Quantum phosphors
Observation of the photon cascade emission process for Pr3+doped phosphors under vacuum ultraviolet (VUV) and X-ray
excitation
A.P. Vink1,2, E. van der Kolk1, P. Dorenbos1 and
C.W.E. van Eijk1
1 Radiation Technology Group, Interfaculty Reactor Institute, Delft University of
Technology, Mekelweg 15, 2629 JB Delft, The Netherlands
2 Chemical Sciences, Netherlands Organisation for Scientific Research, P.O. Box
93470, 2509 AL The Hague, The Netherlands
1
Radiation Technology, Interfaculty Reactor Institute
Outline
1. New generation lighting
2. Quantum cutting
3. Photon cascade emission with Pr3+
4. Selecting materials
5. Two types of emission in one
material
6. Quantum cutting with X Rays
7. Energy transfer 1S0 emission
8. Conclusions
2
New generation lighting
1,0
8.3eV
0,9
7.2eV
0,8
0,7
Intensity (a.u.)
• Commonly used TL lighting,
mercury (254 nm emission) is
used to excite a set of three
phosphors
• Result: white light
• Disadvantages: 1) mercury bad for
environment and 2) start-up time
• Alternative xenon-gas (emission
around 172 nm)
• Result: new set of phosphors
needed
0,6
0,5
0,4
0,3
0,2
0,1
0,0
120
130
140
150
160
170
180
190
200
Wavelength (nm)
3
New generation lighting
• In TL lighting: four lanthanides
used: Y2O3:Eu3+ (red),
BaMgAl10O17:Eu2+ (blue) and
GdMgB5O10:Ce3+,Tb3+ (green)
• Also used in television:
Y2O2S:Eu3+
• Partially filled 4f-shell, shielded
from surrounding (host)
4
Quantum cutting
• Major disadvantage of Xe is low
efficiency
• Comparison:
• Hg 254 nm 50% energy loss (4.9
eV)
• Xe 172 nm 70% energy loss (7.2
eV)
• To increase quantum efficiency:
quantum cutting
• Excitation into high-energy state
gives two step-emission to ground
state: result two photons (visible
region)
5
-1
60
3
• Pr3+: [Xe] 4f2 (praseodymium)
• Energy level scheme: 13 states
• Excitation into 1S0: two photons
• 1S0 level: weak absorption,
excitation into 4f15d1 state,
resulting in 1S0 → 1I6 (400 nm)
and 3P0 → 3H4 (480 nm)
• Predicted by Dexter (1957), but
discovered in 1974 by Sommerdijk
(Philips) and Piper (GE) for
YF3:Pr3+
Energy (10 cm )
Photon cascade emission with Pr 3+
1
1
4f 5d
50
1
S0
40
30
3
P2
3
1
D2
1
10
G4
3
3
H5
H4
3
6
3
F3, F4
3
H 6, F 2
3
0
1
P1, I6
3
P0
20
1,0
em=404 nm
exc=189 nm
-1
1
1
4f 5d
50
1
S0
40
0,8
Intensity (a.u.)
60
3
• Material which shows PCE:
SrAlF5:Pr3+
Energy (10 cm )
Photon cascade emission with Pr 3+
0,6
30
3
P2
0,4
3
second order
1
D2
host
0,2
1
10
G4
3
0,0
100 150 200 250 300 350 400 450 500 550 600 650 700
Wavelength (nm)
3
0
H5
H4
3
7
3
F3, F4
3
H 6, F 2
3
50
1
P1, I6
3
P0
20
Photon cascade emission with Pr 3+
a
A
1
1
3
1
1
3
a: 4f 5d -> H4
Intensity (a.u.)
0,8
1
1
Excitation (em=230 nm)
Emission (exc=190 nm)
3
4f 5d -> F2
1
1
3
4f 5d -> F3
b
b: 4f 5d -> H5
0,6
0,4
1
3
D2-> H4
+
0,2
1
1
1
4f 5d -> G4
1
1
3
1
4f 5d -> D2 4f15d1->3P ,1I (J:0,1,2)
J
6
3
P0-> F2
0,0
100
150
200
250
300
350
400
450
500
550
600
650
Wavelength (nm)
1,0
B
1
Excitation (em=403 nm)
Emission (exc=187 nm)
1
S0-> I6
0,8
Intensity (a.u.)
• Not only fluoride host show
PCE, also oxides!
• Two situations: 4f15d1 below
1S (for CaSO :Pr3+, above)
0
4
1
1
and 4f 5d above 1S0 (for
BaSO4:Pr3+, below)
• What factors determine
position of 4f15d1?
• Predict which material
shows PCE?
1,0
0,6
0,4
1
1
S0-> D2
0,2
1
1
3
S0-> F4
1
1
S0-> G4
3
3
3
P0-> H4
3
D2-> H4
+
3
P0-> F2
0,0
100
150
200
250
300
350
400
450
500
550
600
Wavelength (nm)
8
650
-1
40
0
1
4f 5d
3
• Other lanthanide: Ce3+ ([Xe] 4f1)
4f1 →4f05d1 transition at lower
energy and two 4f1 states
• Scintillator material: position 4f1
→ 4f05d1 known in many
compounds
• 5d1 split in five states
Pr3+ 4f2 →4f15d1
• single 5d electron splits into 5
states remaining 4f1 (Pr4+ or Ce3+)
Energy (10 cm )
Selecting materials
30
20
10
2
0
2
F7/2
F5/2
9
Selecting materials
Intenstity
A
60,0
57,5
55,0
52,5
50,0
47,5 45,0 42,5
3
-1
Energy (10 cm )
40,0
37,5
35,0
32,5
30,0
60,0 57,5 55,0
3
-1
Energy (10 cm )
52,5
50,0
47,5
45,0
42,5
B
Intenstity
• 4fn-15d1 structure of Ce3+ similar
as Pr3+, also crystal field splitting
is roughly the same
(CaSO4:Ce3+/CaSO4:Pr3+
• Energy difference is about 12 240
cm-1 (Dorenbos)
• In principle: extrapolate Pr3+ from
Ce3+ data (scintillator data)
• Differences: splitting of first band
is observed for Pr3+
• Only 4f1 and 5d1 splitting: two
lines, ΔE~ 2 000 cm-1
• 4f15d1 electrostatic interaction
72,5
70,0
67,5
65,0
62,5
10
Selecting materials
exc=216.5 nm
em=285 nm
0,8
Intensity (a.u.)
• In general: which materials show
quantum cutting?
• Determined by position lowest
4f15d1 state
• Position 5d1, centroid energy EC
(determined by type of ligands)
and crystal field splitting εcfs
(mainly by CN)
• Quantum cutters: high EC and
small εcfs
• Host materials: mainly fluorides
(>EC ) and some oxides (<εcfs)
• Example: KY3F10:Pr3+ (low CN)
1,0
second order
0,6
0,4
0,2
0,0
50
100 150 200 250 300 350 400 450 500 550 600 650 700
Wavelength (nm)
11
Two types of emission in one material
1
T= 10K
T= 292K
1
S0 -> I6
0,8
Intensity (a.u.)
• BaSO4
both different
emissions can be found
• Low temperatures: PCE and high
temperatures both PCE and 4f15d1
emission
• Expected: only one emission from
one site, but 4f15d1 near to 1S0
perhaps thermal population?
1,0
:Pr3+
1
0,6
1
1
3
1
S0 -> D2
1
4f 5d -> H4 - G4
0,4
1
1
S0 -> G4
0,2
1
1
1
3
4f 5d -> D2
3
P0 -> H4
0,0
200
250
300
350
400
450
Wavelength (nm)
12
500
Two types of emission in one material
1S
0
Log Intensity
• Decay time
emission becomes
shorter (190 to 56 ns) (4f15d1 →
4f2: 10 ns): extra decay channel
• Equations thermal population:
intensity and decay time
• Determine energy barrier
1
0,1
T= 10K
T= 293K
A
0,01
25
50
75
100
125
150
175
Time (ns)

1
 Atot
  E 
Af  Ad  exp 

kT



  E 
1  exp 

 kT 
R 
If
 E 
 C  exp 

Id
kT


13
Two types of emission in one material
Data points (exc=188 nm)
Data points (exc=205 nm)
Linear fit: ln(R)=a/T+b
ln (R)
-3
-2
-1
0
0,012
0,010
0,008
0,006
0,004
0,002
-1
1/T (K )
20
15
6
-1
1/ (10 s )
• Results on intensity
measurements straightforward
• Lifetime measurements: fitting
Af=6.24*106 s-1, Ad=62.24*106 s-1
(16 ns)
• Determining ΔE: 0.041 eV
(intensity) and 0.040 eV (decay
time)
• ΔE: energy barrier, not ΔE (1S0,
4f15d1)!
-4
10
5
0
0
50
100
150
200
250
300
Temperature (K)
14
Two types of emission in one material
E
• Effect is also found for other
lanthanides with low 4fn-15d1
bands (Eu2+, Sm2+), but not for
trivalent lanthanides
1
1
1
1
4f 5d (2)
4f 5d (1)
1
S0
Q
15
-1
40
0
1
4f 5d
3
• Ce3+ [Xe] 4f1 configuration
• Excitation over the band gap:
direct recombination and Self
Trapped Exciton (STE) formation
• Both emissions give the same
4f05d1 emission to 2F7/2,2F5/2
• Scintillator applications: STE
formation is unwanted; makes the
scintillator slower
• Increase of temperature: more
Ce3+ emission, less STE
• Increase of Ce3+ concentration,
less STE: more efficient transfer
Energy (10 cm )
Quantum cutting with X Rays
30
20
10
2
0
2
F7/2
F5/2
16
Quantum cutting with X Rays
X ray excitation T=100K
exc=111 nm T=10K
0.8
Intensity (a.u.)
• Pr3+ [Xe] 4f1 configuration
• Excitation over the band gap:
direct recombination and STE
formation
• Band gap can be reached with X
rays and VUV (λexc=111 nm)
• SrAlF5:Pr3+ at low temperatures
1.0
0.6
0.4
0.2
0.0
200
300
400
500
600
Wavelength (nm)
17
Quantum cutting with X Rays
• SrAl12O19:Pr3+ material: quantum
cutter
• Concentration dependence of STE
emission! (a: 0.05 %, b: 0.1 %,
c: 0.5% and d: 1.0 %)
• At room temperature 1S0 emission
is present: PCE process
18
Quantum cutting with X Rays
Intensity (*1E9)
0.32
0.24
0.16
0.08
0.00
200
300
400
500
600
700
800
Wavelength (nm)
60
3
-1
Energy (10 cm )
• Two processes: direct
recombination (PCE) and
formation of STE transferring its
energy to Pr3+
• Studied SrAlF5:Pr3+ under X ray
excitation
• STE: 260-545 nm
• < 403 nm 1S0 emissions
• > 487 nm 3P0 and 1D2 emissions
• STE does not overlap with 1S0
level (~215 nm)
T=100K
T=350K
0.40
1
1
4f 5d
50
1
S0
40
30
3
P2
3
1
P1, I6
P0
20
3
1
D2
1
10
G4
3
3
F3, F4
3
H6, F2
3
3
0
H5
3
H4
19
Quantum cutting with X Rays
3P
0
1
1
3
3
1
3
S0-> I6
P0-> H4
Intensity
0.4
D2-> H4
STE
0.3
0.2
0.1
0.0
100
150
200
250
300
350
Temperature (K)
1.0
1
exc=190 nm
exc= 90 nm
1
S0 I6
3
3
P 0 H4
0.8
Intensity (a.u.)
and 1D2 are fed by both STE
energy transfer and second step
PCE process: quench from 300K
• Is the energy transfer STE-Pr3+
efficient?
• Measurements on NaMgF3:Pr3+ at
room temperature
•
0.5
0.6
3
3
P 0 H5
1
1
S0 G4
0.4
3
0.2
1
1
0.0
200
3
D2 H4
3
S0 F3
3
S 0 H4
1
1
3
P 0 H6
1
S 0 D2
300
400
500
600
700
Wavelength (nm)
20
90
CB
11 eV
3
-1
Energy (10 cm )
Quantum cutting with X Rays
75
1
1
4f 5d
1
S
3 eV 0
60
45
3
1
PJ (J:0,1,2), I6
D2
1
30
1
G4
3
H5
15
STE
3
F3, F4
3
3
H6, F2
3
H4
3
Pr
3+
0
VB
21
Quantum cutting with X Rays
Data point
Least squares fit
-1.2
-1.6
Ln I
• Direct recombination is dependent
on temperature: rate determining
step
• Which sequence?
• First Pr3+ + h+→Pr4+ then Pr4+ +
e-→Pr3+ (4f15d1)
• First Pr3+ + e-→Pr2+ then Pr2+ +
VK→Pr3+ (4f15d1)
• Measured Intensity (1S0→1I6) as
function of temperature for
SrAlF5:Pr3+
• Arrhenius behavior: lnI versus 1/T
• Analysis: ΔE= 0.06 eV, 455 cm-1,
2.2kT (RT)
-0.8
-2.0
-2.4
-2.8
-3.2
0.0025
0.0030
0.0035
0.0040
0.0045
0.0050
0.0055
-1
1/T (K )
 E 

kT


I  C * exp 
22
0.0060
-1
CB
90
5b
3
• Energy value is small, a typical
value for a shallow electron trap,
too small for a VK center
• So: first Pr3+ + h+→Pr4+ then Pr4+
+ e-→Pr3+ (4f15d1)
• PCE process is determined by the
recombination rate of electron
trap with Pr4+
Energy (10 cm )
Quantum cutting with X Rays
6
electron trap
75
1
1
4f 5d
2a
1
S0
60
1
7a
45
3
30
1
PJ (J:0,1,2), I6
D2
1
3a
4 7b
3b
1
G4
3
H5
15
STE
3
F3, F4
3
3
H6, F2
3
2b
3
H4
Pr
3+
5a
0
VB
23
60
-1
1S
0
3
→ 1I6 (400 nm, UV) emission
step not suitable for lamp
applications
• Possible solution: co-doping with
other lanthanides or with
transition metal ions
• Possible candidate: Mn2+ (3d5):
1S → 1I overlaps with 6A →4A ,
0
6
1
1
4E (around 400 nm)
•
Energy (10 cm )
Energy transfer 1S0 emission
1
1
4f 5d
50
1
S0
4
T2
T
4 1
A2
4
40
4
T1
E
4
T2
4
2
A1, E
4
T2
4
T1
4
30
3
P2
3
1
P1, I6
20
3
P
D
1 0
2
1
G4
10
3
0
Pr
3+
3
F3, F4
3
3
H , F2
3 6
H5
3
H4
6
Mn
2+
24
A1
Energy transfer 1S0 emission
:Pr3+,Mn2+
• SrAlF5
and SrAlF5
(excitation into Pr3+ at 190 nm)
• No Mn2+ emission visible
• X Ray excitation: Mn2+ built in!
em=523 nm
exc=113 nm
0,8
Intensity (a.u.)
:Mn2+
1,0
Mn
2+
0,6
0,4
STE
0,2
4,5
3+
SrAlF5:Pr ,Mn
4,0
SrAlF5:Pr
SrAlF5:Mn
0,0
50
150
250
350
450
550
650
750
Wavelength (nm)
2+
1,0
3,0
0,8
2,5
2,0
Intensity (a.u.)
Intensity (*1E9)
3,5
2+
3+
scaled
1,5
0,6
0,4
second order
1,0
0,2
0,5
0,0
0,0
200
300
400
500
Wavelength (nm)
600
700
800
250
300
350
400
450
500
550
600
650
Wavelength (nm)
25
700
Conclusions
• Discussed quantum cutting for Pr3+ in a large number of hosts
• Can predict properties Pr3+ from Ce3+ data (scintillation)
• Pr3+ in some hosts can show both 4f15d1 emission and 4f2
emission from the same spectroscopic site
• Excitation with X Rays can also result in quantum cutting, but is
temperature dependent
• Fluoride materials are the most promising materials, have to be
co-doped with another ion
• Energy transfer Pr3+-Mn2+ not visible up till now
26