from the sheer pervasiveness of power in our civilisation and its infrastructure.  The only other resource that even approaches the ubiquity of power is information, which in many senses is related.  For one thing, we are by now conditioned to think of power in terms of electricity, or in terms of fuel for transport if we are a little more sophisticated.   

Think of the effects of altering the costs or availability of either.  
Certainly electricity is our most versatile medium for conveying power.  If we were to replace most of our power generation with power bubbles, it is practically certain that the primary product would be electricity.   
There are other products of course.  For one thing we need energy for chemical processing.  The layman cannot for one moment conceive the sheer variety of chemical products in our lives, let alone the variety of the nature of chemical processing.  Stop chemical processing today and you would starve, paralyse, and poison nearly the whole human world within months.  For one thing you could say goodbye to the production of our fossil transport fuels or the production of metals ranging from lithium to actinoids.  Some people seem to think that one simply pumps diesel and petrol out of the ground (those rare spirits that think at all) but of course that is not quite how things happen.  
Much of the energy we need to drive chemical and many material manufacturing processes is in the form of heat.  Electricity can supply heat with great efficiency, but if we first have to generate that electricity from heat, the overall efficiency drops precipitately.  Accordingly, industries that use a lot of heat are only too happy to buy hot air or other suitable gases at about 1000C, instead of electricity or fuel with which to heat the gases themselves.  Steel-working plants, cement producers, and nitrogen fixation are just three examples of power-intensive processing that could drastically reduce their electricity consumption if they had adequate sources of cheap heat.  
Also, processing of fossil organic material such as tar sand or coal to produce fuels, requires huge amounts of heat.  It requires so much heat in fact that sometimes the resulting yield hardly is worth it in absolute terms.  The real profit in such cases is not in the amount of energy processed into usable transport fuels, so much as in the form of fuel produced.  Not all fuels are of much use for transport. 
Other forms of energy consumption that rely largely on heat or electricity or both, include the production of fuels such as hydrogen by the sulphur-iodine or hybrid sulphur cycles.  

High-temperature electrolysis for production of hydrogen also is efficient, and would have the advantage of producing oxygen as well.  
In general, if we can get heat sufficiently cheaply and plentifully from below the earth, the hydrogen economy might become as familiar as the petroleum economy is today.  Actually, there are other fuel chemicals, such as metals and alcohols or other organic materials, that in some ways would be more valuable than hydrogen, and they also could become economical to use on a large scale.  The details hardly matter.  The point is that, given sufficient clean energy in useful forms we can change the nature of our effect on the planet and in the process we could extend our viability as a species by thousands of years at least.  
Apart from delivering power, power bubbles would be the finest storage medium for power that one could ask for. They would have every desirable attribute except portability. Compressed air is another medium that they could produce for transport or power storage.  Compressed air is beginning to look promising for increasing numbers of applications already.   Use off-peak power to pump cold air into the bubbles, and at peak times the hot air could be tapped off for power. The volumes available for such storage would be huge.

Another emergent consideration might result from the contact between magma and solvents such as superheated water or molten salts. Even under far milder conditions than those in power bubbles, such as in existing geothermal power schemes, we inadvertently extract troublesome minerals that currently we need to dispose of as wastes. In power bubbles however, we might find that the solvent extraction of live magma would yield large quantities of precious or at least valuable elements that are painfully rare on the surface or in the shallower parts of the crust. 

Temptingly possible candidates might include rare earths, cobalt, iridium, platinum and other elements desperately in short supply for industrial purposes. If we strike it rich in some of our power bubbles, their waste streams might play a rewarding role in the economy of the power generation process. They might for example more than pay for all the waste processing. Gross concentrations of the elements would not be necessary; some of them would be attractive to extract from easily processed wastes even at concentrations of a few parts per million.

The obvious choice for where to drill for power bubbles is on land.  It really seems stupid to drill shafts under three to five kilometres of seawater.  Having to work through layers of up to half a kilometre of ooze does not look very encouraging either.  Having to convey terawatts of energy up to the surface for use on land, perhaps hundreds of kilometres from the shaft, seems positively perverse.  

After all, to establish a power bubble would require far more than running a drill string from a surface ship, down a Mohole, to the mantle.  
And yet, one could make a very strong case for submarine power bubbles.  Oceanic lithosphere is possibly half as thick as typical continental lithosphere.  Much of it is more seismically stable than most continental lithosphere as well.  Drilling large shafts many kilometres deep is horrendously expensive, so thin lithosphere is attractive.  In certain areas beneath the ocean the magma is moving comparatively briskly, which implies huge heat flows. 

In short, whoever establishes a workable technology for deep-sea power bubbles will have the world by the tail.  
It is not immediately clear how to balance the risks of digging into mid-oceanic ridges, where vulcanism and instability are threats, against the comparatively trivial depths that one would have to drill to reach enough magma to make bubble blowing worth while.  There certainly are areas of level oceanic floor where the lithosphere is especially thin and yet the geology is acceptably stable for our purposes.  
Without belittling the challenges, it should be practical either to construct an igloo-type structure on the ocean floor, through the floor of which the shaft construction could proceed, or to construct the igloo on land, tow it out, and lower it to the sea floor where desired.  A tall ooze-penetrator might be dropped to establish itself on the hard basement where it could establish a foundation through which to sink a shaft.  
Nor is it clear that drilling power bubble shafts on land is unacceptable.  Hot areas like Yellowstone may or may not be long-lived in comparison to some other sites, but even a productive lifetime of no more than a few centuries should yield a good profit for such a scheme.  In fact, given the uncertainty of the future of Yellowstone, Vesuvius, and various other active spots, it would seem to be grossly irresponsible not to try controlling their development.  Bleeding off their gases and magma for a few decades at a time might forestall even very large explosive accumulations.  An eruption that otherwise might have covered several mid-western states or large cities, would be reduced to a massive source of power and possibly of mineral wealth as well.  
How far to combine such safety punctures with power bubbles is a question for the future. 

There is no doubt that the only option with more disastrous political implications than the successful introduction of power bubbles world wide, would be failure to introduce them.  On this planet the only alternative sources with anything like the same scope would involve importation of fuels or power from outside our planet, which is not the subject of this discussion.  
There are two scenarios to consider:
(a)  Suitable sites for power bubbles might turn out to be so common that most countries could construct as many power bubbles as they desired. This would affect a lot of vested interests and reduce the current local concentrations of fuel sources into few hands.  However, even the wide availability of power sources would lead to vast transitional problems.  Those who got in first and quietly developed the competence and technology would hold a huge advantage for a long time.  Countries that did not seize the opportunity by the forelock might suffer the consequences for generations.  Within each community vested interests of short-sighted parties would resist development, even when it meant destruction of their own interests within a generation. 
(b)  More likely, areas of sufficiently thin and stable continental crust, though plentiful, would be regional, so that many countries would lack economic options for constructing power bubbles.  This would have even more drastic consequences than ubiquitous power bubble schemes.  For one thing the countries that lagged in the race to establish a healthy energy surplus would be left in a very unhealthy economic and social situation.  For another, most of those that currently have very rich oilfields are not situated very promisingly for power bubbles.  For example, much of the middle east is unpromising for several reasons.  If ever there were a time when the major OPEC nations should be quietly preempting a revolution in their own future interests, by buying into power bubble development in promising regions, that time is now.  Conversely, if ever there were a time when the major industrialised countries should be preparing to throw off the constraints of fuel politics, now certainly is the time.  
In either case, there should at the moment be a major move towards developing subterranean power sources, not morsel by morsel, but on a scale to take over the world.  Power bubbles would take time to develop into a routine technology, so there would be intermediate objectives, but those with a view to the future should already have been on the way a long time now if far-sightedness or concern for the future were political strengths.  
As a rule they are not.  
That is the Achilles’ heel of both democracy and short-sighted despotism. 

·       Fossil organic fuels are widely regarded as either practically inexhaustible or sustainably exploitable.  People bandy figures about that differ wildly, partly because they differ in what they prefer to believe, in their choice of definition of what counts as resources, or what proportions of resources will be recoverable at what prices.  

There is one main point on which everyone agrees although they think they differ, namely that the end of organic fossil fuel supplies is urgently at hand.  Some speak of decades, some of centuries, but no rational person speaks even of millennia, let alone significantly long periods.  Bear in mind that in speaking of the survival of our "civilisation", centuries hardly count.  They hardly count even in the history of a country worthy of its name or people.

Some characteristically technically illiterate economists seem to think that because the stone age did not end with the supply of stones, that we need not worry about the energy age ending with the supply of energy.  If we do not prepare for the transition, that is just how it might end. The energy age simply might end with the supply of accessible energy. The foolish virgins among nations that have failed to establish a future for their peoples will have doomed themselves and pauperised their populations by not having prepared alternative power sources for their future economies. And power bubbles or some rival bootstrapping source of geothermal heat are the most promising successors to fossil organic fuels.  

One constant in our history and prehistory that has varied only in its scale, variety, and inefficiency, has been our dependence on energy.  Until economists explain how to transcend the laws of thermodynamics, their only option is to point out alternative sources of energy, or more likely, hope that some stupid technologist somewhere will point out a new source from which they could make money.  Discussion of at least one source is the theme of this essay.  

Even the limited supply of fossil organic fuels is not the whole story.  New sources of energy had better not rely on fuels that are too oxygen consuming.  We are consuming oxygen at a rate that should scare anyone spitless who takes our planetary budget seriously.  After all, where did the oxygen came from in the first place, and how fast?  

So, anyone who for example relies on a long-term hydrogen budget that does not produce as much oxygen as hydrogen, is leading us down the mortuary path.  With due and generous respect for the cap and trade bandwagon, we could multiply our CO2 budget by a factor of ten with no clear risk of harm, but if we consumed just a few percent of our free oxygen, the consequences would be disastrous, in fact murderous.  In comparison global warming would be trivial in consequences and trivial to control.  Recent palaeontological evidence suggests that since the Cambrian period, the oxygen level has never dropped nearly as far as some had supposed.  Even if it turned out that the recent evidence were wrong, it does not follow that we would comfortably survive such a change.  We already see intimations that to increase the rate of photosynthesis is not nearly as simple as people had imagined. 
·       Nuclear energy is particularly attractive in that it consumes oxygen only incidentally and in trivial amounts.  In effect power bubbles consume stored nuclear and gravitational energy for the most part, but in many ways they tempt us to use their power to produce materials such as hydrogen that also consume a fair amount of oxygen.  It would make sense to watch the implications of emergent technologies with caution.

The fashionable current concern about nuclear power is the hysteria about the nearly irrelevant problem of nuclear "waste disposal".  I shall not go into that matter, except to remark that I never have seen any such “waste disposal” that did not amount to criminal waste of irreplaceable resources.  The only rival to such criminality has been irresponsibly incompetent waste accumulation such as by the US and no doubt certain other major nuclear powers, where politicians and management were only interested in their pockets and short term careers. 

Nuclear energy has important roles, for example for heavy-duty transport fuel and transportable generators.  It also will have valuable functions in space, if only for expeditions far from the sun.  On Earth it also has important functions in isotope manipulations and so on.  However, fission power almost certainly will have a shorter fuel supply and less versatile output than power bubbles.  

Fusion power currently seems likely to be even less versatile than fission and it is not clear how much harmful waste it will produce.  Apart from some recent and sophisticated proposals, most claims that fusion power will be clean are exaggerated or ill-informed.  There also is the question of whether it ever will be practical at all, and if so on what scale, in which form, and using which fuels.  Though the technology deserves attention it is too early to rely on its success as a general purpose source of power.  
·       Space based solar energy is notionally attractive, but current proposals look very expensive and doubtfully practical.  Its main attraction is that if it works at all, a well engineered installation should work for a long time.  Some versions look more practical than others, but certain schools of partisans have quaint ideas concerning geosynchronous orbits instead of precessing polar orbits.  One hopes that they will grow out of it before their failures harm the prospects for serious development of a promising technology.  Meanwhile, I refuse to take any large-scale power project seriously before the ISS becomes a success.  If we cannot manage a toy project like ISS, what hope do we have of establishing spaced based solar power generators on a scale to serve New York, let alone the planet?  
·       Space based sources of nuclear fuel are another option.  He3 has been proposed as a fusion fuel of surpassing value.  If so, there are some tempting sources out there, but not yet any clear assurance of our being able to use He3 profitably down here.  Conversely there is no short term prospect of importing fission fuels at a profit. It should be thousands of years at least before we are so short of actinoid elements on Earth that it would pay to prospect for them in space.  
·       Unpredictable discoveries of new power technologies are always possible.  Someone might find out how to harness sunlight at trivial cost, negligible inefficiency, full portability, arbitrary scalability, and with huge storage capacity.  That would drastically affect the attractions of digging for power.  But history does not encourage us to wait for that discovery before considering how to improve our current buggy whips.  
·       An interesting possibility would be to drill for hot rock or localised magma in thin sea floor or near volcanoes, where there certainly is a great deal of energy to be extracted on a non-bootstrap basis.  However it might prove to be prohibitive to install power bubbles in a submarine environment.  All the same, submarine hot magma power might turn out to be a valuable resource.  It certainly would outlast oil and should be much cleaner than coal.  It might be as clean as nuclear energy.  
In comparison to most of those, bubble power offers us indefinite capacity, indefinite lifetime, low running costs, a wide compatibility with existing technologies, and a trivial ecological footprint.  It is not clear how widespread and plentiful sites for exploitation might be.  To ignore the prospects for such a field of technology would be irresponsible. 


A very reasonable question concerns the ecological hazards of soluble inorganic materials such as salts.  Most of them should not be very hazardous under the conditions contemplated.  If the working fluids are gases such as air or argon, which are not solvents at temperatures around 700C-1700C, there should be very little discharge and we could ignore pollution problems stemming from their use.  
Where more powerfully solvent fluids such as water and CO2 are extracted from or pumped through deep hot rock, they are likely to carry substances that one does not normally think of as being soluble, such as silica and various ore chemicals.  Some are harmful and are best removed, preferably for commercial exploitation of their chemical content.  Most are perfectly safe to discard at sea in diluted forms once any toxic components have been removed or immobilised.  Insoluble sludges are harmless as long as they are not allowed to smother large areas.  Soluble chemicals of low toxicity that occur routinely in sea water are harmless as long as they are diluted soon enough; in fact, most common substances to be expected in solutions leached from rocks, such as Ca, Mg and K compounds, would be desirable as nutrients in sea water.  They would have the added attraction of precipitating or neutralising dissolved CO2.  
If molten salts were used in or extracted from the shaft, most of them would be moderately benign and could be discharged into the sea without any ecological threat whatever.  The most obvious candidates would be salts of sodium, magnesium, calcium and so on.  Any breakdown products in the event of overheating or chemical reaction, say chlorine from chlorides, could easily and cleanly be scrubbed and neutralised.  
One certainly would not want too much of most salts to escape into the soil, and certainly not noticeable amounts of sodium and chloride salts.  Soil salination is a widespread problem and to pollute fresh ground water with more than a few parts per thousand of even nutrient salts is unacceptable.  Firstly, this means that effluent must be disposed of with care, and secondly, if the shaft contains any substances undesirable in ground water, suitable precautions must be routine and reliable.  
The necessary precautions vary with the site and conditions.  For example, where ground water is already uselessly salt and has been stagnant in a limited area for a long time, one need not be too fussy about small spills.  Such sites are common.  Also, as a rule, serious drilling or shaft sinking of types relevant to this discussion tends to be many times deeper than any ground water of practical importance.  In fact, ground water could be a serious nuisance, so power schemes of this type probably would shun sites where ground water is plentiful.

Oil drillers must be on the alert for blowouts when the shaft structure or the surrounding material cannot contain the pressure of fluids below ground.  This is a matter of concern for any deep drilling, but really, it is less of a risk for power bubble drilling than for oil wells. 
·       Oil wells tend to be drilled where there are likely to be pockets of compressed gas, such that unexpectedly drilling into one could lead to the drill string being practically lifted out of the shaft.  Power bubble drilling would generally be exactly where one is not likely to find such pockets.  One wants a thin crust in geologically uniform and stable areas, not a salt dome to drill through.  
·       One is not likely to begin a power bubble drilling project at the surface, then drill down to magma and proceed to blow fluid down the hole to make a bubble.  The structure of a power bubble project would be far different.  Excavations for the first three km or so would almost certainly be conventional shaft sinking such as is by now established technique for deep mines.  In fact, it might well be started at the bottom of suitable existing mine shafts.  Much of the infrastructure probably would be constructed within chambers branching off the shaft.  So far there is not much scope for concern.  
·       Deeper down, where the bubble shaft itself begins, it is not yet clear whether conventional drilling would be used at all.  We require bores much wider than simple geothermal injection wells.  Even if we do drill conventionally for a while, it would be for limited depths in solid rock.  Existing studies (e.g. Google for: magma energy engineering feasibility, R. K. Traeger, and Research Project  John L. Colp, Sandia National Laboratories) show that current drilling technology can proceed far deeper into hot rock than we require, but very hot rock is where we might well encounter bubbles of CO2 for example.  
·       When the hot rock phase of the drilling begins, we would need to establish a strongly-anchored well lining.  Once we reach plastic rock we would not only install linings, but anchor them into the walls and cool those walls to render the rock stronger than surrounding material.  At such depths we should be using thermal drilling by burning hydrogen or methane at pressures higher than ambient pressure, so that any gas bubbles we encounter would be at lower pressure than what we apply.  Any really large gas bubble we find probably would be a bonus that could be bled for power or for bubble inflation. 

There are two types of event that might cause a power bubble to wander.  The first would be buoyancy of the bubble within soft rock.  There are several approaches to dealing with that, of which I prefer the idea of designing the injection mechanism to spray the incoming relatively cool working fluid radially or tangentially so that it flows down the walls.  That would cool the upper areas preferentially and create turbulence calculated to circulate the contents of the chamber through any minor diverticula.  It still might happen that a portion of the bubble would rise beyond its most useful level, but within reason that would be a minor problem, something to be avoided as far as practical, but not disastrous in itself.  
The other major type might occur in a long-standing bubble if fairly rapidly drifting magma carries it off to one side faster than the bubble can be inflated.  If this should occur, it should be regarded as a bonus.  Such drift still would probably be slow, centimetres per year, and could be slowed even further by increasing the rate of injection of cool material and extraction of hot.  Another shaft could be sunk into either the leading edge of the bubble, or where the leading edge is anticipated to reach in the near future, or possibly into an upward bulge if a sufficiently large one developed.  
Extra shafts not only would increase the options for pinning the bubble by cooling its surroundings, but greatly increase the rate at which energy could be extracted by passage of the injected fluids down some shafts and up others.  In fact, it is unclear at the time of writing whether such multiple shafts shouldn't be standard practice anyway. 

As I already have mentioned, studies  show that existing drilling technology can penetrate far deeper into hot rock than we require.  All the same I actually am not currently contemplating conventional forms of drilling, except during specific phases where engineering experts might see it as advantageous.  
There are several problems with drilling through deep, hot rock with the conventional drill string.  One is the sheer difficulty and cost of dealing with such a long string, and another is removal of the drilled-out material from the hole.  
Instead I propose approaches based on drilling by application of heat and pressure.  Near the surface this would be costly at least, and might be unpractical in other ways too, but as we go below the zones where the environment is cool enough for humans to work the rock, the shaft sinking would be taken over by automated equipment.  At levels where this too becomes unpractical, we can begin to exploit the very heat that had been an obstacle at shallower levels.  Whether to use chemical or nuclear energy for drilling is an open question.  Possibly first one then the other.
There are two obvious chemical options.  One would be to pump down oxygen and a suitable fuel, possibly hydrogen, carbon, or hydrocarbon fuels. The drill head would be a torch nozzle of a suitable diameter, design, and material, and the whole unit would work under pressurised gas (oxygen?).
I rather suspect that an alternative design would be better, based at least partly on solid fuels such as solid carbon. Oxidisers such as nitrates, perchlorates or peroxides could be dumped in blocks, or pumped down,  perhaps in slurry form. Other fuels such as aluminium or sodium might get their oxygen directly from the ambient molten rock instead.  In general the chemistry of the process would offer an interesting field for study and prudent experiment.  
In any case, the mix would be designed to melt its way down as a pool of molten rock continuously enriched with fuel.  
The molten pool principle could be especially valuable during the early phases of extending the actual bubble downwards.  By continually adding modest amounts of fuel to the molten pool of salts, extra heat could be applied selectively to make sure that the hottest rock and flux would be at the lowest point below the inlet.  If working fluid were injected radially to cool the walls, this should increase the tendency of the bubble to grow downwards rather than sideways.   
By systematically injecting fuels and maintaining pressure, the whole system could melt its way down.  The resulting hole would not be a narrow puncture such as a drill would produce, but perhaps metres in diameter.  
An interesting version of the molten pool approach would be to form a pool of molten iron, manganese, or other suitable metal, feeding it with alternating doses, first of oxidisers such as peroxides, oxygen, and perchlorates, and then of reducing agents such as carbon, aluminium, sodium and hydrogen or hydrides.  The metal pool not only would be designed to be chemically reactive, but also dense.  Properly controlled it should eat and melt its way through most kinds of rock, largely catalytically, because molten slag of various kinds could be reclaimed by feeding in reducing agents such as sodium.  Temperatures could be raised by feeding in oxidisers.  The high density would concentrate the forces on the bottom of the cavity.  Slag of various kinds would float to the surface of the pool, where they could be drawn out onto the sides of the cavity.  Some of the fuels, such those that produce volatile oxides, would react to increase the pressure.  Others, such as reactive metals, would reduce pressures, yielding carbides and carbon instead of CO and CO2.  These reactions could be reversed by adding more oxidisers as required.  
Melting rock instead of drilling through it might sound very inefficient, but it has its compensations.  Firstly there is hardly any drill string problem, which otherwise could be prohibitive in terms of time and mechanical problems.  The heat produced would generally be very effectively contained, stored, and conserved on the way down.  It largely  could be reclaimed from the coolants when they re-emerge, and the cooling would contribute to the congealing of the walls.  
Alternatively, at a suitable depth we could begin to melt our way down by installing a high temperature fission reactor that relies on Doppler broadening to reduce neutron absorption for temperature control by negative feedback.  Such a reactor would be designed to be of considerably greater density than any molten rock it might encounter. It therefore would sink through any rock that it melted. 
If switching the reactor off turned out to be difficult, it could simply be permitted to consume its residual fuel in sinking away from the site of the bubble once we had reached the most desirable depth.  Otherwise it might be possible to keep it at the bottom of the bubble, supplying energy to assist in the inflation.  
As the heat source melted its way down, the hole would be kept open with pressure maintained from above.  Exactly what would happen behind the drill would depend on many factors.  A fair amount of experiment would be necessary to establish the best principles, let alone to design actual equipment.  I doubt that the actual rock itself would be strong enough to withstand the ambient pressures at the depths desired, even if cooled,  but if forming and cooling would do the trick, that would be an attractive option.  
Alternatively, as the shaft progresses, it could be expanded by pressure to somewhat wider than its desired diameter.  Then perhaps a molten refractory lining could be shaped and cooled in place.  The wall of the lining could formed to contain any channels required for carrying down coolants or extracting hot fluids.  
Another option would be to install voussoirs forming the tunnel wall together with any incorporated structures or mechanisms.  The rock around the voussoirs could be allowed to settle in behind them while still soft, holding them in place as it cooled.  Coolant entering the walls would keep the surrounding rock stronger than the rest of the rock in the region.  
A still simpler, but possibly more attractive approach would be simply to stop melting the way down once reaching rock of a suitable temperature, but instead just apply pressure to expand both the walls and the floor.  Pause the process while installing the next set of voussoirs, then apply cooling to the walls.  Once the voussoirs were set in place, apply pressure till the floor once again began to sink and make room for the next set of voussoirs.  Ultimately this process would end in blowing the power bubble itself, installing no more voussoirs, but only cooling the walls enough to prevent expansion of the bubbles from creating buoyancy problems. 

One of the most rational objections concerns the difficulty of applying the necessary pressures for working at such depths at all, let alone for forcing bubbles into the soft rock tens of km down.  
Well, I never said it would be easy.  I am as aware as anyone that forcing such pressures down a tube in the field is not at all the same thing as playing tricks in the laboratory.  
One might of course dismiss that objection with hand-waving reassurances: after all, we have overcome problems in the past that at first had seemed not just equally challenging, but equally absurd.  And yet, many such technologies are nowadays routine, not only taken for granted, but regarded as boring by that minority of the couch potato population as have so much as heard of them.  I am tempted to cite Clarke's law of technologies: that any sufficiently advanced technology is indistinguishable from magic, except that a far more powerful influence is that any effective technology soon seems far more tedious than marvellous to the smuggest and most petulant sector of humanity: the Great Uneducated and Uneducable.  
However, although a direct assault on multi-megapascal pressures might well be forbidding, multistage hydraulic devices of modest performance could be stacked at intervals on the way down the shaft, such that both the mechanisms and the surrounding rocks could handle individual levels of forces with comfortable safety margins.  
Another class of technique could also contribute to the effort.  Filling the hot parts of the shaft with molten salts or similar substances could partly or completely counter the surrounding hydrostatic forces.  
Challenging in general?  Definitely.  And yet, I regard the challenge of achieving of profitable nuclear fusion as far more challenging yet. 

The temperatures in question certainly are challenging, but they are not beyond practical limits.  I doubt that any exit gas temperature beyond about 1200C would be under consideration.  

Temperatures for melting the way down might be higher, perhaps more like 2000C, but those would be contained at the workface.  Perfectly practical alloys could survive such temperatures, and many ceramics could resist them indefinitely.  During the process of melting the way down, powerful oxidising and reducing agents would be used, but afterwards the operating environment would be chemically bland.   This greatly increases the range of candidate materials for the working structures.  In reducing or neutral atmospheres many carbides would work well at very high temperatures.  I could list several pure or composite materials with promising characteristics, but it is a safe bet that whatever we guessed at present, the engineers on the job would prefer their own alternatives for reasons of strength, chemical resistance, flexibility, cost, availability, and so on.  
In particular, I suspect that some of the most useful materials could be partly formed in situ under the influence of the very temperatures and pressures that they were to withstand.  This would hardly surprise anyone who in kitchen, workshop, or laboratory, has ruined an expensive utensil or tool by clogging it with a refractory, resistant, adherent char of one sort or another.  Many promising linings, matrices, cements and fillings might well be applied as polymers or mixes designed to char in place. 

It is well known that among the most worrying problems in drilling really deep shafts, geological faulting and unexpected rock movement and fractures are among the worst and hardest to predict.   Furthermore, some such conditions are particularly common in places where magma is temptingly near the surface.  
However, such a challenge is not as forbidding as it seems.  We do not need a shaft every ten km.  One or a strategically distributed few should be adequate for even a large nation.  This means that in most regions far more usable sites than needed would be available.  We could pick and choose.  For once, commercial considerations would coincide with ecological considerations.  No one in his right mind would want to drill a multi-billion-dollar shaft that needed constant maintenance or threatened immanent shutdown with every earth tremor.  
Geologically unstable regions such as large parts of the Himalayas or the Rocky Mountains/Andes certainly would be poor prospects.  Certainly it is not always obvious from surface inspections whether a given site is suitable, or even whether the best prospects would be on land or submarine abyssal plains. Equally certainly one would not select sites overlying particularly thick crust, but every digging project has its own elements of speculation, and this speculation offers vast rewards. 
When problems match rewards, they are called challenges.  
A very interesting question is whether the rewards are as irregularly scattered as they have turned out to be in the case of prospecting of oilfields.  Either way the political implications could be immense. 

Obviously, without support no shaft could survive at depths of any use in such  schemes.  But then, not even a shallow underground mine of modest scale could survive long enough to be useful without pit props and similar precautions.  There are several kinds of measures to apply in plastic rock shafts.  Which kinds to use would be a matter for engineers to develop during the establishment of the technology and decide on in any given case in the light of experience.  
The most obvious option would be to cool the rock round the shaft to well below the temperature of the surrounding material.  Such cooling should harden it significantly.  At modest depths relatively cool rock walls might be strong enough, depending on the nature of the rock.  The fluids pumped down to cool the walls would of course carry usable heat to the surface.  From the deeper levels this alone could be an indefinite supply of high grade heat.
The next option, and at greater depths, would be to line the hole with strong, refractory materials.  Different materials might be used at different depths, depending on their nature and cost.  For example, silicon carbide would be strong, pressure-resistant, and able to withstand adequate temperatures. In its raw form it should be cheap enough, though I am not at present confident about the costs of fabrication.  As described elsewhere in this document, automated equipment could apply such materials as lining voussoirs by first using fluid pressure to expand the hole as it gets dug, a length at a time, then packing the voussoirs from within and permitting the external rock to settle back, locking the voussoirs into place.  Channels within the voussoirs could carry the cooling fluids that would congeal the rock to cement and hold the blocks in place. 
If a suitable lining could be cast in place on the way down, a sort of high-pressure, high temperature concrete or char moulded in place could result.   
Thirdly, the pressure of the working fluid in the shaft would prevent shaft collapse.  If the shaft was not filled with suitable molten salts or similarly dense fluids, this pressure would of course be too great for the walls of the shallower horizons of the shaft.  There would have to be stages at which some pressure got relieved, probably by driving turbines at set levels in the shaft, or perhaps by bleeding off working gas to drive turbines nearer the surface.        

A major fault could wreck any shaft, no matter how large, no matter what its lining.  However, a suitably flexible or well-enough articulated shaft should be able to survive modest movements; there would be no moving parts running the length of the tube.  To put it mildly, serious damage to anything as valuable and costly as a power bubble shaft would be unwelcome, but not all is doom and gloom.  
By careful selection of sites where the nature of the geology does not encourage drastic movement, we can reduce the risk of major failures, but minor failures are likely to present a more constant, and therefore a more serious problem.  No matter how perfect the shaft, minor shifts are likely to distort and stress the tube from time to time.  If the stressed regions of the shaft can be heated up enough to anneal the surrounding rock to conform and relax the stress, that should be effective.    


By tapping our (by current standards) deep crustal or (by any standards) shallow mantle heat, we could extract energy on a scale to dwarf anything humanity has consumed to date.  As a source of energy it would be clean and versatile, and would outlast our requirements on this planet for many millennia at least, longer than we have used any fuel in any systematic way.  
The nature of the installations, namely bubbles inflated in hot rock in the deepest levels of the lithosphere, lends itself to bootstrapping larger, more productive installations, starting from working installations of modest size.  The bootstrapping facility alone radically distinguishes this type of energy extraction from all other geothermal enterprises or technologies.  So does the scale and viable duration of the installations.    
The nature of the energy is versatile, lending itself to generation of electric power, chemical processing and various industrial physical processes.  Although it should end our dependence on fossil organic fuels, it simultaneously will enable us to redirect the consumption of such materials into production of chemical feedstocks and intermediates.  
It also may develop into a major source of mineral extraction, though that is more speculative, and not for the foreseeable future a prime driver for any such project.  
A new source of power on such a gigantic scale would have intense effects on the international political scene.  Countries that fail to develop competence in the necessary skills and infrastructures will find themselves pauperised or at least at as great a disadvantage as those that currently, for different reasons, are hostage to OPEC. 

What we know about what we want or can get from the earth beneath us, is precious little.  Accordingly, a good way to start would be to drill a lot of moholes at various points around the planet.  That it has not yet been done is a stinging rebuke to the foresight and enterprise of the human community.  That inaction bodes ill for our continued survival, never mind our self-sufficiency.  
A mohole project need not begin very aggressively.  In combination with sufficient numbers of smaller exploratory drillings and a really aggressive seismic programme, two or three moholes on each continent and a couple of dozen on the ocean floor might do to start with.  It should be a sound basis for deciding where to drill the next generation of moholes.  We need to establish a map of the conditions beneath the outer crust, and of the materials and temperatures at various sites and depths. 
Given how little progress we have made with deep drilling, not even one respectable Mohole to date, this might seem to be an extravagant pilot investigation, but really it is long overdue.  Everyone agrees that we are blind ignorant of whole ranges of matters concerning the structures and processes we live above.  If ever there was any doubt that our ignorance is dangerous, there can hardly be doubt now.  Every deep hole presents us with unexpected findings.  It is past time to do something about it, and the nation that acts first might be the one that becomes richest thereby.  
Meanwhile, there are many lines of applied research that we must get underway in parallel with the drilling projects.  We do not yet have even a firm basis for designing large-scale deep-sea igloos, much less excavating multi-metre-wide shafts beneath them or exporting electric or hot-air power from them.  Who knows whether the ocean-floor ooze would be a serious challenge or an aid to the project?  
What about the principle of sinking shafts by melting, whether by nuclear reactor, raw chemical heat, or pressure?  Or the techniques for lining them with voussoirs or with heat-cured char?  All these technologies sound marvellous, almost trivial, but all we really know about them is that if we do not get down to business, we still will not know how or even whether we ever can master them.  
Too expensive to be realistic?  
It is a really disgusting reflection that we could get a good project going for less than the political advertising expenditures of a single presidential race in the US alone. And as for the money wasted on the accommodation of politicians in international conferences of the Kyoto type...!   

In sum, I meant it when I said "stop mucking with geothermal"; it is past time to stop mucking about, and get down to serious business.

Jon Richfield