Stories of the Lithosphere
How Do We Know: Part 2
Exploring Places We Can't Go--What Are Stars Made Of?
Earth Science Essentials
by Russ Colson
Lecture Recap
Electromagnetic Spectrum
Since we can't go to the stars, it's necessary for the stars to come to us—that is, the light from the stars comes to us. It is the light from the stars—the electromagnetic radiation--that reveals the secrets of the star, including its temperature and composition. The electromagnetic spectrum consists of much more than visible light, as shown in the illustration below.
Illustration of the electromagnetic spectrum, showing relative and actual differences in wavelength. Cell phone wavelength is about one-third of a meter.
Temperature of Stars
People have noticed the color in stars for thousands of years. For example, Ptolemy referred to red and yellow stars in his writing of the Almagest, a work that remained a premiere guide for Greek, Arab, and European astronomers from the 2nd century AD to the beginning of the 17th century AD.
The early attempts (1863-1868) of the Jesuit priest Father Angelo Secchi to classify stars based on spectra revealed that the stars segregated into groups of like color. He found that stars could be divided into four types: the white or blue stars, the yellow stars, the red stars, and a second type of red star, the fiery red stars. Modern classifications of stars show similar color groups.
So, does color of a star tell us anything about the nature of the star? Is it possible to tell something about a star simply by what it looks like from a distance?
Common experience indicates that temperature of hot objects and their color are related. Convince yourself. Heat some of the fire-resistant 'cotton' used in gas fireplaces with a propane torch. As you move the torch closer, the temperature of the heat resistant 'cotton' increases and with that increase in temperature the color of light emitted from the heated material changes from dull red, to red, to yellow, to white. Or, watch the colors in your campfire. Relatively cool areas may glow red, hotter areas and flames glow yellow, and, in very hot areas deeper in the fire, coals and flames may appear white or blue. Stop your car in front of your closed garage doors and flip from high beams to low beams. The hotter, high beams appear bluish. The cooler low beams appear reddish (as long as you're using an old-fashioned filament light and not LED lights).
All things 'glow' because of the heat that is in them. The colder an object is, the more of the 'glow' is in the longer wavelengths of light. If an object is sufficiently cool, it will 'glow' in a part of the spectrum that we can't see with our eye. Objects whose temperatures are less than a few hundred degrees Celsius will 'glow' in the infrared region of the spectrum. You can't see infrared radiation. However, notice that you can continue to feel the heat radiating from the burner on your electric stove long after it has stopped glowing red. This heat radiating from the burner is infrared radiation. Even the Earth is 'glowing', radiating energy into space in wavelengths we can't see. Because the wavelength of radiated light is a function of temperature, the color of a star (that is, the light we can see) is a function of its temperature. Based on your own experience with how things glow when hot, and the discussion above, is the blue-white star Vega warmer or colder than our yellow Sun?
The 'glow' of objects due to their temperature is called blackbody radiation. The name 'blackbody' comes from the idea that a theoretical blackbody would not reflect any light, and thus its temperature curve would be due only to the energy it was emitting, not to light it reflected. A chart showing how the peak wavelength of blackbody radiation changes with temperature is shown below.
Our Sun has an effective blackbody temperature of about 6000K.
You might wonder why it doesn't appear blue, since its peak emission is at that wavelength, as seen in the figure above.
The reason has to do with our eye. Because we see a mix of colors together, our eye emphasizes some colors over others, making our sun appear yellow.
In general, the color that our eye sees is toward longer wavelengths from the blackbody peak of a particular star.
In lecture, we looked at some red and blue stars that are visible in the northern hemisphere summer sky. To see red and blue stars in the winter sky, look for the constellation Orion, shown below.
Orion is perhaps the easiest-to-find constellation in the sky.
Composition of Stars
The ancient Greeks could not observe the detailed spectra of the stars because they did not have the tools to do so. To them, the stars appeared geometrically perfect, immutable and unchanging. This suited their view of the heavens as a place without new life or death, without beginning or ending. But because all the things on Earth changed, with beginnings and endings and transitions, it necessarily followed that the stars must be made of something very different from what we find on Earth. The Greeks called this different stuff quintessence (meaning 'fifth element').
We didn't have a better idea of what stars are made of until the invention of the spectroscope in the 1800s. A spectroscope works on the principle of refraction that we studied in the previous unit. When light passes through a prism, the light refracts (bends) because its velocity is different in the prism than in air. Each wavelength of light has a different velocity in the prism, which causes each wavelength to refract by a different amount, splitting the light into its constituent wavelengths. Shorter wavelengths travel slower than longer wavelengths so they get refracted more.
Illustration of the concept of a spectroscope—light is split into its various wavelengths by passing through a prism which refracts different wavelengths differently.
Near the beginning of the 1800's, William Wollaston and Joseph Fraunhofer were trying to make sense of the spectra of a variety of colored flames. As part of their studies, they each examined the spectra of the sun by passing sunlight first through a narrow slit and then through a prism. They both noticed that the spectra from the sun were crossed by a number of dark lines. Fraunhofer went on to realize that these dark lines were the signature of elements in the sun.
Each element absorbs or emits specific wavelengths of light when electrons in the atoms move to different energy states. The fingerprint of each element can be determined in the laboratory by examining the spectrum for that element, as shown below for hydrogen.
Hydrogen absorbs specific wavelengths, leaving dark lines in the spectrum. Each element leaves lines that are different from every other element, allowing the elemental make-up of stars to be determined.
The figure below shows the spectra for different types of stars. This figure comes from a lithograph published in 1870, as reproduced from the Cambridge Illustrated History of Astronomy, M. Hoskin (Ed.) Cambridge Univ. Press, 1997.
Notice the many lines in the spectrum from our Sun. In the 1800s, scientist lacked an understanding of what was causing the lines (that awaited the discovery of quantum mechanics in the 20th century), however, they were able to match up the lines with the lines for known elements and figure out what the stars were made of.
An interesting test of the belief that the dark lines in the spectrum of the sun are related to particular elements was found in the presence of a "mystery fingerprint" in the sun's spectrum, dark spectral lines that did not correspond to any element that was known in the late 1800's. If the dark lines really corresponded to absorption by particular elements, then there had to be some previously unknown element in the sun. This possible new element was named Helium (after helios, the Greek word for the sun). The element helium was later isolated and identified on Earth, and its absorption spectrum matched the mystery lines seen in the sun's spectrum. The mystery element discovered in the sun, where no one has ever been, now fills our balloons here on Earth.
Puzzle 1: Composition of blue star Sirius (similar to Vega, one of the stars in the summer triangle).
Qualitative analysis puzzle:
The spectra for the sun, and stars like it, is quite complex, with many elements. However, the spectrum for Vega, Sirius, and other blue stars is much simpler (at least as long as we restrict ourselves only to the most dominant lines in the spectrum).
Given the information below, can you identify the elements in the blue-white stars? Selected elemental spectra and a spectrum resulting from absorption of light passing through Earth's atmosphere are shown. The spectrum for Sirius is modified from the lithograph published in 1870. Spectral lines are simulated from data in the CRC Handbook of Chemistry and Physics, 63rd and from James B. Kaler, Stars and their Spectra, Cambridge Univ. Press, Cambridge, 1989.
Puzzle 2: First measurement of the atmosphere of an exoplanet
Exoplanet is the name we give to a planet outside our own solar system. Despite all the worlds of Star Wars and Star Trek and all the other science fiction stories, humans had never observed a world outside our own solar system (the 8 planets plus dwarf planets) until the mid-1990s.
It took us a bit longer to observe the atmosphere of an exoplanet. The first atmosphere of a planet outside our solar system was reported in 2002 in the paper "Detection of an extrasolar planet atmosphere" (2002) by David Charbonneau, Timothy Brown, Robert Noyes, and Ronald Gilliland, in The Astrophysical Journal, 568:377-384. The puzzles below have been simplified from that paper.
The methodology for detecting the atmosphere of the planet was to wait for the planet to pass in front of its star and then measure the spectrum of light from the star after it had passed through the planet's atmosphere. Each element in the atmosphere will absorb characteristic wavelengths of light, allowing us to figure out the composition of the atmosphere. The problem is, the planet and its atmosphere are tiny, with only a small influence on the light, while the star itself is big, having the biggest effect on the light. Charbonneau et al solved this problem through careful measurements with the Hubble Telescope combined with careful analysis of the uncertainty in the measurements to make sure that the variations they observed were due to the planet's atmosphere and not due to some random fluctuation in light.
The graph below comes from their paper. The x-axis plots wavelength in nanometers (billionths of a meter) and the y-axis plots a measure of the amount of light arriving on Earth at each wavelength. The graph is centered on the characteristic wavelength of the element that they studied.
Considering the spectrum below showing the characteristic absorption bands of selected elements, and remembering that elements in the atmosphere of the planet—and the atmosphere of its star—absorb light of characteristic wavelengths, what element did Charbonneau et al choose to study?
To measure the effect of absorption by Na in the planet's atmosphere, the authors
1) took the difference between the flux of light at the Na peak and the flux of light to either side of the peak (this compensates for the fact that the light at all wavelengths will decrease when the planet is in front of its star).
2) Then they took the difference for that value between when the planets is not in front of its star and when it is in front of the star (this allowed them to detect a very tiny variation in the darkness of the Na line that would not be big enough to see if they plotted it on their graph above).
Below is figure 4 from the Charbanneau et al. paper, a graph showing their key results. Read it with the goal of understanding what it tells us. The data shown are for multiple measurements made during three separate transits of the planet (a transit is when the planet passes in front of its star). The language 'binned' and 'unbinned' refers to whether multiple data points within a range of time (bins) have been averaged (binned) or not (unbinned).
When you have thought through the graph and what it means, proceed with the puzzles below.
Scientists studying exoplanets use the word 'contact' to refer to specific locations in a planet's orbit. For example one contact is the moment when the transitting planet starts to move off the face of the star. Another is when the planet is completely clear of the star. Another contact is when the planet begins to move back onto the face of the star. Another is when the planet moves completely onto the face of the star.
The detection of Na in the atmosphere of HD2209458b has been confirmed by ground-based observations: "Ground-based detection of sodium in the transmission spectrum
of exoplanet HD209458b", (2008) by I. A. G. Snellen, S. Albrecht, E. J. W. de Mooij, and R. S. Le Poole, in Astronomy and Astrophysics, 487, 357-362.
In order to convert these measurements of Na absorption into a concentration of Na in the planet's atmosphere, it's necessary to
1) make an estimate of how much atmosphere the light passes through (affected by size of planet and presence of clouds, etc)
2) Calculate the thermodynamic state of Na in the atmosphere of the planet (which affects absorption)
3) compare the results with a standard—such as a measurement on Earth of how much light is absorbed when it passes through a known amount of Na (all chemical analyses rest on comparing something that you don't know to something that you do know—a standard).
The complexity and uncertainty in the first two steps mean that we probably can't calculate a concentration yet from these measurements of absorption in the Na band.
Puzzle 3: Organic molecules in the atmosphere of an exoplanet
People are always interested in the possibility of life on other planets. Spectroscopy might eventually provide our first clues to life on distant planets, by allowing us to detect organic molecules or presence of oxygen.
We have begun looking at the atmospheres of exoplanets using spectroscopy to study the molecular composition (different from looking for elements, like Na, which we did above). To look at molecules, we have to examine parts of the spectrum with longer wavelengths than visible light. We look at the infrared.
The puzzle below comes from the paper: Water, methane, and carbon dioxide present in the dayside spectrum of the exoplanet HD 209458b (2009) by M. R. Swain1, G. Tinetti, G. Vasisht, P. Deroo1, C. Griffith, J. Bouwman, Pin Chen, Y. Yung, A. Burrows,L. R. Brown, J. Matthews, J. F. Rowe, R. Kuschnig, and D. Angerhausen, The Astrophysical Journal, 704:1616-1621
Swain et al looked at the emission spectra of giant exoplanet HD 209458b (the same planet we looked at above). An emission spectrum looks at the energy that the planet is giving off (remember that everything is glowing?). The emission spectra is a function of 1) temperature (which affects blackbody radiation and also the speciation and density of the molecules), 2) the kinds of molecules present and the wavelengths they emit, and 3) the kinds of molecules present ant the wavelengths that they absorb.
With Infrared spectroscopy, it isn't usually possible to identify single absorption or emission lines and associate them with particular molecules because all the molecules together, combined with variations in temperature, have to be taken into account to explain the observed spectrum. Thus, scientists often create synthetic (calculated) spectra using various proportions of molecules at various temperatures and see which synthetic spectrum matches the observation. This puts constraints on what can be there, although there are sometimes multiple solutions (such as: 'there might be x amount of methane at temperature y, but if the temperature is y', then the amount of methane will be x').
The results below demonstrate that we can measure the presence of certain molecules in the atmosphere of HD 209458b. Can you read the graphs and figure out what they say?
In this graph, the black bars are the observational data with statistical uncertainties. The colored curves are the synthetic spectra—the black one is for presence of water only, the green is for presence of both water and methane, and the red is for presence of water, methane, and carbon dioxide.
The figure below from the paper shows a wider range of wavelengths (still in the infrared) and more complex synthetic spectra. Due to an error in this figure in the original paper, I have taken this figure from Tinetti et al, (2010) Exploring extrasolar worlds: from gas giants to terrestrial habitable planets, Faraday Discussion, 147, 369-377.
The temperature curves in this figure show blackbody radiation—so you can see how the emission spectrum varies from simple blackbody radiation but follows its general slope. Data from three different sources are plotted as points with uncertainty bars. Synthetic spectra are plotted as colored lines, each taking into account different temperatures, concentrations of methane, water, and carbon dioxide, and tropospheric pressure. Three models include presence of a temperature inversion in the troposphere, and one does not.
These planets are a long ways away, and there are many factors in analyzing the composition of the atmosphere that we can't measure, such as temperature or atmospheric pressure. Thus, we have a ways to go yet before we can reliably and exactly measure the composition of the atmospheres of exoplanets. However, we are on the way!
Last updated Sept 16, 2015. All text and pictures are the property of Russ Colson, except as noted.
.
