With CO present in excess, ONOO reacts to form an adduct in solution and in the solid state, most likely ONOOCO. In solution, the adduct appears within 2 ms and absorbs at 300 with an extinction coefficient, which is either 50% or 100% (preferred) of that of ONOO, 1.70 × 10 M cm, and at 685 nm with an extinction coefficient of 85 M cm. When solid [(CH)N][ONOO] is treated with CO, these two maxima are red-shifted by 30-50 nm. The equilibrium constant for adduct formation in solution is (4.5 ± 0.5) × 10 M. The adduct reacts further with another CO at a rate of (2.6 ± 0.8) × 10 M s and produces 2 CO and NO. Thermochemical calculations show that ΟΝΟΟCO is a strong two-electron oxidizing agent, E°(ONOOCO, H/NO, HCO) = +1.28 V at pH 7 and an even stronger one-electron oxidizing agent E°'(ONOOCO, H/NO, HCO) = +1.51 V at pH 7. The extent of homolysis, that is formation of NO and CO, is small, slightly less than 1% relative to ONOO at the physiological concentration of CO of 1.3 mM in plasma. Thus, ONOOCO is more relevant than CO under in vivo conditions.

In biology, the reaction of ONOO − with CO 2 is the main sink for ONOO − . This reaction yields CO 3•− , NO 2 • , NO 3 − , and CO 2 . There is a long-standing debate with respect to the yield of the radicals relative to ONOO − . The reaction of ONOO − with CO 2 results at first in ONOOCO 2− . According to one hypothesis, ONOOCO 2 − is extremely short-lived and devolves into a solvent cage that contains CO 3•− and NO 2 • . Of these solvent cages, approximately two/thirds result in NO 3 − and CO 2 , and approximately one/third release CO 3and NO 2• that oxidize the substrate. According to our hypothesis, ONOOCO 2 − is formed much faster, is relatively long-lived, and may also be an oxidant; the limited yield is the result of ONOOCO 2− being scavenged by a second CO 2 under conditions of a high CO 2 concentration. We disagree with the first hypothesis for three reasons: First, it is based on an estimated K for the reaction of ONOO − with CO 2 to form ONOOCO 2 − of ∼1 M −1 , while experiments yield a value of 4.5 × 10 3 M −1 . Second, we argue that the solvent cage as proposed is physically not realistic. Given the less than diffusion-controlled rate constant of CO 3•− with NO 2• , all radicals would escape from the solvent cage. Third, the reported ∼33% radical is not supported by an experiment where mass balance was established. We propose here a hybrid mechanism. After formation of ONOOCO 2 − , it undergoes homolysis to yield CO 3 •− with NO 2• , or, depending on [CO 2 ], it is scavenged by a second CO 2 ; CO 3 •− oxidizes ONOO − , if present. These reactions allow us to successfully simulate the reaction of ONOO − with CO 2 over a wide range of ONOO − /CO 2 ratios. At lower ratios, fewer radicals are formed, while at higher ratios, radical yields between 30% and 40% are predicted. The differences in radical yields reported may thus be traced to the experimental ONOO − /CO 2 ratios. Given a physiological [CO 2 ] of 1.3 mM, the yield of CO 3•− and NO 2• is 19%, and lower if ONOOCO 2 − has a significant reactivity of its own.