Not all honey is created equal. When bees feast on the pollen from different flowers, they produce honey that differs greatly in taste and appearance. For example, a study by the National Honey Board of Longmont, Colo., revealed that bees that dine on buckwheat create dark honey with a medicinal taste, while those eating clover produce sweet and transparent honey.Jagdish C. Tewari of Purdue University loads a Fourier transform infrared spectrometer with a sample of honey. The results, together with information from a surface acoustic wave sensor, indicate the floral origin of the sweet substance.Now postdoctoral researcher Jagdish C. Tewari and associate professor in agricultural and biological engineering Joseph M.K. Irudayaraj from Purdue University in West Lafayette, Ind., have used Fourier transform infrared (FTIR) spectroscopy and surface acoustic wave sensing to identify the flowers responsible for the sweet fluid.Honey is a mix of carbohydrates, water, organic acid traces, enzymes, pigments, pollen and other ingredients. Because of the difference in price that various floral honeys command, there is a temptation to adulterate lower-grade honey with additives that mimic the constituents in a more desirable product.It is important for industry to know the floral source of honey because of its economic importance and to maintain the quality of honey,” Tewari said.Analytical methods such as gas chromatography/mass spectrometry that verify a floral source are expensive and involved. They also may require reagents or chemicals that pose environmental concerns. Even more serious, however, is their inability to perform online measurement or field analysis, Tewari said.According to the researchers, the dual approach overcomes these problems in part because the techniques are complementary. FTIR identifies the nonvolatile components of honey via the vibration of molecular bonds. Surface acoustic wave sensing, on the other hand, ferrets out information related to the volatile flavor compounds.The scientists collected samples from various geographical regions and of various floral origins with the help of the National Honey Board. They used a spectrometer equipped with a microattenuated total internal reflectance sampling attachment in the FTIR work. For these measurements, they placed a drop of honey on the attachment’s ZnSe crystal, sent an infrared beam into the crystal at the proper angle and reflected it off the sample/crystal interface. The beam penetrated the honey by about 1.5 µm and picked up its infrared fingerprint.For surface acoustic wave sensing, they used a zNose instrument from Electronic Sensor Technology LP of Newbury Park, Calif. A vibrating crystal sits inside the zNose, and the frequency of the vibration changes as molecules are adsorbed on its surface. The instrument has the speed of an electronic nose but the precision and accuracy of a gas chromatograph, the investigators noted.After analyzing the FTIR spectra and zNose results, they found that a neural network and chemometrics could classify the known honey samples into seven floral categories. They then used this approach to categorize unknown samples, achieving close to 96 percent accuracy on a classification validation set.Tewari said that a fused sensor capable of both FTIR and surface acoustic wave measurements would be needed for industrial and on-site honey grading. Such a tool might save overhead expenses and time while increasing product quality and reliability. However, he added, such a device does not yet exist.The scientists are not working to produce such an instrument because their current focus is on extending the technique to other applications. The technology can spot adulteration in petroleum products as well as identify food contaminants. It also can be used to detect biological and chemical weapons as well as for other security applications.The two scientists performed the research while at Pennsylvania State University in University Park, in collaboration with the National Honey Board. They published their results in the Sept. 7 issue of Journal of Agricultural and Food Chemistry.