The effect of mediator distance from the polymer backbone on the redox behavior, electron transport, electrochemical stability, and electrical communication with redox enzymes was studied with novel linear poly(ethylenimine) redox polymers. To measure these effects, we synthesized two new ferrocene redox polymers having a three-carbon (Fc-C 3 -LPEI) and a six-carbon (Fc-C 6 -LPEI) spacer to complement our previous synthesized polymer (Fc-C 1 -LPEI) having a 1-carbon spacing. Increasing the spacer length to either three or six carbons resulted in a single peak redox behavior over the entire pH tested (3-11), which is in contrast with the multiwave redox behavior observed with a 1-carbon spacing under neutral or basic pH solutions. In addition increasing the spacer also resulted in a six-fold increase in the electrochemical stability of crosslinked redox polymer films. In contrast with previous reports of electron transport increasing with spacer length, we observed no correlation between mediator spacing and electron transport or electrical communication with the enzyme glucose oxidase. Surprisingly, the redox polymer (Fc-C 3 -LPEI) with the lowest electron transport produced the highest enzymatic response (>1 mA/cm 2 ), suggesting that other factors (e.g., polymer-enzyme complexation) were important. These results demonstrate how small changes in the redox polymer structure can affect its properties.

Amperometric biosensors for glucose and hydrogen peroxide have been built by immobilizing glucose oxidase (GOX) and horseradish peroxidase (HRP) in cross-linked films of ferrocene-modified linear poly(ethylenimine). At pH 7, the glucose sensors generated limiting catalytic current densities of 1.2 mA/cm2. These current densities are approximately 4 times higher than those with other ferrocene-based redox polymers and are comparable to the highest reported values for osmium-based redox polymers with GOX. Because of the high sensitivity of these films (73 nA/cm2.microM), glucose concentrations in the micromolar range could be detected. Similarly, sensors were constructed with HRP-generated current densities of 0.9 mA/cm2 under saturation conditions and sensitivities of 500 nA/cm2.microM. The results show that the ability of Fc-LPEI to effectively communicate with a variety of enzymes has potential applications in measuring low substrate concentrations in implantable biosensors and producing high current outputs in enzymatic biofuel cells.

We compared the efficacy of Pickering crystals, a continuous phase crystal network, and a combination thereof against sedimentation and dispersed phase coalescence in water-in-oil (W/O) emulsions. Using 20 wt % water-in-canola oil emulsions as our model, glycerol monostearate (GMS) permitted Pickering-type stabilization, whereas simultaneous usage of hydrogenated canola oil (HCO) and glycerol monooleate (GMO) primarily led to network-stabilized emulsions. A minimum of 4 wt % GMS or 10 wt % HCO was required for long-term sedimentation stability. Although there were no significant differences between the two in mean droplet size with time, the free water content of the network-stabilized emulsions was higher than Pickering-stabilized emulsions, suggesting higher instability. Microscopy revealed the presence of crystal shells around the dispersed phase in the GMS-stabilized emulsions, whereas in the HCO-stabilized emulsion, spherulitic growth in the continuous phase and on the droplet surface occurred. The displacement energy (E(disp)) to detach crystals from the oil-water interface was ∼10(4) kT, and was highest for GMS crystals. Thermal cycling to induce dispersed phase coalescence of the emulsions resulted in desorption of both GMS and GMO from the interface, which we ascribed to solute-solvent hydrogen bonding between the emulsifier molecules and the solvent oil, based on IR spectra. Overall, Pickering crystals were more effective than network crystals for emulsion stabilization. However, the thermal stability of all emulsions was hampered by the diffusion of the molten emulsifiers from the interface.

In this study, we describe the effects of incorporating single-walled carbon nanotubes (SWNTs) into redox polymer-enzyme hydrogels. The hydrogels were constructed by combining the enzyme glucose oxidase with a redox polymer (Fc-C(6)-LPEI) in which ferrocene was attached to linear poly(ethylenimine) by a six-carbon spacer. Incorporation of SWNTs into these films changed their morphology and resulted in a significant increase in the enzymatic response at saturating glucose concentrations (3 mA/cm(2)) as compared to films without SWNTs (0.6 mA/cm(2)). Likewise, the sensitivity at 5 mM glucose was significantly increased in the presence of SWNTs (74 μA/cm(2)·mM) as compared to control films (26 μA/cm(2)·mM). We demonstrate that the increase in the electrochemical and enzymatic response of these films depends on the amount of SWNTs incorporated and the method of SWNT incorporation. Furthermore, we report that the presence of SWNTs in thick films allows for more of the ferrocene redox centers to become accessible. The high current densities of the hydrogels should allow for the construction of miniature biosensors and enzymatic biofuel cells.

Electrodes modified with single-walled carbon nanotubes (SWNTs) offer a number of attractive properties for developing novel electrochemical sensors. A common method to immobilize SWNTs onto the electrode surface is by placing a droplet of a SWNT suspension onto the electrode surface and allowing the solvent to evaporate. In order to maximize the properties of individual SWNTs, surfactants are normally present in these suspensions to provide stable and homogeneous SWNT dispersions. In this study we investigated the effect of different surfactants on the electrochemical and enzymatic performance of SWNT modified glassy carbon electrodes (GCEs). Amperometic biosensors for glucose were fabricated by a two-step procedure. In the first step, SWNT films were deposited onto GCEs by solution casting suspensions of SWNTs in water, Triton X-100, Tween 20, sodium cholate or sodium dodecylbenzenesulfonate (NaDDBS). In the second step, hydrogels containing a redox polymer and the enzyme, glucose oxidase (GOX), were deposited and cross-linked onto the SWNT-modified GCE. Three different redox polymers were tested: 3-ferrocenylpropyl-modified LPEI, (Fc-C3-LPEI), 6-ferrocenylhexyl-modified LPEI, (Fc-C6-LPEI), and poly[(vinylpyridine)Os(bipyridyl)2Cl](2+/3+)(PVP-Os). Biosensors constructed with SWNT films from suspensions of Triton X-100 or Tween 20 generally produced the highest electrochemical and enzymatic responses, with Triton X-100 films producing current densities of ~1.7-2.1 mA/cm(2) for the three different redox polymers. In contrast, biosensors constructed with SWNT films from sodium cholate suspensions resulted in significant decreases in the electrochemical and enzymatic response and in some cases showed no enzymatic activity. The results with SWNT films from NaDDBS suspensions were dependent upon the specific redox polymer used, but in general gave reduced enzymatic responses (~0.05-0.4 mA/cm(2)). These results demonstrate the importance of surfactant type in fabricating SWNT-modified electrode films.