No title

View Article Online View Journal

Nanoscale Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: A. Borodziuk, M. Baranowski, T. Wojciechowski, R. Minikayev, B. Sikora, D. K. Maude, P. Plochocka and . Kopotowski, Nanoscale, 2020, DOI: 10.1039/D0NR04252H. Volume 10 Number 4 28 January 2018 Pages 1549-2172

Nanoscale rsc.li/nanoscale

This is an Accepted Manuscript, which has been through the Royal Society of Chemistry peer review process and has been accepted for publication. Accepted Manuscripts are published online shortly after acceptance, before technical editing, formatting and proof reading. Using this free service, authors can make their results available to the community, in citable form, before we publish the edited article. We will replace this Accepted Manuscript with the edited and formatted Advance Article as soon as it is available. You can find more information about Accepted Manuscripts in the Information for Authors.

ISSN 2040-3372

PAPER

Shuping Xu, Chongyang Liang et al. Organelle-targeting surface-enhanced Raman scattering (SERS) nanosensors for subcellular pH sensing

Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains.

rsc.li/nanoscale

Nanoscale

Page 1 of 10

View Article Online

DOI: 10.1039/D0NR04252H

Excitation efficiency determines the upconversion luminescence intensity of β -NaYF4 :Er3+ ,Yb3+ nanoparticles in magnetic fields up to 70 T† Anna Borodziuk,a Michał Baranowski,b,c Tomasz Wojciechowski,a Roman Minikayev,a Bożena Sikora,a Duncan K. Maude,b Paulina Plochocka,b,c and Łukasz Kłopotowskia,d

Lanthanide-doped nanoparticles enable conversion of near-infrared photons to visible ones. This property is envisioned as a basis of a broad range of applications: from optoelectronics, via energy conversion, to bio-sensing and phototherapy. The spectrum of applications can be extended if magnetooptical properties of lanthanide dopants are well understood. However, at present, there are many conflicting reports on the influence of the magnetic field on the upconverted luminescence. In this work, we resolve this discrepancy by performing a comprehensive study of β -NaYF4 :Er3+ ,Yb3+ nanoparticles. Crucially, we show that the magnetic field impacts the luminescence only via a Zeemandriven detuning between the excitation laser and the absorption transition. On the other hand, the energy transfer and multiphonon relaxation rates are unaffected. We propose a phenomenological model, which qualitatively reproduces the experimental results. The presented results are expected to lead to design of novel, dual-mode opto-magnetic upconverting nanomaterials.

1 Introduction Upconverting nanoparticles (UCNPs) constitute a class of materials, which are able to absorb multiple infrared photons to emit one at a higher energy. 1–6 Typically, UCNPs are based on fluoride or oxide hosts doped with trivalent lanthanide ions. Usually, Yb3+ ions are employed as sensitizers due to relatively high absorption cross sections for near-infrared light. A long lifetime in the Yb3+ excited state facilitates energy transfer to luminescence activators such as Er3+ , Eu3+ , Tm3+ , or Ho3+ ions. Partially filled 4 f shell of these ions give rise to a multitude of electronic configurations and a ladder of energy levels. Optical transitions between these states lead to luminescence spectra extending from the visible to near-ultraviolet and consisting of narrow lines. Crucially, such lanthanide doped UCNPs are photostable and non-blinking, 2 exhibit a low toxicity, 7 and can be synthesized in a broad range of sizes in the range between single nanometers 8,9 to single microns. 10 The above properties lead to a plethora of proposed applica-

a

Institute of Physics, Polish Academy of Sciences, Al. Lotników 32/46, 02-668 Warsaw, Poland b Laboratoire National des Champs Magnétiques Intenses, UPR 3228, CNRS-UGA-UPSINSA, Grenoble and Toulouse, France c Department of Experimental Physics, Faculty of Fundamental Problems of Technology, Wroclaw University of Science and Technology, Wroclaw, Poland d E-mail: [email protected] † Electronic Supplementary Information (ESI) available: [details of any supplementary information available should be included here]. See DOI: 00.0000/00000000.

tions in fields ranging from bio-detection, 8 bio-imaging, 9 and cancer therapy, 11 via nano-thermometry 10,12 and solar light conversion, 13 to RGB displays, 14 optical storage, 15 and security barcoding 16 . (For reviews, see, e.g., Refs. 3–6.) Apart from these widely discussed applications of UCNPs, there have been proposals for other ones such as aircraft guidance, 17 magnetic field detection, 18–20 , DNA detection, 21 and bi-modal magnetic resonance imaging (MRI) 22,23 . These (lesser discussed) proposed applications require simultaneous control over optical and magnetic properties. 17,24 The dual functionality relies in part on tuning of the lanthanide dopant luminescence with magnetic field. It is clear that the first step to achieving the dual magnetooptical functionality of UCNPs is to obtain an in-depth understanding of the influence the magnetic field exerts on the absorption and luminescence properties. The impact of the magnetic field on both the upconverted luminescence (UCL) and ordinary Stokes luminescence of lanthanide ions was investigated for nanoparticles and bulk crystals. However, the reported results and their interpretations widely vary. Several groups observed a monotonic decrease of the luminescence intensity with increasing the magnetic field. 17,19,25,26 This effect was attributed either to a field-induced population of an optically inactive state, 17,19,27 to a field modulated absorption of the exciting laser light, 25 or to a distortion of the dopant site symmetry and an enhanced cross-relaxation rate. 26 Contrary to these observations, other works demonstrated an increase of the upconverted luminescence with increasing the magnetic field and J our na l Na me, [ y ea r ] , [ vol . ] ,1–8 | 1


Published on 21 September 2020. Downloaded by Carleton University on 9/21/2020 8:27:22 PM.

J our nal Name

Nanoscale

Page 2 of 10 View Article Online

attributed the effect to an increased efficiency of energy transfer between the lanthanide dopants. 28,29 Also, a non-monotonic dependence of UCL intensity on magnetic field was observed. 29 Furthermore, apart from the field modulation of the overall luminescence intensity, various authors reported field-induced relative intensity changes between luminescence bands. This led to a suggestion that the magnetic field influences energy relaxation processes. 28,29 One of the major obstacles in achieving a complete picture of the magnetic field impact on UCL is related to the fact that most of the reports cited above employ only a single excitation wavelength. Moreover, different groups employ different excitation sources. These disparities make comparisons of the reported results difficult since the zero-field excitation efficiencies depends on the detuning between the laser line and the absorption transition. 18,25 In this work, we resolve the long standing discrepancy between the above mentioned results. Our studies are performed on the workhorse upconversion system of NaYF4 nanoparticles doped with Yb3+ and Er3+ ions. We present a comprehensive study of UCL intensity in magnetic fields up to 70 T using excitation wavelengths in a wide range between 960 nm and 990 nm. This allows us to demonstrate that the magnetic field influence on the UCL intensity is solely controlled by the detuning between the laser line and the absorbing transition in sensitizer Yb3+ ions. Moreover, we present evidence that even the ultrahigh magnetic field does not disrupt the relaxation pathways within the activator Er3+ ions, nor does it influence the efficiency of the energy transfer between the sensitizer Yb3+ and activator Er3+ ions. Our results are expected to pave the way toward design of novel, bi-modal, magnetooptical upconverting nanoparticles.

2 Experimental methods 2.1 Chemicals and materials All chemicals were analytical grade and used without additional purification. Trifuoroacetic acid CF3 COOH, (99%) and lanthanide oxides: Y2 O3 (99.99%), Yb2 O3 (99.9%) and Er2 O3 (99.9%) used for lanthanide trifluoroacetate precursors synthesis were purchased from Sigma-Aldrich. Sodium trifluoroacetate precursor Na(CF3 COO, 97%) for UCNPs synthesis was purchased from Acros Organics. Solvent used for UCNPs synthesis and UCNPs dispersion: 1-octadecene (ODE, >95%), oleic acid (OA, 99%) and cyclohexane (anhydrous, 99,5%) were purchased from SigmaAldrich. Ethanol (99.8%) used for UCNPs purification was purchased from Chempur. 2.2 Synthesis of UCNPs Our β -NaYF4 :Er3+ ,Yb3+ UCNPs were synthesised via a thermal decomposition method described in detail elsewhere. 30,31 Lanthanide trifluoroacetate precursors were prepared according to Ref. 32 Then, the 2.5 mmol of Na(CF3 COO), 0.78 mmol of Y(CF3 COO)3 , 0.2 mmol of Yb(CF3 COO)3 and 0.02 mmol of Er(CF3 COO)3 were dissolved in 20 mmol of oleic acid and 20 mmol of 1-octadecene in three-necked flask and stirred under argon flow at room temperature for 30 min to remove oxygen. The mixture was then heated to 120◦ C and kept for 60 min to re-

2|

J our na l Na me, [ y ea r ] , [ vol . ] , 1–8

move residual water. After this step, the solution was heated up to 330◦ C and kept for 30 min. Then, the heater was switched off and the solution was left to cool down to room temperature. The oleate-coated UCNPs were precipitated with ethanol and collected via centrifugation. Subsequently, the UCNPs were washed several times with cyclohexane and ethanol. Finally, the UCNPs were dispersed in cyclohexane for solution-phase measurements. 2.3 Structural characterization The shape and size of the synthesised UCNPs, was examined using scanning electron microscopy (SEM, Auriga Cross-Beam Workstation, Zeiss). Crystallographic structure was examined via powder X-ray diffraction using X’Pert Pro MPD (Panalytical) diffractometer. 2.4 Spectroscopic measurements Solution-phase measurements of upconverted luminescence intensity were performed on the sample with a concentration of about 2.4 mM in a quartz cuvette. A diode laser (Lumics LuOcean) emitting at ∼ 980 nm was used as the excitation source. The signal was collected perpendicular to the excitation and coupled to a 100 µm diameter multimode optical fiber. The other end of the fiber was placed at the entrance plane of a 500 mm focal length monochromator (Andor Shamrock) equipped with a 300 grooves/mm grating. The signal was detected with a CCD camera (Andor iDus). For the measurements of the luminescence excitation spectra, the sample was drop-casted onto a clean silicon wafer. A pulsed, tunable Ti:sapphire laser (Coherent Chameleon Ultra II, pulse width ∼ 300 fs) was used as the excitation source. The laser beam was focused onto the sample to a spot with ∼ 0.5 mm in diameter. The signal was collected in back-scattering geometry and focused onto the entrance slit of a 500 mm focal length monochromator (Princeton) equipped with a 600 grooves/mm grating and detected with a liquid nitrogen cooled CCD camera (Princeton). Measurements in pulsed magnetic fields were conducted in back scattering geometry with a sample drop-casted at the end of a low temperature insert, which was placed in a liquid helium cryostat put in the center of a pulsed magnet bore. The magnetic field is generated by discharging a capacitor bank through a resistive coil providing pulsed fields with amplitudes up to 68 T and a temporal pulse width of ' 35 ms. The Ti:sapphire laser beam was delivered to the sample with a multimode fiber. The emitted light was collected by a fiber bundle, surrounding the excitation fiber. The luminescence signal was detected with a 300 mm focal length monochromator (Princeton) coupled with a liquid nitrogen cooled CCD camera (Princeton).

3 Results and discussion A scanning electron micrograph shown in Fig. 1(a) reveals an ensemble of relatively homogeneous nanoparticles, slightly elongated with a diameter/height aspect ratio of about 0.9± 0.1. The size histogram presented in the inset to Fig. 1(a) shows that the average nanoparticle size amounts to 42.7± 2.1 nm. The hexagonal cross-sections seen in Fig. 1(a) suggest that the NaYF4 UCNPs



DOI: 10.1039/D0NR04252H

Page 3 of 10

Nanoscale View Article Online

Fig. 1 (a) Scanning electron micrograph of the investigated sample showing hexagonal nanoparticles. Inset: size histogram obtained from the SEM image. (b) Measured powder X-ray diffraction pattern (red) compared with Inorganic Crystal Structure Database (ICSD) #259179 reference file (black) revealing hexagonal (β -phase) NaYF4 nanoparticles. (c) Upconverted luminescence spectra measured for various excitation powers. (d) Schematic of energy levels of Yb3+ sensitizer ions an Er3+ activator ions together with energy transfer and relaxation pathways. Black thin/dashed arrow – excitation/deexcitation. Yellow arrows – energy transfers. Blue wavy lines – multiphonon energy relaxation. Blue and red dashed lines – cross-relaxation (CR) and back energy transfer (BET), respectively. Thick green and red lines – light emission. (e) Double logarithmic plot of the excitation power dependence of the upconverted luminescence intensity, integrated in three spectral bands around the values given in the legend. Lines are linear fits. Numbers denote fitted slopes. (f) Excitation spectra of the luminescence signals integrated in the three spectral bands.

crystallize in the hexagonal, β -phase. This attribution is corroborated by powder X-ray diffraction (XRD) measurements. The XRD pattern presented in Fig. 1(b) reveals peaks that are identified using the Inorganic Crystal Structure Database (ICSD) reference file #259179 manifesting that, indeed, the UCNPs crystallized in the β -phase. Optical characterization was performed on cyclohexane suspensions of the UCNPs, employing a diode laser emitting at 980 nm as the excitation source. In Fig. 1(c), we show UCL spectra for various excitation powers. The spectra consist of three bands – two in the green part of the visible spectral range, centered at 525 nm and at 545 nm, and one in the red, at 660 nm. As widely discussed for both bulk and nanocrystalline materials doped with Er3+ and Yb3+ ions, 1,2,33,34 these bands are due to the optical transitions between Er3+ manifolds excited via upconversion of near-infrared (NIR) photons absorbed by the Yb3+ ions and subsequent energy transfer. The scheme of the processes involved is depicted in Fig. 1(d). The NIR laser light excites Yb3+ from the 2 F7/2 ground state to the 2 F5/2 excited state (black thin arrow). These excited ions exhibit a lifetime in excess of milliseconds, i.e., an order of magnitude longer than the lifetimes of the Er3+ ions luminescence 34 and thus can be viewed as energy reservoirs. Upon a Förster-Dexter type energy transfer, 5,35 the energy can be transferred to a neighboring Er3+ ion in a process of simultaneous deexcitation of Yb3+ and excitation of Er3+ from the ground state 4 I15/2 to the excited state 4 I11/2 (thin yellow arrows). Occurrence of consecutive energy transfer steps allows to climb up the ladder of the excited Er3+ states, a process which is countered by multiphonon energy relaxation (wavy lines) and accompanied by back energy transfer (red dashed lines) and various cross-relaxations (blue dashed lines). The competition between all these processes determines the relative amplitudes of

the green and red UCL bands. 36–39 As depicted in Fig. 1(d), the green bands at 525 nm and 545 nm originate from the transitions from, respectively, 2 H11/2 and 4 S3/2 states to the 4 I15/2 , while the red band is due to the transition from 4 F9/2 to 4 I15/2 . The excitation power dependence of the three UCL bands is plotted in Fig. 1(e). We find that the UCL intensity I scales with the excitation power P as I(P) ∝ Pn . Both green bands exhibit an approximately quadratic dependence on the laser power described by the exponent n ≈ 2 – see Fig. 1(e). This indicates that absorption of two NIR photons is required to produce one green photon in emission. Indeed, it is widely accepted 34,36,37 that the green UCL of Er3+ ions originates from two energy transfer steps: first one exciting the Er3+ ion to the 4 I11/2 state, the second one exciting it to the 4 F7/2 . Subsequent phonon relaxation brings the system to the green-emitting 2 H11/2 and 4 S3/2 states. On the other hand, the intensity of the red UCL band exhibits a steeper excitation power dependence with n ≈ 2.6. Therefore, the relevant excitation pathways have to involve absorptions of two and three NIR photons. The mechanism leading to the red UCL band of Er3+ ions is a subject of an ongoing debate. 38,39 Since the employed solvent (cyclohexane) does not contain O-H bonds, we assume that the phonon relaxation between the 4 I11/2 and 4 I13/2 is inefficient. 37 Also, we disregard the cross-relaxation mechanisms, which would only be efficient in the presence of a Er3+ ion clusters or concentration exceeding ∼ 10%. We therefore conclude that the red-emitting 4 F9/2 state is populated via two channels. The first one is the phonon relaxation from the green emitting 2 H11/2 and 4 S3/2 states – a process requiring absorption of two NIR photons. The second one is a back energy transfer from Er3+ to Yb3+ ion resulting in the excitation E of the latter – a transition between the ion states Yb3+ , Er3+ : J our na l Na me, [ y ea r ] , [ vol . ] ,1–8 | 3



DOI: 10.1039/D0NR04252H

Nanoscale


λe x = 9 8 0 n m

(c ) 0 3

T

6

T

1 1 2 1 3 9 6 3

5 5 0

6 0 0

6 5 0

T T T T

7 0 0

W a v e le n g th ( n m ) λe x = 9 6 0 n m

(b )

(d ) λe x =

T

0 T 6

T

T

3

1 1 T 2 1 T

9 9 0 n m

λe x = 9 9 0 n m

9 8 5 n m 9 8 5 n m

9 8 0 n m 9 7 5 n m 9 7 0 n m 9 6 5 n m

9 8 0 n m 9 7 5 n m 9 7 0 n m 9 6 5 n m

9 6 0 n m

9 6 0

6 0 0

6 5 0

W a v e le n g th ( n m )

7 0 0

9 9 0

5 4 5 n m 6 6 0 n m

4 0

λe m = 5 4 5 n m

6 3 T

5 5 0

9 7 5

8 0

3 9 T

5 0 0

λe x ( n m )

9 6 0 n m

B m a x (T )

(a )

5 0 0

To better understand this behavior, we performed UCL measurements in magnetic fields under excitation wavelengths ranging from 960 nm to 990 nm. The UCL intensities were spectrally integrated in the red and green bands. The resulting magnetic field dependences of the UCL intensities normalized to the maximum value are plotted in Figs. 2(c) and 2(d), respectively.

N o r m a liz e d U C L In te n s ity ( a r b . u n its )

In te n s ity ( a r b . u n its )

In order to further probe the influence of the excitation efficiency on the UCL intensity, we measured the luminescence excitation spectra (see ESI, Fig. S1†). To perform these studies under similar conditions to the magnetic field measurements (see below), the sample was drop-casted onto a clean silicon wafer. As an excitation source, we employed a pulsed, femtosecond Ti:sapphire laser and, consequently, the excitation line has a spectral width of about 9 nm (100 cm−1 ). In Fig. 1(f), we plot the intensities of the three UCL bands as a function of the excitation wavelength. The linewidths of the observed spectra are a convolution of the (approximately Gaussian) laser line and the excitation spectrum. The peak at 977 nm corresponds to the excitation of the Yb3+ ions from the 2 F7/2 ground state to the 2 F5/2 excited state, as discussed above. 34 The shapes of the spectra obtained for the three UCL bands are roughly similar. However, the linewidth of the red band excitation spectrum is noticeably narrower: its wings fall off faster than for the green bands. This is another indication of its higher sensitivity to the excitation efficiency: since the creation of a red photon require absorption of more NIR photons than the creation of a green photon, the detuning of the laser line from the absorption resonance has a stronger effect on the red band than it has on the green band. We will take advantage of this effect below in the analysis of the magnetic field influence on the UCL intensity.

In te n s ity ( a r b . u n its )


photon process as excitation of an Er3+ ion to the 4 G11/2 state requires three energy transfers from the excited Yb3 ions. The exponent value n ≈ 2.6 indicates that the two- and three-photon processes are approximately equally efficient at the employed excitation powers.

The measurements in magnetic field were performed on the sample drop-casted at the end of a low temperature insert. The same Ti:sapphire laser was used as the excitation source. The magnetic field up to 68 T was generated by discharging a capacitor bank through a resistive coil. Since the magnetic field pulse width is about 35 ms, the UCL integration time is limited to 3 ms. It is therefore important to provide a sufficiently strong UCL signal. To this end, the sample was cooled down with liquid nitrogen to about 100 K, as it was shown that in this temperature range the UCNP luminescence intensity exhibits a maximum. 40,41 Since the temperature is lower than for room temperature measurements reported in Fig. 1(c), the shorter wavelength green band is absent – see Figs. 2(a) and 2(b). This is due to efficient relaxation between the 2 H11/2 and 4 S3/2 states, which leads to their thermal occupations 42 (a property exploited in ratiometric thermometry 12 ). In Fig. 2(a), we plot UCL spectra collected for various magnetic fields under excitation at 980 nm, i.e., roughly in resonance with the Yb3+ transition. In this case, maximum UCL is observed at about 0 T and with increasing magnetic field the intensity decreases. On the contrary, for excitation at 960 nm, i.e. significantly detuned from the resonance, Fig. 2(b), the UCL intensity increases with the field. These results indicate that the excitation wavelength is a crucial parameter controlling the behavior of UCL in magnetic fields.

0

1 0

2 0

λe m = 6 6 0 n m 3 0

4 0

5 0

M a g n e tic F ie ld ( T )

6 0

7 0

8 0

0

1 0

2 0

0

3 0

4 0

5 0

6 0

7 0

8 0

M a g n e tic F ie ld ( T )

Fig. 2 (a) – (b): Upconverted luminescence spectra measured at 100 K for various magnetic fields for two excitation wavelengths: (a) 980 nm and (b) 960 nm. (c) – (d) Magnetic field dependence of upconverted luminescence intensities of the (c) green band and (d) red band normalized to the maximum value and shifted vertically for clarity. Points represent experimental data. Lines are guides to eye (fitted Gaussian functions). Empty triangular arrows denote positions of intensity maxima. The inset in (d) shows excitation wavelength dependence of the magnetic field Bmax , corresponding to the maximum luminescence intensity for the two spectral bands.

4|



DOI: 10.1039/D0NR04252H

E E 2 F7/2 ,4 G11/2 → 2 F5/2 ,4 F9/2 , see Fig. 1(d). 39 This is a three

Page 5 of 10


(a )

U C L In te n s ity ( a r b . u n its )

resulting changes in excitation efficiency are then imprinted in the measured UCL intensities. The field where UCL exhibits a maximum, Bmax , reflects the field at which the resonance is attained. To describe in detail the influence of the Zeeman effect on the excitation efficiency it is necessary to take into account: (i) the crystal-field-induced mixing of the Yb3+ excited states (the spin structure mentioned above), (ii) the values of the Landé g-factors (iii) the g-factor anisotropy, (iv) oscillator strengths of the optical transitions between the 2 F7/2 and 2 F5/2 manifolds, and (v) averaging of the absorption cross-sections over the angle between the quantization axis and the magnetic field direction. Since many of the input parameters are not experimentally determined at present, the modelling of the Zeeman effect requires a separate theoretical study, one clearly beyond the scope of the present report. In order to support our conclusion that the Zeeman-induced changes of the excitation efficiency determine the observed magnetic field dependence of UCL intensity, we developed a simplified, phenomenological model – see Fig. 3(a) and ESI† for details. Namely, we consider optical transitions between two spin1/2 states of Yb3+ ions, i.e., neglect all but the lowest energy mJ states. This assumption is supported by the fact that under our experimental conditions with the sample at 100 K, the ground state population of the 2 F7/2 manifold is at least 2 times larger than the population of the first excited state. Applying a magnetic field B along the quantization axis results in a Zeeman splitting of the transition energies depicted schematically in Fig. 3(a) and given by EYb = 1/2gµB B, where g denotes an effective Landé g-factor and µB is the Bohr magneton. We assume that the absorption transitions are described by Lorentzian lineshapes with a full width at half maximum (FWHM) linewidth ΓL . The UCL intensity of the green band is then proportional to the squared overlap of the absorption spectrum with the Gaussian laser spectrum with a FWHM ΓG = 100 cm−1 determined experimentally. In the discussion below, we refer to excitation wavelength implying the central laser wavelength. We compare the experimental data with the model calculations in Figs. 3(b) and 3(c). In Fig. 3(b), we plot the normalized UCL intensities for the green band IGreen obtained in the experiment for three selected excitation wavelengths. In Fig. 3(c), we plot

9 8 0 n m 9 7 0 n m 9 6 0 n m

1 .0

1 .0

0 .5

0 .5 9 8 0 n m 9 7 0 n m 9 6 0 n m

0 .0 0 2 0 4 0 6 0 8 0 M a g n e tic F ie ld ( T ) 0

2 0 4 0 6 0 M a g n e tic F ie ld ( T )

(b )

(c )

0 .0 8 0

0

2 0 4 0 6 0 M a g n e tic F ie ld ( T )

8 0

Fig. 3 Model analysis of the magnetic field influence on UCL intensity. (a) A schematic showing the relationship between the laser spectrum (left) and the Zeeman split absorption transitions between two spin-1/2 states (see text and ESI† for details). The Gaussians represent the laser spectra, the lines denote the transition energies, and the stars mark the magnetic fields of maximum excitation efficiency. (b) Green band UCL intensities replotted from Fig. 2 for the selected excitation wavelengths. (c) Calculated UCL intensities for the selected excitation wavelengths.

J our na l Na me, [ y ea r ] , [ vol . ] ,1–8 | 5


T r a n s itio n E n e r g y ( a r b . u n its )

(For non-normalized UCL intensities as a function of the magnetic field, see ESI† Fig. S2.) In general, the magnetic field influences the green and red bands in a similar way: UCL maximum occurs roughly at the same field Bmax indicated by the arrows. More importantly, Bmax changes with the excitation wavelength: for 960 nm it around the highest field we can produce, i.e., about 70 T. With increasing the laser wavelength Bmax initially decreases and for resonant excitation at 980 nm Bmax < 10 T (see also ESI† Fig. S3). Further increase in the excitation wavelength results in increasing of Bmax . This behavior is depicted in the inset to Fig. 2(d) in which Bmax is plotted against the excitation wavelength. The results presented in Fig. 2 unambiguously show that the magnetic field dependence of UCL intensities results from tuning of the absorption resonance with respect to the laser wavelength. We will first discuss the physics involved in this process and then provide a simple phenomenological model capturing the essential features of our results. Next, we will show that the magnetic field does not impact neither the energy transfer efficiency between the Yb3+ and Er3+ ions, nor does it influence the energy relaxation within the Er3+ ion states. The Yb3+ and Er3+ dopants in the β -NaYF4 host lattice experience electrostatic interaction with the neighboring atoms. This crystal field potential lifts the degeneracy of the J-multiplets and gives rise to the fine structure of the observed luminescence bands. 42 Since both Yb3+ and Er3+ contain an odd number of electrons on the f -shell, according to the Kramers theorem, the ground state is (at least) doubly degenerate. Assuming that the dopants occupy yttrium sites with C3h symmetry, the ground state of Yb3+ ions is a mJ = ±1/2 doublet, separated from the first excited mJ ± 3/2 state by an energy of about 70 cm−1 as determined by luminescence measurements 43 and calculations. 44 The impact of the crystal field on the excited 2 F5/2 manifold is less known. The reported energy splitting between the first two crystal-fieldsplit states is 40 cm−1 . 34 However, the spin structure and the energy splittings of the remaining states are unknown. The magnetic field lifts the degeneracy with respect to mJ quantum number resulting in Zeeman shifts of the energy levels. Consequently, the absorption resonance is detuned from the excitation laser or tuned to the resonance, depending on the laser wavelength. The

L a s e r P h o to n E n e r g y ( a r b . u n its )


DOI: 10.1039/D0NR04252H

Nanoscale


6|


(a )

5 .5 L o g R e d B a n d I n t e n s it y

5 .0

F itte d s lo p e = 1 .2 1 ± 0 .0 5

4 .5 4 .0 3 .5

d a ta fro m e x c ita tio n s p e c tru m

3 .0 2 .5 3 .0

3 .5

4 .0

4 .5

5 .0

5 .5

L o g G r e e n B a n d In te n s ity 5 .5

(b )

5 .0

F itte d s lo p e = 1 .2 7 ± 0 .0 5

4 .5

S lo p e = 1 .2 1

9 6 0 9 6 5 9 7 0 9 7 5 9 8 0 9 8 5 9 9 0

4 .0 3 .5

d a ta fro m m a g n e tic fie ld

3 .0 2 .5 3 .0

3 .5

4 .0

4 .5

n m n m n m n m n m n m n m

5 .0

L o g G r e e n B a n d In te n s ity Fig. 4 Dependence of the red band UCL intensity on the green band UCL intensity. (a) Data from the excitation spectrum shown in Fig. 1(f). (b) Data from the UCL measurements in magnetic field for the excitation wavelengths given in the legend.

Zeeman splitting of the absorbing Yb3+ ion states. Compared to that effect, the darkening of the UCL due to field-induced population of an optically inactive substate of the erbium 4 S3/2 manifold, proposed in Ref. 27, has a negligible influence. We also do not observe any impact of the magnetic field on the energy transfer efficiency nor on multiphonon relaxation processes proposed in Refs. 28,29 The robustness of the energy transfer against Zeeman detuning can be interpreted within the Förster-Dexter theory. Within this theory, the energy transfer rate WET ∝ fYb fEr s0 r−6 , where fYb and fEr denote the oscillator strengths of the emitting and absorbing transitions related to, respectively, Yb3+ and Er3+ ions, s0 is the overlap integral between the relevant emission and absorption spectra, and r is the distance between Yb3+ and Er3+ ions. 4,5,35,45 From the inspection of the UCL spectra shown in Fig. 2(a) and 2(b), we conclude that the Zeeman splittings of the observed transitions are significantly smaller than both the UCL line broadening and the crystal-field-induced splitting. Thus, the magnetic field influence on the shape of the observed UCL spectra is small, in agreement with prior reports. 28,29 We argue that the same conclusion holds for the absorption and emission spectra which enter into s0 , e.g., those related to transitions 2 F7/2 →2 F5/2 in Yb3+ ions or 4 I15/2 →4 I11/2 in er ions. As a result, the influence of the magnetic field on s0 is negligible. The weak influence of the field on the UCL spectra also indicates that the symmetries of the


the UCL values calculated for the same excitation wavelengths. For approximately resonant excitation at 980 nm, the calculated UCL intensity decreases monotonically with increasing the magnetic field. For slightly off-resonance excitation at 970 nm, the calculated initial, zero field UCL intensity is decreased resulting from decreased excitation efficiency (see Fig. 1(f)) and then increases to attain a maximum at 38 T. For strongly off-resonant excitation, the initial intensity is further reduced and the calculated maximum intensity is shifted to about 70 T. Thus, the calculated field dependences of the UCL intensities reproduce qualitatively the behavior observed experimentally. This allows us to conclude that the main factor determining the UCL intensities of the green band is the magnetic field dependent excitation efficiency of the Yb3+ ions. Note that we cannot expect a quantitative agreement between the simple model and the experimental results since, as discussed above, the model neglects the fine structure of the absorption line and strongly simplifies the treatment of the Zeeman effect. Measurements of magnetoabsorption spectra would provide the information needed to account for these subtle effects. However, due to strong light scattering and relatively low absorption coefficients of our samples, such experiments require long integration times and, thus, might not be infeasible in pulsed magnetic fields. Nevertheless, these measurements would provide a support for our interpretation of the magnetoluminescence data. We will now take a closer look at the properties of the red band in order to determine whether the magnetic field influences the energy transfer upconversion and relaxation pathways within the Er3+ states. In Fig. 4(a), we plot the red band intensity IRed as a function of IGreen as measured for different excitation wavelengths at zero magnetic field (i.e., we replot the data from Fig. 1(f) and intensities of bands at 525 nm and at 545 nm are summed to obtain IGreen ). We observe a superlinear increase of n and fitting the data from Fig. the red band intensity: IRed ∝ IGreen 4(a) yields n = 1.21±0.05. This behavior is in excellent agreement with the measured power dependence of the UCL intensities (see Fig. 1(e)), from which we infer n = 1.2. In Fig. 4(b), we plot IRed versus IGreen with the values taken from the measurements in magnetic fields for all excitation wavelengths (i.e., we replot the data from Figs. 2(c)-(d)). Crucially, fitting the power law dependence of IRed on IGreen yields in this case n = 1.27 ± 0.05. Thus, within the experimental error, the dependence of IRed on IGreen scales with the same exponent regardless of whether we change the excitation power, the excitation wavelength, or the magnetic field. This result demonstrates that detuning the laser from the resonance at zero field has the same effect as detuning the resonance with the magnetic field from the laser. We therefore conclude that the influence of the magnetic field on the intensity ratio between red and green bands is equivalent to detuning the excitation wavelength. The results shown in Fig. 4 indicate that the magnetic field does not influence neither the energy transfer between the Yb3+ and Er3+ ions, nor does it affect the energy relaxation pathways within the Er3+ ions. The results presented above allow us to conclude that magnetic fields up to 70 T influence the UCL intensities of β NaYF4 :Yb3+ ,Er3+ nanoparticles predominantly via the modification of the excitation efficiency, which occurs as a result of the

L o g R e d B a n d I n t e n s it y


DOI: 10.1039/D0NR04252H

Page 7 of 10


wave functions of the states participating in optical transitions, and, consequently, the magnitudes of fYb and fEr are weakly affected. Hence the negligible impact of the magnetic field on WET . Corroboration of this conclusion could be provided by measurements of UCL dynamics, which allow to estimate WET . 36,38 However, such experiments require photon acquisition times of at least tens of seconds and thus can only be performed at static fields limited to about 30 T. Finally, we note that the multiphonon relaxation rate WMPR scales with the energy separation between the initial and final states ∆E as: 5,45 WMPR ∝ exp(−α∆E), where α is a constant specific to the host matrix. The relaxation processes influencing the red band intensity, i.e., between the 4 S3/2 and 4 F9/2 and between 4 I11/2 and 4 I13/2 states (see Fig. 1(d) and the discussion above), occurs across an energy gap ∆E ≈ 3000 cm−1 . This value is significantly larger than the observed Zeeman shifts, which allows us to conclude that the magnetic field influence on the multiphonon relaxation is small.

4 Conclusions and outlook In summary, we have presented evidence that the magnetic field influences upconversion luminescence from β -NaYF4 :Er3+ ,Yb3+ nanoparticles by modifying the excitation efficiency. The modification arises as a consequence of Zeeman detuning of the absorption transition in Yb3+ ions from the excitation laser line. The magnitude of this detuning determines the behavior of luminescence intensity in magnetic field, which depends on the excitation wavelength. Furthermore, we have shown that the magnetic field does not disrupt the energy transfer paths between the Yb3+ and Er3+ ions, nor does it strongly influence the phonon relaxation rates. As a consequence, the intensity ratio between the red and green luminescence bands is only affected via the Zeeman-modified excitation efficiency. These results resolve a long standing discrepancy between results presented in various reports. 17,19,25,25,28,29 We predict that the our conclusions will hold for Yb3+ and Er3+ ions doped into matrices such as CaF2 , YVO4 , or CsCdBr3 . In these materials, the crystal field effect on the dopant ions in these materials is stronger than in NaYF4 . 42,46,47 making WET even less affected by the magnetic field. We note however that our conclusions might not be applicable to other lathanide dopant ions. For example, the luminescence spectra of Eu3 + ions embedded into NaYF4 was shown to exhibit a stronger dependence on the magnetic field than the Er3+ ion spectra presented here. 26 We expect that the presented results will form a basis for design of novel upconverting materials with dual magnetooptical functionalities. One particular path that in our opinion deserves attention is employing gadolinum-based hosts such as Gd2 O3 or NaGdF4 . These matrices exhibit paramagnetic properties related to the Gd3+ ions. The interaction between these ions and the lanthanide dopants providing the optical properties should enable Zeeman-tuning of the optical properties at much smaller fields 17,48 in the same fashion as in transition-metal-doped diluted magnetic semiconductors (DMS). 49 Contrary to DMS however, the physics behind the interaction between localized paramagnetic and non-magnetic lanthanide ions remains to date unexplored.

Conflicts of interest There are no conflicts to declare.

Acknowledgements The authors appreciate the support from National Science Centre Poland within the OPUS program (grant nos. 2019/33/B/ST3/01915 and 2019/35/B/ST3/04235) and also the support from Agence Nationale de la Recherche ANR-10LABX-0037-NEXT, EPSRC (Grant No EP/N01085X/1). This study has been (partially) supported through the grant NanoX n◦ ANR17-EURE-0009 in the framework of the "Programme des Investissements d’Avenir". Ł. K. thanks Mateusz Goryca from University of Warsaw for fruitful discussions.

Notes and references 1 S. Heer, K. Kömpe, H.-U. Güdel and M. Haase, Adv. Mater., 2004, 16, 2102–2105. 2 S. Wu, G. Han, D. J. Milliron, S. Aloni, V. Altoe, D. V. Talapin, B. E. Cohen and P. J. Schuck, PNAS, 2009, 106, 10917–10921. 3 A. Gnach and A. Bednarkiewicz, Nano Today, 2012, 7, 532– 563. 4 B. Zhou, B. Shi, D. Jin and X. Liu, Nat. Nanotechnol., 2015, 10, 924. 5 A. Nadort, J. Zhao and E. M. Goldys, Nanoscale, 2016, 8, 13099–13130. 6 S. Wen, J. Zhou, K. Zheng, A. Bednarkiewicz, X. Liu and D. Jin, Nat. Commun., 2018, 9, 1–12. 7 K. Liu, X. Liu, Q. Zeng, Y. Zhang, L. Tu, T. Liu, X. Kong, Y. Wang, F. Cao, S. A. Lambrechts et al., ACS nano, 2012, 6, 4054–4062. 8 J. Zhao, D. Jin, E. P. Schartner, Y. Lu, Y. Liu, A. V. Zvyagin, L. Zhang, J. M. Dawes, P. Xi, J. A. Piper et al., Nat. Nanotechnol., 2013, 8, 729–734. 9 D. J. Gargas, E. M. Chan, A. D. Ostrowski, S. Aloni, M. V. P. Altoe, E. S. Barnard, B. Sanii, J. J. Urban, D. J. Milliron, B. E. Cohen et al., Nat. Nanotechnol., 2014, 9, 300. 10 P. Rodríguez-Sevilla, Y. Zhang, P. Haro-González, F. SanzRodríguez, F. Jaque, J. G. Solé, X. Liu and D. Jaque, Advanced Materials, 2016, 28, 2421–2426. 11 N. M. Idris, M. K. Gnanasammandhan, J. Zhang, P. C. Ho, R. Mahendran and Y. Zhang, Nat. Medicine, 2012, 18, 1580– 1585. 12 F. Vetrone, R. Naccache, A. Zamarrón, A. Juarranz de la Fuente, F. Sanz-Rodríguez, L. Martinez Maestro, E. Martín Rodriguez, D. Jaque, J. García Solé and J. A. Capobianco, ACS Nano, 2010, 4, 3254–3258. 13 W. G. van Sark, J. de Wild, J. K. Rath, A. Meijerink and R. E. Schropp, Nano. Res. Lett., 2013, 8, 1–10. 14 J. M. Meruga, A. Baride, W. Cross, J. J. Kellar and P. S. May, J. Mater. Chem. C, 2014, 2, 2221–2227. 15 C. Zhang, H.-P. Zhou, L.-Y. Liao, W. Feng, W. Sun, Z.-X. Li, C.-H. Xu, C.-J. Fang, L.-D. Sun, Y.-W. Zhang et al., Adv. Mater., 2010, 22, 633–637. 16 Y. Lu, J. Zhao, R. Zhang, Y. Liu, D. Liu, E. M. Goldys, X. Yang, J our na l Na me, [ y ea r ] , [ vol . ] ,1–8 | 7



DOI: 10.1039/D0NR04252H

Nanoscale


17 18


19

20 21 22 23

24 25 26 27 28 29 30

31

8|

P. Xi, A. Sunna, J. Lu et al., Nat. Photonics, 2014, 8, 32. Y. Liu, D. Wang, J. Shi, Q. Peng and Y. Li, Angew. Chem., Int. Ed., 2013, 52, 4366–4369. R. Valiente, M. Millot, F. Rodríguez, J. González, J. Broto, S. George, S. Garcia-Revilla, Y. Romanyuk and M. Pollnau, High Pres. Res., 2009, 29, 748–753. V. K. Tikhomirov, L. F. Chibotaru, D. Saurel, P. Gredin, M. Mortier and V. V. Moshchalkov, Nano Lett., 2009, 9, 721– 724. Z.-W. Ma, J.-P. Zhang, X. Wang, Y. Yu, J.-B. Han, G.-H. Du and L. Li, Opt. Lett., 2013, 38, 3754–3757. L. Wang and Y. Li, Chem. Comm., 2006, 2557–2559. J. Zhou, Y. Sun, X. Du, L. Xiong, H. Hu and F. Li, Biomaterials, 2010, 31, 3287–3295. X. Chen, J. Vanacken, Y. Liu, L. Meng, J. Ge, J. Hu, S. Wu, Z. Zhong and V. V. Moshchalkov, IEEE J. Sel. Top. in Quant. Elect., 2018, 25, 1–6. G. Wang, Q. Peng and Y. Li, Acc. Chem. Res., 2011, 44, 322– 332. J. Zhang, Z. Ma, J. Han, G. Du, Y. Zhou, L. Li and Z. Cheng, J. lumin., 2013, 144, 53–56. P. Chen, J. Zhang, B. Xu, X. Sang, W. Chen, X. Liu, J. Han and J. Qiu, Nanoscale, 2014, 6, 11002–11006. L. Chibotaru, V. Tikhomirov, D. Saurel and V. Moshchalkov, J. Appl. Phys., 2009, 106, 053502. P. Chen, H. Jia, Z. Zhong, J. Han, Q. Guo, J. Zhou, X. Liu and J. Qiu, J. Mater. Chem. C, 2015, 3, 8794–8798. Q. Xiao, Y. Zhang, H. Zhang, G. Dong, J. Han and J. Qiu, Sci. Rep., 2016, 6, 31327. H.-X. Mai, Y.-W. Zhang, R. Si, Z.-G. Yan, L.-d. Sun, L.-P. You and C.-H. Yan, Journal of the American Chemical Society, 2006, 128, 6426–6436. P. Kowalik, D. Elbaum, J. Mikulski, K. Fronc, I. Kami´ nska, P. C. Morais, P. E. De Souza, R. B. Nunes, F. H. Veiga-Souza, G. Gruzeł et al., RSC Adv., 2017, 7, 30262–30273.


32 J. E. Roberts, JACS, 1961, 83, 1087–1088. 33 F. Auzel, Chem. Rev., 2004, 104, 139–174. 34 J. F. Suyver, J. Grimm, M. Van Veen, D. Biner, K. Krämer and H.-U. Güdel, J Lumin., 2006, 117, 1–12. 35 D. L. Dexter, J. Chem. Phys., 1953, 21, 836–850. 36 J. Zhao, Z. Lu, Y. Yin, C. McRae, J. A. Piper, J. M. Dawes, D. Jin and E. M. Goldys, Nanoscale, 2013, 5, 944–952. 37 C. Würth, M. Kaiser, S. Wilhelm, B. Grauel, T. Hirsch and U. Resch-Genger, Nanoscale, 2017, 9, 4283–4294. 38 R. B. Anderson, S. J. Smith, P. S. May and M. T. Berry, J. Phys. Chem. Lett., 2014, 5, 36–42. 39 M. T. Berry and P. S. May, J. Phys. Chem. A, 2015, 119, 9805– 9811. 40 K. Wu, J. Cui, X. Kong and Y. Wang, J. Appl. Phys., 2011, 110, 053510. 41 W. Yu, W. Xu, H. Song and S. Zhang, Dalton Trans., 2014, 43, 6139–6147. 42 J. Capobianco, P. Kabro, F. Ermeneux, R. Moncorge, M. Bettinelli and E. Cavalli, Chem. Phys., 1997, 214, 329–340. 43 L. Aarts, B. M. van der Ende and A. Meijerink, J. Appl. Phys., 2009, 106, 023522. 44 A. García-Flores, J. Matias, D. Garcia, E. Martínez, P. Cornaglia, G. Lesseux, R. Ribeiro, R. Urbano and C. Rettori, Phys. Rev. B, 2017, 96, 165430. 45 E. M. Chan, D. J. Gargas, P. J. Schuck and D. J. Milliron, The Journal of Physical Chemistry B, 2012, 116, 10561–10570. 46 W.-P. Qin, Z.-Y. Liu, C.-N. Sin, C.-F. Wu, G.-S. Qin, Z. Chen and K.-Z. Zheng, Light: Science & Applications, 2014, 3, e193– e193. 47 P. Goldner, F. Pellé, D. Meichenin and F. Auzel, Journal of luminescence, 1997, 71, 137–150. 48 S. K. Singh, K. Kumar, M. K. Srivastava, D. K. Rai and S. B. Rai, Optics letters, 2010, 35, 1575–1577. 49 Introduction to the Physics of Diluted Magnetic Semiconductors, ed. J. A. Gaj and J. Kossut, Springer-Verlag Berlin Heidelberg, 2010.


DOI: 10.1039/D0NR04252H

Page 9 of 10

Nanoscale

We resolve a long standing discrepancy between various reports on the effect of the magnetic field on up-converted luminescence.

View Article Online



DOI: 10.1039/D0NR04252H

N o r m a liz e d U C L In te n s ity ( a r b . u n its )


0 9 8 5

9 8 9 7 9 7 9 6 9 6

2 0

4 0 6 0 M a g n e tic F ie ld ( T ) 0

5 0

5

0

λe m = 5 4 5 n m

8 0



Page 10 of 10

DOI: 10.1039/D0NR04252H

λe x ( n m ) = 9 9 0

[No title available]

[No title available]

[No title available]

[No title available]

[No title available]

[No title available]

Title index.

Misleading title.

Incorrect table title.

Title X: a critical difference.

CT scanners vie for title.

What makes a good title?

What's in a title--Mr or Dr?

Writing style: what's in a title?

[Funktionsoberarzt. Pseudo-title or meaningful position?].

Futurist Art: Motion and Aesthetics As a Function of Title.

The anatomy of an article: title, abstract, and introduction.

"The chronic patient": in search of a title.

Citations-be sure to have a good title.

World title boxing: from early beginnings to the first bell.

Peer bids for protection of title 'veterinary nurse'.

Barriers to effective outreach in Title VII nutrition programs.

Article title misstates the role of pavement sealers.

Title and Abstract Screening and Evaluation in Systematic Reviews (TASER): a pilot randomised controlled trial of title and abstract screening by medical students.

No title

Follow Us

Suggest Documents

[No title available]

[No title available]

[No title available]

[No title available]

[No title available]

[No title available]

Title index.

Misleading title.

Incorrect table title.

Title X: a critical difference.

CT scanners vie for title.

What makes a good title?

What's in a title--Mr or Dr?

Writing style: what's in a title?

[Funktionsoberarzt. Pseudo-title or meaningful position?].

Futurist Art: Motion and Aesthetics As a Function of Title.

The anatomy of an article: title, abstract, and introduction.

"The chronic patient": in search of a title.

Citations-be sure to have a good title.

World title boxing: from early beginnings to the first bell.

Peer bids for protection of title 'veterinary nurse'.

Barriers to effective outreach in Title VII nutrition programs.

Article title misstates the role of pavement sealers.

Title and Abstract Screening and Evaluation in Systematic Reviews (TASER): a pilot randomised controlled trial of title and abstract screening by medical students.

No title

Recommend Documents

Follow Us