20 SUN ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 80, No. 1,1997 AGRICULTURAL MATERIALS Determination of Boron in Plants and Plant-Derived Foods by Ultrasonic Nebulization-Inductively Coupled Plasma Atomic Emission Spectrometry with Addition of Mannitol DA-HAI SUN, JAMES K. WATERS, and THOMAS P. MAWHINNEY University of Missouri-Columbia, Experiment Station Chemical Laboratories, Room 4 Agriculture Building, Columbia, MO 65211 An ultrasonic nebulization-inductively coupled plasma (ICP) atomic emission spectrometry proce- dure was developed to determine boron in plants and plant-derived foods. Samples were ashed at 550°C for 4 h, and the resulting ash was dissolved into a mannitol-nitric acid solution to form a boron- oron is an essential micronutrient for higher plants (1). Boron is also an unavoidable component of plant-de- rived foods, although its specific role in human nutrition is still unclear (2). Understanding the role of boron in plant growth and in the human diet requires precise and accurate de- termination of this element. Analytical methods for determin- ing boron include spectrophotometry (3), atomic absorption spectrometry (4), neutron activation mass spectrometry (NAMS; 5), inductively coupled plasma atomic emission spec- trometry (ICP-AES; 6-9), and inductively coupled plasma mass spectrometry (ICP-MS; 10, 11). Among these, NAMS is the most sensitive for boron trace analysis. But NAMS is not a common technique and is very time-consuming. ICP-MS is also very sensitive for elemental analysis, but instrumental costs and spectral interference of' C limit its extensive appli- cation in boron determination. Spectrophotometry and AAS Received June 17, 1996. Accepted by RN August 16, 1996. either include a complicated chemical sample pretreatment or have very poor sensitivity. ICP-AES is a powerful and time-saving method for mul- tielement analysis and is widely used to determine boron (6-9). However, the sensitivity of ICP-AES with typical pneumatic nebulizers is often not sufficient when sample amount is lim- ited or boron concentration is relatively low (<1 \ig/g). Exten- sive efforts have been expended to improve detection limits (DLs) of ICP-AES. Some auxiliary techniques, such as flow injection analysis, hydride generation/cold vapor, and ultra- sonic nebulization (USN), have been developed to improve de- tection for trace analysis (12). Compared with standard pneu- matic nebulization, USN has advantages. It can generate a homogeneous aerosol of fine (<5 Lim) droplets, the carrier gas flow rate and generation rate of aerosol can be varied inde- pendently, and more important, USN has a very high nebuliz- ing efficiency, which results in an order of magnitude increase in sensitivity (13). Despite these advantages, USN has not been accepted for boron analysis. The main reason may be sample memory ef- fects. These effects are attributed to renebulization of residual droplets of previously nebulized sample and to existence of dead volumes for most elements (12). Memory effects for most elements have been reduced markedly with modified systems in the past few years (13). With most commercial USNs, only 35^10 s is needed to return signals to blank values for most elements when their concentrations are about 10 000 times higher than DL (12). The mechanism of the memory effect for boron seems to be different from that of other elements, be- cause the emission signal of boron does not return to a blank value even after 25 min of flush time after a solution of 100 mg B/L is introduced into an ICP via USN (14). The key to boron analysis with USN is to greatly reduce or eliminate the serious memory effect. In 1994, Evans and Krahenbiihl (11) indicated that for sam- ple introduction with a pneumatic nebulizer the memory effect of boron could be reduced by rinsing with a slightly acidified (0.02M HN0 3 ) NaF solution (2 mg/g). They assumed that the borate, which tends to adsorb to the sampling system surface, is complexed by high excess of fluoride to stable boron tri- fluoride and then washed out. But NaF could not significantly reduce the memory effect of boron with USN. Mannitol, a mannitol complex. Solutions were directly nebu- lized into ICP by an ultrasonic nebulizer. The mem- ory effect—the main drawback for determining bo- ron with the ultrasonic nebulizer—was minimized with addition of mannitol. Two National Institute of Standards and Technology (NIST) standard refer- ence materials (SRM 1515 apple leaves and SRM 1547 peach leaves) and 25 food and plant ma- terials spiked with NIST SRM 1515 were analyzed to verify method accuracy. Boron contents found in NIST standard samples were in excellent agree- ment with certified values. A detection limit (2o) of 0.5 (ig/L and typical recoveries of 93 to 106% from spiked selected samples were obtained. Downloaded from https://academic.oup.com/jaoac/article-abstract/80/1/20/5684432 by guest on 25 July 2020