Chemiluminescence esters

Peroxyoxalate chemiluminescence is the most efficient nonenzymatic chemiluminescent reaction known. Quantum efficiencies as high as 22—27% have been reported for oxalate esters prepared from 2,4,6-trichlorophenol, 2,4-dinitrophenol, and 3-trif1uoromethy1-4-nitropheno1 (6,76,77) with the duorescers mbrene [517-51-1] (78,79) or 5,12-bis(phenylethynyl)naphthacene [18826-29-4] (79). For most reactions, however, a quantum efficiency of 4% or less is more common with many in the range of lO " to 10 ein/mol (80). The inefficiency in the chemiexcitation process undoubtedly arises from the transfer of energy of the activated peroxyoxalate to the duorescer. The inefficiency in the CIEEL sequence derives from multiple side reactions available to the reactive intermediates in competition with the excited state producing back-electron transfer process. 

Alternatively, weak acids and certain salts have been found to extend the lifetimes of inherently rapid reactions which occur with highly reactive esters, such as bis(2,4-dinitrophenyl) oxalate (95). A chemiluminescent demonstration based on the oxaUc ester reaction has been described (96) and the reaction has been developed iato a practical lighting system. 

A number of other chemiluminescent reactions appear to be related to peroxyoxalate chemiluminescence although thek mechanistic details may vary. For example, various chlotinated esters and ethers react with and a fluorescer to emit light (98—101). Other examples have been given... 

The theoretical limit of light capacity has been estimated for an ideal reaction that provides yellow light with a photopic factor of 0.85 in a quantum yield of one at 5 Af concentration as 173,000 (Im-h)/L, equivalent to the light output of a 40-W bulb burning continuously for two weeks (237). The most efficient formulation available, based on oxaUc ester chemiluminescence, produces about 0.5% of that limit, with a light capacity of 880 (Im-h)/L (237). 

ImmunO lSS iy. Chemiluminescence compounds (eg, acridinium esters and sulfonamides, isoluminol), luciferases (eg, firefly, marine bacterial, Benilla and Varela luciferase), photoproteins (eg, aequorin, Benilld), and components of bioluminescence reactions have been tested as replacements for radioactive labels in both competitive and sandwich-type immunoassays. Acridinium ester labels are used extensively in routine clinical immunoassay analysis designed to detect a wide range of hormones, cancer markers, specific antibodies, specific proteins, and therapeutic dmgs. An acridinium ester label produces a flash of light when it reacts with an alkaline solution of hydrogen peroxide. The detection limit for the label is 0.5 amol. 

In the chemiluminescence-based HPLC detection system, illustrated schematically in Figure 6, the oxalate ester and hydrogen peroxide are introduced to the eluent stream at postcolumn mixer Mj, which then flows through a conventional fluorescence detector with the exciting lamp turned off or a specially built chemiluminescence detector. The two reagents are combined at mixer Mj, rather than being premixed, to prevent the slow hydrolytic reactions of the oxalate ester. 

Scheme 2. Mechanism for hydrogen peroxide-oxalate ester generation of chemiluminescence. Ar = aryl B = a base.

Table VI. RP-HPLC Chemiluminescence Sensitivity in Previous Studies with Oxalate Esters...

To understand how these parameters affected the efficiency of the chemiluminescent reaction, we examined the mechanism originally proposed by Rauhut (26). As shown in Scheme 2, hydrogen peroxide reacts with an oxalate ester, such as 2,4,6-trichlorophenyl oxalate (TCPO), in a two-step process to form a reactive intermediate for which Rauhut suggested structure 1, the 1,2-dioxetanedione. The dioxetanedione then interacts with an acceptor (ACC) to produce two molecules of COj and the excited state of the acceptor. The last stage of the sequence is fluorescence emission from the acceptor. 

Figure 13, indicates that the first mole of phenol is released in <30 s, the same elapsed time for the chemiluminescence to reach a maximum intensity. In fact, the measured rate constant r, for the rise in the chemiluminescence emission, is identical to the rate of the first phenol s release from the oxalate ester. Furthermore, the slower rate of release of the second phenol ligand has a rate constant that is identical to the chemiluminescence decay rate f. Thus, the model allows a quantitative analysis of the reaction mechanism, heretofore not available to us. We intend to continue this avenue of investigation in order to optimize the chemiluminescence efficiencies under HPLC conditions and to delineate further the mechanism for peroxy-oxalate chemiluminescence. 

We have also investigated other oxalate esters as a potential means to improve the efficiency. The most commonly used oxalates are the 2,4,6-trichlorophenyl (TCPO) and 2,4-dinitrophenyl (DNPO) oxalates. Both have severe drawbacks namely, their low solubility in aqueous and mixed aqueous solvents and quenching of the acceptor fluorescence. To achieve better solubility and avoid the quenching features of the esters and their phenolic products, we turned to difluorophenyl oxalate (DFPO) derivatives 5 and 6 (Figure 14). Both the 2,4- and the 2,6-difluoro esters were readily synthesized and were shown to be active precursors to DPA chemiluminescence. In fact, the overall efficiency of the 2,6-difluorophenyl oxalate 5 is higher than for TCPO in the chemical excitation of DPA under the conditions outlined earlier. Several other symmetrical and unsymmet-rical esters were also synthesized, but all were less efficient than either TCPO or 2,6-DFPO (Figure 14). 

Figure 14. Oxalate esters synthesized (A) and oxalate dependence on the chemiluminescence quantum efficiency (B).

Though we and others (27-29) have demonstrated the utility and the improved sensitivity of the peroxyoxalate chemiluminescence method for analyte detection in RP-HPLC separations for appropriate substrates, a substantial area for Improvement and refinement of the technique remains. We have shown that the reactions of hydrogen peroxide and oxalate esters yield a very complex array of reactive intermediates, some of which activate the fluorophor to its fluorescent state. The mechanism for the ester reaction as well as the process for conversion of the chemical potential energy into electronic (excited state) energy remain to be detailed. Finally, the refinement of the technique for routine application of this sensitive method, including the optimization of the effi-ciencies for each of the contributing factors, is currently a major effort in the Center for Bioanalytical Research. 

The chemiluminescence reaction of esters of oxalic acid can proceed within a wider pH range than for luminol. However, the most efficient oxalate derivatives are only soluble in organic solvents such as ethyl acetate, acetonitrile, dioxane or dimethoxyethane and dissolution problems of these solvents in aqueous media are encountered. This can limit the use of this chemiluminescence reaction for a direct coupling to an H202-generating enzymatic reaction. 

Me Capra in particular proposed n> that the chemiluminescence reactions of a large number of organic compounds had this concerted dioxetane decomposition step as key reaction in the production of electronically excited products, namely acridinium salts 25,26,27) indolylperoxides 28>, activated oxalic esters 29>, diphenyl carbene 30>, tetrakis-dimethylamino-ethylene 31 32>, lucigenin 33>, and substituted imidazoles 23>. 

Numerous autoxidation reactions of aliphatic and araliphatic hydrocarbons, ketones, and esters have been found to be accompanied by chemiluminescence (for reviews see D, p. 19 14>) generally of low intensity and quantum yield. This weak chemiluminescence can be measured by means of modern equipment, especially when fluorescers are used to transform the electronic excitation energy of the triplet carbonyl compounds formed as primary reaction products. It is therefore possible to use it for analytical purposes 35>, e.g. to measure the efficiency of inhibitors as well as initiators in autoxidation of polymer hydrocarbons 14), and in mechanistic studies of radical chain reactions. 

As mentioned in Section II. C., the concerted bond cleavage of 1.2-dioxetane derivatives has been proposed to be of general importance in respect of the excitation step of a large number of chemiluminescence reactions. The first experimental results concerning simple dioxetanes were obtained by M. M. Rauhut and coworkers in their work on activated oxalic ester chemiluminescence 24>. From experimental data on the reaction of e.g. bis (2.4-dinitrophenyl)oxalate with hydrogen peroxide in the presence of rubrene, they concluded that 1.2-dioxetanedione... 

The leaving group of the oxalic ester has a strong effect on the efficiency of the peroxyoxalate chemiluminescent system. The electron-attracting power of the substituents on the phenyl rings of the substituted diphenyl oxalates is important to the overall efficiency of the chemiluminescent reactions. Steric effects... 

Several N-methyl-9-acridinecarboxylic acid derivatives (e.g., 10-methyl-9-acridinecarboxylic chloride and esters derived therefrom [39]) are chemiluminescent in alkaline aqueous solutions (but not in aprotic solvents). The emission is similar to that seen in the CL of lucigenin and the ultimate product of the reaction is N-methylacridone, leading to the conclusion that the lowest excited singlet state of N-methylacridone is the emitting species [40], In the case of the N-methyl-9-acridinecarboxylates the critical intermediate is believed to be either a linear peroxide [41, 42] or a dioxetanone [43, 44], Reduced acridines (acridanes) such as N-methyl-9-bis (alkoxy) methylacridan [45] also emit N-methylacridone-like CL when oxidized in alkaline, aqueous solutions. Presumably an early step in the oxidation process aromatizes the acridan ring. 

Figure 6 Chemiluminescent reaction of an acridinium ester and hydrogen peroxide in the presence of a base.

Following the disclosure of the outstanding catalytic ability of imidazole compared to other bases, the catalysis of the POCL reaction by imidazole was studied in more detail [163], and it was concluded that the POCL reaction mechanism included the concurrent catalysis by two imidazole molecules, by what was described as general-base and nucleophilic pathways, respectively. The mechanism for this was suggested to be a base catalysis of imidazole catalysis by imidazole itself as previously reported for imidazole-catalyzed reaction of esters [164, 165], Despite this, it was not until the introduction [151] of 1,1 -oxalyldiimidazole (ODI) as a chemiluminescence reagent, and the postulation of its intermediate appearance in the imidazole-catalysed POCL reaction, that the... 

---