Development of a heat-sensor for the measurement of reaction enthalpy based on laser deflection

Article abstract of DOI:10.1016/S0040-6031(00)00641-9

A new method for the measurements of reaction enthalpies based on a laser deflection heat-sensor was developed. The heat released or absorbed from a chemical reaction in an aqueous phase causes the formation of thermally driven refractive index gradients in an adjacent organic phase. This effect is monitored by the change in the deflection angle of a laser light in the organic phase. The instrument was tested in the complexation reaction of EDTA with magnesium (endothermic) and calcium (exothermic). It is shown that the direction of the deflection angle is indicative of the sign of the enthalpy change. The magnitude of this enthalpy change can be calculated using the intensity of the maximum deflection and a calibration slope obtained using the reaction enthalpy of the NaOH/HCl neutralization. The heat limit of detection (HLOD) was calculated to be 60 mJ. Possible improvements in the HLOD are discussed. (C) 2000 Elsevier Science B.V.

Full text of DOI:10.1016/S0040-6031(00)00641-9

Thermochimica Acta 362 (2000) 179±184
Development of a heat-sensor for the measurement
of reaction enthalpy based on laser de¯ection
*
Thomas Schneider, M a rio J. Politi, Mauricio S. Baptista
Departamento de Bioqu Âõ mica, Instituto de Qu Âõ mica, Universidade de S aÄ o Paulo, C.P. 26077,
C.E.P. 05599-970 S aÄ o Paulo, SP, Brazil
Received 26 May 2000; received in revised form 15 August 2000; accepted 21 August 2000
Abstract
A new method for the measurements of reaction enthalpies based on a laser de¯ection heat-sensor was developed. The heat
released or absorbed from a chemical reaction in an aqueous phase causes the formation of thermally driven refractive index
gradients in an adjacent organic phase. This effect is monitored by the change in the de¯ection angle of a laser light in the
organic phase. The instrument was tested in the complexation reaction of EDTA with magnesium (endothermic) and calcium
(
exothermic). It is shown that the direction of the de¯ection angle is indicative of the sign of the enthalpy change. The
magnitude of this enthalpy change can be calculated using the intensity of the maximum de¯ection and a calibration slope
obtained using the reaction enthalpy of the NaOH/HCl neutralization. The heat limit of detection (HLOD) was calculated to be
6
0 mJ. Possible improvements in the HLOD are discussed. # 2000 Elsevier Science B.V. All rights reserved.
Keywords: Laser; De¯ection; Heat-sensor; Enthalpy
1
. Introduction
optical properties of materials (solids, liquids and
gases), as well as photophysical properties of excited
states [7]. The so-called photothermal techniques are
generally very sensitive, i.e. detection limits of pico-
joules and single molecule determinations have been
reported [8,9]. However the application of laser tech-
niques for the study of thermo-properties of chemical
reactions has not yet been properly explored.
Wu and coworkers were the ®rst to demonstrate that
the heat released in chemical reactions causes the
de¯ection in a laser beam positioned in an adjacent
medium. This phenomenon was used in several ana-
lytical applications including acid base titration, the
analytical determinations based in enzymatic reac-
tions in tandem with ¯ow injection devices, and gas
diffusion measurements [10±13]. We have recently
shown its application for the quantitative determina-
tion of singlet oxygen [14].
Microcalorimetric techniques are of major impor-
tance to almost every ®eld of science [1,2]. Nowadays,
isothermal, reaction and temperature scanning calori-
meters reach detection limits lower than microjoule,
allowing sensitive investigations of protein denatura-
tion processes, phase diagram determinations and
study of adsorption phenomena at interfaces [3,4].
Although providing good sensitivity these instruments
are yet expensive, and are invasive in nature [5].
Laser techniques have revolutionized several ®elds
of science specially physical-chemistry [6]. The inter-
action of laser light with heat-induced temperature
gradients has allowed the characterization of thermo-
*
Corresponding author. Fax: ‡55-11-8155579.
E-mail address: baptista@iq.usp.br (M.S. Baptista).
0040-6031/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 0 - 6 0 3 1 ( 0 0 ) 0 0 6 4 1 - 9

1
80
T. Schneider et al. / Thermochimica Acta 362 (2000) 179±184
Given the fact that analytical determinations, i.e. to
nd out analyte concentrations, are just an indirect
cw He±Ne laser (Oriel-79309) is focused in the sam-
®
ple cuvette (1 cm quartz) in the CCl phase (ꢁ3 mm
4
application of this essentially calorimetric method, our
goal in this work is to demonstrate the application of
this laser de¯ection heat-sensor for the determination
of enthalpy changes of chemical reactions. It will be
shown that by using a low-cost, non-contact detector,
it is possible to perform the determination of exother-
mic and endothermic enthalpies of complexation reac-
tions.
below the interface with water). A knife-edge blocks
around 50% of the beam and a converging lens is used
to refocus the beam in the photodiode (OPT101-DIP,
Burr±Brown). The signal is digitized by a 12 bit A/D
board (CIO-AD08, Computer Boards) installed in a
386 PC computer (TRI). The total volumes of the
aqueous and organic phases are 0.2 and 1.0 ml, respec-
tively. The CCl is chosen because of its large dn/dT
4
value, which maximizes the refractive index gradient
g2 e0 n8 Ce r. ated [10,11]. The experiments are run at T ꢁ
2.3. Methods
2
. Experimental
2
.1. Materials
Carbon tetrachloride was spectroscopic grade
Merck). CdCl , MgCl , were obtained from Merck
Chemical reactions (HCl/NaOH neutralization and
2
‡
2‡
(
Ca /Mg complexation with EDTA) are initiated by
adding a small volume (few microliters) of a solution
that has one of the reactants to the aqueous phase
containing the other reactant. The solutions are pre-
pared with a large molar excess of one of the reactants.
Considering the concentration excess and that the
equilibrium constants of these chemical reactions
2
2
and were dried for 5 h at 1008C. Ethylenediaminete-
traacetic acid disodium salt dihydrate (ꢀ99%,
Aldrich) (EDTA) was used without further treatment.
Water was doubly distilled in an all glass apparatus,
and further deionized and ultra-®ltrated (Milli-Q).
11
� 1
2
.2. Instrumentation
are 10
M or higher [15], the forward chemical
reaction is completed in a pseudo-®rst order regime.
The heat liberated or absorbed by the chemical
reaction in the aqueous phase (depending on the sign
of the enthalpy change) propagates to the organic
phase causing a temperature gradient, which ulti-
mately causes a refraction index gradient. The laser
beam passing through the organic phase senses this
refraction index gradient, which acts as a thermal
prism and causes the de¯ection of the laser beam
The heat-sensor based on laser de¯ection is sche-
matically depicted in Fig. 1. The beam from a 1.4 mW
(
Fig. 1). Depending on the intensity of de¯ection,
the laser beam is more or less blocked by the knife-
edge. The laser light intensity is monitored by a
photodiode (Fig. 1). Therefore, this laser based detec-
tion method differs fundamentally from the ones
present in commercially available calorimeters, which
are usually based in thermistors' responses.
The amplitude of the de¯ection signal is calibrated
with a chemical reaction whose enthalpy value is
Fig. 1. Scheme of the laser de¯ection heat-sensor, where S is He±
Ne laser, L1 is a verging lens (f0 ˆ 11 cm), C is a 1  1 quartz
cuvette, A is a knife-edge, L2 is verging lens (f0 ˆ 5 cm), D is a
photodiode. The lower panel represents an ampli®cation of the
cuvette where the aqueous phase is black, the organic phase is clear
and the laser beam is passing in the organic phase below the
0
known (neutralization of NaOH by HCl, DH ˆ
�
56:5 kJ/mol) [15], i.e. a calibration curve with Volts
per Joule (V/J) units is obtained. This calibration
curve, which is obtained just before or after the actual
experiment is performed, to keep the same laser and
4
aqueous/CCl interface.

T. Schneider et al. / Thermochimica Acta 362 (2000) 179±184
181
detector alignments, is used to the enthalpy determi-
nation of other chemical reactions as expressed by
Eq. (1).
duction between the two phases. For slower chemical
�
1
reactions (rate in the order of min ) kinetic informa-
tion can also be obtained [10±14]. Different ways of
injecting the sample at different velocities, at different
positions of the sample surface and by using different
sample volumes (keeping number of mols constant)
were tested and the results turn out the same within the
experimental error, suggesting that mixing did not
affect substantially the de¯ection signal.
DI
Slo  N
DH ˆ
(1)
where DH is the enthalpy change calculated for the
reaction, DI the maximum de¯ection signal, Slo the
calibration curve slope obtained with the acid/base
neutralization, and N the number of mols of the
limiting reactant.
By increasing the number of mols of HCl and
consequently the amount of heat released there is a
proportional increase in the DI (initial voltage minus
voltage at 35 s) (Fig. 2). This result is in agreement
with the acid/base titration performed in a similar
apparatus by Wu and coworkers [10,11]. The heat
detection limit calculated as three times the blank
standard deviation divided by the calibration curve
slope was ꢁ60 mJ.
3
. Results and discussion
In Fig. 2, typical laser de¯ection signals measured
after the addition of X mmol of HCl to a NaOH
solution, are shown. These signals are composed of
the following features: at ꢁ25 s after the addition of
HCl the voltage measured in the photodiode starts to
decrease, indicating that a higher percentage of the
beam is being blocked and, therefore, that the beam is
being de¯ected. The signal change reaches its max-
imum value at around ꢁ35 s and then starts to return to
the original value leveling off at ꢁ150 s.
In order to test whether this instrument can measure
the enthalpy change related with other chemical pro-
cesses, the complexation reactions between MgCl2
and CdCl with EDTA were investigated. It is shown
2
in Fig. 3 that the de¯ection signals measured after the
addition of 15 mmols of MgCl and of 15 mmol of
2
CdCl in a 1 M EDTA solution. It is clearly seen that a
2
For fast reactions like the ones investigated in this
work, the heat liberated from the reaction works as an
instantaneous heat source and the temporal signal
behavior can be explained based only on heat con-
positive and a negative DI were measured for the
EDTA/Mg and EDTA/Cd complexation, respectively.
This can be understood considering that the enthalpy
change related with these processes have opposite
Fig. 2. De¯ection signal obtained after the reaction of 3, 6, 9 and 12 mmol of HCl with a 1 M NaOH aqueous solution.

1
82
T. Schneider et al. / Thermochimica Acta 362 (2000) 179±184
2 2
Fig. 3. De¯ection signals obtained after the reaction of 15 mmol of CdCl (~) and MgCl (&) with a 1 M EDTA solution. The ®gure inserts
show the laser de¯ection signals of HCl/NaOH neutralization reactions (A) used to construct the calibration curve (B).
signs, i.e. � 38 and 23 kJ/mol for EDTA/Cd and
EDTA/Mg, respectively [16]. Therefore, in the case
of the EDTA/Cd reaction a negative refractive index
gradient is created, de¯ecting the laser beam to the
same direction as observed for the HCl/NaOH neu-
tralization reaction. It is important to emphasize that
NaOH reaction series (Fig. 3 insets Ð slope ˆ
0:775 V/J) and the heat changes were calculated
(Q). The reaction enthalpies were obtained by dividing
the Q values by the number of mols of CdCl and
2
MgCl added to the EDTA solution (Eq. (1)).
2
It is shown in Table 1 that the reaction enthalpy
values calculated agree, within the experimental error,
with the literature values of these reaction enthalpies,
obtained by Jordan and Aleman using titrimetric
enthalpy determinations [15]. The chelation of EDTA
with magnesium has also been studied by other
authors and con¯icting results were found. Charles
�
4
� 1
the dn/dT value of CCl is � 6:1 Â 10
K , conse-
4
quently with the increase in the temperature there is a
decrease in the refractive index [7]. In the case of the
EDTA/Mg reaction, the aqueous phase removes heat
from its boundaries creating a negative heat and a
positive refractive index gradients in the organic
phase, which de¯ect the laser beam to an opposite
direction.
0
reported an endothermic DH of ꢁ3 kJ/mol [16],
while Carini and Martel obtained an exothermic
0
In order to calculate the enthalpy changes related
with these reactions, the measured DI values were
compared with the calibration curve slope of an HCl/
DH of approx. � 12 kJ/mol [17]. Our result is in
0
quantitative agreement with the DH reported by
Jordan and Alleman [15].

T. Schneider et al. / Thermochimica Acta 362 (2000) 179±184
183
Table 1
DH for the CdCl
2
and MgCl
2
complexation with EDTA
Q (kJ)
a
DI (V)
DH (experimental) (kJ/mol)
DH (literature) (kJ/mol)
CdCl
MgCl
2
� 0.41
� 0.53
� 35 Æ 5
23 Æ 3
� 38
2
0.26
0.34
23
a
Values were obtained from [15].
The reproducibility of thermometric titrations is
characteristically good, usually around 1% or better.
Linde et al. reported reproducibility lower than 1% for
several kinds of reactions including, neutralization
titrations, complex formation and precipitation [18].
Similar results were obtained by Jordan and Alleman
the heat wave using an interferometer in the laser
beam pathlength.
4. Conclusion
[
15]. High sensitivity calorimeters in development
It has been shown that the laser de¯ection sensor is
capable of measuring enthalpy changes related with
endothermic and exothermic processes. Depending
upon the sign of the enthalpy change of the chemical
reaction, refractive index gradients of opposite signs
will be generated in the adjacent medium, de¯ecting
the laser beam to opposite directions. This instrument
can be calibrated to obtain the value of the enthalpy
change. The HLOD and the ®nal uncertainty in DH
(ꢁ60 mJ and 15%, respectively) are still too high
compared with the ones obtained using detection
schemes available in nowadays microcalorimeters.
It should be mentioned, however, that the absolute
value of these parameters are dependent on cuvette
design and size and other instrumental variables that
can be improved. These modi®cations would allow us
to adapt this detector in the sample compartment of a
differential scanning microcalorimeter to test its per-
formance in temperature scanning thermograms.
stage, that use smaller sample volumes tend to have
reproducibility worse than the ones cited above
(
around 5%) [19]. In our case, the ®nal uncertainty
obtained in the DH calculation was around 15%
Table 1). The high ®nal uncertainty values observed
(
in the DH calculation are the result of several errors
along the procedure, including sample volume deli-
very in the microliter range and problems with the
interface of the two liquids (meniscus position is a
factor that contribute for the scattering of values
around the average). We are con®dent that these
dif®culties can be worked out by designing a better
sample compartment and sample delivery device.
The high HLOD values obtained (60 mJ) can be
explained by the fact that in our sample compartment
the heat is diffused to all spatial directions and only a
minor fraction is felt by the probe laser and also by the
fact that there is a large magnitude difference between
the size of the reaction compartment (0:2 cm  1 cm
of the sample cell and 0.2 cm of the liquid phase
height) and the diameter of the probe laser (ꢁ50 mm).
In this context, micromechanical technology can
improve our instrument as has been shown to other
newly devised integrated circuit calorimeters [19]. In
fact, a micrometer sample compartment is being built
for us by the Engineering School of our University. In
this sample compartment, the size differences between
the laser beam and the sample compartment is mini-
mized and the heat generated in the aqueous phase will
be driven to the laser beam compartment by the
differences in thermal conductivities of the materials
used. We have also been working to improve the
sensitivity of this detection scheme by measuring
Acknowledgements
We thank the ®nancial support of the Brazilian
funding agencies FAPESP and CNPq. T.S. is an under-
graduate fellow from the Physics Institute of USP with
fellowships from the PIBIC program (99-07/00 CNPq)
and from FAPESP (08/00-12/00). We also would like
to express our gratitude to M. Gugliotti for helpful
discussions.
References
[
1] L. Dejean, B. Beauvoit, B. Guerin, M. Rigoulet, Biochim.
Biophys. Acta 1457 (2000) 45.

1
84
T. Schneider et al. / Thermochimica Acta 362 (2000) 179±184
[
[
2] S.A. Potekhin, O.I. Loseva, E.I. Tiktopulo, A.P. Dobritsa, [10] X.-Z. Wu, H. Shindoh, M. Yamada, T. Hobo, Anal. Chem. 65
Biochemistry 38 (1999) 4121.
3] M.G. Fonseca, C. Airoldi, J. Chem. Soc., Dalton Trans. 21
(1993) 834.
[11] X.-Z. Wu, T. Hobo, Anal. Chim. Acta 316 (1995) 111.
[12] X.-Z. Wu, H. Shindoh, T. Hobo, Anal. Chim. Acta 299 (1995)
333.
(
1999) 3687.
[
[
4] P.J. Reucroft, D. Rivin, Thermochim. Acta 328 (1999) 19.
5] R. Denoyel, M.T.J. Keene, P.L. Llewellyn, J. Rouquerol, J.
Therm. Anal. 56 (1999) 261.
[13] X.-Z. Wu, T. Morikawa, K. Uchiyama, T. Hobo, J. Phys.
Chem. 101 (1997) 1520.
[
6] H.P. Lang, M.K. Baller, R. Berger, C. Gerber, J.K.
Gimzewski, F.M. Battiston, P. Fornaro, J.P. Ramseyer, E.
Meyer, H.J. Guntherodt, Anal. Chim. Acta 393 (1999) 59.
7] S.E. Bialkowski, Photothermal Spectroscopy Methods for
Chemical Analysis, Wiley, New York, 1996.
[14] T. Schneider, M. Gugliotti, M.J. Politi, M.S. Baptista, Anal.
Lett. 33 (2000) 297.
[15] J. Jordan, T.G. Alleman, Anal. Chem. 29 (1957) 9.
[16] R.G. Charles, J. Am. Chem. Soc. 76 (1954) 5854.
[17] F.F. Carini, A.E. Martell, J. Am. Chem. Soc. 76 (1954) 2153.
[18] H.W. Linde, L.B. Rogers, D.N. Hume, Anal. Chem. 25 (1953)
404.
[
[
[
8] J. R Barnes, R.J. Stephenson, M.E. Welland, C. Gerber, J.K.
Ginzewski, Nature 372 (1994) 79.
9] M. Tokeshi, M. Uchida, K. Uchiyama, T. Sawada, T.
Kitamori, J. Luminesc. 83 (1999) 261.
[19] D. Capary, M. Schr oÈ pfer, J. Lerchner, G. Wolf, Thermochim.
Acta 337 (1999) 19.