Polymer-bound sensor systems

Operative lifetimes of the PVC entangled sensor systems so far described are considerably shorter than their solid-state counterparts, e.g. the lanthanum fluoride model. The principal cause is the loss of active component(s) from the polymer matrix. Thus, the deterioration of PVC potassium electrodes relates to the leaching of potassium membranes with an initial resistance of 5.8MD left in contact with deionized water for four days became paler and the resistance rose to 30MD compared with 50MQ for a PVC-only membrane. Similar, but slower, detrimental processes arise with PVC nitrate electrodes (74). 

The stiffening of calcium ISE membranes based on neutral carrier (4) and DOPP after use in 0.1 M sodium dodecyl sulphate indicates a similar loss of 

PVC T z Resistance Slope (mV Limit of Selectivity coefficient, (separate solution method)  

Models to predict operational lifetimes have been made on the kinetics of loss of mediator or sensor from the PVC into the sample solution. To ensure a continuous lifetime of one year, or more, the partition coefficient for both these components between sample and membrane should exceed 10 (30). 

Several grafted calcium ISEs have been reported (73,76,77). In one instance membranes were synthesized by cross-linking the styrene-b-butadiene-b-styrene (SBS) triblock elastomer with triallylphosphate, followed by alkaline hydrolysis of the resultant covalently bound trialkyl phosphate to give a pendant dialkyl phosphate. Good calcium electrodes with lifetimes greater than 6 months were obtained, but the limited selectivity for magnesium and sodium could relate to the absence of a solvent mediator (76). 

---

James and co-workers have developed polymer sensors by grafting a solution based D-glucose selective receptor 27d to a polymer support [78]. The major difference between the polymer-bound system 86 and solution-based system 27d is the D-glucose selectivity, which drops for polymer 86 (whereas the selectivity with other saccharides is similar to those observed for compound 27d). However, the polymeric system still has enhanced D-glucose selectivity (nine times) over the monoboronic acid model compound. The reduced binding of 86 for D-glucose has been attributed to the proximity of the receptor to the polymer backbone. 

The sensors discussed so far are based on ligands covalently bound to the polymer backbone. Other methods of detection - often referred to as mix and detect methods - work by simple noncovalent incorporation of the polymer with the ligand of interest. Reichert et al. generated liposomes of polydiacetylene with sialic acid for the same purpose of detection as Charych s surface-bound polymers, but realized that covalent functionalization of the polymer was not necessary [17]. Through simple mixing of the lipid-bound sialic acid with the polymer before sonication and liposome formation, they were able to form a functional colorimetric recognition system (Fig. 8). 

Based on the observation that chloramphenicol (CAP)-imprinted polymer possessed a modest affinity for chloramphenicol-methyl red (CAP-MR), Levi et al. [65] designed an intriguing MIP sensor to monitor the change of CAP in patients blood (Fig. 6). The presence of CAP in blood leads to a competitive displacement of CAP-MR from the imprinted cavities. The displaced composite is subsequently monitored at 460 nm. After optimizing the flow rate and concentration of CAP-MR in acetonitrile mobile phase, the response of this system to CAP, thiamphenicol (TAM), and chloramphenicol diacetate (CAP-DA) was determined (Fig. 7). As observed for CAP, there was a linear correlation over the range 1-1000 pg/mL. However, for CAP-DA almost no appreciable response was achieved, even if it was injected to 1000 pg/mL. As also observed, the value for CAP was about 40% higher than that for TAM at the same concentration. This revealed that CAP could compete more efficiently with the bound CAP-MR than TAM did. Further information showed that this method was adequate for detection below and above the recommended therapeutic range (10-20 pg/mL serum, potentially toxic above 25 pg/mL). 

---