PubMed2001
It was found that combining capillary electrophoresis (CE) and electrospray ionization mass spectrometry (ESI-MS) overlays two controlled current techniques to form a three-electrode system (CE inlet, CE outlet/ES emitter, and MS inlet electrodes) in which the CE outlet electrode and the ES emitter electrode were shared between the CE and the ESI-MS circuits. Depending on the polarities and magnitudes of the voltages at the CE inlet, CE outlet/ES emitter, and MS inlet electrodes, the nature of the two redox reactions at the shared electrode was the same or different (both reduction, both oxidation, or one oxidation and the other reduction). Several redox buffers were introduced for controlling electrochemical reactions at the shared electrode. By reacting at this electrode, redox buffers were able to maintain electrode potentials below the onset of water electrolysis, thereby eliminating gas bubble formation and/or pH drift. The volume of the gas generated due to water electrolysis was used to quantitate water oxidation or reduction at this electrode. Two types of redox buffers were used. A reactive electrode with an oxidation potential below that of water was used as the electrode under anodic conditions. Also, a reactive compound with a redox potential below that of water was added to the CE and/or ESI running buffer. When the shared electrode was the anode of both CE and ESI-MS circuits, the use of iron or etched and sanded stainless steel (ss) wire, instead of platinum wire, suppressed bubble formation at the shared electrode. Under these conditions, corrosion of the Fe wire and formation of Fe2+ replaced oxidation of water, eliminating O2 gas bubble and H+ formation. When mixtures of peptides were analyzed, iron adducts of peptides were observed. For a fresh wire, however, the intensities of adduct ions were less than 3% of the protonated molecules. After a few days of operation, the intensities of the adduct ions increased to approximately 50%, due to rust formation on the Fe wire. On-column rinsing with a 40% solution of citric acid rejuvenated the Fe wire and reduced the adduct peak intensities to less than 3%. Unmodified ss wire did not quench bubble formation, which was attributed to its passivated surface. When Fe, ss, and Pt wires were used as the shared electrode under forward polarity CE and positive ESI mode, where the shared electrode acted as a cathode with respect to CE inlet and as an anode with respect to MS inlet, reduction of water at the cathodic end of the electrode and, in the case of ss and Pt wires, oxidation of water at the anodic end of the shared electrode produced a significant amount of bubbles. Under these conditions, however, a buffer containing 50 mM p-benzoquinone completely suppressed both cathodic reduction and anodic oxidation of water for CE currents up to 4 microA. Reduction of p-benzoquinone at the cathodic end of the shared electrode to hydroquinone, and oxidation of this hydroquinone at the anodic end of the electrode, replaced reduction and oxidation of water, eliminating bubble formation. A 0.1% acetic acid solution saturated with I2 was also found to suppress bubble formation at the cathode for CE currents up to 3 microA; however, strong iodine adduct ions were observed under CE/ESI-MS when a mixture of peptides was analyzed. The application of iron as an in-capillary electrode for the analysis of a peptide mixture and a protein digest demonstrated a high separation efficiency similar to when hydroquinone was used as a redox buffer.