Honey is a complicated food to analyze. Because it is difficult to know from where, exactly, the bees get their nectar, it is difficult to label honey accurately. Some honeys claim to be unifloral, with distinct tastes and characteristics, but most are, in fact, polyfloral, made by bees that have collected nectar from a variety of sources.Researchers from ETH Zurich and the Swiss Bee Research Centre evaluated three types of spectroscopy for honey authentication. Several analytical methods exist for distinguishing between unifloral and polyfloral honeys, but they are time-consuming and costly. Also, specialized expertise is needed to interpret the pollen spectrum used to determine the geographic origin of honey. Therefore, there is a need for an analytical technique that will enable rapid and reproducible authentication of the botanical and geographic origins at low cost.Kaspar Ruoff believes that spectroscopy could be the answer. Working at ETH Zurich and at the Swiss Bee Research Centre in Bern, both in Switzerland, Ruoff collaborated with colleagues from various European institutions to evaluate three types of spectroscopy for honey authentication: front-face fluorescence, Fourier transform near-IR (FT-NIR) and Fourier transform mid-IR (FT-MIR) spectroscopy using an attenuated total reflectance (ATR) cell. The scientists reported their findings in three papers published online Aug. 12 in the Journal of Agricultural and Food Chemistry.“ATR-MIR spectroscopy is the most promising method for determining the botanical origin of honey,” Ruoff said. He used the technique to identify unifloral honeys and to distinguish them from polyfloral honeys. The technique is nondestructive, rapid, easy to use and inexpensive. It needs neither particular sample preparation nor special qualification of laboratory personnel.“Our results show that the authentication of the botanical origin of honey by using ATR-MIR spectroscopy and chemometrics is in agreement with the determination using classical criteria.” He added that another advantage ATR-MIR spectroscopy has over other types of spectroscopy is that the same spectra can be used to obtain quantitative information about several measurands used for the routine quality control of honey.In their ATR-MIR experiment, the investigators recorded FT-MIR spectra using a Bio-Rad FTS-7 along with an MKII Golden Gate single-reflection ATR accessory from Specac Inc. of Woodstock, Ga. The spectrometer was equipped with a deuterated triglycine sulfate detector and was operated at 4 cm–1 spectral resolution. The scientists liquefied the honey samples in a water bath at 55 °C for eight hours and allowed them to cool to room temperature before analysis.Ruoff determined that front-face fluorescence spectroscopy, while a sensitive technique, would be difficult to automate because it requires the use of neat honey. FT-NIR spectroscopy combined with chemometrics showed promise for authentication of certain unifloral honeys, and Ruoff believes that problems related to the determination of the polyfloral honeys could be handled by the successive use of at least two mathematical models. In fact, he noted that the choice of mathematical models could be even more important than the choice of analytical technique.“A drawback of the spectroscopic techniques we studied is the considerable work involved in building the classification models before they can be used in analytical practice,” Ruoff said. “However, in laboratories with a high sample throughput, the investment will probably pay back soon, as labor is the most cost-intensive factor in honey analysis.”Spectroscopic techniques allowed the botanical origin, as well as the most important physical and chemical properties of honey, to be determined within a few minutes instead of the hour or more required for traditional methods. ATR-MIR spectroscopy also correctly classified honey samples according to their geographic region, whereas front-face fluorescence- and FT-NIR spectroscopy could not.The researchers successfully combined ATR-MIR spectroscopy with multivariate calibration algorithms such as partial least squares regression to set up calibrations that enabled them to accurately predict the concentrations of the main components in honey. “The fact that ATR-MIR spectroscopy allows quantitative analysis is important,” Ruoff said. “Interestingly, we also managed to accurately measure characteristics that are not infrared-active, such as pH and electrical conductivity.” Unfortunately, infrared spectroscopic methods did not enable the research team to quantitatively determine hydroxymethylfurfural and enzyme activities — two criteria important for the honey trade because they indicate storage and/or heat damage. The substances were too low in concentration and lacked infrared absorptionRuoff has founded a company, Inframiel, to share his knowledge with other analysts and to offer chemometric data evaluation and spectroscopic analysis of honey. Contact: Kaspar Ruoff, Inframiel, Brugg, Switzerland; +41 56 442 45 29; e-mail: kaspar.ruoff@inframiel.ch.