Peroxide value reversion

PTFE polytetrafluoroethylene PUFA polyunsaturated fatty acid PV peroxide value PVDF polyvinylidene difluoride PVP polyvinylpyrrolidone PVPP polyvinylpolypyrolidone RAS retronasal aroma stimulator RDA recommended dietary allowance RF radio frequency RFI relative fluorescence intensity RI retention index RNU relative nitrogen utilization ROESY rotational nuclear Overhauser enhancement spectroscopy RP-HPLC reversed-phase HPLC RPER relative protein efficiency ratio RS resistant starch RT retention time RVP relative vapor pressure S sieman (unit of conductance)... 

Fig. 45 Reversed-phase HPLC of autoxidized trilinolenin (peroxide value = 236.4 meq/kg). Nova-Pak C18 cartridge column (Waters, Milford, MA) (3.9 X 150 mm, 60 A, 4 yam), mobile phase acetonitrile/ dichloromethane/methanol (80 10 10). Ultraviolet (UV) detector (235 nm) and evaporative light-scattering detector (ELSD). Primary oxidation products, double peak at 3.6 min secondary oxidation products elute before primary oxidation products.

Fig. 46 Reversed-phase HPLC of autoxidized triolein (peroxide value = 149.7 meq/kg). See Fig. 45 for abbreviations and chromatographic conditions. Primary oxidation products, peak at 18.6 min, secondary oxidation products elute before primary oxidation products.

The development of a characteristic, objectionable, beany, grassy, and hay-like flavor in soybean oil, commonly known as reversion flavor, is a classic problem of the food industry. Soybean oil tends to develop this objectionable flavor when its peroxide value is still as low as a few meq/kg, whereas other vegetable oils, such as cottonseed, com, and sunflower, do not (15, 51). Smouse and Chang (52) identified 71 compounds in the volatiles of a typical reverted-but-not-rancid soybean oil. They reported that 2-pentylfuran formed from the autoxidation of linoleic acid, which is the major fatty acid of soybean oil, and contributes significantly to the beany and grassy flavor of soybean oil. Other compounds identified in the reverted soybean oil also have fatty acids as their precursors. For example, the green bean flavor is caused by c/i-3-hexenal, which is formed by the autoxidation of linolenic acid that usually constitutes 2-11% in soybean oil. Linoleic acid oxidized to l-octen-3-ol, which is characterized by its mushroom-like flavor (53). 

Soybean oil and other linolenate-containing oils such as canola oil are notorious for their flavor deterioration noted at unusually low levels of oxidation, sometimes at peroxide values below 1. This flavor defect is known as reversion , an obsolete term derived from the flavor characteristic of cmde soybean oil described as beany or grassy . Oils susceptible to this flavor defect contain oxidatively derived dimers and oligomers produced during deodorization by thermal decomposition of hydroperoxides present initially... 

Sensory evaluation is a specialized discipline, using trained panels to measure and analyse the characteristics of food lipids evoked by the senses of taste, smell, sight and mouth feel. Sensory analyses are those most closely associated with the quality of food lipids, but their usefiilness is limited because they are costly and require a well-trained taste and odor panel and the proper facilities. However, sensory analyses provide sometimes ausefiil approach to identifying flavor or odor defects in the processing of food lipids that caimot be detected by other more objective chemical or instrumental analyses. For example, certain flavor defects characterized as grassy or fishy in linolenate-containing oils such as soybean and low-erucic rapeseed (canola) oils (Chapter 1) occur at such low levels of oxidation that they can only be detected by sensory analyses. The old term flavor reversion for soybean oil is based on the characteristic of this oil undergoing flavor deterioration at unusually low levels of oxidation that cannot be measured by peroxide value determination. Oils derived from fish... 

The reported and claimed activities of different antioxidants and formulations are difficult to evaluate because of the different testing conditions used. For example, when tested at 45°C, ascorbic acid was less effective in stabilizing soybean oil than ascorbyl palmitate, which was in turn more effective than BHA and BHT, but not as active as PG and TBHQ. TBHQ was the most active of the antioxidants tested (Table 9.4). However, when tested at 98°C under conditions of the active oxygen method (AOM), the relative antioxidant activities of ascorbic acid and ascorbyl palmitate were reversed and ascorbic acid became more active than PG and TBHQ. These results must be interpreted with caution. The peroxide value used to follow oxidation is not reliable when oils are heated at 98°C because a large fraction of the hydroperoxides is decomposed at this temperature (Chapter 7). The peroxide value of 70 used as an end-point is also questionable since flavor deterioration occurs in soybean oil at peroxide values below 10. 

Soyabean oil tends to develop an undesirable flavour and odour known as reversion when peroxide value is still as low as a few meq/kg. This reversion flavour is characterized as beany and grassy, and is often found in the oil after light... 

Han et al. [48] also studied the effects of various antioxidants such as ascorbic acid, a-tocopherol, and rosemary extract on the oxidation of fish oil and soybean oil using phosphatidylcholine reverse micelles. Peroxide values of the oils indicated that only ascorbic acid (0.02%) solubilized in the reverse micelles was an effective antioxidant in both fish oil and soybean oil. Also, as in the other studies, a combination of ascorbic acid and a-tocopherol was shown to act synergistically in the fish oil. 

The value obtained (8 x 10 1 mol-1 s-1 in di-t-butyl peroxide at 40°) is in fair agreement with that (1.5 x 10 ) reported by Janzen and Evans (1973) both figures seem surprisingly large in view of the rapid reversal of the process (, = 0.14 s"1 at 40°). 

If we make the assumption that the reverse of reaction 15.5 is diffusion-controlled and assume that the activation enthalpy for the acyl radicals recombination is 8 kJ mol-1, the enthalpy of reaction 15.5 will be equal to (121 - 8) = 113 kJ mol-1. This conclusion helps us derive other useful data. Assuming that the thermal correction to 298.15 K is small and that the solvation enthalpies of the peroxide and the acyl radicals approximately cancel, we can accept that the enthalpy of reaction 15.5 in the gas phase is equal to 113 kJ mol-1 with an estimated uncertainty of, say, 15 kJ mol-1. Therefore, as the standard enthalpy of formation of gaseous PhC(0)00(0)CPh is available (-271.7 5.2 kJ mol-1 [59]), we can derive the standard enthalpy of formation of the acyl radical Af//°[PhC(0)0, g] -79 8 kJ mol-1. This value can finally be used, together with the standard enthalpy of formation of benzoic acid in the gas phase (-294.0 2.2 kJ mol-1 [59]), to obtain the O-H bond dissociation enthalpy in PhC(0)0H DH° [PhC(0)0-H] = 433 8 kJ mol-1. 

Because the accuracy of the data for three of the diacyl peroxides is in question, we will attempt to derive enthalpies of formation for them from the reverse of equations 15 and 16. The enthalpy of reaction 15 for dibenzoyl peroxide, using enthalpy of formation values of unquestioned accurac)f, is —400.8 kJmor. This is the same as the ca —398 kJmol for the hquid non-aromatic diacyl peroxides discussed above. Using the solid phase enthalpy of reaction for dibenzoyl peroxide and the appropriate carboxylic acid enthalpies of formation, the calculated enthalpies of formation of bis(o-toluyl) peroxide and bis(p-toluyl) peroxide are —432.2 and —457.6 kJ moU, respectively. From the foregoing analysis, it would seem that the measured enthalpy of formation is accurate for the bis(p-toluyl) peroxide but is not for its isomer. The analysis for dicinnamoyl peroxide is complicated by there being two enthalpies of formation for frawi-cinnamic acid that differ by ca 12 kJmoU. One is from our archival source (—336.9 12 kJmoU ) and the other is a newer measurement (—325.3 kJmol ). The calculated enthalpies of formation of dicinnamoyl peroxide are thus —273.0 and —249.8 kJmoU. Both of these results are ca 80-100 kJmol less negative than the reported enthalpy of formation. 

A wide range of transition metal complexes will carry dioxygen reversibly.1 In all cases, electron transfer takes place from the metal centre to the antibonding ir-orbitals of the bound dioxygen. In some cases the bound dioxygen appears to be present as superoxide and in others as peroxide [as shown by (02) values around 1100 and 800 cm 1 respectively]. 

Aging of bleached pulps in presence of high humidity results in higher values of peroxide formed than aging at low humidity. The total amount of peroxide increases linearly with the time of aging.. . . Moisture was found to promote not only peroxide formation but, to some extent, also, brightness reversion, particularly in absence of air. Aldehyde groups formed by hydrolysis were found to be involved in peroxide formation. 

Fig. 2. Thermodynamics of the reduction of oxygen species Potential diagram at pH 7. Figures are apparent standard redox potentials versus NHE, at pH 7. Figures used for the alkyl peroxide species at the top of the diagram are approximations derived from computations by Koppenol [19]. An upper value for the ROO /ROOH redox couple would be 1.1 V, i.e. the value derived from the trichloromethyl peroxyl radical, the strongest oxidant known among peroxyl radicals. See the discussion in the text. Note (see a) A aq -0.16 V (relative to 1 M 02) is a better reference than —0.33 V for the reversible...

---