THE SPECTROGRAPH Volume 23, Number 2 Spring 2007

Home , Graham Fleming, Mildred Dresselhaus

George R. Harrison Spectroscopy Laboratory Massachusetts Institute of Technology

Good Vibrations! Graham Fleming to Give Lord Professor Moungi Bawendi, an MIT MIT Institute Professor and Spectroscopy Lecture April 24 chemistry professor and Spectroscopy Laboratory core researcher Mildred Dres- Laboratory core researcher, has been se- selhaus is the North American winner of a lected as one of eight winners of the 2007 2007 L’Oréal-UNESCO Award for Women Ernest Orlando Lawrence Award of the in Science. Department of Energy, Secretary of En- She and four other recipients, each repre- ergy Samuel W. Bodman has announced. senting a different continent, were named at The award honors scientists and engineers a February 22 ceremony at UNESCO House at mid-career for exceptional contributions Dresselhaus, continues on page 3 in research and development that support its mission to advance the national, economic and The Executive Committee of the MIT energy security of the United States. Bawen- Corporation has announced that William di is cited for “chemical synthesis and char- H. Green, a professor in MIT’s Chemical Bawendi, continues on page 2 Engineering Dept and a core Spectros- copy Laboratory researcher, has been promoted to full professor. Green is known for his work in formulating accurate chemical kinetic models using state-of- the art quantum chemistry techniques to quantitatively predict the time evolution Graham Fleming of chemical mixtures, and for methods for Professor Graham R. Fleming of UC Berke- constructing and solving the associated ley will deliver the 2007 Richard C. Lord large reaction schemes. These models re- lecture on April 24, on “Two dimensional late to combustion, pyrolysis and other ultrafast electronic spectroscopy”. complicated, technologically important, Fleming was born in Barrow, England in reactive chemical processes. The results 1949. He received his Bachelor of Science are used to test and predict experimental degree in 1971 from the University of Bris- data he obtains using laser techniques to tol and his Ph.D. in chemistry in 1974 from probe free radicals in the gas phase. Green London University. He conducted postdoc- is well known as an excellent teacher, toral research at the California Institute of and has made signifi cant contributions to Technology, the University of Melbourne, Professor Moungi Bawendi the revitalizing the Chemical Engineer- and the Royal Institute, UK. In 1979 Dr. Also in this issue ing curriculum, including a new course Fleming joined the faculty at the Univer- “Introduction to Chemical Engineering”. sity of Chicago where he became the Arthur Research Report: ~ He was a member of MIT’s Energy Re- Holly Compton Distinguished Service Pro- Chirality Distribution of Carbon search Council, and plays a leadership fessor in 1987. In 1997 he accepted a posi- Nanotube Materials role in several industry-sponsored energy tion at the University of California-Berkeley ~ Research Report: research projects at MIT. where he is currently the Melvin Calvin Dis- A New Hybrid Multivariate tinguished Professor of Chemistry. Calibration Method Page 1 Fleming is one of the world’s foremost authorities on ultrafast processes and con- ~ Spring Seminar: tinues to make revolutionary discover- Modern Optics and Spectroscopy ies in biophysics. Currently, his group is ~ Workshop studying dynamical processes in a range of Lester Wolfe Workshop : Optical complex systems, such as liquids, solutions Methods in Breast Cancer and proteins, with the goal of understanding the primary steps of photosynthesis. ~ Spectral Lines: This may lead to a breakthrough in artifi cial Pop Spectrum Professor William H. Green, Jr. Lord, continues on page 3 Page 1 Bawendi, continued from page 1 Personality versity, both national universities. Ram- acterization of functional semiconducting Ramachandra Dasari achandra’s graduate research was on the nanocrystals, also known as quantum dots.” electronic spectroscopy of simple mole- Bawendi, a materials chemist, developed cules. After graduation, in 1962, he joined a synthesis of semiconductor nanocrystals the Physics Department of the Indian In- that was the fi rst to enable precise control of stitute of Technology, Kanpur (IIT), a new their size and precise determination of their university established in collaboration properties. Using the Bawendi synthesis, nanocrystals are now routinely made to or- with nine US universities, led by MIT. He der. One of the world’s leaders in nanotech- was one of the fi rst new faculty and one of nology, his research focuses on the synthesis, the few faculty members with an all Indian electronic properties and optical properties education, more than 90% of the new fac- of semiconductor nanocrystals (quantum ulty coming from the nine US universities. dots) for applications as diverse as biol- Ramachandra’s connection to MIT ogy, optoelectronics and nanoelectronics. dates back to 1966, when he came for two Moungi G. Bawendi was born in 1961 years as a fellow under the US AID pro- in Paris, France. He received his A.B. in Ramachandra Dasari, Associate Direc- gram with an allowance of $8 per day. For Chemistry from Harvard University in 1982, tor of the Spectroscopy Laboratory, is the two years he worked in the research group his A.M. in Chemical Physics from Harvard glue that binds the Laboratory together. of Professor Ali Javan, who had just ar- in 1983, and his Ph.D. in Chemistry from Confi dant to Spectroscopy Laboratory rived at MIT from Bell Laboratories after The University of Chicago in 1988. He was graduate students and professors, expert having invented the helium-neon laser. a Postdoctoral Member of the Technical negotiator, project organizer and trouble- During this period Ramachandra gained Staff at AT&T Bell Laboratories from 1988- shooter, Ramachandra is always there valuable experience in fabricating lasers 1990. In 1990, he began teaching at MIT. when he is needed. And when it comes to and conducting research based on the new The Lawrence Award was established equipment purchases, Ramachandra al- laser spectroscopy. With this background, in 1959 in honor of Ernest O. Lawrence, a ways knows how to get those special dis- he returned to IIT and established one of UC Berkeley physicist and Nobel Laureate counts. Sometimes the equipment manu- the largest laser laboratories in India, and who invented the cyclotron and after whom two major DOE laboratories at UC Berke- facturers even pay him for the privilege of trained a large number of Ph.D. students ley and in Livermore, California are named. placing their equipment in laser research. Gradu- in Spectroscopy Labo- “...Ramachandra did well ate students were trained THE SPECTROGRAPH ratory laboratories—at and was sent on to high to fabricate then-novel Published by the George R. Harrison Spec- least it seems that way! school, although this re- and (then considered by troscopy Laboratory at the Massachusetts Ramachandra was some) esoteric lasers such Institute of Technology, Cambridge, MA born in a village of 50 quired walking a few miles as molecular nitrogen and 02139-4307. Comments, suggestions, and houses or less in the each day without slippers... argon ion, in addition to inquiries can be directed to the editor. in Krishna District of helium-neon. Few if any Editors: Geoff O’Donoghue & Mei-Hui Liu Andhra Pradesh, probably in 1933. His lasers could be purchased at that time, and GEORGE R. HARRISON exact birth date is not recorded. His par- there were no company reps. And besides, SPECTROSCOPY LABORATORY ents owned a small farm. Neither his fa- laser equipment, even if it did exist, would Director: Michael S. Feld ther nor his mother had any schooling, be beyond the tiny budget of a fl edgling Assoc. Director for Scientifi c Coordination: although his mother could read and write. Indian university. So, as Ramachandra ex- Robert W. Field His grandfather, a deeply religious man plains, building lasers became the way of Associate Director: who lived in a neighboring village, had life of his laboratory, and this provided ex- Ramachandra R. Dasari the strong belief that his family must start cellent training for the graduate students. The Spectroscopy Laboratory houses two laser to become educated. Overcoming enor- Ramachandra’s laboratory was also research resource facilities. The MIT Laser Re- mous diffi culties, he built an elementary noted for developing interactions with search Facility provides shared facilities for core school in his village and selected Ram- R&D laboratories throughout India. And researchers to carry out basic laser research in the achandra as one of the students. Ramach- using the home built lasers and classi- physical sciences. The MIT Laser Biomedical andra moved to his grandfather’s village cal sources, Ramachandra continued his Research Center, a National Institutes of Health where his grandfather raised him. Ram- studies in molecular spectroscopy, which Biomedical Research Technology Center, is a resource center for laser biomedical studies. The achandra did well and was sent on to high led to the fi rst observation of the elec- LBRC supports core and collaborative research school, although this required walking tronic spectra of the NSe molecule. He in technological research and development. In a few miles each day without slippers. was promoted to full professor at IIT in addition, it provides advanced laser instrumenta- Ramachandra’s early education was in 1973. As a member of the Physics Panel tion, along with technical and scientifi c support, Telugu, his mother tongue, and thus his of the University Grants Commission, he free of charge to university, industrial, and medical researchers for publishable research projects. English was weak until he entered college. introduced new initiatives for improv- Call or write for further information or to receive His undergraduate studies were at Andhra ing undergraduate education and orga- our mailings. University, 15 miles from his village. He nized workshops for training teachers. received his Masters degree in 1956 from Ramachandra left IIT in 1978, and spent (617) 253-4881 Benaras Hindu University and his Ph.D. a year as a visiting Senior Research Of- http://web.mit.edu/spectroscopy degree in 1960 from Aligarh Muslim Uni- fi cer at the National Research Council Page 2 of Canada, Ottawa, and another year as graduates, and a poster session from cur- Lord, continued from page 1 a visiting scientist at the Department of rent IIT students. Ramachandra spoke on photosynthesis that can provide effi cient Physics, University of British Columbia “Spectroscopy for diagnosis of disease.” and sustainable energy for mass consump- before coming to MIT in 1980 as a Vis- Michael Feld was the inaugural lecturer. tion. A second research goal is to develop iting Professor of Physics. He has been Ramachandra has been happily mar- complex pulse sequences in order to ma- a Principal Research Scientist in the ried for 56 years to Suhasini Dasari, who nipulate and modify molecular dynamics. Spectroscopy Laboratory since 1981. He works at the MIT Medical Department. Throughout his career, Dr. Fleming has was appointed Assistant Director of the Following typical Indian tradition they been an active and successful scientifi c re- Spectroscopy Laboratory in 1984 and married young, in 1951, when he was 18 searcher, having co-authored over 345 pub- promoted to Associate Director in 1992. and she was 15. Many believe that much lications. In addition to his cutting-edge Ramachandra has played a major role of Ramachandra’s wisdom actually ema- research, Dr. Fleming has proved to be a in the Spectroscopy Laboratory over nates from Suhasini. They have two chil- capable leader. He served as Chair of the the past 27 years. He oversees project dren, a son, Satish, who is a physician Chemistry Department for three years at the University of Chicago, during which coordination and facility development in Indiana, and a daughter, Lakshmi, an he oversaw the creation of the Institute for of the NIH-supported Laser Biomedi- M.S. in computer sciences who teaches Biophysical Dynamics. After moving to cal Research Center, and coordinates high school mathematics in Seattle, and UC Berkeley, he initiated and directed the research programs of the physical sci- two grandchildren, Siddharth and Vivek. Physical Biosciences Division at Lawrence ence-based MIT Laser Research Facility. Ramachandra will be semi-retiring at Berkeley National Laboratory and also Ramachandra’s research has covered the end of the summer. But he will con- became co-director of the California Insti- a wide range. His early studies include tinue to oversee the Spectroscopy Labo- tute for Quantitative Biomedical Research classical high-resolution spectroscopy of ratory’s fi scal and strategic activities. So (QB3). In 2005 Dr. Fleming was appointed simple molecules, atomic and molecular we won’t be saying adieu, only au revoir. as Deputy Director at Lawrence Berkeley collisions, vibrational-rotational relax- Dresselhaus, continued from page 1 National Laboratory. Dr. Fleming has re- ation, laser frequency measurements in ceived numerous awards for his contribu- the far infrared, and laser spectroscopy of tions, including his election as a fellow of rare-earth ions in single crystals. He then both the Royal Society of London and the progressed to study Dicke narrowing in American Academy of Arts and Sciences. infrared transitions, multiphoton ioniza- Dr. Fleming has recieved numerous tion, laser-nuclear studies, molecular col- awards for his contributions. He is a Fel- lisions and dynamics, cavity-QED, the low of the Royal Society and a Fellow of the single atom laser, and surface-enhanced American Academy of Arts and Sciences. Raman scattering. More recently, his re- He is recipient of Marlow, Tilden and Cen- search emphasis has switched to laser- tenary Medals from the Royal Society of Professor Mildred Dresselhaus received a 2007 biomedical studies, where he has pursued Chemistry, the Nobel Laureate Signature L’Oréal-UNESCO Award for Women in Science Award for Graduate Education Chemistry, spectral diagnosis of atherosclerosis and for her research on carbon nanotubes, images early-stage cancer in various organs of the the Peter Debye Award and the Harrison of which she points to on a monitor. Photo cour- Howe Award of the American Chemical body using light scattering, refl ectance, tesy / L’Oréal ©Micheline Pelletier/Gamma Society. He has also received Sloan, Drey- fl uorescence, and Raman spectroscopy. in Paris where Sir Lindsay Owen-Jones, fus, and Guggenheim Fellowships, as His most recent research is in the fi eld of chairman of L’Oréal, and Koïchiro Matsu- well as the Coblentz Award and the Inter- low coherence interferometry for detect- ura, director-general of UNESCO, present- American Photochemical Society Award. ing nanometer motions in red blood cells ed each laureate with her $100,000 award. Now in its 15th year, the Lord lecture com- and nerves. He has given numerous lec- Dresselhaus was selected for “conceptu- memorates the achievements of Richard C. tures at universities in the US, Canada and alizing the creation of carbon nanotubes,” Lord, a pioneer in infrared and biochemical India. His research publications, which according to L’Oréal and UNESCO (United spectroscopy and director of the Spectros- number well over 200, have appeared in Nations Educational, Scientifi c and Cul- copy Laboratory for 30 years. Each year’s most of the major physics and chemis- tural Organization). Due to their small lecturer is selected by a committee of chem- try journals. Under his guidance, twelve size, high strength and electrical conduc- ists, physicists and engineers at MIT who MIT students have received Ph.D. de- tivity, carbon nanotubes are ideal for new are active in various fi elds of spectroscopy. grees, and several others Masters degrees. materials used in objects such as light- On February 28, 2005, the Center for weight bicycles and fl at-panel screens. Dresselhaus was the fi rst tenured Laser Technology/Laser Technology Pro- A native of Bronx, N.Y., Dresselhaus has woman professor at MIT’s School of En- gram at IIT created a distinguished annual conducted scientifi c research for more than four decades. An MIT Professor of Physics gineering and one of the fi rst women symposium to honor Ramachandra. (Feb- and Electrical Engineering, she received her to receive a Fulbright Fellowship. She ruary 28 is National Science day in India, Ph.D. from the University of Chicago. She has received numerous awards, includ- as well as C.V. Raman’s birthday.) Each began her MIT career at Lincoln Labora- ing the U.S. National Medal of Science symposium is centered on a presentation tory studying superconductivity; she later and 19 honorary doctorates worldwide. from a distinguished scientist from India switched to magneto-optics, carrying out Now in its ninth year, the L’Oréal- or abroad. The inaugural symposium was a a series of experiments that led to a fun- UNESCO Award for Women in Science is day-long program of talks from professors damental understanding of the electronic the only one of its kind to honor eminent and senior scientists who were former IIT structure of semi-metals, especially graphite. women scientists at the international level. Page 3 Research Report bration methods is the requirement of com- this compacted and simplifi ed form of the plete knowledge of the model components. data, Eq.(3) may be inverted to arrive at b. A New Hybrid Multivariate In many cases, such as in biological spec- PLS is similar to PCR with the exception troscopy, this is not feasible. Therefore, that the matrix decomposition for PLS is Calibration Method implicit calibration methods may be used. performed on the covariance matrix of the Implicit calibration methods require only spectra and the reference concentrations, Kate L. Bechtel, Wei-Chuan Shih, Michael S. Feld a set of calibration spectra, Sc, associated while for PCR only spectra are used. PLS with several known concentrations of the and PCR have similar performance if G. R. Harrison Spectroscopy Laboratory, Mas- analyte of interest that are expressed as a noise in the spectral data and errors in the sachusetts Institute of Technology, Cambridge, column vector, c. The forward problem reference concentration measurements are Massachusetts 02139 for implicit calibration is similar to Eq.(1): negligible. Otherwise, PLS generally pro- c STb , (3) vides slightly better analysis than PCR [3]. Multivariate calibration is a powerful c Although these are powerful methods, chemometric technique for extracting an- with the objective to use the set alyte concentrations in complex chemical they are not without their limitations. As of calibration data [Sc,c] to obtain pointed out in the literature [4], spuri- systems that exhibit linear response. Mul- an accurate b by inverting Eq.(3). tivariate techniques are particularly well ous effects such as instrument drift and suited to analysis of spectral data because There are two primary diffi culties in di- co-variations among components can be information about all of the analytes can rectly inverting Eq.(3). First, the system incorrectly interpreted as arising from be collected simultaneously at many wave- is usually underdetermined, i.e., there are the analyte of interest. Thus, when the lengths. The goal of multivariate calibra- more variables (e.g., wavelengths) than calculated b is applied to a future spectrum in which one or all of those correlation is to obtain a spectrum of regression Experiment - Clear coeffi cients, b, such that an analyte’s con- tions are not present, an erroneous result Uncorrelated may be obtained. It may be possible to centration, c, can be accurately predicted 1 by taking the scalar product of b with a improve calibration and limit spurious prospective experimental spectrum, s: 0.5 correlations by incorporating additional

SEP (a.u.) SEP information about the system or analytes. T 0 c s b . (1) PLS HLA CR PLS HLA CR GCThe MIT Spectroscopy Laboratory has (Lowercase boldface type denotes a col- a history of developing hybrid methods, Correlated umn vector, uppercase boldface type a 1 which combine elements of both explicit matrix; and the superscript T denotes and implicit calibration by including prior transpose.) In the absence of noise and 0.5 information. The ﬁ rst such method, hybrid

other component correlations, the regres- (a.u.) SEP 0 linear analysis (HLA) was developed by sion vector, b, is unique and is equivalent PLS HLA CR PLS HLA CR G C Berger, et al. [5]. In HLA, the separately to the spectrum of the analyte of interest. Figure 1: SEP values normalized to PLS measured pure component spectrum of the Under realistic experimental conditions, results for glucose (G) and creatinine (C) in analyte of interest, p, is used together with however, b is more complicated and only clear sample experiments without (Uncorrelat- the reference concentrations, c, to subtract an approximation to the ideal b can be ed) and with (Correlated) analyte correlations the spectral component attributed to the found. The “accuracy” of b is judged by analyte of interest, cpT, from the set of cal- its ability to correctly predict concentra- equations (e.g., number of calibration ibration spectra, S. Principle components tions via the standard error of prediction samples). Thus, direct inversion does not are derived using the modifi ed spectra and (SEP) with independent data. generally yield a unique solution. Second, these principle components are then sub- even if a pseudo-inverse exists and results There are two categories of multivari- tracted from p to form b. In this manner, in a unique solution, such a solution tends ate techniques: explicit and implicit. Ex- only the portions of p that are orthogonal to be unstable because all measurements plicit techniques are those in which the to the other components in the system are contain noise and error. That is, small vari- pure component spectra are known or can retained. As a result, HLA is insensitive ations in c or S can lead to large variations be extracted from existing information. to spurious correlations in a well-defi ned in b. Therefore, data reduction methods The most common of these techniques system. However, because HLA relies on are usually applied to arrive at an equiva- is ordinary least squares (OLS), in which the subtraction of the analyte spectrum lent data set that can be easily inverted. a spectrum is fi t with a matrix of pure from the calibration data, it is highly sensi- component spectra, the fi t coeffi cients Principal component regression (PCR) tive to the accuracy of the analyte spectral being directly related to the concentra- [1] and partial least squares (PLS) [2] are shape and intensity. For complex turbid tions. Mathematically, this is equivalent two frequently used implicit calibration samples in which absorption and scatter- to deriving regression vectors for all com- methods. PCR decomposes the matrix of ing are likely to alter the analyte spectral ponents simultaneously, B, directly from calibration spectra into orthogonal prin- features in unknown ways, we fi nd that the pure component spectral matrix, P: ciple components which best capture the the performance of HLA is impaired. variance in the data. These new variables BT (PTP)1PT (2) Motivated by advancing transcutane- eliminate redundant information and by us- ous in vivo blood analysis, we have de- and then applying Eq. (1). ing a subset of these principle components, veloped a new calibration method, confi lter noise from the original data. With One obvious limitation of explicit cali- strained regularization (CR), recently Page 4 published in Analytical Chemistry [6], Experiment - Turbid spectrum used as the spectral constraint which is more robust against inaccura- only guides the inversion process, allowing cies in the previously measured pure ana- the minimization algorithm to arrive at the lyte spectra. In the following, we show 1 optimal solution, thereby reducing its de- that with CR the prediction error is lower 0.5 pendency on the accuracy of the spectrum. than methods without prior information, SEP (a.u.) These results demonstrate that there is such as PLS, and is less affected by ana- 0 PLS HLA CR PLS HLA CR a tradeoff between maximizing prior in- lyte co-variations. We further show that GCformation utilization and robustness con- CR is more robust than our previously Figure 2: SEP values normalized to cerning the accuracy of such information. developed hybrid method, HLA, when PLS results for glucose (G) and creati- Multivariate calibration methods range there are inaccuracies in the applied con- nine (C) in the turbid sample experiment from explicit methods with maximum use straint, as often occurs in complex or of prior information (e.g. OLS, least robust The results from the fi rst experiment, turbid samples such as biological tissue. when accurate model is not obtainable), in which analyte concentrations were Multivariate calibration can be con- hybrid methods with an infl exible con- uncorrelated, are summarized in Fig. 1 sidered as the minimization of a least- straint (e.g. HLA), hybrid methods with a (Uncorrelated). OLS results are not listed squares quadratic cost function, Ф: fl exible constraint (e.g. CR), and implicit because the three-constituent model does 2 methods with no prior information (e.g. ) STb c , (4) not account for all experimental varia- c PLS, most robust, but is prone to be mis- tions, e.g. low amounts of fl uorescence with ||x|| the Euclidean norm (i.e., mag- led by spurious correlations). We believe from the quartz cuvette; therefore, OLS nitude) of x [7]. CR is based on a tech- that CR achieves the optimal balance be- no longer provides the best achievable nique known as regularization [8] in which tween these ideals in practical situations. performance. It is observed that substan- instabilities associated with the inversion tial improvement over PLS is observed process are removed by damping the ef- using the hybrid methods. CR and HLA fect of smaller principle components with References generate similar SEP values, suggesting the aid of a regularization parameter, Λ: that these two methods have comparable 1. Gunst, R.F. & Mason R.L. Regression T 2 2 analysis and its application : a data- )(/) Sc b c / b . (5) performance under highly controlled experimental conditions with clear sam- oriented approach. M. Dekker (1980). Thus, b is limited to a fi nite value based ples and without analyte correlations. 2. Wold, S., Martin H. & Wold H. on a proper choice for the value of Λ. In Lecture Notes in Mathemat- CR, an additional piece of information is The second experiment, in which corre- ics. Springer-Verlag (1983). included in the cost function. Rather than lations between analytes were introduced, 3. Thomas, E.V. & Haaland. D.M. Analyt- simply limit the norm of b itself, the cost demonstrates that CR is less susceptible ical Chemistry 62, 1091-1099 (1990). function is minimized based on the differ- than PLS to spurious correlations among 4. Arnold, M.A., Burmeister J.J. & co-varying analytes. We modifi ed the cali- ence between b and a spectral constraint, b0: Small G.W. Analytical Chem- bration data set such that the concentration 2 istry 70, 1773-1781 (1998). ( , b ) STb c b b 2 . (6) 2 ) / 0 c / 0 of glucose correlated to creatinine with R 5. Berger, A.J., Koo T.W., Itz-

~ 0.5. The prospective set remained un- kan I. & Feld, M.S. Analytical The insight for CR is that a well-chosen correlated. The results are displayed in Chemistry 70, 623-627 (1998). Fig. 1 (Correlated). Again, CR and HLA b0, namely, the spectrum of the analyte of 6. Shih, W.-C., Bechtel K.L. & interest or the elements of which that are are observed to have similar performance. Feld M.S. Analytical Chem- orthogonal to other system components, In principle, HLA should be less affected istry 79, 234-239 (2007). can constrain b to a more accurate solution. by analyte correlations than CR, however, 7. Bertero, M. & Boccacci P. Introduc- In other words, this constraint can help the possible explanations include imperfect ex- tion to inverse problems in imag- minimization algorithm arrive at the “cor- perimental conditions and the higher sen- ing. Institute of Physics Pub (1998). rect” local minimum when performing a sitivity of HLA to inaccurate constraints. 8. Tikhonov, A.N. & Arsenin calibration in which other analytes may par- The third experiment was performed V.I.F.A.F. Solutions of ill-posed tially correlate with the analyte of interest. with turbid (scattering and absorbing) problems. Winston (1977). We have studied the effectiveness of samples, created by addition of intralipid CR using numerical simulations and ex- and India ink. The results are summa- perimental Raman spectra. In the inter- rized in Fig. 2. Substantial improvement est of space, only the experimental re- over both PLS and HLA is observed us- sults are presented here. In all studies, ing CR. As expected, the performance glucose and creatinine were chosen as of HLA is signiﬁ cantly impaired as a re- analytes of interest, while urea was al- sult of the turbidity-induced distortions ways present as an additional active Ra- in the analyte spectra. The distortions man spectral interference. Because the render the pure component spectrum an goal of these studies is not to champion inaccurate representation of the analyte spectrum as exists in the turbid medium. Cartoon courtesy Tom Swanson’s Science Car- detection limits, results are normalized toons (http://home.netcom.com/~swansont/sci- to PLS results, an objective baseline. In CR, however, the pure component ence.html).

Page 5 Research Report tion recognition. The dimensions of the energy). We have overcome this challenge markers are 0.5~1μm and the distance with our home-built Raman spectrometer. Characterizing the Chirality between them is ~20μm. The black lines In a typical measurement, the sample, indicate nanotubes which can be either di- depicted in Fig 1(a), is placed on an au- Distribution of Single-Walled rectly grown on the substrate or dispersed Carbon Nanotube Materials tomatic scanning stage and the laser spot from solutions of a bulk material. The is scanned over a 40μm by 20μm area with Tunable Raman Spectros- samples are then put onto an automatic with a 0.5μm X 1μm step size, and 1701 copy scanning stage of a home-built confocal Raman spectra are taken for the area be- Raman microscope. Figure 1(b) is the ing mapped. The laser excitation is then H. B. Son1,4, A. Reina2,4, M. S. Dresselhaus1,4, Kataura-plot calculated by the extended- tuned to a different wavelength and the and J. Kong1 tight binding method [3], where the op- same area is mapped. For SWNTs that tical transition energies E for different 1 Department of Electrical Engineering and ii are resonant at a particular laser energy, Computer Science (n,m) structural indices are plotted against their Raman spectra will then appear. If 2 Department of Material Science and Engi- the Raman radial breathing mode (RBM) the nanotube is long enough, the Raman neering frequency. The blue points correspond to signal along the nanotube can also be im- 3 Department of Physics semiconducting SWNTs and red ones cor- aged in a spatial mapping. This mapping 4 G.R. Harrison Spectroscopy Laboratory, Mas- respond to metallic SWNTs. These cal- method allows us to follow and identify sachusetts Institute of Technology, Cambridge, culations match the experimental data of Massachusetts, 02139-4037 each nanotube, regarding its spatial position. The tuning of the excitation energy (a) Single-walled carbon nanotubes allows the recognition of a SWNT go- (SWNT) are fascinating one-dimensional marker ing in and out of resonance. Therefore a materials with remarkable properties and resonance profi le of each SWNT can be many promising applications[1]. They can obtained and the optical transition energy be considered as tiny cylinders rolled up can be more accurately determined [6]. from a single layer of graphite and their With automatic scanning and data collec- structure can be indexed by two integers tion, hundreds to thousands of nanotube resonant Raman spectra can be obtained, (n,m) which, are called the chirality of nanotubes SWNTs. A very unique feature of SWNTs and this will allow us to perform the statis- is that their properties depend sensitively (b) tical analysis of the chirality distribution. on their chiralities; for example, a SWNT Figure 2 shows a typical result obtained is metallic if n-m equals 3 times an inte- with our tunable Raman spectrometer. ger, or semiconducting otherwise [2]. The In Fig 2(a) the substrate is Si with 300 bandgap of a semiconducting nanotube is nm thermally grown SiO2, and trenches inversely proportional to its diameter. This (which run horizontally in Fig 2(a)) of unique characteristic gives rise to many width 1.5, 3, 6μm are etched into the SiO2. different applications; on the other hand, SWNTs are grown on the substrates lying it poses a signifi cant challenge for the pro- across the trenches aligned in the vertical duction of SWNT material. With current direction [7]. Fig 2(a) demonstrates the synthetic methods, there is always a mix- Figure 1: Method for obtaining a chirality mapping result of Raman signals of a par- ture of SWNTs with a variety of chiralities distribution of a SWNT material. (a) Sche- ticular SWNT lying across the trenches. matic diagram of the sample used in these obtained, whereas for many applications Three panels are shown: the left one is the experiments. (b) Kataura plot calculated it is highly desirable to have SWNTs with by the extended-tight binding method [3]. G band Raman signal of the nanotube at only one type of structure. In order to have 1600 cm-1, the middle one is the RBM sig- -1 better control of the synthesis process, SDS-wrapped SWNTs very well [4]. The nal of this particular nanotube at 173 cm , - more understanding of how each synthetic 2n+m = constant families are denoted by and the right is the Si signal at 520 cm 1 parameter can be used to tune or control the dotted lines and the 2n+m family [5] . It can be seen that at the trench region, the chirality distribution of the SWNT numbers are also indicated. As the optical the Si signals are relatively lower, which material is needed. Therefore a good transition energy and the RBM signal of a is due to the fact that the auto-focusing method of characterizing the chirality dis- SWNT are obtained simultaneously in the is on the substrate surface. This allows tribution of a synthesized material is the resonant Raman process, the nanotubes’s us more capability for position recogni- fi rst step to enable such an investigation. chirality (n,m) can therefore be identi- tion making use of surface irregularities. Here we present our methodology of fi ed from the Kataura Plot [4]. However, From the left and middle panels, one can characterizing the chirality distribution since the optical transition energy spans a see that the same nanotube has a different of SWNT materials with tunable Raman wide range (from <1.5eV to >2.6eV), in peak intensity for the RBM and G-band Spectroscopy. Figure 1 explains the basic order to access most of the nanotubes with signals depending on whether the tube is procedure. Figure 1(a) is a schematic dia- the resonant Raman process, the Raman suspended or in contact with the substrate. gram of the sample used in our method. spectrometer needs to be tunable over a It has been proposed that the substrate The gray area is part of the substrate and similarly large range (1.5eV~2.6eV corre- interactions alter the optical transition the yellow squares are markers for posi- sponds to 470nm ~ 840nm laser excitation energy and Raman frequencies [4]; this

Page 6 is called the “environmental effect”. For effectiveness of the basic principle. In (a) different substrates and environments (for addition, we can see that there is a fairly example, solutions), the effects are differ- large environmental effect: the experi- ent, which complicates the chirality as- mental data points appear at lower excita- signment, a topic still under investigation. tion energy and at a slightly higher RBM Figure 2(b) demonstrates the exam- frequency compared with the calculated ple of the resonant profi le of a metallic ones. Since these calculated results fi t the nanotube, which is identifi ed as a (10,4) SDS-wrapped nanotubes in water solu- nanotube in the Katuara plot (it is circled tions fairly well, it is reasonable for the in Fig 1(b)). Due to the trigonal warping nanotube lying on Si to have a different effect [8], the optical transition energy for environmental shift. However, this com- non-armchair metallic nanotubes splits, as plicates the chirality assignment and ren- (b) can be seen in the Kataura plot. When the ders the (n,m) identifi cation not defi nitive RBM signal (240 cm-1) intensity versus when two (n,m) points are very close to the excitation energy is plotted, an inter- each other. Further in-depth investigations esting two-peak resonance can be seen in of the environmental effect will be carried the resonant profi le, confi rming the theo- out in order to achieve clear chirality as- retical prediction. The red dots are experi- (a) G band 3 RBM 2 Si 4 x10x 10 x10x 10 40 40 x10 mental data and the black curve shows the 4 4 440 9 2.525 8 8 calculated resonance profi les for the RBM 220 220 100100 220

7 intensity for both transitions obtained 220 0 0 0 Figure 3: Preliminary results for the compar- 0 0 800 0 8 6 from time-dependent third-order pertur- 6 ison of two samples grown at two different tem- -20 -20 -20 -2 1.515 -2 -2 5 bation theory [9]. The blue diamonds in- 6006 peratures (the rest of the growth conditions are

4 dicate the predicted resonant energy. From -4-40 -4-40 -4-40 4 all the same). (a) Growth carried out at 650ºC. 1 10 400 our results we concluded that there is a 4 3 Green dots are the data points. (b) Growth -6-60 -6-60 -6-60 2 2 at 900ºC. Yellow dots are the data points. downshift of ~70 meV for the resonant 0.55 2002 -80 -80 -80 -8 -8 -8 1 energy between the experimental data and at blue excitation energies, small ω 0 0 0 RBM -100 0 -100 0 -100 0 theoretical prediction, which is due to the -10-20 0 20 -10-20 0 20 -10-20 0 20 -2 0 2(Pm) -2 0 2(Pm) -2 0 2(Pm) are diffi cult to observe). Among these, environmental effect mentioned above. 130 correspond to metallic SWNTs, 126 With the spatial mapping and energy (b) Experiment correspond to semiconducting ones. The profi le capabilities confi rmed for our Ra- 0.15 ETB points Calculated same SWNT can be observed at different man spectrometer, we investigated two excitation energies if the laser energies are SWNT samples grown with the same cat- 0.10 close enough to be within the nanotube’s alyst and same chemical vapor deposition resonance window. We include this factor (CVD) conditions except for the growth into our consideration by using a ±0.1eV temperature. The catalyst is Al2O3 sup- 0.05 resonance window and a 50 meV environ- Intensity (arb. units) (arb. Intensity ported Fe/Mo nanoparticles [10] directly mental downshift (these parameters work spun onto Si/ SiO substrates and the 2 0.00 well for our experimental data points) and CVD conditions at the growth tempera- 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 found that it is expected to observe 177 Energy (eV) ture are methane with a fl ow rate of 1000 RBM signals corresponding to metallic P Figure 2: Typical results obtained with standard cubic centimeter per minute nanotubes and 166 RBM signals corre- (sccm) mixed with hydrogen of 500 sccm. the tunable Raman spectrometer. (a) Spa- tial mapping capability allows the reg- sponding to semiconducting nanotubes. The growth temperatures for the two sam- istering of each nanotube with regard to This means if there is no preference in ples are 650ºC and 900ºC, respectively. its position. The laser excitation wave- the chirality distribution, there should be In Figure 3 the data of the samples are length is 579 nm in this example. (b) Reso- a roughly 1:1 ratio of metallic to semi- shown. The calculated Kataura plots are nant energy profi le of a metallic nanotube. conducting nanotubes in these laser exci- also displayed for reference. Figure 3(a) signments in the future. Nevertheless, it tation regions. Our results give 130:126, corresponds to the data obtained with the which falls within this energy range. For sample grown at 650ºC (green dots), and can be seen from Fig. 3 that once the rest of the laser excitation energies are avail- the 650 ºC sample, we observed 29 RBM Fig 3(b) displays the data of the sample signals corresponding to metallic SWNTs grown at 900 ºC (yellow dots). The blue able, it will be straightforward to obtain a diameter distribution and metallic/semi- and 53 for semiconducting nanotubes in shaded areas indicate the regions where -1 -1 conducting ratio characterization easily. the region 160 cm ≤ ωRBM ≤ 340 cm . we have laser excitation energies avail- The number of data points at the current able. The size of the dots is proportional With these preliminary results, we performed a rough estimate of metallic to stage is not enough for making a statisti- to the logarithm of the intensities of the cally meaningful estimate for this sample. Raman signal. Currently we are limited semiconducting nanotube ratio. For the by the available laser sources, and thus 900ºC sample, we have observed 256 are not able to perform the chirality dis- RBM signals within the current excitation Acknowledgements We thank the -1 - tribution for these two samples. However, ranges for 160 cm ≤ ωRBM ≤ 340 cm Intel Higher Education Program for sup- 1 -1 these preliminary results demonstrate the (we exclude ωRBM < 160 cm because Carbon Nanotubes, continues on page 15 Page 7 Seminar on MODERN OPTICS AND SPECTROSCOPY Spring 2007

February 27 Dana Dlott, University of Illinois Ultrafast vibrational spectroscopy with high time and space resolution

March 6 Moungi Bawendi, MIT Nanocrystal quantum dot science and technology: The importance of spectroscopy

March 13 Daniel Murnick, Rutgers University Counting carbon 14 atoms for health improvement

March 27 Mriganka Sur, MIT Imaging cells, synapses, and molecules in the live brain

April 3 Xi-Cheng Zhang, Rensselaer Polytechnic Institute Recent development of THz wave technology for sensing and imaging applications

April 10 Andrei Tokmakoff, MIT The fluctuations and switching of hydrogen bonds in water

April 17 János Hebling, MIT Generation and application of high field THz pulses

April 24 16th Annual Richard C. Lord Lecture: Graham Fleming, University of California Berkeley Two-dimensional ultrafast electronic spectroscopy

May 1 Rebecca Richards-Kortum, Rice University Point-of-care optical diagnostics to improve global health

May 8 Adam Steeves, MIT Acetylene: What happens when a well-behaved molecule gets bent out of shape?

Tuesdays, 12:00 - 1:00 p.m., Grier Room (34-401)

Refreshments served following the seminar

Sponsored by the George R. Harrison Spectroscopy Laboratory, Department of Electrical Engineering and Computer Science and School of Science, MIT. GEORGE R. HARRISON SPECTROSCOPY LABORATORY TELEPHONE: (617) 253-7700 MICHAEL S. FELD, DIRECTOR FAX: (617) 253-4513 77 MASSACHUSETTS AVENUE, ROOM 6-014 URL: http://web.mit.edu/spectroscopy/ CAMBRIDGE, MASSACHUSETTS 02139-4307

Lester Wolfe Workshop in Laser Biomedicine

Optical Methods in Breast Cancer

Tuesday, April 17, 2007, 4:00-6:00 pm Richard B. Simches Research Center, Room 3110 185 Cambridge St. Boston (Located in Charles River Plaza next to Whole Foods Market and CVS)

*********************************************************************** Increasing incidence rates of breast cancer amongst US women and the importance of early detection in predicting outcome has led to searches for better screening methods. Mammography, although useful, has been called a double-edged sword due to the occur- rence of many false positives. This workshop will cover some applications of optics to the problem of breast cancer detection and diagnosis and monitoring response to therapy. ***********************************************************************

Introduction and clinical overview Barbara L. Smith, Massachusetts General Hospital

Shedding light on breast cancer: Advances in spectroscopic diagnosis Maryann Fitzmaurice, Case Western Reserve University

Diffuse optical tomography for breast cancer detection and characterization Arjun Yodh, University of Pennsylvania

Intra-operative optical biopsy of breast cancer Stephen Boppart, University of Illinois at Urbana-Champaign

Refreshments served at 3:30 pm

Sponsored by G.R. Harrison Spectroscopy Laboratory, MIT MGH Wellman Center for Photomedicine Harvard-MIT Division of Health Sciences and Technology and Center for the Integration of Medicine and Innovative Technology PLEASE POST Spectroscopy Laboratory Published Articles, 2006

Reprints of articles can be obtained by downloading from the Spectroscopy Laboratory and/or core faculty websites, and/or upon e-mail request ([email protected]). Angheloiu, G. O., Arendt, J. T., Muller, Dasari, R. R. and Feld, M. S., “Observa- J. A., Nazemi, J., Lyons, J., Hicks, D., M. G., Haka, A. S., Georgakoudi, I., tion of sub-poisson photon statistics in the Fitzmaurice, M., Dasari, R. R., Crowe, J. Motz, J. T., Scepanovic, O. R., Kuban, cavity-QED microlaser,” Phys Rev Lett P. and Feld, M. S., “In vivo margin assess- B. D., Myles, J., Miller, F., Podrez, E. A., 96 (9), 093603 (2006) ment during partial mastectomy breast Fitzmaurice, M., Kramer, J. R. and Feld, surgery using raman spectroscopy,” Can- M. S., “Intrinsic fl uorescence and diffuse Chung, I., Witkoskie, J. B., Cao, J. S. and cer Res 66 (6), 3317-3322 (2006) refl ectance spectroscopy identify superfi - Bawendi, M. G., “Description of the fl uo- cial foam cells in coronary plaques prone rescence intensity time trace of collections Halpert, J. E., Porter, V. J., Zimmer, J. P. to erosion,” Arterioscler Thromb Vasc of CdSe nanocrystal quantum dots based and Bawendi, M. G., “Synthesis of CdSe/ Biol 26 (7), 1594-1600 (2006) on single quantum dot fl uorescence blink- CdTe nanobarbells,” Journal of the Amer- ing statistics,” Physical Review E 73 (1), ican Chemical Society 128 (39), 12590- Anikeeva, P. O., Madigan, C. F., Coe-Sul- 011106 (2006) 12591 (2006) livan, S. A., Steckel, J. S., Bawendi, M. G. and Bulovic, V., “Photoluminescence DeCamp, M. F. and Tokmakoff, A., “Sin- Hodgkiss, J. M., Damrauer, N. H., Presse, of CdSe/ZnS core/shell quantum dots gle-shot two-dimensional spectrometer,” S., Rosenthal, J. and Nocera, D. G., “Elec- enhanced by energy transfer from a phos- Optics Letters 31 (1), 113-115 (2006) tron transfer driven by proton fl uctuations phorescent donor,” Chemical Physics Let- in a hydrogen-bonded donor-acceptor as- ters 424 (1-3), 120-125 (2006) Dresselhaus, M. S., Villalpando-Paez, F., sembly,” Journal of Physical Chemistry B Samsonidze, G. G., Chou, S. G., Dres- 110 (38), 18853-18858 (2006) Barros, E. B., Souza Filho, A. G., Son, H., selhaus, G., Jiang, J., Saito, R., Souza Mendes Filho, B. J., Jorio, A., Dressel- Filho, A. G., Jorio, A., Endo, M. and Kim, Hodgkiss, J. M., Rosenthal, J. and Nocera, haus, G. and Dresselhaus, M. S, “Double Y-A, “Raman Scattering from One-Di- D. G., “The Relation Between Hydrogen resonance Raman spectroscopy of gra- mensional Carbon Systems,” Physica E Atom Transfer and Proton-Coupled Elec- phitic foams,” Phys. Rev. B (2006) j.physe.2006.07.048, (in press) (2006)* tron Transfer in Model Systems,” In Hand- book of Hydrogen Transfer. Physical and Bechtel, H. A., Steeves, A. H. and Field, Endo, M., Kim, Y. A., Hayashi, T., Mu- Chemical Aspects of Hydrogen Transfer, R. W., “Laboratory measurements of the ramatsu, H., Terrones, M., Saito, R., Vol. 1: J.T. Hynes, R.L. Schowen, H.H. hyperfi ne structure of (HNC)-N-14-C-12 Villalpando-Paez, F., Chou, S. G. and Limbach, Eds.; Wiley VCH: Weinheim, and (DNC)-N-14-C-12,” Astrophysical Dresselhaus, M. S., “Nanotube coales- Germany (2006) Journal 649 (1), L53-L56 (2006) cence-inducing mode: A novel vibrational mode in carbon systems,” Small 2 (8-9), Hunter, M., Backman, V., Popescu, G., Brokmann, X., Bawendi, M. G., Coolen, 1031-1036 (2006) Kalashnikov, M., Boone, C. W., Wax, A., L. and Hermier, J-P, “Photon-correlation Gopal, V., Badizadegan, K., Stoner, G. D. Fourier spectroscopy”. Optics Express, Esswein, A. J., Veige, A. S. and Nocera, D. and Feld, M. S., “Tissue self-affi nity and 14:6333-6341 (2006) G., “Photocyle for hydrogen production polarized light scattering in the born ap- from homogeneous solution,” Abstracts proximation: a new model for precancer Caruge, J. M., Halpert, J. E., Bulovic, V. of Papers of the American Chemical So- detection,” Phys Rev Lett 97 (13), 138102 and Bawendi, M. G., “NiO as an inorgan- ciety 231 (2006) (2006) ic hole-transporting layer in quantum-dot light-emitting devices,” Nano Letters 6 Fang-Yen, C., “Quantum trajectory stud- Kojima, M., Tada, S., Umenmoto, S., Ishii, (12), 2991-2994 (2006) ies of many-atom and fi nite transit-time Y., Mizusaki, H., Muramatsu, H., Shima- effects in a cavity QED microlaser or mi- moto, D., Kim, Y.A., Hayashi, T., Endo, Cavalleri, A., Wall, S., Simpson, C., Statz, cromaser,” Optics Communications 262 M., Terrones, M. and Dresselhaus, M. S., E., Ward, D. W., Nelson, K. A., Rini, M. (2), 224-228 (2006) “In-situ Raman study on electrochemical and Schoenlein, R. W., “Tracking the mo- Li insertion into multiwalled carbon nano- tion of charges in a terahertz light fi eld by Fang-Yen, C., Yu, C. C., Ha, S., Choi, tubes,” Carbon (2006) femtosecond X-ray diffraction,” Nature W., An, K., Dasari, R. R. and Feld, M. S., 442 (7103), 664-666 (2006) “Observation of multiple thresholds in Liu, S. Y. and Nocera, D. G., “A simple the many-atom cavity QED microlaser,” and versatile method for alkene epoxida- Chan, Y. T., Snee, P. T., Caruge, J. M., Physical Review A 73 (4), 4 (2006) tion using aqueous hydrogen peroxide and Yen, B. K., Nair, G. P., Nocera, D. G. and manganese salophen catalysts,” Tetrahe- Bawendi, M. G., “A solvent-stable nano- Green, W. H., “Predictive Kinetics: A dron Letters 47 (12), 1923-1926 (2006) crystal-silica composite laser,” Journal of New Approach for the 21st Century,” Ad- the American Chemical Society 128 (10), vances in Chemical Engineering (in press) Liu, S. Y., Soper, J. D., Yang, J. Y., Ry- 3146-3147 (2006) (2006) bak-Akimova, E. V. and Nocera, D. G., “Mechanistic studies of hangman salo- Choi, W., Lee, J. H., An, K., Fang-Yen, C., Haka, A. S., Volynskaya, Z., Gardecki, Publications, continues on page 12 Page 11 Publications, continued from page 11 of human atherosclerosis and vulnerable S., Kastner, M. A. and Bawendi, M. G., phen-mediated activation of O-O bonds,” plaque,” J Biomed Opt 11 (2), 021003 “Temperature-, gate-, and photoinduced Inorganic Chemistry 45 (19), 7572-7574 (2006) conductance of close-packed CdTe nano- (2006) crystal fi lms,” Physical Review B 73 (15), Muzzey, D. and van Oudenaarden, A., (2006) Loparo, J. J., Roberts, S. T. and Tokmak- “When it comes to decisions, myeloid off, A., “Multidimensional infrared spec- progenitors crave positive feedback,” Cell Poulin, P. R. and Nelson, K. A., “Irrevers- troscopy of water. I. Vibrational dynamics 126 (4), 650-652 (2006) ible organic crystalline chemistry moni- in two-dimensional IR line shapes,” Jour- tored in real time,” Science 313 (5794), nal of Chemical Physics 125 (19), (2006) Oluwole, O. O., Bhattacharjee, B., Tols- 1756-1760 (2006) ma, J. E., Barton, P. I. and Green, W. H., Loparo, J. J., Roberts, S. T. and Tokmak- “Rigorous valid ranges for optimally re- Reece, S. Y., Hodgkiss, J. M., Stubbe, J. off, A., “Multidimensional infrared spec- duced kinetic models,” Combustion and and Nocera, D. G., “Proton-coupled elec- troscopy of water. II. Hydrogen bond Flame 146 (1-2), 348-365 (2006) tron transfer: the mechanistic underpin- switching dynamics,” Journal of Chemi- ning for radical transport and catalysis in cal Physics 125 (19), (2006) Park, Y. K., Popescu, G., Badizadegan, K., biology,” Philosophical Transactions of Dasari, R. R. and Feld, M. S., “Diffraction the Royal Society B-Biological Sciences Lu, J., Kopley, T., Dutton, D., Liu, J., phase and fl uorescence microscopy,” Op- 361 (1472), 1351-1364 (2006) Qian, C., Son, H., Dresselhaus, M. S. and tics Express 14 (18), 8263-8268 (2006) Kong, J., “Generating suspended single- Reece, S. Y., Seyedsayamdost, M. R., walled carbon nanotubes across a large Pedraza, J. M. and van Oudenaarden, Stubbe, J. and Nocera, D. G., “Electron surface area via patterning self-assembled A. “Noise in gene regulatory networks”, transfer reactions of fl uorotyrosyl radi- catalyst-containing block copolymer thin book chapter in ‘Complex Systems Sci- cals,” Journal of the American Chemical fi lms,” J. Phys. Chem. B 110 10585— ence in BioMedicine’ (Kluwer Academic, Society 128 (42), 13654-13655 (2006) 10589 (2006) New York, 2006) Rosenthal, J., Hodgkiss, J. M., Young, E. Lue, N., Popescu, G., Ikeda, T., Dasari, Ploeger, J. M., Bielenberg, P. A., Dinaro- R. and Nocera, D. G., “Spectroscopic de- R. R., Badizadegan, K. and Feld, M. S., Blanchard, J. L., Lachance, R. P., Taylor, J. termination of proton position in the pro- “Live cell refractometry using microfl uid- D., Green, W. H. and Tester, J. W., “Mod- ton-coupled electron transfer pathways ic devices,” Opt Lett 31 (18), 2759-2761 eling oxidation and hydrolysis reactions in of donor-acceptor supramolecule assem- (2006) supercritical water-free radical elementary blies,” Journal of the American Chemical reaction networks and their applications,” Society 128 (32), 10474-10483 (2006) Matan, K., Grohol, D., Nocera, D. G., Combustion Science and Technology 178 Yildirim, T., Harris, A. B., Lee, S. H., (1-3), 363-398 (2006) Rosenthal, J., Luckett, T. D., Hodgkiss, J. Nagler, S. E. and Lee, Y. S., “Spin waves M. and Nocera, D. G., “Photocatalytic ox- in the frustrated kagome lattice antiferro- Ploeger, J. M., Green, W. H. and Tester, idation of hydrocarbons by a bis-iron(III)- magnet KFe3(OH)(6)(SO4)(2),” Physical J. W., “Co-oxidation of methylphosphonic mu-oxo Pacman porphyrin using O-2 and Review Letters 96 (24), (2006) acid and ethanol in supercritical water - II: visible light,” Journal of the American Elementary reaction rate model,” Journal Chemical Society 128 (20), 6546-6547 McKee, T. D., Grandi, P., Mok, W., Al- of Supercritical Fluids 39 (2), 239-245 (2006) exandrakis, G., Insin, N., Zimmer, J. P., (2006) Bawendi, M. G., Boucher, Y., Breakefi eld, Samadani, A., Mettetal, J. and van Oude- X. O. and Jain, R. K., “Degradation of fi - Popescu, G., Badizadegan, K., Dasari, R. naarden, A., “Cellular asymmetry and brillar collagen in a human melanoma xe- R. and Feld, M. S., “Observation of dy- individuality in directional sensing,” Pro- nograft improves the effi cacy of an onco- namic subdomains in red blood cells,” J ceedings of the National Academy of Sci- lytic herpes simplex virus vector,” Cancer Biomed Opt 11 (4), 040503 (2006) ences of the United States of America 103 Research 66 (5), 2509-2513 (2006) (31), 11549-11554 (2006) Popescu, G., Ikeda, T., Dasari, R. R. and Mettetal, J. T., Muzzey, D., Pedraza, J. M., Feld, M. S., “Diffraction phase micros- Scepanovic, O. R., Fitzmaurice, M., Gar- Ozbudak, E. M. and van Oudenaarden, copy for quantifying cell structure and dy- decki, J. A., Angheloiu, G. O., Awasthi, S., A., “Predicting stochastic gene expression namics,” Opt Lett 31 (6), 775-777 (2006) Motz, J. T., Kramer, J. R., Dasari, R. R. dynamics in single cells,” Proceedings of and Feld, M. S., “Detection of morpholog- the National Academy of Sciences of the Popescu, G., Ikeda, T., Goda, K., Best- ical markers of vulnerable atherosclerotic United States of America 103 (19), 7304- Popescu, C. A., Laposata, M., Manley, S., plaque using multimodal spectroscopy,” J 7309 (2006) Dasari, R. R., Badizadegan, K. and Feld, Biomed Opt 11 (2), 021007 (2006) M. S., “Optical measurement of cell mem- Motz, J. T., Fitzmaurice, M., Miller, A., brane tension,” Phys Rev Lett 97 (21), Seyedsayamdost, M. R., Reece, S. Y., Gandhi, S. J., Haka, A. S., Galindo, L. 218101 (2006) Nocera, D. G. and Stubbe, J., “Mono-, H., Dasari, R. R., Kramer, J. R. and Feld, di-, tri-, and tetra-substituted fl uorotyro- M. S., “In vivo Raman spectral pathology Porter, V. J., Mentzel, T., Charpentier, sines: New probes for enzymes that use

Page 12 tyrosyl radicals in catalysis,” Journal of tional Edition, 45 5796-5799 (2006) the American Chemical Society 128 (5), Wong, B. M., Steeves, A. H. and Field, 1569-1579 (2006) Steckel, J. S., Yen, B. K. H., Oertel, D. C. R. W., “Electronic signatures of large am- and Bawendi, M. G., “On the mechanism plitude motions: Dipole moments of vi- Seyedsayamdost, M. R., Yee, C. S., Reece, of lead chalcogenide nanocrystal forma- brationally excited local-bend and local- S. Y., Nocera, D. G. and Stubbe, J., “pH tion,” Journal of the American Chemical stretch states of S-0 acetylene,” Journal rate profi les of FnY356-R2s (n=2, 3, 4) Society 128 (40), 13032-13033 (2006) of Physical Chemistry B 110 (38), 18912- in Escherichia coli ribonucleotide reduc- 18920 (2006) tase: Evidence that Y-356 is a redox-active Tharp, W. G., Yadav, R., Irimia, D., Upad- amino acid along the radical propagation hyaya, A., Samadani, A., Hurtado, O., Wong, B. M., Thom, R. L. and Field, R. pathway,” Journal of the American Chem- Liu, S. Y., Munisamy, S., Brainard, D. M., W., “Accurate inertias for large-amplitude ical Society 128 (5), 1562-1568 (2006) Mahon, M. J., Nourshargh, S., van Oude- motions: Improvements on prevailing ap- naarden, A., Toner, M. G. and Poznansky, proximations,” Journal of Physical Chem- Sheng, W. C., Kim, S., Lee, J., Kim, S. M. C., “Neutrophil chemorepulsion in de- istry A 110 (23), 7406-7413 (2006) W., Jensen, K. and Bawendi, M. G., “In- fi ned interleukin-8 gradients in vitro and situ encapsulation of quantum dots into in vivo,” Journal of Leukocyte Biology 79 Wong, H. W., Cesa, M. C., Golab, J. T., polymer microspheres,” Langmuir 22 (8), (3), 539-554 (2006) Brazdil, J. F. and Green, W. H., “Kinetic 3782-3790 (2006) modeling to estimate fundamental yield Tracy, J. B. and Bawendi, M. G., “Defects bounds for selective propylene oxidation Singer, A. B., Taylor, J. W., Barton, P. I. in CoO in oxidized cobalt nanoparticles over bifunctional catalysts,” Applied Ca- and Green, W. H., “Global dynamic op- dominate exchange biasing and exhibit talysis a-General 303 (2), 177-191 (2006) timization for parameter estimation in anomalous magnetic properties,” Physical chemical kinetics,” Journal of Physical Review B 74 (18), (2006) Yang, J. Y., Bachmann, J. and Nocera, D. Chemistry A 110 (3), 971-976 (2006) G., “Hangman salen platforms contain- Tsang, J. and van Oudenaarden, A., “Ex- ing two xanthene scaffolds,” Journal of Snee, P. T., Somers, R. C., Nair, G., Zim- citing fl uctuations: monitoring compe- Organic Chemistry 71 (23), 8706-8714 mer, J. P., Bawendi, M. G. and Nocera, D. tence induction dynamics at the single- (2006) G., “A ratiometric CdSe/ZnS nanocrys- cell level,” Molecular Systems Biology tal pH sensor,” Journal of the American (2006) Yu, C. C., Lau, C., Tunnell, J. W., Hunter, Chemical Society 128 (41), 13320-13321 M., Kalashnikov, M., Fang-Yen, C., Ful- (2006) Van Geem, K. M., Reyniers, M. F., Marin, ghum, S. F., Badizadegan, K., Dasari, R. G. B., Song, J., Green, W. H. and Matheu, R. and Feld, M. S., “Assessing epithe- Son, H. B., Reina, A., Samsonidze, G. G., D. M., “Automatic reaction network gen- lial cell nuclear morphology by using azi- Saito, R., Jorio, A., Dresselhaus, M. S. eration using RMG for steam cracking of muthal light scattering spectroscopy,” Opt and Kong, J., “Raman characterization of n-hexane,” Aiche Journal 52 (2), 718-730 Lett 31 (21), 3119-3121 (2006) electronic transition energies of metallic (2006) single-wall carbon nanotubes,” Physical Yu, J., Sumathi, R. and Green, W. H., Review B 74 (7), (2006) Vaughan, J. C., Feurer, T., Stone, K. W. “Accurate and effi cient method for pre- and Nelson, K. A., “Analysis of replica dicting thermochemistry of Furans and Son, H. B., Reina, A., Dresselhaus, M. S. pulses in femtosecond pulse shaping with ortho-arynes: Expansion of the bond-cen- and Kong, J., “Characterizing the chirality pixelated devices,” Optics Express 14 (3), tered group additivity method,” Journal distribution of single-walled carbon nano- 1314-1328 (2006) of Physical Chemistry A 110 (21), 6971- tube materials with tunable Raman spec- 6977 (2006) troscopy,” Physica Status Solidi B-Basic Villalpando-Paez, F., Zamudio, A., Elias, Solid State Physics 243 (13), 3161-3165 A. L., Son, H., Barros, E. B., Chou, S. Zalc, J. M., Green, W. H. and Iglesia, E., (2006) G., Kim, Y. A., Muramatsu, H., Hayashi, “NOx-mediated homogeneous pathways T., Kong, J., Terrones, H., Dresselhaus, for the synthesis of formaldehyde from Souza, A. G., Endo, M., Muramatsu, H., G., Endo, M., Terrones, M. and Dressel- CH4-O-2 mixtures,” Industrial & Engi- Hayashi, T., Kim, Y. A., Barros, E. B., haus, M. S., “Synthesis and characteriza- neering Chemistry Research 45 (8), 2677- Akuzawa, N., Samsonidze, G. G., Saito, tion of long strands of nitrogen-doped 2688 (2006) R. and Dresselhaus, M. S., “Resonance single-walled carbon nanotubes,” Chemi- Raman scattering studies in Br-2-adsorbed cal Physics Letters 424 (4-6), 345-352 Zimmer, J. P., Kim, S. W., Ohnishi, S., double-wall carbon nanotubes,” Physical (2006) Tanaka, E., Frangioni, J. V. and Bawendi, Review B 73 (23), 235413 (2006) M. G., “Size series of small indium arse- Weiss, D. N., Brokmann, X., Calvet, L. nide-zinc selenide core-shell nanocrystals Steckel, J. S., Snee, P. T., Coe-Sullivan, S. E., Kastner, M. A. and Bawendi, M. G., and their application to in vivo imaging,” A., Zimmer, J. P., Halpert, J. E., Anikeeva, “Multi-island single-electron devices Journal Of The American Chemical Soci- P. O., Kim, L-A, Bulovic, V. and Bawendi, from self-assembled colloidal nanocrystal ety 128 (8), 2526-2527 (2006) M. G.,“Color-Saturated Green-Emitting chains,” Applied Physics Letters 88 (14), QD-LEDs,” Angewandte Chemie Interna- 143507 (2006)

Page 13 Spectral Lines analyzing unknown substances. But the trappings of science fi ction but not aspir- instruments are known, if at all, by their ing to anything resembling scientifi c ac- Pop Spectrum casings, in the current laboratory equip- curacy or technical rigor. So we shouldn’t by Stephen R. Wilk ment style, and to the general public it blame him for his errors in science, but Textron Defense Systems doesn’t matter whether the instrument is a rather credit him for knowing what he Cambridge, MA Fourier Transform Infrared Spectrometer does and using it as a plausible-sounding or an Atomic Absorption Spectrometer or rationale for his background. a Mass Spectrometer. It’s a magical de- In Burroughs’ universe, there are at vice that spits out correct and infallible least two rays beyond the visible spectrum answers in no time at all and without that of seven colors, and these “eighth” and troubling ambiguity. (I know I would’ve “ninth” rays have wonderful properties. loved to have had one of those when I The ninth ray is used to manufacture oxy- found a very regular absorption structure gen in the Martian atmosphere factory. in one of my superoxide-doped laser crys- This ray is separated from the other rays tals that was turning the normally clear of the sun by means of fi nely adjusted crystals brown. It took me weeks to track instruments placed upon the roof of the down the source of that alien absorption. I huge building, three-quarters of which is Stephen R. Wilk fi nally tumbled to the fact that it was due used for reservoirs in which the ninth ray to ozonide only by analogy with the simi- is stored. Red, yellow, green, red, blue, blue, blue lar unexpected absorbing species S3- seen The eighth ray is also remarkable: This Red, purple, green, yellow, orange, red, red in sulfur-doped crystals. I sure could’ve ray, like the ninth ray, is unknown on Earth, Red, yellow, green, red, blue, blue, blue used a computer-assisted analysis routine but the Martians have discovered that it is Red, purple, green, yellow, orange, red, red with a built-in reference library.) an inherent property of all light no matter Usually one only hears the results, but from what source it emanates. They have Blend them all and what do you get? in the fi lm Outland you actually see a learned that it is the solar eighth ray which Cerise, chartreuse, and aqua, spectrum with peaks. This the doctor im- propels the light of the sun to the various Mauve, beige, and ultramarine, and every color mediately interprets as a particular drug on planets, and that it is the individual eighth in between, sight, without having to consult any refer- ray of each planet which “refl ects,” or pro- Hazo ka li ka no cha lum bum ences. But I’ll let that slide in the interests pels the light thus obtained out into space of compact and concise story-telling. At once more. Color has its harmony, and just like I have least you saw an analytical mind at work Burroughs doesn’t say where these rays said, there. Even in most science fi ction, ironi- lie relative to the visible spectrum (in- Red, yellow, green, red, blue, blue, blue, cally, you generally don’t see the mechan- deed, he suggests that they are themselves Red, purple, green, yellow, orange, red, red ics of analysis – you get instant results. visible, just unknown to terrestrial vision -The Spectrum Song (Ludwig von Drake) There’s very little of the physical nature and science). One naturally suspects they One of the questions that must greet of light or the spectrum in most pop litera- are infrared, but the atmosphere-making anyone working at the George Harrison ture. There are a very few exceptions, but properties suggest the high energy pho- Spectroscopy Laboratory – or any labora- they’re from the end of the nineteenth and tons of the ultraviolet. If one were feeling tory doing spectroscopy – is one addressed the beginning of the twentieth centuries. generous, one could argue that UV rays by spouses, children, relatives, and friends In 1894 Ambrose Bierce published the could provide both photochemical effects not in the sciences: “What do you do in short story The Damned Thing, about an and some form of laser propulsion, but there all day?” And so all of the profes- invisible creature that attacks people. The that’s more thought than Burroughs prob- sors, students, and technicians have to try creature is invisible because, as Bierce’s ably put into it. and explain in layman’s terms just what narrator tells us, Probably the most extravagant use of spectroscopy is, and why they do it. Not “At each end of the solar spectrum the the spectrum in pop literature is in H. P. everyone has the advantage of a friend or chemist can detect the presence of what Lovecraft’s 1927 story, The Colour out of relative working at a spectroscopy labora- are known as ‘actinic’ rays. They repre- Space. Lovecraft was an extremely cre- tory, however. What do they think is going sent colours – integral colours in the com- ative and infl uential author, whose works on? What is the pop culture perception of position of light – which we are unable to are still in print and avidly read. In The the spectrum and spectroscopy? discern. The human eye is an imperfect Colour out of Space he managed to make I’m not going to consider the rainbow, instrument; its range is but a few octaves a color the Monster in his story. It all be- or interference bands in soap bubbles and of the real ‘chromatic scale’. I am not mad; gins with the fall of a strange meteorite at oil fi lms. How is the practice and art of there are colours that we cannot see. the Gardner farm in the fi ctional town of splitting light into its constituent colors And, God help me! The Damned Thing Arkham, Massachusetts. Professors from and interpreting their intensities perceived is of such a colour!” nearby Miskatonic University examine it by the general public? It’s an explanation that makes sense, as and take samples, which subject to a bat- Pretty vaguely, for the most part. Spec- long as you don’t think about it too much. tery of tests. Lovecraft was an amateur troscopy is one of those tools seen in Edgar Rice Burroughs’ novel, A Prin- astronomer, and his descriptions of these forensic science television series such cess of Mars, is science fantasy, rather tests are pretty accurate. as CSI, one of a battery of tests used in than science fi ction, a work having the “…and when upon heating before the Page 14 spectroscope it displayed shining bands ter the fi rst atomic tests, radiation became Carbon Nanotubes, continued from page 7 unlike any known colours of the normal the source of B-movie monsters and other port on computer resources. Alfonso spectrum there was much breathless talk wonders. (Spiderman originally resulted Reina acknowledges MARCO intercon- of new elements, bizarre optical proper- from a student bitten by a radioactive nect focus center for fi nancial support. ties, and other things which puzzled men spider). As new effects and technologies of science are wont to say when faced by come into the limelight, they take over the unknown……They had uncovered this role. (In the recent movies, Spider- References what seemed to be the side of a large co- man was the result of a student being bit- 1. M. S. Dresselhaus, G. Dresselhaus, loured globule embedded in the substance. ten by a genetically engineered spider. and Ph. Avouris, Springer (2001) The colour, which resembled some of the Radiation is passé.) Spectroscopy was 2. C. L. Kane and E. J. Mele, Phys. bands in the meteor’s strange spectrum, already an old and established discipline Rev. Lett. 78, 1932 (1997). was almost impossible to describe; and by the time Bierce wrote, but the fl edgling 3. G. G. Samsonidze, R. Saito, N. Kobayashi, et.al, Appl. it was only by analogy that they called it fi eld of science fi ction and science fantasy Phys. Lett. 85, 5703 (2004). colour at all.” literature hadn’t yet had a chance to em- 4. C. Fantini, A. Jorio, M. Souza, et. al, Eventually, the meteorite dissolves and brace it. Over the next few years Bierce, Phys. Rev. Lett. 93, 147406 (2004). the evil color seems to invest all things Burroughs, Lovecraft, and a few others 5. S. M. Bachilo, M. S. Strano, C. Kit- growing on the farm, poisoning them. In used the still unfamiliar discipline to give trell, et.al, Science 298, 2361(2002). fact, Lovecraft’s description of the failing an air of verisimilitude to their fantasies. 6. A. Jorio, R. Saito, J. H. Hafner, et. health of the crops, animals, and people But now the technology is old and estab- al, Phys. Rev. Lett. 86, 1118 (2001). living on the farm strongly suggests to lished. There was a fl urry of interest in la- 7. Y. Y. Zhang, J. Zhang, H. B. the modern mind contamination by some sers when they were new, but it focused Son, J. Kong, and Z. F. Liu, JACS 127, 17156 (2005). chemical substance: on the potential destructive power of these 8. R. Saito, G. Dresselhaus, and M.S. Dres- “Specimens could be analyzed….under coherent beams. Laser spectroscopy was selhaus, Phys. Rev. B 61, 2981 (2000). the spectroscope both samples gave off an too esoteric to grab the public interest, 9. H.B. Son, A. R. Cecco, G. unknown spectrum, in which many of the and couldn’t really compete with real-life G. Samsonidze, et. al, manu- baffl ing bands were precisely like those “Death Beams”. script submitted to Phys. Rev. B. which the strange meteor had yielded in 10.J. Kong, H. T. Soh, A. M. Cas- the previous year.” Reference sell, C. F. Quate, and H. J. But in Lovecraft’s universe it was the 1.) Music and Lyrics by Richard M. Dai, Nature 395, 878 (1998). color itself that caused the blight, and and Robert B. Sherman for the Sept. 24, which had a sort of life of its own. “It 1961 inaugural episode of Disney’s The was just a colour out of space – a frightful Wonderful World of Color. Performed by messenger from unformed realms of infi n- Paul Frees as Ludwig von Drake. ity beyond all Nature as we know it…” Many years later, Lovecraft’s story was Answers to last issue’s crossword turned into a feature fi lm starring Boris The winner of the Fall 2006 Spectrograph crossword challenge was Scott Karloff, but the aspect of a monstrous Mohr, Visiting Fellow, MIT Engineering Systems Division. color was completely dropped. The title The winner of this issue’s challenge will be published in the Fall 2007 was changed to the more lurid and salable issue of The Spectrograph. Die, Monster, Die. The meteorite produc- es its effects by a physical contamination, probably radioactive in nature. There is no spectral analysis in the fi lm, but at the end Karloff’s character does glow with a weird greenish light. (When pressed to actually show a weird or unusual color, the default is invariably greenish.) A 1987 adaptation of the story bore the title, The Curse, and was even less memorable. It might seem odd that appearances of spectroscopy and odd colors in popular literature are confi ned to the fi rst quarter or so of the twentieth century, and virtu- ally ignored or taken for granted since, but it’s actually pretty typical. When a new technology or phenomenon appears, it becomes the item of interest in popular stories. When television fi rst became common, there were pulp magazine stories of evil televisions (I kid you not). Af-

Page 15 The Back Page

Optics Lab Essentials

1 2 3

4 5 6 7 8 9

10 11 12 13 14 15 16

17 18 19 20 21

23 24 25

27 28

Across Down 2. holographic grating type 1. optic mount with orthogonal rotation 6. multi-wavelength wave plate 3. shiny filter 7. stand-alone post 4. grating type commonly used in lasers 8. laser table screw 5. no spherical aberration 14. plano-plano 9. optic mount screw 15. grey-black filter 10. holds base to table 17. the bane of screw-hunters everywhere 11. colored glass filter 20. a degree of freedom 12. connects to optic mount 21. optic that gets you 2 for 1 13. true half-wave plate 22. safer than your eye 16. motion along an axis 23. colorblind lens 18. motion around an axis 26. 1-axis lens 19. optic mount with independent axis control 27. most common lens 21. essential yellow- or orange-handled tool 28. famously depicted on Pink Floyd album 24. reflective optic

MASSACHUSETTS INSTITUTE OF TECHNOLOGY Nonproﬁ t Organization G. R. HARRISON SPECTROSCOPY LABORATORY US Postage CAMBRIDGE, MA 02139-4307 PAID Cambridge, MA Permit No. 54016

Page 16