The Effect of Annealing Conditions and Concentration on

1
The Effect of Annealing Conditions and
Concentration on 5D3 ? 7FJ Emission in
Terbium-doped Sol-gel Glasses Colleen
Gillespie and Dan Boye, Davidson College,
Davidson, NC Ann Silversmith, Hamilton College,
Clinton, NY
Abstract Sol-gel glasses have optical properties
similar to those of traditional melt glasses, but
are appealing because they can hold a higher
concentration of dopants due to their lower
processing temperatures. In silicate sol-gel
doped with trivalent terbium, the intensity of
fluorescence from the 5D3 level to the 7FJ (J
06) ground state manifold levels is highly
dependent on both terbium concentration and
annealing conditions. 5D3 emission is observed
in glasses annealed at 750?C and increases in
intensity with increasing annealing time and with
higher annealing temperature. The relative
intensity of emission from the 5D3 state
decreases with increasing Tb3 concentration. A
cross-relaxation process involving two nearby
Tb3 ions depopulates the 5D3 level and causes
this concentration quenching.
Motivation Sol-gel glasses Sol-gel synthesis
is a low temperature was to prepare optically
transparent materials. Because the sol-gel is
prepared at room temperature, it is easy to
include metallic, organic, and inorganic
additives. The optical properties of a sol-gel
that has been heated to 1000ºC are similar to
those of traditional melt glasses, but since this
temperature is below the melting point of the
material, the sol-gels can hold more dopants than
melt glasses. Sol-gels with rare earth ions as
dopants are used in phosphors, solid-state
lasers, and amplifiers. We can use the optical
properties of these materials to characterize the
interactions between the dopants and the
surrounding material.
Picture of Samples
Theory Concentration Dependence
Theory Annealing conditions
Which sample looks bluer? So, which sample has
the lower concentration? The sample on the left
has a concentration of 0.5, and the one on the
right is 0.1. The 0.5 sample has very little
emission from the 5D3 level (blue light) while
the 0.1 sample has a substantial amount of
emission from this level, and therefore looks
blue-ish.
Hydroxyl groups are present in the sol-gel
materials. They have an absorption band between
2000 and 4000 cm-1. This allows electrons in the
5D3 level to relax to the 5D4 level by losing
energy to another hydroxyl group combined with a
lattice vibrational mode. This means that there
is virtually no fluorescence from the 5D3 level.
However, the number of hydroxyl grounds in the
material can be significantly reduced by
annealing the samples at a high temperature.
Consequently, the 5D3 emission becomes comparable
to 5D4 emission. When the samples are taken to
750ºC, the 5D3 peak becomes detectable. Taking
the samples to higher temperatures and for longer
dwell times at those temperatures, the intensity
of the 5D3 peak becomes stronger. However, if
the samples are annealed at too high a
temperature, they can start to crystallize or
crack. Therefore, we wanted to find an ideal
annealing temperature and time to have a high 5D3
emission without any damage to the sample.
Exposure to Humidity
Sol-gel Recipe

• Dissolve 11.8 mg of terbium nitrate
  (Tb(NO3)35H2O) and 20.3 mg of
• aluminum nitrate (Al(NO3)39H2O) in 7.8 mL
  deionized water.
• ?Add 20 µL concentrated nitric acid and 4.00 mL
  TMOS
• (tetramethylorthosilicate, 99) to solution
  and stir for 10 minutes
• ?Put sol into 4 tightly capped polystyrene
  disposable test tubes
• ?Gel at room temperature for 72 hours
• ?Ramp at 5ºC/hr to 60ºC, then sit for 48 hours
• ?Ramp at 5ºC/hr to 90ºC, then sit for 48 hours
• ?Ramp at 2ºC/hr to 110ºC, then sit for 48 hours
• ?Let cool to room temperature

A cross-relaxation process involving two Tb3
ions depopulates the 5D3 level and reduces 5D3
emission intensity. At higher terbium
concentrations, the terbium atoms are closer
together, so there is stronger cross-relaxation.
Therefore, we predict that as terbium
concentration increases, the intensity of the 5D3
peak will decrease.
Results Annealing Conditions
Results Concentration Dependence
After the sample is taken out of the oven, it
begins to reabsorb hydroxyl groups, and the 5D3
peak begins to shrink. Here is a close up of the
5D3 peaks at three consecutive times. This decay
is enhanced by increasing the hydroxyl
concentration by putting the sample in water for
a few minutes.
Terbium Energy Levels
Conclusions

• We determined that heating the samples to 900ºC
  with a 12 hour dwell gives optimal optical
  properties without damaging the sample.
• We confirmed that the 5D3 emission intensity
  decreases with increasing terbium concentration
• Now that we have determined the best protocol
  for processing these samples, we can use our
  knowledge to begin working with erbium-doped
  sol-gels.

I heated the sample to 800, 900, and 1000ºC for
various dwell times. I then plotted the ratio of
the 5D3 ? 7F4 transition to the 5D4 ? 7F6
transition. At higher temperatures and for
higher dwell times, the ratio increases. The
sample shattered after being at 1000ºC for 4
hours.

• Project supported by a grant
• from the NSF-NRI program

This is a partial energy level diagram of
trivalent terbium, with labeled transitions
corresponding to observed fluorescence lines.
The samples were excited with 240 nm excitation
light. The 5D4 ? 7F5 transition dominates the
emission spectrum and produces a green color.
Plotting spectra of four different
concentrations, it is clear that 5D3 emission
decreases as terbium concentration increases.