Journal of Luminescence 27 (1982) 455—462 455
North-Holland Publishing Company
FLUORESCENCE QUANTUM YIELDS OF SOME RHODAMINE
DYES
R.F. KUBIN and A.N. FLETCHER
Chemistry Division, Research Department, Naval Weapons Center, China Lake, California,
USA
Received 10 February 1982
Fluorescence quantum yields of seven rhodamine dyes were measured relative to quinine
sulfate dihydrate (QSH) in 1.0 N H2S04. The values obtained were rhodamine 6G (0.95), B
(0.65), 3B (0.45), 19 (0.95), 101 (0.96), 110 (0.92), 123 (0.90) at 25.0°C. Effects of
temperature on the quantum yields of rhodamine B and QSH show a large temperature
coefficient for rhodamine B and a significant one for QSH. Dye concentration was found to
be critical in reporting observed fluorescence wavelength maxima.
1. Introduction
Fluorescence quantum yields of several rhodamine dyes (i.e., B, 101, 6G)
have recently been remeasured [1—3].For some of these dyes there has been
some question of the value of the quantum yield. For example, it has been
noted that the value for rhodamine B ranges from 0.69 to 0.97 [2a] in ethanol.
Two values have been reported for rhodamine 6G, 0.94 [1] and 0.88 [3] in
ethanol.
At this laboratory we have at present a continuing need for reliable
quantum yield determinations on new dyes; e.g., energy transfer agents and
laser dyes. Against this background we have measured the fluorescence quan-
tum yields of rhodamine B, 3B, 6G, 19, 101, 110, 123 using quinine sulfate
dihydrate (QSH) in 1.0 N H2S04 as the primary standard. We have also
employed rhodamine 6G as a second standard in order to be able to detect any
change that might occur in the standard solutions with time. Our results are
summarized in table 1. We have also investigated the effects of temperature
and concentration.
There are some differences in the literature between the reported fluores-
cence wavelength maxima and some of our observed values. We ascribe these
differences to the various concentrations used in making these fluorescence
measurements. Our feeling is that the reported maxima should be those
observed for essentially infinite dilution.
0022-231 3/82/0000—0000/$02.75 © 1982 North-Holland
456 R. F. Kubin, A. N. Fletcher / Fluorescence quantum yields of some rhodamine dyes
Table I
Quantum yields at 25°C
Compound (mol/l) A5 (nm) A5 peak ~ Remarks
c(Xl0
7) (nm)
QSH 9.24 350 458 0.55 Standard
Rhodamine 6G 1.05 248—528 558 0.95
Rhodamine B 1.58 259—542 568 0.65 Basic EtOH
Rhodamine 3B 2.48 257—556 580 0.45
Rhodamine 19 1.47 245—516 544 0.95 Basic EtOH
Rhodamine 101 1.34 265—563 588 0.96 Basic EtOl-1
Rhodamine 110 1.85 267—501 524 0.92 Basic EtOH
Rhodamine 123 1.76 245—510 534 0.90
2. Experimental
All rhodamine dyes were obtained from either Eastman Kodak or Exciton
and were used as received. The QSH was recrystallized from water and dried as
described earlier [4]. The QSH was dissolved in 1.0 N H
2S04 because that was
the solvent used by Melhuish [5] when he determined the accepted value of
0.546 for the quantum yield of QSH.
The absorbances of all the dye solutions were determined on a Cary 14
recording spectrophotometer operated at a lowered dynode voltage setting.
This procedure was used to widen the slits and hence the bandpass of the Cary
in order to approach the 4 nm bandpass used on the Perkin Elmer MPF-44B
Fluorescence Spectrophotometer. This was the narrowest bandpass used for a
good signal to noise ratio. The MPF-44B is used in conjunction with a Perkin
Elmer DCSU-1 computer for making corrected spectra. The excitation system
(xenon lamp, optics, and grating) is calibrated by using an optically dense
solution of rhodamine B in ethylene glycol for the wavelength region 200 to
600 nm. A triangular cuvette is used so that the fluorescence of the rhodamine
B is centered in the detector field of view. All the exciting light is assumed to
be reemitted with the same efficiency independent of wavelength. The instru-
ment is in essence a quantum counter. The detector itself is calibrated using a
secondary standard tungsten lamp. The computer has stored in a read only
mode the spectral data for the tungsten lamp and is supposed to cover the
400—900 nm range. However, we have found the data beyond approximately
700 nm unreliable. There is a blank in the memory from 724 to 736 nm and the
output becomes unusually noisy and goes off scale beyond 800 nm. The
manufacturer is presently working on resolving these problems. The detector
response can also be calibrated using the internal xenon lamp. The latter
procedure has a long wavelength limit of 600 nm. The quantum yield of
R. F Kubin, A.N. Fletcher / Fluorescence quantum yields ofsome rhodamine dyes 457
rhodamine 110 was measured relative to QSH using the two different calibra-
tions of the detector and the results differed by about 8%, the upper error limit
of the results reported here. The summary given in table 1 is for the tungsten
lamp calibration.
The instrument has a very good 4-cell turret head constant temperature
cuvette holder as an accessory. This accessory is indispensable for making
reasonably quick quantum yield measurements. Cuvettes in which the fluores-
cence area of QSH has been measured under identical conditions are used. The
cuvettes are then used to hold QSH, dye solution, and pure solvents. It is
necessary each time quantum yields are determined to keep the QSH standard
in the instrument so that frequent checks can be made to compensate for small
amounts of dynode voltage drift that occur. If this voltage changes by ±4V in
650, the observed change in the output signal is approximately 3 to 4%.
Using fresh standard solutions, the quantum yield of rhodamine 6G was
measured as 0.95 assuming a value of 0.55 for QSH. These same solutions,
using fresh samples, were then used as comparison standards to determine
unknown quantum yields. In this procedure the quantum yield of each
standard was determined in terms of the other and provided a check on the
integrity of the standard solutions. For QSH the range of values of 25.0°Cwas
0.53 to 0.56 assuming a value of 0.95 for rhodamine 6G. For rhodamine 6G the
range of values of 25.0°Cwas 0.92 to 0.96 assuming a value of 0.55 for QSH.
Thus, the reproducibility of this procedure was about ±5%. Repeat determina-
tions of dye quantum yields using new solutions have a reproducibility of
±8%. Within a given set, the quantum yields of various dyes calculated using
the two standards were observed to vary from agreement to ±5% difference.
We have reported the average of these determinations. This difference, at least
for the few dyes thus far determined, does not seem to depend on the
excitation wavelength range or on the standard used. For instance, for rhoda-
mine B the value of the quantum yield obtained using QSH or rhodamine 6G
was the same. In the case of rhodamine 3B, the quantum yield value based on
rhodamine 6G was about 4% lower than the value obtained using QSH as the
reference standard.
All dye solution dilutions were made by weight. The pure and basic ethanol
solvents were stored under argon in inert atmosphere vessels. Withdrawal of
the solvents was assumed to air saturate them. Absorption spectra taken over a
period of time showed that dyes in pure ethanol were stable for at least 2
months. Dye solutions in basic ethanol had noticeably increased absorption in
the UV region (220 to 260 nm) after as short a time as 2 days. A major fraction
of this increase is due to the ethanol alone because once the basic ethanol is air
saturated the increase in absorption becomes noticeable.
When using the square cuvettes, dye concentrations were held to a value
yielding an absorbance of ~ 0.014 in a 1 cm path. At these absorbances
self-absorption due to overlap of the principal (S1 ~— S0) excitation band and
458 R. F. Kuhin, A.N. Fletcher / Fluorescence quantum yields of some rhodamine dyes
the emission band is greatly minimized. This is very important for small Stokes
shift dyes such as the rhodamines. Because the exciting light is attentuated by
only about 3%, the assumptions of constant light flux in the detector solid
angle and isotropic emission are valid. In the MPF-44B, the fluorescence is
viewed at 90°to the incident excitation light. Solution concentrations are the
molarities corrected to 25°C.
The absorbance criterion above applied to the largest absorption band,
generally the S1 ~— S0 band for the dyes studied here. Because of the problems
encountered in basic ethanol, a standard procedure was used to calculate
absorbance of other than the principal band. The absorbance of the principal
band which is stable for long periods even in basic ethanol was measured using
10 cm cells on the Cary. An excitation spectrum of the dye at the concentration
used for the quantum yield determination was then used to obtain peak height
ratios for the much smaller peaks at other wavelengths. These ratios were then
used to calculate the absorbances at these other wavelengths from the one
measured value. It is recognized that this procedure in fact assumes that the
quantum yield is constant as a function of wavelength.
For the temperature studies a Neslab RTE4 circulating bath was used. This
bath has the capability of holding to better than ±0.05°C.Initially the thermal
cell holder was probed with a copper constantan thermocouple as were the
individual cuvettes. Temperatures throughout the block were constant to better
than 0.1°C,the reading limit of our thermocouple readout, a Doric DS-lOO-T3.
In the range 0 to 60°Cthermal gradients at equilibrium were small between
cell block and sample. At 60°C the gradient was approximately 4°C. In
making the measurements, equilibrium was defined as 5 mm with no tempera-
ture change. Temperatures recorded were those measured in the dye solutions.
The effects of concentration were studied using triangular cuvettes. These
cuvettes insure that the exciting beam strikes the center of the detector solid
angle. The cuvette is so oriented that the fluorescence is filtered by the
solution.
Fluorescence wavelengths were read directly from the MPF-44B corrected
dye spectrum. The excitation wavelengths used were the observed maxima read
from the Cary 14. Under the run conditions used, using a steel rule, the
maximum reading error in these wavelengths was 4.7 A. Comparison of
absorption spectra and excitation spectra taken on the Cary 14 and the
MPF-44B, respectively, showed a precision of ±1 nm for reading the MPF-44B
.x—y recorder.
3. Results and discussion
Table 1 gives the quantum yields found for the standard rhodamine dyes.
The data are in good agreement with the previous recent literature values and
R.F Kubin, AN. Fletcher / Fluorescence quantum yields ofsome rhodamine dyes 459
thus add substantiation to these values. For rhodamine 6G, the value of 0.95
agrees with Drexhage [6] (0.95) and Butenin et al. [1] (0.94 ±0.01) and is the
recommended value over the value of 0.88 [3]. However, the precision of
quantum yield values of ±5to 10% does not make the lower value that far out
of line. The quantum yield determined for rhodamine B is in excellent
agreement with the results of Schwerzel and K.losterman [2a]. From their graph
we read a value of 0.64 at 25°C.The quantum yields for rhodamine 101 and
3B again agree with Drexhage’s results [6]. Only in the case of rhodamine 110
is there a large difference. Our results indicate a higher quantum yield of
0.92 ±0.03 over the value of 0.85 [6]. The values differ by the maximum error
observed in the reproducibility of our results. These results, using in essence
two comparison standards, demonstrate that there is no difficulty in spectral
range compatibility in using QSH in 1.0 N H2S04 as a primary standard, a
fear that has been expressed [2b]. Rhodamine 6G is an excellent candidate as a
I I
0.80— -
0.70— -
0.60— o -
O 0—..El...
~~-0-~~
0.50— —----.--...- -
~~~~0~
0.40— -
0.30 - 0 RHODAMINE B -
0 QUININE SULFATE HYDRATE
020 I0 10 20 30 T °~ 40 50 60 70
Fig. I. Temperature variation of the fluorescence quantum yield of quinine sulfate dihydrate and
rhodamine B.
460 R. F. Kuhin, A. N. Fletcher / fluorescence quantum vield,~of arnie rhodamme £]ve.v
100 I I I I I I I I
90 - CONCENTRATION + -
lx1o~M
~\t1~1 I ———— ~xio~
80 - i~\~ \ \ sxio4 -I 5X105
- — 2X1
II! 1 I
70 - / II ~ \ \ 1.5X10
ii II 1 I
60 — I 1 \
(~) I I~II I
2 I j I
wH. I
2 I
_50 - i ~ -
:1 I
40 — I ~ I —
I I
0 ~. ‘
30 fi I
20- / i
1/
I I .4L-”I I
400 440 480 520 560 600 640 680 720 760
WAVE LENGTH rn,
Fig. 2. Concentration dependence of the fluorescence spectrum of rhodamine 6G. Spectra taken
in a triangular cell except for 1.5 X l0 M solution which was in a square cell.
primary standard. Its carboxyl group is esterified: thus the dye is not subject to
the acid-base equilibria in polar solvents [6] that we also have observed with
those dyes having a nonesterified carboxyl group.
Schwerzel and Klosterman [2a] made a study of the temperature depen-
dence of several rhodamine dyes. We were particularly interested tn the
temperature dependence of rhodamine B and QSH. Fig. I shows the observed
temperature dependence of these dyes. The results point up the necessity for
good temperature control in order to make precise measurements. Rhodamine
B has a strong temperature dependence. The results are in essential agreement
with those of Schwerzel and Klosterman. The temperature dependence of
QSH, while less marked, is still significant. Over the 50° range of the
R. F Kubin, A. N. Fletcher / Fluorescence quantum yields of some rhodamine dyes 461
measurements the slope of the line is 0.0020/deg. Thus at 25°Cthe quantum
yield changes by approximately 0.37% per degree. This result is somewhat
higher than the range given by Melhuish [5]. It is also noted that the limited
data suggest that the least squares line drawn for QSH should be a curve.
The final point is that the quantum yield should be determined on as dilute
a solution as possible. This consideration is especially important for small
Stokes shift dyes such as the rhodamines. Absorption of fluorescence can
become a serious problem. This is shown dramatically in fig. 2 for rhodamine
6G. As the concentration is increased there is an apparent shift in the peak
fluorescence of almost 30 nm over the concentration range 1.5 x l0~ to
I x i03 M. Lowering the concentration below about 1 X 106 M does notfurther affect the apparent peak fluorescence of the rhodamines studied. The
values for the maximum peak fluorescence given in table 1 are somewhat
shorter than reported elsewhere in the literature [1,3,7] and are for concentra-
tions in the l0~ M range as shown.
Differences in quantum yields reported in the literature can be ascribed not
only to temperature but to concentration errors as well. Unless measurements
of relative quantum yields at higher concentrations; i.e., where absorbances
greatly exceed 0.0 14, are made under the conditions of matched absorbance of
the test material and the standard at the exciting wavelengths, the quantum
yield values will not be valid. When large amounts of the excitation light are
absorbed, fluorescence within the cuvette will not be uniform. The emission
will be high at the entrance of the excitation light into the cuvette and decrease
with distance from the entrance face. In addition, if the Stokes shift is not
large, reemission errors will occur. For small Stokes shift dyes, if the quantum
yield is low, absorption of fluorescence would further lower the measured
value. However, as is the case with the rhodamine dyes, if the quantum yield is
high, then anomalously large values can be obtained due to reemission of light
within the detector field of view from volumes not directly illuminated by the
exciting light beam.
‘the quantum yields in this work were calculated using the formula given by
Demas and Crosby [8] including the index of refraction term and using the
fraction of light absorbed rather than the absorbance.
References
[II A.V. Butemn, B. Ya. Kogan and NV. Gundobin, Opt. Spectrosc. (USSR), 47 (1979) 568.
12a] RE. Schwerzel and N.E. Klosterman. NatI. Bur. Standards, SP 526, 1978. (Private communi-
cation. Complete data to be published.)
[2bl T. Korstens and K. Kobs. J. Phys. Chem. 84 (1980) 1871.
(3] J. Olmsted III, J. Phys. Chem. 83 (1979) 2581.
[4] AN. Fletcher, Photochem. Photobiol. 9 (1969) 439.
[5] W.H. Melhuish. J. Phys. Chem. 65 (1961) 229.
462 R. F. Kuh,n. A. N. Fletcher / Fluorescence quantum yields ofsome rhodamine dye.s
[6] K.H. Drexhage. Structure and Properties of Laser Dyes, in Dye Lasers. F.P. Schafer. ed., 2nd
edition (Springer, 1977).
[7a] Laser Dyes. Exciton Chemical Company, Inc.. Dayton. OH.
[7b3 Kodak Laser Products. Pub. No. JJ-169. Eastman Kodak Company, Rochester. NY.
[8] J.N. Demas and GA. Crosby. J. Phys. Chem. 75 (1971) 991.