New Water-Soluble Poly(propylene imine) Dendrimer Modified with 4-Sulfo-1,8-naphthalimide Units: Sensing Properties and Logic Gates Mimicking

A new water-soluble poly(propylene imine) dendrimer (PPI) modified with 4-sulfo-1,8-naphthalimid units (SNID) and its related structure monomer analog (SNIM) has been prepared by a simple synthesis. The aqueous solution of the monomer exhibited aggregation-induced emission (AIE) at 395 nm, while the dendrimer emitted at 470 nm due to an excimer formation beside the AIE at 395 nm. Fluorescence emission of the aqueous solution of either SNIM or SNID was significantly affected by traces of different miscible organic solvents, and the limits of detection were found to be less than 0.05% (v/v). Moreover, SNID exhibited the function to execute molecular size-based logic gates where it mimics XNOR and INHIBIT logic gates using water and ethanol as inputs and the AIE/excimer emissions as outputs. Hence, the concomitant execution of both XNOR and INHIBIT enables SNID to mimic digital comparators.


Introduction
Dendrimers are three-dimensional star-shaped supramolecular architectures having various functional groups in the structure. They have recently been attracting the attention of scientists as an alternative to linear and branched polymers due to their unique structural features, including large surface area and the flexibility to incorporate different compounds into their periphery or interior parts [1][2][3]. Over the last two decades, numerous structural scaffolds for dendrimers have been reported, ranging from pure organic molecular frameworks to organometallic and biomaterials [4][5][6]. Exploration of dendrimers applications in supramolecular chemistry is still ongoing. Recently, many reports have presented their potential in drug delivery [7], tissue engineering [8], bio-imaging [9], catalysis [10], cancer therapy [11], and a variety of other applications. Luminescent dendrimers are indispensable components in high-technology industries, particularly optoelectronics, light-harvesting antennae in solar cells, sensors for detecting pollutants in the environment, biology, and medicine [12][13][14].

Design and Synthesis of the Probe
The synthesis of 4-sulfo-1,8-naphthalimide-modified PPA dendrimer SNID and its related monomer SNIM is presented in Scheme 1. Their chemical structures were confirmed by UV-Vis absorption, fluorescent, FT-IR, and NMR spectra (Figures S1-S6). The π-π stacking of 1,8-naphthalimide units is favoured in water, and hence, aggregation-induced emission AIE is possible. The function of the dendrimer scaffold is to stick close to the 4-sulfo-1,8-naphthalimde moieties, thus enabling the aggregation-induced excimer formation in water.

Photophysical Characteristics
The influence of solvent polarity on the absorption and emission spectra of SNIM and SNID has been investigated, and the respective data have been summarized in Table  1. The absorption spectra of monomer and dendrimer have an absorption band in the range of 300-370 nm corresponding to the 4-sulfo-1,8-naphthalimide chromophore group, Scheme 1. Synthesis of SNIM and SNID.

Photophysical Characteristics
The influence of solvent polarity on the absorption and emission spectra of SNIM and SNID has been investigated, and the respective data have been summarized in Table 1. The absorption spectra of monomer and dendrimer have an absorption band in the range of 300-370 nm corresponding to the 4-sulfo-1,8-naphthalimide chromophore group, Figure 1. While the absorption band was characterized by a well-developed vibrational fine structure in most solvents, the structure is almost blurred in the hydroxylic solvents due to hydrogen bonding with the solvent molecules and π-π stacking that restricts the vibrational transitions. The position of the absorption band is not affected significantly by the solvent polarity, suggesting that these compounds in the ground state are not sensitive to the polarity of the environment. The solvent polarity has an impact only on the vibrational transitions. On the other hand, the absorption spectra of the dendrimer SNID in different solvents are similar to the ones of the monomer SNIM, except the molar extinction coefficients at the absorption maxima, which are approximately four times higher than those of the monomer SNIM, which indicates the full substitution of the primary amino groups in the dendrimer periphery by 4-sulfo-1,8-naphthalimide units [70]. Regarding fluorescence emission, after excitation at 340 nm, the monomer gives a strong fluorescence emission centred at 392 nm only in water, Figure 2A. It is attributed to the monomer molecules aggregation that is induced by the π-π stacking. This stacking restricts the nonradiative vibrational de-excitations processes of the excited molecules. Moreover, the monomer SNIM gives a weak emission in ethanol and DCM due to the vague formation of aggregates in these solvents. The fluorescence emission observed in THF, despite the well-developed vibrational fine structure of the absorption band, refers to the aggregation favoured in the excited state rather than in the ground state. Strikingly, the behaviour of the dendrimer in water is different from that of the monomer, where besides the emission at 395 nm, which is weaker, a strong emission centred at 475 nm is observed ( Figure 2B). This is confirmed by the photograph of the CNID and SNIM compounds dissolved in water, DMF, and ethanol and irradiated with monochromatic UV light at 366 nm. The figure shows the blue-green fluorescence emission of CNID in an aqueous solution, while SNIM emits blue fluorescence ( Figure 2C). The former emission, as mentioned above, is caused by the excimer formation of 4-sulfo-1,8-naphthalimide units, while that of the latter is due to the aggregation of dendrimer molecules [71]. Regarding fluorescence emission, after excitation at 340 nm, the monomer gives a strong fluorescence emission centred at 392 nm only in water, Figure 2A. It is attributed to the monomer molecules aggregation that is induced by the π-π stacking. This stacking restricts the nonradiative vibrational de-excitations processes of the excited molecules. Moreover, the monomer SNIM gives a weak emission in ethanol and DCM due to the vague formation of aggregates in these solvents. The fluorescence emission observed in THF, despite the well-developed vibrational fine structure of the absorption band, refers to the aggregation favoured in the excited state rather than in the ground state. Strikingly, the behaviour of the dendrimer in water is different from that of the monomer, where besides the emission at 395 nm, which is weaker, a strong emission centred at 475 nm is observed ( Figure 2B). This is confirmed by the photograph of the CNID and SNIM compounds dissolved in water, DMF, and ethanol and irradiated with monochromatic UV light at 366 nm. The figure shows the blue-green fluorescence emission of CNID in an aqueous solution, while SNIM emits blue fluorescence ( Figure 2C). The former emission, as mentioned above, is caused by the excimer formation of 4-sulfo-1,8-naphthalimide units, while that of the latter is due to the aggregation of dendrimer molecules [71].
light at 366 nm. The figure shows the blue-green fluorescence emission of CNID in an aqueous solution, while SNIM emits blue fluorescence ( Figure 2C). The former emission, as mentioned above, is caused by the excimer formation of 4-sulfo-1,8-naphthalimide units, while that of the latter is due to the aggregation of dendrimer molecules [71].  The discriminated fluorescence emission of the monomer and its dendrimer in water encouraged us to investigate the applicability of these compounds as probes for quantitative measurements of the purity of water contaminated with another miscible organic solvent. We used ethanol, DMF, and dioxane as representatives for polar protic and aprotic and nonpolar solvents, respectively. Moreover, we investigated the influence of water traces in the solvents on the emission response of the SNIM and SNID.

Solvatochromism of SNIM
It has been found that ethanol has no effect on the emission of the aqueous solution of SNIM till 60% (v/v) of ethanol. Higher amounts of ethanol (>60%) led to emission quenching at 395 nm due to the dissociation of aggregates by ethanol molecules. On the other hand, fluorescence emission at 395 nm of ethanol solution of SNIM has enhanced by adding water, Figure 3. The limit of detection (LOD) for water presence in ethanol was found to be 0.09% by volume. LOD was calculated using LOD = 3σ/b [38], where b is the slop and σ is the standard deviation. The increase in the emission by adding water is ascribed to the aggregation of SNIM molecules induced by π-π stacking of nonpolar 1,8naphthalimide moieties in the presence of water. The low LOD of SNIM towards water The discriminated fluorescence emission of the monomer and its dendrimer in water encouraged us to investigate the applicability of these compounds as probes for quantitative measurements of the purity of water contaminated with another miscible organic solvent. We used ethanol, DMF, and dioxane as representatives for polar protic and aprotic and nonpolar solvents, respectively. Moreover, we investigated the influence of water traces in the solvents on the emission response of the SNIM and SNID.

Solvatochromism of SNIM
It has been found that ethanol has no effect on the emission of the aqueous solution of SNIM till 60% (v/v) of ethanol. Higher amounts of ethanol (>60%) led to emission quenching at 395 nm due to the dissociation of aggregates by ethanol molecules. On the other hand, fluorescence emission at 395 nm of ethanol solution of SNIM has enhanced Sensors 2023, 23, 5268 6 of 14 by adding water, Figure 3. The limit of detection (LOD) for water presence in ethanol was found to be 0.09% by volume. LOD was calculated using LOD = 3σ/b [38], where b is the slop and σ is the standard deviation. The increase in the emission by adding water is ascribed to the aggregation of SNIM molecules induced by π-π stacking of nonpolar 1,8-naphthalimide moieties in the presence of water. The low LOD of SNIM towards water presence in ethanol indicates that it can be used as a low-cost reagent for the detection of traces of water in alcohol. The required volume of water to reach saturation of the fluorescence response of SNIM in the ethanol solution was found to be ≈24% (v/v) ( Figure S7). and nonpolar solvents, respectively. Moreover, we investigated the influence of water traces in the solvents on the emission response of the SNIM and SNID.

Solvatochromism of SNIM
It has been found that ethanol has no effect on the emission of the aqueous solution of SNIM till 60% (v/v) of ethanol. Higher amounts of ethanol (>60%) led to emission quenching at 395 nm due to the dissociation of aggregates by ethanol molecules. On the other hand, fluorescence emission at 395 nm of ethanol solution of SNIM has enhanced by adding water, Figure 3. The limit of detection (LOD) for water presence in ethanol was found to be 0.09% by volume. LOD was calculated using LOD = 3σ/b [38], where b is the slop and σ is the standard deviation. The increase in the emission by adding water is ascribed to the aggregation of SNIM molecules induced by π-π stacking of nonpolar 1,8naphthalimide moieties in the presence of water. The low LOD of SNIM towards water presence in ethanol indicates that it can be used as a low-cost reagent for the detection of traces of water in alcohol. The required volume of water to reach saturation of the fluorescence response of SNIM in the ethanol solution was found to be ≈24% (v/v) ( Figure S7).   Moreover, SNIM exhibited the ability to investigate the contamination of water in DMF, as a representative for polar aprotic solvents, by its fluorescence emission, Figure 4. Similar to ethanol, the presence of DMF decreased the emission of SNIM in water due to the dissociation of aggregated molecules by DMF solvation. The limit of detecting DMF contamination was found to be 0.08%, refereeing to the applicability of SNIM to detect traces of DMF in water. The saturation of emission response was reached after the addition of 5% of DMF to the water solution, after which the decrease in fluorescence with increasing DMF content up to 10 % is negligible ( Figure 4B).   Moreover, SNIM exhibited the ability to investigate the contamination of water in DMF, as a representative for polar aprotic solvents, by its fluorescence emission, Figure 4. Similar to ethanol, the presence of DMF decreased the emission of SNIM in water due to the dissociation of aggregated molecules by DMF solvation. The limit of detecting DMF contamination was found to be 0.08%, refereeing to the applicability of SNIM to detect traces of DMF in water. The saturation of emission response was reached after the addition of 5% of DMF to the water solution, after which the decrease in fluorescence with increasing DMF content up to 10% is negligible ( Figure 4B).
Moreover, the effect of dioxane, as a representative of nonpolar solvents, on the emission of a SNIM solution in water has also been investigated ( Figure 5). In this case, the emission is quenched by the presence of dioxane traces due to the dissociation of π-π stacking between 1,8-naphthalimide moieties. The LOD and dioxane volume required to reach saturation was found to be 0.05% and 10%, respectively. Moreover, the influence of water presence on the emission of SNIM solution in dioxane has been investigated. Contrarily, the presence of water traces enhanced the fluorescence emission. The limit of detection of water in dioxane was found to be 0.14%. Hence, SNIM has a dual sensitive sensory applicability for investigating the purity of both water and dioxane in the presence of the other as a contaminant. In other words, SNIM is able to detect the presence of dioxane traces in a water sample and water traces in a dioxane sample.
its emission at 395 nm. Influence of water content on (C) the emission spectrum of SNIM solution in ethanol and (D) its emission at 395 nm. c = 10 −5 M excitation at 340 nm.
Moreover, SNIM exhibited the ability to investigate the contamination of water in DMF, as a representative for polar aprotic solvents, by its fluorescence emission, Figure 4. Similar to ethanol, the presence of DMF decreased the emission of SNIM in water due to the dissociation of aggregated molecules by DMF solvation. The limit of detecting DMF contamination was found to be 0.08%, refereeing to the applicability of SNIM to detect traces of DMF in water. The saturation of emission response was reached after the addition of 5% of DMF to the water solution, after which the decrease in fluorescence with increasing DMF content up to 10 % is negligible ( Figure 4B).   Moreover, the effect of dioxane, as a representative of nonpolar solvents, on the emission of a SNIM solution in water has also been investigated ( Figure 5). In this case, the emission is quenched by the presence of dioxane traces due to the dissociation of π-π stacking between 1,8-naphthalimide moieties. The LOD and dioxane volume required to reach saturation was found to be 0.05% and 10%, respectively. Moreover, the influence of water presence on the emission of SNIM solution in dioxane has been investigated. Contrarily, the presence of water traces enhanced the fluorescence emission. The limit of detection of water in dioxane was found to be 0.14%. Hence, SNIM has a dual sensitive sensory applicability for investigating the purity of both water and dioxane in the presence of the other as a contaminant. In other words, SNIM is able to detect the presence of dioxane traces in a water sample and water traces in a dioxane sample.

Solvatochromism of Dendrimer SNID
The effect of water traces on the emission of dendrimer solution was examined in ethanol solution. As shown in Figure 6, water leads to emissions enhancement at both 395 nm and 470 nm, and the limits of detection were found to be 0.5% and 1%, respectively. The fluorescence enhancement at 395 nm was observed till 50% water fraction; after that, the emission quenched by further water addition, Figure 7, due to the higher rate of excimer formation and to the fact that more 1,8-naphthalimides unites become included in the excimer formation. In concomitance, the emission at 470 nm increased slowly till 50 % water fraction, then further addition of water increased the rate. Behaviour of the dendrimer in the presence of both water and ethanol solutions as inputs and the emission at 395 nm (λex. = 340 nm) as output and using the initial case of 50% water fraction mimics XNOR logic gate, Figure 7C, where at the initial state (water coded as 0 and ethanol as 0),

Solvatochromism of Dendrimer SNID
The effect of water traces on the emission of dendrimer solution was examined in ethanol solution. As shown in Figure 6, water leads to emissions enhancement at both 395 nm and 470 nm, and the limits of detection were found to be 0.5% and 1%, respectively. The fluorescence enhancement at 395 nm was observed till 50% water fraction; after that, the emission quenched by further water addition, Figure 7, due to the higher rate of excimer formation and to the fact that more 1,8-naphthalimides unites become included in the excimer formation. In concomitance, the emission at 470 nm increased slowly till 50 % water fraction, then further addition of water increased the rate. Behaviour of the dendrimer in the presence of both water and ethanol solutions as inputs and the emission at 395 nm (λ ex. = 340 nm) as output and using the initial case of 50% water fraction mimics XNOR logic gate, Figure 7C, where at the initial state (water coded as 0 and ethanol as 0), the output is high (coded as 1). Addition of ethanol till water fraction = 20% (ethanol coded as 1 and water coded as 0) gets the emission at 395 nm low (coded as 0). Moreover, the addition of water till it reaches a water fraction of 80 % (ethanol coded as 0 and water coded as 1) gets the emission low and coded as 0. Finally, the addition of both ethanol and water in equal amounts (both coded as 1) retains the initial state (emission gets high and coded as 1). On the other hand, using the emission at 470 nm as output and the emission threshold shown in Figure 7B, SNID mimics INHIBIT logic gate where the emission can be considered high (coded as 1) only in the case of adding water alone and otherwise the emission is low (coded as 0). Moreover, a combination of XNOR and INHIBIT logic gates works as a digital comparator, Figure 7D. Furthermore, the applicability of the dendrimer SNID for detecting DMF contamination in water has been studied, Figure 8. The addition of DMF traces to SNID solution in water was associated with quenching the emissions at 395 nm and 470. The limits of detection were found to be 0.09% and 0.2% using emissions at 395 nm and 470 nm, respectively. The quenching of the emissions by DMF contamination is linear in the range of 0-1% of a DMF fraction. On the other hand, the emission spectrum of SNID solution in DMF was affected only by large volumes of water, Figure 9, due to the good solvation of DMF to dendrimer molecules.  Furthermore, the applicability of the dendrimer SNID for detecting DMF contamination in water has been studied, Figure 8. The addition of DMF traces to SNID solution in water was associated with quenching the emissions at 395 nm and 470. The limits of detection were found to be 0.09% and 0.2% using emissions at 395 nm and 470 nm, respectively. The quenching of the emissions by DMF contamination is linear in the range of 0-1% of a DMF fraction. On the other hand, the emission spectrum of SNID solution in DMF was affected only by large volumes of water, Figure 9, due to the good solvation of DMF to dendrimer molecules.    Moreover, the applicability of SNID for detecting dioxane contamination in water samples has been investigated. The presence of dioxane traces quenched both the emissions at 395 and 470 nm linearly in the range of 0-2%. The LOD values were found to be 0.3% and 0.7% using emissions at 395 nm and 470 nm, respectively, Figure 10. On the other hand, the addition of water to SNID solution in dioxane is associated with enhancing the emissions at 395 nm and 470 nm, like the case of adding water to SNID solution in ethanol, Figure 11. Hence, SNID can act as a digital comparator using water and dioxane as inputs and the emissions at 395 nm and 470 nm as outputs.    Moreover, the applicability of SNID for detecting dioxane contamination in water samples has been investigated. The presence of dioxane traces quenched both the emissions at 395 and 470 nm linearly in the range of 0-2%. The LOD values were found to be 0.3% and 0.7% using emissions at 395 nm and 470 nm, respectively, Figure 10. On the other hand, the addition of water to SNID solution in dioxane is associated with enhancing the emissions at 395 nm and 470 nm, like the case of adding water to SNID solution in ethanol, Figure 11. Hence, SNID can act as a digital comparator using water and dioxane as inputs and the emissions at 395 nm and 470 nm as outputs.  Moreover, the applicability of SNID for detecting dioxane contamination in water samples has been investigated. The presence of dioxane traces quenched both the emissions at 395 and 470 nm linearly in the range of 0-2%. The LOD values were found to be 0.3% and 0.7% using emissions at 395 nm and 470 nm, respectively, Figure 10. On the other hand, the addition of water to SNID solution in dioxane is associated with enhancing the emissions at 395 nm and 470 nm, like the case of adding water to SNID solution in ethanol, Figure 11. Hence, SNID can act as a digital comparator using water and dioxane as inputs and the emissions at 395 nm and 470 nm as outputs.

Conclusions
This work presents the synthesis of a new water-soluble poly(propylene amine) dendrimer from the first generation, modified with 4-sulfo-1,8-naphthalimide SNID and its monomer analog SNIM for detecting water contamination by different organic solvents. Both the monomer and dendrimer aggregate in the aqueous solution because of the π-π stacking of 4-sulfo-1,8-naphthalimide moieties that allows the formation of excimers between the excited and non-excited 4-sulfo-1,8-naphthalimide fragments of dendrimer molecules. Moreover, the incorporation of 4-sulfo-1,8-naphthalimide units into the dendrimer scaffold improves their tolerance towards strong bases. Furthermore, the dependence of the emissions, caused by the aggregation and excimer formations, on water presence enables these molecules to detect the presence of traces of various organic solvents in water and vice versa. It has been shown that the dendrimer SNID mimics both XNOR and INHIBIT logic gates which work in combination to execute the function of the digital comparator.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1; Supplementary data associated with this article can be found in theonline version. Figure S1. 1 H-NMR of the monomer SNIM. Figure S2. 13 C-NMR of the monomer SNIM. Figure S3. FTIR spectrum of the monomer SNIM. Figure S4. 1 H-NMR spectrum of the dendrimer SNID. Figure S5. 13 C-NMR spectrum of the dendrimer SNID. Figure S6. FTIR spectrum of the dendrimer SNID. Figure S7

Conclusions
This work presents the synthesis of a new water-soluble poly(propylene amine) dendrimer from the first generation, modified with 4-sulfo-1,8-naphthalimide SNID and its monomer analog SNIM for detecting water contamination by different organic solvents. Both the monomer and dendrimer aggregate in the aqueous solution because of the ππ stacking of 4-sulfo-1,8-naphthalimide moieties that allows the formation of excimers between the excited and non-excited 4-sulfo-1,8-naphthalimide fragments of dendrimer molecules. Moreover, the incorporation of 4-sulfo-1,8-naphthalimide units into the dendrimer scaffold improves their tolerance towards strong bases. Furthermore, the dependence of the emissions, caused by the aggregation and excimer formations, on water presence enables these molecules to detect the presence of traces of various organic solvents in water and vice versa. It has been shown that the dendrimer SNID mimics both XNOR and INHIBIT logic gates which work in combination to execute the function of the digital comparator.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/s23115268/s1; Supplementary data associated with this article can be found in theonline version. Figure S1. 1 H-NMR of the monomer SNIM. Figure S2. 13 C-NMR of the monomer SNIM. Figure S3. FTIR spectrum of the monomer SNIM. Figure S4. 1 H-NMR spectrum of the dendrimer SNID. Figure S5. 13 C-NMR spectrum of the dendrimer SNID.