Click for next page ( 168


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 167
CHAPTER V Instrumentation in Chemistry All scientific knowledge is rooted in our abilities to observe and measure the world around us. Thus, science benefits enor- mously when more sensitive measuring techniques come on the scene. This is the situation in chemistry today. The discussions to follow will identify a number of powerful instrumental methods that are now the everyday tools of re- search chemists. We will focus on the capabilities of today's instruments and on how much they have changed over the last decade or two. ~67

OCR for page 167
Al Hi, A Laser Flashlight A laser flashlight? Sounds like something out of Buck Rogers or Star Trek! What would that be? Well, to bring this down to Earth, let's first think about what a laser is and then how to make it into a flashlight. Lasers are very special light sources. They put out pencil-sha~p beams of pure color end so intense they can be used to cut patterns in steel. Also, they can be focused so sharply they are better than a surgeon's knife in mending the retina in your eye. Finally, they can give light pulses as short as a millionth of a millionth of a second! That's what's called a picosecond. With shutter speeds that fast, chemists can now `'photograph'' the fastest chemical changes known. And how do lasers work? It all begins with a whole bunch of atoms or molecules all ready to emit light of exactly the same color. Atoms and molecules usually absorb light, not emit it, so somehow we've not to DUmD them UD in energy SO Hat ~ ,&. ~ &. ~ ~ they're more inclined to emit than absorb. This is called a "population inversion.' Once we've got a population inversion, there are some tricky things to do with mirrors to make a laser out of it but we don't have to go into everything. So how do we pump up the molecules to get that population inversion? One good way to do it is to use electrical energy, like we do In a fluorescent light. That's what you might use to light up a dark closet to look for your missing sneaker. And it works fine as long as you have a long enough extension cord. But think of looking for your jacl: ~ the back of your car when you have a flat out on a dark highway. That's where a flashlight comes In handy. In a flashlight, the energy comes Tom a chemical reaction. That's what the batteries are all about. Could we use a chemical reaction to pump a laser? If so, it would be a "chemical laser,'' our laser flashlight. But that would require a chemical reaction that produces a population inversion. Trying to find out whether there are such reactions led chemists to the discovery of the first chemical laser. O[course, it's hardly news that chemical reactions can emit light. Candles do it all the fume. And thirds about a fireflyhe (or she) Carl do it without an extension cord. These emissions show that a reaction follows special pathways, and when energy Is released, these preferred pathways might be a super ~ ~ way to get a population inversion. Mar W '599~ ._ . _ 'it '~_- ~ ._ 168 I, The suspense, though, was that chemical lasers were not discovered by looking at bit flames or by copying the firefly. Chemical lasers were discovered to operate best In the infiared, where the human Or eye is blind. This spectral region is where molecular vibrations ` ~ can cause molecules to absorb or emit light. From these lasers cat . . . ` we reamed that quite a few reactions prefer reaction path- .~ Sways that put most of the available energy into vibrational `~' motions of the final products. Why this happens still isn't "I clear, but we're working on it. In the meantime, we have a whole bunch of fine chemical lasers. They can be very efficient, which has made at least one chemical laser a candidate to be the match to lift the nuclear fire of nuclear fusion. If that worked, chemical lasers would help us get '~clean" nuclear energy for the rest of tune. Chemical lasers can also be very intense, as shown with the fluonne-hydrogen flame laser. What good is that? Well? that gets us back to Buck Rogers, Star Trek. and Star Wars. If you want a laser out in space, you'll either need a chemical laser or a mighty long extension cord. So wraths next? We're still after that firefly.

OCR for page 167
V-A. INSTRUMENTATION FOR STUDY OF CHEMICAL REACTIONS V-A. Instrumentation for Study of Chemical Reactions Section IV-A indicated how the chemist's use of the most modern instrumenta- tion is making it possible to investigate even the fastest chemical processes in intimate detail. We are witnessing a quantum jump ahead in our understanding of the factors that control the rates of chemical reactions. Among the tools respon- sible for this rapid advance are lasers, computers, molecular beams, synchrotrons, and, on the horizon, free-electron lasers. We will consider each in turn. LASERS Chemical lasers have been discussed in "A Laser Flashlight" on page 168. For a laser to work, it needs a "population inversion" in which there are more molecules that have enough energy to emit light than there are molecules ready to absorb light. To maintain such an inversion, energy must be injected somehow. Some energy-releasing reactions do this (resulting in chemical lasers), but the energy can be injected in other ways. The simplest way is through irradiation with a conventional light source. However, electrical energy input is probably the most convenient way to establish a population inversion. The ap- paratus need not be much dif- ferent from a fluorescent light fixture. Whatever the manner of en- ergy input (the "pumping" method), the special qualities of laser light arise from "stimu- lated emission," which can be regarded as the inverse of light absorption. A photon of light of the exact energy needed to excite a molecule from one energy level to another, higher level can stimulate emission of a second photon from a molecule already in the higher level. The second photon that is thus produced turns out to be perfectly in-phase ("coherent") with the electromagnetic wave of the first photon that started it all. This coherence gives lasers their distinctive character. It accounts, for example, for the pencil- sharpness that permitted us to reflect a laser `'searchlight" beam offa mirror placed on the Moon by the Apollo astronauts. The remaining feature of a laser is a set of accurately focussed mirrors that cause any stimulated emission to go back and forth many times through the population inversion. These mirrors are called an optical cavity; they permit and cause the buildup of the special qualities of laser light. Lasers bring to mind a brilliant beam of light cutting through a sheet of steed or shining deep into space. But to a scientist, the beauty of the laser lies in its ability to deliver light of extremely high intensity, extremely high power, extremely high spectral purity, and/or extremely short duration. For a given experiment, laser design is dictated by the one of these features of greatest value to the experiment at hand, and usually at some sacrifice in the others. Some of this trade-off is PUMP I NO LIGHT hv~ coo o Font ax' 0 0 0 o E2 LIGHT ABSORPTION CAN ESTABLISH A POPULATION INVERSION 169

OCR for page 167
170 MIRROR F LUORESCENT M I RROR Dl SCHARGE Z=~=~ ~ .. ~ AM I TG.NT ~ _ _ HIGH VO LTAGE ~t ~ ~ A LASER NEF.D NOT BE; COMPLICATED INSTRUMENTATION IN CHEMISTRY imposed by the Uncertainty Principle. This fundamental premise of quantum mechanics states that the duration of a light pulse is related to and limits the spectral purity. Thus, the Uncertainty Princi- ple tells us that if a pulse is as short as one picosecond (10-~2 seconds), there will be an uncertainty in the frequency (the color) at least as large as 5 cm-. With this much frequency spread, most information is lost about molecular rotations of gaseous molecules. On the other hand, if a line width of 0.005 cm~~ is needed to detect individual rotational states, then the molecule of interest must be examined by a light pulse at least as long as one nanosecond (10-9 seconds). This limitation depaves us of time information about species or events with shorter lifetimes than the nanosecond probe. PULSE DURATION ~ > SPECTRAL PURITY ,- ONE MICROSECOND = .000 OO1 SECOND ONE NANOSECOND S .000 OOO OO 1 SECOND ONE PICOSECOND = .000 OOO OOO OO 1 SECOND ONE FEMTOSECOND - .000 OOO OOO OOO OO 1 SECOND - .ooooo5 cm l .005 cm l < - 5 cm 5OOO cm l PULSE DURATION LIMITS FREQUEN CY ACCURACY AND V ICE VERSA Developments in the Last Decade There were three crucially important developments in laser technology that took place during the 1970s and they are having a great impact on chemistry. First, several types of tunable lasers were developed and became commercially available. A "tunable laser" is one whose color (wavelength) can be selected according to need. The wider the range of the spectrum that lasers can work over, the more valuable they are as a research tool. The most important of these was the dye laser, which gave continuous color tuning throughout the visible region of the spectrum and a bit beyond into the near-infrared and near-ultraviolet. Dyes are chemical compounds whose intense color causes them to absorb light efficiently so that they can then emit coherent laser light. Second, the invention of efficient ultraviolet lasers gave scientists access to the photochem~cally important ultraviolet region at wavelengths shorter than 300 mm. These include "excimer lasers'' that are based upon light emitted from molecules formed from electronically excited reactants. An example is the krypton fluonde laser. Krypton is an inert gas that does not form bonds in its ground state. After one of its valence electrons is excited, however, the

OCR for page 167
V-A. INSTRUMENTATIO1V FOR STUDY OF CHEMICAL REACTIONS resulting krypton atom has the chemistry of rubidium. Thus, the molecule formed between Kr and F has the bond strength and stability of RbF. This is a desirable factor in building up concentration to reach a popu- lation inversion, so that it can emit laser light. The third de- velopment was the discovery of methods of laser operation that gave short-duration light puIses~ne picosecond or Kr electron Krypton ' coll~S'~ (ground state) 5s O 4p Is 3d ~ ~ 3) ~ fin -17~1 Kr. Krypton (excited state) Hi, ~0 }fir (gas) + F2 (gas) ~ (KrF)~(gas) + F(gas) Pb~gas) ~ F2(gas) - > RbF(gas) + F(gas3 Rb Rubidium (ground state) 0 AFTER EXCITATION KRYPTON REACTS LIKE RUBlI)IUM THAT MAKES EXCIMER LASERS POSSIBLE less. In 1970, the tunable dye laser did not exist except as a laboratory curiosity. In the early 1980s, almost every chemistry research laboratory had more than one tunable laser source. Tunable lasers can now be conveniently operated over the wavelength range from 4 microns in the infrared (40,000 A) to 1,600 A in the ultraviolet beyond the wavelength at which air becomes opaque (i.e., into the range called the "vacuum ultraviolets. Already in the state-of-the-art stage are lasers that extend the wavelength range to beyond 20 microns (200,000 A) in the infrared and to less than 1,000 A in the vacuum ultraviolet. Chemical Applications Table V-A-l lists many chemical applications of lasers. It is important to note that most of the more powerful lasers are not continuously tunable; they have only particular output wavelengths. They are most useful in the study of solid matenals, which will usually absorb a wide range of wavelengths of light. For most chemical applications, tunable sources are critically important, and these lasers are often TABLE V-A-1 Some Research Areas Utilizing the Laser Area Research Application Solar energy research, photosynthesis Uranium, plutonium isotope purification Trace element analysis, environmental monitoring Probing flames, explosions Monitoring industrial processes Cell discrimination and separation Photochemistry within biological cells Gas-phase deactivations, Chemical reactions Laser Used - Excimer, dye Excimer, dye, TEA CO2 Continuous ion, color center Photochemistry Isotope separation Atomic absorption, fluorescence Combustion diagnostics Atmospheric gas analysis Biological cell sorting Cell bleaching Microsecond kinetics (1-100) x 10-6 see Nanosecond kinetics 10-6 to 10-9 see Picosecond kinetics 10-9 to 10-'2 see Subpicosecond kinetics <10-'2 see Excited state lifetimes, very fast reactions Fast electronic state deactivation, coherence decay in liquids Vibrational deactivation in solids and liquids Solid-state, dye Semiconductor diode Ion laser Dye laser Flashlamp dye, TEA CO2, chemical Solid-state, excimer Ion, solid-state Ion, solid-state

OCR for page 167
172 INSTRUMENTATION IN CHEMISTRY excited with another powerful, single-frequency laser. Having the best suited laser system is essential for work at many of today's most exciting chemical research frontiers. COMPUTERS The use of computers by chemists has paralleled the tremendous computer development of the last three decades. The size of this growth is reflected in the number of industrial installations of the largest IBM computers over this same time penod. In the mid-1950s, there were 20 or 30 such machines (IBM 701s). By the mid-1960s, the much more powerful 7094 and 360 systems numbered about 350. Today, there are perhaps 1,700 indus- tnal installations of IBM 3033s. This numerical growth has been accompanied by a phe- nomenal increase in computer power. The extent to which chemis- try has benefited from this growth can be seen by compar- ing two landmark calculations. For polyatomic molecules, the first theoretical calculations based upon the Schroedinger Wave Equation without any simplifying assumptions (an ab initio calculation) appeared in the 1960s. Of special impor- tance was the study of rotation around the carbon-carbon bond of ethane, C2H6. As the hydrogen atoms at one end rotate past the hydrogen atom at the other end, the energy rises to a maximum. To learn the height of this internal rotation barrier, theoretical cal- TABLE V-A-2 Relative Computing Speeds of Computer Levels 2000 ,_ 1 500 <5: I: cat z ~ Too lo: - o ~ 500 :~ MA I NFRAME COMPUTERS _ _ t ~ 1 1 1 1 1 1 1 1 t I I I 1 960 1 970 1 980 YEAR INDUSTRIAL USE OF LARGE COMPUTERS Computing Superminicomputers Mainframes Supercomputers Example DEC VAX 11/780 IBM 3033 CRAY IS (1) 10-15 80-120 culations ("self-consistent field" method) were based upon a ba- Relative SiS set of 16 functions. Speed This can be contrasted with a recent and similar study of deca- methy! ferrocene, ECs(CH31512Fe. This calculation used a basis set of 501 functions. Since such stud- ies require computing effort proportional to the fourth power of the number of basis functions, the decamethy! ferrocene computation involves (501/1614 or one million times more computation than the ethane problem! \

OCR for page 167
V-A. INSTRUMENTATION FOR STUDY OF CHEMICAL REACTIONS Superminicomputers This level of computer has become a workhorse in chemistry. Instruments like the DEC VAX 11/780 are comparable to the worId's largest mainframe computers available in the late 1960s. They have revolutionized computing in chemistry because of their substantial capacity, high speed, and lowered cost, which is now in the range $300,000 to $600,000. The last 20 years have also seen three important develop- ment phases for the use of com- puters in chemical expenments. In the first, computerization phase, advances in both hard- ware and software greatly im- proved our ability to accumulate measurements (data acquisi- tion). Then an automation phase increased the possibilities for experiment control through continuous morutonug of cntical parameters. F~nady, a "knowledge engineer- ing" phase ushered in an era in which computers perform high-level tasks to interpret collected information. An excellent example is the Fourier Transform aIgonthm which permits us to record spectral data over a long time period, thereby to achieve high speck resolution. Because this allows detection of quite weak signals, this aIgonthm is now routinely used to record ~3C NMR sign~s and to transform infrared inte~erograms. Because of the success of these instruments, the Fourier Transform aigonthm is now being incorporated into ad sorts of equipment: electrochemical, microwave, ion cyclotron resonance, dielectnc, and solid-state NMR instrumentation. \ ASH C_H me, ETHANE 1 6 FUNCTIONS CH3` C ~ _CH~ - ~ -C CH3 C - C CCH3 HA ~ , HI ~ CH3Cx C ACHE ARC . '-Can CH , CH3 3 (50 1)4 _ 1 0 DECAMETHYL FERROCENE 50 I FUNCTIONS A MILLION-FOLD ADVANCE IN TWO DECADES Mainframe and Supercomputers Some needs for computation in chemistry can be met only with the greater capacity and capability of the largest scientific computers (Cray/M and X-MP or CYBER 205), coupled with specialized resources such as software libranes and graphics systems. This is most notably true for electronic structure studies for many atom molecules beginning with the complete Schroedinger Equation and without approximations tab initio calculations). Another area that will benefit from supercomputers is computational biochem- istry. Most dynamical simulation procedures applicable to biological molecules require calculation of the simultaneous motions of many atoms. A conventional LOO-picosecond molecular dynamics simulation of a small protein in water would require about 100 hours on a DEC VAX Il/780 or 10 hours on an IBM 3033. Calculations of the rate constant for a simple activated process require a sequence of dynamical simulations to determine the free energy barner, and additional simulations to determine the nonequilibrium contributions; the times can now 173

OCR for page 167
174 INSTRUMENTATION IN CHEMISTRY reach 1,000 hours on a DEC VAX 11/780. More complicated processes or longer simulations become impossible without the much higher speeds of supercomputers. MOLECULAR BEAMS The advances of vacuum technology over the last three decades have made it possible to reduce the pressure in an experimental apparatus to a point at which molecular collisions become quite improbable (e.g., at pressures below 10-9 torr). Under these conditions, molecules that enter the vacuum chamber stream to the opposite chamber wall without deflection. Such a situation is called a "molecular beam." This provides a special opportunity to study chemical reactions. The most obvious application is to cause two such molecular beams to intersect. When a molecular collision does occur, it is almost always in this intersection zone. If the collision causes a chemical reaction, the product fragments leave the reaction zone with energies and directions that provide information about the reactive collision. By measuring the spatial distribution and fragment energies, we can learn intimate details about single-collision chemistry. Capabilities A typical, crossed molecular beam apparatus can contain as many as eight differentially pumped regions provided by various high-speed and ultrahigh-vacuum pumping equipment. It may be necessary to maintain a pressure differential from one atmosphere of pressure behind the nozzle of the molecular beam source to 10-i ' torr at the innermost ionization chamber of the detector. What is glibly ceded the "detector" is likely to be an extremely sensitive mass spectrometer with which to measure the velocity and angular distributions of products. By replacing one of the beams by a high-power laser, molecular beam systems are now giving new kinds of information on the dynamics and mechanism of primary photochem~cal processes. In the past 5 years, molecular beam experiments have played a crucial role in advancing our fundamental understandings of elementary chemical reactions at the microscopic level. These advances provide deeper insights with which to build our explanations of macroscopic chemical phenomena from the information gathered in microscopic experiments. The pervasive importance of these deeper insights was recognized in the award of the 1986 Nobel Prize in Chemistry to those responsible for bringing molecular beams into chemistry. SYNCHROTRON LIGHT SOURCES Characteristics of Synchrotron Sources The most intense, currently available source of tunable radiation in the extreme ultraviolet and X-ray region is synchrotron radiation, which is produced when energetic electrons are deflected in a magnetic field. That happens, of course, all the time in a synchrotron, which is an instrument that accelerates electrons to very high energies for particle physics studies. To reach these high energies, the electrons must be "recycled" through the accelerating zone many, many times. "Recycling" requires bending their trajectories through four successive 90-degree turns. At each

OCR for page 167
V-A. INSTRUMENTATION FOR STUDY OF CHEMICAL REACTIONS of these turns, the acceleration needed to change direction causes intensive radiation over the entire spectral range from the far-infrared to the X-rav range. This has, in the past, been looked on as an imtating energy loss. Now, however, synchrotrons are running out of things to do in high-energy physics. Hence, attention has turned from synchrotrons as accelerators (with radiation seen as undesired energy loss) toward snychrotrons as sources of light. Devices are placed inside the accelerator that increase the number of sharp bends in the electron trajectories to increase these radiative properties. These devices are descriptively called "wigglers" or "undulators." They show potential for intensity increases by several powers of 10 over the already bright radiation emitted by an ordinary synchrotron. Principal current use of tunable synchrotron radiation fails in the X-ray energy range, 1-100 keV. ~.~ ~ ^'D ' Applications of Synchrotron Sources in Chemistry Extended X-ray Absorption Fine Structure (EXAFS) has been one of the more fruitful applications of synchrotron radiation to solid substances. When one of an atom's inner-sheD electrons is excited by an X-ray photon, the atom emits light that is then diffracted by neighboring atoms. The result is a diffraction pattern that contains information about the interatom- ic spacings of these neighbors. ~ ~ Much attention has been di- ~ BW5 FREE ELECTRON . C 10 . "' i; LASERS . ~ r~ 1 o1 o 1 0 LO Con 15 10 ol 4 OCR for page 167
176 INSTRUMENTATION IN CHEMISTRY stimulated emission can occur to produce laser light. Such a device is called a Free-Electron Laser (FEL). Potential Capabilities Experience to date indicates that high-efficiency wavelength tunability and high average and peak power will all be forthcoming over a wavelength range extending from microwave frequencies through the infrared and visible to the vacuum ultraviolet spectral ranges. Average bnghtnesses several powers of 10 greater than those provided by conventional tunable lasers or synchrotron sources might be possible, particularly in the ultraviolet. An FEL has been operated at Los Alamos National Laboratory, based on a linear accelerator 2 or 3 meters long. Once a second, the device provides a train of pulses of tunable infrared radiation currently, in the 9- to 11-micron wavelength range, with 30-picosecond pulses, peak power of 5 megawatts, and 50-nanosecond spacing between pulses. Such perform- ance extended over the mid-infrared spectral region (4 to 50 microns) would open the way to many novel applications in chemistry. Examples are vibrational relaxation, multiphoton excitation, nonlinear processes in the infrared region, fast chemical kinetics, infrared study of adsorbed molecules, and light-catalyzed chemical reactions. As the wavelength is moved through the visible and toward the ultraviolet, a variety of novel chemical applications could be explored in photo- chemistry and fast chemical kinetics, as well as multiple photon and other nonlinear processes. SUPPLEMENTARY READING Chemical & Engineering News "Laser Vaporization of Graphite Gives Sta- ble 60-Carbon Molecules" by R.M. Baum (C.&E.N. stab, vol. 63, pp. 20-22, Dec. 23, 1985. "Imaging Method Provides Mass Transport' (C.&E.N. staff, vol. 63, p. 29, Sept. 23, 1985. '~Computers Gaining Fiery Hold in Chemical Labs" by P. Zurer (C.& E.N. staff), vol. 63, pp. 21-31, Aug. 19, 1985. "Supercomputers Helping Scientists Crack Massive Problems Faster" by R. Dagani, , . vol. 63, pp. 7-14, Aug. 12, 1985. "Spectroscopic Methods Useful in Inorganic Labs" (C.&E.N. staff), vol. 63, pp. 33-39, Jan. 14, 1985. "Technique Allows High Resolution Spec- troscopy of Molecular Ions" by R.M. Baum (C.&E.N. staff), vol. 62, pp. 34-35, Feb. 20, 1984. "Extreme Vacuum Ultraviolet Light Source Developed" by R.M. Baum (C.&E.N. stab, vol. 61, pp. 28-29, Feb. 7, 1983. "Synchrotron Radiation" by K. O. Hodgson and S. Doniach, vol. 56, pp. 26-27, Aug. 21, 1978.

OCR for page 167
The Ant That Doesn't Like Licorice While touring through the Costa Rican jungle recently I stumbled on a terribly wide path completely devoid of plant life. The path must have been 6 feet wide, and as I strolled along it I tried to keep out of the way of the native ants who were bustling past me. Each of the ones going the other way was carrying a big piece of leaf overhead; the whole bunch looked like a fleet of Chinese junks sailing along. Suddenly, I was oversalted by this very attractive native ant. `'Hi there!" I introduced myself, '`My name is Red Ant. What's your name?" Blushing, she answered,"My last name is Formicidae, but they call me L`eafeutter.'' "Say, that's a pretty name. Why do they call you that?" Giggling, she said, `'Everyone knows why it's because that's my business." She gracefully pointed an antenna at a pitifill-looking tree up the path. "See that?" she asked, "My sisters and I did that. We cut every one of the leaves off that tree in only 5 ~~_ days. Enough to feed the whole family for 2 months.'' ~ She turned to leave. "Don't go!" I exclaimed, "I'll Go' ~ , _ get you a leaf from this tree right here." Reaching ~ ~~' toward a lush tree that all the other ants were passing by, pulled oiT a leaf and presented it to Leafcutter. "Pew!" she ~~''~'~' Am' said, holding her nose, "Take it away~ hate licorice." Sure enough, the leaf ~ held smelled just like liconce. I wondered what was wrong with licorice Leafcutter explained `'I'm not sure why, but Mama doesn't like it when we bung leaves that smell like that into the anthill." I was still poled so I asked her if she'd show me her home. Leafeutter lived in this gorgeous anthill along with her 5 Anion sisters, 500 brothers, her Mama, and, believe it or not, a Angus! Her lazy brothers never lifted a feeler to bring in even one leafall they seemed to do was amuse ~ Mama. And guess what? The ants didn't even eat all ~7: ~~' those leaves they brought home at Soothe fungus did! Apparently, the ants don't have the right enzymes to metabolize carbohydrates. But the fungus thrives off those leaves, and in gratitude to the ants for supplying them, it coIlveIts their carbohydrates into delicious sugars that the ant family lives on. '`Mama says we're symbiotic," Lealcutter exploded. Scientists have also taken an interest in Lealcutter and her family. They have concentrated on the leaves that Leaicutter doesn't like, trying to find out what protects these leaves over the others. Using liquid chromatography, they've extracted 10 to 15 milligrams of about 50 different compounds from great piles of the rejected leaves. Then they've worked to purify and identify these compounds. NMR studies have shown that every one of those trees that L~eaicutter dislikes contain compounds with molecular structures like that of carophyllene oxide, the compound that gives Sconce its flavor. They've adso got this notion that it's the fungus that gets sick on those leaves. And when the fungus gets sick there's no sugar for the ant family. So it looks as though the liconce-flavored trees have learned to synthesize their own fungicide to protect themselves from the Leatcutters The next step for those scientists will be to try to synthesize some similar compounds to combat harmful fungi elsewhere. The next step for me is Into a cozy little anthill with Leaicutter we're getting hitched in the spring. 177

OCR for page 167
192 INSTRUMENTATION IN CHEMISTRY first few layers. Two or more complementary methods used together can greatly enhance the significance of any single measurement used alone. TABLE V-C-1 Instrumentation Relevant to Chemistry on Surfaces . . Method Electron energy loss EELS spectroscopy Bombard or Irradiate Acronym with: Electrons, 1-10 eV Physical Basis Information Obtained Molecular structure, surface bonding of adsorbed molecules Molecular structure of adsorbed molecules Energy of surface binding Vibrational excitation of surface molecules Infrared spectroscopy IRS Thermal desorption Auger spectroscopy Low-energy electron diffraction Secondary ion mass spectroscopy TDS Auger LEED SIMS Heat Infrared light Vibrational excitation of surface molecules Thermally induced desorpiion of adsorbates Electron emission from surface atoms Back-scattenng, diffraction Ejection of surface Electrons, 2-3 keV Electrons, 10-300 eV Ions, 1-20 keV atoms as ions Surface composition Atomic surface structure Surface composition Electrons are useful surface probes because their energies, and hence, their wavelengths, can be accurately controlled with their accelerating voltage. At low energy, near 25 electron volts, the wavelength of an electron is close to the atomic spacings in a metal, so a beam of such electrons reflected from the surface will show diffraction ejects. Thus, low-energy electron diffraction (LEED) can play the same role in determining bond distances and bond angles in surface chemistry as X-ray diffraction plays in the structural chemistry of solids. LEED reveals the atomic structure of clean surfaces, as well as any regulanty in the packing of atoms and molecules adsorbed on the surface. In the Auger (pronounced "Ohjay'') effect, high-energy electrons (2,000-3,000 eV) striking an atom cause the atom to eject a secondary electron from an inner shell. The energy of the ejected electron is determined by the energy levels of the atom it came from, so measurement of the electron energy identifies the atom. Since the bombarding electrons do not penetrate deeply, these secondary electrons reveal, with high sensitivity, the composition of the first few surface layers. This information can be important because surface impurities and i'Te~ulanties can . , , ~ ,~ ~ _ _ ~ _ dominate surface chemistry. Hence, the combination of Auger and LEED is used routinely to verify the cleanliness and perfection of the surface under study. Electron energy loss spectroscopy (EELS) is of particular value because it detects the resonant vibrational frequencies of atoms and molecules bound to the surface. Chemists routinely use such vibrational frequencies for gaseous molecules to decide which atoms are hooked to which, how strong the bonds are, and their molecular geometry (see Infrared Spectroscopy later in this section). In EELS, an electron beam of known energy is bounced off the metallic surface into an energy analyzer. If the electrons hit an area where a molecule is adsorbed, the molecule can be left vibrating in one of its characteristic motions. The energy needed to do this, determined by the frequency of the motion, is taken away from the kinetic

OCR for page 167
V<. INSTRUMENTATION AND THE NATIONAL WELL-BEING energy of the electron. The measurement of these electron energy losses of the reflected beam gives a vibrational spec- trum of the adsorbed mole- cules. Ion scattering from surfaces has been used for surface com- positior~ analysis with great sensitivity, 109 atoms/cm2. In secondary ion mass spectros- copy (SIMS), neutral and ion- ized atoms and molecular frag- ments are ejected by bombard- ment with high-energy (1-20 keV) inert gas ions. Ton scat- tenng spectroscopy determines the surface composition by the energy change of inert gas ions upon surface scattering. Ton etching removes atoms from surfaces layer by layer. The combined use of ion etching and electron spectroscopy yields a depth profile analysis of the chemical composition in the near-surface region. This combination of instrumental methods is called "dynamic SIMS." The availability of high-intensity laser sources is now awakening the develop- ment of a new set of surface-sensitive techniques. Surface infrared spectroscopy, laser Raman spectroscopy, and second harmonic generation surface spectroscopy all provide information about the surface chemical bonds of adsorbed atoms and molecules. All of these emerging surface science techniques will permit us to watch chemical reactions as they occur on well-charactenzed and clean surfaces. This is an important development in chemistry because surfaces provide the two-dimen- sional reaction domain that accounts for heterogeneous catalysis. reflected electrons E2-E1-1200 \ incident Of electrons p' / ;~ ~,~ ASH \ He C / \ ~ ~ ~ ~ M E, / EELS Vibrations of Surface Molecules The Energy Loss }dentifies the Vibrational Mode SURFACE ANALYSIS Any sensitive measurement technique can be used as an analytical tool. This is the case in the surface sciences. Every one of the capabilities listed in Table V-C-1 can be put to analytical use in the pursuit of questions that may be only remotely connected to the surface sciences. As an example, a state-of-the-art laser micro- probe device designed to desorb (remove) molecules from a solid surface can be used to detect the presence of a pesticide on the leaf of a plant. Such a capability was quite impossible only 10 years ago; today it permits us to contemplate tracking the amount, stability, weathering, and chemistry of a pesticide in field use. Of course, the analytical technique may just as well be concerned with monitonug or clarifying chemical changes that take place on a surface or with a surface. Many of these analytical studies relate to catalysis. In Section IV-C, examples were given of the use of EELS to determine the molecular structures that exist on a catalyst surface as it functions. Such applications have given rise to surface analysis, a new subdivision of analytical chemistry. 193

OCR for page 167
194 INSTRUMENTATION IN CHEMISTRY The elective sampling depth is a most important feature of any surface analytical technique. Sampling depth is important, because the measuring technique must be appropriate to the phenomenon under study. For example, bonding to the surface, wettability, and catalysis in- OOO LATERS valve only a few atomic layers, whereas surface hardening treatments involve 10 to 1,000 atomic layers. Typical sam- pling depths for the primary surface analytical techniques are one or two atomic layers for low-energy ion scattering, 5 A depth for SIMS, 20 A for the Auger technique, and 100 ~ for ion etching coupled with SIMS. Laser mass spectrome- try, the Raman microprobe, and scanning electron micros- copy (SEM) reach from 1,000 to 10,000 A (i.e.' to one micron). l he shallower the sampling depth of the tech- nique, the more finely it is able to define the surface composi- tion of a sample. A major challenge in the de- velopment of surface analytical instrumentation is the reinforcement of its quantitative dimension. Most of the examples given have been concerned with what is there. We must also be able to determine how much. Another important problem is the development of micro- probes which can provide both chemical and positional information about surface species. Currently, Auger and ion microorobes are useful in this respect for _ it, ,,, ,,,, ~ ,,, ~ ~ CATA LTSI S ~ //// WETTAB I L I TT _~ URFACE BOND I NG cn ::~ E EECTR I CA ~ CONDUCT I V I TTl l ~/~< //////././//, . <~//,//~/.~,///////////////////////~/////~ PASSIVATION, SURFACE TREATMENTS I ~7~s~//////~ _OPT I CA L AB SORPT ~ ON //D ~,/,~,~i'~'~6,Y~/,~,~Y,~,~,.,^~/,~,,Y/~/i~//~//////~/y~y~///~/~ CORRO S I ON ~_~ ~~ ~~__~ - Low ILM STRUCTURES ~ ~'~///~////~///~/~///////~/////////////////////////////~/////~///////////////~/~ VISUAL EFFECTS, COLOR ISS SIMS ESCA SIMS LASER SEM (STATIC ) AUGER (DYNAMIC ) MS RAMAN HOW DEEP IS THE SURFACE? IT DEPENDS ON WHAT YOU CARE ABOUT At. . .. .. mapping elemental composition, as in revealing both the presence and location of the trace contaminants phosphorus and lead in silicon chips. However, they are not yet able to detect and map large organic molecules such as carcinogens or therapeutic drugs. Characterization of small particles is another important chal- lenge for surface analysis; this is particularly important in environmental monitor- ing where the analysis of carcinogenic hydrocarbons on atmospheric dust and other particulates is a current problem. CHROMATOGRAPHY Chromatography separates molecules or ions by dividing species between a moving phase and a stationary phase. A liquid or a gas flowing continuously through a tube (called a "column") provides the moving phase. The stationary phase can be either small solid particles packed in the tube or, for a small-diameter

OCR for page 167
V-C. INSTRUMENTATION AND THE NATIONAL WELL-BEING tube (a capilIary), the walls of the tube itself. If a pulse or a squirt of soluble substance enters the tube at one end, that substance will have some tendency to stick to the stationary surface, becoming adsorbed. However, the continuing flow of fresh solvent keeps acting to redissolve this adsorbed matenal, moving it forward in the tube. How fast the process of adsorbing and desorbing takes place depends sensitively on the composition and structure of the substance. Conse- quently, different substances that entered the tube together in the same pulse will move at different speeds through the tube, so they will exit at different times. This separation technique takes advantage of small differences in properties such as solubility, absorbability, volatility, stereochemistry, and ion exchange, so that understanding the fundamental chemistry of these interactions is basic to progress in the field. Liquid chromatography has shown an impressive growth since 1970. The current $400 million annual sales are mainly by U.S. manufacturers. This growth has come through innovations such as high pressure and moving phases of changing composition ("gradient moving phases") to give greater speed and resolution. "Bonded-molecule" stationary phases are chemically designed to increase selectivity and to extend the useful lifetime of a column. Detection also has improved with electrochemical, fluorometric, and mass spectrometric detectors, reaching sensitivities as low as 1o-~2 grams. Although gas chromatography is a more mature field by perhaps a decade, important advances continue to appear. High-speed separations can now be accomplished in a few tenths of a second; portable instruments the size of a matchbox are in use outside of the laboratory. A complex mixture can be separated into literally thousands of components, using fused-silica capillary columns that are a direct spin-off from optical fiber technology for communications. It is even possible to separate compounds that differ only in isotopic composition. , . _ High-Performance Liquid Chromatography (HPLC) Dunng the 1970s, theoretical understandings of the complex flow and mass transfer phenomena involved in chromatographic separation helped perfect column design. During this same period, small-diameter (3-10-micron) silica particles with controlled porosity were introduced. Synthetic advances in silica chemistry led to the tailoring of particle diameter, pore diameter, and pore size distribution. Today, 15-cm columns with efficiencies exceeding 10,000 distillation steps ("theoretical plates") are routine. Still another major advance of the 1970s was the introduction of chemically bonded phases in which surfaces of porous silica are covalently coated with organic molecules containing silicon (organosilanes). Especially important is the use of hydrocarbon attachments (such as n-octy} and n-octadecyI) to make the surface look like an organic solvent. Then, the mobile liquid phase is typically an organic-aqueous mixture. This is called reversed phase chromatography (RPEC), and it currently provides well over 50 percent of all HPEC separations. It is especially well suited to substances that are at least partially soluble in water (drugs, biochemicals, aromatics, etc.~. Finally, the microprocessor/computer is playing an increasing role. "Smart" HPEC instruments are under development to program the performance. New 195

OCR for page 167
196 INSTRUMENTATION lN CHEMISTRY detectors of greater sensitivity and selectivity are on the horizon. In particular, laser spectroscopy promises to yield highly sensitive devices for subpicogram detection (less than 1o-~2 grams). Because of performance improvements, HPEC is having a major impact on diverse fields of biochemistry, biomedicine, pharmaceutical development, environ- mental monitonng, and forensic science. Today, peptide analysis and isolation requires HPEC because of its separating power and speed. Analysis of amino acids in protein/peptide sequencing is conventionally accomplished by RPEC. In clinical analysis, therapeutic drug monitoring can be accomplished by HPEC. The analysis of catecholamines (important as "neurotransmitters") is typically accomplished by RPEC with electrochemical detection. Isoenzyme analysis, which is important, for example, in assessment of heart damage after an attack, can be rapidly accom- plished by HPEC. The analysis of polar and high-molecular-weight organic species in waste streams in sewage or factory treatment can be performed by HPEC, while the separation and analysis of phenols in water supplies by RPEC is recommended. Analysis of narcotics, inks, paints, and blood represent only a few of the forensic applications in police laboratories. Capillary Chromatography This version of chromatography uses an open capillary tube with a thin liquid layer on its inner wall. It began with capillary gas chromatography (GC), but the fragility of glass as an inert matenal for GC capillary columns discouraged many potential users. Now we have flexible, fused-silica capilIanes with a polymer overcoat, a spin-off of fiber optics technology. These advances in capillary column technology led to intensive commercialization dunng the 1970s. Today's capillary columns exhibit efficiencies between ]05 and 106 distillation steps ("theoretical plates") and are capable of separating literally hundreds of components within a narrow boiling point range. Direct introduction of samples at the nanogram (IO-9 grams) levels has been developed, and much effort has been directed at perfection of gas-phase ionization detectors. Combined advances in the column and detector areas now make trace analytical determinations below 10-~2-gram levels practical by capillary gas chromatography. Of particular note is the combination of capillary GC with powerful identification methods such as mass spectrometry and Fourier Transform infrared spectroscopy, as mentioned in Section IV-C. The combined techniques are now routinely capable of identifying numerous compounds of interest that are present in complex mixtures in only nanogram quantities. They have been used in identification of new biologically important molecules, as well as in drug metabolism studies, forensic applications, and identifications of trace environmental pollutants. Every fluid has a characteristic temperature and pressure above which its gas and liquid phases become indistinguishable. Above these critical conditions, the "supercritical" fluid displays exceptionally low viscosity, and it can become a much better solvent. Consequently, the use of supercntical fluids in capillary chromatography has recently emerged as a promising approach to the analysis of complex nonvolatile mixtures. As the solute diffusion coefficients and viscosities of supercritical fluids are more favorable than those of normal liquids, chromato-

OCR for page 167
V-C. INSTRUMENTATION AND THE NATIONAL WELL-BEING graphic performance is substantially enhanced. Furthermore, the optical transpar- ency of supercritical fluids makes them attractive for certain optical detection techniques. Field-Flow Fractionation (FFF) Chromatography becomes more difficult to apply as molecular size grows, and it becomes ineffective in separating macromolecules and colloidal particles in the size range 0.01 to 1 micron in diameter. A recent innovation, field-flow fractionation, may fill this need. In FFF, a liquid sample flows through a thin (0.~-0.3 mm), nbbon-like flow channel. A temperature difference or electric field is maintained across the ribbon. Each constituent in the sample distributes itself in a way that is determined by its diffusional properties and its response to the applied thermal or electncal field. Since flow through the channel is fastest near the middle of the ribbon, substances that are pulled close to the ribbon wall move more slowly than substances that reside near the middle of the flow channel. Separations are thus achieved. A useful aspect of this technique is that the strength of the applied field can be varied in a deliberate and programmed way by a computer during the course of the separation. Such thermal gradients are effective in separating most synthetic polymers. The mass range of molecules and particles to which FFF has been applied extends from molecular weights of 1,000 up to 10~8, that is, up to particle sizes of about 100-micron diameters. FFF appears to be applicable to nearly any complex molecular or particulate material within that vast range. Applications have so far included macromolecules and particles of biological and biomedical relevance (proteins, viruses, subcellular particles, liposomes, artificial blood, and whole celIs), of industrial importance (both nonpolar and water-soluble polymers, coal liquid residues, emulsions, and colloidal silica), and of environ- mental significance (waterborne colloids and the tiny particles called fly ash in smoke plumes). INFRARED SPECTROSCOPY A molecule can be pictorially, but accurately, viewed as a collection of wooden balls held together in a fixed geometry by springs. The masses of the balls are proportional to the atomic masses, and the strengths of the springs are proportional to the strengths of the chemical bonds. Such a "ball-and-spring" mode! will have resonant vibrational frequencies in which the wooden balls move back and forth ire regular patterns. These frequencies are determined by the masses, the spnug constants, and the geometry. A molecule is exactly the same. If measured, the resonant fre- quencies give direct informa- tion about the molecular archi- tecture. Consider, for example, the water molecule. This bent, Resonant Vibrational Motions of H2O it'd \~ /~/ All or ~ '0 'A %~` it\\ 3587 cm~ 1 35(30 cm~ 1 1600 cm~ Vibrational Frequencies Reveal Bond Strengths and Bond Angles 197

OCR for page 167
198 At A: o a: INSTRUMENTATION IN CHEMISTRY tnatomic molecule has three resonant vibrations. In one of these, the two bonds stretch back and forth in phase, and in another, the two bonds both stretch, but this time out of phase. In the third characteristic vibration, the bond angle alternately opens and closes. Such molecular vibrations do not break bonds, so they require little energy. Absorption of light is one way to excite these vibrations, but photons of appropriate energy are in the infrared spectral region, far beyond the visual sensitivity of the human eye. A typical molecular vibration, such as the bending motion of the water molecule, has a frequency of 4.S x 10~3 vibrations per second. This unwieldy number is usually brought into reasonable magnitude by dividing it by the speed of light, which changes the dimensions to I/cm or cm- ("reciprocal centimeters". 4.8 X 10~3 vibrations/sec _ = 3 x 10~ cm/see 1,600 cm-~. Infrared vibrational frequencies are always expressed in reciprocal centimeters (cm-~) (sometimes called "wave numbers". The measurement of these molec- ular frequencies is called vibrational spectroscopy or infrared spectroscopy. These vibrational frequencies are so characteristic that they furnish a distinctive and easily measured "fingerprint" for each molecule. This spectral fingerprint, once measured for a particular molecule, can be used to determine whether that molecule is present in a sample and, if so, how much. The vibrational frequencies also reveal the molecular structure and bond strengths in the molecule, so they can be used to learn about the molecular architecture. When an unknown compound is under study, the infrared spectrum provides one of the easiest ways to decide what the compound is likely to be. Because infrared spectroscopy is so informa- tive, it has become one of the routine diagnostic tools of chemistry. A large, research-oriented chemistry department might operate 5 to 10 infra- red spectrometers with capabilities ranging from rugged, low-resolution instrument for instruction in an advanced first-year chemistry course, to high-resolution Fourier Transform Infrared Spec- trometers (FTIR), suited to molecular structure determination and specialized research use. DIFLUOROPROPENE 1R SOLID KRYPTON, ~ 2 Mel cis and gauche em_ JE - ~ PBOlOLTSIS (hi me_ ~1 ~ gaucaet l ~_J ~ By- --or- ~ cis1 . I I I' I t 1400 1200 1000 V(cm l) _~ "E. (hi)BE=RE (hi) ~""""'B~"~\~"'~"'~ FTIR DIFFERENCE SPECTROSCOPY SHOWS ROTAMER INTERCONVERSION Computer-Aided Spectrometers Modern research infrared spectrometers incor- porate computers to permit programmed opera- tion, data collection, and data manipulation. The major impact of computers, however, has been their influence on the performance of Fourier Transform interferometers. The perfection of the

OCR for page 167
V<. INSTRUMENTATION AND THE NATIONAL WELL-BEING Fourier Transform algorithm (program), plus the reduction in accompanying computer costs, brought the interferometer from a trouble-plagued, research- only instrument to a routine, high-performance workhorse. A notable capability brought by the computer is the ease and accuracy with which one spectrum can be subtracted from another to emphasize small changes. This is called a difference spectrum. One important application relates to infrared spectra of biological samples in which evidence of a chemical change associated with a certain specific biological function can be completely covered up by the heavy infrared spectrum of the inactive substrate in which the sample is located. The digitized data permit precise spectral subtraction so that the back- ground spectrum can be virtually eliminated to reveal the spectral changes of interest. Another vivid display of the value of the difference capability is provided by photolysis of molecules suspended in a cryogenic solid ('`matrix isolations. If the digitized spectrum before photolysis is subtracted from the spectrum after photolysis, only the features that change are seen. Any molecule that is being consumed presents its spectral features downward, while spectral features of the growing product extend upward. This has been used, for example, to distinguish the two forms (cis- and gauche-) of the 2,3,-difluoropropene in the cluttered spectrum of a complex mixture. A laser tuned to an absorption frequency of one of the "rotamers," say, the cis- form, is used to irradiate the cold sample. Absorption of this light adds enough energy to the absorbing cis- molecule to permit it to convert to the gauche- form. Then in the difference spectrum, the cis- molecule spectrum appears as a negative spectrum and the gauche- molecule spectrum as a positive one. Absorptions due to other molecules do not change, so they simply do not appear at all. Applications The coupling of FTIR with gas chromatographic separations in a variety of analytical uses has been dis- cussed. Also, as noted earlier, infrared spectroscopy is a spe- cially effective method for monitoring and studying at- mospheric chemistry. This is because gaseous molecules of Laser magnetic low molecular weight are im- portar~t, including formaide- hyde, nitric acid, sulfur diox- ide, acetaldehyde, ozone, ox- ides of chlorine and nitrogen, nitrous oxide, carbon dioxide, and the Freons. These sub- stances are influential partici- pants in photochemical smog production, acid rain, strato- TABLE V-C-2 Additional Instrumental Techniques in Modern Chemistry Instrument Information Obtained Reaction rates of gaseous molecular ions Precise molecular structures, gaseous free radicals Vibrational spectrum Lifetimes9 electronically excited molecules Stereo confol~llations Laser-activated cell sorter Automated analysis of protein sequence Automated synthesis of designed DNA segments Molecular structure, gases Tracking radiotracers Ion cyclotron resonance resonance Laser Raman Fluorimeter Circular dichroism Flow cytometer Protein sequencer Oligonucleotide synthesizer Electron diffraction Scintillation counter 199

OCR for page 167
200 INSTRUMENTATION IN CHEMISTRY spheric disturbance of the ozone layer, and the "greenhouse effect.'' Infrared spectroscopy shows where they are and how much is there. OTHER INSTRUMENTATION In Sections V-A, V-B, and V-C, there has been detailed discussion of over a dozen different classes of instrumentation which are important in defining and advancing the current frontiers of chemistry. By no means, however, is the list all-inclusive. Table V-C-2 lists additional types of equipment and what kinds of chemical information each one provides. The length of this table is only one more signal of the crucial importance of instrumentation in modern chemistry. SUPPLEMENTARY READING Chemical & Engineering News "Instrumentation '86~0ptical Spectroscopy" (C.& E.N. staff), vol. 64, pp. 3442, Mar. 24, 1986. "Instrumentation 'chromatography" (C. & E.N. staff), vol. 64, pp. 52-68, Mar. 24, 1986. "Instrumentation '8~Mass Spectrometry" (C.& E.N. staff), vol. 64, pp. 70-72, Mar. 24, 1986. "Low Cost FTIR Microscopy Units Gain Wider Use in Microanalysis" (C.& E.N. stab, vol. 63, pp. 15-16, Dec. 9, 1985. "pity Chromatography" by Parikh and P. Cuatrecasas, vol. 63, pp. 17-31, Aug. 26, 1985. "GC Detector Uses Gold Catalyst for Oxi- dation Reactions" by W. Worthy (C.& Em. staff), vol. 63, pp. 42-44, June 24, 1985. '~X-Ray Technique May Provide New Way to Study Surfaces, Films" by W. Worthy (C.& E.N. stab, vol. 63, pp. 28-30, April 8, 1985. "Centrifugal Force Speeds Up Countercur- rent Chromatography" by S.C. Stinson (C.& E.N. staid, vol. 62, pp. 35-37, Nov. 26, 1984. Science "A New Dimension in Gas Chromatogra- phy" by T.H. Maugh II (Science stair), vol. 227, pp. 1570-1571, Mar. 29, 1985. "Ion Beams for Compositional Analysis" (SIMS), by A.L. Robinson (Science stab), vol. 227, pp. 1571-1572, Mar. 29, 1985.

OCR for page 167
CHAPTER VI The Risk/Benefit Equation in Chemistry

OCR for page 167
-I ~- ~o',~ ~ f ~ ~ ~<~ ~~ OF <~lnvestigat~ng Smog Soup ~c ~ . I,=, ~0 ~ o'N`o ~ ~ Air pollution Is a visible reminder of the price we sometimes pay for progress. Emissions from thousands of sources pour into the atmosphere a myriad of molecules that react and re-react to forth a "smog soup." We are already aware of some of the potential dangers of leaving these processes unstudied and unchecked: respiratory ailments, acid rain, and the greenhouse eject. Sur~ns~ngly, you and ~ are the principal cuipnts in generating much of this unpleasant brew~veryt~me we start our cars or switch on our air conditioning or central heating! Transportation, heating, cooling, and lighting account for about two-thirds of U.S. energy use, almost all derived from combustion of petroleum and coal. Pinpointing cause and effect relationships begins, inevitably, with the identification and measurement of what is up there, tiny molecules at parts-per-billion concentrations in the mixing bowl of the sky. Finding out what substances are there, how they are reacting, where they came from, and what can be done about them are all matters of chemistry. The first two questions require accurate analysis of trace pollutants. Physical and analytic chemists have successfully applied to such detective work their most sensitive techniques. An example is the Fourier Transform Infrared Spectrome- ter. This sophisticated device can look through a mile or so of city air and identify all the chemical substances present and tell us their concentrations down to the parts-per- billion level. Recognizing a substance at such a low concentration is comparable to asking a machine to recognize you in a crowd at a rock concert attended by the entire U.S. population. ~ . . . How does this superb device work? "Infrared" means light just beyond the red end of the rainbow visible to the human eye. Hence infrared light is invisible, though we can tell it is there by the warmth felt under an infrared lamp. But molecules can '~see" inhered light. Every polyatomic molecule absorbs infrared "colors" that are uniquely characteristic of its molecular structure. Thus each molecular substance has an infrared absorption "fingerpr~nt"~i~erent from any other substance. By examining these finge~pnnts, chemists cart identify the molecules that are present. Boa An example of what can be done is the measurement of formaldehyde and nitric acid as trace constituents in Los Angeles smog. Unequivocal detection, ~ using almost a m~le-long path through the polluted air, revealed the growth _ ~ \ dunug the day of these two bad actors and tied their production to photochemical processes initiated by sunlight. Continuing experiments led ~'J \ to detailed characterization of the simultaneous and interacting concentra- @_: ~ tions of ozone, peroxyacetyl nitrate (PAN), formic acid, formaldehyde, and nitric acid ir1 the atmosphere. These detections removed an obstacle to the complete understanding of how unburned gasoline and oxides of iitrogeI1 leaving our exhaust pipe end up as eye and lung irritants in the atmosphere. This advance doesn't eliminate smog =~ soup, but it is a bin steD toward that desirable end ~ n' ~ ~ _ ~ i1~ =~w 202