Chemiluminescence of 5-(azo -para -phenylene- N -aza-15-crown-5)-phthalhydrazide

5-(Azo -para -phenylene- N -aza-15-crown-5)-phthalhydrazide presents weak fluorescence and (in the presence of aqueous phosphate buffer at pH = 10.1 with H 2 O 2 , with or without Fe 3+ or Cu 2+ ions) a significant chemiluminescence (CL) at λ CL = 425 nm. Linear correlations exist between the hydrogen peroxide concentration and the intensity of the CL. For similarly intense chemiluminiscence, in the presence of cupric ions, the H 2 O 2 concentration range is ten times lower than in their absence. This observation allows the determination of low concentrations of H 2 O 2 and of Cu 2+ ions


Introduction
The chemiluminescence (CL) of organic compounds 1 has both theoretical and practical interest.][10][11] Based on its CL, luminol is being used for the bioanalytical determination of enzymatically-formed oxidizing species such as the superoxide anion and the hydroxyl radical.[12][13] It is also possible to determine cholesterol via the H 2 O 2 concentration formed by enzymatic oxidation.14 Electro-chemiluminescence determinations have recently been reported using luminol or its chemical modifications such as deposition on gold 15,16 or silver nanoparticles, 17 on graphite-methacrylate 18 or glass-carbon composite electrodes 19 on platinum electrodes 20 or as a film formed by copolymerization with aniline. 21 Drivatives of luminol with azo-calixarenes 22 have also been reported to present weak FL and significant CL.Also phthalhydrazide-ionophores such as styrylphthalhydrazides with crown ethers 23 and aza-crowned isoluminol 24 were obtained.
5-(Azo-para-phenylene-N-aza-15-crown-5)-phthalhydrazide 1 was designed and synthesized for its pH-dependent chromogenic (indicator) properties 25 due to its azo group, and for its chromoionophoric behavior in the presence of lithium and sodium cations 26 due to its crown ether moiety.The tunable CL is caused by the luminol moiety, and the azo bridge between it and the crown ether moiety allows electronic communication between these two moieties.In the present paper we report a study of the CL when 1 is oxidized in alkaline medium by hydrogen peroxide, with or without Fe 3+ or Cu 2+ ions.

Chromogenic, solubility, and fluorescence properties of coumpound 1
In Scheme 1 one can see the various chemical species formed from 1 in acidic media (only two lactam tautomers are shown, 1a and 1b although lactim forms are also posible) and in alkaline media; the resonance forms of the monoanion 1c and of the dianion 1d influence chromogenic properties. 25Acidification of 1 is accompanied by a bathochromic shift due probably to the paraquinone-diimidic tautomer 1b, whereas in basic medium a hypsochromic effect takes place, with phthalhydrazide resonance structures of 1c and 1d.
The amphoteric nature of 1 determines its solubility.Therefore neutral 1 is soluble in organic solvents, whereas its alkaline salts 1c and 1d are soluble in water as in the case of luminol, 7,9 and the ratio between mono-and dianion (Scheme 1) is strongly dependent on the pH.In sodium phosphate buffer at pH = 10.1, under photoexcitation with λ ex = 420 nm, the FL intensity of 1 was 30 times lower than that of luminol.This effect may be attributable to electron-attracting effect of the azo group, similarly to the congeneric azo-calixarene derivatives. 22emiluminescence First, we performed qualitative investigations for finding optimal reaction conditions, then experiments for confirming the putative reaction mechanism, and finally we collected quantitative data.These results are presented below sequentially.

Qualitative experiments and reaction mechanism
The pH of the buffer (Na 2 HPO 4 + NaOH) was selected to be 10.1, by analogy with literature values: pH = 10.15 for luminol, 10,23 and pH = 10.5 for the phthalhydrazide-azo-calixarene derivative. 22Experiments were carried out at room temperature (25ºC).Higher pH values than 10.1 cause precipitation when using Fe 3+ and Cu 2+ salts.At an optimal concentration of 10.3 µM with or without Fe 3+ or Cu 2+ ions, compound 1 with the phosphate buffer presented CL on treatment with H 2 O 2 , after 5 s at λ CL = 425 nm, like luminol [7][8][9] but differing slightly from the azo-calixarene congener which had λ CL = 420 nm. 22ollowing Merényi and coworkers, 13,27,28 it is generally accepted [29][30][31] that the CL of luminol starts with electron transfer processes between redox couples. 32,33][10] (Scheme 2).Monitoring the intensity of CL in time, the following results were obtained: (i) when only H 2 O 2 was employed, the I CL value decreased for 50 s reaching a plateau (Figure 1A); even after 5 minutes I CL did not decrease to a half; (ii) with a high ratio [H 2 O 2 ]/[M n+ ] (M n+ = Fe 3+ or Cu 2+ , the I CL value decreased for 50 s and then starts to increase ( Figure 1B, 1C); even after 5 minutes I CL did not decrease to a half; (iii) inversely, with a low ratio [H 2 O 2 ]/[M n+ ] (M n+ = Fe 3+ or Cu 2+ ), I CL started higher (Figure 1D) but decreased for 50 s, and had reached half of its value after 15 s.After monitoring at λ CL = 425 nm, it was not posible to evidence by TLC the reaction products because less than 5% of the compound 1 had reacted, as determined from the UV-Vis spectra.This situation is similar to that observed for the CL of luminol. 9hen transition metal cations were also present, the processes responsible for oxidizing 1 are represented by equations ( 1) and (2) (where: M n+ = Fe 3+ or Cu 2+ and M (n-1)+ = Fe 2+ or Cu 1+ ), which afford hydroxyl radicals via reaction (2).(2) For checking that Cu 2+ behaves like Fe 3+ in the presence of H 2 O 2 , we used EPR spectroscopy for spin-trapping with 5,5-dimethyl-1-pyrroline N-oxide [34][35][36] (DMPO, 4) any shortlived free radicals such as HO• (Scheme 4).We did obtain EPR quartets with 1:2:2:1 intensities only in the presence of Fe 3+ or Cu 2+ cations, and we ascribe these EPR spectra (Fig. 2) to the persistent free radical 5, formed from 4 and a hydroxyl free radical.

Quantitative experiments Determination of the hydrogen peroxide concentration with or without Cu 2+
We investigated the possibility of determining the concentration of H 2 O 2 in the absence or in the presence of Cu 2+ ions using compound 1, and we obtained the following results (Figure 3): (i) In the absence of Cu 2+ ions (Figure 3A), there is a linear correlation 37 (equation 3) for hydrogen peroxide concentrations ranging from 12 µg•mL -1 to 90 µg•mL -1 ; (3) where: I CL = intensity of CL; [X] = concentration of H 2 O 2 , Fe 3+ and Cu 2+ .(ii) In the presence of excess Cu 2+ ions (Figure 3B), a similar linear correlation allows the determination of H 2 O 2 in 20-times lower concentration range between 0.5 to 5 µg•mL -1 .
In the above correlations, slightly lower coefficients R would result if a = 0 in equation ( 3) for the cases presented in Figures 3A, 3B, and 4A, where the linear correlation is close to the origin of the Cartesian coordinates.However, in the case presented in Figure 4B, the origin is appreciably farther, so that a cannot be ignored.
We tested the effect of alkali metal cations, and found that in the presence of Li + salts λ max = 501 nm and the CL has λ max,CL at 436 nm, whereas in the presence of Na + salts λ max = 492 nm and the CL has λ max,CL at 437 nm.

Conclusions
In alkaline solution (pH=10.1),compound 1 presents at 25°C a weak FL.In the same conditions, in the presence of H 2 O 2 , with or without Fe 3+ or Cu 2+ cations, compound 1 displays after 5 s a significant CL at 425 nm.After about 50 s secondary phenomena occur due to various reactive oxygen species.By spin trapping using DMPO and EPR spectrometry, it was possible to prove the formation of HO• in such conditions.The linear correlation between I CL of 1 and oxidizing species (H 2 O 2 , Fe 3+ or Cu 2+ ) encourages us to propose using 1 (with chromogenic properties depending on pH, and chromoionophoric properties depending on alkali metal cations) for analytical and bioanalytical assays (Figure 5).

Solutions
The phosphate buffer was prepared from bidistilled water and Na 2 HPO 4

Methods
CL measurements were performed with TD 20/20 Turner Design, USA (the points on the plot were obtained by integrating the light signal over periods of 5 s for five measurements, then average values were calculated, obtaining a maximum 10% relative scattering of the results from the mean value).EPR spectra were recorded at room temperature after 8 minutes with compound 1 (1.03•10 -4 M) and H 2 O 2 (60 µg•mL -1 ) in phosphate buffer (pH=10.1) in the presence of DMPO (10 -1 M) with a JEOL FA 100 spectrometer with 100 kHz modulation frequency, at 0.998 mW microwave power, with 120 s sweep time, 0.7 G modulation amplitude, time constant 0.1 s The DMPO solution was freshly prepared; the simulated EPR spectra were performed with Winsim-program standard.Simulations yielded the hyperfine splitting constants with Fe 3+ (44•10 -3 µg•mL -1 ) and with Cu 2+ (2.3•10 -3 µg•mL -1 ).The experimental spectra had sharper lines for Fe 3+ than with Cu 2+ (Figure 2). ) and with a given volume of phosphate buffer so that on rapidly adding 0.5 to 5 µg•mL -1 of the 0.03 % H 2 O 2 solution one obtains a final volume of 1000 µL.After stirring for 5 s, I CL was measured; the average of 5 measurements was obtained, as shown in Figure 3B.Determination of Cu 2+ and Fe 3+ concentrations.An aqueous solution of 1 (100 µL) was admixed at 25 °C with 2 to 8 µL of Fe(NO 3 ) 3 •9H 2 O solution or with 8 to 40 µL of CuCl 2 •2H 2 O solution; then a corresponding volume of the phosphate buffer was added so as to reach a total volume of 1000 µL after the rapid ddition of 20 µL (60 µg•ml -1 ) 0.3 % H 2 O 2 solution.After stirring for 5 s, I CL was measured; the average of 5 measurements was obtained, as shown in Fig. 4.

Scheme 2 .
Scheme 2. The probable process for CL of compound 1.

5 Scheme 4 .Figure 2 .
Scheme 4. Reaction of diamagnetic DMPO 4 with HO• radical yielding the paramagnetic nitroxide 5 with practically equal hyperfine splitting constants a H and a N .

Figure 5 .
Figure 5. CL properties of the compound 1.
Determination of H 2 O 2 concentrations without Cu 2+ .An aqueous solution of 1 (100 µL) was admixed at 25 °C with a given volume of phosphate buffer (pH=10.1)and then rapidly with 4 to 20 µL of the 0.3 % H 2 O 2 solution yielding a final volume of 1000 µL.After stirring for 5 s, I CL was measured; the average of 5 measurements was obtained, as shown in Figure 3A.(ii) Determination of H 2 O 2 concentrations in the presence of Cu 2+ .An aqueous solution of 1 (100 µL) was admixed at 25 °C with 20 µL of CuCl 2 •2H 2 O solution (0.596 µg•mL -1 •12H 2 O (17.956 g in 250 mL); into this solution under continuous stirring and pH measurement (with pH Meter 315 i/SET, WTW) a 0.2 M solution of NaOH was added till the pH reached 10.1.Buffer solutions were used immediately after preparation.In all experimental procedures fresh aqueous solutions were used.Compound 1 was dissolved at room temperature (25°C) in phosphate buffer (pH=10.1).The 0.3% (g/v) H 2 O 2 solution was obtained by diluting 30%, H 2 O 2 with bidistilled water at 25°C, and determining the concentration by titration with KMnO 4 and H 2 SO 4 .Solutions of transition metal salts (0.08% Fe(NO 3 ) 3 •9H 2 O and 0.008% CuCl 2 •2H 2 O) were obtained with bidistilled water at 25°C.