88

౪

4 Smell and Taste

Figure 4.5: Cells are immobilized onto graphene and used as an NO sensor. NO is detected via chronoam-

perometry [19]. (A) Photograph of graphene biofilm. Side-view (B and C) and top-view (D) SEM images of

the graphene biofilm. (E) IR spectra of (a) pyrenebutyric acid functionalized graphene film and (b) RGDpep-

tide covalently bonded graphene biofilm. (F) Fluorescence staining of graphene biofilm.

A common signal is shown in Figure 4.6 [20]. A concentration of 10 ppm of the alcohols

(and possibly less) could be identified in about 20 s. Different sensors gave repeatable

results within an acceptable error. When measuring mixtures of alcohols, the different

alcohols could not be identified but the mixtures could still be measured. So, the sensor

is not specific, but could be used in combination with specific sensors to increase the

sensitivity of detecting rotting meat.

In another example, mouse G-protein-coupled receptors (GPCR) were used [21].

Three different receptors were coupled to carbon nanotube resistors. Since GPCRs are

membrane proteins, they need to be stabilized inside a membrane. In this case, a “nan-

odisc” was used (Figure 4.7) (in nanodiscs, a lipid bilayer is surrounded by membrane

scaffolding proteins that stabilize a small area of lipid bilayer). A change in current is

measured when an odorant binds the receptor.

These devices are stable for at least a month. There is a significant drift in the re-

sponse baseline, but that can be accounted for if one records the percent change from the

baseline current (% ΔI/I). Since the variability between different sensors is also a change