 |
Description  |
|
|
A phase shifting optical polarimeter is composed of commonly available,
inexpensive, compact parts and is completely solid state. The disclosed
polarimeter uses a technique which is capable of accurately resolving very
small, biologically significant, polarization vector rotations in the
presence of large noise, such as the concentration of certain sugars in
human blood.
BACKGROUND OF THE INVENTION
Glucose is an optically active substance. Its concentration in a given
solution can be quantitatively measured using polarimetric techniques. The
rotation of plane polarized light by an optically active substance is well
known. The concentration of an optically active substance is related to
the magnitude of the rotation observed, .theta., the wavelength of the
light, .lambda., the material's thickness, D, specific rotation,
[.alpha.], and its temperature, t.
##EQU1##
If plane polarized light of incident intensity, I.sub.1, is passed through
a polarizing material (analyzer) having its transmission axis at some
angle, .alpha., to the plane of the light's polarization, the transmitted
intensity, I.sub.2, is related to the square of the angle's cosine by
Malus' law.
I.sub.2 =I.sub.1 Cos.sup.2 (.alpha.) (2)
Where .alpha. is the observed angle of rotation.
Human blood normally has a glucose concentration of between 80 and 120
mg/dl. The amount of polarization vector rotation imparted to plane
polarized light with a wavelength of 670 nm by 100 mg of glucose dissolved
in 1 dl of solution having a thickness of 1 cm and a temperature of
98.6.degree. F. is on the order of 0.004.degree.. Resolving rotation
angles this small has become commonplace. However, in the presence of
large noise, like that encountered in live human tissue, resolving angles
of this magnitude becomes a formidable problem.
With an analyzer's transmission axis placed such that its parallel to the
polarization axis of a beam of plane polarized monochromatic light which
has a transparent optically active solution in its path, the intensity of
light through the analyzer can be used to determine solution
concentration, C. Combining equations 1 and 2:
##EQU2##
Equation (3) is completely amplitude dependant and assumes that there is no
absorption, scattering or other amplitude related noise. The device
described herein employs the use of a phase shifting technique which is
amplitude independent, eliminating the largest source of noise.
Additionally, the unique optics arrangement facilitates a technique for
linearly magnifying this phase shift, providing improved instrument
sensitivity. I also employ the use of sophisticated electronics capable of
recovering a very small signal from large noise.
SUMMARY OF THE INVENTION
Two linearly polarized beams of monochromatic laser light are amplitude
modulated with low frequency sinusoids (for example less than 100 Hz). The
first laser is modulated with a sinusoid 90.degree. out of phase from the
second laser. These two beams are then collimated and combined into one
composite beam using simple optics. This composite beam is then allowed to
pass through an optically active substance of a calibrated path length.
The laser radiation exiting the sample is passed through a optical notch
pass filter, reducing any signal that may be present from any polarized
ambient light by passing only light of the laser's frequency. The
composite beam then passes into a beam splitting polarizing cube. This
cube has the function of separating the beam into two orthogonal
polarization components. The two beams of light exiting the cube are
focused onto silicon photoelectric detectors. The cube is oriented such
that one of the cube's transmission axis is aligned to coincide with the
polarization axis of one of the lasers. The current from the detectors is
converted into a voltage. The resulting detector voltages are then AC
coupled to remove any DC component of the voltage, leaving only signal.
These voltages are then subtracted using an instrumentation grade
operational amplifier. This subtraction removes any component of signal
resulting from common mode electrical noise, randomly polarized light, or
scattering in the sample. The subtracted signal is phase shifted in direct
proportion to solution concentration. The phase shift is measured and the
solution concentration calculated. Setting the ratio of laser signal
amplitudes greater than unity can further amplify this phase shift to
produce high instrument sensitivity.
BRIEF DESCRIPTION OF THE DRAWINGS
Other objects, features and advantages of this invention will be more
apparent after referring to the following specification and attached
drawings in which:
FIGS. 1 is a schematic of the instrument of this invention;
FIG. 2A and 2B are phase diagrams of the modulated laser output;
FIGS. 3A, 3B, 3C and 3D are respective plots of a composite of the detected
waveform showing noise, the modulated signal with the noise added, the
demodulated wave form with symmetric noise ready to be filtered, and
finally the filtered output signal;
FIG. 4 is a plot of a sample having three degrees of rotation (3.degree.)
illustrating the increased phase shift possible by the amplitude ratio of
the laser signals being greater than or equal to unity; and,
FIG. 5 is a plot of degrees of polarization rotation versus observed
instrument waveform phase shifting for the differentially amplified
detector signals resulting in increased instrument sensitivity.
DESCRIPTION OF THE PREFERRED EMBODIMENT
Referring to FIG. 1, two lasers 14, 16 driven by conventional modulating
electronics 18 emit linearly polarized beams of monochromatic laser light.
These lasers 14, 16 are amplitude modulated with low frequency sinusoids.
(See FIGS. 2A and 2B). The first laser 14 is modulated with a sinusoid
90.degree. out of phase from second laser 16.
The signal on laser 14 will be denominated as a sine wave; the signal on
laser 16 will be denominated a cosine wave. Laser 16 has its polarization
vector set such that it is rotated 45.degree. from the polarization vector
of laser 14. The polarization vectors of the two lasers are thus fixed and
do not change. The only time variant quality of the light exiting the
lasers are their amplitude modulation.
Signal impressed upon the carrier wave modulating laser 14 can be
mathematically set forth:
A.sub.1 Sin(.omega.t), Plane polarized at angle .theta.
Signal impressed upon the carrier wave modulating laser 16 can be
expressed:
A.sub.2 Cos(.omega.t), Plane polarized at angle .theta.+45.degree.
Where A.sub.1 and A.sub.2 are signal amplitudes and .theta. is an arbitrary
reference to describe the laser's plane of polarization.
These two beams are then collimated and combined into one composite beam
using simple optics including plano-convex lens 19 and plano-concave lens
20. This composite beam is then allowed to pass through an optically
active sample of a calibrated path length measured at distance device 30.
This has the function of rotating the polarization vectors of the
composite beam in direct proportion to the concentration of optically
active constituent present.
The laser radiation exiting the sample is passed through a optical notch
pass filter 35, reducing any signal that may be present from any polarized
ambient light by passing only light of the laser's frequency. The
composite beam then passes into a beam splitting polarizing cube 40. This
cube has the function of separating the beam into two orthogonal
polarization components.
The two beams of light exiting the cube 40 are focused by respective lenses
50, 52 onto silicon photoelectric detectors 60, 62. The cube 40 is
oriented such that one of the cube's transmission axis is aligned to
coincide with the polarization axis of laser 14.
Presuming the absence of sample S, all of the light from laser 14 will pass
through the cube 40 to impinge upon detector 60. It follows that the light
from laser 16, with its polarization vector set 45.degree. from the first,
will be split in half and impinge equally upon both detectors. The current
produced by the incident light on the detectors is directly related to the
light's intensity. The current from the detectors is converted into a
voltage V.sub.1, V.sub.2. The resulting detector voltages are then AC at
capacitors C.sub.1, C.sub.2 coupled to remove any DC component of the
voltage, leaving only signal.
Presuming that a sample S is present causing rotation: From equation (2):
Detector 62 demodulated and filtered signal:
Cos.sup.2 (.alpha.)[A.sub.1 Sin(.omega.t)]+Cos.sup.2
(45.degree.+.alpha.)[A.sub.2 Cos(.omega.t)] (4)
Detector 60 demodulated and filtered signal:
Sin.sup.2 (-.alpha.) [A.sub.1 Sin(.omega.t)]+Cos.sup.2 (45.degree.-.alpha.)
[A.sub.2 Cos(.omega.t)) (5)
Where .alpha. is the rotation of polarization vectors.
These voltages are then subtracted using an instrumentation grade
operational amplifier 70. This subtraction removes any component of signal
resulting from common mode electrical noise, randomly polarized light, or
scattering in the sample S. Detector 62--Detector 60 demodulated and
filtered signals:
A.sub.1 Sin(.omega.t)[Cos.sup.2 (.alpha.)-Sin.sup.2 (-.alpha.)]-A.sub.2
Cos(.omega.t)[Cos.sup.2 (45.degree.+.alpha.) -Cos.sup.2
(45.degree.-.alpha.)]
Scattering in the sample would produce a component of signal which is
randomly polarized and thus equally distributed to the detectors.
Therefore, when the signals from the detectors are subtracted, only the
signal due to the fraction of light retaining its original polarization,
though perhaps rotated, remains.
With no sample S in the path of the composite beam, the resultant
subtracted signal is in phase with the signal on laser #1 (Sin(.omega.t)).
However, placing an optically active substance in the path of the
composite beam imparts a rotation of polarization vectors. Now, with this
polarization rotation, subtracting the detector voltages yields a
resultant sinusoid wave form that is phase shifted from the signal on
laser 14.
Referring to FIGS. 3A-3D, the signal produced by this invention can be
viewed at various stages of the signal processing. Noise is schematically
set forth in the form of sinusoids at 80 in FIG. 3A. Signal plus noise is
shown at 85 in FIG. 3B. A demodulated wave form with symmetric noise,
ready to be filtered is shown at 90 in FIG. 3C. Finally, the recovered
demodulated and filtered wave form is shown at 100 in FIG. 3D. The reader
will understand that wave form phase analysis can occur through a phase
detector 72 and computer 74.
Referring to FIG. 4 with a laser signal amplitude ratio (A.sub.2 /A.sub.1)
of unity, the shift in phase will be equal to twice the angular rotation
of the polarization vectors. This phase shift magnification is a
consequence of this polarimeter's design and aids in its sensitivity.
Setting the ratio of laser signal amplitudes greater than unity, typically
A.sub.2 /A.sub.1 =5, can further magnify this phase shift. In the example
here, what would be a phase shift on the order of 3.degree. observed in a
conventional polarimeter, can be a phase shift in the order of
27.72.degree. for A.sub.2 /A.sub.1 =5. Thus, for low concentrations of
optically active material in the path of the polarimeter, greater
sensitivity is achieved.
For small polarization rotations, the total phase shift amplification
produced by these techniques can be approximated to be linear and equal to
2 (A.sub.2 /A.sub.1). (See FIG. 5) For example, using a conventional phase
shifting polarimeter, a polarization rotation of 4.5 milli-degrees would
result in a signal phase shift of 4.5 milli-degrees, an expected rotation
for the glucose concentration present in human blood using a path length
of 1 cm. The polarimeter described in this paper, with its phase
amplification techniques, will exhibit a signal phase shift of about 45
milli-degrees (e.g. 10 times 4.5 milli-degrees).
The function yielding the magnitude of signal phase shift for a given
positive rotation:
##EQU3##
Where .phi.=observed signal phase shift and .alpha. is the polarization
rotation of the composite beam.
It will be understood that the present invention has been illustrated with
respect to the measurement of sugar, preferably in human blood. The reader
will understand that this disclosure is applicable to polarimeter utilized
in other fields.
I have used the term "linearly" polarized. This will be understood to
include so-called "elliptical polarization" in which linear polarization
along one axis significantly exceeds polarization along the remaining
axis.
* * * * *
|
|
 |
|
 |
Description |
|
