for frequency, and 
for wavelength. The symbol `c' is used for the velocity of
light, thus:
Until the development of the space program, virtually all of the astronomers' information came in the form of light from distant objects. It is natural that they would have played an important role in the investigations of the nature of light, just as they did in the development of Newton's mechanics. Many of the pioneers in the development of the modern theory of light were keenly interested in astronomical problems.
The true nature of light wasn't really understood until the 20th century. Experiments done in the 19th century indicated that light was a form of wave motion. For many purposes, it is sufficient to describe light as a combined electrical and magnetic wave, or an electromagnetic wave.
In the decade following 1964, the great English physicist James Clerk Maxwell was able to describe all electrical and magnetic phenomena with the help of four differential equations. They are now called Maxwell's equations, and every student of physics must learn to work with them.
It's quite amazing that phenomena as varied as starlight, and children's magnets could be described by four relatively short equations, but this is the case.
Perhaps the simplest way to think of these waves is to first picture ``lines of force'' about an electrical charge. Nearly everyone has seen the lines of force about the poles of a magnet demonstrated with the help of iron filings and a glass plate. The concept of lines of force arose as a way of eliminating the problem that arose with the attraction of two bodies separated by some distance.
It's easy enough to understand how something will move if you grab it and push or pull. On the other hand, how can two bodies separated by "nothing" attract or repel one another? This difficulty is known as the problem of action at a distance. It can be solved, after a fashion, with the notion of lines of force. Think of electrical charges or magnetic poles as being surrounded by lines of force, like those demonstrated for a bar magnetic. Then the lines of force fill in the void between the bodies. They will grab the other body, and there is no more action at a distance.
Maxwell's equations described these lines of electrical or magnetic force. They showed that if you accelerated a charge, for example, if you wiggled it up and down, a wave would run out the electrical line of force.
The same equations showed that the electrical wave would have to be accompanied by a magnetic wave. That's a little hard to see, and we won't go into it here. Take our word for it. But the electrical wave can be pictured as something similar to the wave that would travel down a rope. Stretch a rope out horizontally, and wiggle one end of it, and a wave will run down the rope away from your hand. Pretend that you fasten the far end way away, so you don't have to worry about what happens when the wave gets to that end.
The electromagnetic wave is a form of light. It turns out that light is most conveniently described as a wave when the wavelength is relatively long, say of the order of a centimeter or more. Radio waves can be tens or even hundreds of meters in wavelength.
All wave motion travels with a velocity equal to it's frequency
(units are per second, or sec-1) multiplied by its
wavelength. Astronomers traditionally use the Greek letter
for frequency, and
for wavelength. The symbol `c' is used for the velocity of
light, thus:
= c
When the wavelength of light is shorter, a millimeter or less, say, a rather different picture is better--the photon picture.
According to the quantum theory, the energy in electromagnetic waves comes in little bundles that are called photons. The shorter the wavelength, the better it is to think of the photon as a kind of particle--rather than a wave. It isn't that the wave picture becomes invalid, it's just that for many purposes, it's better to think in terms of photons.
Light, X-rays, 
-rays, and radio waves are thus all forms of the
same phenomena--electromagnetic radiation.
In 1900, the German physicist Max Planck was trying to understand the nature of radiation that filled an enclosure that had come to equilibrium. When temperature within the enclosure was the same everywhere, people measured the amount of energy in the radiation as a function of wavelength. Their results could be displayed as a kind of histogram, with energy plotted against small wavelength intervals. What they found was quite surprising. The shapes of the curves depended only on the temperature of the enclosure, and not on the material inside. Planck himself made many of these measurements, and the energy distribution became known as the universal Planck radiation law.
The German word for this kind of a distribution of electromagnetic radiation literally means "radiation from a cavity." The equivalent English term is black body radiation. One can blacken a surface with paint, or better, with soot, and it will emit radiation that is (very nearly) in the form of Planck's universal law.
Planck found that he could account for this law with the help of
Maxwell theory provided he assumed that the energy in the radiation
came in multiples of the frequency: E = h.
The constant quantity `h' is now known as Planck's constant, and it is one
of the three fundamental constants of nature. The other two are the
constant of gravitation G, and the velocity of light, c.
The most energetic forms of electromagnetic radiation are called
-rays. These photons are characteristic of the emissions of
atomic nuclei, and have energies of millions of electron volts
(MeV). X-rays are slightly less energetic, with energies of
thousands of electron volts, kilo-electron volts, or KeV. The
photons of visible light have energies of the order of a few electron
volts (eV).
Name Energy A Characteristic Wavelength ---- ------ ---------------------------- Gamma rays Mev 0.01 Angstroms X-rays KeV 1 Angstrom visible light 0.2 eV 5000 Angstroms infrared light 0.003 cm microwaves 1 cm radio waves 100 meter
There are other forms of "radiation" that are not electromagnetic.
The famous "rays" of radioactive material were called 
,
,
and . Of these, only
rays are electromagnetic. The
and rays are helium nuclei
and electrons, respectively.
The first astronomical telescopes were made by Galileo. They used a lens and an eyepiece. Such telescopes are called refractors because the light is bent by these lenses. The bending of light is called refraction. Galileo's eyepieces allowed him to see the images right side up, just as binoculars do. It turns out to be a little more efficient if one can live with an upside down image. The traditional astronomical telescope inverts the image. This is not a disadvantage when viewing celestial objects, for which we may have no preconceived notion of how they should be oriented.
For many years, astronomers made upside down maps of the Moon, because that was the way the Moon appeared in their telescopes. Even in recent textbooks on astronomy, one can see photographs of the Moon that are upside down. Planetary astronomers prefer to have the north at the top of a map, however, and gradually most lunar maps have appeared in that orientation.
The largest refracting telescope is the 40" at the Yerkes Observatory of the University of Chicago. It, and the Lick 36" were built at the end of the 19th century. Early in the 20th century these instruments were surpassed by large reflecting telescopes. The 60" reflector Mount Wilson went into operation on Mount Wilson, California in 1908, followed by the 100" reflector in 1917. The latter was the world's largest telescope until the 200" reflector was completed at mid-century, and placed on Palomar Mountain. As our century ends, numerous larger telescopes have gone into operation. Currently the largest are the twin Keck telescopes in Hawaii, which have the light-gathering equivalent of 10 meter telescopes.
Telescopes have 3 roles or functions
is the ratio
/D, where D
is the diameter of the primary. This ratio comes out in radians, and is
often converted to seconds of arc. There are 206265 seconds of arc in a
radian. Higher resolving power traditionally means the ability to
see objects separated by a smaller angle.
Telescopes may be characterized by 2 numbers, the diameter of the primary (mirror or lens) and the focal length of the primary. The focal length is the distance from the primary at which an image is formed by light rays coming from infinity.
All of the newer major telescopes are reflectors, because it is possible to make them bigger than lenses. The current new telescopes are of the order of 8 meters in diameter. By putting many mirrors together, the effective aperture has been increased to 10 meters. Reflectors are free from chromatic aberration, which distorts images formed by lenses. If a telescope used a simple lens, the blue light would come to a focus before (closer to the lens) than the red.
Mirrors have distortions too. In particular, light rays near the axis of the mirror may not come to the same focus as rays near the edge. This effect is called spherical aberration. It was the problem that affected the Hubble Space telescope.
Our Earth's atmosphere is transparent to light only from about 3200 to 10,000 Angstrom units. One Angstrom is 10-8 cm. Shorter and longer wavelengths are blocked by various absorbers. At much longer wavelengths, there is an important radio window, from about a centimeter to perhaps 10 meters.
The atmosphere not only absorbs radiation at wavelengths astronomers would like to observe at, it also distorts images--the twinkle of stars results in a distortion. So space observations are important, and the Hubble Space Telescope has been making many important discoveries.
The outer rim of the Hubble primary is too low, giving rise to spherical aberration. After three years of sub-standard performance, a servicing mission installed corrective optics, and the instrument is now performing to expectations.
Photography came into its own in the latter half of the 19th century. All of the early astronomers before this time were visual observers. They had to record their observations in the form of sketches, and some of them got to be pretty fanciful. The Earth's atmosphere distorts the images of stars and planets, so in a very real sense, observers couldn't really see the same things.
The problem came to a head in the case of observations of Mars, where the American astronomer Percival Lowell claimed to see elaborate markings, which he interpreted as artifacts from an advanced civilization. Other astronomers saw nothing like the amount of detail in Lowell's sketches.
Photographic plates of Mars showed even less detail than visual observers could see. This is because the plates were essentially long exposures; the flickering images of the planet were blurred. Visual observers could wait for moments of exceptional "seeing," and it was at those times that Lowell claimed to see things no one else could.
For most of the 20th century, the traditional astronomical detectors were photographic plates. They were widely used in astronomy toward the end of the last century, continuing to well past the middle of the current one. Quite wide fields may be photographed, and this is still one of the major strengths of astronomical photography. However, only about 1 to 2% of the light falling on a photographic plate is detected. This drawback not present with modern electronic detectors; the most common is called the CCD, or charge coupled device.
CCD's were invented at Bell Labs in the mid 1970's. Unlike photographic plates, they detect 90% or more of the light that falls on them. However, they are still small in size, the largest ones being about 2 inches on a side. They are also expensive.
CCD's operate on the basis of the interaction of light photons and silicon crystals. Photons in a certain wavelength interval are capable of freeing electrons within the crystal so they can be made to move as in ordinary electrical current. The freed electrons are maintained in the original locations by electric fields generated as a part of the detector. These fields are said to generate potential wells, which we may conceptually think of as little bowls in which the electrons collect. The more electrons, the more light photons that hit the silicon. It is possible to localize these potential wells in to very small areas, so that the resolution of modern CCD's approaches that of photographic plates--10 to 20 microns (1 micron is 10-4cm).
After a time, the charge in the wells represents a digitized version of the image. It is then possible to manipulate the wells in such a way that the charge in each of them may be measured. In essence, the charge (or number of electrons) in each well is read one by one, in such a way that the location of the charges is remembered. The image may then be reconstructed by a computer. To do this by hand would take quite a while. A modern CCD might contain 20 Megabytes of data!
With a photographic plate, or with a CCD, the most elementary technique
is called imaging--just a fancy way to say "taking a picture."
Images of star fields were traditionally measured carefully for the
positions of stars, to reveal their proper motions or parallaxes.
The images of extended objects, such as bright or dark gas clouds, or of
external galaxies would simply be examined and their characteristics
noted. Photographic images of galaxies led Edwin Hubble to his famous
tuning-fork classification scheme.
Extensive imaging from space probes have been used to advance
our knowledge of the solar system. Even before the lunar missions,
planetary geologists had begun to map the surface of the Moon using
techniques that had been developed by terrestrial geologists.
We had to wait for space probes before the same methods could
be applied to the planets. Telescopic images of Mars,
both visual and photographic, had led
to more confusion than enlightenment. Mariner and Viking images
of Mars in the 1970's, revealed a surface intermediate in complexity
between those of the Earth and Moon.
A standard technique for the analysis of an extended source such
as a planetary surface is to make plots that indicate areas with
equal brightnesses. Astronomers call such plots isophotes,
and they resemble contour maps, except that brightness rather than
altitude are indicated. With digital information, it is possible
to display brightness with color rather than contours. Such
false color maps are now used in a wide variety of scientific
applications.
It was known to Isaac Newton that if sunlight was passed through a
polished glass prism, it would be dispersed into a rainbow of colors.
This is illustrated in Figure 13-2
During the 19th century, physicists learned how to use this technique
to identify chemical elements in a laboratory sample. At the same
time, similar methods were employed as a means to analyze the
atmospheres of the sun and the stars. If the dispersed light is
examined with the eye, the instrument is called a
spectroscope.
If that light is first recorded with a photographic plate, or
electronic detector, the term spectrograph is used.
The light itself, after dispersion, is called the spectrum
of the object that emitted it. The plural of spectrum is spectra.
It was already recognized that the nature of the spectrum
of an object depended on physical conditions.
Gustav Kirchhoff's
laws of
spectroscopy are
valid today:
It was soon realized that one could identify the chemical element
by its bright or dark spectral lines. There is a story that Kirchhoff
once told his banker that he had identified the chemical element gold
in the spectrum of the sun. The banker was unimpressed, because
there was no way to mine that gold. Later, Kirchhoff was awarded
a gold medal from the British Royal society, along with some gold
coins. He is said to have returned to his banker, saying "this is
gold from the sun."
Modern spectrographs generally employ gratings rather than prisms
as dispersing elements. Gratings can be used either in transmission
or reflection. In the latter case, a mirror is closely ruled with
parallel scratches. The light that bounces off such a surface is
dispersed into little rainbows on either side of a central image,
called orders.
It has become possible to make these scratches so that most of the
light is thrown into one of these orders. Such a grating is said
to be blazed, and can be a very efficient dispersing element.
Atoms, electrons, and nuclei no longer obey Newton's "classical" laws.
The modifications are not intuitive. You have to learn the rules, and
that takes some effort. For our purposes, we can examine the case of
a very
simple system, a hydrogen atom consisting of an electron and a proton.
The two relevant charges, are the same in this case, and
the attractive force is is an inverse square one, just as for
gravitation. The potential energy curve is proportional to 1/r,
and looks just like the one for the two-body problem [Figure 11-2(b)].
Quantum rules prevent the electron from being at any (old) distance
from the proton. If we drop the electron straight at the proton, then
there are a series of distances where the electron will "hang" on its
way down the well.
It is the wave nature of the electron that makes it "hang" at certain
depths in the potential well. It turns out that when you examine the
wave nature of the electron, its wave has just one loop at the
lowest possible energy level,
2 loops at the next one up, and so on. The picture is a
little more complicated if we add angular momentum, but we don't
need that here. There must be an
integral number of waves in the well or the corresponding energy
isn't allowed. This is why the energies
are said to be quantized.
On the way to higher energies, the allowed levels crowd closer and
closer together.
Electrons can get from one level to another by emitting or
absorbing a photon of just the right energy. This is an example
of the famous quantum jump. For a jump down, a photon
must be emitted, and for a jump up, one must be absorbed. These
photons are created when the electron jumps down, and are destroyed
when it jumps up. So the energy of the photon, h If a photon of frequency We can understand the Kirchhoff laws with the help of this quantum
picture. First [see K1 above], we get a continuum when the energy levels get
very close together. This can happen in solids (like a tungsten
filament) and liquids. When the atoms are very near one another, the energy
levels get smeared out and their average values can also get close together.
This will also happen to a gas under high pressure,
such as the visible photosphere of the sun.
When [K3] a continuous spectrum
shines through a cool, low pressure gas, the atoms in the gas selectively
absorb out the discrete wavelengths determined by their special
energy levels: E1, E2, E3, etc.
If a hot gas is observed without a continuous source behind it,
some fraction of the atoms in this gas will always be in higher energy
levels. They can get into these upper levels in a variety of ways.
They can bump into one another, and get boosted to higher internal
energy states. What happens in that some of the kinetic energy of
motion in transformed into internal energy. Also the hot gas
may absorb occasional photons from some other source. These
photons may boost the electrons to upper levels, or even ionize
the atoms. Eventually, the electrons will recombine and/or just
jump to lower energy levels, with the emission of a photon. When
these photons are observed, they show a bright-line, or emission
spectrum [K2].
Radio telescopes function much like optical ones. Their primaries
are often parabolic reflectors, but much larger. The same formula,
The large, often parabolic receivers that one sees in pictures of
radio telescopes function rather like the antennae of ordinary radios.
The electromagnetic waves from space are made to drive electrical
currents which
are then subject to ingeneius and powerful amplification. You can often
read about radio astronomers "listening" to sounds from space. The
signals received are electromagnetic, and not acoustical. The popular
notion of listening comes by analogy to "listening" to an ordinary
radio. The electrical currents can be made audible with the help
of speakers. Often, radio astronomers employ speakers, and do listen
to their sounds. Unlike the sounds from an ordinary radio, the
sounds from speakers at a radio telescope mostly resemble static.
Astronomers are keenly interested in signals from deep space that
are carried by high energy photons, X-rays and In
planetary astronomy, this region of the electromagnetic spectrum is
chiefly employed in the overall technique called remote sensing,
which we will discuss in detail later in the course.
Electromagnetic radiation comes in many forms. In order of increasing
wavelength, and decreasing energy and frequency, we have:
Light is analyzed with the help of spectrographs. Astronomers
have used both prisms and gratings to disperse light.
Each atom and ion has a set of unique energy levels. Transitions
or quantum jumps among these levels give rise to the emission or
absorption of photons. Kirchhoff's laws describe the conditions for
continuous, absorption, or emission spectra. Chemical elements
can be identified by measuring the wavelengths of light that
are emitted or absorbed by some source.
What does it mean to analyze cosmic materials? What information
would we have after such an analysis that we didn't have before?
There are a variety of ways cosmic materials may be analyzed.
The simplest kind of question we might ask is for the number
of atoms of each chemical element in the sample. So we can
think of the periodic table, and having a number for each element.
A chemist would think in terms of the number of "moles" rather
than numbers of atoms, but if we know Avogadro's number
(Na) we
can give the information either way.
If we take a sample and divide it into several parts, the
number of atoms in each part will be different, in general.
If the sample is uniform--not different in one place than
another--the relative numbers of atoms in each of the
divided sample will be the same. Cosmochemists almost always
work with relative numbers of atoms, rather than absolute
numbers.
In the analysis of terrestrial and moon rocks, or meteorites
the element silicon is usually taken as a standard, and numbers
of atoms of other elements are given relative to the number of
silicon atoms. A standard ploy is to quote the number of atoms
of elements relative to a million silicon atoms.
Geochemists or cosmochemists will commonly speak of
abundances of the elements in some sample. They almost
always mean by `abundances' the relative number of atoms
or moles of some element to the number of atoms or moles
of silicon. Since we are taking ratios we are free to use
either atoms or moles, since a possible Avogadro's number would
cancel from numerator and denominator of the fraction.
Analyses by number of atoms are very useful when it comes
to investigating isotopes. Isotopic abundances are essential
for radioactive dating of cosmic materials as well as for
determining their history. It has been found that the relative
abundances of the three stable oxygen isotopes, 16O,
17O, and 18O are characteristic of different
portions of the solar system. Since lunar rocks show the
same isotopic abundances as terrestrial, we must conclude that
the lunar impactor (Big Whack) must have originated from very
nearly the same regions of the primitive solar system as the
Earth.
An analysis by number of atoms alone obscures much of the
history of cosmic materials. All of the chemistry, all of the
processes that caused the atoms to fall into the potential wells
we call chemical bonds, is lost. It is therefore also necessary
to analyze samples in such a way that as much of this chemistry
as possible is revealed. In the case of moon rocks and meteorites,
much of the relevant chemistry is essentially mineralogy. We
will get to mineralogy in detail in the next lecture. Here,
we will discuss methods used to determine relative
amounts of minerals in a sample.
For more than a hundred years the typical technique for
the chemical analysis of a sample involved dissolution in
acids. The solutions might then be treated with reagents (chemicals)
to produce an insoluble precipitate which would then be
separated and weighed. Today's techniques often involve dissolution
in acids, but usually as a first step to a variety of methods
quite different in nature from those that ended with measurements
being made on a chemical balance.
We will discuss some of these methods in turn, starting with
an instrument called a mass spectrograph.
Mass spectrometry may be the most powerful technique of modern
cosmochemical analysis. The
basic instrument
was invented in 1919
by the British chemist and physicist F. W. Aston.
A standard mass spectrograph would work in the following way.
First, the sample would be vaporized. This might be done by
simple heating, or the sample might be dissolved in acid first,
and then vaporized. The chemical treatment would have to be
carefully chosen not to interfere with the analysis. The vaporized
atoms or molecules would then be ionized. There are a variety of
ways to do this, both chemical and physical. A simple method is
to bombard the sample with electrons, perhaps from a hot filament.
The ions in the sample are then accelerated in an electric field,
and collimated into a beam, often with simple mechanical baffles
or other means, called ``ion optics.'' The beam then passes into a
region with a magnetic field.
It is one of the fundamental laws of electromagnetism, that a
moving, charged particle is acted upon by a force. The force depends
on the velocity and the electrical charge, which we shall call q.
The direction of the force is perpendicular to both the
direction
of the field and the velocity. This means if the velocity is
in the x-direction, and the field in the y-direction, the force
is in the z-direction. Complicated, but that's the way it is.
By Newton's second
law, the acceleration (vector), which determines the trajectory,
then depends on the ratio of the charge to the mass:
The ion beam is therefore bent in a curved path. The ions are eventually
collected in a detector, and made to generate an electrical current
that can be measured. Ions with different m are accelerated differently
and so follow slightly different paths. If multiple collectors are
available, the relative numbers of ions with different paths is
immediately determined by the difference in electrical currents
generated in the collectors.
With one collector, only one
kind of ion can be measured at a time. However, it
is possible to change the
strength of the magnetic field in such a way that a new ion, with
a different m, will enter the detector. The relative numbers of ions
can be determined as before, by the different electrical currents
generated.
It is also possible to tell the differences in the masses of particles
in the beam by the magnetic field strength change that was necessary to
make the new ions enter the detector.
Since F is directly proportional to the charge, q,
the mass spectrograph strictly measures the ratio, q/m. If we can be
sure that the charges are the same on all of the ions, one may
determine the mass of particles entering a detector from the
geometry of the beam.
One of the major uses of mass spectrographs is to measure the
relative abundances of different isotopes of the same or other chemical
elements. The age determinations of rock samples rely on such
determinations. We will discuss this technique in Lecture 16.
Mass spectrographs can measure ratios of species with such
accuracy that they can also be used to get absolute amounts of
elements using a trick called isotope dilution. Here,
isotopic ratios, say 16O to 17O are
measured in an unknown sample. Then that sample is mixed
thoroughly with a known amount of material with a different
16O to 17O ratio, but for which the
percentage of all oxygen is completely known. When the
isotopic ratio of the mixture is again measured, one can tell
how much total oxygen was in the unknown from the amount
that the isotopic ratio changed.
Chromatography gets its name from a technique used to separate
different substances in mixtures used in dying. If a cloth were
dipped into such a solution, the liquid would begin to wet the cloth,
but different constituents of the dye would climb the cloth at
different rates. The result would be a cloth with stripes of
different colors.
Chromatography today is used in a variety of ways having little to
do with the color of anything. The basic mechanism upon which it is
based is the mobility of atoms, molecules, or ions as a function of
their physical and chemical properties. Chromatographs commonly
employ either the gas or liquid phases of matter. In either cases
there are similar basic constituents:
Figure 14-2 illustrates these components for a gas chromatograph.
The supply for the carrier gas is shown on the left. This is often an
inert gas, such as helium or nitrogen. The stationary phase is chosen
so that the unknown species will be temporarily trapped by it. This
trapping may take place by a variety of interactions, but they generally
involve weak chemical bonding. A simple kind of trapping is by a process
known as adsorption, in which a solid surface attracts a layer of
gas molecules. If the stationary phase is liquid, the unknown samples
can dissolve in the liquid, ultimately to evaporate from it.
Different "unknown" species will interact with the stationary phase
in different ways and so reach the detectors at different times.
A variety of detectors can be used to detect the presence of the
(unknown) foreign species in the carrier gas. A common detector
measures the thermal conductivity of the gas--the rate at which heat
will flow across it. This depends on the composition of the gas.
It is these times that are used to analyze an unknown mixture.
In space probes, a common detector is a mass spectrograph. The
combination, mass spectrograph-gas chromatograph, is common enough
to be known by the abbreviation GCMS.
This
combination is an integral part of the experiment to investigate
the large moon of Saturn known as Titan.
One very powerful technique for the analysis of cosmic materials
in the laboratory is irradiate an unknown sample with neutrons.
The neutrons strike the nuclei within the sample, which then react
in characteristic ways. The method is called neutron activation
analysis, or NAA.
The neutrons used in this technique are typically obtained from
a nuclear reactor. Many such reactors are available are available
at national laboratories and universities. Neutron activation
analyses are performed at the U of M with the help of the facilities
of the Michigan Memorial
Phoenix Project.
Each atomic nucleus is unique. It has its own
energy levels, typically of the order of MeV apart. Since neutrons
are uncharged, they are readily absorbed by most nuclei, to produce
an isotope of the same element with one additional unit of mass.
Not all chemical elements are easily analyzed by neutron activation
methods, but many are. Let us take the lanthanide rare Earth cerium
as a typical example.
Cerium has 4 stable isotopes:
136Ce, 138Ce, 140Ce, and
142Ce. The most
common of which is 140Ce. When a thermal neutron is
absorbed by a 140Ce nucleus, the isotope
141Ce is created. This isotope is unstable, and eventually
decays to the stable 141Pr (praseodymium), with the
emission of electron, a process we have seen before, called
beta decay. The half-life for this transition is 32.5 days.
The 141Pr nucleus that is created by the beta decay
of 141Ce is not in the ground state for that nucleus.
It is in an excited state, 0.145 MeV above the ground state. The
141Pr nucleus quickly gets rid of this extra energy by
the emission of a photon with 0.145 Mev of energy. It is this
photon that is detected and measured in neutron activation analysis.
The gamma ray photon from the decay of 141Pr may
be
measured with a gamma-ray detector that operates on the principle
of the photoelectric effect. One detector uses a germanium crystal
which absorbs the gamma rays. Free electrons are then produced
within the crystal, and they can generate a current, that can be
made proportional to the energy of the incident gamma rays. In
this way, the gamma-ray spectrum from the sample can be measured.
Ultimately the interactions that take place in an experiment
of this kind are sufficiently complicated that it is necessary
to calibrate the instrument with the help of a
standard sample whose
composition is already known. Such a standard is exposed to the
neutrons of the reactor in a way that is as close as possible to
that of the unknown material. Then signals from the known and
unknown are compared.
When we may assume the responses of the instrument are linear,
we can say the following. If the detector gives twice the current
from 0.145MeV gamma rays from the unknown sample as from the standard,
then there are twice as many 140Ce atoms in the sample
as in the standard.
This is the essence of the neutron activation method. It is well
suited to the analysis of whole-rock samples, as opposed to the analysis
of some portion of a rock that might be isolated by physical or
chemical means. Some 60 or so elements may be analyzed in this way,
but one does not obtain direct information on the chemical or mineralogical
composition of the sample.
In practice,
in this as in most instrumental methods, there are many refinements
and complications. We must leave them to those who actually perform
the analyses.
Spectroscopy came into use in astronomy as well as analytical
chemistry in the late 1900's. The technique is basically the same
although it is more common in the laboratory to use emission
spectroscopy for quantitative work than in stellar work.
Astronomers have also analyzed many emission sources, from hot,
diffuse gases. These gases may be excited by the ultraviolet
photons from hot stars, or by shock waves from exploding stars.
Both in laboratory and stellar spectroscopy, there is a
light source, and a spectrometer which produces a spectrum.
The spectrum may consist of either bright or dark lines.
The essence of the technique of spectroscopic analysis is that
the strength of the lines is related to the number of atoms
that produce them.
Laboratory analysts have a definite advantage over astronomers.
They can compare the spectra of an unknown material with those
of one or more samples whose compositions are completely known.
If the strength of spectral lines of an element in some unknown sample
are the same as those in a standard, then it may be safely assumed
the number of atoms in both sources are similar.
Astronomers must know the detailed conditions of the
stars or nebulae emitting the light they must analyze. They
must know the temperature and density of the relevant gases.
By contrast, the laboratory spectroscopist only needs to be
sure that the unknowns and standards have been analyzed in the
same way. This is not to say that the job of the laboratory
analyst is a cinch. It may give some insight into why
laboratory results are generally more accurate than astronomical
ones, and why it is so essential for us to have "returned samples"
from the planets.
When atoms combine chemically to form molecules, the relevant
energy levels change in fundamental ways. Molecules have their
own unique spectral lines. Moreover, these lines may be produced
in ways not available to atoms.
As in atoms, the electrons in a molecule may jump from one
energy level to another, with either the emission or absorption
of a photon. Unlike atoms, molecules may store energy either
in the form of rotation or vibration. The energies associated
with molecular rotation or vibration are typically lower than those
associated with electronic levels. The latter are, as in atoms,
of the order of electron volts (eV).
Molecular vibration and rotation energies are one to many orders
of magnitude smaller than electronic energies. Characteristic
vibrational energies lead to spectra typically in the range of
one to tens of microns (10-4 cm or 10-6
meter). Rotational spectra lie typically in the microwave region
of the electromagnetic spectrum, with wavelengths of a millimeter
to a meter.
Molecular spectra may be observed in emission or absorption from
free molecules in the interplanetary medium or planetary atmospheres.
In the Earth's atmosphere, there is extensive absorption due to water
vapor. Because of that, much of the infrared spectra of
stars and planets cannot be observed from the ground.
Very important advances have been made over the last several decades
in the realm of reflectance spectroscopy. In this case, one examines
the spectra of light reflected from a sample. This sample might be
a planetary surface, in which case the source of light would be the
sun. We will return to this method in Lecture 27, when we discuss
asteroids and techniques of remote sensing.
One of the most useful analytical methods of the geologist
is called optical mineralogy. The method goes back to
Henry C. Sorby who examined the first thin section with
a microscope in the mid 1800's. The technique requires grinding
and polishing a sample of a rock or mineral to a thickness of
0.003 cm (or 30 microns micro-meters). These are mounted on
slides, and examined with a
petrographic microscope. The light from the source is
made to pass through polarizing filters before it goes through
the thin sections. The results are typically beautiful
kaleidoscopic
images generated because the light interacts differently with
minerals in different ways.
People trained in optical mineralogy
not only recognize the different minerals--a kind of quantitative
analysis--but can also tell quantitatively how much of each mineral
is in the sample. The relative fractions of different minerals
is a primary factor that distinguishes rock types. Rough estimates
may often be made from hand specimens, but the ultimate arbiter is
the thin section. It is interesting that this general method was
one of the most enlightening used in the study of the lunar samples.
In this course, we do not need to become experts in optical mineralogy.
We only need to know that thin sections can be made, minerals
identified from them, and their relative amounts evaluated.
The information from optical mineralogy is quite different in nature
from that obtained from the techniques discussed earlier. While one
may deduce some information on mineralogy from these methods, one
gets it directly from the geologist's microscope. As we will see in
Lecture 14, the mineral content of a rock is an important clue to its
history.
Several methods of laboratory analysis involve
firing beams of particles at samples and observing the X-ray photons that
are then emitted. The particles may be electrons, protons, or even
ions, for example, O- (oxygen with an extra electron
attached). It is also possible to use X-rays photons in the exciting
beam.
In all these instances a small volume of the sample is affected,
typically, a micron or two in diameter. This means that it is possible
to examine individual minerals in a rock sample, so these probes can
give important information about the mineralogy as well as the
elemental composition of a sample. Often, when a crystal forms
by solidification within a melt, the composition of that melt will
change while the crystal is forming. This can make the composition
of the crystal change from the inside to the outside, a phenomenon
known as zoning. Probes are ideal for investigating such changes
in the composition of a zoned crystal.
Typically, when fast electrons from a probes strikes the atoms
in a source, an electron is ejected from an inner shell. This creates
a hole in the inner shell, into which an electron from an outer shell
may drop. The jump of the outer electron into the hole is deeper into
its potential well, and the excess energy is emitted as a photon.
Inner shell electrons have energies in the X-ray region, so X-rays
are emitted.
It is possible to distinguish among chemical elements by the
wavelengths of these X-rays. Generally speaking, the lightest
chemical elements are not easily studied from their X-rays.
Ion beams are used in a rather different way. They
strike very small areas of a specimen, and deposit sufficient
energy to create a small plasma (ionized gas) atmosphere over
the point of impact. This plasma may then be analyzed using
mass spectrographic methods. In this way one can get isotopic
information, often lacking in other methods.
An important technique that has been used for the analysis of
lunar and Martian rocks in situ has been to fire
The common analytic techniques that are used in the analysis
of cosmic materials are (1) optical and mass spectroscopy, (2)
neutron activation analysis, (3) electron and ion microprobes,
and (4) optical mineralogy with a polarizing microscope.
Remote sensing by reflectance spectroscopy will be covered in
Lecture 27.
Optical
spectroscopy, in the laboratory or at astronomical observatories,
is based on the unique pattern of absorption or emission lines
from the chemical elements. In neutron activation, one observes
the varied reactions of atomic nuclei that have absorbed neutrons.
These nuclei emit gamma rays which are characteristic of the individual
nuclei. Microprobes fire beans of particles (or photons) at a
source. The analyst either examines X-rays, or in the case of
ion probes, uses a mass spectrograph to analyze the microplasma
created over the focus of the beam. The thin section and
polarizing microscope is a classical tool in geology that allows
the skilled observer to
recognize minerals and determine their relative proportions.
NASA Pathfinder Image
Matthew P. Golombeck
Project Scientist, Mars Pathfinder
In Lecture 4 we pointed out that Earth's crust is
only about 0.4% of its total mass. The oceans and the atmosphere
are only about 6% of the mass of the crust. Most of the Earth
is rock and metal. Apart from the liquid outer core, these materials
are minerals.
There is little reason to believe that the bulk structure of Venus is
significantly different from that of the Earth. Mercury appears to have
a somewhat larger portion of metal than rock, while with Mars, there
is more rock and less metal. As far out as the asteroids, we believe
the denizens of the solar system are primarily mineralic in nature.
If we are to understand the current structure and history of these
objects, we must learn something about the special chemicals known as
minerals. This is the purpose of the present lecture.
We shall define minerals as naturally occurring materials that are
usually solids with a definite crystalline structure
and chemical composition. The classical definition of minerals omits
the qualification "usually" and must draw a distinction between ice
and liquid water. It must also make some provision for mercury, which
occurs rarely in its elemental, liquid form. Biological processes are
capable of producing crystalline solids that would not qualify as
minerals under some definitions.
Books on mineralogy usually provide an description of
calcite (CaCO3), whether or not the atoms were ever a part
of a living creature. On the other hand, they do not describe
crystallized protein, which surely has a regular spatial structure and
chemical composition. The latter is clearly organic in origin,
while the biological origins of the former are often shrouded by
time, physical, and chemical processes.
There are problems with most definitions, as we pointed out in connection
with the undefined terms of physics. However, these difficulties are
rarely a bother to anyone other than pedants.
References on mineralogy list several thousand mineral names.
Most of these minerals are rare, and of little relevance to
the non specialist. There are two kinds of mineral names
in common use, specific and general, and it is important to
realize the difference. Mineral family names are
often used. "The feldspars" provide a common example. The
family name, "feldspar," includes the common pink potassium
feldspar, and a sodium (albite) and calcium (anorthite)
feldspar, as well as additional varieties that we shall
not be concerned with.
Only
a small number of minerals dominate the bulk chemistry of
the Earth and probably all cosmic solids. We therefore set out
a highly simplified classification of minerals, as follows.
The refractory oxides did not all vaporize when the solar system
formed. Some of them were formed out in interstellar space, perhaps in
the atmosphere of a red giant star. Their composition therefore
is not the same as that of the solar system as a whole, the
SAD. Here,
we speak of the isotopic composition, the relative proportions of
isotopes in the minerals. Since corundum is always
Al2O3, its elemental composition is always the
same. But there may be a different mixture of the three isotopes of
oxygen: O-16, O-17, and O-18.
Mineral fragments that never vaporized when the solar system formed
are called pre solar grains. Their investigation is one of the
most exciting areas in the chemistry of cosmic materials. We will
return to the topic of pre solar grains in Lecture 35.
Silicon forms many compounds for much the same reason that
carbon
does. Recall that the entire domain of organic chemistry is the
chemistry of carbon compounds. Both carbon and silicon have 4 electrons
in the outermost shell, 2 s-electrons and 2 p-electrons. These
different subshells form a mixture in such a way that all 4 electrons
are similar in nature. Chemists call such a mixture a hybrid. The flat
formula for methane, CH4 is
This is supposed to show the 4 bonds are similar (the | and -- are as
close as I could get them to look with the font available). Actually,
methane is a 3 dimensional structure, with the bonds directed to the
corners of a regular tetrahedron. This geometrical figure has four
faces made of equilateral triangles. Each bond angle is equidistant
from the other bond angles.
Silicon also typically forms bonds directed to the vertices of a
regular tetrahedron. Silicon tetrahedra are SiO4,
and this complex ion
requires four additional positive charges for electrical neutrality.
A good example of a mineral where the silicon bonds to 4 oxygens
with tetrahedral bonding is the family of olivines. These are
Mg2SiO4 and Fe2SiO4,
and mixtures
with intermediate compositions, which we might write
(FeMg)2SiO4.
The mixture can form a solid solution of the two "end members." It is a
miscible solution, that is, the solution is like alcohol and water
rather than oil and water.
The next major family is the pyroxenes. This time there are
four end members, but we shall only need the names of two of them,
enstatite and diopside. The most common pyroxene of terrestrial
rocks, augite, is a mixture of all four end members, but closest
in composition to diopside.
This can be thought of as the bottom part of a triangular diagram
with Ca2Si2O6 at the top. However,
this calcium silicate is not called a pyroxene because of the structure
of its crystals.
The feldspars can be described with the help of a full triangular
(sometimes called a ternary) diagram:
The two feldspars on the bottom form a miscible solid solution the
general term for which is plagioclase. Anorthite is very common in
lunar rocks. Rocks dominated by plagioclase feldspar are called
anorthosite. The suffix "site" usually indicates a rock rather than a
mineral--"ite" is a common ending for a mineral. Lunar anorthosites are
dominated by the calcic feldspar. Much of the ternary field between
K-feldspar and anorthite is not filled by natural minerals.
The x's in the above diagram illustrate this area, but very
crudely. Between
albite and K-feldspar one gets an "unhappy" solid solution. Given
enough time, the two feldspars will try to separate out from one
another in the solid state. This phenomenon does not happen for an
albite-anorthite mixture (plagioclase).
A mnemonic for the three common silicate families is to add an
SiO2. Thus, enstatite
Mg2Si2O6
has one more SiO2
than forsterite, Mg2SiO4
and albite NaAlSi3O8
has one more SiO2
than enstatite. The mnemonic isn't perfect.
You'll sometimes see enstatite written MgSiO3,
you still must
remember that the feldspars have aluminum, and that the Al's
and Si's change from albite to anorthite.
Use the mnemonic if it helps.
The olivines, pyroxenes, and feldspars are the major minerals of the
Earth's crust. They explain to a large extent why the crust is
dominated by 8 elements: O (62%), Si (21.2%), Al (6.5%),
Na (2.6%), Fe (1.9%),
Ca (1.9%), Mg (1.8%), and K (1.4%). The percentages here are by
numbers of atoms.
You can see that these are the
elements in the dominant minerals.
Note also that the elements Na, Al, and K, which all have odd Z are
more abundant in the crust than you would expect them to be from their
abundances in the SAD. In particular, compare
these abundances with that of the even-Z element, sulfur, which is not a
major element of the Earth's crust.
Clearly Na, Al, and K are abundant in the crust because they occur in
the feldspar minerals. Next question: why are the feldspars so
abundant in the crust of the Earth. They are not the major minerals of
the far more massive mantle.
When a heterogeneous solid,
such as a rock is heated, the first liquid that appears will not have
the same chemical composition as the rock itself. Indeed, the melt is
enriched in those minerals with the lowest melting temperature. If this melt
is separated from its parent material, and then refrozen, the new rock will
be enriched in those more easily melted minerals. If the new rock is again
subject
to partial melting, the new melt will be still richer in minerals that
melt easily.
The rocks that make up the terrestrial crust have a complicated
history of partial melting, freezing,
melting and refreezing. Petrologists often speak of the "distillation" of
the material. As any student of booze knows, if you distill a mixture of
alcohol and water, the vapor is first more enriched in alcohol than the
parent liquid. This same situation holds for
mineral mixtures, except that in this case
the relevant phases are solid and liquid, rather than liquid and vapor.
There are
phase diagrams that are used to describe this
process in detail. The upper phase diagram describes the behavior of a
mixture of the feldspars albite and anorthite. Pure albite
NaAlSi3O8 is at the
left, and pure anorthite, CaAl2Si2O8
is at the right. A solid solution of these two minerals is called
plagioclase. Note that the melting point of pure anorthite is
considerably higher than that of pure albite. Anorthite is tough stuff,
a fact which explains much of the chemistry of the lunar highlands.
We'll come to that in Lecture 23.
If a solid plagioclase with the
composition A' is melted, the first liquid to appear has the composition
A. With further melting, the solid will move up the curve labeled
solidus, toward the point C, and the liquid composition will move to the
upper right, toward the composition A' again. When all of the solid
has melted, the liquid will have the original composition, A', again, of
course.
Partial melting by definition, means that not all of the parent
material is melted. Then the liquid that may be removed will have a
different composition--it will be chemically differentiated from its
parent. In processes that take place on the planets and their moons,
solids may be partially melted, and the liquids forced toward
the surface through
fissures or vents of various kinds.
The albite and anorthite are miscible in both liquid and solid
solution. The lower phase diagram shows what happens when a solid
containing diopside and anorthite crystals is partially melted. This
time, there is no solidus. When the solid is heated, the first melt has
the composition indicated by E. If the parent material has the
composition A, that is, 80% diopside, and 20% anorthite, then some
diopside will remain solid until all solid is melted. The liquid will
begin with the composition E, and the melting will go on until all of
the anorthite is melted. Then the liquid will gradually move up toward the
composition A again.
It is not necessary for you to remember the details of the partial
melting. For various mixtures, it can get very complicated. What you
need to remember is that during partial melting,
the most easily melted minerals will be
the first to liquefy. This is the reason they work
their way upwards--toward the surfaces of planetary bodies.
This is the situation for the sodic and
potassium feldspars, as well as for quartz. They are more
easily melted than the olivines and pyroxenes that form the
bulk of the Earth's mantle. In addition, their densities are
lower. These two factors insure that the feldspars and quartz will
work their way upward as a result of melting and partial melting
processes that go on in the Earth.
What processes cause partial melting? We know from seismology
that most of the volume of the Earth is solid. But heat is generated
within the Earth, probably mostly by radioactive decay, and it works
its way non uniformly. Most of the melting in the present Earth
is associated with plate tectonic activity, which we shall
take up in Lecture 17.
Solid Earth materials are subject to two main processes, partial melting
and weathering. While these can be very complicated processes, certain
regularities do emerge, so that it is possible to say from the mineral
content of a rock, if its constituents have had a complex history or
not.
The American
geochemist N. L. Bowen summarized the overall geochemical trends with
a diagram that
has come to be known as the "Bowen Series." There are two series,
a continuous and a discontinuous one:
The series on the left is called discontinuous because the
olivines and pyroxenes are immiscible as solid solutions.
The feldspars are all at least partially miscible, hence that
branch of the series is called continuous.
Natural processes cause minerals
at the top of the series to be transformed by a variety of chemical
reactions to ones lower down. These processes may take place at
low temperatures, as in the case of the weathering of feldspars
to clay minerals. Important geochemical reactions also take place
at the higher temperatures of magmas.
In this course we will call the minerals
at the top "early" and the ones at the bottom "late." A rock with
mostly "late" minerals may be assumed to have had a complex history of
partial melting and weathering.
We shall be concerned primarily with how easily minerals are
melted and how dense they are. Generally speaking, these properties
are closely related to positions in the Bowen series. The
early minerals,
at the top of the series are both denser, and less easily melted than
the late ones. The properties are not parallel in the continuous and
discontinuous branches. This is shown in the following
table.
We can see that forsterite has both a higher melting point and
density than it's opposite number in the continuous series, anorthite.
However, anorthite has a higher melting point but a lower density
than the common pyroxene diopside. This has important consequences
for the formation of the lunar highlands, as we have already
mentioned.
We have introduced four categories of minerals: native elements
and alloys, oxides, silicates, and a miscellaneous category containing
only a few carbonates, phosphates, or sulfides.
Of these, the silicates
are the most complicated. We discussed the olivines, pyroxenes,
and feldspars. There are also hydrated and complex silicates,
such as the amphiboles, and micas. Chemical formulae for
14 minerals were written. You need not explicitly memorize
all of these formulae and names, but you should be able to
recognize them, and
put them in the appropriate mineral families. Use the mnemonic
trick for the olivines, pyroxenes, and feldspars.
The Bowen Series relates geochemical processes on the Earth
to mineral content of rocks. Rocks with complicated geochemical
histories contain minerals that are late in the series.
Typically, minerals that are early in the series are denser
and have higher melting temperatures than those that are late.
The preponderance of the mass of the terrestrial planets is either
metal or rock. To a geologist, any mineral aggregate is a rock, and
if there is only one mineral in the mass, it may be called a
"monomineralic" rock. The study of rocks is called petrology,
from the Greek word for rock, "petro."
The three main divisions of rocks are
igneous, metamorphic, and sedimentary.
Igneous rocks formed from a melt, which might have cooled slowly
within a planet, or rapidly on its surface.
Sedimentary rocks were deposited from a fluid. In the case of terrestrial
rocks, the fluid is almost always liquid water. An interesting case
occurs when volcanic ash is deposited in layers from the air. It
can be compacted into rocks with the mineralogy of igneous material.
Such rocks, called tuffs, often appear layered, are light in color,
and can be mistaken for limestones. Large volumes of tuff may be seen
in Yellowstone Park. The most common sedimentary rocks are limestones
and sandstones. None of the lunar rocks are sedimentary.
Metamorphic rocks are either igneous or sedimentary rocks that have been
modified by heat, pressure, or shock. To metamorphose is to change.
Common examples are gneiss and schist. Gneisses are typically squeezed
igneous rocks that exhibit foliation or layering. Schists
come from shales subjected to pressure, and for them too, the layering
is apparent. Marble is metamorphosed limestone.
In this course we shall omit most of the complex processes
of metamorphic and sedimentary petrology. Thus far, these processes
belong (almost) entirely to Earth scientists. On the other hand,
Moon rocks have been formed (almost) exclusively by igneous
processes. When samples are returned from Mars, this situation may
change completely. There is good evidence that liquid water
was once common on the surface of that planet, so the probability
of finding sedimentary or metamorphosed Martian rocks is reasonably
high.
There is one metamorphic process that is highly relevant for lunar
rocks. It is called shock metamorphism. Almost all of the lunar
samples show evidence of extensive fracturing due to impacts of
meteoroids during the last phases of lunar formation. Rocks that
are made up of broken fragments are called breccias. Most
of the lunar samples are brecciated.
The other source of extraterrestrial rock samples available to us
are meteorites. These may contain igneous materials as well as
those that have been subjected to aqueous alteration, that
is, metamorphosed by water and water solutions. Other meteorites
contain materials that are not accurately described by any of the
terminology developed for terrestrial or lunar petrology. We
postpone a discussion of these materials until Lecture 35.
We must consider one category of sedimentary rocks, the
limestones. They are of great importance in the history of the
Earth's atmosphere, because they now hold the Earth's complement of
CO2. In Lecture 26, we will explain why there is so
much CO2 in the atmosphere of Venus, and so little on
the Earth.
With the complexities of most metamorphic and sedimentary
processes avoided, we now turn to a consideration of igneous rocks.
Igneous rocks are described in a variety of ways.
There are words to describe:
The following figure gives a simplified, two-dimensional classification
of igneous rock types. The horizontal coordinate is related to the
rock chemistry or mineralogy. The vertical coordinate is related to
texture.
The names mafic and felsic are derived from chemistry. 'Mafic'
derives from magnesium, and
ferric.
'Felsic' comes from feldspar and
silica.
Important chemical and mineralogical trends that take place from
mafic to felsic rocks:
Some special rock categories that shown in small type in
Figure 16-1 are:
Even though the mantle is mostly olivine and pyroxene, there is some
extra SiO2 and feldspar. If mantle rocks are partially melted, the
silica and feldspar can come squirting up through fissures and veins.
We think of the crust as a kind of distillate. As we mentioned
in Lecture 15, we use the
notion of distillation, which commonly involves transformation of a
liquid to a vapor. The recondensed vapor is called the distillate.
In the present case, we are talking about solid and liquid material.
Partially melting of mantle rock produces a liquid relatively rich in
SiO2 and feldspar. When this freezes, it is a sort of distillate,
and this "distillation" process is one reason why the crust of the Earth
is chemically very different from the mantle. The other main reason is
"weathering."
Sea floor spreading, typified by the mid-Atlantic ridge, builds
an oceanic crust rich in mafic minerals that have been only slightly
modified from mantle materials. Similar mafic rocks are thought to
form the lower parts of the thicker, continental crust, which are
often referred to as the sima, for silicon and magnesium.
The upper continental crust, called the sial, is also enriched
in aluminum, primarily from the feldspars.
The oceanic crust, mostly sima, is some 10 to 12 km in thickness.
The continental crust, sima + sial may be 25 to 35 km thick.
We now have the background to understand the results of the analysis
of Martian rocks by the Pathfinder mission. In July of 1997, the lander
deployed on the surface of the planet, and a roving laboratory and scout
called the Sojouner began to analyze some of the nearby rocks. At the
height of the mission, some of the names given to the rocks by the mission
scientists, such as Barnacle Bill, and Yogi, became household words.
The principle instrument to probe the mineralogy was called an
Alpha Proton X-ray Spectrometer. It is a kind of ion probe, as
discussed in Lecture 13, but in this case, the ions were alpha particles,
or helium nuclei. The particles came from radioactive nuclei that
emit alphas, such as plutonium. This was a part of the instrumental package.
When these alphas hit the rock, they interact with the atoms and nuclei
of the rock to produce X-rays, protons, and simply backscattered
alphas. From the energy spectra of these particles and photons,
it is possible to determine relative atomic abundances in the sample.
Unfortunately, an analysis for the relative proportions of the
chemical elements leaves the mineralogical composition of the rocks
open. Geologists have a way of getting from the atomic percentages
to plausible mineralogical compositions. Efforts of this
kind are shown in Figure 16-2
What we see from these pie charts is that the geologists believe
there is a lot of quartz and feldspar in these rocks in addition to
the mafic mineral pyroxene.
A plot that does not involve assumptions about the mineralogy is
shown in the next figure.
The two Mars rocks are indicated by the large stars, near the
bottom-center of the plot. Barnacle Bill is A-7, and Yogi is A-3.
What we need to notice about this plot is that the composition
of these rocks does not resemble that of the terrestrial
ultramafics. Indeed, our best guess at the bulk composition of the
Earth's mantle would plot rather high and to the left in the
broad area labeled "Terrestrial Ultramafic Rocks."
Putative meteorites from Mars, including the notorious AH 84001
with possible evidence of past microbial life, plot to the left of
the terrestrial samples.
Press releases have described the "surprising results" of
these analyses as indicating an overall composition for the rocks
at the landing site as resembling the bulk composition of the
Earth's crust--andesitic. We now understand what this term
means, from Table 16-1. Moreover, we can also make some inferences
about the maturity of the Martian rocks from the point of view
of the Bowen series. The rocks have been subject to some
distillation and perhaps weathering.
The overall analyses of these rocks have been a little confused
by a dust coating. Attempts have been made
to correct some analyses for contamination by this dust. The
overall conclusion is that all of the rocks in the vicinity of
the Rover had compositions resembling an `average' for the continental
crust. Therefore, not so felsic as the sial, but definitely more
differentiated than mantle ultramafics.
Two common words that are used to describe rock types are 'volcanic'
and 'plutonic'. These words can mean very nearly the same thing as
'fine grained' and 'coarse grained'. Rocks from volcanoes are extruded
on the surface of the Earth, and cool rapidly. Their grain size is
therefore small. Anyone who has tried to grow crystals of sugar or
salt knows it takes time to form large ones.
By this time, we have three terms to describe rocks that cooled
rapidly: fine grained, volcanic, and extrusive. There's at least
one more that the reader will be spared.
Some rocks cool slowly, because the magma from which they form
did not erupt on the Earth's surface, but was intruded into a layer
beneath it. These rocks are also called intrusive, and
because large aggregates of rock formed in this way are also called
plutons, such rocks are also called plutonic. Slow cooling allows
for grains to grow in size, hence we have coarse grained, plutonic,
or intrusive rocks. Again, we spare the reader additional names.
The professional geologist may use these words with shades of
meaning beyond those of grain size. Usually, volcanic rocks
are mafic, and often plutons are granitic or andesitic in
composition. Thus chemistry as well as history might be
implied.
Some rocks cool so rapidly no grains form at all, and the frozen
material is said to be glass rather than crystalline.
One may define
a glass to be a solid lacking crystalline structure. Physical
chemists describe glasses as supercooled liquids.
Glassy rocks are called obsidian if they are granitic
(= rhyolitic) in composition. Obsidians are common roadside finds
in Oregon, where volcanoes of the Cascade Mountains have ejected
felsic materials.
Pegmatites fall at the opposite extreme. These are rocks
with large grains, of the order of a centimeter or more. The most
common pegmatites occur along with minerals late in the Bowen series,
and are therefore granitic in composition. This means that pegmatites
have had a complex chemical history as well as an extended cooling
time.
Many of the trace chemical elements, the rare Earths, and the
radioactive elements uranium and thorium have no major
minerals of their own. They also have rather large ions, and have
to force their way into rock crystals. This means that they tend
to be preferentially retained in a melt, and that they are among
the first materials liquefied upon partial melting. Thus, they
tend to work their way upward, along with the typical felsic
materials. Thus, granites tend to have more radioactivity than
basalts, and pegmatities are even richer in these trace elements.
Will any pegmatites be returned from Mars or Venus? If so,
we will know some of the chemical and physical processes that
may be inferred from such a find.
An important trend in the properties of cosmic materials as the
chemistry changes from mafic to felsic is the viscosity of the melt.
Viscosity is the property of a fluid that makes it sticky. Tar is
a viscous fluid that becomes less viscous as it is heated.
There are two properties of mafic lavas that make them less
viscous than felsic lavas. First, they melt at characteristically
higher temperatures, and most fluids become less viscous at higher
temperatures. (Modern multi-weight engine oils are an important
exception to this rule.) Second, the presence of the SiO2
in the magma is known to be an important factor in the viscosity,
the more SiO2, the higher the viscosity. The nature
of the interactions are rather complicated, and we must be
satisfied with the general notion that liquid SiO2
makes for a sticky magma.
There are many
volcanoes
known in the solar system.
Impressive volcanoes are known on Mars, and there may
be more volcanoes on Venus than on any other planet in the
solar system. As far as we know, none of these volcanoes are
active,
but on the Galilean satellite, Io, there is active volcanism.
A very
simple division of volcanoes uses only two types:
The main difference in these extreme volcanic types is
due to the viscosity of the lavas. If the lavas are mafic,
shield volcanoes form. Felsic lavas tend to stick in
vents, often forming plugs. When the pressure builds to the
point that the plugs are ejected, an explosion often follows
driven by the release of steam.
Volcanic explosions occur for reasons similar to explosions of
chemical bombs--there is a rapid transformation in the phase of
material from solid or liquid to vapor. In the case of the volcanos,
water is dissolved in the magma. Because of the high pressures under
the Earth, much more water can be dissolved in the liquid than would
be possible at atmospheric pressures. The sticky felsic lavas seal
the vents of the volcanos, and allow pressure to build. When cracks
develop so the magma is open to the lower pressures above ground, the
water comes rapidly out of solution, and because of the high
temperatures, it comes out as a vapor rather than a liquid. This
water vapor requires a much larger volume than when it was dissolved.
The expansion is the source of the explosion.
The mechanism resembles what happens
when a carbonated drink is shaken in the bottle and the cap
suddenly taken off. Here, CO2 dissolved under high
pressure comes out of solution when the pressure is released.
The volume of the CO2 is much greater than the volume
of the bottle, once the pressure is released.
Volcanologists have many tales of destruction by explosive
volcanoes. In a notorious case,
the town of St. Pierre on Martinique island was destroyed
along with some 20000 inhabitants by the eruption of
Mt. Pele in 1902.
Rocks may be igneous, metamorphic, or sedimentary. The former
are most important for astronomy.
Rocks names may describe the chemistry, texture, history of a rock
or a mixture of these. Mafic rocks contain ferromagnesian
minerals while felsic are rich in feldspars and silica. From
mafic to felsic, the fine-grain types are basalt, andesite,
and rhyolite. The corresponding coarse-grained types are gabbro,
diorite, and granite. Anorthosites, dominated by plagioclase
feldspar, are common on the Moon.
Explosive volcanoes are associated with viscous, felsic lavas. Mafic
lavas flow more readily, and tend to form shield volcanoes.
Geology is the study of a planet.
We have discussed how waves can travel down a rope when we
described
a model for electromagnetic waves. That kind of a wave is called
a shear wave.
When the material a wave is running through is moving
perpendicular to the direction of wave motion, the wave is said to
be a shear wave.
Shear waves are unable to travel through gases and liquids.
All three phases are capable of transmitting pressure waves.
When pressure waves run through matter, the particles oscillate
in the direction of propagation of the wave itself.
In both pressure and shear waves, there is no net motion
of particles of the medium. The waves travel, but the particles only
oscillate over a limited distance.
The velocity of a wave through matter depends on two main factors.
Both pressure and shear waves travel more slowly through a denser medium
than a less dense one. The velocity also depends on the strength with
which the medium resists deformation. Most solids are a little more
resistive to a compression, which reduces the volume, than a shear, in
which the volume is distorted in shape, but not changed in size.
This latter property of wave motion makes pressure waves travel faster
than shear waves through the same medium, and gives rise to the
geophysicist's designation of the pressure waves as `P', and the
shear waves as `S'.
It is a useful mnemonic that pressure waves are designated with
a P, and shear waves with an S. The origin of these letters came about
in an entirely different way.
Geologists use instruments called
seismographs to detect wave motions in the Earth
that are generated by earthquakes. The first waves to reach the
instrument are called primary, or P. These are characteristically
followed by slower waves, the secondary, or S waves.
The instrument in Figure 17-3 illustrates the general principle
on which all seismometers work. Part of the apparatus will move
along with the earth. In the figure, this is the base. A second
component is constrained by inertia so that it will not respond
immediately to short-timescale Earth movements. This is the arm,
weighted in the figure with a brick! The motion of the Earth may
be recorded with a pen on moving paper, or in modern instruments,
an electrical signal is generated, amplified, and eventually displayed.
The
speed of the two kinds of seismic waves depends on the composition
and state
of the material through which they travel. It is possible to
use them to gather
information about the unseen interior of the Earth. The simplest
example of this is the location of the Earth's core. It is obvious
that the Earth's composition must be a mixture of both metal and
rock from its mean density. Even if we allow for compression, the
decompressed density of the Earth is higher than any plausible rocky
composition it might have.
The Earth's core was discovered in 1906 by the Irish geologist
Richard Oldham. It is interesting that he did not get a clue to
the presence of the core from the S waves, which are actually incapable
of being transmitted through the liquid of the outer core.
Rather he noted the existence of a shadow zone in which
P waves from an quake in the opposite hemisphere of the Earth
failed to appear.
Waves traveling through the body of the Earth are bent outward,
as shown in the figure because of refraction. This is the same
phenomena that causes light to be brought to a focus by a lens.
The difference is that when light enters glass from the air, it moves
into a medium where its velocity is lower. The ray is therefore
bent
toward the normal the the surface.
The velocity of seismic waves, both S and P, increase as
they move into the interior of the Earth, and therefore their
trajectories are
bent
away from normal, or outward, as shown. The reason for the increase
in the wave speed is related to the difficulty of compression (or
shearing) of the material as it is subjected to ever increasing
pressures of the overlying layers of the Earth. There is, actually,
a competition between the increase in density and the increase in
the forces that resist deformation. The former would make the waves
travel more slowly, the latter more rapidly. In this case, the
latter forces win out, and the wave velocities increase with depth
in the Earth.
At the boundary to the outer core, the phase of the material changes
from solid to liquid, and the resistance to deformation changes
accordingly. The S waves are not transmitted through the outer core
at all, and the velocity of the P waves drop significantly.
Whenever waves encounter a medium in which the velocity changes,
they may be reflected as well as refracted. In general, this happens
at every surface. Astronomers coat the glass surfaces of lenses
with a special material that reduces the amount of light that is
reflected from them. This increases the efficiency of their instruments
since the reflected light is typically lost. The geologist cannot
do this with seismic waves, of course. Therefore every P wave that
strikes the core is partially reflected from it, and partially transmitted
into it.
S waves cannot travel into the core, but their energy can be
transformed into a P wave at the surface of the core, and the
resultant wave can propagate through the core, to emerge as
partially P and partially S.
The
zone of rapid rise of seismic velocities for waves descending from
the crust of the Earth
is called the Mohorovicic discontinuity
or Moho. It marks the transition from the Earth's mantle
to the crust.
The Moho is the step to the first long shelf of the upper mantle.
The asthenosphere is indicated by the dashed lines, indicating
(in a highly schematic way) lowered velocities in the mantle due
to softening of the material.
Figure 17-6 is an expansion of Figure 17-5, with some annotations.
It shows the basis
for divisions of the
mantle into inner and outer zones. The jumps in wave velocity
correspond to changes in the crystalline structure of the olivines
that make up the bulk of the mantle. Two transitions occur,
one very near 400 km depth, and another at a depth of about 650 km.
As the pressures of the Earth's layers increase, the olivines
first change the arrangements of their ions into a structure
that resembles that of another family of minerals known as
spinels. This change is a physical and not a chemical change.
At greater depth, both a
physical and chemical change takes place:
The olivine changes into a silicate with the composition of
a pyroxene, plus an oxide. The structure of the MgSiO3,
the arrangement of the ions, is not the same as that of the common
pyroxenes, but resembles the oxide perovskite, CaTiO3.
This mineral form has only recently been understood as a result of
experiments that have been carried out with
diamond anvil presses,
capable of reaching the necessary pressures to bring about this
phase change. We must leave the fascinating technology and
experimental results with diamond anvils to web surfers.
Just enter "diamond anvil", and away you go!
The new geometrical arrangements of the ions as a result of
pressure changes both the density and the resistance of the
materials to deformation. Generally speaking, the resistance
to deformation has a larger influence on the velocities of the
seismic waves, so these velocities increase, causing the
upward steps to the right (toward the center of the Earth)
seen in Figures 17-5 and 17-6.
Modern geophysicists analyze a broad network of seismometers
with the help of sophisticated analytical programs. The results
are three-dimensional images with far more detail than the classical
picture of shells. The basic technique is not unlike that employed
in medical cat scans of the human body. This technical term for
building higher dimensional images is called tomography.
The seismic waves, for example, travel along a one-dimensional path
from an earthquake to a given seismometer. However, information
from enough of them gives two and three dimensional pictures of the
Earth's interior.
With the help of these methods it has become possible to study
in detail the modes of transfer of heat from the center of the Earth
to its surface.
The planets all have both internal and external heat sources. The
most important external heat source now is the sun, but in the past
meteoroid bombardment supplied a good deal of heat. Internal heat
sources derive from radioactive decay, chemical energy, and
gravitation.
The average heat coming from the Earth's interior is estimated to
be about 4 x 1013 Joules/sec, or 4 x 1013 watts.
Radioactive elements within the Earth are expected to provide
just about this amount. We shall make more quantitative estimates
of this heat in Lecture 29.
Heat may be transported by three classical mechanisms: conduction,
radiation, and convection.
Heat is transported by conduction when there is no net motion of
the molecules through which the heat is flowing. Molecules in regions
where the temperature is high pass their energy to regions where the
temperature is lower by collisions or by oscillations. If you stick
a poker in a fire, eventually the end you are holding gets hot.
There is no transfer of iron atoms down the poker. The ones in or near
the fire oscillate more rapidly than those where it is cooler, and
pass their energy down the rod of the poker.
In radiation, photons transport energy from regions where it is hot
to those where it is cooler. Energy from the sun comes to us by
radiation.
In convection, there is a physical transport of hot material to
regions where it is cooler. This transport often comes about because
hotter materials are less dense than their surroundings, and they
tend to rise. In the interior of the Earth, it is thought that
very slow convection currents move hotter rocky material from deep
regions toward the surface.
In classical laboratory experiments done at the end of the
1800's, convection cells were observed in laboratory fluids
that were heated from below. These cells lasted for long periods
of time relative to the turnover time for the fluid, and had a
regular, honeycomb-like appearance when viewed from above.
If the heating from below was increased, the cells became irregular
in shape, and individual cells would persist only for one or several
turnover times for the fluid.
Convection cells are common in the atmospheres of planets.
The Earth is no exception. Such cells
can often be seen in cloud layers from an airplane flying above them.
We have mentioned this in connection with the sun's atmosphere. In the
Earth, the motions are very slow indeed. They must take place in
a medium that transmits seismic shear waves. However the distortions
in these waves are very rapid compared to the motions thought to
occur in mantle convection. Here the motion is comparable to that
of the continental plates. Indeed, their motion is thought to
derive from the circulatory pattern of mantle convection.
It is not certain how deep the convective cells really go.
Figure 17 - 7 shows a cell that involves both the upper and lower
mantle. Some geophysicists believe only the upper mantle is involved
in convection.
The crustal plates are thought to be driven from below by
convection on
a time scale of about 100 million years. The rate of motion of the
plates is about an adult's height in one lifetime. In round numbers,
say about 2 meters in 100 years. That makes 2000 km in 100 Million
years--roughly correct. Continental drift, an older term with much
the same meaning as plate tectonics, was strongly advocated by
Alfred Wegener in the early decades of the 20th century. It was not
accepted until convincing observations were made of sea-floor spreading
in the mid-Atlantic ridge.
The plate boundaries on the Earth are well
delineated by
earthquakes and active volcanoes. Many of the Earth's gross
features may be accounted for in terms of the motions and interactions
of these plates.
When the plates move toward one another, one of the plates may
sink beneath the other. In this way, crust is destroyed, and we
have the opposite of what occurs in sea floor spreading, where
crust is created. The region where one plate is moving beneath
another is called a subduction zone. There are several
possibilities when plates collide, because the plates may be
made of either (thin) oceanic or (thicker) continental crust.
Here are some classical processes that take place at plate boundaries.
Until the spring of 1999, there was little evidence for
plate tectonic activity on any
body other than the Earth. But in April of that year, NASA announced
results from the Mars Global Surveyor, of a pattern of magnetic
"stripes" running from east to west in Mars's southern hemisphere.
These stripes appeared to be similar to the ones found on either side of
the mid-Atlantic ridge, which played a key role in the final acceptance
of the theory of plate tetonics. The consequences for the overall
history of Mars could be extensive, but as of this writing, the
information is too new for definitive statements.
Thus far, we know of no evidence for plate tectonics on Venus.
The adjective ``tectonic'' may apply to a variety of processes,
including the building of mountains or volcanoes, or
localized (not global) motions of
broken crustal blocks.
There has been extensive "tectonic" activity, on all of the solid
surfaces of planets and satellites in the solar system.
We shall explore some of them in detail.
When rocks freeze, any radioactive elements slowly transform themselves,
from parent to daughter atoms. Some of the daughter element may
have been present at the time the rock froze. If we can determine
the amounts of the parent and daughter both now and when the rock
froze, we
can tell how long ago that freezing took place. This is because
the parent decays to the daughter at a rate that can be measured
in a laboratory experiment.
Let PF be the amount of the parent present when
the rock freezes, and Pt be the amount of the parent
at the time t. Then the law of radioactive decay gives
The constant Of the 92 elements between hydrogen and uranium, all after bismuth
Z=83 have only radioactive isotopes. Moreover, two elements much
lighter than bismuth have no stable isotopes. These two elements are
technetium (Z = 43), and promethium (Z = 61). Several other elements
have radioactive isotopes that are useful for dating materials.
These elements can also supply energy to the interiors of planets and
satellites.
A few examples of important radioactive isotopes are:
C-14 or 14C is useful for dating materials that are no
older than some 30 to 50,000 years. Here is what happens.
Cosmic rays--mostly fast protons--smash into atoms in the
upper atmosphere, and split their
nuclei. In some cases, free neutrons are produced, and these bump into
nitrogen atoms. The most common isotope of nitrogen, N-14, absorbs a
neutron, and emits a proton, becoming C-14. The C-14 then decays with
a half-life of 5730 years.
If the production rate of C-14 remains constant over several
half-lives, then it is straightforward to show that an equilibrium ratio
is set up of radioactive C-14 to the other atmospheric constituents.
Plants take in all isotopes of carbon when they process
CO2, and therefore, a fixed proportion of this carbon is
C-14. After the plant dies, the C-14 begins to decay. We can tell how
old the plant is by how much of its carbon is C-14. The same situation
holds for animals, since they must eat plants, or maybe they eat other
animals that eat plants. Ultimately, all energy to run most life forms
comes from plants or bacteria via photosynthesis, which is the conversion of
CO2 + H2O to carbon-containing molecules.
The C-14 method is not good for deep geological time, that is,
times of the order of hundreds of millions of years or more.
The 50,000 years
mentioned above is only a small fraction of the way, for example, to the
KT (Cretaceous-Tertiary) boundary, corresponding to the extinction of the
dinosaurs.
A method that is much more satisfactory for deep time uses Rb-87,
which decays to Sr-87. We
would expect all rocks to have some Sr-87 initially, that is, before
the Rb-87 started to decay. Some of the Sr-87 therefore comes from the
Rb-87 and some was present to begin with. How do we tell the
difference?
When several minerals are present, as in most rocks, it is possible to
determine both the age of the rock and the initial Sr-87 present in it.
The resulting age is called a "sample age."
In practice, geologists use mass spectrographs (Lecture 13), and
they measure ratios of the parent and daughter isotopes to an isotope
that is not involved in the decay. Sr-86 is usually chosen.
At the time a rock freezes from a liquid, the ratio of Sr-87 to Sr-86
will be the same for all minerals, because these isotopes have the same
chemical properties. What is important in this instance is the radii of
the isotopes, because this determines how they will fit into the
crystals of the minerals making up the rock. While the two strontium
isotopes will have the same ratios in the rocks, the ratio of Rb-87 to
Sr-86 will be different from one mineral to the next because the radii
of Rb and Sr ions are very different. For example, Rb and K have
about the same ion sizes. So the Rb-87/Sr-86
ratio will tend to be high in potassium feldspar
(KAlSi3O8). But in an olivine,
(Mg2SiO4) there wouldn't be much room for the
large Rb-87 ions, and the Rb-87/Sr-86 ratio would be low.
Consider a plot of Sr-87/Sr-86 (y-axis) vs Rb-87/Sr-86 (x-axis).
When
the rock first freezes, the y-points will be the same because of the
identical chemical properties of the strontium isotopes. The x-points
will be different for each mineral.
As time goes on, the x-values of points along the x-axis will decrease
as the Rb-87
turns to Sr-87. The points on the y-axis will increase by exactly the
same amount, since the Sr-87 comes from the Rb-87. In the ideal case,
the points for each mineral will continue to fall on a single line, but
the slope of that line will increase with time. Therefore, the slope
gives the sample age of the rock, that is, the time since it froze.
We can get the initial Sr-87/Sr-86 ratio by looking at the intercept
of the points, that is, the y-value for x=0. This is essentially the
y-value for a hypothetical mineral with no Rb-87 to decay to Sr-87,
so its initial Sr-87/Sr-86 ratio is its ratio for all times.
As time goes on, the total amount of Rb-87 decreases, and the total
amount of Sr-87 increases in the rock. If the rock is softened by heat,
the radioactive "clock may be reset." This can happen if the ions in
the rock can move freely--essentially, the rock becomes a solution
again. The next time the rock freezes, the minerals will again have
the same Sr-87 to Sr-86 ratio, but that ratio will be greater than when
the rock last froze.
Crustal rocks with complicated histories typically have higher values
of the initial Sr-87 to Sr-86 ratios than mantle rocks. And they also
have shorter sample ages.
Geologists also give another kind of age called a
model age. In this kind of age, the
initial Sr-87/Sr-86 ratio is assumed. You may read about the
method from the link.
The USGS has a great page on radioactive dating
and the history of the Earth.
The Earth is a large body, and it certainly seems to be flat. It's
only when there is some way to examine enough of it that becomes
clear it's round (spherical). Many aspects of the physical universe
become fundamentally different when it is possible to view them from
a broad perspective. Our ideas about the nature of living things
changed fundamentally with the use of the microscope. Similarly,
powerful telescopes have enabled us to see that our universe extends
well beyond our solar system and Galaxy.
There is an interesting chapter in the history of geology that illustrates
this. It concerns speculations on the age of the Earth. During the
Middle Ages, most educated people tried to discern information about the
age of the Earth from the scriptures. The famous Irish cleric James
Ussher concluded in the mid 1600's that the
world began in 4004BC. Some 200 years later geological wisdom was that
there was no beginning to the world at all!
The exorcism of the idea of a beginning of the world is often attributed
to two British geologists, James Hutton (1726--1797) and Charles Lyell
(1797--1875). Lyell's Principles of Geology was the definitive
work for many years, and echoes of it remain in geology texts today.
Those scholars whose outlook was based largely on notions of the
``creation'' and the ``flood'' came to be known as catastrophists.
They thought there was a time when the world was different in most ways
from what it was in their time.
At one point, for example, it was ``without form and void.''
Today, relatively few who call themselves
scientists take the this view, but those who do are not difficult
to find on the internet.
Hutton and Lyell were uniformatarianists. The uniformitarian
point of view is probably best expressed in Hutton's poetic words.
He presented a summary of his geological ideas in papers read to
the Royal Society of Edinburgh in 1785, which concluded as follows:
The result, therefore, of our present enquiry is, that we find no
vestige of a beginning,--no prospect of an end. Lyell was somewhat
more cautious. He wrote that any time when the Earth was fundamentally
different from it's present state was outside the bounds of what he
considered to be legitimate science.
There is an irony in this. Hutton and Lyell may be considered
true scientists, who laid the foundations of modern geology. Nevertheless,
from the modern point of view, they were very wrong about beginnings
and endings. Why?
When we review the kinds of information available to geologists in the
eighteenth and nineteenth century, we find it severely limited.
Radioactive dating of rocks did not occur until the early 1900's. Tectonic
activity and erosion erased most of the evidence of the early Earth.
What Hutton and Lyell saw, then, resembled a manuscript, with erasures
superimposed upon
erasures. They had no basis to conclude that the manuscript had once
been empty.
Most of Lyell's geology deals with the most recent 0.6 billion years
of the Earth's history. That time represents the interval in which
bones and shells could be found in the form of fossils. Prior to this
period, soft-bodied life left few traces. Prior to the use of
radioactive dating methods, fossils were the most common tools used
to give relative sequences for layered structures of the Earth.
It is therefore not surprising that geology concentrated on the
rather short time interval for which this tool was available.
It is quite clear that the uniformitarians simply had too restricted
a view to be able to see back to the birth of the Earth. They were
in some ways like the people who looked around and concluded the Earth
must be flat. This is what we mean by a flat-Earth view of Earth history.
The contemporary author John McPhee has written a number of popular
books on geology, and has contrasted views of the past history of
the Earth. Most who based their estimates of Earth history on
religion have rarely thought the Earth was as old as 10,000 years!
Hutton and Lyle--and Darwin--thought in terms of tens and hundreds
of millions of years. McPhee has said that they grasped deep time.
In fact, their perspective was limited. We think the Earth is
some 4.5 thousand million years old.
The late Harvard professor of geology and zoology Stephen
J. Gould wrote a rather severe criticism of the 19th
century stalwarts Hutton and Lyell. The criticisms appear in
one of Gould's lesser-known works, entitled Times Arrow,
Time's Cycle.
The uniformitarian theory of Earth history had no ultimate beginnings
or endings, but there were definitely processes at work. These
processes built and eroded mountains, and created lakes and streams.
With no beginning or end, there was nothing else for these processes
to do but cycle.
The problem with this approach is that it was invalidated by
evidence already known to 19th century geologists in the
form of the fossil record. Indeed, William Smith, by the end of
the 18th century, had made use of fossils to identify geological
strata, and put them in time sequences relative to one another.
Smith used the principle of superposition which goes
back at least to Nicolas Steno (1638-1686). This principle
seems almost like common sense, but was a formidable intellectual
step centuries ago. It says, for example, that of two horizontal
geological strata, say one of sandstone and one of limestone,
the one on top is the youngest. Similarly, if a dike of quartz
passes through these layers, that dike must be younger than
either of them.
In the illustration to the right, the plaque reads: Pegmatite
dikes of different ages which have invaded granite. Note that
the horizontal dike must be younger then the two more nearly
vertical dikes across which it cuts. Clearly all dikes must
be younger than the host rock. The word "pegmatite" refers to
the grain size rather than the chemistry of a rock, but pegmatites
are typically granitic in composition. The pink color is surely
due to potassium feldspar, which is probably more abundant in
the dike than the host granite.
Steven J. Gould made the point that the fossil record, unlike
that of rocky layers and dikes, was monotonic
in time. It was easy enough for Hutton and Lyell to see that
rocky layers could become inverted or melted in such a way that
similar patterns could recur in time. But
fossils in the oldest layers did not recur in the
younger. Indeed, it was possible to use the fossils to estimate
relative ages of geological strata. This provided an "arrow"
for time--a unique direction. Hutton and Lyell thought the
various processes taking place on the Earth went through endless
cycles.
Hutton, and Lyell--until the end of his life--never
accepted the fossil record as evidence that Earth's history
did not go through cycles. For this
Gould takes them to great task.
The story is interesting because Hutton and Lyell are generally
seen as giants, standing at the foundation of modern geology.
One can view their apotheosis in the following way. Check a
modern geology text, and look up the principle of uniformitarianism.
You will probably read that it means that the laws of physics (!)
have not changed in time. This sometimes reminds me of the
one-time adversary of Galileo, Cardinal Bellarmino, who said that if
the scriptures seemed to be at odds with our interpretation, we
need to modify our interpretation.
We have pointed out earlier that belief in the constancy
of physical laws is
virtually essential to the scientific method. Undoubtedly
Hutton and Lyell embraced it too. But they surely meant more
than that by uniformitarianism, and Gould has a valid criticism.
It's one thing to be correct, and yet another to be fair.
It is fair to say that the geological record was still very
incomplete in the time of Hutton and Lyell. This incompleteness
may well have left legitimate room
to believe that eventually cycling might be found in the fossil
record. On more than one occasion scientists have gone out on
a limb, and made assumptions that were necessary to preserve some
concept they felt too dear to relinquish. This happened when
Wolfgang Pauli postulated the neutrino to save conservation of
energy and momentum. It is happening today, when dark matter
is postulated to save the law of gravitation. We know that Pauli
was right. The jury is still out on dark matter.
Seismic waves probe the interior of the Earth. S and P waves
propagate through the mantle, and reveal changes in the physical
state with depth. With the help of tomographic techniques, seismic
data has revealed three dimensional patterns consistent with
convective motions that drive the surface plate tectonic motions.
Plate interactions, including subduction, account for many of the
gross surficial features of the Earth.
Convection cells carry heat that is generated by radioactive
decay. It is not yet clear whether these cells extend into the
lower mantle, or are confined to the upper mantle.
The age of the Earth is now well established by radioactive
dating of rocks. Sample and model ages of rocks are determined
from well-understood principles of radioactive decay.
James Hutton and Charles Lyell established
modern geological principles. They firmly grasped ``deep time''
even though their view was limited.
Their principle of uniformatarianism--a kind of steady state
picture of Earth history--requires some reinterpretation to make it
consistent with modern views of the age of the Earth.
What happened during the formative phase of Earth history, the
hundreds of millions of years missing from the geologic record? The
Moon holds the answer. The Moon is too small to have plate tectonics or
own an atmosphere. There is no mountain building, it never rains, and
rocks don't weather like they do on Earth. Though churned by meteorites
over the ages, the otherwise pristine lunar surface retains a record of
its embryonic development. The scarred and cratered moonscape reveals
what a horrendous time this was.
--J. William Schopf (Cradle of Life, 1999, Princeton Univ. Press)
Physiographic provinces are regions with similar geomorphology,
that is, they are similar in form. A few of the physiographic
provinces of the continental US are shown in Figure 18-1.
It is not necessary to remember all of the names of these provinces.
There are at least 24 of them, but it will be necessary to remember
enough so you get the idea. If you remember the names of the
central lowlands, where we live, and the basin and range
provinces, that should be enough. Basin and Range is the name
of a popular book on geology by John McPhee, who introduced the colorful
phrase ``deep time,'' that we used in the last lecture.
On the Moon, there are only two physiographic provinces, the
(1)highlands and the (2)lowlands. These two provinces
may be seen in a striking
false-color lunar image made by the
Galileo spacecraft on its way to Jupiter. As is explained in the
caption to the figure the highlands show up pink, while the lowlands
are either blue or blue-green. The colors are part of a remote
sensing experiment, which uses techniques that we mentioned briefly in
Lecture 13. We will have more on remote sensing in Lecture 27,
when we discuss the asteroids.
The lunar physiographic provinces can be seen with the naked eye.
They are what makes the features sometimes said to be the "man in the
moon." With only mild optical aid, with binoculars, for example, the
difference in the colors of the Moon's surface are easy to see. The
dark areas are called maria or seas.
Ironically, they are dryer than terrestrial deserts. The highlands
are much whiter, and it is easy to see where the highlands start and
the maria stop.
Here are the names of the major lunar basins,
along with the locations of the Apollo landing sites. The Fall 1996
class had a contest for a mnemonic for the first 8 maria. The
winner was,
going clockwise from Imbrium: I'm Sure That Frogs Never Need Hair Pieces.
To these, you need to add Mare Crisium, not included at that time.
We shall have another contest, which includes Crisium, winner to
receive the usual reward.
The
Fra Mauro crater, within the rectangular outline, marginally
visible on this image, is an important site. We will make a
figure of it below, but you should use your browser to get a
better view.
Prior to the return of the samples from the Moon, no one was sure
what the rocks would be like. There was some speculation that they
would be pristine in their composition, that is, they would be like
the SAD, but without the hydrogen and helium. Other speculations
ran the gamut of possible rock types, from mafic to felsic, but
with a slight preference for the latter, based on the properties
of light reflected from the Moon's surface.
Astronomy books written in the pre-Apollo era described in
exquisite and boring detail the various lunar features, the mountains,
the rills, the ridges, the craters, mostly with no hints about
how these features came to be. One school of thought was that the
maria were low regions that had been filled with dust or chips.
These people thought the first astronauts might sink out of sight
into dust layers possibly up to a mile in thickness.
The returned samples from Apollo 11 (the first)
told much of the tale. It
landed in Mare Tranquillitatis, so the majority of their samples
were characteristic of the lowlands, or mare. These rocks are
basalts, similar in nature to flood basalts known on the Earth.
Terrestrial flood basalts are found in eastern Washington
state, as well as at the Snake River plain in Idaho. The mafic
lavas flowed over great distances because of their low viscosity.
This is true for both the Earth and the Moon.
Among the Apollo samples were some fragments of nearly pure
plagioclase feldspar. The American astronomer John Wood then predicted
that this kind of rock, an anorthosite, would dominate the composition
of the highlands. Subsequent missions confirmed his prediction.
Some rock classifications group the anorthosites with the gabbros.
Gabbros, generally speaking, are the coarse-grained counterparOptical Spectrometers
Energy Levels and Line Spectra of Atoms and Molecules
must be
just equal to the difference in the energies of the two levels
involved in the jumping. If we have two allowed energies,
En and Em > En, then the frequency of a photon,
that will be emitted will be given by the relation
=
Em-En/h strikes
an atom in the lower level n, the atom may absorb the photon, and jump
to the level m.
Radio Telescopes
/D holds for their resolving power, but it is
much lower than for
optical telescopes. This is because the wavelength of
radio waves is many orders of magnitude greater than optical light, and
the large primaries do not compensate. So /D
is generally much
bigger for a radio telescope than an optical one. Radio astronomers
compensate for this by combining observations from telescopes separated
by large distances--sometimes thousands of miles. The technique is
called interferometry, and by using it, radio instruments can
resolve down to 0.0001 seconds of arc. This is now much smaller than
the resolution of optical telescopes!
Specialized Detectors
-rays.
They have built a variety of special instruments to detect these
photons, which we shall only mention briefly. X-ray telescopes
can resemble optical telescopes in that the light may be brought
to a focus and analyzed. Depending on the energy of the
-rays,
these telescopes make use of the interaction between the
-rays
and atoms or nuclei. The most energetic -rays
cause showers
of secondary particles to form in the atmosphere. These may be detected
at ground level with specialized instrumentation that need not concern
us in this course.
Summary
-rays, X-rays, visible light, infrared radiation,
and radio waves. Astronomers traditionally used optical telescopes,
first refractors, and then reflectors to catch visible starlight. With
the help of satellites and radio telescopes they now investigate the
entire electromagnetic spectrum. The light-gathering and resolving
power of a telescope depends on the aperture (size) and in the latter
case also the wavelength of the light received. Radio telescopes
receive and amplify cosmic radiation at long wavelengths. The
light-gathering and resolving power of these instruments are the
same as for optical telescopes.
Lecture 14 Laboratory Analysis of Cosmic Materials
Wet Chemical Analysis
The Mass Spectrograph
Chromatography
Neutron Activation Analysis (NAA)
Spectroscopy of Flames and Hot Sources
Molecular and Microwave Spectroscopy
Thin Sections and the Polarizing Microscope
Electron and Ion Probes
particles at a small region, and examine the
results. Many of the 's will orbit the nuclei
of atoms in the sample, and return in nearly the direction they
came from. This is called backscattering.
The energies of the backscattered 's depend
on the mass and charges of the nuclei they interact with. One can
use this technique to determine the composition of the lighter
elements in the sample. This method was used to analyze martian
rocks.
Summary
backscattering has recently been used to analyze Mars rocks.
Lecture 15 - Down to Earth I: Minerals
Rocks, Rocks, look at those rocks... We had gone to Mars to look at
rocks... Why did we want rocks? Every rock carries the history of its
formation locked in its minerals, so we hoped the rocks would tell us
about the early Martian environment. The two-part Pathfinder payload,
consisting of a main lander with a multispectral camera and a mobile
rover with a chemical analyzer, was suited to looking at rocks.
Although it could not identify the minerals directly--its analyzer could
measure only their constituent chemical elements--our plan was to
identify them indirectly based on the elemental composition and the
shapes, textures, and colors of the rocks. By landing Pathfinder at the
mouth of a giant channel, where a huge volume of water once flowed
briefly, we sought rocks that had washed down from the ancient, heavily
cratered highlands. Such rocks could offer clues to the early climate
of Mars and to whether the conditions were once conducive to the
development of life.
The Terrestrial Planets and Asteroids: Rock and Iron
A Simplified Classification
Three Silicate Families
H
|
H--C--H
|
H
Mg_2SiO_4 Fe_2SiO_4
Forsterite(Fo) Fayalite(Fa)
| |
-------------------------------------------
100%Fo 100%Fa
0%Fa <-- Olivine--> 0%Fo
Partial Melting: A Form of Magmatic Differentiation
Geochemical Maturity: The Bowen Series
Olivines Anorthite
Pyroxenes Albite
Amphiboles K-spar
Micas
Quartz (SiO_2)
clay minerals
(soil)
Physical Properties of the Minerals
Mineral Formula Melting T(K) Density
Water=1
Forsterite Mg SiO 2163 3.21
2 4
Enstatite Mg Si O 1830 3.19
2 2 6
Diopside CaMgSi O 1668 3.30
2 6
Anorthite CaAl Si O 1830 2.76
2 2 8
Albite NaAlSi O ~1313 2.63
3 8
Summary
Lecture 16 - Down to Earth II: Rocks
Classification of Igneous Rocks
The Pathfinder's Analysis of Martian Rocks
Rock Names and History
The Chemistry Volcanic Lavas
Summary
Lecture 17 - Down To Earth III: Earth Structure
Waves and Seismology
The Earth's Core and the Shadow Zone
The Inner Core
Cat Scanning the Earth, Convection, and Plate Tectonics
Radioactive Dating
t)
in this formula
is simply related to the
half-life. Put Pt/PF = 0.5 in the above
formula, and solve for t = t1/2, the half-life. We
readily find t1/2 = 0.693/
Isotope half-life (years)
C-14 5730
K-40 1.28 x 10^9
Rb-87 4.8 x 10^10
U-235 7.04 x 10^8
U-238 4.47 x 10^9
Deep Time and the Age of the Earth
Time's Arrow and Cycle
Summary
Lecture 18 - Earth and Its Nearest Cosmic Neighbor, The Moon
Physiographic Provinces of the Earth and Moon
Numbers refer to x I - Mare Imbrium
Apollo landings x x S - Mare Serenitatis
x I15 S x T - Mare Tranquillitatis
x 17 x F - Mare Foecunditatis
x P 11T c x N - Mare Nectaris (to east)
x 12 14 16 x N - Mare Nubium (to west)
x N F x H - Mare Humorum
x H N x P - Oceanis Procellarum
x x c - Mare Crisium
x
The Simplified Petrology of the Moon
