Highly-luminescent Eu,Sm,Mn-doped CaS up/down conversion nano-particles: application to ultra-sensitive latent fingerprint detection and in vivo bioimaging

Jikai Wang a, Ni He a, Yanli Zhu a, Zhengbin An a, Ping Chen b, Craig A. Grimes c, Zhou Nie *a and Qingyun Cai *a
aState Key Laboratory of Chemo/Biosensing and Chemometrics, Hunan University, Changsha, Hunan 410082, China. E-mail: qycai0001@hnu.edu.cn; niezhou.hnu@gmail.com
bCollege of Life Sciences, Hunan Normal University, Changsha, Hunan 410006, China
cFlux Photon Corporation, Garner, North Carolina 27529, USA

Received 10th October 2017 , Accepted 1st December 2017

First published on 5th December 2017

Due to their unique properties, rare-earth doped upconversion luminescence (UCL) nanomaterials are of considerable scientific interest. Meanwhile, alkaline-earth sulfide materials based on a completely different electron trapping (ET) mechanism demonstrate extremely high UCL efficiencies, which are several dozens of times more than those of conventional fluoride UCL nanomaterials. However, the large particle size, easy hydrolysis, and difficulty in achieving uniform dispersion have precluded bioassay applications. Herein, we have synthesized super-bright Eu,Sm,Mn-doped CaS nanoparticles of ∼30 nm average particle size using a reverse microemulsion technique. The UCL quantum yield was up to nearly 60%. Modification of the nanoparticles with an organic layer allows their stable dispersion throughout aqueous solutions without significant loss of the fluorescence intensity. We demonstrate the application of the novel UCL materials to latent fingerprint detection, deep tissue imaging, and in vivo bioimaging.

While lanthanide doped inorganic materials, such as lanthanide-doped NaYF4, have attracted wide interest for their upconversion luminescence (UCL) properties,1–3 successful application has been limited by drawbacks that include a low upconversion luminescence efficiency and insufficient brightness.4,5 High efficiency UCL emission from europium and samarium doped alkaline-earth sulfides, such as CaS:Eu,Sm and SrS:Eu,Sm, is well known;6,7 notably, the UCL quantum efficiency of CaS:Eu,Sm runs up to 76%,8,9 while conventional UCL nanomaterials based on fluoride, even with core–shell structures, are usually no higher than 5%.1 However, application of these materials has been predominately limited to bulk or powder form.10–12 While it is of considerable interest to explore the application of lanthanide doped alkaline-earth sulfides in biological assays due to their much higher UCL efficiency, bioassay applications have been precluded due to their easy hydrolysis, large particle size, and poor dispersibility.

In this work, Eu, Sm, and Mn doped CaS nanoparticles (ESM-CaS NPs) with an average diameter of ∼30 nm were successfully prepared using a reverse microemulsion strategy. The as-prepared nanoparticles were encapsulated by amphiphilic ligands, as depicted in Fig. 1, allowing them to be uniformly dispersed in water without loss of the fluorescence efficiency; modification with 1-dodecanethiol, through the reaction between calcium cations and the thiol group of 1-dodecanethiol, formed a hydrophobic layer that protected the ESM-CaS NPs from hydrolysis. In order to improve the dispersibility of the ESM-CaS NPs in water, 11-mercaptoundecanoic acid was introduced to the particle surface through the competitive replacement of 1-dodecanethiol with 11-mercaptoundecanoic acid, forming a mixed monolayer with exposed hydrophilic carboxyl groups; these same groups could also be used as active groups for further molecular covalent attachment. Through the co-modification of 1-dodecanethiol and 11-mercaptoundecanoic acid on the ESM-CaS NPs (the encapsulated particles are represented as DT/MUA@ESM-CaS NPs), we successfully solved two major problems: (1) fluorescence quenching in water due to particle hydrolysis, and (2) particle aggregation.

image file: c7cc07790d-f1.tif
Fig. 1 Schematic illustration of the CaS:Eu,Sm,Mn nanoparticle modification.

Transmission electron microscopy (TEM) images show that the as-synthesized ESM-CaS NPs are comparatively uniform with a mean size of ∼30 nm (Fig. 2a). The smaller size in Fig. 2b and c than that in Fig. 2a is due to the higher dispersity. After the mechanical dispersion and surface modification, the aggregated particles are dispersed uniformly. The high-resolution TEM (HRTEM) image and selected-area electron diffraction (SAED) pattern of the ESM-CaS NPs (Fig. 2d) show high crystallinity in accordance with the calcium sulfide lattice. Energy dispersive X-ray analysis reveals the presence of Ca, S and the co-doped elements Eu, Sm, and Mn in the ESM-CaS NPs (Fig. S1, ESI). The DT, DT/MUA-modified and unmodified ESM-CaS NPs show the same X-ray diffraction (XRD) patterns which are consistent with that of the cubic phase of CaS (Fig. 2e and the ESI). Dynamic light scattering (DLS) analysis reveals that the average hydrodynamic diameters of the ESM-CaS NPs, DT@ESM-CaS NPs and DT/MUA@ESM-CaS NPs are 110, 95 and 147 nm, respectively (Fig. S2, ESI). The DT modification formed a hydrophobic surface resulting in a smaller hydrodynamic diameter, while the size increase (∼52 nm) after MUA modification is ascribed to the H2O molecules associated with the DT/MUA@ESM-CaS NPs. Zeta potential measurements (Fig. S3, ESI) show that the ESM-CaS NPs carry a slight negative charge, and become positive after 1-dodecanethiol modification. The further modification with 11-mercaptoundecanoic acid makes the zeta potential shift to −28 mV, indicting the presence of carboxyl groups and the successful immobilization of MUA on the ESM-CaS NPs. These results are consistent with the measured hydrodynamic diameters of the particles. Fourier transform infrared spectroscopy (FT-IR) was employed to determine the specific bonds formed in the nanoparticle modification. As shown in Fig. S4 (ESI), the disappearance of the characteristic thiol group absorption peaks at 2700–2600 cm−1 indicates complexation between calcium cations and the thiol groups of the 1-dodecanethiol and 11-mercaptoundecanoic acids. The bands at 3000–2800 cm−1 are attributed to DT and MUA alkyl chain vibrations. The nuclear magnetic resonance spectroscopy (HNMR) and thermogravimetric analysis (TGA) of the DT@ESM-CaS NPs and DT/MUA@ESM-CaS NPs are presented in Fig. S5 and S6 (ESI), which also indicate binding of DT and MUA.

image file: c7cc07790d-f2.tif
Fig. 2 TEM images and (inset) histograms of the particle size distribution of the (a) ESM-CaS, (b) DT@ESM-CaS, and (c) DT/MUA@ESM-CaS NPs. (d) HRTEM image and (inset) SAED pattern of the ESM-CaS NPs. (e) XRD spectra of the ESM-CaS, DT@ESM-CaS, and DT/MUA@ESM-CaS NPs. (f) DCL excitation spectrum measured at 610 nm (1); DCL spectrum under 355 nm excitation (2); UCL spectrum under 980 nm excitation (3); and (inset) photographs of the ESM-CaS NPs under UV (left) and NIR excitation (right).

The DT/MUA@ESM-CaS NPs show both high UCL and down conversion luminescence (DCL) intensities (Fig. 2f). The DCL excitation spectrum in Fig. 2f (line 1, measured at 610 nm) indicates that the absorption band in the range from 425 to 600 nm is attributed to 4f7(8S7/2) → 4f65d(τ2g) transitions of Eu ions, and covers almost the entire visible region. The absorption peak at 355 nm is ascribed to the 4f7(8S7/2) → 4f65d(εg) transition of Eu ions. The red DCL emission peak at 615 nm (line 2) is the characteristic emission peak of Eu ions corresponding to 4f65d → 4f7 transitions, while the UCL emission peak at 645 nm (line 3) is characteristic of the emission from Eu2+ ions under 980 nm laser excitation.

Fig. 3a shows the UCL spectrum of the ESM-CaS NPs dispersed in ethanol and, for the sake of comparison, NaYF4:Yb,Er,Mn NPs dispersed in cyclohexane,13,14 both at a concentration of 1 mg mL−1 under excitation with a 980 nm laser (power ≈ 200 mW). The ESM-CaS NPs (line 1) show a much higher UCL emission; so much so that it can easily be distinguished with the naked eye (inset of Fig. 3a). In the NaYF4:Yb,Er,Mn NPs, the low absorption cross section of the Yb3+ ions and the non-radiative decay of the excited states lead to a significant decline in the upconversion luminescence efficiency.15,16 The absolute quantum yield (Fig. S7, ESI) of the CaS nanoparticles was measured to be 59.3%. The higher UCL efficiency of the ESM-CaS NPs can be ascribed to their UCL electron-trap mechanism.17–19 As depicted in Fig. 3b, europium ions act as the luminescence center, Eu2+ 8S7/2 state or valence band electrons are promoted into the Eu2+ 4f65d excited state band under visible/UV light, then this is followed by radiative decay and produces DCL (Stokes luminescence), or the electrons are trapped by Sm3+ ions in the 4G5/2 state where the ions act as the electron-trapping center. Meanwhile, Mn2+ plays a critical role in enhancing the trapped electron number in the Sm3+ 4G5/2 state and increasing the UCL emission. Mn2+ might not only facilitate the energy transfer from Eu2+ due to the overlapped excited levels with the Eu2+ 4f65d state, but also improve the probability of excitation transfer to the Sm3+ 4G5/2 state, because the energy of the Mn2+ 4T1 (4G) state is slightly above the Sm3+ 4G5/2 energy level, thus the 4T1 (4G) state will relax by energy transfer to a neighboring Sm3+ ion. As the trap-conduction band energy gap is consistent with the NIR wavelength, under NIR illumination the trapped electrons in the Sm3+ 4G5/2 state are excited into the conduction band and relax rapidly to the lowest Eu2+ 4f65d excited state, then recombine with holes in the Eu ion level (luminescence center) to produce a red UCL emission (anti-Stokes luminescence).

image file: c7cc07790d-f3.tif
Fig. 3 Under 980 nm 200 mW excitation, (a) the UCL spectrum of the ESM-CaS NPs dispersed in ethanol (1), and the NaYF4:Yb,Er,Mn NPs dispersed in cyclohexane (2), both at a concentration of 1 mg mL−1; and (inset) photographs of the ESM-CaS NPs (left) and NaYF4:Yb,Er,Mn NPs (right). (b) Proposed UCL/DCL mechanisms of the ESM-CaS NPs. (c) and (d) are the time-dependent UCL spectra of the ESM-CaS (c) and DT/MUA@ESM-CaS NPs (d) dispersed in PBS solution (pH = 6.8) for 24 h. (e) The corresponding time-dependent UCL intensities of the DT/MUA@ESM-CaS NPs (1) and ESM-CaS NPs (2). (f) The UCL spectrum and (g) photographs of the ESM-CaS (1), DT@ESM-CaS (2), and DT/MUA@ESM-CaS NPs (3) dispersed in anhydrous ethanol, cyclohexane/water and water, respectively.

The naked ESM-CaS NPs are stable only in organic solutions, and are readily hydrolyzed in aqueous solutions, or even air, with a corresponding loss in fluorescence properties (Fig. 3c and e). In contrast, the DT/MUA@ESM-CaS NPs are stable in aqueous solution without a significant decrease in the UCL intensity (Fig. 3d and e). As seen in Fig. 3f and the UCL photographs in Fig. 3g, there is no significant difference in the UCL intensity observed from the ESM-CaS NPs in anhydrous ethanol, the DT@ESM-CaS NPs in cyclohexane/water, and the DT/MUA@ESM-CaS NPs in water.

The UCL efficiency is primarily dependent upon the number of trapped electrons, the amount of doped metal ions, as well as the annealing temperature/time which were optimized by L9(45) orthogonal tests. The selected factors and results are presented in Table S1 and Fig. S8 and S9 (ESI), from which it was determined that the optimal amount of doping was 0.3% Eu, 0.3% Sm and 0.8% Mn, and annealing was at 850 °C/60 min. The UCL intensity (∼9863) achieved for these conditions was found to be superior to all of the others. It appears that the key factors affecting the UCL efficiency are the amount of europium, and the annealing temperature and time. The XRD patterns (Fig. S10, ESI) of the samples synthesized under different conditions show identical peaks at the characteristic angles, implying the same cubic phase of these samples.

The DT/MUA@ESM-CaS NPs were compared with NaYF4:Yb,Er,Mn NPs in applications including latent fingerprint detection and bioimaging. 3-Mercaptopropionic acid (MPA) modified NaYF4 showed a more optimal hydrophilicity than that with MUA modification. Therefore, the hydrophobic NaYF4:Yb,Er,Mn NPs were first converted into hydrophilic NPs by replacing the surface oleylamine with MPA.20,21

For latent fingerprint detection, both the DT/MUA@ESM-CaS NPs and MPA-NaYF4:Yb,Er,Mn NPs were covalently linked with arginine (Arg). FT-IR analysis indicates that arginine was successfully grafted onto the NP surface (Fig. S11, ESI). Here, arginine was designed as the ligand to target the residual amino acids in the latent fingerprint, based on the affinities of the hydrogen bonds and electrostatic interactions. The Arg-DT/MUA@ESM-CaS NPs provided images showing a higher contrast between the ridges and furrows (Fig. 4b and c) than that in the DT/MUA@ESM-CaS NPs (Fig. 4a).

image file: c7cc07790d-f4.tif
Fig. 4 Digital photograph of latent fingerprints stained with the (a) DT/MUA@ESM-CaS NPs, and (b) Arg-DT/MUA@ESM-CaS NPs. (c) Fluctuations in the red value over a few papillary ridges, indicated by the yellow line in (a) and (b).

Fig. 5 presents photographs of latent fingerprints detected on glass, foil, tile and plastic substrates. The Arg-DT/MUA@ESM-CaS NPs display the clearest UCL fingerprint pattern with a high contrast, high selectivity and no background interference on all tested substrates (Fig. 5a), and relatively clear DCL fingerprint patterns (Fig. 5b). This is of particular interest since reported DCL latent fingerprint detection generally suffers from high background interference.22,23 The UCL fingerprint images can be easily distinguished with the naked eye and captured by an ordinary smart-phone; there is no need for expensive image-capturing equipment such as electron multiplying charge-coupled devices. In stark contrast, the UCL fingerprint images obtained using the Arg-MPA-NaYF4:Yb,Er,Mn NPs are almost invisible to the eye (Fig. 5c).

image file: c7cc07790d-f5.tif
Fig. 5 Digital photographs of latent fingerprints imprinted on different substrates including glass, foil, tile and plastic. (a) UCL imaging and (b) DCL imaging of the Arg-DT/MUA@ESM-CaS NPs. (c) UCL imaging of the Arg-MPA-NaYF4:Yb,Er,Mn NPs. (d) DCL imaging of commercial red fluorescent powders. All of the images were acquired with a commonly used smart-phone under 980 nm excitation (power ≈ 200 mW).

The fingerprint images obtained using a commercial DCL red fluorescent powder, Fig. 5d, offer results comparable to those of the Arg-DT/MUA@ESM-CaS NPs DCL images, Fig. 5b. However, the auto-fluorescence and background, especially on the plastic surface, seriously impair fingerprint pattern details necessary for identification. The UCL mode Arg-DT/MUA@ESM-CaS NPs provide clear, highly visible fingerprint images for all of the surfaces.

Tissue imaging depth was investigated by observing the DT/MUA@ESM-CaS NPs arranged to form the letter ‘Z’ laid under chicken breast tissue of various thickness. Ethanol dispersed DT/MUA@ESM-CaS NPs were dropped into a Z-shaped groove, formed by etching a wax plated glass substrate, followed by solvent evaporation. Also, two dots, spaced 4 mm apart, were made in the same manner with the DT/MUA@ESM-CaS NPs or NaYF4:Yb,Er,Mn NPs for deep tissue optical imaging. As shown in Fig. 6a, under excitation of a 980 nm laser (power ≈ 200 mW), the Z image can be captured even at a depth of 10 mm indicating excellent prospects for deep tissue optical imaging. Fig. 6b shows images of the two dots; the DT/MUA@ESM-CaS NP dot is visible through almost 10 mm of tissue, while the NaYF4:Yb,Er,Mn NP dot is barely visible at 4 mm.

image file: c7cc07790d-f6.tif
Fig. 6 (a) UCL images of a letter ‘Z’ pattern incorporating the DT/MUA@ESM-CaS NPs through chicken breast tissue of variable depth. (b) UCL images of two dots incorporating the DT/MUA@ESM-CaS nanoparticles (1) and NaYF4:Yb,Er,Mn NPs (2) through chicken breast tissue of varied depth. All of the images were acquired with a commonly used smart-phone.

Finally, the nanoparticles were applied for in vivo imaging. For this purpose, PEG was linked to the DT/MUA@ESM-CaS NPs. A cell viability higher than 95% indicated minimal cytotoxicity (Fig. S12, ESI). As seen in Fig. 7a, after subcutaneous injection into a Balb/c mouse, a much brighter UCL signal from the PEG-DT/MUA@ESM-CaS NPs is observed as compared with the signal from the MPA-NaYF4:Yb,Er,Mn NPs. Fig. 7b shows a distinguishable signal in the liver region 0.5 h after intravenous injection. The ex vivo UCL image in Fig. 7c exhibits the distribution of the PEG-DT/MUA@ESM-CaS NPs in detail; accumulation of the nanoparticles in the liver, spleen and kidneys is identified by UCL emission.

image file: c7cc07790d-f7.tif
Fig. 7 In vivo imaging of a Balb/c mouse after (a) subcutaneous injection (1 mg mL−1, 0.1 mL) of the DT/MUA@ESM-CaS NPs (1) and the MPA-NaYF4:Yb,Er,Mn NPs (2). (b) Intravenous injection of the PEG-DT/MUA@ESM-CaS NPs (2 mg mL−1, 0.2 mL); and (c) ex vivo UCL imaging 3 h after injection.

In summary, this is the first report on the fabrication and use of electron trap-based UCL materials. CaS NPs with nearly 60% UCL quantum yield were synthesized. Through encapsulation with an organic compound layer, the particles show excellent stability and dispersibility in water. The applications to latent fingerprint detection, deep tissue imaging, and in vivo imaging reveal the great potential of the described materials in biolabeling and bioimaging.

This work was supported by the National Science Foundation of China (No. 21235002, and No. 31670838).

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. W. Zheng, P. Huang, D. Tu, E. Ma, H. Zhu and X. Chen, Chem. Soc. Rev., 2015, 44, 1379 RSC.
  2. X. Li, F. Zhang and D. Zhao, Chem. Soc. Rev., 2015, 44, 1346 RSC.
  3. L. Cheng, C. Wang and Z. Liu, Nanoscale, 2013, 5, 23 RSC.
  4. Y. Liu, D. Tu, H. Zhu and X. Chen, Chem. Soc. Rev., 2013, 42, 6924 RSC.
  5. X. Zhu, Q. Su, W. Feng and F. Li, Chem. Soc. Rev., 2017, 46, 1025 RSC.
  6. L. Lu, X. Zhang, Z. Bai, X. Wang, X. Mi and Q. Liu, Adv. Powder Technol., 2006, 17, 181–187 CrossRef CAS.
  7. W. Jiang, Z. Xu, F. Zhang, X. Zhang and L. Wang, J. Mater. Process. Technol., 2007, 184, 93–96 CrossRef CAS.
  8. W. H. Fan, X. Hou, W. Zhao, X. J. Gao, W. Zou and Y. Liu, Appl. Phys. A: Mater. Sci. Process., 2001, 73, 115–119 CrossRef CAS.
  9. K. Ye and G. Zhao, Acta Photonica Sin., 2001, 30, 487–491 CAS.
  10. H. Nanto, Proc. SPIE, 1999, 3802, 258–265 CrossRef CAS.
  11. C. Chen, K. L. Teo, T. C. Chong, D. M. Newman and J. P. Wu, Phys. Rev. B: Condens. Matter Mater. Phys., 2005, 72, 153–200 Search PubMed.
  12. C. Guo, D. Huang and Q. Su, Mater. Sci. Eng., B, 2006, 130, 189–193 CrossRef CAS.
  13. G. Tian, Z. Gu, L. Zhou, W. Yin, X. Liu, L. Yan, S. Jin, W. Ren, G. Xing and S. Li, Adv. Mater., 2012, 24, 1226 CrossRef CAS PubMed.
  14. Y. Wang, F. Nan, Z. Cheng, J. Han, Z. Hao, H. Xu and Q. Wang, Nano Res., 2015, 8, 2970–2977 CrossRef CAS.
  15. F. Auzel, Chem. Rev., 2004, 104, 139–174 CrossRef CAS PubMed.
  16. S. Heer, K. Kömpe, H. U. Güdel and M. Haase, Adv. Mater., 2004, 16, 2102–2105 CrossRef CAS.
  17. K. Takahashi, K. Kohda, J. Miyahara, Y. Kanemitsu, K. Amitani and S. Shionoya, J. Lumin., 1984, 31–32, 266–268 CrossRef CAS.
  18. Z. Y. He, Y. S. Wang, L. Sun and X. R. Xu, J. Phys.: Condens. Matter, 2001, 13, 3665–3675 CrossRef CAS.
  19. J. Wu, D. Newman and I. Viney, Appl. Phys. B: Lasers Opt., 2004, 79, 239–243 CrossRef CAS.
  20. R. Kumar, M. Nyk, T. Y. Ohulchanskyy, C. A. Flask and P. N. Prasad, Adv. Funct. Mater., 2009, 19, 853–859 CrossRef CAS.
  21. M. Nyk, R. Kumar, T. Y. Ohulchanskyy, E. J. Bergey and P. N. Prasad, Nano Lett., 2008, 8, 3834 CrossRef CAS PubMed.
  22. C. Xu, R. Zhou, W. He, L. Wu, P. Wu and X. Hou, Anal. Chem., 2014, 86, 3279 CrossRef CAS PubMed.
  23. F. Gao, C. Lv, J. Han, X. Li, Q. Wang, J. Zhang, C. Chen, Q. Li, X. Sun and J. Zheng, J. Phys. Chem. C, 2011, 115, 21574–21583 CAS.


Electronic supplementary information (ESI) available. See DOI: 10.1039/c7cc07790d

This journal is © The Royal Society of Chemistry 2018