Efficient and color-tunable Mn-doped ZnSe nanocrystal emitters: Control of optical performance via greener synthetic chemistry

Article abstract of DOI:10.1021/ja068360v

Formation of Mn-doped ZnSe quantum dots (Mm:ZnSe d-dots) using nucleation-doping strategy was studied systematically and optimized through greener approaches. The resulting d-dots were with high (～50%) photoluminescence (PL) quantum yield (QY), which was achieved by the controlled formation of small-sized MnSe nanoclusters as the core and a diffused interface between the nanocluster core and the ZnSe overcoating layers. Synthesis of the d-dots under high temperatures (240-300 °C) was achieved by varying the structure of the metal carboxylate precursors, concentration of the inhibitors, free fatty acid, and concentration of the activation reagents, fatty amines. Highly emissive d-dots synthesized under desired conditions were found to be extremely stable upon thermal treatment up to the boiling point of the solvent (about 300 °C), which was quantitatively studied using in situ measurements. The PL peak of the d-dots was controllably tuned in a surprisingly large optical window, from 565 to 610 nm. These highly emissive and stable d-dots possess characteristics of practical emissive materials, especially for applications requiring high power, high concentration of emitters, and under tough conditions.

Full text of DOI:10.1021/ja068360v

Published on Web 02/21/2007
Efficient and Color-Tunable Mn-Doped ZnSe Nanocrystal
Emitters: Control of Optical Performance via Greener
Synthetic Chemistry
Narayan Pradhan and Xiaogang Peng*
Contribution from the Department of Chemistry & Biochemistry, UniVersity of Arkansas,
FayetteVille, Arkansas 72701
Received November 21, 2006; E-mail: xpeng@uark.edu
Abstract: Formation of Mn-doped ZnSe quantum dots (Mn:ZnSe d-dots) using nucleation-doping strategy
was studied systematically and optimized through greener approaches. The resulting d-dots were with
high (∼50%) photoluminescence (PL) quantum yield (QY), which was achieved by the controlled formation
of small-sized MnSe nanoclusters as the core and a diffused interface between the nanocluster core and
the ZnSe overcoating layers. Synthesis of the d-dots under high temperatures (240-300 °C) was achieved
by varying the structure of the metal carboxylate precursors, concentration of the inhibitors, free fatty acid,
and concentration of the activation reagents, fatty amines. Highly emissive d-dots synthesized under desired
conditions were found to be extremely stable upon thermal treatment up to the boiling point of the solvent
(about 300 °C), which was quantitatively studied using in situ measurements. The PL peak of the d-dots
was controllably tuned in a surprisingly large optical window, from 565 to 610 nm. These highly emissive
and stable d-dots possess characteristics of practical emissive materials, especially for applications requiring
high power, high concentration of emitters, and under tough conditions.
beads-based bar-coding,11 and others requiring significant power
and/or high density of nanocrystals.
Introduction
Doped semiconductor nanocrystals (d-dots), specifically ones
not containing heavy metal ions, have the potential to become
a class of mainstream emissive materials. As illustrated recently,1
Mn- and Cu-doped ZnSe (Mn:ZnSe and Cu:ZnSe) d-dots can
cover an emission window similar to that of the current
workhorse of intrinsic quantum dot (q-dots) emitters, CdSe
nanocrystals.2,3 Besides their low toxicity by replacing cadmium
in CdSe quantum dots with zinc, d-dot emitters can also
overcome a couple of intrinsic disadvantages of undoped
quantum dot emitters, that is, strong self-quenching caused by
their small ensemble Stokes shift (energy difference between
absorption spectrum and emission band)4,5 and sensitivity to
thermal, chemical, and photochemical disturbances.1,6,7 These
two intrinsic properties of undoped quantum dots may make
them not ideal for several potential applications being actively
pursued in the field, such as LEDs,8 lasers,9 solid-state lighting,10
As compared to common emissive materials, organic dyes
and inorganic phosphor powders, undoped q-dot emitters possess
apparent advantages, such as narrow and symmetric emission
with tunable colors, broad and strong absorption, zero scattering,
reasonable stability, and solution processibility. d-Dot emitters
can potentially retain more or less all of these advantages. They
overcome the self-quenching problems of undoped q-dots by
their substantial ensemble Stokes shift. This large Stokes shift
in d-dots was due to their relatively small emission energy gap,
in comparison to the absorption band gap of the host nano-
crystals, in their atomic-like emission states. These atomic-like
states are inner core electronic states of the dopant centers, which
thus do not couple to the lattice phonons strongly. Consequently,
the emission of d-dots can be thermally stable.12 For these
reasons, d-dots may become a new class of practical emissive
materials, which complement organic dyes, inorganic phosphors,
and undoped q-dots.
(1) Pradhan, N.; Goorskey, D.; Thessing, J.; Peng, X. J. Am. Chem. Soc. 2005,
127, 17586-17587.
Doped semiconductor nanocrystals are not new to the field.
Many different synthetic approaches have been reported13-24
(2) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993,
115, 8706-15.
(3) Hines, M. A.; Guyot-Sionnest, P. J. Phys. Chem. 1996, 100, 468-71.
(4) Kagan, C. R.; Murray, C. B.; Nirmal, M.; Bawendi, M. G. Phys. ReV. Lett.
1996, 76, 1517-20.
(10) Lee, J.; Sundar, V. C.; Heine, J. R.; Bawendi, M. G. AdV. Mater. 2000,
12, 1102-1105.
(5) Achermann, M.; Petruska, M. A.; Crooker, S. A.; Klimov, V. I. J. Phys.
Chem. B 2003, 107, 13782-13787.
(11) Han, M.; Gao, X.; Su, J. Z.; Nie, S. Nat. Biotechnol. 2001, 19, 631-635.
(12) Kim, J. H.; Holloway, P. H. AdV. Mater. 2005, 17, 91-96.
(13) Bhargava, R. N.; Gallagher, D.; Hong, X.; Nurmikko, A. Phys. ReV. Lett.
1994, 72, 416-19.
(6) Empedocles, S. A.; Norris, D. J.; Bawendi, M. G. Phys. ReV. Lett. 1996,
77, 3873-3876.
(7) Li, J. J.; Wang, Y. A.; Guo, W.; Keay, J. C.; Mishima, T. D.; Johnson,
M. B.; Peng, X. J. Am. Chem. Soc. 2003, 125, 12567-12575.
(8) Colvin, V. L.; Schlamp, M. C.; Allvisatos, A. P. Nature 1994, 370, 354-
7.
(14) Sooklal, K.; Cullum, B. S.; Angel, S. M.; Murphy, C. J. J. Phys. Chem.
1996, 100, 4551-5.
(15) St. John, J.; Coffer, J. L.; Chen, Y.; Pinizzotto, R. F. J. Am. Chem. Soc.
1999, 121, 1888-1892.
(9) Klimov, V. I.; Mikhailovsky, A. A.; Xu, S.; Malko, A.; Hollingsworth,
J. A.; Leatherdale, C. A.; Eisler, H.; Bawendi, M. G. Science 2000, 290,
314-7.
(16) Mikulec, F. V.; Kuno, M.; Bennati, M.; Hall, D. A.; Griffin, R. G.; Bawendi,
M. G. J. Am. Chem. Soc. 2000, 122, 2532-2540.
9
10.1021/ja068360v CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 3339-3347
3339

A R T I C L E S
Pradhan and Peng
Results
for developing doped emissive materials as well as other types
of functional materials, such as dilute magnetic doped semi-
conductor materials.25 The traditional synthetic approaches have
been generally based on a reaction system with both dopant
ions and competitive host ions in it, which is difficult to control
and optimize. As a result, the resulting d-dots are often mixed
with a significant portion of undoped nanocrystals, judged by
their emission spectra. This situation was substantially improved
by our recent report,1 which reported tunable and pure dopant
emission (>99% in intensity) from about 470 to 590 nm with
PL QY around 10-30%. The key feature of the related synthetic
chemistry is decoupling the doping from nucleation and/or
growth through nucleation-doping and growth-doping strate-
gies.1 This allows, in principle, the placement of the dopant
ions at a desired radial position in the host nanocrystals. During
the preparation of this Article, we noticed that Cao’s group26
realized decoupling doping from nucleation and growth by
isolating CdS core nanocrystals prior to the doping process.
Although cadmium was still included in the resulting Mn-doped
CdS/ZnS core/shell structure and might not be ideal for practical
applications, the PL QY of their doped nanostructures was
impressive, as high as 50%. From a synthetic point of view,
their method is somewhat similar to the growth-doping strategy.
Control of the size of the MnSe core is the first key issue
for the nucleation-doping process. Ideally, a small-sized nucleus
would place the dopant ions as close as possible to the center
of the d-dots. This shall result in emission centers as far away
as possible from the potential surface trap states of the
nanocrystals, and thus improve the optical performance of the
d-dots. A small nucleus also means a relatively uniform
environment of the doping ions in a d-dot. Furthermore, a small-
sized nucleus might be more suited for formation of a better
interface between the MnSe core and ZnSe shell, even a
mutually diffused interface, in comparison to a large-sized MnSe
nanocrystal. The latter hypothesis is based on the fact that the
chemical potential of particles increases dramatically as their
size decreases in the few nanometer size range.28 A more
practical issue needed to be addressed in this section is the slow
formation rate (a few hours) of MnSe nuclei reported in the
previous communication.1
One structural feature for the targeted d-dots is that the anionic
precursor, specifically the Se precursor, is common for both
host growth and nucleation. This offers a convenient handle
for tuning the reaction conditions. In principle, if the Se
precursor is used in large excess, the cationic precursor, Mn
fatty acid salts, should be consumed rapidly, and small-sized
MnSe nanoclusters would be the resulting products. Such small
MnSe nanoclusters should be stable in the reaction solution
because of the existence of an excessively high Se monomer
concentration in the solution.29
When the precursor ratio was relatively low, formation of
large-sized MnSe nanocrystals by continuous growth was
observed. For the example in Figure 1a (Mn to Se precursor
equals 1:8), the absorption edge gradually shifted to over 400
nm, which is similar to what was observed in the previous
communication for MnSe particles larger than 4 nm in size.1
When the Mn to Se precursor was 1:32 (Figure 1b), however,
the absorption edge of the MnSe clusters stayed below 300 nm
even after the reaction mixture was heated for about 1 h. The
corresponding FTIR spectra (Figure 1c) revealed that Mn
carboxylate was consumed within about 2 min, indicated by
the disappearance of the asymmetric vibration peak of the
-COO- group at about 1550 cm-1 (Figure 1c). Here, the IR
peaks for the -CdC- double bond of octadecene as the solvent
(labeled in Figure 1c) were used as the internal references.
Practically, the Se to Mn precursor ratio between 25:1 and
35:1 was found to be adequate for the formation of small and
stable MnSe clusters. When this ratio was higher than 35, the
large excess of Se precursor might cause some complication in
the subsequent overcoating stage.
The potential of the new synthetic strategies, nucleation-
doping and growth-doping, seems to be promising. However,
several fundamental properties of the resulting d-dots need to
be greatly improved for practical applications, such as PL QY,
color tunability, color purity (emission peak width), surface/
ligand chemistry, durability and stability, etc. For the nucleation-
doping strategy to be discussed here, both energy transfer and
emission processes should occur at the interface of the nuclei
with the dopant ions as the emission centers and the pure host
overcoating layer as the absorption zone. For this reason, the
primary emphasis in this work was the controlled formation of
the dopant-containing nuclei as well as the interface between
the doped core and pure host shell. The surprising tunability of
the emission wavelength of Mn:ZnSe d-dots observed previously
was also studied further in this work. Overall, the synthetic
schemes were developed with consideration of possible greener
methods, that is, simpler, less time-consuming, and less danger-
ous than that reported previously.1 A separate report27 will deal
with chemical and photochemical stability, biocompatibility, and
bioaccessibility of these highly efficient d-dot emitters for
biomedical applications. Although the current report concen-
trates on key issues of synthetic chemistry, some structural
characterization of the resulting d-dots will also be provided
and discussed.
The quality of the Mn fatty acid salt precursor, such as Mn
stearate (MnSt2), played a determining role for the reproduc-
ibility of the synthesis. If a stoichiometric amount of base was
mixed with the fatty acid prior to the addition of MnCl2, the
resulting MnSt2 (after purification and drying) should be a white
powder. If the reaction was with excess base, or the base was
added after the mixing of the fatty acid and MnCl2, the resulting
fatty acid salts would be brownish (see Supporting Information,
(17) Norris, D. J.; Yao, N.; Charnock, F. T.; Kennedy, T. A. Nano Lett. 2001,
1, 3-7.
(18) Bol, A. A.; Meijerink, A. J. Phys. Chem. B 2001, 105, 10203-10209.
(19) Azad Malik, M.; O’Brien, P.; Revaprasadu, N. J. Mater. Chem. 2001, 11,
2382-2386.
(20) Radovanovic, P. V.; Norberg, N. S.; McNally, K. E.; Gamelin, D. R.
J. Am. Chem. Soc. 2002, 124, 15192-15193.
(21) Raola, O. E.; Strouse, G. F. Nano Lett. 2002, 2, 1443-1447.
(22) Schwartz, D. A.; Norberg, N. S.; Nguyen, Q. P.; Parker, J. M.; Gamelin,
D. R. J. Am. Chem. Soc. 2003, 125, 13205-13218.
(23) Yang, H.; Santra, S.; Holloway, P. H. J. Nanosci. Nanotechnol. 2005, 5,
1364-75.
(24) Erwin, S. C.; Zu, L.; Haftel, M. I.; Efros, A. L.; Kennedy, T. A.; Norris,
D. J. Nature 2005, 436, 91-94.
(27) Pradhan, N.; Battaglia, D.; Liu, Y.; Peng, X. Nano Lett. 2006, ASAP,
DOI: 10.1021/nI062336y.
(28) Peng, X. Chem.-Eur. J. 2002, 8, 334-339.
(29) Peng, X.; Wickham, J.; Alivisatos, A. P. J. Am. Chem. Soc. 1998, 120,
5343-5344.
(25) Radovanovic, P. V.; Gamelin, D. R. J. Am. Chem. Soc. 2001, 123, 12207-
12214.
(26) Yang, Y.; Chen, O.; Angerhofer, A.; Cao, Y. C. J. Am. Chem. Soc. 2006,
128, 12428-12429.
9
3340 J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007

Color-Tunable Mn-Doped ZnSe Nanocrystal Emitters
A R T I C L E S
Figure 1. (a and b) Temporal evolution of the UV-vis spectra of the MnSe nanoparticles grown with different precursor ratios. (c) Temporal evolution of
FTIR spectra for the case with Mn to Se precursor ratio being 1:32. The three IR peaks were assigned to CdC of ODE and the asymmetric stretching
vibration of the -COO- group.
Figure 2. (a) UV-vis spectra of the d-dots in different growth stages. (b,c,d) PL spectra of the d-dots at different stages and different growth temperatures.
The Mn:Zn ratio in each plot is the precursor ratio at a given growth stage. Excitation wavelength is 350 nm.
Figure S1). This brownish color was most likely associated with
the formation of Mn hydroxide or oxide. The existence of such
impurities made the formation of MnSe nanoclusters compli-
cated and irreproducible, particularly when the Se concentration
was low. Another issue related to the precursor was, if a
significant amount of free fatty acid was not purified away, the
formation of small-sized MnSe nanoclusters would become
difficult because of the decreased reactivity of the metal
carboxylate precursors as we observed previously.30 It should
be pointed out that the existence of free fatty acids could be
easily detected using FTIR, with a carbonyl vibration peak
pot fashion, meaning no isolation of the MnSe nanoparticles,
the complete consumption of the Mn precursor and stability of
the resulting MnSe nanoclusters were critically important,
especially for the relatively high reaction temperatures to be
discussed below. In this sense, a large excess of Se precursor
for the formation of MnSe particles would be ideal.
All overcoating reactions were monitored with UV-vis
spectroscopy. After each addition of zinc fatty acid salt
precursor, the reaction was allowed to proceed until no
observable changes were found by UV-vis within about 15
min intervals. Such a cycle, from the injection of zinc precursor
to the completion of the ZnSe growth caused by the injection,
is called a one growth step in this report. The final UV-vis
spectrum after each growth step (Figure 2a) for the reaction
carried out under different growth temperatures tested was about
the same as long as the amount of zinc precursor injected was
the same for a given growth step. Overall, the UV-vis spectra
of the resulting d-dots are similar to those of quantum shell
samples,32 instead of quantum dots, as reported in the previous
communication.1 This is consistent with the structural feature
of the core-doped d-dots, with the host semiconductor layer
epitaxially grown onto the surface of the doped core particles.
In fact, observation of sharp exciton absorption peaks similar
to quantum dots was noted as a sign for the formation of
undoped ZnSe nanocrystals.
between 1700 and 1720 cm-1
.
Overall, with the properly synthesized precursors and a
relatively large excess of Se precursor, formation of small-sized
MnSe nanoclusters was rapid (shorter than 10 min) and
reproducible. The exact size of the excessively small MnSe
nanoclusters could not be directly determined when using a
transmission electron microscope (TEM). An indirect method
will be discussed later.
The overcoating of ZnSe layer onto the MnSe particles was
studied systematically by varying the reaction temperature for
the overcoating process. Based on a general understanding of
epitaxial growth,31 the quality of the interface should be greatly
influenced by the reaction temperature. However, the method
reported in our previous communication was limited to 180 °C
and with tributylphosphine as the coordinating solvent for
limiting the reactivity of the zinc acetate precursor. It was
expected that the composition of the reaction mixture must be
significantly changed to increase the overcoating temperature.
Because all overcoating processes were carried out in a one-
PL QY of the resulting d-dots was found to be strongly
dependent on the overcoating temperature (Figure 2b-d),
although the UV-vis spectra for a given overcoating step were
found to be almost independent of the growth temperature.
Generally speaking, the higher was the overcoating reaction
temperature, the higher was the PL QY (Figure 2). For the first
growth step, however, the PL QY was generally low and not
strongly temperature dependent (data not shown).
(30) Yu, W. W.; Peng, X. Angew. Chem., Int. Ed. 2002, 41, 2368-2371.
(31) Herman, M. A.; Sitter, H. Molecular Beam Epitaxy: Fundamentals and
Current Status, Second ReVised and Updated Edition; 1996; 453 pp.
(32) Battaglia, D.; Li Jack, J.; Wang, Y.; Peng, X. Angew. Chem., Int. Ed. Engl.
2003, 42, 5035-9.
9
J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007 3341

A R T I C L E S
Pradhan and Peng
Figure 3. Zn to Mn ratio of d-dots and TEM images of the d-dots. The representative electron diffraction pattern of the d-dots is illustrated as the inset in
the right plot.
It should be pointed out that d-dots with high PL QY, between
40% and 70%, could only be made with small MnSe nanoclus-
ters and under relatively high overcoating temperatures. If the
overcoating temperature was below 200 °C, the PL QY was
found to be generally low, in the range between 10% and 30%
for the highest values, similar to that reported in the previous
communication.1 In this temperature range, the size of the MnSe
nanoparticles did not seem to make a definite difference. When
the MnSe nanoparticles were the large-sized nanoparticles as
shown in Figure 1b, the highest PL QY was found to be around
20-30%.
For all growth reactions, the PL QY of the d-dots increased
significantly as the thickness of the ZnSe layer increased (Figure
2). As the ZnSe layer grew thicker, the emission centers, Mn
ions, would be embedded deeper inside a d-dot and more
isolated from the surface defects. Consequently, an enhanced
PL QY should be the result.
Controlling the reactivity of zinc precursor was the key
for boosting the overcoating reaction from 180 °C reported
previously for the same type of d-dots1 to 290 °C. The targeted
Figure 4. Top: Tunable emission peak position of d-dots grown under
different conditions. Bottom: Color change upon the addition of thiol
reactivity of the zinc precursor should be sufficiently high to
ensure the overcoating of the ZnSe layer, but low enough to
avoid homogeneous nucleation of ZnSe nanocrystals. This
balanced reactivity was achieved by using amines as activation
reagents, tuning the chain length of the fatty acid salts, and
adding free fatty acids as inhibitors. A brief discussion is offered
below, and details can be found in the Experimental Section.
When zinc undecylenate (ZnUet2) was used as the precursor,
no free undecylenic acid (UeA) should be added if the
overcoating temperature were chosen to be below 210 °C. Using
the same precursor, UeA in a molar ratio to ZnUet2 of 1:5 would
be needed if the overcoating temperature were selected to be
240 °C. Even more UeA (molar ratio to ZnUet2 being 2.5:1)
would be necessary for a reaction with 260 °C as the overcoating
temperature. ZnSt2 has a longer hydrocarbon chain than does
ZnUet2, and reactions with ZnSt2 as the precursor could thus
be carried out under higher reaction temperatures with less
corresponding free acid. Specially, for ZnSt2-based reactions,
the zinc precursor to stearic acid ratio was 5:1 for 260 °C, 2.5:1
for 290 °C, and 2:1 for 300 °C as the overcoating temperature.
Other zinc carboxylates were also used for achieving desired
overcoating temperatures. For example, zinc acetate was used
for below 180 °C, and a zinc myristate/myristic acid mixture
could be an option for overcoating temperature between 220
and 260 °C. Fatty amines were exploited as both ligands for
the nanocrystals and for the activation reagents. However, a
large excess of amine would sometimes cause the formation of
ligands.
ZnO33 or ZnSe34 nanoparticles, especially for the reactions at
above 250 °C. Hence, periodical injections of amine were
essential as the amines were consumed continuously (see details
in Experimental Section).
In the previous communication,1 tributylphosphine (TBP) was
used for preventing the formation of pure ZnSe nanocrystals as
mentioned above. The current system successfully eliminated
the usage of TBP as a part of zinc precursor, and at the same
time enabled a broad range of reaction temperatures.
The size and composition of the resulting d-dots were
determined by TEM measurements, atomic absorption spec-
troscopy (AA), and X-ray photoelectron spectroscopy (XPS).
For the samples used in the composition measurements in Figure
4, the average sizes of the d-dots were determined by the TEM
images (Figure 3) as 4.3((0.2) and 6.2((0.3) nm, respectively.
The d-dots for both samples were found to possess the same
electron diffraction pattern that matched the pattern of zinc
blende ZnSe nanocrystals, which is the same as what was
reported in the previous communication.1 This is structurally
reasonable as MnSe can exist in zinc blende structures at the
synthetic temperature (>180 °C). The lattice constants for the
zinc blende MnSe and ZnSe are almost identical.
(33) Jana, N. R.; Chen, Y.; Peng, X. Chem. Mater. 2004, 16, 3931-3935.
(34) Li, L. S.; Pradhan, N.; Wang, Y.; Peng, X. Nano Lett. 2004, 4, 2261-
2264.
9
3342 J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007

Color-Tunable Mn-Doped ZnSe Nanocrystal Emitters
A R T I C L E S
Figure 5. Thermal stability of PL of the d-dots grown with different overcoating temperatures. The peak at about 360 and the other small peaks beyond 700
nm are background noise.
The corresponding Mn to Zn ratio in the d-dots determined
by AA followed the expected values calculated from the
amounts of injected precursors (Figure 3). The Zn to Mn ratio
of the smaller d-dots determined by XPS was slightly higher
than that of the AA measurements. The Zn to Mn ratio of the
larger d-dots determined by XPS, however, was found to be
about 3 times higher than that from the AA measurements. This
result is consistent with the targeted d-dot structure, with Mn
ions located within the center of a d-dot. Because AA measure-
ments provide the overall ratio of the two cations in a sample,
so the observed Mn to Zn ratio should reflect the real value.
However, XPS is mostly sensitive to the surface atoms.35 As a
result, it showed more Zn than the real value in a core/shell
d-dot. For the same reason, the slightly higher Zn to Mn ratio
for the d-dots formed by the low dose of zinc precursor implies
that there was likely some interface diffusion between the
preformed MnSe nanoclusters and the ZnSe coating layer for
these small d-dots. This issue will be further discussed later.
their ligands were changed from a neutral species, such as
aliphatic amines, to a charged one. Thiolate ligands, deproto-
nated thiols, are known as stronger ligands to II-VI semicon-
ductor nanocrystals in comparison to amines.36 Following this
hypothesis, we added thiol ligands into the reaction mixture after
the overcoating process was completed. Presumably, the added
thiols shall be deprotonated to form thiolates by the excess
amines existing in the reaction mixture.
Upon heating the reaction mixture with the thiolates at above
220 °C, the PL color of the d-dots in the reaction flask indeed
changed gradually from bright yellow (PL peak at around 582
nm) to orange red (with a peak being around 610 nm) for the
example in Figure 4 (bottom panel). The initial and final PL
spectra of this treatment are shown in Figure 4c. Importantly,
this treatment could stop at any moment and obtain a stable
sample with a defined emission peak position, as long as there
was no further thiol treatment. It should be pointed out that the
thiol treatment process described here did not change the
absorption properties of the d-dots.
However, the incorporation of sulfur anions into the d-dot
lattice might also play a role for the red-shift observed with
thiol ligands. As reported in literature, Mn:ZnS nanocrystals
showed their PL peak at around 600 nm.26 Furthermore, the
red-shift of the PL peak observed with thiol ligands needed an
excessively high temperature, above 220 °C as stated above.
At such high temperature, decomposition of thiol ligands might
have occurred.37 Further studies, both theoretical modeling and
experimental exploration are needed to definitely explain this
interesting color tunability.
By combining the sizing information obtained from TEM
images of the d-dots and the Mn to Zn ratio determined by AA,
one could actually calculate the size of the initial MnSe
nanoclusters that were invisible under TEM. Assuming a
spherical shape for both MnSe nanoclusters and the resulting
d-dots, the obtained diameter of the initial MnSe nanocluster
was estimated to be roughly around 1.5 nm. This value is
consistent with the extremely high-energy absorption edge
illustrated in Figure 1b.
The color tunability of the PL of Mn:ZnSe d-dots, from
about 575 to 595 nm, was identified in the previous com-
munication1 as the thickness of the ZnSe shell increased. Our
initial interpretation was that, as the ZnSe layer grew thicker,
the lattice field around a given Mn center became more
symmetric. As a result, the d-orbitals of the Mn center shall
experience less electric field difference along different directions,
which results in a smaller splitting of the energy levels.
A preliminary experiment was carried out for Mn:ZnSe d-dots
using the growth-doping strategy to see if there is any possibility
to further tune the emission color of this d-dot system. As shown
in Figure 4a, the dopant emission peak of the resulting d-dots
(35) Atkins, P. W. Physical Chemistry, 6th ed.; 1998; 1014 pp.
(36) Aldana, J.; Lavelle, N.; Wang, Y.; Peng, X. J. Am. Chem. Soc. 2005, 127,
2496-2504.
(37) Sigman, M. B., Jr.; Ghezelbash, A.; Hanrath, T.; Saunders, A. E.; Lee, F.;
Korgel, B. A. J. Am. Chem. Soc. 2003, 125, 16050-16057.
If the above interpretation was reasonable, one would expect
a significant red-shift of the PL spectra of a d-dot sample if
9
J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007 3343

A R T I C L E S
Pradhan and Peng
was at about 565 nm, which was about 10 nm blue-shifted in
comparison to the low limit for the corresponding d-dots
synthesized using nucleation-doping. Unfortunately, we have
not optimized the synthesis using the growth-doping strategy,
and a small host emission was observed in the samples (Figure
4a).17,24 The PL QY of these d-dots was quite low as well, not
more than 10%.
Overall, efficient and pure dopant emission from Mn:ZnSe
nanocrystals could be tuned from about 575 nm to about 610
nm (Figure 4). The green-yellow emission at about 565 nm from
the growth-doping (Figure 4a) needs further optimization for
practical applications. The overall tunable window was expanded
to about twice as broad as the one reported in the early
communication.1
Figure 6. PL spectra of the d-dots overcoated with different temperatures.
and the low-energy shoulder became visible at room temperature
(Figure 6b), which can be fitted into two PL peaks, one for the
4
6
Thermal stability of PL of d-dots is a key attractive feature
for them to be considered as the next generation of nanocrystal
emitters. Because of the strong coupling between excitons and
lattice phonons for undoped quantum dots,6 a strong thermal
response has been considered as an intrinsic nature of their
exciton emission. For this reason, the exciton emission of
undoped quantum dots may not be ideal for high power
applications, such as LEDs, solid-state lighting, and solid-state
lasers, for which strong thermal effects are inevitable.
The PL of Mn:ZnSe d-dots grown by nucleation-doping
strategy was found to be still bright up to about 220 °C in the
previous communication.1 This thermal stability is consistent
with the nature of the emission states, which are inner core states
(d orbitals split by the crystal field) and not coupled with the
lattice phonons. Because of its importance, thermal stability of
the d-dots synthesized using the new approaches discussed above
was further studied in a more systematic and quantitative
manner.
For the d-dots grown with an overcoating temperature at
280 °C, a slight increase of intensity from room temperature to
180 °C followed by a slight decrease for the further heating to
300 °C (Figure 5, left) was reproducibly observed by the PL
spectra recorded in situ (see details in Experimental Section).
Generally speaking, a PL intensity change within (10% was
observed in this temperature range. However, this intensity
change was hardly visible by bare eye or the pictures taken by
the digital camera (Figure 5, left) in the temperature range up
to 300 °C (roughly the boiling point of the reaction mixture).
The d-dots grown with overcoating temperature at 180 °C
(Figure 5, right), however, behaved quite differently. A dramatic
decrease of PL brightness (in picture) or PL intensity (in PL
spectra) was observed after the temperature increased beyond
about 180 °C. Although the PL almost diminished at about
280 °C, it recovered reversibly if the d-dots were cooled down
to 180 °C and below.
regular d-d transition (commonly called T1 to A1 transi-
tion)13,38,39 and the other peak centered at about 640 nm. The
640 nm peak is similar to the PL peak position from the emission
centers with a Mn ion next to the Mn emission center (Mn-
Mn centers), instead of all zinc ions, observed in the bulk
samples.40 Such Mn-Mn emission centers, with a lower
efficiency and shorter lifetime,41 are different from the single
emission center for the 4T1 to 6A1 transition. It should be pointed
out that, because both transitions are optically not allowed, their
extinction coefficients are extremely low. The UV-vis mea-
surements with very high concentration of MnSt2 and Mn:ZnSe
nanocrystals, however, did show different absorption bands in
the energy range expected for single Mn center and a mixed
case, respectively (see Supporting Information, Figure S3).
Mn-Mn emission centers only occur when Mn dopant
concentration was very high for bulk samples40 and d-dots grown
by traditional approaches.41 However, the d-dots grown by
nucleation-doping should have a great chance to possess such
Mn-Mn emission centers because the Mn centers should be
all from the central core in a d-dot in this specific case, and the
Mn to host cation ratio was very high (Figure 3) as compared
to typical doped semiconductors. More discussion on this issue
will be given in the Discussion.
An interesting PL peak shift was observed upon thermal
treatment of d-dots. The emission peak position of d-dots blue-
shifted for about 5-8 nm reversibly by heating the solution
from room temperature to about 300 °C. q-Dots, however, are
known to red shift with the increase of temperature (Supporting
information, Figure S5). At this moment, we do not have a
reasonable explanation for this interesting phenomenon.
The PL peak width of an emissive material determines its
color purity. Mn:ZnSe d-dots were always found to possess a
full width at half-maximum (fwhm) around 52((5) nm. This
value is significantly broader than that of the exciton emission
from high-quality CdSe quantum dots, around 22-25 nm
fwhm,42 in the similar optical window. However, the lifetime
of dopant emission should be very long as reported in
literature43,44 and our own results (to be published separately)
because the related transition is optically forbidden. Conse-
quently, the intrinsic PL peak width for the dopant emission in
A close examination of the PL spectra at high temperature
for the d-dots overcoated at different temperatures showed an
interesting difference. As temperature increased, the contour of
the PL spectra did not change for the d-dots overcoated at
280 °C, but the PL spectra for the d-dots overcoated at 180 °C
became significantly asymmetric, with a visible shoulder at the
high wavelength side (Figure 5). To better illustrate this
difference, the PL spectra of the two samples at 250 °C were
plotted together after normalization (Figure 6a).
(38) Jones, G.; Woods, J. J. Phys. D: Appl. Phys. 1973, 6, 1640-51.
(39) Eno, E. E.; Grasser, R.; Scharmann, A. J. Lumin. 1981, 24-25, 261-4.
(40) Chen, W.; Sammynaiken, R.; Huang, Y.; Malm, J.-O.; Wallenberg, R.;
Bovin, J.-O.; Zwiller, V.; Kotov, N. A. J. Appl. Phys. 2001, 89, 1120-
1129.
(41) Suyver, J. F.; Wuister, S. F.; Kelly, J. J.; Meijerink, A. Phys. Chem. Chem.
Phys. 2000, 2, 5445-5448.
(42) Qu, L.; Peng, X. J. Am. Chem. Soc. 2002, 124, 2049-2055.
Interestingly, if the overcoating process took place at an even
lower temperature (150 °C), the PL QY was low (less than 10%)
9
3344 J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007

Color-Tunable Mn-Doped ZnSe Nanocrystal Emitters
A R T I C L E S
Figure 8. Schematic structures of the d-dots formed under different reaction
temperatures.
synthetic chemistry will be also considered to explore possibili-
ties for synthesizing d-dots, which are not only efficient, durable,
and with zero self-quenching but also with a narrow emission
line width.
Figure 7. PLE spectra of the d-dots at different emission wavelengths.
Inset: Normalized PLE spectra at different emission wavelengths, and each
emission wavelength is indicated by a dashed line in the PL spectrum.
Discussion
Although the main goal of this report is on developing
synthetic chemistry of high performance d-dot emitters, some
discussions on the structure of the resulting d-dots should help
for understanding such composition engineered nanocrystals.
At present, complete structural characterization of doped nanoc-
rystals has not been reported. Needless to say, the key structural
feature for a doped nanocrystal is the exact location and
concentration of the dopant ions in a given nanocrystal, which
may be partially assessable with those high-end X-ray absorption
techniques. However, TEM, XPS, and AA results, along with
spectroscopy data, including UV-vis, PL, and electron spin
resonance, can provide some evidence to determine the structure
of the d-dots.
Combining the XPS and AA measurements (Figure 3), one
can conclude that the d-dots reported in this work did have a
quasi core/shell structure, with the Mn ions mostly located close
to the center of the structure. This is consistent with the targeted
structure for nucleation-doping (Figure 8). There is some
additional evidence to support such a core/shell structure. For
instance, the increased PL QY (Figure 2) and improved
chemical/photochemical stability (to be published separately)27
of the d-dots upon the increase of the overcoating thickness
are strong evidence to support the core/shell structures. Both
sets of data indicate that the emission centers were becoming
farther away from the surface as the thickness of the ZnSe
coating layer increased. In addition, both absorption (Figure 2)
and PLE (Figure 7) spectra of the d-dots were found to be
similar to that of quantum shells,32 instead of quantum dots.
This implies that ZnSe likely formed a shell structure, instead
of a uniformly mixed alloy with MnSe for the entire particle.
A core/shell structure indicates that the d-dots structures are
significantly different from that grown by using doped nano-
clusters as the precursors.21 In the latter case, a large portion of
the doped nanoclusters was destroyed and re-formed into large-
sized doped nanocrystals probably through a ripening mecha-
nism. In the nucleation-doping here, the nanoclusters acted as
the nuclei for the growth of the host semiconductor.
d-dots should be smaller, instead of larger, in comparison to
that of the undoped quantum dots. This implies that the large
fwhm for the d-dot PL is from inhomogeneous broadening.45,46
The inhomogeneous line width of the PL of undoped quantum
dots can be revealed by photoluminescence excitation (PLE)
studies if it comes from the size distribution of the corresponding
nanocrystals. In principle, as the emission wavelength monitored
in PLE shifts along the PL spectrum of a quantum dot sample
from high energy to low energy, the PLE spectrum also shifts
from high energy to low energy, if a broad PL peak width was
caused by the size distribution.
To clarify the contribution of the size inhomogenity for PL
peak width, PLE spectra of a d-dot sample were recorded at a
set of emission wavelengths along the entire PL spectrum
(Figure 7). Evidently, the intensity of the PLE spectra changed
as the emission wavelength changed. However, the normalized
PLE spectra were practically the same as shown in the inset in
Figure 7. Therefore, it is reasonable to state that the emission
at any wavelength was contributed from the entire ensemble of
d-dots in a given sample. These results clearly indicate that,
although the size distribution of the d-dots was not monodisperse
(Figure 3), it did not contribute to the PL peak width.
Another type of commonly known spectral inhomogeneous
line width is caused by the fluctuation and variation of the
environment of emissive species. This is why the PL spectrum
of an organic dye is often quite broad, and its PL position and
width can change substantially upon simply changing its
solvents, although an organic dye has a definite molecular
structure. Likely, PL peak width of d-dots was also caused by
the variation of the local environment of the Mn emission
centers. Although the nucleation-doping chemistry demonstrated
here has reached a reasonable degree of control, it was not
difficult to imagine that a significant environment distribution
for Mn emission centers was still in place. Single molecular
spectroscopy experiments are planned to reveal the intrinsic line
width of PL from d-dots, and further optimization of the
The next question is how sharp the interface is between the
MnSe core and ZnSe overcoating layer. The substantial PL
signal from Mn-Mn emission center for the d-dots overcoated
at 150 °C (Figure 6b) indicates that the integrity of the MnSe
nanoclusters maintained reasonably well during the overcoating
process at this low temperature. As the reaction temperature
(43) Bol, A. A.; Meijerink, A. Phys. ReV. B: Condens. Matter Mater. Phys.
1998, 58, R15997-R16000.
(44) Yan, K.; Duan, C.; Ma, Y.; Xia, S.; Krupa, J.-C. Phys. ReV. B: Condens.
Matter Mater. Phys. 1998, 58, 13585-13589.
(45) Albe, V.; Jouanin, C.; Bertho, D. Phys. ReV. B: Condens. Matter Mater.
Phys. 1998, 57, 8778-8781.
(46) Salvador, M. R.; Hines, M. A.; Scholes, G. D. J. Chem. Phys. 2003, 118,
9380-9388.
9
J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007 3345

A R T I C L E S
Pradhan and Peng
increased to above 280 °C, however, the Mn-Mn emission was
decreased to a nondetectable level, and only a strong single Mn
center emission was observed (Figure 6a). This implies that
MnSe nanoclusters were less retained as the overcoating
temperature increased, which indicates a diffused interface
between the core and external shell. In other words, the exact
structure of the Mn:ZnSe d-dots grown by nucleation-doping
at a reasonably high temperature is unlikely a pure MnSe
nanocluster core coated with a pure ZnSe shell. This diffusion,
however, seems to be limited to a relatively thin ZnSe shell as
indicated by the XPS results discussed above (Figure 3), the
ligand modification experiments, and stability tests (to be
published separately).27 Figure 8 (bottom) provides a modified
picture for the resulting Mn-doped ZnSe nanocrystals through
nucleation-doping strategy, a MnxZn1-xSe core coated with pure
ZnSe overcoating layer.
The reversible thermal response of the d-dots overcoated at
180 °C indicates that the diffusion process could not occur for
the formed d-dots with a relatively thick ZnSe shell upon thermal
annealing up to 300 °C. Instead, such diffusion should occur
during the growth process. The reversibility along with the
stability at a given temperature further imply that the d-dots
structure formed in the temperature range studied here was
thermally stable, without significant motion of the cations.
The picture suggested above (Figure 8) is important for the
synthesis of high-quality d-dots through nucleation-doping and
is thus worth to be further confirmed. Ideally, if the interface
were the important region, the overcoating temperature for the
first dose of zinc precursor would be important, at least more
important than the following overcoating processes for thick
ZnSe shells. The experimental results (Supporting Information,
Figure S4) did confirm this prediction, that a higher temperature
for the initial growth of ZnSe was indeed more important than
a higher temperature for the subsequent overcoating for obtain-
ing thermally more stable and more efficient d-dot emitters.
It is known that MnSe has a zinc blende structure identical
to that of ZnSe. This metastable phase of MnSe was known to
be stable above 180 °C. This is probably why we observed a
maximum PL QY at 180 °C (Figure 5). The electron diffraction
patterns of the resulting d-dots (Figure 3) were consistent with
this common structure, and high-resolution TEM results revealed
the single crystalline nature of the d-dots and did not show any
crystal interface between the core and shell. It should be pointed
out that MnSe was well known to be in zinc blende structure
in an epitaxial system if the substrate was in zinc blend, such
as GaAs.47 In our case, MnSe was a small volume portion of
the entire structure (Figure 3), and it is reasonable for MnSe to
adopt a zinc blende structure. Furthermore, the pure 4T1 to 6A1
emission for the d-dots synthesized under a relatively high
overcoating temperature (Figure 5) implies the Mn ion emission
centers for these highly emissive d-dots are likely located on
the common lattice positions and reasonably isolated from each
other.
disappearance of the Mn-Mn emission centers (Figure 6) just
considering the structural imperfection of the ZnSe shell and
without taking interface alloying into account. Furthermore, the
PL QY of the d-dots with large-sized MnSe particles was found
to be always low, below 30%, even when a high overcoating
temperature was applied (see Results). This indicates that a high
growth temperature alone, supposedly resulting in a better ZnSe
shell, is not the sole determining factor for a high PL QY.
Conclusion
In conclusion, synthesis of highly efficient (40-70% PL QY)
and thermally ultrastable (up to boiling point of the solvent,
300 °C) Mn:ZnSe d-dots was studied systematically using
greener approaches by nucleation-doping strategy. The main
discovery of the current work, in comparison to the existing
report,1 is that small size of the MnSe nanoclusters (about 1.5
nm) and a diffused interface between MnSe core and ZnSe shell
were identified as two key parameters for obtaining the high-
quality d-dots. Furthermore, methods were developed to satisfy
these two key parameters by optimizing nucleation temperature,
overcoating temperature, structure of the precursor, and com-
position of reaction solutions at different stages. d-Dots
synthesized under optimized conditions were found to be
extremely thermally stable, up to the boiling point of the solvent
(around 300 °C), which was quantitatively verified by in-situ
measurements. The tunability of the PL color of the Mn:ZnSe
d-dots was expanded to a quite broad window, from 565 to 610
nm. The emission color purity of the d-dots, the PL peak width,
was identified not to be related to the size distribution of the
d-dots. Applications27 of d-dots as biological labeling reagents
and lighting devices offered encouraging evidence for exploiting
these non-heavy metal-containing nanocrystals as practical
emissive materials.
Experimental Section
Materials. Zinc stearate (ZnSt2, 12.5-14% ZnO), stearic acid (SA,
95%), selenium powder (∼200 mesh, 99.999%), tetramethylammonium
hydroxide pentahydrate (TMAH), manganese chloride (MnCl2), doda-
canethiol (DDT), and octadecylamine (ODA, 95%) were purchased
from Alpha Aesar. 1-Octadecene (ODE), mercaptopropionic acid
(MPA), and tributyl phosphine (TBP, 95%) were purchased from
Aldrich. Zinc undecylenate (ZnUet2) was purchased from Gelest.
Undecylenoic acid (UeA) was purchased from TCI, and Rhodamine
3B dye was purchased from Eastman. All chemicals were used without
further purification.
Synthesis of Manganese Stearate (MnSt2). (A) In a typical
synthesis, SA (20 mmol) was dissolved in 30 g of methanol and heated
to 50-60 °C, or sonicated, until it became a clear solution and was
then allowed to cool to room temperature. The solution of TMAH was
prepared by taking 20 mmol of TMAH in 10 g of methanol and mixing
with the SA solution. The mixture was stirred for 15 min to ensure the
reaction had gone to completion. To this solution, 10 mmol of MnCl2
dissolved in 10 g of methanol was added dropwise with vigorous
stirring, and a white precipitate of MnSt2 slowly flocculated. The
precipitates were repeatedly washed with methanol and dried under
vacuum. (B) Alternatively, MnSt2 was also prepared by mixing SA
and MnCl2 solution together in a 2:1 molar ratio, and the base, TMAH,
was added dropwise to obtain a brownish precipitate of MnSt2. Digital
pictures of the two types of synthesized MnSt2 are shown in
Figure S1.
One may argue that the high PL QY and enhanced thermal
stability of the d-dots grown under relatively high overcoating
temperatures (Figures 2 and 5) was due to the high quality of
the ZnSe shell formed at a higher temperature. At this moment,
it is impossible to completely exclude the contribution from
this possibility. However, it would be difficult to explain the
Preparation of Stock Solutions. TBPSe solution was prepared
inside a Glovebox by adding 1.9 g of Se to 8 g of TBP. Several types
of zinc precursor solutions were prepared according to the type of
(47) Eunsoon, O. h.; et al. Phys. ReV. B 1993, 47, 7288.
9
3346 J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007

Color-Tunable Mn-Doped ZnSe Nanocrystal Emitters
A R T I C L E S
zinc carboxylate salt used. For example, when ZnUet2 was used, the
zinc precursor solution was prepared by mixing 1.9 g of ZnUet2 and
0.6 g of UeA into 8 g of ODE. For the case of ZnSt2, ZnSt2 (1.8 g) and
0.4 g of SA were dissolved into 7 g of ODE. The amount of free acid
for each was varied depending on the injection temperature as described
in the text.
(C) Tunable d-Dot Emission from ∼580 to ∼610 nm. For any of
the aforementioned reaction systems, the emission peak position can
be tuned by the addition of thiol. To maintain a high PL QY, the
thickness of the ZnSe layer must reach a reasonable thickness, with its
absorption band-edge beyond 440 nm, before the addition of the thiol.
Injection of the thiol was performed at a temperature above 220 °C.
For a reaction with 12 g of ODE, 2 g of ODA, 0.03 g of MnSt2, and
0.5 g of ZnSt2, nearly 2 g of dodecanethiol was injected. After the
thiol injection, the reaction was allowed to continue until the PL peak
position shifted from about 580 nm to a designated wavelength, but
not beyond 610 nm. For a reaction in a 50 mL reaction flask, it took
about an hour for the emission to shift to 610 nm at 260 °C. The reaction
was monitored by UV-vis and PL measurements. During this process,
the UV-vis spectra remained unaltered, although the PL position shifted
from ∼580 to 610 nm.
(A) Typical Synthesis of d-Dots. ODE (25 mL) and 0.1 g of MnSt2
were loaded into a 100 mL three neck flask and degassed at 100 °C
for 20 min by bubbling with argon. The temperature was then raised
to 280 °C. In a separate vial, ODA (1 g) and 1 mL of TBPSe stock
solution were mixed and heated to ∼70 °C, and injected into the above
reaction flask at 280 °C. After the injection, the color of the solution
rapidly turned faint yellow, which indicated formation of MnSe
nanoclusters. The reaction was cooled to 260 °C and annealed for 60
min. The reaction temperature was then set at 240 °C (or another
designated overcoating temperature) for ZnSe overcoating. Next, heated
ZnUet2 stock solution in ODE (3 mL) was injected into the reaction
flask. Immediately after the ZnUet2 injection, the solution started to
glow yellow under UV-light, showing the growth of ZnSe on MnSe
nanoclusters. The temperature was further cooled to 230 °C, and 0.5 g
of heated ODA in 0.5 g of ODE was injected into the reaction mixture
to activate the zinc carboxylates. The remaining ZnUet2 stock solution
was injected by adding 3 mL after 15 min intervals, followed by ODE/
ODA injection. The growth process was monitored through successive
UV-vis and PL measurements. Finally, the reaction was cooled to
room temperature, and the nanocrystals were precipitated using acetone.
Nanocrystals (about 200 mg) produced using this procedure were
reproducible, demonstrating 25-40% QYs. If the overcoating temper-
ature was set as 250 °C with calculated amounts of ZnUet2 and UeA,
the resulting d-dots typically had 30-45% QYs.
(B) d-Dots Formed by High Overcoating Temperature. To reduce
the possibility of forming pure ZnSe particles, free acids were mixed
with zinc precursors as inhibitors. Periodic introduction of amines as
discussed in the above paragraph must be performed to avoid
accumulation of excess free acid introduced by the injections, which
maintained a reasonable reactivity for the zinc fatty acid/salt fatty acid
mixture in the reaction solution.
Formation of MnSe nanoclusters was performed according to the
procedure discussed in section A. After the MnSe solution was annealed
at 260 °C for 60 min, the reaction temperature was increased to
290 °C for overcoating of ZnSe. The zinc precursor solution was
prepared by dissolving 2 g of zinc stearate and 0.4 g of stearic acid
into 8 g of ODE. This solution was maintained warm using a heat-
gun, degassed with argon, and injected into the reaction flask, in 3
mL/15 min intervals as discussed above, which was set at 290 °C.
Immediately after the injection, the reaction mixture was able to emit
bright yellow under UV-light. The reaction was cooled down to
260 °C, and 0.5 g of ODA in 0.5 g of ODE was injected into the
reaction mixture to maintain the amine concentration. Other procedures
were similar to the ones described in section A. The resulting d-dots
had a measured QY of 35-55%. The fwhm of the emission peak was
52((2) nm for both systems.
Optical Measurements. UV-vis spectra were recorded on an HP
8453 UV-visible spectrophotometer. Photoluminescence spectra were
taken using a Spex Fluorolog-3 fluorometer. The PL QY was measured
following a suggested protocol in the manual of the Fluorolog-3 (see
Supporting Information). Infrared spectra were obtained on a Bruker
Tensor 27 spectrophotometer. IR samples were prepared by dropping
hot aliquots directly taken from the reaction solution onto a NaCl disc.
X-ray photoelectron spectra (XPS) were recorded using a Kratos Axis
HSI spectrometer. Samples were prepared by dropping purified d-dot
solution onto platinum films. Atomic absorption (AA) measurements
were carried out using a GBC 932 Plus spectrophotometer after
digesting the samples in concentrated nitric acid.
In Situ Measurements. Photoluminescence of d-dots at high
temperature was measured directly from the reaction flask using an
Ocean Optics USB 2000 spectrophotometer. Positions of the UV-source
and fiber optic probe were fixed throughout the measurements.
Successive photoluminescence spectra were recorded during cooling
and heating of the reaction system from 290 to 100 °C and vice versa.
Transmission Electron Microscopy (TEM). TEM images were
taken on a JEOL X-100 electron microscope using a 100 kV
accelerating current at 50k magnification. Specimens were prepared
by dipping a Formvar-coated copper grid into a toluene solution of the
nanocrystals, and the grid with the nanocrystals was dried in air.
Selected area electron diffraction pattern (SAED) was taken with a
camera length of 120 cm.
Acknowledgment. Financial support of this work was
provided by the National Science Foundation. Synthesis of Mn
fatty acid salt precursors was developed by Dr. Yongfen Chen
in Peng’s research group. Discussions with David Battaglia at
NN-Labs are also acknowledged.
Supporting Information Available: Supporting results men-
tioned in the text. This material is available free of charge via
the Internet at http://pubs.acs.org.
JA068360V
9
J. AM. CHEM. SOC. VOL. 129, NO. 11, 2007 3347