Synthesis of Dansyl Cyclen and Preliminary Study of Its Fluorescent Properties

The synthesis of a dansyl cyclen-based compound as a potential chemical sensor has been carried out. The initial study of its fluorescent properties has also been conducted. This study aims to synthesize a cyclen-based compound comprising three identical pendant arms and another different arm carrying a dansyl fluorophore. Producing these heterogenous pendant arms, a-three pendant arm cyclen 9 was reacted with dansyl aziridine 10. The synthesis products were characterized using 1H NMR, 13C NMR, IR, and elemental analysis. In addition, a Fluorescent Spectrophotometer has been used to assess the fluorescent intensity changes of the synthetic ligand in a range of pH 2–13 and when it was titrated with some metal ions. Based on the results of characterization with 13C NMR for compound 2 and additional characterization with IR and elemental analysis for its hydrochloric form 11, it is wisely said that the proposed compound has been successfully synthesized, giving 66% yield as viscous brown oil 2. Moreover, the fluorescent property showed that the higher the pH employed, the higher the fluorescent intensity observed. Meanwhile, the addition of some cationic metals revealed that cadmium (II) gave the highest increase in the fluorescent intensities among other cationic metals.


Introduction
The development of chemical sensors has been quite remarkable during the last few decades. Many studies have been conducted and recently reported related to this field, for instance, Wahyuni et al., who investigated uric acid sensor [1]. Nonetheless, the field of chemical sensor is very broad. Tetraazamacrocycles have been the subject of many research groups to study and design to create a class of molecular receptors [2]. Successful development of molecular receptor complexes, such as 1 by Smith [3], has enabled the evolution in the synthesis of a new generation of analogous cyclen-based receptor complexes. Evidence from 1 that four identical pendant arms N-bonded constructs it to the cyclen framework. Therefore, the methodology responsible for its synthesis is quite simple.
Meanwhile, synthesizing a new generation of four pendant-armed receptors incorporating three identical pendant arms and a different fourth pendant arm is quite challenging because a significant synthetic effort dealing with N-protection and deprotection steps is required. Bradbury [4] has reported that an anthracene moiety has been included to replace one of the pendant arms of the Smith [3] compound. In addition, Hodyl [5] has also reported a similar compound as Bradbury [4], but instead of using anthracene as the substituent, he attached a silica moiety. Inspired by Bradbury [4] and Hodyl [5], the current study also addresses this issue.
The mainframe of the current study was to replace one of the pendant arms of 1 with a functional group bearing a fluorescent agent, in this case, a dansyl moiety, to produce a molecular ligand 2, as depicted in Figure 1. Dansyl group has been commonly utilized as fluorescence signals in chemical sensors [6], Gu [7], Omer [8], and Kim [9]. Nonetheless, to the best of our knowledge, the synthesis of dansyl cyclen-based receptors and the study of their fluorescent properties have not been reported. Due to the presence of the dansyl group, it was expected that the synthesized compound would be the potential for being a molecular fluorescent sensor. This report focuses on the synthesis of the free receptor ligand 2; secondly, the assessment of the fluorescence emission intensity of the free ligand against a range of pH and the fluorescent emission of the free ligand on its own and with the addition of metals.
This study proposed the schematic diagram to achieve the synthetic compound target, as shown in Figure 2. The synthesis would be initiated by using a cyclen 3 as the primary precursor in the reaction, after which protection and deprotection reactions would take place. Moreover, finally, a fluorescence ligand 2 would be attained by a reaction between a three-identical substituted arm of cyclen 9 with dansyl aziridine 10.

Methodology
All reactions were conducted under a nitrogen atmosphere unless specified otherwise. Solvents were always purified prior to utilization by standard methods. Cyclen (98%) was purchased from Strem Chemicals, U.S.A. The stationary phase of column chromatography utilized Merck Silica gel 60 (230-400 mesh). The progress of a reaction or separation during the chromatographic process was monitored by thin-layer chromatography (TLC) No. 5554 (60 PF254) from Merck, and N-Dansyl aziridine was purchased from Sigma-Aldrich. CD3OD at 49.00 ppm; and D2O at δ 67.19 ppm by the addition of 1,4-dioxane. In order to examine the functional groups of synthetic compounds, a HORIBA FT-720 FT-IR Spectrometer was used. The measurement of fluorescence emission spectra was conducted with a Varian Cary Eclipse Fluorescence Spectrophotometer using quartz cells with 1.0 cm path length, at a wavelength ranging from 400-650 nm at 0.15 nm intervals, with a scan rate of 40 nm/min. A blank solution was always run first at each measurement. (3C, -CHO), 53.10-40.00 (8C, cyclen -CH2). This data was in good agreement with the reference of Bradbury [4] and Hodyl [5]. Benzyl chloroformate (710 mg, 4.15 mmol) was added to a solution of 4 (682 mg, 2.7 mmol, pH 9) dissolved in deionized water. The resultant mixture was then stirred for 1 h at room temperature after adjusting the pH from 4 to 10 using saturated Na2CO3 solution. A second aliquot of benzyl chloroformate (710 mg, 4.15 mmol) was added using an identical procedure as the first aliquot after the pH was again adjusted from 6 to 10 with saturated Na2CO3 solution. Then, the third aliquot of benzyl chloroformate (710 mg, 4.15 mmol) was added. The resultant mixture was stirred overnight at room temperature under a nitrogen atmosphere. The mixture was extracted with dichloromethane (5 x 20 ml), and the combined organic layers were washed with saturated NaHCO3 (1 x 10 ml), dried over MgSO4, and concentrated in vacuo to obtain crude 5 as a yellow oil (583 mg) or 56% yield. It was used in the triformyl deprotection step without further purification [10].

Preparation of the mono-protected cyclen
This data was in good agreement with the reference of Bradbury [4] and Hodyl [5]. Purification was conducted by suspending the crude 5 in the cold (4℃) diethyl ether for 7 days. The white solid was obtained and filtered under nitrogen to produce 5 as a hygroscopic white powder in a 95% yield. 13  . This data was in good agreement with Bradbury [4] and Hodyl [5].

pH Titration of Free Ligand
Solution of protonated ligand 34 (at 5 x 10 -6 M) in 30% aqueous acetonitrile, at a constant ionic strength of I = 0.1 M, NEt4ClO4, was titrated with 0.1 M NEt4OH with the range spectra was 400-650 nm. The range of pH applied was from 2 to 13, and it was indicated by using a combination glass electrode that was calibrated against an aqueous buffer solution for pH 2-12. While pH 13 was indicated using litmus paper.

Metal Titration of Free Ligand
The effect of the addition of metal ions on fluorescence emission of the ligand was studied by conducting a series of titrations using several cationic metals. Solution of the ligand was at 10 -5 M buffered at

Results and Discussion
All the synthetic precursor compounds ran according to procedures obtained from previous reports and gave results as expected. They were all in good agreement with the references. The results of the syntheses of 1-[(5-(Dimethylamino)-naphthalene-1sulfonyl-amino)ethyl]-4,7,10-[(2S)-2-hydroxy-3phenoxypropyl)-1,4,7,10-tetraazacyclododecane and its protonated form were also successfully achieved. Due to a huge amount of hydrogen in the compound, using 1 H NMR to identify the resulting product's protons was not useful. The appearance of proton NMR was messy and noninformative. That is why the 13 C NMR was utilized instead of 1 H NMR to characterize the product.

Synthesis of Free Ligand
This reaction is based on the ring-opening of aziridine, which is similar in principle to epoxide ringopening except that a nitrogen atom has replaced the oxygen atom. The synthetic approach employed in this reaction was adapted from Yang [12], as shown in Figure  12.

Figure 12. Synthesis of the fluorescent ligands
A mixture of dansyl aziridine 10 and compound 9 was refluxed in acetonitrile at 80℃ for five days. Since the dansyl fluorophore is light-sensitive, the reaction apparatus was covered with aluminum foil to exclude light. Progress of the reaction was monitored by thinlayer chromatography to detect the disappearance of dansyl aziridine. The product was purified by column chromatography on silica gel using acetone/ triethylamine (95:5) eluent to obtain 2 in 66% yield. Besides the 13 C NMR spectrum of the synthetic product presented in the subheading 2.4, the ring-opening was confirmed by 13 C NMR spectroscopy due to the loss of resonances at about 28 ppm, originating from the aziridine carbon atoms was also crucial evidence that the compound target 2 was successfully achieved. In order to produce a sample suitable for microanalysis, which occurs as a viscous oil, 2 was converted to its pentahydrochloride salt form, 11, by treatment with an excess of 32% hydrochloric acid [13]. It was found that C, 54.53; H, 6.87; N, 7.56% in which the proposed compound consisting of pentahydrogenchloride, C49H71Cl5N6O8S, requires C, 54.42; H, 6.62; N, 7.77% [14]. This indicates that the proposed compound was achieved. The absorption peak at 1318 cm -1 represents the infrared frequency of the asymmetric SO2 stretching frequency (υasSO2). This is in good agreement with Wang [15] and Xue [16].

Effect of pH on the fluorescent properties of the receptor-ligand
Several titration experiments were conducted to investigate the influence of the pH on the fluorescence properties of the receptor-ligand. The emission spectra of ligand 2 were recorded as a function of pH, as depicted in Figure 13. The pH values recorded between 2 and 13 were indicated by a pH meter employing a combination glass electrode calibrated against an aqueous buffer. For pH values above 12, indicator paper was used. In general, this involved the addition of aliquots of a solution tetraethylammonium hydroxide (NEt4OH) to a dilute solution (10 -5  The fluorescence emission curve against pH of ligand 2 provided an apparent increase with the increase of pH, as shown in Figure 13. This fact agrees with a report by Aoki et al., [17] stating that cyclen moiety will deprotonate as the pH increases causing the emission intensity also increases. As shown from Figures 13 (A) and (B), there are groups of emission spectra having close to each other. For example, at pH 3-8 the emission spectrum is quite close to another, so that of pH 10-12, while in comparison, the fluorescence emission spectrum at pH 9 is relatively isolated. This titration curve indicates that pH around the pKa value is sensitive to the acid and base environment. It is also clearly shown from Figure 13 (B) that the quantum yield of fluorescence tended to increase while the pH of ligand solution increased.

The effect of metal complexation on fluorescence
In this study, the potentially eight-coordinating metal ions Cd(II), Zn(II), Hg(II), Pb(II), La(III), Y(III), and Ca(II) were investigated for their influence on the fluorescence properties of the ligand. The previous work showed that when the metal ions under consideration bind to ligands similar to 2, they bound to all four nitrogen atoms in the macrocycle residue. If the pendant dansylamine did not coordinate with the metal atom, it could exist in the protonated or deprotonated form, depending on the prevailing pH. The previous section showed obviously from fluorescence pH curves of titration that minor pH change around the pKa value of the pendant dansylamine led to significant shifts in the fluorescence emission intensity of the ligand. Therefore, it was necessary to investigate the fluorescence of metal complexes of 2 to buffer the solution at working pH to avoid bias recorded fluorescence emission intensities due to metal hydrolysis. For this reason, it was decided to buffer the solutions at pH 10.5 using CAPS (N-cyclohexyl-3aminopropanesulfonic acid). CAPS was preferred as it is known to be resistant to metal ion complexation and would then control the pH without interfering with the investigation [18,19,20]. The effect of metals on free ligand 2 can be seen in Figure 14.
From Figure 14(A), it is clear that the most significant increase in the intensity of fluorescence of the solution of ligand occurred when Cd(II) solution was added, followed by Zn(II). The former gained up to 150% compared to the free ligand, while the latter contributed 130%, as shown in Figures 14 (A) and (B). The presence of Y(III), Ca(II), Pb(II), and La(II), however, relatively did not change the intensity of the fluorescence. The fluorescence intensity quenched quite significantly while Hg(II) was added. Nonetheless, all the metalligand complexation followed the 1:1 ratio, as the titration curve indicated in Figure 14 (A). It was evident that the maximum point was achieved when the ligand and cationic metals ratio was identical at 1:1 [21,22].

Conclusion
The synthesis of the dansyl cyclen-based receptor has been successfully achieved. The characterization results confirmed this synthetic product using 13 C NMR, IR spectrophotometer, and microanalysis. The physical properties, in particular, the fluorescent emission trend of the receptor-ligand against the pH tended to increase as the pH increased and it experienced a hypsochromic shift. From metal ions' titration, it revealed that Cd(II) ion had a significant impact on increasing the fluorescence of the ligand to almost two-fold, followed by Zn(II). Meanwhile, Hg(II) ion caused a significant decrease in the fluorescence intensity of the ligand. The synthetic receptor-ligand likely has a great potency to develop it as a chemical sensor.