Selective Colorimetric Detection of Mn2+ and Cr2+ Ions using Silver Nanoparticles Modified with Sodium Dodecyl Sulfonate and β-Cyclodextrin

The present study was conducted to determine manganese (Mn2+) and chromium (Cr2+) ions in aqueous systems with a simple, rapid, sensitive, selective and cost-effective colorimetric approach. Sodium dodecyl sulfonate (SDS) and β-cyclodextrin (CD) were used as both stabilizer and surface functionalizing agents for the synthesis of silver nanoparticles (AgNPs). Synthesized AgNPs were characterized by FT-IR spectroscopy, UV-visible spectroscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD) and dynamic light scattering (DLS) techniques. The effect of pH on the stability of nanoparticles was investigated. SDS modified silver nanoparticles (SDS-AgNPs) and SDS with β-cyclodextrin modified silver nanoparticles (SDS-CD-AgNPs) demonstrated sensitive and selective colorimetric detection of Mn2+ and Cr2+ ions at ppm level, respectively. The prepared SDS-AgNPs and SDS-CDAgNPs solution showed a color change visible to the naked eye from yellow to orange upon adding Mn2+ and Cr2+ ions, respectively; however, other metal ions did not induce such a change. In addition, the results of SEM TEM and DLS” for both sensors showed the aggregation of nanoparticles after adding Mn2+ or Cr2+. Overall, the results of this study show the successful synthesis of AgNPs, SDS-AgNPs, and SDS-CD-AgNPs as simple, rapid, sensitive, selective colorimetric sensors with high potential for rapid and on-site detection of Mn2+.


Introduction
Manganese (Mn) and Chromium (Cr) are essential trace elements for many life processes of the human body such as enzyme activity, bone growth, and fat metabolism. 1,2 However, excessive levels of these elements in biological activities can be toxic for health and may lead to fatal diseases. 3,4 Essential quality standards for Mn (0.05 mg/L) and Cr (0.1 mg/L) have been adopted at the EU level. 5,6 Therefore, the determination of these ions has become inevitable for remedial processes. In this regard, many efforts have been done to detect Mn and Cr ions with high selectivity and sensitivity.
Conventional methods for detecting metal ions are mostly spectroscopy-based techniques such as fluorescent probes, 7,8 atomic absorption, 9,10 surface-enhanced Raman spectroscopy, 11 inductively coupled plasma atomic emission, 12 and inductively coupled plasma mass spectroscopies. [13][14][15] However, some of these methods are not widely used due to high cost, multi-step sample Pretreatments, and time-intensiveness. 16,17 The attempts to overcome these limitations led to the development of alternative colorimetric sensors. Several colorimetric assays for detecting metal ions in aqueous solutions based on the use of sensitive chromophores or fluorophores, 18 polymers, 19,20 oligonucleotides, 21 DNA, 22,23 and metal nanoparticles 24,25 have been developed. These assays have advantages such as high sensitivity, simplicity, speed and ease of use without the need of any special apparatus. 26 Nanostructured materials have extensive applications in biosensors and ecosensors, especially for detecting metal ions. [27][28][29][30][31][32] Noble metal nanoparticles, especially gold and AgNPs, have attracted considerable attention in preparing metal ion sensors because of their size and morphology dependent optical properties, strong surface plasmon resonance properties, high extinction coefficient at the visible region, highly stable dispersions, good biocompatibility and chemical inert-Akhondi and Jamalizadeh: Selective Colorimetric Detection of Mn 2 + and Cr 2+ Ions ... ness. 33,34 Notably, AgNPs have attracted much attention because of lower prices and higher extinction coefficients than those of gold nanoparticles. 16,35,36 The performance of these sensors is based on the surface plasmon resonance (SPR)-induced color variance of nanoparticles, as a result of signal-analyte interaction. [37][38][39][40] Nanoparticles surface functionalization affects the interaction between nanoparticles and metal ions, 38,41 which is critical in the development of noble nanoparticles as a colorimetric sensor for metal ions. For example, among the works conducted in this regard, one can name Ag nanoparticles surface functionalization by octamethoxy resorcin [4]arene tetrahydrazide as a reducing and stabilizing agent for colorimetric detection of cadmium, 42 synthesized AgNPs by cysteic acid as a capping agent for detection of manganese ions, 17 Glutathione-stabilized Ag-NPs as a colorimetric sensor for nickel ions, 43 thiodisuccinic acid-functionalized AgNPs as a sensor for lead ions, 38 β-cyclodextrin-functionalized AgNPs for detection of mercury and sulfide ions, 44 glutathione-modified AgNPs as a sensor for cobalt ions, 45 dopamine-AgNPs as a sensor for copper ions, 46 AgNPs without any surface modification as a sensor for copper ions, 47 and N-acetyl-L-cysteine-stabilized AgNPs for detection of iron ions. 48 The purpose of this study is to develop a simple, rapid, efficient and cost-effective method for the determination of some metal ions by functionalized AgNPs. Two new AgNPs based colorimetric sensors were prepared with SDS and SDS-CD as the stabilizing agents without further modification and investigated as colorimetric sensors for Mn 2+ and Cr 2+ ions, respectively. There are many published works on colorimetric detection of Mn 2+ ion based on functionalized AgNPs system; e.g., Pyrophosphate capped AgNPs (Na 4 P 2 O 7 ), 49 Sodium pyrophosphate (Na4P2O7) and hydroxy propyl methyl cellulose-stabilized AgNPs, 50 5-sulfoanthranilic acid dithiocarbamate-stabilized AgNPs, 51 L-tyrosine stabilized AgNPs, 52 l-arginine-stabilized AgNPs 16 and alginate-stabilized Ag-NPs. 53 Although several efforts have been devoted to the development of colorimetric detection of both Cr 3+ and Cr 6+ cations, 54-57 based on the functionalized AgNPs system, the detection of Cr 2+ species has been overlooked. To the best of authors' knowledge, only, Rull-Barrull et al. reported a selective rhodamine-based chemosensor for colorimetric and fluorimetric detection of Cr 2+ under aqueous conditions. 58 Herein, the SDS was chosen as a candidate for modifying the AgNPs to design a selective and sensitive colorimetric sensor for detecting Mn 2+ in the wide range pH (3-13) with a suitable response rate. Also, SDS and CD were chosen to modify the AgNPs to design a selective and sensitive colorimetric sensor for detecting Cr 2+ in a wide range of pH (i.e., 3-13) with a suitable response rate. Consequently, simple, rapid colorimetric assays for detecting Mn 2+ and Cr 2+ were established. These detection methods showed successful detection under various pH conditions. The colorimetric detection mechanism of two sensors, which is based on the accumulation of AgNPs, is illustrated in Figure 1.

Instrumentation
UV-vis spectra were recorded on a spectrophotometer (Varian, Cary50), using a 1 cm of quartz cell. Scanning Figure 1. Schematic illustration of detecting mechanism of Mn 2+ and Cr 2+ ions based on the aggregation of SDS-AgNPs and SDS-CD-AgNPs, respectively electron microscopy (SEM) images were obtained using a field-emission scanning electron microscope (KYKT, model EM-3200). Transmission Electron Microscope (TEM) images were recorded from JEOL 2100F TEM. XRD patterns were obtained in reflection mode using a powder X-ray diffractometer (X′Pert Pro MPD Philips Xray diffractometer, Ni-filtered Cu-Kα radiation, λ = 0.154 nm). The FT-IR spectra were recorded using the FT-IR spectrometer (Brucker, model Tensor 27) in the region 4,000-400 cm -1 using KBr pellets. To determine the mean hydrodynamic diameter (d H ) of the AgNPs in aqueous solution, dynamic light scattering (DLS) investigations were performed without filtering AgNPs solutions at 25 °C using the new non-invasive back-scatter (NIBS) technology by a 1-cm quartz cell. The pH measurements were performed using a pH-meter (METROHM, Model 713).

3. Synthesis of AgNPs
AgNPs were prepared by reducing AgNO 3 with NaBH 4 in the presence of SDS or mixture of CD and SDS as stabilizing agents. Briefly, 0.1 g of the stabilizing agent was dissolved in 50 ml deionized water. Then, 1.0 mL of AgNO 3 aqueous solution (0.001M) was slowly added into the above solution under vigorous stirring for 30 min. Finally, 0.001 g NaBH 4 was added into the solution, which caused the color change from yellow to orange.

4. pH-Dependent Stability Study of AgNPs
There are several important factors that affect the stability of the AgNPs. 59 In this section, we will investigate the effect of pH on the stability of AgNPs by monitoring the color variation and UV-vis intensity changes of the Ag-NPs solution upon the addition of HCl or NaOH solution (0.1 M). In order to study the effect of pH on the stability of AgNPs, the absorption spectra were recorded after 30 min of mixing.
For all samples, after 2 min of mixing, some photographs were taken using a digital camera under daytime light and absorption spectra were recorded using a spectrophotometer. The limit of detection (LOD) and limit of quantitation (LOQ) were determined from the standard deviation of the slope of the calibration curve using the following expressions: LOD = 3.3 σ a /b and LOQ = 10 σ a /b; where σ a is the standard deviation of the blank signals and b is the slope of the calibration curve.

1. Characterization of Silver Nanoparticles
The synthesized AgNPs were characterized by UV-visible spectroscopy, FT-IR spectroscopy, SEM, TEM, TEM, XRD and DLS techniques. AgNPs absorption spectra in the UV-visible region show a specific absorbance band due to the excitation mode of their surface plasmons by incident light. 60 Based on Mie's theory, spherical nanoparticles have only one SPR band depending on their size. 61,62 Thus, the UV spectrum was used to confirm the synthesis of AgNPs ( Figure S1). 61,62 The UV-Visible spectrum shows an absorption maximum that appears at 410 and 400 nm for SDS-AgNPs and SDS-CD-AgNPs, respectively. These values are in good agreement with the literature values for AgNPs. 60 The difference in UV-vis absorption spectra between the SDS-AgNPs and SDS-CD-AgNPs spectrums is due to the variation in the size distribution of the AgNPs as confirmed using SEM and TEM ( Figures S2  and 2). The results of SEM and TEM images show that both nanoparticles are polygonal. But, compering these nanoparticles show that SDS-AgNPs have a smaller size with a narrower size distribution than SDS-CD-AgNPs. Based on the TEM results, the average diameter of SDS-AgNPs and SDS-CD-AgNPs are 8 nm and 15 nm, respectively. According to the lectures, using the same reduction reagent, nanoparticle size can be varied by changing the synthesis conditions. 63,64 Probably, the size difference is due to the different rates of diffusion of silver ions into the seed of SDS-AgNPs and SDS-CD-AgNPs. According to previous studies, it is very likely that for both types of nanoparticles (i.e., SDS-AgNPs and SDS-CD-AgNPs), the anionic SDS was adsorbed on the surface of nanoparticles through hydrophobic bonding with a tail surface/headwater orientation. Furthermore, in the presence of cyclodextrin, the hydrophobic carbon chain of SDS was wrapped into the CD cavity-forming host-guest molecules complex. 65 Based on these results for the SDS-CD-AgNPs, the AgNPs were capped by the host-guest inclusion of SDS-CD. And it can be stated that the diffusion of silver ions into the primary seed coated with SDS is more difficult than penetration of silver ions into the primary seed coated with SDS-CD. As a result, the size of SDS-AgNPs is smaller than the size of SDS-CD-AgNPs. In addition, based on the TEM results SDS-CD-AgNPs are more agglomerate than SDS-AgNps. According to the pka values, the SDS surface groups are more deprotonated than the CD groups at neutral pH, consequently, more electrostatic repulsions between the negatively charged groups of SDS-AgNPs, leading to more dispersion of SDS-AgNPs than SDS-CD-AgNPs. 66 In addition, the binding affinity (hydrogen and hydrophobic interactions) between cyclodextrin molecules on the surface of AgNPs leads to some agglomeration of SDS-CD-AgNPs. These agglomeration structures for SDS-CD-AgNPs can be seen in the TEM image.
The effective diameter of the AgNPs in the aqueous solution (pH = 7) was measured by DLS ( Figure S3). Unlike the results of TEM, the hydrodynamic diameter of SDS-CD-AgNPs was significantly larger than the hydrodynamic diameter in DLS analysis. Indeed, this analysis shows how a particle diffuses within a fluid while TEM just shows the shape and size of AgNPs. 67 The FTIR spectra of the SDS-AgNPs and SDS-CD-Ag-NPs are also recorded to identify the functional groups of the SDS and CD involved in the synthesis of the AgNPs. According to the FTIR spectrums in Figures 3a and 3b, the sharp peaks belonging to the SDS and CD were observed in the spectrum of the AgNPs. In Figure 3a, the FTIR spectra of the SDS-AgNPs and SDS-CD-AgNPs show the peaks at 2943, 2915, 2848, and 1465 cm -1 corresponding to the CH 2 stretching and bending modes, the 1220 cm -1 peak corresponding to skeletal vibration involving the bridge S-O stretch, and the 1075 cm -1 peak corresponding to C-C band stretching of SDS. Also, the peaks at 827 and 579 cm -1 correspond to asymmetric C-H bending of the CH 2 group of SDS. 68 The FTIR results are in good agreement with the values reported in literature for SDS. In addition, the spectra of SDS-CD-Ag-NPs in Figure 3b shows such peaks at 3400 cm -1 , 2930 cm -1 ,    70,71 Moreover, the spectrums of AgNPs contain the XRD characteristic peaks of Na 2 SO 4 that corresponded well with those of the standard spectrum. As reported in the literature, 72 SDS undergoes hydrolysis to yield dodecanol and sodium hydrogen sulfate when heated and then it is converted to Na 2 SO 4 .

2. pH-Dependent Stability Study of AgNPs
As the pH of the solution may influence the surface charge and aggregation of NPs, the pH-dependent stability of SDS-AgNps and SDS-CD-AgNps were studied by UV-visible absorption measurements at the pH range of 1-13 Figure S4. The results show that the SDS-AgNPs are stable in the pH range of 3-11. However, according to a) b)  significant changes in the SPR absorption intensity and color of solutions, nanoparticles are unstable at pH < 3 and pH > 11. At higher pH values (> 11), SDS-AgNPs aggregate shows a sudden change in the solution color from yellow to orange, a decrease in SPR absorption intensity, and a new peak at 547 nm. According to Mie's theory, 73 due to the exhibition of two bands, SDS-AgNPs may not be spherical and have an anisotropic shape at this range of pH.
From absorption spectra and image of the SDS-CD-AgNPs, it is seen that the particles are stable when pH value changes from 3 to 13. In addition, according to significant changes in the SPR absorption intensity and color of solutions, nanoparticles are unstable at pH < 3.  Co 2+ , Cu 2+ , Ni 2+ , Zn 2+ , Cd 2+ , Fe 2+ , Hg 2+ , and Mn 2+ ions) at pH 7. In addition, the results indicate that the solution contacting both Cr 2+ and Hg + ions changes from yellow to brown. Also, a change in the SPR band was observed (Figure S5). In addition, the absorption ratios of this solution are much lower than the other solutions. These results show that the sensing of Cr 2+ ions is also possible in the presence of other metal ions and SDS-CD-AgNPs has a good selectivity to Cr 2+ ions. Furthermore, it is possible to use the SDS-CD-AgNPs solution containing Cr 2+ ions as an Hg 2+ sensor. Similar research has been done in this regard. 11,35 The sensitivity of SDS-AgNPs toward Mn 2+ was evaluated by varying the concentration of Mn 2+ ions ( Figure  7). According to this Figure S6, by adding Mn 2+ ions (1.0 mM to 6.0 µM) to SDS-AgNPs solution, the color of SDS-AgNPs solution changed, which was detected by the naked eye. But, based on the UV-vis spectrum (Figure 7), the changes in the absorption intensity at the maximum wavelength (λ 410 ) showed a linear relationship only in the concentration range from 6.0 µM to 9.0 µM. And in this concentration range the color of the solutions changed from yellow to orange, Figure S6. The LOD and LOQ for Mn 2+ were 0.02 µM and 0.04 µM, respectively. Similarly, as can be seen from Figure S7, by adding Cr 2+ ions (1.0 mM to 3.0 µM) to SDS-CD-AgNPs solution, the color of solution changed. But, the changes in the absorption intensity showed a linear relationship with the concentration of Cr 2+ only in the range from 3.0 µM to 9.0 µM. As Figure 8a Shows, by increasing Cr 2+ concentration from 3.0 µM to 9.0 µM, the UV-vis absorption peak intensity at 400 nm decreased and the color of the solutions changed from yellow to orange ( Figure S7). The LOD and LOQ for Cr 2+ were 0.01 µM and 0.03 µM, respectively. Figure S8 shows the changes in the SDS-AgNPs absorbance spectrum after the addition of Mn 2+ ions. As can be seen, after about 1 min, the colorimetric response of the sensor is observed. Also, after 14 min, the aggre- Similarly, Figure S9 shows that after the addition of Cr 2+ to the SDS-CD-AgNPs solution, first, there is a decrease in the SPR peak intensity and then it reaches relatively constant values within 4 min.

5. Optimization of the Conditions for the Mn 2+ or Cr 2+ Measurement
In the following, the effect of the concentration of the AgNps and pH on the detection of Mn 2+ or Cr 2+ ions was investigated.

5. 1. Effect of the Concentration of the AgNps
The effect of the dosage of AgNps on their colorimetric sensing ability is examined as follows: First, different concentrations of AgNPs were prepared with diluting Ag-NPs solutions in different volumes of deionized water as blank solutions (2:1, 1:1, and 1:2 V:V). Then, different volume ratios of AgNPs solutions to the desired ion solution are prepared (2:1, 1:1, and 1:2).
The absorption spectra and image of Figure 9 reveal that by varying the dosage of AgNps, the colorimetric sensing ability of AgNPs was preserved; however, by increasing the volume ratio of AgNPs solutions to the desired ion solution, a decrease in analytical signal occurs. This reduction could be due to the complexing action of SDS or CD-SDS surface functionality with metal ions (Mn 2+ or Cr 2+ ), leading to the aggregation of NPs. Also, the Figure shows that by increasing the volume ratio of the AgNPs to the desired ion solution, a higher red shift in the AgNPs SPR band occurred. Since desired ions could complex with surface functionality of several nanoparticles, it can be stated that the presence of more nanoparticles around the desired ions in the solution results in agglomeration of more nanoparticle, forming bigger agglomerates, and a higher redshift in the AgNPs SPR band. Thus, the volume ratio of AgNPs to ion solution was selected to be 2:1 with maximal changes in SPR peaks and the color of the solution was selected as the optimal ratio for detecting desired ions by both sensors.

5. 2. Effect of pH on the Detection of Mn 2+ or Cr 2+ Ions
The effect of pH on the performance of the sensors was tested as follows. About 1 mL AgNPs solution was mixed with 0.1 ml HCl or NaOH solutions with different concentrations. Next, 1 mL 10 μM Mn 2+ or Cr 2+ ions solution was added to the AgNPs solutions. Figure 10 present the effect of pH (1 to 13) on the colorimetric response of the sensors. According to Figure, by adding desired ions to all solutions, a color change and a decrease in the absorbance ratio is clearly observed. These results indicate that both sensors keep their sensing capability at a pH range of 3 to 13. The results show that at the investigated pH range (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13), pH values of 5 and 11 have more UV-vis intensity changes of the AgNPs solution toward Mn 2+ detection than the other pHs. Thus, the pHs of 5 and 11 were selected as the best values for detecting manganese ion by SDS-AgNPs in acidic and basic media, receptively. Meanwhile, the pHs of 3 and 9 showed more UV-vis intensity changes toward Cr 2+ detection than the other pHs. The effect of pH on the colorimetric response of AgNPs can be ascribed to the four factors: (I) AgNPs dissolution (II) forming hydroxide for metal ions, (III) size of AgNPs, and (IV) the effect of electrostatic inter- The results reported in the literature show that particle size has generally an inverse effect on AgNPs dissolution: small AgNPs release more Ag + than large ones because smaller particles are energetically unfavorable due to higher surface-to-volume ratio and consequently are more soluble 76 . Based on the TEM results, SDS-AgNPs have a smaller size and a narrower size distribution than SDS-CD-AgNPs. Thus, SDS-CD-AgNPs are more stable than SDS-AgNPs. It was shown at low pH values, AgNPs are dissolved. 76 As can be seen in Figure 10, at pH = 1, both AgNPs are dissolved and no colorimetric response was observed for them. Also, at pH = 3, the SDS-AgNPs have less UV-vis intensity changes toward Mn 2+ detection than the pH = 5, which could be due to the lower concentration of SDS-AgNPs at pH = 3 than pH = 5. However, at pH = 3, the SDS-CD-AgNPs have the best UV-vis intensity changes toward Cr 2+ detection, which is due to the larger size of SDS-CD-AgNPs than SDS-AgNPs and their more stability at this pH.
Overall, it can be stated that the performance of SDS-AgNPs and SDS-CD-AgNPs between pH = 5-11 and pH = 3-9 is influenced by two factors, which compete with each other: (I) the size of AgNPs and (II) the electrostatic interaction between the negatively charged groups of Ag-NPs and positively charged matal ions. Thus, at lower pH, size may control the sensing ability of AgNPs, so that by decreasing pH, the size of particles decreased, as well. And at upper pH, the negative charge of the SDS or CD groups of AgNPs is a more effective factor. So, by increasing pH, SDS surface groups are more deprotonated and, consequently, the electrostatic repulsions between the negatively charged groups increased, leading to an increase in the dispersion of AgNPs. Thus, the desired ions could complex more with surface functionality of nanoparticles than lower pH. Although proper dispersion of nanoparticles enhances the responsiveness of the sensors under alkaline conditions, at pH = 11 for Mn 2+ and pH = 10 for Cr 2+ , the sensor performance is reduced due to the formation of manganese hydroxide and chromium hydroxide and, consequently, a decrease in the concentration of Mn 2+ and Cr 2+ ions. 50,52 However, the results indicate that the colorimetric response of SDS-CD-AgNPs at pH = 13 is higher than pH = 11, probably due to the deprotonation of hydroxyl groups of CD bound at the surface of nanoparticles at pH > 12. Therefore, dihydroxylation of OH-groups of cyclodextrin molecules plays a key role in enhancing the formation of Cr 2+ -CD complex and agglomeration of the nanoparticles. 66

6. SEM, TEM and DLS Analysis
To explain the mechanism of sensing of the AgNPs to detect desired ions, the nanoparticles were examined before and after desired exposure using UV-vis, TEM, SEM, and DLS observations. According to the above results, the color of both nanoparticles changes from yellow to orange in the presence of desired ions. The results of TEM, SEM, and DLS for SDS-AgNPs sensor show that SDS-AgNPs are dispersed with small size (Figures 2a, S2a and S3a); but, after adding Mn 2+ ions, a significant aggregation of nanoparticles is observed (Figures 11a, S 10a (Figures 11 b, S 10b and S 11b). According to the UV-vis results, with increasing the metal ions concentration (Mn 2+ or Cr 2+ ), a red shift in the AgNPs SPR band appears. This shift, which could be due to the complexing action of SDS or CD-SDS surface functionality with metal ions (Mn 2+ or Cr 2+ ), leads to the aggregation of NPs. Thus, for both sensors, the aggregation resulted in reflecting the ion-induced aggregation of the functional Ag-NPs. As mentioned, according to the information obtained from previous studies, 68,77 it is very likely that the SDS was adsorbed on the surface of nanoparticles through hydrophobic bonding with a tail surface/headwater orientation. The high selectivity of the SDS-AgNP toward Mn 2+ can be attributed to two phenomena: (I) the cooperative effect of electrostatic interaction between the negatively charged SO 4groups of SDS and positively charged Mn 2+ ions and (II) the complexation of Mn 2+ ions as hard acids with oxygen donors of SDS. Accordingly, the SDS-AgNPs tend to come closer, which decreases the interparticle distance and makes them aggregated. According to the lectures, 78,79 cyclodextrin stabilizes the surface of AgNPs through chemisorption of hydroxyl groups and the hydrophobic-hydrophobic interactions. Thus, the CD has been applied as both stabilizer and surface functionalizing agents for AgNPs synthesis. Furthermore, as mentioned earlier, in the SDS-CD-AgNPs, the AgNPs are capped by the hostguest inclusion of SDS-CD. According to the obtained results, SDS-AgNPs do not respond to the Cr 2+ ions. In comparison, the SDS-CD-AgNPs demonstrates sensitive and selective colorimetric detection of Cr 2+ ions. Probably, Cr 2+ ion as a harder acid than Mn 2+ ion in addition to bind sulfate groups and oxygen donors of SDS, has tendency to bind ether and hydroxyl groups of cyclodextrin. Thus, selectivity of the SDS-CD-AgNPs toward Cr 2+ can be ascribed to the three facators: (I) The cooperative effect of electrostatic interaction between the negatively charged SO 4 groups of SDS and positively charged Cr 2+ ions; (II) The complexation of Cr 2+ ions as hard acids with oxygen donors of SDS; and (III) The complexation of Cr 2+ ions as hard acids with ether and hydroxyl groups of cyclodextrin as hard base. Owning to these reasons, the SDS-CD-Ag-NPs tend to keep closer, decreasing the interparticle distance and gets aggregated. Based on TEM results, the aggregates of the SDS-CD-AgNPs in the presence of Cr 2+ ions are larger than those of the SDS-AgNPs in the presence of Mn 2+ ions. This can be attributed that the SDS-CD-AgNPs are more agglomerate than SDS-AgNps, Figure 2. In addition, as mentioned, the interaction between SDS and CD surface groups with desired ions are different.  study, each of them has some limitations. For example, colorimetric sensors reported by Wu et al. 50 Chen et al. 49 , He et al. 16 and Badri et al. 53 had a detection limitation with respect to pH conditions for Mn 2+ ions. Compared with these reports for the colorimetric detection of Mn 2+ , our method provided a good detection performance of Mn 2+ for the pH range of 1 to 13. Moreover, according to Table 1, the sensors reported by Annadhasan et al. 52 and Vaibhavkumar et al. 51 suffered from selectivity limitations of detecting Mn 2+ ions. Compared with these works, the sensors in our work are selective only for Mn 2+ detection. In some studies reported by Wu et al. 50 and Vaibhavkumar et al, 51 the AgNPs sensor preparation process is complicated. Our approach to synthesize SDS-AgNPs is not time-consuming and costly. Table 2 summarizes information of Cr 2+ detection method based on the colorimetric sensing probe. According to Table 2, our method provided a better sensing ability for detecting Cr 2+ for the pH range of 1 to 13 and a good response rate. Compared to Rull-Barrull's work, our sensor has a higher sensitivity. As well as, in Rull-Barrull's study, the Ag-NPs sensor preparation process is complicated. 58

Conclusion
In summary, two types of stable silver nanoparticles (AgNPs) were synthesized with SDS and a combination of SDS and CD, which were successfully utilized as a colorimetric sensor for Mn 2+ and Cr 2+ ions at ppm level, respectively. Upon adding desired ions, both AgNPs solutions changed from yellow to orange, which was detected by the naked eye and UV-vis spectroscopy. Indeed, the desired ions can be detected rapidly based on the color change of the system, due to the aggregation of Ag nanoparticles by adsorbing metal ions on their surface.