Synthesis, quantitative structure–property relationship study of novel fluorescence active 2-pyrazolines and application

A variety of fluorescence-active fluorinated pyrazolines 13–33 was synthesized in good yields through cyclocondensation reaction of propenones 1–9 with aryl hydrazines 10–12. Some of the synthesized compounds provided promising fluorescence properties with quantum yield (Φ) higher than that of quinine sulfate (standard reference). Quantitative structure–property relationship studies were undertaken supporting the exhibited fluorescence properties and estimating the parameters governing properties. Five synthesized fluorescence-active pyrazolines (13, 15, 18, 19 and 23) with variable Φ were selected for treating two types of paper sheets (Fabriano and Bible paper). These investigated fluorescence compounds, especially compounds 19 and 23, provide improvements in strength properties of paper sheets. Based on the observed performance they can be used as markers in security documents.


Introduction
Fluorescence-active compounds are important materials due to their potential application in various fields, such as medicine,

Quantitative structure-property relationship studies
2D-QSPR studies were undertaken using comprehensive descriptors for structural and statistical analysis (CODESSA-Pro) software. Florescence active pyrazolines 13-33 were used as training set for constructing the 2D-QSPR model. Geometry of the training set compounds was initially optimized by AM1 technique [29,30], then exported to CODESSA-Pro for the final geometry optimization (MOPAC software). CODESSA-Pro calculated 821 molecular descriptors (constitutional, topological, geometrical, charge-related, semi-empirical, thermodynamical, molecular-type, atomic-type and bondtype descriptors) for the exported 21 training set fluorescence-active pyrazolines. Different mathematical transformations (including property (quantum yield), 1/property, log(property) and 1/log(property)) of the experimentally observed training set compounds were used for searching for the best QSPR model. The best multi-linear regression (BMLR) technique was used which is a stepwise search for the best nparameter regression equations (where n stands for the number of descriptors used), based on the highest R 2 (squared correlation coefficient), R 2 cvOO (squared cross-validation 'leave one-out, LOO' coefficient), R 2 cvMO (squared cross-validation 'leave many-out, LMO' coefficient), F (Fisher statistical significance criteria) values and s 2 (standard deviation). The QSPR models up to 5-descriptor model describing the properties of the fluorescence-active agents were generated (obeying the thumb rule of 5 : 1, which is the ratio between the data points and the number of QSPR descriptors).
Another validation technique was also considered to examine the robustness of the 2D-QSPR model, where part of the available data (two-thirds of the data) was used for determining a QSPR model and the remaining data points (one-third of the data) was used as a test set (external validation) [31].
(i) All the available data points (21 fluorescence-active pyrazolines 13-33) were arranged in the descending order of Φ s values and separated into three subsets (A, B and C) by selection of every third point from the original dataset in order to obtain fair distribution of the investigated property values for each subset.

UV-visible absorption and fluorescence spectra 3.2.1. UV-visible spectra
The absorption spectra of the synthesized pyrazolines 13-33 were determined in chloroform with constant concentration (4 mg l −1 ). The data of maximum absorption wavelength (λ max ) and molar extinction coefficient (ε max ) are recorded in table 1. All synthesized pyrazolines 13-33 revealed two prominent peaks around 238 and 358 nm, attributed to the π-π* and n-π* transitions, respectively. The effect of introducing different electronic donating (EDG) and withdrawing groups (EWG) in R, R' and R' groups of 1,3,5-triaryl-4,5-dihydro-1H-pyrazoline had a remarkable effect on the differences in both the position and intensity of absorption peaks. The relatively higher absorption intensity is noticed for compound 29. This is probably ascribed to the combination of 4-chlorophenyl and 2,4-dichlorophenyl at 1-and 5-positions of pyrazoline heterocycle. The three chlorine atoms (auxochromes) in addition to the presence fluorine atom at phenyl groups attached to the 1-, 3-and 5-positions of the pyrazoline may explain this observation. It is also noticed that the presence of 2-naphthyl group in compound 18 provided red shifted absorption peak relative to the other synthesized analogues.

Fluorescence spectra
The maximum excitation wavelengths and the emission wavelengths for fluorescence spectra of the synthesized pyrazolines 13-33 were also measured in chloroform with constant concentration (1 × 10 −5 mol l −1 ) (table 1). The spectra of the representative fluorescence compounds, with different quantum yields are illustrated in figure 1. From the observed data (table 1), it has been noticed that the synthesized pyrazolines are excited at 360-370 nm (corresponding to the high wavelength absorption band) affording fluorescence emission in the blue to green regions (emission peak wavelengths 438-471 nm). It has also been noticed that the quantum yield (Φ sample ) value is greatly affected by the substitution type at 1-, 3-and 4-positions of the synthesized pyrazolines. The fluorescence quantum yield of the synthesized pyrazolines 13-33 was compared to that of quinine sulfate (used as reference compound, Φ ref = 0.546), and was calculated using the following equation (table 1): From the experimental data calculated and recorded in table 1, it is evident that the different substituents with EDG and EWG on both R' and R'' groups, as well as their different positions at the 1,3-pyrazoline moiety had a great effect on both the intensity and the emission maxima wavelengths compared to the effect of the substituents on R groups. As can be seen, the presence of EWG and bulky groups in R' such as present in compounds 13, 14, 18, 19, 20, 27 and 32 led to higher emission intensity together with blue shifting of emission spectra than the other synthesized compounds. On the contrary, the presence of EDG on both R' and R'' (e.g. compounds 25, 16, 28 and 33) led to remarkable decreasing of the emission intensity without remarkable effect on the position of emission spectra.
From the foregoing data, it could be concluded that all the synthesized fluorescence-active fluorinated pyrazolines 13-33 exhibited vivid fluorescence properties with different quantum yields, ranging from 0.155 to 0.858. Five compounds with quantum yields of 0.858, 0.844, 0.819, 0.802, 0.553 (13, 15, 18  these active fluorinated pyrazolines for production of functional paper sheets (strength and fluorescence performance).

Modelling
Quantitative structure-property relationships have been a major part of many important scientific studies attracting attention of researchers not only for designing and developing agents of better behavioural manifestation but also to validate the experimental observed properties. This is due to the capability of QSPR to represent mathematical relationships between the property of interest and descriptors (physico-chemical parameters) based on the molecular structure. This study deals with QSPR study of fluorescence properties for the synthesized pyrazolines to explore the controlling parameters governing properties. CODESSA-Pro software was used for conducting the present QSPR study employing the 21 synthesized fluorescence-active fluorinated pyrazolines 13-33 which exhibit variable properties (quantum yields). The BMLR-QSPR model obtained for the present study is statistically significant (table 2 and

Molecular descriptors
The first descriptor controlling the BMLR-QSPR (t = 7.403) is LUMO + 1 energy, which is a semiempirical descriptor. Lowest unoccupied molecular orbital (LUMO) energy is determined by where φ LUMO stands for lowest unoccupied molecular orbital andF for Fock operator. (t = −5.609) is bond-type descriptor. Coulombic force is determined by the following equation: where q i and q j represent the point charges on atoms i and j, respectively, with r ij being the distance between them. D denotes the dielectric constant of the medium. Positively charged part of partial charged surface area (MOPAC PC) is a charge-related descriptor. The partial positively charged surface area (PPSA) is determined by where S A stands for the positively charged solvent-accessible atomic surface area (electronic supplementary material, table S1, shows the descriptor value for each tested compound).   Another validation technique was also considered to examine the robustness of the 2D-QSPR model, where part of the available data (two-thirds of the data) were used for determining a QSPR model and the remaining data points (one-third of the data) were used as a test set (external validation). The observed 3 descriptor BMLR-QSPR models due to this technique (N = 14) are statistically significant (R 2 = 0.  blank blank + solvent S13 S15 S18 S19 S23

Validation
blank blank + solvent S13 S15 S18 S19 S23 blank blank + solvent S13 S15 S18 S19 S23 blank blank + solvent S13 S15 S18 S19 S23 tensile index (Nm g -1 ) Fabriano paper Bible paper Strength properties of paper sheets treated with representative fluorescent compounds. S13, S15, S18, S19 and S23 for Fabriano (SF) and Bible (SB) treated by compounds 13, 15, 18, 19 and 23. (c, d) Strength properties of paper sheets treated with representative fluorescent compounds. S13, S15, S18, S19 and S23 for Fabriano (SF) and Bible (SB) treated by compounds 13, 15, 18  These treated paper sheets were denoted with SF13, SF15, SF18, SF19 and SF23 for Fabriano papers; while for Bible paper were denoted as SB13, SB15, SB18, SB19, SB23, respectively. Table 4 shows that the treated paper sheets are excited at the range of 279-380 nm that corresponds to the absorption band in the solutions. It can be noticed that the decrease in the intensity of emission spectra is in the order of SF4 > SF5 > SF3 > SF1 > SF2 for the Fabriano papers; while for Bible sheets this order becomes SB3 > SB4 > SB1 > SB2 > SB5. On comparison with the emission bands of fluorescent compound solutions, it is noticed that the presence of the 4-chlorophenyl group at pyrazoline moiety for SF1, SF3, SB1 and SB3 leads to a red shift of the emission spectra; however, a blue shift of emission band is clear in the case of paper treated with compounds including 4-methylphenyl, as the case of SF2 and SB3. For SF4, SF5, SB4 and SB5, which were treated with fluorescent compounds including 2-naphthyl, no marked change in emission band is observed, in comparison to the data obtained in solution. Table 4 also shows that the trend of emission intensities data for the treated Bible paper sheets is different from that for the Fabriano sheets, which may be ascribed to the change in cellulose substrates, basis weight and degree of sizing of the two types of paper sheets.
With regard to strength properties of treated paper sheets, figure 3a,b shows that all investigated compounds provide the improvement in quality numbers (Q z , indication to all strength properties). The improvement in the case of Bible paper sheets is greater than in the case of Fabriano paper sheets, especially for samples S19 and S23. The explanation of these observed data is probably ascribed to the basis weight of paper sheets. Bible paper is lower in grammage (45 g m −2 ), which promotes the diffusion of fluorescent solutions through the fibres and interact with the free hydroxyl groups of paper substrates, i.e. it leads to reformation of bonds but with higher strength hydrogen bonds. This view is emphasized from decreasing the strength of Bible paper sheets treated by chloroform solution only. The treating of Bible paper by chloroform leads to decrease in the energy of hydrogen bonds between fibres. It means that more free OH groups in cellulose substrate are possibly created.
The active ultraviolet area for the treated paper sheets was screened via a video spectral comparator (VSC ® 6000) as shown in figure 4. This figure shows that the treated paper sheets, denoted as SF13, SB13, SF18, SB18, SF19, SB19, SF23 and SB23, exhibit a green fluorescence light; while those denoted as SF15 and SB15 exhibit a blue fluorescence light. The Fabriano paper sheets show a greater intensity than the treated Bible paper sheets. This trend is related to the foregoing reason of lower grammage Bible paper sheets than Fabriano paper. The higher grammage paper sheets of Fabriano type (approx. 80 g m −2 ) may lead to a concentration of the fluorescent compounds on the surface of paper. As can be seen, the florescence intensity followed the order of quantum yields of compound solutions.

Conclusion
Synthesis of novel pyrazolines with fluorescence properties and application to paper sheets is the main objective of this study. 2D-QSPR study was undertaken for validating the fluorescence behaviour of the synthesized heterocycles. Some of the synthesized heterocycles (13, 15, 19 and 23) possessed promising fluorescence properties with quantum yield value reaching 0.86. The quantum yield values were estimated relative to the standard reference used (quinine sulfate). The greatest improvement in strength properties was observed in Bible paper sheets upon treatment with the fluorescence-active compounds 19 and 23; while the greatest improvement in the fluorescence intensities was observed in treated Fabriano paper sheets. Both promising fluorescence and strength properties of treated paper sheets persuade us to examine, in forthcoming work, their safety behaviour towards resistance to counterfeiting and forgery, as well as ageing.
Ethics. This work is delivered from the PhD thesis (in preparation) of our student working at Forgery Research Department, Medico-legal Sector, and Ministry of Justice. Because this external registration in Al-Azhar University,