Skip to main content
Chemistry LibreTexts

Spectroscopy: Infochemistry using Atomic Emission Beacons

  • Page ID
    85524
  • \( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \) \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)\(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\) \(\newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\) \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\) \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\) \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\) \( \newcommand{\Span}{\mathrm{span}}\)\(\newcommand{\AA}{\unicode[.8,0]{x212B}}\)

    Article: C.N. LaFratta, I. Pelse, J.L. Falla, M.A. Palacios, M. Manesse, G.M. Whitesides, and D.R. Walt, “Measuring atomic emission from beacons for long-distance chemical signaling,” Anal Chem. 2013, 85, 8933-8936.

    This paper describes a method of sending information via atomic emission signals from emergency beacons consisting of alkali-metal doped string soaked in methanol. The authors describe a telescope for reliably quantifying atomic emission from three alkali metals at a distance of up to 1.7 km from the flame. This novel application is engaging to students while also addressing instrument design and fundamentals of atomic emission. Consequently, the assignment is designed to be used after previous in-class discussions of AES theory and instrumentation. Students who have taken linear algebra may appreciate the mathematics used to determine the instrument response matrix on p. 8935; however, the questions are designed so that it is not critical to understand these mathematical sections of the paper as long as students reach some conceptual understanding of why the response matrix is used (in-class question Q2).

    Out-of-Class Questions

    The out-of-class questions ensure students understand new terms (Q1, Q5) and review concepts from previous class periods on atomic emission (Q2, Q3, Q6).

    Q1. Define “infochemistry”.

    According to the abstract, infochemistry is “encod[ing] and transmit[ting] information using chemistry instead of electronics.”

    Q2. In this paper, the salts are atomized by combustion (i.e., a flame). What other methods for atomization for atomic emission are available?

    Individual instructors may vary in their coverage of atomization methods for emission, but common methods presented in textbooks include arcs, sparks, lasers, and of course, inductively coupled plasma (ICP).

    Q3. A methanol and air flame like the one used here typically obtains a maximum temperature of 2000 °C. What is the maximum temperature of an ICP? Why are high temperatures necessary for AES?

    The maximum temperature of an ICP is approximately 10,000 °C. The high temperature is needed to excite atomic electrons (since atomic emission, unlike atomic fluorescence, does not use light to do this). The high temperatures of an ICP also reduce the formation of complexes, minimizing interference from broad molecular absorption bands.

    Q4. Read section 10C-1 on p. 273 of the textbook. Why did the authors use alkali metals, instead of other elements, for this application?

    The section and page number reference given are for Skoog, Holler, and Crouch, Principles of Instrumental Analysis, 6th ed., Brooks/Cole, © 2007. Alternatively, students can be referred to "10.7.3 Quantitative Applications: Choice of Atomization and Excitation Source" in Chapter 10 of David Harvey’s freely available text Analytical Chemistry 2.0. Both texts address the fact that most other elements are not excited by the low temperatures of a methanol flame.

    Q5. What do the authors mean when they say the signal is “isotropic”? How is isotropic emission an advantage for this application?

    Isotropic processes occur in all directions equally. In this application, isotropic emission means that light will be emitted in all directions, so that the person transmitting the signal does not need to consider in which direction the receiver might be located.

    Q6. In the custom telescope in Figure 2, what components act as wavelength selectors? What wavelength selector is typically used in AES instrumentation? Explain the advantages of these two types of wavelength selectors for each application.

    The telescope uses band pass filters (BP) and dichroic mirrors (DC) as wavelength selectors. Band pass filters pass a range of wavelengths above and below specific thresholds. Dichroic mirrors typically reflect light with a wavelength lower than a threshold and transmit light of longer wavelengths. More conventional atomic emission instrumentation uses a monochromator to disperse all wavelengths so that many different elemental lines can be measured. Filters are good for applications where a small number of specific wavelengths will be measured but tend to be impractical for applications that scan across a wide wavelength range.

    [In my course, we discuss both filters and monochromators as wavelength selectors. For students who have not previously discussed the advantages and disadvantages of these options, this question could be moved to the in-class portion of the assignment.]

    Q7. What signal processing method was used on the data in Figure 3? What are the advantages of this method for this application?

    The signal is processed using boxcar averaging. [This question can be omitted from classes that do not discuss signal processing.] Students who are familiar with these techniques, however, should note that this method can be done in real-time without intensive computational resources; it does not require a repeatable signal (as would ensemble averaging); and although it can result in loss of temporal resolution, this is not an issue for this application since the beacons should burn steadily, and no useful information is encoded in their variation in time.

    Q8. Estimate the signal-to-noise ratio of the processed cesium signal in the top right panel of Figure 6.

    [This question is a review of signal-to-noise calculations and can be a useful reference for the in-class discussions of the distance LOD (in-class question Q1).]

    The average signal, S, is approximately 0.8 V. The peak-to-peak noise is approximately 1.3-0.6 V = 0.7 V. Assuming that the root mean square noise, N, is approximately one fifth the peak-to-peak noise gives N ≈ 0.7 V/5 = 0.14 V.

    \[\dfrac{S}{N}=\dfrac{0.8\: V}{0.14\: V}=6\]

    Q9. Go to the NIST Handbook of Basic Atomic Spectroscopic Data and look up the most intense persistent strong line(s) between 700-900 nm for Na, Li, and Ca in air. Lines indicated “P” next to the intensity refer to persistent lines, which are detectable even for low concentrations of the element in the presence of other species. http://www.nist.gov/pml/data/handbook/index.cfm#

    [Along with out-of-class Q10, this question ensures that students have the information needed to answer in-class Q4.]

    Sodium: 819.4824 nm

    Lithium: 812.6453 nm

    Calcium: 854.2089 and 866.2140 nm

    Q10. Peruse the bandpass filters available from Spectrofilm.com (the vendor used for parts in this manuscript). What is the narrowest range of wavelengths that can be passed by these commercial filters? How does this compare to the width of an atomic emission line?

    In addition to the lines above, the elements used in the paper were detected at 766, 852, 780 nm for K, Cs, and Rb, respectively. The closest lines are Cs at 852 nm and Ca at 854 nm. The Spectrofilm bandpass filters are available with bandwidths as narrow as 1 nm. The width of the atomic emission lines is 1.5-5 pm (even after Doppler broadening and other contributions beyond the natural line width), which means the two wavelengths will be well-separated.

    In-Class Questions

    Q1. Consider Figure 4.

    This question addresses a fundamental property of light, LOD calculations for non-linear relationships, and the practical considerations needed in designing beacons based on emission.

    1. From Figure 4, the equation for the calibration plot is \(y = 50760x^{-1.94}\), an inverse square relationship as expected based on the decrease in intensity of light with distance. The standard deviation of the blank, sbl = 9 mV, which means the signal LOD = 3*9 mV = 27 mV. Plugging this value in for y in the calibration plot gives us the distance at which the signal will drop to LOD:

      \[0.027=50760x^{-1.94}\]

      \[ x=\left(\dfrac{0.027}{50760}\right)^{ \Large{{}^1/_{-1.94} } } \]

      Solving for x gives 1700 m, the reported LOD.

    2. The intensity is related to distance by an inverse square relationship (rounding from -1.94 to ‑2 for the exponent, as predicted by theory). As a result, increasing the distance by a factor of 2 means that the intensity of the signal must be increased by a factor of 4.

      [Students may initially try to solve this problem using the calibration curve equation from (a). It is helpful to explain that an increase in the signal means that this calibration curve will change since the same distance will give a higher signal value.]

      Some students with good mathematical intuition will readily appreciate the explanation above. Students who value a more concrete mathematical explanation may find the following helpful:

      \[2x=2*1700\: m=3400\: m\]

      We need the signal at 3400 m to be 27 mV, assuming that the LOD is improved by increasing the signal rather than decreasing the noise. The current signal at 3400 m is 7 mV, as shown below:

      \[y=50760(3400)^{-1.94}=0.007\]

      If the signal LOD is 27 mV, then this means that the signal must increase by a factor of 27/7, or 3.9. (The discrepancy between the 3.9-fold increase in intensity calculated here and the 4-fold increase estimated above arises because the exponent used in these explicit calculations is -1.94 rather than -2.)

    3. For this question, I encourage the students to visualize the process of preparing the flares, burning them, and detecting the light. Possible complications could involve solubility limits of the salts being used and self-absorption of the emission at high concentrations.

      If desired, students can investigate this question more quantitatively. The supporting information states that between 15-375 µL of 1 M solutions of the nitrate salts were used to prepare the beacons. Students can research the solubility limits for the three salts, which are very near 1 M. At 20 °C, the limits for nitrate salts of cesium, potassium, and rubidium are 23 g, 33 g, and 53 g per 100 g of water respectively. Ignoring changes to solvent volume, these concentrations correspond roughly to 1.2 M, 3.3 M, and 3.6 M. These values mean that increasing the concentration of each salt in the beacon by a factor of 4 would be impossible to do simply by increasing the concentrations of the solutions used to prepare the beacons. Depending on the time allotted for this question, the supporting information and solubility limits can be given to the students so that they can do these calculations themselves, or this information can be presented by the instructor once all the groups have reported on their discussions.

    Q2. In the top panels of Figure 6, why is it that the cesium signal can be the highest magnitude in the raw data, but is at the “low” level after decoding in the processed data?

    This question is meant to lead into Q3, which asks students to explain the instrument response matrix. The emission of Cs is higher at all concentrations than the emission of the other elements. The high and low levels of Cs relative to the other elements are only apparent after normalization using the instrument response matrix. The higher intensity of the Cs emission compared to the K and Rb emission at equal concentrations (as seen in Figure 3) is one of the more easily grasped difference between the three “channels” used to encode information in the beacons.

    Q3. What is the purpose of the instrument response matrix and what factors affect it?

    While the mathematics on p. 8935 of the article may not be accessible to students who have not taken linear algebra, the text does provide some guidance here, noting on the same page that “The 3 × 3 instrument response matrix, R, corrects for differences between the three channels of the detector and any crosstalk.” The key is for students to consider what these two items mean and explain them in their own words. Difference between the three channels include differences in the intensity of the emission from the three metals, as highlighted in the previous question, Q2. Other differences might include the sensitivity of the PMTs to specific wavelengths and other imperfections in the instrumentation. In my experience, this may be the first time that students consider, for example, that filters may not be perfect. If students have not looked up the term “crosstalk” or learned about it in a previous class, a short conversation about what this term means is helpful. In particular, students will probably need to be told that crosstalk can be spectral as well as electrical.

    Q4. The authors suggest that greater information density could be achieved by adding Na, Li, and Ca to the signal. Do you think that this would be feasible? Support your answer using the information you looked up for out-of-class questions 9 and 10.

    See notes for out-of-class Q10 above. Make sure that the students consider both the width of the bandpass filters and the width of the emission lines. Encourage students to consider the added complexity of the instrument with the addition of three more channels. For students who have not taken a physics class with an optics component, it may be helpful to tell them that light is lost at each interface (mirror, filter, etc.) and have them consider the effect of additional signal channels on LOD. Some students may suggest redesigning the telescope to include a monochromator, leading to a discussion of cost and the increased complexity of instruments that involve moving parts.

    Q5. The authors suggest that this beacon could be used to transmit data in resource-poor environments, such as natural disaster sites. Evaluate the feasibility of this application given the information in the manuscript.

    Students should consider the distance LOD, the information density, the susceptibility of the technique to sunlight and adverse weather conditions. When students have concluded fairly quickly that the telescope is not practical for emergency situations, I often start a conversation about whether this research and the idea of “infochemistry” in general might have other applications or value.


    This page titled Spectroscopy: Infochemistry using Atomic Emission Beacons is shared under a CC BY-NC-SA 4.0 license and was authored, remixed, and/or curated by Contributor.