U.S. patent application number 14/176536 was filed with the patent office on 2014-08-14 for organic semiconductor lasers by triplet managers.
This patent application is currently assigned to The Regents of the Universtiy of Michigan. The applicant listed for this patent is The Regents of the Universtiy of Michigan. Invention is credited to Stephen R. Forrest, Yifan Zhang.
Application Number | 20140226692 14/176536 |
Document ID | / |
Family ID | 48172404 |
Filed Date | 2014-08-14 |
United States Patent
Application |
20140226692 |
Kind Code |
A1 |
Forrest; Stephen R. ; et
al. |
August 14, 2014 |
ORGANIC SEMICONDUCTOR LASERS BY TRIPLET MANAGERS
Abstract
A first device is provided. The device includes an organic
semiconductor laser. The organic semiconductor laser further
includes an optical cavity and an organic layer disposed within the
optical cavity. The organic layer includes: an organic host
compound; an organic emitting compound capable of fluorescent
emission; and an organic dopant compound. The organic dopant
compound may also be referred to herein as a "triplet manager." The
triplet energy of the organic dopant compound is lower than or
equal to the triplet energy of the organic host compound. The
triplet energy of the organic dopant compound is lower than or
equal to the triplet energy of the organic emitting compound. The
singlet energy of the organic emitting compound is lower than or
equal to the singlet energy of the organic host compound.
Inventors: |
Forrest; Stephen R.; (Ann
Arbor, MI) ; Zhang; Yifan; (Ann Arbor, MI) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
The Regents of the Universtiy of Michigan |
Ann Arbor |
MI |
US |
|
|
Assignee: |
The Regents of the Universtiy of
Michigan
Ann Arbor
MI
|
Family ID: |
48172404 |
Appl. No.: |
14/176536 |
Filed: |
February 10, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13446928 |
Apr 13, 2012 |
8654806 |
|
|
14176536 |
|
|
|
|
13283284 |
Oct 27, 2011 |
|
|
|
13446928 |
|
|
|
|
Current U.S.
Class: |
372/45.01 |
Current CPC
Class: |
H01S 5/183 20130101;
H01S 5/041 20130101; H01S 5/36 20130101; H01S 3/168 20130101; H01S
5/12 20130101 |
Class at
Publication: |
372/45.01 |
International
Class: |
H01S 5/36 20060101
H01S005/36 |
Goverment Interests
STATEMENT AS TO RIGHTS TO INVENTIONS MADE UNDER FEDERALLY SPONSORED
RESEARCH OR DEVELOPMENT
[0003] This invention was made with government support under
FA9550-10-1-0339 awarded by the AFOSR (AirForce Office of
Scientific Research). This invention was made with government
support under DE-SC0001013 awarded by the U.S. Department of
Energy, Energy Frontier Center at the University of Southern
California. The government has certain rights in the invention.
[0004] The claimed invention was made by, on behalf of, and/or in
connection with one or more of the following parties to a joint
university corporation research agreement: Regents of the
University of Michigan, Princeton University, The University of
Southern California, and the Universal Display Corporation. The
agreement was in effect on and before the date the claimed
invention was made, and the claimed invention was made as a result
of activities undertaken within the scope of the agreement.
Claims
1. A first device, further comprising: an organic semiconductor
laser, further comprising: an optical cavity; an organic layer
disposed within the optical cavity, the organic layer comprising:
an organic host compound; an organic emitting compound capable of
fluorescent emission; and an organic dopant compound; wherein: the
triplet energy of the organic dopant compound is lower than or
equal to the triplet energy of the organic host compound; the
triplet energy of the organic dopant compound is lower or equal to
than the triplet energy of the organic emitting compound; the
singlet energy of the organic emitting compound is lower than or
equal to the singlet energy of the organic host compound.
2. The first device of claim 1, wherein the singlet energy of the
organic emitting compound is lower than the singlet energy of the
organic host compound.
3. The first device of claim 1, wherein the singlet energy of the
organic emitting compound is lower than the singlet energy of the
organic dopant compound.
4. The first device of claim 1, wherein the organic dopant compound
does not strongly absorb the fluorescent emission of the organic
emitting compound.
5. The first device of claim 1, further comprising an optical pump
optically coupled to the organic layer.
6. The first device of claim 1, wherein the organic semiconductor
laser further comprises: an anode; a cathode; wherein the organic
layer is disposed between the anode and the cathode; a hole
transport layer disposed between the organic layer and the anode;
and an electron transport layer disposed between the organic layer
and the cathode; wherein the organic dopant compound is present
only in the emissive layer.
7. The first device of claim 1, wherein the triplet decay time of
the dopant compound is shorter than the triplet decay time of the
emitting compound
8. The first device of claim 1, wherein: the concentration of the
dopant compound is 10 wt %-90 wt %; the concentration of the
emitting compound is 0.5 wt %-5 wt %.
9. The first device of claim 1, wherein the organic emitting
compound is capable of fluorescent emission at room
temperature.
10. The first device of claim 1, wherein the dopant compound is
selected from the group consisting of: anthracene, tetracene,
rubrene, and perylene and their derivatives.
11. The first device of claim 1, wherein the dopant compound is
selected from anthracene and its derivatives.
12. The first device of claim 11, wherein the dopant compound is
ADN.
13. The first device of claim 1, wherein the dopant compound is a
phosphor.
14. The first device of claim 1, wherein the dopant compound is a
fluorophore.
15. The first device of claim 1, wherein the first device is a
consumer product.
16. A method, comprising: providing an organic semiconductor laser,
further comprising: an optical cavity; an organic layer disposed
within the optical cavity, the organic layer comprising: an organic
host compound; an organic emitting compound capable of fluorescent
emission; and an organic dopant compound; wherein: the triplet
energy of the organic dopant compound is lower than or equal to the
triplet energy of the organic host compound; the triplet energy of
the organic dopant compound is lower or equal to than the triplet
energy of the organic emitting compound; the singlet energy of the
organic emitting compound is lower than or equal to the singlet
energy of the organic host compound; pumping the organic
semiconductor laser to achieve lasing.
17. The method of claim 16, wherein the singlet energy of the
organic emitting compound is lower than the singlet energy of the
organic host compound;
18. The method of claim 16, wherein the pumping is optical
pumping.
19. The method of claim 16, wherein the pumping is electrical
pumping.
20. The method of claim 16, wherein lasing is achieved for at least
1 microsecond.
21. The method of claim 16, wherein the organic semiconductor laser
is pumped at a power that exceeds the pulse threshold.
22. The method of claim 16, wherein the organic semiconductor laser
is pumped at a power that exceeds the continuous wave
threshold.
23. The method of claim 16, wherein the organic semiconductor laser
is pumped at a power that exceeds the continuous wave threshold for
at least 1 microsecond.
24. The method of claim 16, wherein the organic semiconductor laser
is pumped at a power that exceeds the continuous wave threshold for
at least 100 microseconds.
25. The first device of claim 6, wherein the organic semiconductor
laser further comprises: a hole injection layer disposed between
the anode and the hole transport layer; and an electron injection
layer disposed between the cathode and the electron transport
layer.
26. The first device of claim 6, wherein the organic semiconductor
laser further comprises a feedback structure.
27. The first device of claim 6, wherein the feedback structure
comprises anyone of, or some combination of: a planar waveguide
structure; a distributed feedback structure; a Bragg reflector
feedback structure; or a vertical cavity structure surface emitting
structure (VCSEL).
28. The first device of claim 6, wherein the organic semiconductor
laser further comprises a substrate, wherein: the anode is disposed
over the substrate; and at least one mirror is disposed between the
substrate and the anode.
Description
CROSS-REFERENCES TO RELATED APPLICATIONS
[0001] This application claims priority to, is a
continuation-in-part of, and incorporates by reference in its
entirety U.S. Ser. No. 13/283,284 filed on Oct. 27, 2011.
[0002] This application is related to U.S. application Ser. No.
12/117,926, filed May 27, 2011, which claims priority to U.S.
61/396,862, filed Jun. 3, 2010, and U.S. 61/398,627, filed Jun. 29,
2010. These related applications are incorporated by reference in
their entireties.
FIELD OF THE INVENTION
[0005] The present invention relates to lasers using organic
materials.
BACKGROUND
[0006] Opto-electronic devices that make use of organic materials
are becoming increasingly desirable for a number of reasons. Many
of the materials used to make such devices are relatively
inexpensive, so organic opto-electronic devices have the potential
for cost advantages over inorganic devices. In addition, the
inherent properties of organic materials, such as their
flexibility, may make them well suited for particular applications
such as fabrication on a flexible substrate. Examples of organic
opto-electronic devices include organic light emitting devices
(OLEDs), organic phototransistors, organic photovoltaic cells, and
organic photodetectors.
[0007] Various ways to deposit the organic materials used to
fabricate organic devices are known, such as vacuum thermal
evaporation, solution processing, organic vapor phase deposition,
and organic vapor jet printing.
SUMMARY OF THE INVENTION
[0008] A first device is provided. The device includes an organic
semiconductor laser. The organic semiconductor laser further
includes an optical cavity and an organic layer disposed within the
optical cavity. The organic layer includes: an organic host
compound; an organic emitting compound capable of fluorescent
emission; and an organic dopant compound. The organic dopant
compound may also be referred to herein as a "triplet manager." The
triplet energy of the organic dopant compound is lower than or
equal to the triplet energy of the organic host compound. The
triplet energy of the organic dopant compound is lower or equal to
than the triplet energy of the organic emitting compound. The
singlet energy of the organic emitting compound is lower than or
equal to the singlet energy of the organic host compound.
[0009] In some embodiments, in the first device as described above,
the singlet energy of the organic emitting compound is lower than
the singlet energy of the organic host compound.
[0010] Preferably, the singlet energy of the organic emitting
compound is lower than the singlet energy of the organic dopant
compound.
[0011] Preferably, the organic dopant compound does not strongly
absorb the fluorescent emission of the organic emitting
compound.
[0012] In one embodiment, the first device further includes an
optical pump optically coupled to the organic layer.
[0013] In one embodiment, the organic semiconductor laser further
includes an anode and a cathode. The organic layer is disposed
between the anode and the cathode. A hole transport layer is
disposed between the organic layer and the anode. An electron
transport layer is disposed between the organic layer and the
cathode. The organic dopant compound is present only in the
emissive layer.
[0014] Preferably, the triplet decay time of the dopant compound is
shorter than the triplet decay time of the emitting compound.
[0015] Preferably, the concentration of the dopant compound is 10
wt %-90 wt %, and the concentration of the emitting compound is 0.5
wt %-5 wt %.
[0016] Preferably, the organic emitting compound is capable of
fluorescent emission at room temperature.
[0017] Preferably, the dopant compound is selected from the group
consisting of: anthracene, tetracene, rubrene, and perylene and
their derivatives. More preferably, the dopant compound is selected
from anthracene and its derivatives are particularly preferred.
More preferably, the dopant compound is ADN.
[0018] In some embodiments, the dopant compound is a phosphor. In
some embodiments, the dopant compound is a fluorophore.
[0019] In some embodiments, in the first device as described above,
the organic semiconductor laser may further comprise a feedback
structure. In some embodiments, the feedback structure may comprise
anyone of, or some combination of a planar waveguide structure; a
distributed feedback structure; a Bragg reflector feedback
structure; or a vertical cavity structure surface emitting
structure (VCSEL).
[0020] In some embodiments, in the first device as described above,
the organic semiconductor laser may further comprise a substrate.
The anode of the device may be disposed over the substrate and at
least one mirror is disposed between the substrate and the
anode.
[0021] In some embodiments, in the first device as described above,
the organic semiconductor laser may further comprise a hole
injection layer disposed between the anode and the hole transport
layer, and an electron injection layer disposed between the cathode
and the electron transport layer.
[0022] The first device may be a consumer product.
[0023] A method is provided. An organic semiconductor laser is
provided. The organic semiconductor laser further includes an
optical cavity and an organic layer disposed within the optical
cavity. The organic layer includes: an organic host compound; an
organic emitting compound capable of fluorescent emission; and an
organic dopant compound. The organic dopant compound may also be
referred to herein as a "triplet manager." The triplet energy of
the organic dopant compound is lower than or equal to the triplet
energy of the organic host compound. The triplet energy of the
organic dopant compound is lower or equal to than the triplet
energy of the organic emitting compound. The singlet energy of the
organic emitting compound is lower than or equal to the singlet
energy of the organic host compound. The organic semiconductor
laser is pumped to achieve lasing.
[0024] In some embodiments, in the method as described above, the
singlet energy of the organic emitting compound is lower than the
singlet energy of the organic host compound.
[0025] In one embodiment, the pumping is optical pumping.
[0026] In one embodiment, the pumping is electrical pumping.
[0027] When the organic semiconductor laser is pumped, lasing is
achieved for at least 1 microsecond.
[0028] In some embodiments, the organic semiconductor laser is
pumped at a power that exceeds the pulse threshold.
[0029] In some embodiments, the organic semiconductor laser is
pumped at a power that exceeds the continuous wave threshold.
[0030] In some embodiment, the organic semiconductor laser is
pumped at a power that exceeds the continuous wave threshold for at
least 1 microsecond, and more preferably for at least 100
microseconds.
BRIEF DESCRIPTION OF THE DRAWINGS
[0031] FIGS. 1a-1c show separate channels for singlet (S) and
triplet (T) formation and transfer in triplet managed lasers.
Singlets are generated (circles) on both Alq.sub.3 and ADN, and
Forster transferred (solid arrows) to DCM2. Triplets are generated
by intersystem crossing (ISC) or singlet fission and collected by
ADN through Dexter transfer (dashed arrows).
[0032] FIG. 2(a) shows Photoluminescence (PL), and FIG. 2(b) shows
lasing transients measured at 1.6 kW/cm.sup.2 pump intensity for
different host blends. The PL transients are normalized by the peak
intensities, and lasing transients are normalized to 1 for x=0, 10,
and 30 ADN blends, and to 5 for x=50 and 70. The fits are obtained
by the model described in text with parameters summarized in Table
1. Inset: Lasing spectrum of an x=70 OSL.
[0033] FIGS. 3a-3c show simulated threshold singlet population,
S.sub.TH(t) with (x=70, stars) and without (squares) a triplet
manager, and S(t) (lines) for the x=70 OSL. Lasing occurs when
S.gtoreq.S.sub.TH. The dashed lines correspond to lasers that have
exceeded their pulsed threshold singlet population (S.sub.PS), but
not their CW threshold (S.sub.CW). Left Inset: Streak camera image
of laser emission for a triplet-managed OSL (x=70) measured at 2.4
kW/cm.sup.2 pump intensity, 18 .mu.s pulse width. Right Inset:
Simulated lasing duration evolution with increasing pump power for
x=70 triplet managed OSL.
[0034] FIG. 4(a) shows Photoluminescence (PL) spectra of thin film
ADN (squares), Alq.sub.3 (circles), and DCM2 doped in an x %
ADN/(100-x)% Alq.sub.3 blend with x=0 (stars) and 70 (triangles).
The shaded region corresponds to where the DCM2 absorption
coefficient is >4.times.10.sup.4 cm.sup.-1, implying efficient
energy transfer from both ADN and Alq.sub.3. FIG. 4(b) shows
phosphorescence of Alq.sub.3, DCM2, and ADN measured at 14 K, the
fits are to double-peak Gaussian for Alq3 and single-peak Gaussians
for the rest.
[0035] FIG. 5(a) shows a probe pulse PL spectrum with and without
the pump pulse for x=0; and FIG. 5(b) shows a probe pulse PL
spectrum with and without the pump pulse for x=70 films. The 200 nm
film was grown on SiO.sub.2 (2 .mu.m)/Si substrate.
[0036] FIG. 6 illustrates the calculation of an absorption
coefficient for x=0 and 70 from the data in FIGS. 5(a) and
5(b).
[0037] FIG. 7 shows a streak camera image of laser emission for
x=70 triplet managed OSL measured at 2.4 kW/cm.sup.2 pump intensity
and 100 .mu.s pulse width integrated over 10 pulses. Note that the
color map uses logarithm scale. Lasing duration (characterized by
narrow spectrum) is approximately 55 .mu.s. The lasing wavelength
is slightly different from the inset of FIGS. 3a-3c because these
two lasers are not fabricated at the same time and their gain
medium thicknesses or grating periods can be slightly
different.
[0038] FIG. 8(a) illustrates EL transients of OLEDs at various
concentrations (0, 10%, 30%, 50%, 70% from bottom to top) of ADN as
a triplet manager pumped at 2 A/cm.sup.2 using rectangular current
pulses. Lines are fits to the model. FIG. 8(b) shows a Triplet
absorption coefficient, .alpha..sub.r, for 2% DCM2 doped ALq.sub.3
and ADN films versus wavelength.
[0039] FIG. 9 illustrates intensity peak-to-peak-steady-state ratio
(.rho. on left axis) as a function of current density for OLEDs
with different ADN concentrations. The corresponding guest triplet
populations (N.sub.GT on right axis) are calculated below.
[0040] FIG. 10(a) illustrates the EQE or triplet-managed OLEDs with
different ADN concentrations (x). The filled symbols are from
steady-state measurements, and open symbols are from transient
measurements. The inset shows current density vs. voltage (J-V)
characteristics for x.ltoreq.50% (almost identical) and x=70%
OLEDs. FIG. 10(b) illustrates triplet (.eta..sub.r) and charge
(.eta..sub.c) induced external quantum efficiency (EQE) roll-off
for x=0 and 50% OLEDs.
[0041] FIGS. 11(a)-11(d) illustrates exemplary feedback structures
in accordance with the present invention.
DETAILED DESCRIPTION
[0042] A model has been developed for organic lasers that predicts
two threshold pump intensities in optically pumped organic
semiconductor lasers (OSLs); one for pulsed lasing, I.sub.PS, and
another for continuous-wave (CW) lasing, I.sub.CW. The theory
predicts a decrease in I.sub.CW from 32 kW/cm.sup.2, or well above
the damage threshold, to 2.2 kW/cm.sup.2, for a laser employing
4-(dicyanomethylene)-2-methyl-6-julolidyl-9-enyl-4H-pyran (DCM2)
doped tris(8-hydroxyquinoline)aluminum (Alq.sub.3) if the triplets
can be effectively removed from the emissive guest. Based on this
analysis, it has been demonstrated that the lasing duration can be
extended to nearly 100 .mu.s, ultimately limited by degradation of
the lasing medium when a "triplet manager" molecule,
9,10-di(naphtha-2-yl)anthracene (ADN), is blended into the gain
region of a distributed feedback OSL. The triplet manager
facilitates radiative singlet transfer while suppressing
non-radiative triplet transfer to the emitter molecule, thus
reducing the triplet-induced losses. Our theory conclusively shows
that these lasers have entered the CW lasing regime.
[0043] Optically pumped organic semiconductor lasers (OSLs) with
low thresholds and wide spectral tuning ranges have attracted
interest since their demonstration 15 years ago. However, a
significant obstacle to the application of OSLs has been their
limitation to only pulsed operation with a maximum duration of
several tens of nanoseconds. This limitation is imposed by the
build-up of triplet (T) excitons in the gain region that are
generated from intersystem crossing (ISC) of radiative singlets
(S). Since the relaxation from T to the ground state is quantum
mechanically forbidden, the lifetime of T exciton is large
(.about.ms) compared to S (.about.ns), allowing the T population
can accumulate over time. The high T population, together with
overlapping S emission and T absorption, results in singlet and
photon losses that ultimately shut down lasing, thereby preventing
continuous-wave (CW) operation.
[0044] While triplet losses in liquid dye lasers can be mitigated
by using quencher molecules with triplet energies lower than that
of the dye, no CW operation has been realized without dye
circulation. For OSLs, gain medium circulation is not possible;
however, several efforts have been made to mitigate, although not
eliminate triplet losses to the extent that CW operation can be
achieved. Bornemann, et al., Opt. Lett. 31 (11), 1669-1671 (2006),
have used a rapidly rotating substrate to demonstrate a CW solid
state dye laser, but the output was unstable. Schols, et al., Chem
Phys Chem 10 (7), 1071-1076 (2009), have shown that "scavengers"
can be used to de-excite triplets, but no lasing improvement was
demonstrated. Rabe, et al., Appl. Phys. Lett. 89 (8), 081115
(2006), and Leahnhardt, et al. Org. Electron. 12 (8), 1346-1351
(2011), demonstrated a polymer OSL pumped by very low duty cycle
(<0.1%) pulses to extend the total duration to 400 .mu.s,
although this is not true CW operation.
[0045] Here, we introduce a "triplet manager" into the gain region
along with the guest emitter and host molecules. The manager
reduces the emitter triplet population, thus extending the lasing
duration. The inset of FIG. 1 shows the triplet management concept.
The manager has lower triplet energy and higher singlet energy than
the emitter. When either the host or manager molecules are excited,
Forster transfer of S-states to the emitter is highly efficient.
Furthermore, Dexter transfer of triplets lead to their trapping on
the manager since it has lower triplet energy than both the guest
and the host. The manager triplet absorption is shifted from the
guest emission and thus the trapped triplets do not contribute to
optical losses or singlet quenching.
[0046] The 200 nm thick OSL active region consists of the manager,
9,10-di(naphtha-2-yl)anthracene (ADN), co-deposited into the
conventional guest-host gain medium consisting of 2 vol % of the
red emitting
4-(dicyanomethylene)-2-methyl-6-julolidyl-9-enyl-4H-pyran (DCM2) in
tris(8-hydroxyquinoline) Al (Alq.sub.3). The S and T energies are
determined from fluorescence at room temperature, and
phosphorescence at 14K respectively, see EPAPS. Here, ADN has lower
T (1.69 eV) and higher S energy (2.83 eV) than Alq.sub.3 (T=1.99 eV
and S=2.38 eV). Furthermore, S=2.03 eV and T=1.74 eV for DCM2. This
system, therefore, is energetically consistent with FIG. 1.
[0047] The manager concentration in (100-x) vol % Alq.sub.3 is x
vol % ADN (x=0, 10, 30, 50, 70, 100). Blended films were deposited
by thermal evaporation in high vacuum (10.sup.-7 Torr) on quartz,
Si, and 2 .mu.m thick SiO.sub.2-on-Si substrates for characterizing
absorption, photoluminescence (PL), and triplet absorption,
respectively. The same films were deposited on gratings with a
period of 430 nm.+-.5 nm and a 50 nm depth on the SiO.sub.2-on-Si
to form distributed feedback (DFB) OSLs. Output from a 0.6 W laser
diode at wavelength .lamda.=405 nm was focused to a 150
.mu.m.times.250 .mu.m spot to optically pump the thin film.
Alq.sub.3 and ADN pure film absorption coefficients were measured
to be 4.8.times.10.sup.4 cm.sup.-1 and 9.1.times.10.sup.-4
cm.sup.-1 at .lamda.=405 nm, and are assumed to contribute to the
total blend film absorption proportionate to their volume. All
measurements were performed in N.sub.2 ambient to minimize film
degradation.
[0048] FIGS. 2(a) and (b) shows the PL and lasing transients pumped
at 1.6 kW/cm.sup.2. From FIG. 2(a), the Alq.sub.3 host undergoes a
55% reduction in PL to its steady state value within 30 .mu.s of
the onset of the pump. Previous studies have shown that this
intensity roll-off is due to singlet quenching from S-T
annihilation. That is, following the onset, the S density rapidly
reaches a peak and subsequently decays due to annihilation by the
slowly increasing T population. The existence of the long-term
steady state PL intensity below its peak suggests saturation of the
guest triplet population. By including the ADN manger into the host
blend with x=10 to 70, the PL transient quenching is reduced to
17%. Further increasing to x>70 can eliminate quenching
entirely. We infer, therefore, that triplets are transferred from
Alq.sub.3 to DCM2, while the transfer from ADN to DCM2 is
forbidden, consistent with the triplet energy relation
T(Alq.sub.3)>T(DCM2)>T(ADN). Note, however, that morphology
degradation under high pump intensity occurs for x>50,
consistent with the previous observation of morphological
instability of ADN.
[0049] In FIG. 2(b), more than a ten-fold increase in lasing time
(from approximately 400 ns to 4.5 .mu.s) is observed when x
increases from 0 to 70. Lasing is not observed for x=100 due to
degradation. The inset shows a typical lasing spectrum of a 70% ADN
OSL centered at .lamda.=687.9 nm, with a full width half maximum of
0.15 nm limited by the spectrometer resolution. The threshold
pulsed pump intensity, I.sub.PS (characterized by an abrupt
spectral narrowing from >30 nm to <0.5 nm, and a significant
increase in the slope efficiency) was obtained using a 30 ns pump
pulse (Table 1).
TABLE-US-00001 TABLE 1 Parameters for PL and lasing transients
fits, and the corresponding, measured pulsed (I.sub.PS) and
calculated CW (I.sub.CW) lasing threshold intensities Device
I.sub.TH.dagger. k.sub.HG* k.sub.ISC* N.sub.0*
.sigma..sub.TT.dagger. .sigma..sub.stim* I.sub.CW (% ADN)
(kW/cm.sup.2) (10.sup.10/s) (10.sup.7/s) (10.sup.18/cm.sup.3)
(10.sup.-17 cm.sup.2) (10.sup.-16 cm.sup.2) (kW/cm.sup.2) 0 0.93
4.0 3.3 5.0 .+-. 0.4 4.0 .+-. 0.3 1.9 32 10 0.75 3.5 2.6 3.9 .+-.
0.3 3.8 .+-. 0.3 2.0 19 30 0.72 13 2.3 2.8 .+-. 0.3 3.6 .+-. 0.4
2.4 8.8 50 0.45 3.0 .times. 10.sup.3 1.7 1.5 .+-. 0.2 4.3 .+-. 0.6
2.1 3.7 70 0.43 5.0 .times. 10.sup.5 1.3 0.92 .+-. 0.08 4.1 .+-.
0.4 2.3 2.2 .dagger.Parameters from measurement *Parameters from
fits to data
[0050] To understand the transient PL and lasing dynamics, we
extend previous work'_ENREF.sub.--8 to include triplet Dexter
transfer from host blend to guest, and guest triplet saturation.
Hence, the coupled laser rate equations are:
S t = .eta. I e p d - k S S - k ISC S - k ST ST G - .gamma. c n eff
P , ( 1 ) T H t = k ISC S - k HG exp ( - 2 L 1 N 0 - T G 3 ) T H ,
( 2 ) T G t = k HG exp ( - 2 L 1 N 0 - T G 3 ) T H , ( 3 ) P t = (
.GAMMA..gamma. - .alpha. CAV - .GAMMA. .sigma. TT T G ) c n eff P +
.GAMMA. .beta. k S S , ( 4 ) ##EQU00001##
where S, T.sub.H, T.sub.G, P are guest singlet, host blend triplet
(including both ADN and Alq.sub.3), guest triplet, and lasing mode
photon densities, respectively, t is time, .eta. is the fraction of
the pump emission absorbed by the organic film, I is the pump
intensity, e.sub.p=3.06 eV is the pump photon energy, d=200 nm is
the OSL gain medium thickness, k.sub.s=(6.7.+-.0.5).times.10.sup.8
s.sup.-1 is the guest S natural decay rate (measured from a 2%
DCM2:Alq.sub.3 film excited by 1.5 ns width N.sub.2 laser pulses),
k.sub.ISC is the host ISC rate, k.sub.ST is the guest S-T
annihilation rate, .gamma.=.sigma..sub.stimS is the gain,
.sigma..sub.stim is the stimulated emission cross-section, c is the
speed of light, and n.sub.eff=1.6 and .GAMMA.=0.69 are the
effective refractive index and optical confinement factor for the
SiO.sub.2 (n=1.48)/organic (n=1.82)/air (n=1) waveguide..sup.24
Also, k.sub.HG is the host-guest Dexter transfer coefficient, L is
the guest-host van der Waals radius (.about.1 nm), N.sub.0 is the
guest triplet saturation population, .alpha..sub.CAV is the cavity
loss without contributions from triplet absorption, .alpha..sub.TT,
and .sigma..sub.TT is the guest triplet absorption cross-section,
and .beta..apprxeq.10.sup.-4 is the spontaneous emission
factor.sup.25. For PL, we have P=0, and the intensity is
proportional to S. We assume that the host triplet population does
not interact with S or P; this is tested by fits to the data as
well as via direct triplet absorption measurements. Now, N.sub.0 is
determined from the saturation of PL quenching, and due to the
balance of triplet transfer from Alq.sub.3 to DCM2, and triplet
trapping on ADN. Introducing N.sub.0 avoids the complication of
treating individual triplet transfer in the ternary blend.
[0051] Free parameters k.sub.ST, k.sub.ISC, k.sub.HG, and N.sub.0
are used in fitting the PL transients. To test for model
consistency, transients at four different pump intensities (1.6,
1.3, 0.93, and 0.56 kW/cm.sup.2) yield a single set of parameter
values summarized in Table 1. For all films,
k.sub.ST=2.0.times.10.sup.10 cm.sup.3/s, as expected for guest S-T
annihilation due to resonant energy transfer that is only dependent
on DCM2 S emission and T absorption. As x increases, fewer triplets
are transferred from Alq.sub.3 to DCM2 and more are trapped on ADN;
thus N.sub.0 decreases from 5.0.times.10.sup.18 cm.sup.-3 to
9.2.times.10.sup.17 cm.sup.-3 when x=70, leading to decreased PL
transient roll-off. The .about.10.sup.5 increase in k.sub.HG seems
surprising, however, the Dexter transfer rate is determined by
k Dex = k HG exp [ - 2 L 1 N 0 - T G 3 ] . ##EQU00002##
Thus, for T.sub.G=0.7N.sub.o, k.sub.Dex only increases from
7.6.times.10.sup.4/s (x=0) to 4.4.times.10.sup.5/s (x=70),
consistent with the shorter PL quenching time for higher x.
[0052] To model the lasing transients in FIG. 2(b), three
additional parameters, .sigma..sub.TT, .sigma..sub.stim, and
.alpha..sub.CAV are required. Here, .alpha..sub.TT(.lamda.) was
measured following Lehnhardt, et al., Phys. Rev. B 81 (16), 165206
(2010). (see EPAPS), and .sigma..sub.TT=.alpha..sub.TT/N.sub.0 is
shown in Table 1 at .lamda.=680 nm. The nearly constant
.lamda..sub.TT(.lamda.) spectra and .sigma..sub.TT for all x are
consistent with the assumption that only guest triplet absorbs the
lasing emission (i.e. host and manager absorptions are negligible).
Furthermore, .alpha..sub.CAV=.GAMMA..sigma..sub.stimS.sub.PS, where
S.sub.PS=.eta.I.sub.PS/(e.sub.pdk.sub.s) is the pulse threshold S
population, where T build-up under short excitation pulses is
negligible. With these measurements and assumptions, the lasing
transients are fit using only a single free parameter,
.sigma..sub.stim (Table 1). We note that the effect of ADN as a
triplet manager lies in its ability to decrease N.sub.0, while
k.sub.ST and .sigma..sub.TT remain unchanged since they are
intrinsic to DCM2.
[0053] The net gain
g(t)=.GAMMA..sigma..sub.stimS(t)-.alpha..sub.CAV-.GAMMA..sigma..sub.TTT.s-
ub.G(t)=0 determines the threshold S population dynamics,
S.sub.TH(t), which are plotted in FIG. 3 for the Alq.sub.3 host
(squares) and the optimized (x=70) blend host (stars), using
parameters in Table 1. Surprisingly, two distinct threshold S
populations emerge from the fits, with a CW threshold population
(S.sub.CW) occurring at a density larger than that needed for
pulsed lasing (S.sub.PS). As t.fwdarw.0, triplet loss
.GAMMA..sigma..sub.TTT.sub.G<<.alpha..sub.CAV, giving
S.sub.PS=.alpha..sub.CAV/(.GAMMA..sigma..sub.stim). With time,
T.sub.G increases, concomitantly increasing the associated loss
until .GAMMA..sigma..sub.TTT.sub.G>.alpha..sub.CAV. Finally,
T.sub.G reaches its saturation density, N.sub.0, at which point the
triplet loss can no longer increase, giving
S.sub.CW=(.alpha..sub.CAV+.GAMMA..sigma..sub.TTN.sub.0)/(.GAMMA..sigma..s-
ub.stim). FIG. 3 also shows S(t) for several pump intensities, I,
for the host-manager blend, with the lasing duration vs. I plotted
in the inset. Due to the saturation of T.sub.G and thus S.sub.TH,
at I greater than the CW pump intensity threshold, I.sub.CW=2.2
kW/cm.sup.2, the lasing duration is no longer affected by triplet
loss, and approaches infinity. Table I also gives I.sub.CW for all
x in blended hosts. With larger N.sub.0 and thus increased triplet
loss, the Alq.sub.3-host OSL has I.sub.CW=32 kW/cm.sup.2. Due to
organic film damage at such high intensities, the CW lasing
threshold in the absence of a manager has not been previously
reported.
[0054] Neglecting the change in singlet population due to
stimulated emission (c.f. Eq. (1)), the CW threshold is
approximately:
I CW ( N 0 ) = e p d ( k S + k ISC + k ST N 0 ) .alpha. CAV +
.GAMMA..sigma. TT N 0 .eta..GAMMA. .sigma. stim , ( 5 )
##EQU00003##
compared to the pulse threshold
I.sub.PS=I.sub.CW(N.sub.0.apprxeq.0). From Eq. (5), I.sub.CW is a
quadratic function of guest triplet saturation population:
k.sub.STN.sub.0 is from S-T quenching that reduces the gain; and
.GAMMA..sigma..sub.TTN.sub.0 is due to triplet absorption,
increasing loss.
[0055] To test the existence of this CW regime, we excited an
Alq.sub.3/ADN (x=70)/DCM2 laser at 2.4 kW/cm.sup.2, or just above
the calculated value of I.sub.CW using the parameters in Table I.
FIG. 3, left inset, shows a streak camera image of this emission
over 20 .mu.s duration. Lasing becomes weaker (leading to the
apparent spectral narrowing) but does not turn off at the end of
the long pulse, consistent with theory. Indeed, we observed nearly
100 .mu.s lasing duration (see EPAPS) when pumped by a single
pulse, although film degradation due to high optical pump
intensities ultimately limits the lasing duration. Hence, while
this OSL has clearly exceeded its CW threshold, the laser operates
quasi-CW due to material degradation.
[0056] Interestingly, the lasing wavelength blue shifts from
.lamda.=688.1 nm to 687.7 nm during the lasing period, shown in
FIG. 3, left inset. Wavelength shifts have been observed in liquid
dye lasers where they have been attributed to the competition of
triplet absorption with the gain spectrum. In thin film single mode
DFB OSLs where the triplet absorption is largely constant (see
EPAPS), the shift is more likely due to changes in the effective
refractive index with increasing T density.
[0057] We note that I.sub.CW can be further reduced and the lasing
time extended by using a more stable, lower triplet energy manager,
and with a better match between the manager emission and the guest
absorption than is achieved with ADN. Then, the smaller saturation
guest triplet population contributes negligible loss, in which case
I.sub.CW.fwdarw.I.sub.PS. The design concept can be applied to the
eventual development of electrically pumped organic semiconductor
lasers, where 75% of the injected electrons result in triplets
compared to only a few percent in optical pumping.
[0058] In conclusion, we show the existence of a CW threshold at a
higher pump intensity than the pulsed threshold observed in all
previous OSL studies. Based on our analysis, we demonstrate a
lasing duration of up to 100 .mu.s by introducing a triplet manager
into the OSL gain medium. The reduced triplet-induced loss of the
triplet managed OSL decreases I.sub.CW from 32 kW/cm.sup.2 to a
more practical value of 2.2 kW/cm.sup.2 observed here.
Measurements
[0059] The singlet energy is derived from the peak of PL spectrum
in FIG. 4(a) for ADN (S=2.83 eV), Alq3 (S=2.38 eV), and DCM2
(S=2.03 eV). The spectral overlap of the ADN and Alq.sub.3 PL and
DCM2 absorption, together with DCM2 emission in the lightly doped
films, indicates efficient and complete Forster transfer of
singlets generated in both host and manager to the guest. The 17 nm
bathochromic shift in PL for the doped film is due to the
solid-state solvation effect.
[0060] Triplet energies of the three fluorescence molecules are
measured using the method of Tanaka et al., Phys. Rev. B 71 205207
(2005). Due to the negligible triplet emission of the fluorophores,
the two phosphors: tris(2-phenylpyridine)Ir(III) (Ir(ppy).sub.3),
T=2.4 eV) and bis(2-phenylquinoline) (acetylacetonate)Ir(III)
(PQIr), T=2.1 eV) are co-doped with the material under study to
enable efficient transfer of photogenerated triplets from the
phosphor to the fluorophore, thereby overcoming its low intersystem
crossing rate. The triplet energy of the phosphor is lower than the
fluorophore singlet, but higher than its triplet for transfer to
occur. Photoluminescence at 14K from the blends:
Alq.sub.3(25%)/Ir(ppy).sub.3(75%), DCM2(50%)/PQIr(50%), and
ADN(50%)/Ir(ppy).sub.3(50%) using a N.sub.2 pump laser (1 ns pulse)
are measured using a streak camera (Hamamatsu C4334). The spectra
at 0.4 ms to 9 ms after the pump pulses are shown in FIG. 4(b).
Triplet energies of Alq.sub.3 (T=1.99 eV), DCM2 (T=1.74 eV), and
ADN (T=1.69 eV) are extracted from fits to Gaussians (solid lines).
The Alq.sub.3 triplet energy is consistent with that obtained by
Tanaka, et al., Phys. Rev. B 71 205207 (2005).
[0061] The triplet absorption is measured through spatially
separated pump probe experiment proposed by Lehnhardt et al., Org.
Electron. 12 486 (2011). The pump pulse from a .lamda.=405 nm laser
diode has 50 .mu.s duration, saturating the guest triplet; the
probe pulse from a N.sub.2 (1.5 ns width) laser is applied 100 ns
after the turn-off of the pump. The film PL is collected from the
film edge by either a streak camera (Hamamatsu C4334) for
absorption spectrum .alpha..sub.TT(.lamda.) measurement, or through
a .lamda.=680.+-.5 nm band-pass filter by an avalanche photodiode
(C5658) for .alpha..sub.TT near the lasing wavelength with a
smaller error. FIG. 5 shows the PL intensity from the probe with
and without the pump pulse for x=0 and 70 films. We can see the
triplet absorption is significantly reduced with introducing the
triplet manager. From FIG. 5, two similar broad and featureless
.alpha..sub.TT(.lamda.) are obtained in FIG. 6, confirming that the
absorptions are from the same excited state (guest triplet). As
described above, the .about.5 times difference is from the
different guest triplet saturation population, N.sub.0.
Enhanced Efficiency in High-Brightness Fluorescent OLEDs
[0062] The inventors have also demonstrated enhanced efficiency in
high-brightness fluorescent organic light emitting diodes through
triplet management.
[0063] In particular, the inventors have demonstrated
singlet-triplet (S-T) quenching, and hence increased quantum
efficiency, in high-brightness fluorescent organic light emitting
diodes (OLEDs) by reducing the guest triplet population through the
introduction of a triplet manage molecule into the emission layer
(EML). As an example, an OLED whose EML consists of the red
fluorophore,
4-(dicyanomethylene)-2-methyl-6-julolidyl-9-enyl-4H-pyran doped
into the host, tris(8-hydroxyquinoline) Al (Alq.sub.3) is blended
with the triplet manager, 9,10-di(naphtha-2-yl)anthracene. The
manager triplet energy is less than that of the host or dopant,
leading to efficient triplet removal from the dopant without
affecting the radiative singlet population. Measurements suggest
the complete suppression of S-T quenching using the triplet
management strategy, leading to grater than 100% increase in the
steady state OLED external quantum efficiency.
[0064] Efficiency loss in fluorescent organic light emitting diodes
(OLEDs) has been variously attributed to singlet polaron (S-P)
quenching charge imbalance, and singlet-triplet (S-T) quenching; In
the case of S-T quenching, the efficiency reduction is proportional
to the product of the triplet (T) population and the quenching
rate. Since 75% of the injected charge results in the formation of
non-radiative triplets with relatively long lifetimes (typically
>1 ms) compared to radiative singlets (.about.1-10 ns), triplets
in OLEDs can reach a high density (>10.sup.18 cm.sup.-3). Also.
fluorescent OLEDs often utilize a laser dye as the emitter, which
usually has a resonance between the S emission and T absorption
spectra, leading to a large S-T quenching rate. The combined large
T population and high S-T quenching rate can reduce the OLED
efficiency.
[0065] The conventional emission layer (EML) of a fluorescent OLED
consists of a conductive host and an emissive guest. Because the
guest often has lower S and T energies than the host, both
excitonic species formed on the host transfer to the guest where
S-T quenching occurs. One strategy to reduce quenching is to blend
a "triplet manager" molecule into the EML to collect triplets, as
shown again in FIG. 1. If the T manager has a higher S and lower T
energy than the guest, it facilitates Forster transfer of S
excitons to the guest and Dexter transfer of T excitons from the
guest to the manager, thereby potentially eliminating guest S-T
annihilation. Here, we demonstrate the effectiveness of such a
strategy in OLEDs with EMLs comprising a tris(8-hydroxyquinoline)
Al (Alq.sub.3) host, doped with 2 vol. % of
4-(dicyanomethylene)-2-methyl-6-julolidyl-9-enyl-4H-pyran (DCM2).
Various concentrations (O.ltoreq.x.ltoreq.70 vol. %) of 9,10
di(naphtha-2-yl)anthracene (ADN) are blended in the EML for T
management.
[0066] The OLEDs were fabricated in vacuum (.about.10.sup.-7 Torr)
by thermal evaporation on pre-patterned indium-tin-oxide
(ITO)-coated glass substrates following standard procedures. The 25
nm-thick EML is sandwiched between a 35 nm thick
4,4'-bis[N-(1-naphthyl)-N-phenyl-amino]-biphenyl (NPD) hole
transport layer (HTL) and a 25 nm thick bathocuproine (BCP)
electron transport layer. The array of 1 mm.sup.2 devices is
completed by depositing 0.8 nm thick LiF and 100 nm thick Al
through a shadow mask to define the cathodes. The steady state
current density-voltage-luminance (J-V-L) characteristics were
measured at <0.3 A/cm.sup.2 using a parameter analyzer and a
calibrated Si photodetector. At higher currents of 0.1-2.5
A/cm.sup.2, the electroluminescence (EL) intensity was measured
using a pulse generator (100 .mu.s pulse width and an avalanche
photodiode.
[0067] With reference again to FIG. 4(a), which shows the
photoluminescence (PL) specta of ADN and Alq.sub.3 films and the EL
spectra for OLEDs with EMLs with x=0 and 50% ADN. The PL spectral
overlap with the DCM2 absorption (shaded box) and the pure DCM2
OLED emission confirm the complete Forster-transfer of singlets to
the guest. The triplet energies of the three molecules are measured
at 14 K using the method of triplet energy transfer from a
sensitizing iridium complex. From FIG. 4(b), the inventors measured
triplet energies of Alq.sub.3 (T=1.99 eV), DCM2 (T=1.74 eV), and
ADN (T=1.69 eV), which are consistent with the requirements shown
in FIG. 1.
[0068] Evidence that ADN suppresses triplet transfer to DCM2 is
inferred from the triplet absorption coefficients (x.sub.r) for 200
nm thick films of 2% DCM2 doped in ADN and Alq.sub.3 on SiO.sub.2,
in FIG. 8 (inset). The triplet state is populated by a 1.6
kW/cm.sup.2, 50 .mu.s, .lamda.=405 nm wavelength pump laser diode
through intersystem crossing from the singlet, and the absorption
is measured by transmission of the PL excited by a 5
.mu.J/cm.sup.2, 1.5 ns .lamda.=337 nm N.sub.2 laser pulse delayed
100 ns from the pump. Since x.sub.r=.sigma..sub.TT.sub.G, where
.sigma..sub.T is the guest triplet absorption cross-section, and
T.sub.G is its density, the elimination of x.sub.r by ADN is due to
the successful removal of triplets from DCM2. Further, the
negligible excited state absorption for the DCM2-doped ADN film
also indicates that the ADN triplet absorption vanishes in the DCM2
PL spectral range.
[0069] FIG. 8(a) shows the transient EL for a series of OLEDs with
different ADN fractions, x, following the onset of a J=2 A/cm.sup.2
current step. For the control device (x=0), the EL rapidly reaches
a peak and then decays to a steady state intensity .about.50% of
its initial value during the following 20 .mu.s. Since the EL
intensity is proportional to the S density, the transient decay is
an indication of S-T quenching. Following the onset of the current
pulse, the S population rapidly approaches its peak in the absence
of triplets. As current is maintained, the long lifetime
(.about.ms) triplet population increases, resulting in increased
S-T quenching, and thus a decreased S population. Finally, when the
guest T population approaches steady state, S-T quenching also
stabilizes, giving rise to a reduced steady-state S population.
[0070] In the presence of the T manager, the magnitude of the EL
transient decay reduces with increasing x and disappears for
x.gtoreq.50%, indicating the complete suppression of S-T quenching.
FIG. 9 shows the EL intensity peak-to-steady-state ratio, .rho.(J)
for OLEDs in FIG. 8(a). It was determined that .rho.(J) decreases
with increased manager concentration, and .rho.(J)=1 for
x.gtoreq.50%, corresponding to zero transient decay.
[0071] Two processes dominate energy transfer in triplet-managed
OLEDs, namely guest S-T quenching, S+T.sub.G.fwdarw.S.sub.o+T.sub.G
(So is the ground state), and host-to-guest triplet Dexter
transfer, T.sub.H includes both Alq.sub.3 and ADN triplets). The EL
may be modeled using:
S ( t ) t = h 1 4 J ? - S ( t ) ? - ? , ( 1 ) T ? t = 6 3 4 J c d -
? exp ( - 2 L 1 N ? - T G ( t ) ) T ? , ( 2 ) T G ( t ) t = ? exp (
- 2 L 1 N GT ( J ) - T G ( t ) ) ? ? indicates text missing or
illegible when filed ( 3 ) ##EQU00004##
where b is the charge imbalance factor, e is electron charge, d is
the EML thickness assuming that excitons are uniformly distributed
across this layer, .tau..sub.s is the singlet natural decay
lifetime, and k.sub.st is the S-T quenching rate. Also, k.sub.HG
and L are the host-to-guest triplet transfer rate coefficient and
van der Waals radius, respectively, and N.sub.GT(J) is the
saturation guest T population; i.e. when
T.sub.G(t).fwdarw.N.sub.GT(J), the transfer stops and
dT.sub.G/dt=0. This treatment assumes 25% of injected charges from
singlets. The exponential factors are due to diffusive transfer of
excitons between molecules consistent with the Dexter process.
Eventually, T.sub.H saturates due to triplet-triplet (T-T)
quenching and natural decay, but these processes are much slower
(.about.1 ms) than the time window of interest (.about.30 .mu.s)
and are thus neglected. Singlet-singlet (S-S) quenching is not
considered since it is negligible at such low densities. Finally,
S-P quenching is also possible; however, it cannot result in the
observed EL transients since the polaron density reaches steady
state within tens of ns following the onset of the current
pulse.
[0072] Assuming the boundary condition that T.sub.G=0 at t=0, and
T.sub.G=N.sub.GT as t.fwdarw..infin. in Eq. (1), .rho.(J) is
.rho.(J)=1+.tau..sub.Sk.sub.STN.sub.GT(J).
The external quantum efficiency roll-off is then given by
EQE(J)=.eta..sub.o.eta..sub.C(J).eta..sub.T(J), where .eta..sub.o
is external quantum efficiency (EQE) with perfect charge balance
and no singlet quenching, .eta..sub.C(J) is the roll-off due to
charge imbalance and S-P quenching, and .eta..sub.T(J)=1/.rho.(J)
is the roll-off due to S-T quenching.
[0073] From the PL transients measured for 50 nm thick films using
an N.sub.2 laser pump, .tau..sub.s=1.5.+-.0.2 ns for
DCM2:ADN/Alq.sub.3, independent of the manager concentration. For a
2 vol. % DCM2 concentration, L.apprxeq.3 nm as determined from the
average distance between dopant molecules. With the constraint on
N.sub.GT(J) given by Eq. (4), and for perfect charge balance (b=1),
the EL transients are fit by Eqs. (1)-(3) as shown by the solid
lines in FIG. 8(a). From the fits, k.sub.ST=1.5.times.10.sup.-10
cm.sup.3/s, independent of host and manager concentrations, and is
similar to previous values obtained for DCM:Alq.sub.3. Also, it was
found that k.sub.HG=2.5.times.10.sup.7 s.sup.-1 for x=0 and 10%,
and k.sub.HG=3.times.10.sup.8 s.sup.-1 for x=30% ADN.
[0074] With these fits, we obtain N.sub.GT using Eq. (4), with
results also indicated in FIG. 9. Clearly, the introduction of the
triplet manager substantially reduces N.sub.GT. For example, at
J=0.5-2.5 A/cm.sup.2, N.sub.GT(J) is reduced from 4.times.10.sup.18
cm.sup.-3 (x=0) to <3.times.10.sup.17 cm.sup.-3 (x=30%) and O
cm.sup.-3 (x.gtoreq.50%).
[0075] FIG. 10(a) shows EQE(J) for different ADN concentrations.
For x=0-50%, the EQEs at low J (<10.sup.-4 A/cm.sup.2) are
similar at .about.3%, while the EQE roll-off with J is reduced at
higher x. At high currents (J>0.1 2.5 A/cm.sup.2), the EQE at
x=50% is enhanced by more than 100% from x=0 due to reduced
roll-off. The separate contributions from polaronand
triplet-induced roll-off are plotted in FIG. 10(b). Here,
.eta..sub.T(J) is obtained from .rho.(J), while .eta..sub.C(J) is
solved assuming .eta..sub.o=3% obtained at J.fwdarw.O. From FIG.
10(b), reduced S-T quenching is the major contribution for the
enhanced EQE at x=50%, while polaron-induced roll-off is also
slightly reduced at x=50% due to bipolar charge transport in ADN.
That is, while both ADN and AIq.sub.3 are good electron
transporting materials, ADN has a higher hole mobility. This can
result in more efficient recombination, and thus a smaller free
carrier density in the EML. In turn, this reduces the S-P quenching
and thus slightly increases .eta..sub.C(J).
[0076] In contrast, EQE decreases when x increase from 50% to 70%,
though in both cases, .rho.=1. This is attributed to reduced
.eta..sub.C(J) at x=70% due to charge imbalance. Since the NPD
lowest unoccupied molecular orbital (LUMO) is at 2.3 eV, the
electron blocking barrier at the HTL/EML interface is smaller for
ADN (with a LUMO at 2.6 eV) than for Alq.sub.3 (LUMO=3.1 eV). Thus,
at high ADN concentrations electrons preferentially transport on
ADN and more readily escape from the EML without recombination.
This is evident from FIG. 10(a), inset, where x=70% shows
significantly higher current at x<50%. Thus, the host
(Alq.sub.3) cannot be completely replaced by the triplet manager to
ensure that charge balance is maintained.
[0077] T.-H. Liu, C.-Y. Iou, and C. H. Chen, Curr. Appl. Phys.
5(3), 218 (2005), which is incorporated herein by reference in its
entirety, used ADN and rubrene as co-hosts in an OLED based on the
fluorophore,
4-(dicyanomethylene)2-t-butyl-6-(1,1,7,7-tetramethyljulolidyl-9-enyl)-4H--
pyran (DCJTB), doped into Alq.sub.3 to reduce the high intensity,
steady state EQE roll-off. The enhanced EQE was explained by
reduced singlet-charge (Alq.sub.3.sup.-) quenching. The inventors
analysis differs from that of Liu et al., where the transient
efficiency indicates that reduced S-T quenching is the principal
cause of the enhanced EQE. It should be noted that rubrene (T=1.1
eV (shown in M. Montalti, A. Credi, L. Prodi, and M. T. Gandolfi,
Handbook of Photochemistry, 3.sup.rd ed. (CRC. Boca Raton, Fla.
2006) p. 83, which is incorporated herein by reference in its
entirety)) can also be effectively employed as a triplet manager
although data for rubrene:Alq.sub.3:DCM2 devices are omitted for
brevity.
[0078] In conclusion, the inventors have demonstrated that triplet
managers blended into the EML of fluorescent OLEDs can lead to a
significant increase in quantum efficiency at high current density.
The manager molecules promote efficient triplet transfer from the
guest, thereby reducing S-T quenching at high currents. Such a
management strategy can be generally applied to fluorescent OLEDs
that suffer S-T quenching to achieve very high brightness, and has
also been shown to be effective in allowing organic semiconductor
lasers to enter the continuous wave operation regime.
[0079] Previous studies have shown that triplet induced losses
including both singlet-triplet quenching and triplet absorption are
significant obstacles for electrically pumped organic semiconductor
lasers (OSLs). In view of the above and provided herein, a triplet
management method that could eliminate (or reduce) the triplet
induced losses in OSLs is provided and thus opening a new route for
the development of electrically pumped OSLs.
[0080] The triplet induced losses in conventional OSLs (including
both small molecules and polymers) originate from the spectrum
overlap of the laser emission and triplet state absorption. The
triplet management strategy involves doping the laser gain medium
by an additional molecule, called triplet manager, which may have
the following features: 1. the manager triplet energy is lower than
all other molecules already existed in the gain medium; 2. the
manager triplet absorption is negligible at the lasing wavelength.
Once these two features are satisfied, triplet induced loss may be
eliminated or significantly reduced. The effectiveness of using
9,10-di(naphtha-2-yl)anthracene (ADN) or rubrene as triplet
managers in the conventional small molecule guest-host gain medium
of 4-(dicyanomethylene)-2-methyl-6-julolidyl-9-enyl-4H-pyran (DCM2)
doped tris(8-hydroxyquinoline) aluminum (Alq.sub.3) was
demonstrated above. By optical pumping, the ADN managed laser has a
lasing duration of nearly 100 .mu.s compared to only 400 ns of the
conventional non-managed laser.
[0081] Using the same triplet management strategy in the emission
layer (EML) of a fluorescent organic light emitting diode (OLED),
the inventors discovered complete suppression of triplet induced
loss and thus more than 100% efficiency enhancements. This was also
demonstrated by using ADN or rubrene as triplet manager in DCM2
doped Alq.sub.3 EML of an OLED.
[0082] As provided herein, the inventors found that the two
approaches described above could be combined for an electrically
pumped OSL. A laser consists of a gain medium and a feedback
structure. The gain medium of the electrically pumped OSLs utilizes
a triplet management strategy, e.g. a triplet manager molecule
doped in an otherwise conventional polyfluorene or small molecule
guest-host system. The feedback structure can have any suitable
form, such as the examples provided in FIGS. 11(a)-(d). For
example, FIG. 11(a) shows a planar waveguide feedback, with
feedback provided by reflection from the edge surface. FIG. 1(b)
shows the distributed feedback (DFB), with feedback given by the
diffraction of the grating. FIG. 11(c) shows the planar distributed
Bragg reflector feedback (DBR), with feedback provided by the
reflection from the DBR parallel to the substrate. FIG. 11(d) shows
the vertical cavity structure surface emitting structure (VCSEL),
with the feedback given by the reflection from front and back
mirror. However, as would be appreciated by one of ordinary skill
in art after reading this disclosure, any suitable feedback system
or apparatus may be used.
[0083] In general, the material choice and thickness for the anode,
hole injection layer (HTL), electron injection layer (ETL) and
cathode in an OSL may be the same as for conventional OLEDs. The
gain medium thickness may be in the range of 50 nm-300 nm. The
substrate should preferably have a refractive index less than
organic films (n.about.1.6-1.8), and is typically quartz or
SiO.sub.2, but may comprise any suitable material. In the exemplary
feedback structures shown in FIGS. 11(a)-(c), the feedback is
parallel to substrate; whereas in the exemplary structure shown in
FIG. 11(d), the feedback is perpendicular to substrate.
[0084] The lasing threshold current density for the exemplary
electrically pumped OSLs may be calculated as follows. The
threshold condition is given by:
g(t)=.GAMMA..sigma.S(t)-.alpha.-.GAMMA..sigma..sub.TTT(t)>0,
where t is time, g is the net gain, .GAMMA. is the optical
confinement factor, .sigma. is the stimulated emission cross
section, .alpha. is the cavity loss, .sigma..sub.TT is the triplet
absorption cross-section, and S and T are singlet and triplet
densities, respectively. Now, with triplet managers, the triplet
induced losses can be eliminated, thus:
g ( t ) = .GAMMA..sigma. S ( t ) - .alpha. > 0 and S ( t ) t = I
4 ed - S ( t ) .tau. , ##EQU00005##
where d is the gain medium thickness (EML thickness in OLEDs), e is
the electron charge, and .tau. is the singlet lifetime. Then,
threshold current density is
J.sub.th=4ed.alpha./.GAMMA..tau..sigma. (note that .alpha. and
.GAMMA. are not independent). With typical values of d=50 nm,
.GAMMA.=0.3, .alpha.=10 cm.sup.-1, .tau.=2 ns, and
.sigma.=4.times.10.sup.-17 cm.sup.2, then it can be calculated that
J.sub.th=1.3 kA/cm.sup.2. According to previous studies, such high
current density is generally reachable in organic semiconductor
materials. If triplet losses exist, the threshold current densities
will be significantly higher (See N. C. Giebink and S. R. Forrest,
"Temporal response of optically pumped organic semiconductor lasers
and its implication for reaching threshold under electrical
excitation," Phys. Rev. B 79 (7), 073302 (2009) at FIG. 3(b), which
is incorporated by reference herein in its entirety).
[0085] Besides triplet loss, contact loss and polaron loss are also
considered as obstacles for electrically injected OSLs. One recent
study disclosed in M. Reufer, S. Riechel, J. M. Lupton, J.
Feldmann, U. Lemmer, D. Schneider, T. Benstem, T. Dobbertin, W.
Kowalsky, A. Gombert, K. Forberich, V. Wittwer and U. Scherf,
"Low-threshold polymeric distributed feedback lasers with metallic
contacts," Appl. Phys. Lett. 84 (17), 32623264 (2004) at p. 8,
which is incorporated herein by reference in its entirety, showed
that the contact loss can be eliminated by carefully tuning the
waveguide mode. Polaron absorption was previously measured to be
similar to triplet absorption, but it can be mitigated by using
high conductivity materials to reduce the polaron concentration.
This is because the equilibrium polaron population decreases with
increasing the mobility by: n .varies. 1/ {square root over
(.mu.)}, where .mu. is the mobility of the film.
Exemplary Embodiments
[0086] A first device is provided. The device includes an organic
semiconductor laser. The organic semiconductor laser further
includes an optical cavity and an organic layer disposed within the
optical cavity. The organic layer includes: an organic host
compound; an organic emitting compound capable of fluorescent
emission; and an organic dopant compound. The organic dopant
compound may also be referred to herein as a "triplet manager." The
triplet energy of the organic dopant compound is lower than or
equal to the triplet energy of the organic host compound. The
triplet energy of the organic dopant compound is lower than or
equal to the triplet energy of the organic emitting compound. The
singlet energy of the organic emitting compound is lower than or
equal to the singlet energy of the organic host compound.
[0087] The energy relationships described allows the organic dopant
compound to drain triplets from the organic host compound and the
organic emitting compound, and to avoid transfer of triplets from
the organic dopant compound to the other compounds. The organic
dopant compound also preferably has a triplet lifetime that is
short relative to the organic host and organic emitting compounds,
so that the organic dopant compound can drain many triplets over a
period of time.
[0088] Triplet and singlet energies for a wide variety of compounds
are readily available in the literature. Some of that literature is
described herein. These energies may be measured, and they may also
be calculated. There may be some differences in the values obtained
from various measurement and calculation techniques. For purposes
of comparing energies to see whether one is greater or less than
another, it is preferred that the same or similar measurement or
calculations be used to obtain the values to be compared. If a
particular measurement or calculation tends to overestimate or
underestimate energy values, using the same measurement or
calculation to obtain all values being compared will minimize any
such effect.
[0089] The use of the term "or equal to" with regard to the singlet
energy of the organic emitting compound in comparison to the
singlet energy of the organic host compound is intended to take
into consideration the possibility for endothermic transfer. These
endothermic processes are described in more detail in U.S. Pat.
Pub. No. 2005/0214576 filed on May 3, 2005, which is incorporated
herein by reference in its entirety. Thus, while it may generally
be preferred that the singlet energy of the organic emitting
compound is lower than the singlet energy of the organic host
compound, embodiments are not necessarily so limited.
[0090] In some embodiments, in the first device as described above,
the organic semiconductor laser may further comprise a hole
injection layer disposed between the anode and the hole transport
layer, and an electron injection layer disposed between the cathode
and the electron transport layer. As noted above, the device
(and/or the OSL) may comprise some or all of the organic layers
that are typically utilized in conventional OLEDs.
[0091] Preferably, the singlet energy of the organic emitting
compound is lower than the singlet energy of the organic dopant
compound.
[0092] This energy relationship minimizes transfer of singlets from
the organic emitting compound to the organic dopant compound, so
that the singlets on the organic emitting compound may desirably
emit a photon.
[0093] Preferably, the organic dopant compound does not strongly
absorb the fluorescent emission of the organic emitting
compound.
[0094] By stating that the dopant compound "does not strongly
absorb" the fluorescent emission of the emitting compound, it is
meant that the ground, singlet excited and triplet-triplet excited
states all have low absorption of the spectrum emitted by the
fluorescent emitting compound. One of skill in the art can
ascertain whether there is low absorption by observing plots of the
emission and absorption spectra on the same graph, and checking
whether there is significant overlap--some small overlap in the
tail regions is permissible.
[0095] In one embodiment, the first device further includes an
optical pump optically coupled to the organic layer.
[0096] In one embodiment, the organic semiconductor laser further
includes an anode and a cathode. The organic layer is disposed
between the anode and the cathode. A hole transport layer is
disposed between the organic layer and the anode. An electron
transport layer is disposed between the organic layer and the
cathode. The organic dopant compound is present only in the
emissive layer.
[0097] Preferably, the triplet decay time of the dopant compound is
shorter than the triplet decay time of the emitting compound.
[0098] Preferably, the concentration of the dopant compound is 10
wt %-90 wt %, and the concentration of the emitting compound is 0.5
wt %-5 wt %.
[0099] In one embodiment, the host may be completely absent from
the organic layer, i.e., the organic layer includes only the
emitting compound and the dopant compound. In this situation, the
concentration of the emitting compound is preferably 0.5 wt %-5 wt
%, with the remainder of the layer being the dopant compound. The
preferred energy level relationships described herein for the
dopant compound and the emitting compound still apply to this
embodiment. The device may still include other organic layers, such
as hole transport and electron transport layers--this embodiment
relates to a two-component emitting layer having the dopant and the
emitting compound, but the rest of the device may have any of the
other components described herein.
[0100] Preferably, the organic emitting compound is capable of
fluorescent emission at room temperature.
[0101] Preferably, the dopant compound is selected from the group
consisting of: anthracene, tetracene, rubrene, and perylene and
their derivatives. More preferably, the dopant compound is selected
from anthracene and its derivatives are particularly preferred.
More preferably, the dopant compound is ADN.
[0102] In some embodiments, the dopant compound is a phosphor. In
some embodiments, the dopant compound is a fluorophore.
[0103] It is preferable to maximize the transfer of singlets from
the host and dopant to the emitting compound. The co-system of host
and dopant (without an emitting compound) may have different
emission spectra for different dopant concentrations. For an
optimized concentrations, the overlap between this emission spectra
and the guest absorption spectrum is maximized. The result of such
maximization is increased or possibly complete singlet transfer to
the emitting compound. This is a more detailed condition than the
condition that the singlet energy of the emitting compound is
simply less than the singlet energy of the host and the singlet
energy of the dopant.
[0104] It is preferable to maximize the transfer of triplets from
the host and the emitting compound to the dopant. A higher dopant
concentration will result in more efficient triplet transfer from
the emitting compound and the host to the dopant. This is a more
specific condition than the condition that the triplet energy of
the dopant is less than or equal to the triplet energy of the host
and the emitting compound. However, if the dopant concentration is
too high, transfer of singlets to the emitting compound may be
compromised. Also, too high a dopant concentration can result in
faster device degradation.
[0105] For embodiments where the laser is pumped electrically,
i.e., where the emissive layer is between an anode and a cathode, t
is also preferable to select concentrations of the emitting
compound, the host and the dopant that result in the highest
emitting layer mobility.
[0106] In some embodiments, the triplet energy of the dopant
compound is preferably greater than 1.3 eV, and more preferably
greater than 1.6 eV. 130 kJ/mol is around 1.3 eV If the triplet
energy of the host material is sufficiently high, the triplet sink
dopant triplet energy may also be high. For example, anthracene
(1.6 eV triplet exciton) as a triplet sink in Alq.sub.3 is an
example. The use of high energy materials as described herein
enables hosts that are capable of supporting high energy
fluorescent emitters, such as blue-emitting emitters. Because
technology for organic emitting compounds is generally weaker for
blue emitters, the use of high energy hosts and triplet sinks to
improve the performance of blue-emitting fluorescent laser devices
is a particularly desirable result. In a particularly preferred
embodiment, the triplet energy of the dopant compound is at least
1.3 eV, preferably at least 1.6 eV, and not more than 1.7 eV for
the reasons described herein.
[0107] The first device may be a consumer product. In the context
of this application, such a consumer product include any product
available for purchase that includes a laser. Examples include
printers, communications equipment, CD/DVD devices, and many other
well known products. One of skill in the art can readily substitute
the lasers described herein for the conventional lasers already
used in such consumer products.
[0108] In some embodiments, in the first device as described above,
the organic semiconductor laser may further comprise a feedback
structure. The feedback structure may be configured to reflect (or
otherwise divert or direct) photons having a specific wavelength
and phase to travel back through the gain (i.e. lasing) medium. The
feedback structure may comprise, on one or both ends, a partially
transmissive surface that allows a fraction of the photons to
transmit through the interface (whereas the remaining photons may
be reflected or otherwise directed back through the gain medium. In
some embodiments, the feedback structure may comprise one or more
mirrors (e.g. a half silvered mirror) and or diffractive grating.
In some embodiments, the feedback structure may comprise anyone of,
or some combination of a planar waveguide structure; a distributed
feedback structure; a Bragg reflector feedback structure; or a
vertical cavity structure surface emitting structure (VCSEL).
However, any suitable feedback structure may be used in embodiments
disclosed herein as would be understood by one of ordinary skill in
the art. For example, in some embodiments, in the first device as
described above, the organic semiconductor laser may further
comprise a substrate. The anode of the device may be disposed over
the substrate and at least one mirror is disposed between the
substrate and the anode.
[0109] A method is provided. An organic semiconductor laser is
provided. The organic semiconductor laser further includes an
optical cavity and an organic layer disposed within the optical
cavity. The organic layer includes: an organic host compound; an
organic emitting compound capable of fluorescent emission; and an
organic dopant compound. The organic dopant compound may also be
referred to herein as a "triplet manager." The triplet energy of
the organic dopant compound is lower than or equal to the triplet
energy of the organic host compound. The triplet energy of the
organic dopant compound is lower or equal to than the triplet
energy of the organic emitting compound. The singlet energy of the
organic emitting compound is lower than or equal to the singlet
energy of the organic host compound. The organic semiconductor
laser is pumped to achieve lasing.
[0110] In some embodiments, in the method as described above, the
singlet energy of the organic emitting compound is lower than the
singlet energy of the organic host compound.
[0111] In one embodiment, the pumping is optical pumping.
[0112] In one embodiment, the pumping is electrical pumping.
[0113] When the organic semiconductor laser is pumped, lasing is
achieved for at least 1 microsecond.
[0114] In some embodiments, the organic semiconductor laser is
pumped at a power that exceeds the pulse threshold.
[0115] In some embodiments, the organic semiconductor laser is
pumped at a power that exceeds the continuous wave threshold.
[0116] In some embodiment, the organic semiconductor laser is
pumped at a power that exceeds the continuous wave threshold for at
least 1 microsecond, and more preferably for at least 100
microseconds.
[0117] It is understood that the various embodiments described
herein are by way of example only, and are not intended to limit
the scope of the invention. For example, many of the materials and
structures described herein may be substituted with other materials
and structures without deviating from the spirit of the invention.
The present invention as claimed may therefore includes variations
from the particular examples and preferred embodiments described
herein, as will be apparent to one of skill in the art. It is
understood that various theories as to why the invention works, and
modeling of specific configurations, are not intended to be
limiting.
* * * * *