Dual-mode colorimetric/fluorescent chemosensor for Cu2+/Zn2+ and fingerprint imaging based on rhodamine ethylenediamine bis(triazolyl silsesquioxane)

A novel dual functional and visual rhodamine ethylenediamine bis(triazolyl silsesquioxane) (RBS) chemosensor was successfully synthesized using “click” chemistry. The results have unambiguously demonstrated that RBS can act in fluorescent and colorimetric sensing of Cu2+ and Zn2+ by their respective coordination with triazole structures and, more importantly, it has also been found that triazole-amide of RBS could turn on chelation-enhanced fluorescence (CHEF) of Cu2+. Remarkably, the addition of Cu2+ triggered an enhanced fluorescent emission by 63.3-fold (ϕF = 0.41), while Zn2+ enhanced it 48.3-fold (ϕF = 0.29) relative to the original RBS (ϕF = 0.006) in acetonitrile (MeCN) solvent. The fluorescent limit of detection for Cu2+ and Zn2+ is similar and fall within 3.0 nM, while under colorimetric sensing the responses were 2.14 × 10–8 and 4.0 × 10–8 mol L−1, respectively. Moreover, the effective sensing profile of RBS and extended applications of RBS–Cu2+ and RBS–Zn2+ for fingerprinting detection and imaging were observed with adequate sensitivity, stability and legibility under the dual visual responses.


Introduction
The design and synthesis of novel nanoscopic molecules that can recognize and detect cations and show specificity for biologically relevant metal ions have gained attention during the last decade [1]. Particularly, Cu 2+ , Zn 2+ and Fe 3+ play vital roles in several biological, environmental and chemical systems. These metal ions are essential trace elements for both plants and animals, including humans. Literature medical data reiterates that excess accumulation of Zn 2+ has the propensity to alter Cu 2+ and Fe 2+ /Fe 3+ absorption in humans and eventually lead to a series of health problems [2]. While several expensive and time-consuming spectroscopic analytical techniques could detect and discriminate Zn 2+ , Cu 2+ and Fe 2+ /Fe 3+ among many other metals [3,4], a more accurate and sensitive probe would be beneficial for detecting these biologically relevant cations.
Within this context, fluorescent and colorimetric sensors are appropriate alternative probes for Cu 2+ and Zn 2+ sensing [5][6][7][8][9][10]. Among common fluorophores, rhodamine derivatives are widely employed for detection of metal ions due to their variably distinguished properties such as (i) the ability to generate cell-and tissue-permeant labels for biological imaging and sensing, despite the photophysical equilibrium between the non-fluorescent lipophilic closed form and the open form of strongly fluorescent rhodamine derivates [11][12][13][14] and (ii) the ability to re-design a red-shifted longer wavelength absorbance/emission spectrum from a green to pinkish red colour for a fluorogenic and/or chromogenic response [13].
To turn on the fluorescence emission of rhodamine probes, a spirolactam ring must be opened [14]. One of the methods to achieve this consists in introducing a strongly chelating metal ion [15]. Several photoactive moieties, including rhodamine, have been effectively conjugated with appropriate binding units using "click"-generated 1,2,3-triazole-based chemosensors for Zn 2+ , Cu 2+ and Fe 3+ [16][17][18][19][20]. For effective and efficient practical application of rhodamine triazolyl-based fluorogenic probes, photostability is a highly required feature that, unfortunately, remains unresolved. Indeed, we recently pioneered the synthesis of fluorescent-labeled triazolyl polyhedral oligomeric silsesquioxanes (POSS) for applications in forensic science [21,22]. Taking advantage of the great versatility of the Cu(I)catalyzed 1,3-dipolar cycloaddition of azides and alkynes (CuAAC), the obtained POSS nanohybrids showed excellent photostability and thermal properties, and were able to image fingerprints with good selectivity, sensitivity and legibility. Several factors that underpin the candidature of POSS include the photostability showed by its triazolyl hybrids, together with their biocompatibility, facile functionalization and nano-dimensionality (0.5-0.7 nm) [23][24][25][26][27]. As we engaged in unprecedented advances vis-à-vis proof of concept of our ongoing work, we present herein the strategic design and synthesis of novel rhodamine-labeled bis(triazolyl)-POSS (RBS) using the popular CuAAC for selective sensing of Cu 2+ and Zn 2+ and extended the application for fingerprint imaging via (RBS-M 2+ )-amino acid interaction. As far as we are aware, this is also the first example utilizing two binding profiles of bis-triazolyl-POSS rhodamine for dual functionality and visuality.

Materials
All chemicals (i.e., rhodamine B, propargyl bromide, sodium azide, ethylenediamine, CuBr, N,N,N′,N″,N″pentamethyldiethylenetriamine (PMDETA) and 3-chloropropyl)hepta(i-butyl)octasilsesquioxane) were purchased from Sigma-Aldrich.  29 Si NMR was recorded in the same manner. Chemical shifts are reported in ppm relative to either tetramethylsilane (TMS) (1H) (δ = 0 ppm) as an internal standard or the residual solvent peak of CDCl 3 ( 1 H NMR: δ = 7.26 ppm; 13 C NMR: δ = 77.16 ppm). FT-IR spectra were obtained with an Agilent Technologies Cary 630 FT-IR spectrometer equipped with Golden Gate Diamond attenuated total reflection (ATR) accessory. Mass data for RBS were obtained at Wurzburg University with an Exactive Plus Orbitrap Mass Spectrometer from Thermo Scientific. All reactions were monitored by TLC using Merck silica gel plates 60 F254, and visualization was accomplished with short-wavelength UV light (254 nm) and/or upon staining with appropriate stains (anisaldehyde, orthophosphomolybdic acid).

Absorption measurements
Steady-state absorption spectra were recorded in a JASCO V-630 spectrophotometer. Quartz cells with 1 cm optical path length and 3 mL capacity were employed. Molar coefficient extinction was determined according to the Lambert-Beer law (Eq. 1): where Abs is the absorbance of the sample, C the concentration (M), and L the optical path length (cm).

Fluorescence experiments
Emission spectra were recorded at 22 °C on a JASCO FP-8500 spectrofluorometer system, provided with a monochromator in the wavelength range of 200-850 nm. From the intersection between normalized excitation and emission spectra, the singlet energy was determined. Fluorescence quantum yields were determined using 9,10-dimethylantracene as standard (ϕ F(std) = 0.95, EtOH) (Eq. 2): where A i is the fluorescence area of the sample, A std is the fluorescence area of standard, Abs and Abs std correspond to the absorbance intensity at excitation wavelength of the sample and standard, respectively, and n is the refraction index of the solution employed. Fluorescence lifetimes were recorded on a Photon Technology International (PTI) fluorometer which includes a pulsed LED excitation source (310 nm), a sample holder and a lifetime detector. For lifetime analysis, EasyLife X software, was used.

Binding constant
The binding constant was determined from a Job plot using Benesi-Hildebrand (B-H) equation [28] (Eq. 3): where F 0 is the emission intensity of the probe RBS at maximum (λ = 580 nm), F is the observed emission intensity at that particular wavelength in the presence of a certain concentration of the analyte (C), F max is the maximum emission intensity value that was obtained at λ = 580 nm during titration with varying analyte concentration, [C] is the concentration of Cu 2+ or Zn 2+ and K a is the apparent binding constant

RBS, RBS-Cu 2+
and RBS-Zn 2+ solutions in THF (20 μM) were irradiated with a monochromatic xenon arc lamp at irradiances of 0.039 W/cm 2 and 0.052 W/cm 2 for 60 min. Photostability was evaluated by monitoring absorption and emission spectra as a function of time.

Fingerprint development and imaging
Fingermarks were collected from two voluntary anonymous donors and deposited on various selected substrate surfaces (i.e., smooth cell phone surface, cellotape paper, glass bottles) (Fig. S1). After the donor rubbed with their thumb or forehead/nose, it was press-stamped on the selected substrate. The sample was then immersed in RBS, RBS-Cu 2+ and RBS-Zn 2+ solution and rinsed with water or some organic solvents (acetone or ethyl acetate). The developed fingerprint was illuminated with UV lamp (365 nm) and the image recorded with a Samsung smartphone camera.
Similarly, 13 C NMR, 29 Si NMR and mass spectra (Figs. S8, S9 and S10, respectively) also supported the structure of RBS. However, it draws attention the presence of a signal centered at 65.9 ppm in the 13 C NMR spectrum, which indicates the closed form of the spirolactam ring (Fig. S8) [20]. The presence of 1,2,3-triazole groups in the RBS probe was also corroborated by FT-IR, which prominently exhibited -N=N group and -C-N bonds in addition to the characteristic Si-O-Si of the POSS (Fig. S11).
We explored its photophysical properties in different solvents which included absorption/emission bands, molar absorption coefficients, Stokes shifts, fluorescent quantum yields, emission rate constants, singlet energies and lifetimes ( Fig. S12 and Table S1). The absorption band in the UV region was clearly dominated by the rhodamine-type chromophore. Values of Stokes shifts were found to be solvent dependent, being markedly higher for polar media. Regarding fluorescence quantum yields (ϕ F ), RBS presented extremely poor values in all solvents in comparison with the parental rhodamine B (for instance, ϕ F = 0.40 in acetonitrile) [12]. This fact could be attributed to the nearby POSS moieties that may influence the radiative and non-radiative pathways of the material, with the occurrence of aggregate formation. In this sense, the emission decay traces of RBS were satisfactorily fitted by a biexponential function. This was in full agreement with previously reported data for similar POSS nanohybrids containing dansyl chromophores [22]. Thus, the shorter lifetimes would be ascribed to an intramolecular charge transfer (ICT) state, in which an electron transfer from the amino group to the xanthene π-system, followed by rotation of the alkyl groups was occurring [29]. Meanwhile, the formation of aggregates could be responsible for the longer lifetimes. As previously established for the rhodamine dyes that can easily form aggregates in solution [30], when a locally excited (LE) state cannot be mixed with a charge transfer (CT) (or electron transfer) state, the interconversion from the LE to the CT results in fluorescence properties that are highly sensitive to steric environments, i.e., aggregation. This fact was indeed reflected in the contribution of the longer lifetime component, being the same or even higher, at moderately polar solvents, strengthening a major participation of aggregate formation. With the photophysical data in hands, RBS might be a potential candidate to be successfully applied as an "OFF-ON" metal sensor, since the possible coordination could evade formation of the aggregates and enhance the radiative pathways.

Metal ion sensing capability of the RBS probe
To establish the sensing capacity of RBS, we first evaluated its photophysical properties, focusing mainly on the emission property changes due to the low sensitivity that UV-visible (absorbance) measurements usually exhibit. A solution (20 μM) of RBS was visually colorless and almost non-fluorescent in common solvents such as THF, MeCN or n-hexane (HEX) (   1 3 [20]. Upon separate addition of several metal ion solutions to RBS, only Cu 2+ and Zn 2+ exhibited a strong fluorescent band (580 nm) with Cu 2+ having stronger emission band (Fig. 2). Under incremental addition of Cu 2+ to RBS, the absorption spectra supported the fluorescence spectra as shown in Fig. S13.
This observation indicates that only Cu 2+ and Zn 2+ from among the tested metal ions could coordinately chelate with RBS and, eventually, emit reddish pink and yellowish green  (Table 1). These results provide a platform for Cu 2+ and Zn 2+ sensing in either biological or environmental condition accompanied by structural transformation occasioned by ring opening of the spirocyclic unit via a "turn-on" chelation-enhanced fluorescent (CHEF) mechanism [31]. It is worth mentioning that we also tested additional monovalent and trivalent ions (i.e., Al 3+ , Fe 3+ , Cr 3+ , Na + , Ag + and K + ). However, titration experiments resulted in non-coordinated uncomplexed colloidal suspensions, which made further photophysical experiments rather difficult. Therefore, we limited in this study the probe responses to representative divalent metal ions shown in Fig. 2A.
Due to the overriding sensing performance of RBS for Cu 2+ , we performed titration experiments. The results indicated that the probe displays significant "turn-on" responses upon incremental addition of Cu 2+ (0-25 μM ranging from 0 to 10 equiv) (Fig. 3A). The corresponding absorption spectra are shown in Fig. S14.
The saturation point was easily reached with the addition of just 7 equiv of Cu 2+ , indicating tight binding. However, the reversibility of, in particular, RBS-Cu 2+ formation is underscored by a reversible color transition from reddish pink back to colorless upon addition of EDTA which brought down the emission intensity to the same level as in the original uncomplexed RBS (Fig. 3B). Thereby, the loss of the spectroscopic features of the system due to the induced decomplexation process would be indicative of the restoring of the spirocyclic form. The equilibrium interchange [14] between the open and closed spirolactam ring in reversible motif prompted by EDTA addition could be an index for monitoring metal complex formation and subsequent sensing of M 2+ .
Among the divalent metal ions tested for RBS sensing, Cu 2+ and Zn 2+ are, undoubtedly, notable in this work. However, where both metal ions co-exist, selectivity should be an important parameter for testing the performance of a novel probe like RBS. In this regard, we conducted a competitive experiment by monitoring the change in fluorescent intensities at 580 nm upon addition of 5 equiv of Cu 2+ to a solution of different metal ions (5 equiv). The results are displayed in the chart (Fig. 4A). High selectivity of RBS for Cu 2+ from among all the metals tested was underscored by the significant fluorescent profile in a co-existing multi-metallic interference environment. Furthermore, confirmation of the continued selectivity and sensitivity dominance of Cu 2+ sensing among multi-metallic interferences was further proved by the addition of RBS-Zn 2+ to various cations, which did not induce any noticeable changes in the relative intensities of other metal ions except Cu 2+ (Fig. 4B).

Dual sensing profile of RBS
Fluorescent transition from colorless to reddish pink and colorless to yellowish green favoring RBS-Cu 2+ and RBS-Zn 2+ falls within the addition of 1-7 equiv of metal ions to RBS. From Benesi-Hildebrand plots, we could determine the binding constants of 5.12 × 10 8 and 3.6 × 10 8 , respectively, which are high enough to justify stable RBS-M 2+ bindings (Fig. 5).
The detection limit of Cu 2+ and Zn 2+ via fluorescent response was the same as 3.0 nM, which allows sensing of both metal ions in the nanomolar concentration range. It is noteworthy that colorimetric responses capable of being detected with the naked eye proceeded with the incremental addition of metal ions (8-10 equiv) to RBS (Fig. 6A). The colorimetry responses were supported by a linear relationship between the absorption intensity at 452 nm and various concentration of metal ions in the range 1-100 μM (Fig. 6B)  B Reversibility of fluorescence spectra of complexed (7 equiv of Cu 2+ ) and de-complexed RBS following the addition of excess EDTA 2.14 × 10 -8 and 4.0 × 10 -8 mol L −1 , respectively. Hence, the results obtained, to a large extent, validate the reliability and practicality of the established dual-mode colorimetry and fluorescent sensing of Cu 2+ and Zn 2+ . Furthermore, the dual-mode sensing of Cu 2+ and Zn 2+ by RBS was compared with a series of previously reported probes (Table 2). In general, results shown in this work are encouraging and superior in some cases.

Stability profile of RBS in sensing Cu 2+ and Zn 2+
Photo and chemical stability are major pre-requisites for any chemosensor to be considered for biological and environmental applications [35]. Thus, we determined the stability of RBS-Cu 2+ and RBS-Zn 2+ over the physiologically relevant pH range (1-12) (Fig. 7). Only RBS-Cu 2+ was found to be strongly pH dependent and recorded relatively weak fluorescent decline up to pH 5. However, there was an exponential increase in fluorescence up to pH 11, indicating that proper pH is a pre-condition for bio-application of RBS-Cu 2+ . On the other hand, neither RBS nor RBS-Zn 2+ was pH dependent, showing that RBS could not be used for Zn 2+ detection under physiologically aqueous conditions. More detailed experiments are yet necessary to unequivocally correlate protonation/deprotonation rate constants and/or induction of molecular aggregation with the emission intensity profile over the pH scale for the different complexes. Secondly, to establish the photostability profile, we irradiated a solution of RBS-Cu 2+ and RBS-Zn 2+ at 356 nm up to 60 min. The absorption and emission as a consequence of the irradiation are comparatively illustrated in Fig. 8. As it can be observed, RBS in its complex formation is adequately photostable in sensing both Cu 2+ and Zn 2+ , as there are no significant absorption and emission changes at their respective maximum band peaks.  Along with the adequate photostability shown by RBS in sensing Cu 2+ and Zn 2+ , the fast sensing response time toward these metal ions should also be highlighted, within 11 and 13 min, respectively, and keeping constant thereafter for a reasonable period of time (Fig. 9).
Additionally, the excited state of RBS-Cu 2+ and RBS-Zn 2+ showed adequate lifetime within 3.20 ns and 1.68 ns, respectively, which equally signified sensing stability ( Table 1) (Table 1), corroborating the good stability, reliability and practicality of the dual sensing properties of RBS for these metal ions. These results underpinned to a large extent earlier studies [21,22], reiterating triazolyl-POSS with scavenging photodegradation. Our designed bis-triazolyl-POSS anchoring rhodamine enhanced the photostability as a result of the strong binding profile of bis-triazoles to metal ions. The folding tendency of the triazole unit toward metal coordination depends on the bulkiness of the group at the terminal. In this sense, the inclusion of large POSS units facilitates the folding over the diamino linker favoring a stronger metal coordination.

Coordination sensing mechanism
The observed dual-mode visuality prompted by fluorescent and colorimetric responses of RBS to Cu 2+ and Zn 2+ occasioned, undoubtedly, the plausible binding modes vis-à-vis structural orientation as detailed in Fig. 10. Job's plot indicates that RBS-Cu 2+ and RBS-Zn 2+ have preferences for 2:1 and 1:1 binding stoichiometry, respectively (Fig. 11). The observed discrepancy in the binding stoichiometry suggests different binding orientations with respect to bistirazole (-N=N-Cu-N=N-) and open spirolactam carbonyl of the amide/triazole (O-Cu-N=N). Plausibly, once the Cu 2+ or Zn 2+ is fixed in the cavity of RBS via the coordination-bonding interaction, the electron cloud on the electronrich imine fragment will transfer to the large rhodamine ring system resulting in the ring opening of spirolactam amide to form a five-member or six-member ring [36]. The exclusive sensing of Cu 2+ and Zn 2+ owes to the stronger ability of these ions toward the ligand due to the optimal size of RBS cavity. We can further explain this phenomenon based on hard and soft acid and base interaction (HSAB) theory [36,37] and chelation size [38], since Cu 2+ and Zn 2+ have comparable chelating sizes. The predominating selectivity of RBS for Cu 2+ is based on an analogy that the metal ion is a hard acid, which preferentially opts for a stronger "N" hard base in the triazole moiety [35].
As predicted, RBS was much more selective for Cu 2+ than Zn 2+ and free from cross reactivity with other competing metal ions. It should also be considered that an interference of Zn 2+ could be significant where Cu 2+ and Zn 2+ co-exist in analytical samples for physiological applications. This prompted us to carry out further dual metal system experiments. The addition of Zn 2+ to the RBS-Cu 2+ resulted in a 2.1-fold fluorescent intensity (see Fig. 10, inset). Such enhancement may be due to the addition of Zn 2+ , which possesses the tendency to displace Cu 2+ from the triazole-triazole (N=N-Cu-N=N) coordination and, eventually, opt for amide-triazole (O-Cu-N=N) coordination. As previously proposed [39], and supporting our analogy, the folding tendency of the triazole unit moves toward the amide segment and is favorable for Zn 2+ coordination. Furthermore, the dangling bulky triazolyl-POSS could facilitate Zn +2 coordination during Cu 2+ /Zn 2+ exchange.

Deployment of the resulting RBS-Cu 2+ and RBS-Zn 2+ for fingerprinting imaging
Inspired by our most recent work [21,22] where we successfully took advantage of fluorophore-triazolyl-POSS for fingerprint imaging, coupled with various reported M +X complexes [40][41][42], we considered it worthwhile to adopt the fluorescent and colorimetric attributes of RBS-M 2+ in fingerprint imaging. For these studies, fingerprints of selected donors were imaged and detected via both fluorescent and colorimetric responses of RBS-Cu 2+ and RBS-Zn 2+ with high sensitivity, selectivity and legibility (Fig. 12). Detailed observation of the ridge features showed Whorf, bifurcation and ridge ending, which to a large extent fulfill the requirements of fingerprint identifications in forensic science.
A judicious balance of hydrophobic and π-π interactions have been recognized in previous reports [21,22]. Accordingly, RBS-M 2+ was designed to possess multiple binding profiles which are subsequently available for interaction with the -C=O and -NH 2 of amino acids in fingerprint oils (Fig. 13). The whole ensemble, which consists of RBS-M 2+ in non-covalent interactions with -C=O and -NH 2 , formed a ternary complex that cooperatively work to provide stabilized fingerprint detection and eventual imaging. Some fluorescent chemosensor systems have been developed to detect amino acids based on probe-metal ion ensembles [15]. Importantly, brightness, contrast and visual legibility remained unchanged up to 1 year, when developed with RBS-Cu 2+ under fluorescent response (2 equiv of Cu 2+ ). However, dimmer and less contrast pink and green images surfaced at the end of the 6th month for RBS-Cu 2+ and RBS-Zn 2+ , respectively, when colorimetric responses (10 equiv) predominate. We could ascribe the apparently lesser short-lived and dimmer image features obtained under colorimetric responses to unhidden open images that are frequently exposed to light, which eventually fade the pink and yellowish green fingerprint images over time. Obviously, the latent nature of fingerprint developed by fluorescent RBS-Cu 2+ (2 equiv) forestalls frequent exposure to light and eventually confers extra stability to the reddish pink fingerprint image which spanned over 12 months.

Conclusion
We have established a "turn-on" type of CHEF and colorimetric sensor for Cu 2+ and Zn 2+ , respectively, using a novel RBS derivative. The binding pattern for RBS-Cu 2+ (2:1) and RBS-Zn 2+ (1:1) exist by coordination of the corresponding metal ions with triazole-triazole and triazole-amide moieties. Competitive experiments with RBS-Cu 2+ showed that the addition of Zn 2+ could displace Cu 2+ by triggering a fluorescent enhancement (2.1-fold). The synthetic design of the RBS sensor was modular, being possible to further attach a number of other fluorophores using "click" chemistry while creating a library of M 2+ sensors with different excitations, absorptions and emission wavelength. This is the first example of a bis-triazolyl-POSS conjugate anchoring on rhodamine. Therefore, one can envisage other uses for such a generic system in chemical sensing Funding Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature.
Data availability All data generated or analysed during this study are included in this published article and its supplementary information files. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.