Note: Descriptions are shown in the official language in which they were submitted.
r
- 1 20~6630
53594/009.528
"Sensor"
This invention concerns a novel sensor for
determination of absorption of electromagnetic radiation
by an analyte and a method of determination using such a
sensor.
Absorption of electromagnetic radiation, typically
visible light, is commonly used for the detection and/or
quantification of chemical substances
or acquisition of information concerning such
substances. In general, photometric methods used
previously have relied on measuring the transmission or
the incident radiation and relating this to a standard
transmission value. However, such methods are sensitive
to radiation scattering and are often unsuitable for
analysis of particulate samples. Recently, however,
methods have been proposed which measure absorption
directly by determining the temperature increase in the
analyte caused by absorption of the incident radiation,
and thus seek to avoid problems caused by scattering.
Tanaka et al (J. App. Phys. 63(6) p.1815, Ig88)
have described a system of photothermal spec~roscopy fo~
thin solid films wherein the thin sample is mounted on a
transparent temperature sensor and irradiated with
pulsed light to measure increases in sample temperature
caused by absorption. However, due to the incomplete
transparency of the thermal sensor, this itself becomes
heated. This aberration is worsened where the sample
scatters the incident light and thus both light
scattering and light absorption will increase the siqnal
thereby giving an anomalous result. Further, the
transparent sensor comprises a sandwich of a
thermosensitive material between electrode films. The
latter have to be extremely thin to reduce absorption
and are conse~uently very susceptible to both mechanical
... ' ' ':
,:
, . :, ' :
.
.
. :- , ., ' : .
'
2 0 ~
2 --
and chemical degradation. Additionally, the sample is
in contact with one of the electrodes whereas it is
important to isolate the sample from the electronic
circuitry. In some cases, the sample can act as an
antenna and pick up disturbances.
USP 3948345 describes a photoacoustic method of
spectroscopy wherein a gas contained in a resonant
container and surrounding the analyte to be investigated
is irradiated with pulsed light. The absorption of
this light by the analyte and the resulting increase in
temperature creates a pulsed elastic expansion, i.e.
elastic waves, in the gas which can be detected by a
conventional acoustic detector such as microphone. USP
4303343 using the same principle optimises the
relationship between the pulse-frequency, the wavelength
of the incident light and other parameters.
European Patent 49918 typifies a development of the
technique in which absorption of pulsed light by the
analyte sample produces pulsed expansion and contraction
of a solid element which is transformed into an
electrical signal by means of a piezoelectric transducer
attached to the solid element. However, such
photoacoustic methods are extremely sensitive to local
vibrations and have proved difficult to use in some
cases.
Wo 86/05275 discloses a method for the measurement
of sedimentation of a sample of particles in a rlui~
(e~g. erythrocytes in blood) whereby an intensity
modulated light beam is directed onto the exposed
surface of the sample (i.e. the air/sample interface) to
produce a thermal response which may be detected by
means such as a photoacoustic cell, infra-red detector
or piezo-electric crystal
We have now found that by irradiating the sample
through a solid element which is transparent to the
incident radiation, and thus not heated by it, and yet
is highly conductive to heat, and by providing a thermal
: ' .
.
: . - .
' : . ' ' ' '
~i 2~g6~30
detector on the solid element close to the sample, it is
possible to detect directly increases in temperature of
the sample caused by irraaiation.
According to the present invention therefore we
provide a sensor for the detection or quantification of
absorption of electromagnetic radiation by a sample
wherein the increase of temperature induced in the
sample by the radiation produces a signal proportional
to said temperature increase, characterised in that the
sensor comprises a heat conducting solid element which
is transparent to said electromagnetic radiation and
which has a first surface for contacting with said
sample, a radiation input surface and a radiation path
between said surfaces, thermooptic or thermoelectric
thermal detector means being provided in thermal contact
with said solid element close to said first surface to
receive conducted heat therefrom without obstructing
said radiation path.
It will be appreciated that by placing the thermal
detector substantially outside the path of the incident
radiation, the effect of radiation scattering by the
sample may be minimised since, as described in more
detail hereinafter, it is possible substantially
completely to shield the detector from such scattered
light, thereby enhancing the sensitivity of the sensor.
In general, the sensor will be provided with means
for irradiating the sample through the solid element.
It is particularly advantageous to irradiate with
radiation which is modulated with respect to amplitude
and/or wavelength since this enables background errors
such as overall temperature variations largely to be
eliminated. The radiation may be ultraviolet, visible
or infrared light.
Amplitude modulation or pulsing of the incident
radiation can conveniently be achieved by a conventional
mechanical light chopper placed in the collimated light
path. Variation of the wavelength of the incident
.
.. . . . : ' ' .
,
.
2~66~
- 3a -
light, e.g. between an absorption maximum and a minimum,
may, for example, be effected by a laser diode. In
general, the modulation frequency should be low e.g.
below 50 Hz.
The frequency of signal amplification or other
periodic means of electronic sampling can be
synchronised or locked onto the modulation frequency of
the incident radiation so that extraneous temperatur~
: .
.; ~
., . . ~
W~90/08952 PCT/EPgO/00211
- 4 -
variations occurring between the pulses are not
amplified. Apparatus for effecting such modulation
and sampling are described in USP 3948345.
Furthermore, the pulse frequency can be related
to the rate of conduction of heat from the sample
to the sensor. Thus, the amplitude of the signals
produced by the temperature fluctuations depends
in part on the transfer of heat from irradiated
sections of the sample at a given distance from
the surface of the transparent solid element.
Heat generated at points deeper into the sample
is not transferred to the sensor in the time between
incidence of the radiation and sampling of the
signal from the thermal detector. The maximum
depth within the sample from which heat contributes
to the signal is termed the 'thermal diffusion
length' and defines the volume of the sample which
is analysed. This definition of the volume makes
quantification of an absorbing substance possible.
Incident light is conveniently led to the
sensor by means of an optical fibre system. The
light source may be a laser or a strong lamp. In
general, it should be possible to produce incident
radiation in the wavelength range 250 nm to 2500
nm.
The thermal detector may, for example, be
a thermoelectric d~vice such as thermistor or thermo-
couple or a thermooptical device such as a temperature
responsive laser.
The solid heat conductive element may conven-
iently be made of diamond, which has a heat conductance
six times that of copper, or sapphire or quartz,
all of which are substantially completely transparent
to ultraviolet, visible and infrared light. The
solid element is conveniently in the form of a
block with two opposed ends and at least one side
onto which a thermal detector can be mounted.
~ W090/08952 PCT/EPgO/00211
~ _ 5 _ 2~6~
The sample can then be mounted on or thermally
contacted with one of the ends (the "sampling end")
while the incident radiation enters the block through
the opposite end, the path between the radiation
source and the sample thus being unobstructed.
It may be advantageous for the sampling end
of the block to be rounded to some extent, since
this will increase the surface area of the block
which will contact a given volume of sample material.
The sampling end of the block may if desired
be given a thin protective coating, e.g. of a plastics
material such as an epoxy resin. The thickness
of the coating should be such that there is no
undue reduction in thermal contact between the
sample and the block (the use of a rounded sampling
end as described above may assist in offsetting
any such reduction). The use of protective films
of plastics materials such as polycarbonates, polyacrylates,
polyamides, polyesters, polyalkylenes and polyhaloalkylenes,
especially if extended to protect the thermal detector,
may be particularly advantageous where hazadous
(e.g. infectious or toxic) or chemically highly
reactive samples are to be investigated. The films
may advantageously be designed to be disposable,
especially where infectious or toxic samples are
to be encountered.
The solid heat conductive element may if
desired comprise more than one component. Thus,
for example, a block may have a thin disc of similar
material transparently adhered to one face so that
one side of the disc ~orms the sampling end of
the element. The thermal detector in such arrangements
may be attached to the block or the underside of
the disc as appropriate, and will be particularly
well protected against contamination by sample
material.
WO90/08952 PCT/EP90/00~
- 6 - 2~
In most applications, however, the thermal
detector is advantageously mounted on a surface
of the heat conducting solid element which extends
parallel to the radiation path. Substantially
total internal reflection of the incident radiation
at the said parallel surface should then prevent
the radiation from reaching the detector. Such
internal reflection may be enhanced by attaching
the thermal detector to the solid element using
an adhesive having a smaller index of refraction
than the material of the solid element. Since
materials such as sapphire and diamond have a high
index of refraction, a wide range of adhesives
may be used, including epoxy adhesives, cyanoacrylate
adhesives and polyester adhesives. The adhesive
may additionally be used to coat the remaining
sides of the solid element to minimise egress of
light therefrom. Particularly suitable adhesives
include electrically conductive glues such as metal
epoxy glues, for example a silver epoxy such as
~po-tek H 20 E (manufactured by Epoxy Technology
Inc., Mass., U.S.A), since these ensure maximum
light retention while also having good thermal
and electrical conductivity. Alternatively, the
surface of the transparent solid element may be
coated with a reflective layer, e.g. a thin layer
of aluminium or silver, before attachment of the
thermal detector, such treatment being particularly
~- suited to measurements in the ultraviolet and infrared
regions.
~ here a thermistor is employed as the thermal
detector this may, when the scale of the apparatus
permits, be formed by thick film technology, i.e.
by_printing a paste of thermistor material onto
the solid element after any necessary pretreatment
to ensure maximum internal reflection and then
sintering the paste at high temperature.
,
.
- .
.
~V90/08~52 PCT~EP9~
The distance between the sample and the thermal
detector is preferably as small as possible, in
order to minimise the time for conduction of heat
from the sample to the detector and thereby achieve
S maximum sensitivity. In general, the specific
conductivity of the solid element will be many
times that of the sample- Typically, the dista~ce
of the thermal detector from the sampling end of
the element. will be of a similar order of magnitude
to the dimensions of the sampling end. Thus, for
example, the thermal detector might be mounted
about 1 mm from a sampling end which itself is
about lmm across. Alternatively the surface of
the sampling end may extend further along the axis
passing through the detector to provide a larger,
essentially oblong area in contact with the sample.
I~here the sample absorbs the incident radiation
strongly, the latter will readily be absorbed within
the thermal diffusion length and produce a strong
signal. Where absorption is low, only a part of
the incident light may be absorbed within the thermal
diffusion length. It will be appreciated that
in general, the thickness of the sample should
exceed the thermal diffusion length and is preferably
at least twice that length.
Sensors according to the invention may, if
desired, be very small. The thermal detector can
readily be made of the same size or smaller than
the heat conducting solid element. It is particularly
convenient to mount the solid element on the end
of an optical fibre; the signal from the thermal
detector can be conducted by electrical wires or
an optical fibre, conveniently mounted parallel
to _the optical fibre for the incident radiation.
Sensors so arranged can readily be used to
detect or quantify samples in a wide range of situations,
for example not only in in vitro experiments but
-~WO90/08952 PCT/EP90/00211
- 8 -
also in vivo. Thus, for example, such a sensor
may be inserted into a blood vessel for continuous
measurement of haemoglobin content. It is particularly
useful to be able to immerse such a sensor in a
liquid sample to e~amine the analyte at different
depths, in particular at points distant from the
surface; for example erythrocytes will absorb oxygen
from the atmosphere when near to the surface of
a liquid sample containing them and thus may alter
their absorption spectrum.
In certain applications it may be necessary
to shield the sensor from thermal influences or
chemical corrosion. One or more protective layers,
e.g. of any appropriate polymer material, may, for
example, be applied over the whole sensor, excluding
the surface in contact with the sample, in order
to achieve this end.
Shielding against electrical influences or
disturbances may also be desirable in particular
applications and may, for example, be effected
by surrounding the sensor (again excluding the
surface in contact with the sample) with a metal
shield. Thus, for example, the sensor may be situated
in an appropriately earthed metal container, e.g.
a tube of a material such as acid-resistant steel,
and/or may be coated with an electrically conductive
glue such as a metal epoxy glue.
According to a further feature of the invention
we provide a method for detection or quantification
of absorption of electromagnetic radiation by a
sample wherein a sensor according to the invention
is irradiated to cause modulated radiation to pass
along said radiation path to the said first surface
and tbence to said sample, heat produced by absorption
of radiation by the sample being conducted to the
thermal detector of said sensor to produce signals
the amplitude of which is indicative of the heat
produced by said absorption.
.
, ,' . , ~ :
W090~08952 PCT/EP90~0021l
_ g _ '
The method of the invention is particularl~ 3
useful in detecting or quantifying suspensions
of particles, e.g. cells or aggregates, which
are difficult to assay using older methods due
5 to the problem of light scattering.
The sensor and method described herein may
also be utilized in the measurement of colour intensity
of samples immobilised on solid supports; the signal
is not subject to the disturbance by the mechanical
10 contact between the sensor and the support which
one may experience when using photoacoustic methods.
The principle is similar to that employed in solution.
The light is chosen at a wavelength suitable for
absorption by the material in question. The increase
15 in temperature is proportional to the colour strength
and may be measured as described above. Since
the technique is based on absorption rather than
reflection, it is more sensitive than reflectometric
methods. Furthermore, a quite small area of colour
20 is sufficient to obtain a good signal. Coloured
areas less than 1 mm2 are normally sufficient.
Some analytical techniques are based on formation
of colours on a surface, either by chemical reactions
leading to formation of insoluble or immobilized
coloured material, or filtration of coloured agglutinates
formed by coupling of receptor-ligand pairs, or
by selective filtration with one member of a receptor-
ligand pair immobilized in a porous material.
The sensors of the invention are of particular
-use in all these methods.
The sensor and method according to the invention
may in particular be used to detect or quantify
analytes in a test sample based on alterations
in_the rates of sedimentation of particles due
- 35 to chemical or physical interactions, as measured
by, for example, determining the radiation absorption
of the sample at time intervals. Further applications
WOsO/08952 PCT/EP90/00211
- 10 ~
include analysis of blood by determination of haemoglobin
in haemocytes.
In view of the small dimensions to which
sensors may be made, measurements may readily be
made in flow systems, e.g. where the sample contact
surface of the sensor is situated within the interior
of a flow chamber~ Surprisingly the optometric
signal is substantially unaffected by the flow
of a ~lowing liquid sample.
The invention is now more particularly described
with reference to the accompanying drawings in
which:
Fig. 1 shows a thermal sensor according to
the invention.
Fig. 2 shows an arrangement for using the
thermal sensor according to Fig. 1.
Fig. 3 shows a complete optical sensor system
wherein a sensor according to Fig. 1 is positioned
in one end of an optical fibre.
Fig. 4 shows a plot of the signal from a
device according to Fig. 3 for various concentrations
of a coloured substance dissolved in water.
Fig. 5 shows a sensor consisting of a series
of heat conducting elements according to Fig. 1
positioned close to each other, but not in thermal
contact.
Fig. 6 shows a fIow chamber incorporating
a thermal sensor according to the invention.
Fig. 7 shows an alternative embodiment of
a thermal sensor positioned on an optical fibre
and a method of assembling the same.
Fig. 8 shows a thermal sensor according to
the invention protected by a plastic material.
Fig. 9 shows a thermal sensor according to
the invention which has a rounded sampling end
and is protected by a plastic material.
.
.
' , . ' ,, ' ' . " ' , . '. '
: .- : .
- ~O90/0~952 PCT/EP90/~211
11- 2~
Fig. lO shows a thermal sensor according
to the invention in which thé heat conductive element
comprises a thin disc adhered to a block.
Fig. ll shows a plot of the results obtained
by optothermal spectrometry and re1ectometry in
the measurement of colloidal gold immobilised on
a porous membrane.
Fig. 12 shows a plot of the results obtained
from measurement of haemoglobin ~n blood ~galnst
those from a standard method (Coulter S-880).
In the sensor shown in Fig. l the heat-conducting
element l is a cube of transparent material with
a high ability to conduct heat. Light-pulses 4
are sent through the heat-conducting element l
and into the sample 3 mounted thereon. A proportion
of the heat which is generated in the sample is
conducted to the interface between the sample 3
and the heat-conducting element l. The increase
in temperature at this interface de~ends on the
light absorbant properties of the sample. Because
of the high heat-conductivity of the element l,
generated heat is conducted from the surface of
the sample 3 to the thermoelectrical detector 2.
The heat-conducting element l is of a size which
allows the sample and the thermoelectrical detector
to be located at a distance apart from each other
which is less than or equal to the thermal diffusion
length of the actual material of the heat-conducting
element l. Since the thermal diffusion length
is dependent on the pulse frequency of the incoming
light, the size of the heat-conducting element
must be chosen with respect to the highest theoretically
used frequency. As the frequency increases, the
distance between the sample and the thermoelectrical
- 35 detector should be decreased. In the case shown
in Fig. l, the thermoelectrical detector is a thermistor.
A constant voltage is applied to the thermistor
20~3~
WO90/~8952 PCT/EP9~/00211
- 12 -
through cable leads 5. When the temperature varies,
the current through thermistor, conducted via cable
leads 5, will vary due to altered resistance.
Using a suitable electronic arrangement the variations
!; in current may be amplified and recorded.
In the arrangement shown in Fig. 2, light
from a lamp 6 is focussed through lenses 7 and
7A. Light pulses are created using a chopper 8
(a rotating disc), and the light passes through
a filter 9 in order to select a required wavelength
before passing to the sample 3 through a transparent
heat conducting element l carrying a thermistor
2 connected via cable leads 5. The wavelength
and pulse frequency of the light are chosen with
respect to the sample to be analysed. The electronics
are locked on to the frequency of the modulated
light source and the signals are then amplified.
This reduces the noise and ensures that the sensor
will not register variations of the temperature
of the surroundings.
In the arrangement shown in Fig. 3, a heat-
conducting element l is positioned on the end of
an optical fibre ll. The light source is a laser
diode 10 with constant intensity and variable wavelength.
The light is led from the laser diode l0 to tbe
sample 3 through the optical fibre ll. The recorded
variations in temperature depend on the variation
in absorbed light at different wavelengths. One
may for example change the wavelength from an absorbance
~0 maximum to a minimum. A laser diode 12 is used
as thermooptical detector, its output and frequency
varying with temperature. The radiation from this
laser is lead through another optical fibre 13
tD an optoelectrical transformer 14 where the optical
signal is transformed into an electrical signal
which may then be recorded. The entire sensor,
except for the part which should be in contact
.
.
WO90/08952 PCT/EP90/00211
) - 13 - '~ 3
with the sample, is covered by a protective material.
The sensor is characterized by being substantially
insensitive to electrical disturbances since it
produces an optical output signal from the laser
S diode 12.
The plot shown in Fig. 4, obtained using
appar,atus as described in connection with Fig.
3, shows a substantially linear correlation between
optothermal signal and the concentration of various
samples of black ink in water.
The arrangement shown in Fig. 5 illustrates
the possibility of combining several sensors together.
The heat conducting elements l carry thermoelectrical
detectors 2 connected to amplifiers (not shown)
by cable connectors 5. They are thermally isolated
from each other. Lisht at different wavelengths
is applied via optical fibres ll to the elements l
which may then measure the absorbance at various -
wavelengths in a sample, thus providing knowledge
about the absorbant properties of the various components
in a sample. The concentration of each component
may thereby be calculated based on the measured
signal at each of the wavelengths.
Another possibility is to use different modulated
~e.g. pulsed) frequencies for the various sensors.
Using low frequencies one may analyse a rather
thick layer of a sample compared to the thin layer
analysed at high frequencies. Using a proper mathematical
treatment of the measured signal one may be able
to analyse the concenteation profile of substances
situated some distance into a sample.
A further possibility is to analyse a sample
- which shows variations from point to point. In
this case the same frequency and wavelength are
- 35 used in all of the sensors. The measured signals
may be utilized for evaluation of variations from
point to point, or they may provide a mean value
for a larger surface.
WO90/08952 PCTtEP9~/00211
- 14 -
In the flow chamber shown in Fig. 6, solid
structure 15 is formed with a flow chamber 16 having
an inlet 17 and an outlet 18. A recess 19 in the
structure 15 is adapted to receive a thermal sensor
20 which rests on an O-ring 21 abutting against
a flange 22. The sensor 20 is pressed into contact
with the O-ring 21 by springs 23 held in position
by a cap 24. The sensor 20 comprises a body of
cruciform vertical cross-section provided with
a central, vertical, cylindrical hole into which
is set a light path 25 leading to a sapphire window
26. A thermistor 27 is provided laterally to the
sapphire window 26 and is connected by electrical
leads 28 to the signal sensing device (not shown).
In the embodiment shown in Fig. 7, the heat
conducting element 1 may for example be a sapphire
rod polished to good optical quality on all surfaces.
The thermal detector 2 is a thermistor preferably
coated on its larger lateral faces with thin films
of silver or gold to ensure good electrical connection.
One such lateral face of the thermistor 2 is affixed
to a vertical face of element 1 by means of silver
epoxy glue. The remainder of this vertical face
of element 1 and the other larger lateral face
of thermistor 2 are covered with silver epoxy glue
29 whereby electrical cable connectors 5 may be
attached, one to element 1 and one to thermistor
2. The remaining three vertical faces of element
1 are preferably also covered with silver epoxy
glue. Element 1 may be affixed to an optical fibre
11 using a drop 30 of W -curable glue and subjecting
the resulting assembly to W irradiation 31.
Typical dimensions for such a sensor include
eLement 1 ~lxlx6 mm) and thermistor 2 (0.5x0.5x0.35mm).
Applying a constant voltage to thermistor 2 via
leads 5, resistance changes of the order of 4%
per C may, for example, be observable.
:
--W090/03952 - 15 - 2 0 ~ ~ 6 3 ~
In the arrangement shown in Fig. 8 the element
1, thermistor 2 and electrical connections 5 are
protected by enclosure in a tube of epoxy resin
33, leaving only the sampling end 32 of element
1 exposed. This minimises interference and consequent
noise which may otherwise occur if a sample comes
into electrical contact with the thermistor 2.
An alternative method of protecting the sensor
is shown in Fig. 9, where element 1 has a rounded
sampling end 32 which, together with thermistor
2 and electrical connections 5 is protected by
thin flexible plastic film 34 which in use is in
thermal contact with sampling end 32 and on which
sample 3 is placed.
In the embodiment shown in Fig. 10, the heat
conducting element is a two component system comprising
a rod 35 and disc 36. These may conveniently be
made of sapphire, representative dimensions including,
for example, lxlx6 mm for rod 35 and diameter 3-
5 mm and thickness 0.1-0.3 mm for disc 36. Rod
35 and disc 36 are glued together using a transparent
glue and thermistors 2 are glued to the latter
using silver epoxy glue. The undersides of disc
36 and thermistors 2 and the sides of rod 35 are
coated with a layer of silver epoxy glue 37, a
small ring at the edge of disc 36 being left uncoated.
~ Electrical connections 5 are attached in the usual
- manner. Disc 36 is adhered by glue 37 to a metal,
e.g. acid-resistant stainless steel, tube 38, which
electrically screens or shields the sensor, and
a protective coating 39 is applied.
Sample 3 is irriadated by light pulses 4
passing through rod 35. By virtue of the nature
of the construction of the sensor contact between
sample 3 and thermistors 2 is minimal, especially
when highly impermeable materials such as sapphire
are used for disc 36.
.,
WO9OJ08952 PCT~EP90/00211
, . . .
- 16 -
The following Examples illustrate the method
according to the invention:
Example 1
5An optothermic spectrophotometer system as
shown in Fig. 2 had a transparent, conductin~
eleme~t 1 comprising a sapphire having a surface
of 1 x lmm . The sapphire was connected to thermal
sensors, and light pulses tfrequency 2 Hz) were
led to the sapphire thro~gh an optical fibre.
The light source was a halogen lamp, and the light
was filtered to give a wavelength of 540 ~ 40 nm.
1 ~9 of anti-C-reactive protein monoclonal
antibody formed by murine hybridoma cells was added
to an activated porous membrane to immobilize the
antibodies (Hybond N nylon membrane, Amersham,
UK).
The surface area of the membrane was 10 mm2
in each of the measurements peformed. Solutions
of C-reactive proteins varying from 0.5 to 15 ~g/ml
were added and sucked through the membrane by a
negative pressure. Thereafter, a solution containing
about 1 ~g of another anti-C-reactive protein antibody
coupled to colloidal gold with an average diameter
of 4.5 nm was added and sucked through the membrane.
An increasing amount of colloidal gold was arrested
in the membrane as the amount of C-reactive protein
was increased.
The intensity of the coloured surface was
measured both by reflectometry (Macbeth 1500 Plus,
Reflectometer) and optothermic spectroscopy as
described above. Each optothermic measurement
was performed for ten seconds. The results obtained
us ng the two methods are shown in Fig. 11 and
can be seen to correlate well.
.
, ~ :
' ~
.
W090~08952 PCT/EP9~ ~2~ 3
- 17 -
Example 2
This example demonstrates how an optothermal
sensor may be used for the measurement of haemoglobin
in blood.
100 ~1 blood was added to a conically shaped test-tube
containing 10 ~1 20% Sterox SE. The blood was
haemolyzed immediately by the detergent.
10 The haemolyzed blood samples were measured using
an instrument as described in Example 1, but now
using a frequency of 16 Hz.
The results from 75 blood samples were compared
15 to a standard method (Coulter S-880) resulting
in a correlation coefficient of 0.99 (Fig. 12).
Repeated analyses of the same sample showed a coefficient
of variation of 0.5-1.7%
20 Ex~
The instrument o~ Example 2 was used with
a frequency of 16 Hz. The sensor was equipped
with a plastic cup which enabled blood to stay
in contact with a horizontally positioned sensor.
25 When blood was measured directly without haemolysis,
the results correlated with the standard method
about as well as described in Example 2. Thus,
the sensor also makes direct measurements of haemoglobin
in blood samples possible.
Example 4
This example illustrates how the senæor
may_be used for the measurement of haemoglobin
- 35 in a flow system.
: ;
W090/08952 PCT~E~/00211 -)
-- 18 - 2~6~
The instrument illustrated in Fig. 6 was used.
The optothermal sensor 20 used had a sensitive
area of 1 mm and an outer diameter of 3 mm. The
flow rate of blood through the chamber 16 where
the sensor was positioned was 2 ml per minute.
Between each sample, the chamber 16 was rinsed
with a hypochlorite solution. The instrument
was connected via the light path 25 to a 20 W halogen
lamp and was operated at a frequency of 16.7 Hz.
Each sample was measured 2-4 times over a period
of 20 seconds.
Twenty-six blood samples were tested using the
described method, and the results were compared
lS to a standard method for the measurement of haemoglobin
using a Coulter instrument. The correlation coefficient
obtained was 0.990 and the linear regression line
was y = 1.04x -4.4 where y is the optothermal value,
and x is the value from the standard method.
2~
The signal at the output of the amplifier attached
to the sensor was also observed with an oscilloscope.
No disturbances could be detected due to the blood
flow.
;
~' ,
.
,
, ,
'~
'
