Seelye, M., Schaare, P., & Künnemeyer, R. (2014). Evaluating the mineral nutrient status of fresh pasture herbage using laser-induced breakdown spectroscopy. Presented at the The 8th International Conference on Laser Induced Breakdown Spectroscopy (LIBS 2014), Tsinghua University, Beijing, China, 08 - 12 September 2014.
Permanent Research Commons link: https://hdl.handle.net/10289/9047
Comprehensive determination of the mineral nutrient status of pasture or horticultural crops currently requires leaf or herbage samples to be harvested and taken to a laboratory for analysis, which is both labour-intensive and time-consuming. This study examined the potential of laser-induced breakdown spectroscopy (LIBS) to provide immediate, in-field assessment of the mineral nutrient status of standing plants. Success would offer the prospects of spatially and temporally improved knowledge of plant status and of real-time remediation of nutrient deficiency. The mineral nutrient status of fresh pasture herbage has been evaluated using an Applied Photonics Ltd？ LIBS-6 system, comprising a Qswitched Nd:YAG laser, a SpectroModule-6 spectrometer (covering wavelengths in the range ~185 – 1064 nm) and an ICE 450 water-cooled Laser power supply. LIBS spectra obtained from the fresh pasture samples were collected with the aim of determining the nutrient status of the herbage. An enclosed Modular Sample Chamber housed the Qswitched Nd:YAG laser, generating energy in the near infrared region at λ = 1064nm. The laser power was set to 100mW per pulse. Samples of approximately 50g of fresh ryegrass and ryegrass/clover mixed pasture were loaded into a sample holder which allowed a flat ‗carpet‘ of individual leaves to be presented at a relatively uniform distance from the laser. The optimum vertical distance between sample and laser was determined by maximising spectra amplitudes. Automated control of the translation stage supporting the sample holder was used to ensure each shot of the laser was acquired from a fresh location. Spectrum sets comprising both 100 individual spectra and 100 shot accumulated spectra were obtained from each sample under both air and under the inert gas, argon. Following the spectral analysis of the fresh pasture, each sample was sent to a commercial laboratory for standard nutrient analysis, providing elementary composition on the most common plant elements (N, P, K, S, Ca, Mg, Na, Fe, Mn, Zn, Cu and B). Once standard nutrient analysis was completed a second set of spectra was taken from tablets formed from compressed dried ground powder remaining after laboratory analysis. To date, initial results from simple chemometric analysis have shown limited success in predicting the nutrient content of fresh pasture with slightly improved results in estimating tablet composition. While initial analyses have concentrated on trends and correlations between actual spectra and laboratory analysis, known emission spectra, have for the meantime, been ignored. Further in-depth analysis using more robust chemometric analysis, utilizing known emission spectra and adjusting for variations in plasma intensity are currently underway and are expected to improve the accuracy of composition measurement.
© 2014 copyright with the authors.