 |
Description  |
|
|
FIELD OF THE INVENTION
This invention relates to improved evanescent wave sensors for use in
spectrophotometric assays of analytes in fluids, and more particularly to
such sensors having a shell with inner and outer wall surfaces which
propagate radiation between them by total internal reflection.
CROSS-REFERENCES
The following applications, filed concurrently herewith are incorporated
herein by reference:
Slovacek et al, U.S. Ser. No. 07/711,783 entitled "Multiple Surface
Evanescent Wave Sensor System"; and
Slovacek et al, U.S. Ser. No. 07/712,304 entitled "Multiple Output
Referencing System for an Evanescent Wave Sensor".
TECHNICAL DISCLOSURE
There are a number of optical devices which propagate radiation by total
internal reflection to generate an evanescent wave at the interface of the
device and a medium having a lower index of refraction. See Harrick, N.J.,
Internal Reflection Spectroscopy, Harrick Scientific Corp., Ossining, N.Y.
(Third Printing 1987). The evanescent wave is an electromagnetic waveform
which typically extends less than a wavelength into the surrounding
medium. However, this penetration is sufficient to permit substantial
optical interaction between the evanescent wave component and target
substance(s) or analytes in the medium. Analyte, as used herein, shall be
understood to include any of a variety of chemical and biochemical
substances. The analyte sources may include physiological, scientific and
industrial (toxic and nontoxic) test media; where the presence, absence or
quantity of the analyte in the test medium is sought to be determined; and
where, for example, analysis of a physiological analyte is relevant to
diagnosis and/or treatment of disease.
One use of optical devices is in the area of fluorescent immunoassays.
Presently, such applications include optical waveguides, for example, in
the form of fiber optic rods typically coated with either an antibody or
an antigen which binds the corresponding antigen or antibody, respectively
from a medium or test sample. This coating is applied typically prior to
the performance of an immunoassay measurement. In a "sandwich"
immunoassay, an antibody is bound to the surface of the fiber optic rod to
form a reactant coating, and the device is immersed in a test sample
comprising the antigen to be analyzed, if present, and a second antibody,
previously labeled by a fluorescent tag. Alternatively, in a "competitive"
assay, the fluorescent labeled antigen is first mixed with the antigen in
the sample, and the mixture is brought into contact with the rod and the
first, bound antibody coating. In either technique, the fluorescently
labeled antibody attached to the antigen to form a tagged complex bound to
the fiber optic rod by the first antibody or the fluorescently labeled
antigen itself combines with the first antibody coated on the fiber optic
rod. Light is introduced into the fiber optic rod at a predetermined
wavelength band and is propagated along the rod by total internal
reflection. The reflection is, of course, not completely total since the
fluorescent tag absorbs a small amount of radiation. Alternatively, the
attached fluorescent tag, referred to as a fluorophore, is excited by the
evanescent wave electromagnetic fields at a first wavelength and
fluoresces at a second, longer wavelength. Fluorescence from the excited
fluorophore passes into the optical waveguide via a tunneling effect and
the portion of the fluorescent radiation which occurs at an angle greater
than the critical angle is propagated through the optical waveguide to
emerge from an output end.
The use of fiber optic rods in fluorescent immunoassays provides several
advantages over the use of glass or plastic microtiter plates into which
fluid suspected of containing an analyte is placed. Rather than serving as
an optical waveguide, the plate simply contains the fluid and radiation is
passed directly through the fluid and the plate to excite a tagged
complex. The plate requires separate washing steps to remove unattached
analyte and tagged binder, because unbound fluorophores still present in
the medium would otherwise be detected. By comparison, the use of the
fiber optic rod eliminates the additional washing steps and is therefore
faster and easier.
However, fiber optic rods have a number of disadvantages including
difficulty in handling the rod during manufacture and its mounting within
a test apparatus. Fiber optic rods must be carefully handled because even
a tiny nick or scratch in the surface defeats total internal reflection at
that region. In defective or damaged rods, radiation escapes from the rod
rather than propagating down the entire length of the rod.
Manufacturing fiber optic rods requires a number of processing steps.
Typically, fiber optic rods of silica are drawn from a furnace and then
cut. These are bundled and potted in wax to enable handling without
scratching the surface of the rods. The rods are then recut to length,
polished, rewaxed and cleaned. Finally, the rods are coated with a
hydrophobic silane to assist attachment of a reactant coating, and the rod
is then cured. An antibody or other reactant is then coupled or attached
to the rod to form the reactant coating, typically by dipping the rod in
an antibody solution.
Another disadvantage of the fiber optic rods is that light must be
introduced at an angle into the rod to achieve total internal reflection.
The rod must therefore be carefully aligned with an optical system to
achieve total internal reflection, resulting in waveguiding of the light.
The reflection is, of course, not completely total since the fluorophore
absorbs a small amount of radiation. Further, the evanescent wave is
generated only at the points where the radiation reflects from the surface
of the rod. A 1 mm diameter rod having a length of 50 mm typically
achieves only about twenty-two reflections along its length for light near
the critical angle, and even fewer reflections for light propagated at
lower angles.
Several improved waveguides are described in U.S. Pat. No. 4,880,752; which
is incorporated herein by reference. In one construction, the waveguide
has an elongated, rod-shaped core having an opening within the core
material. A reactant coating is disposed about the opening within the
core. The elongated core has the disadvantages mentioned above for solid
fiber optic rods, namely difficulty in handling during manufacture,
coating and mounting.
SUMMARY OF THE INVENTION
The preferred evanescent wave sensor of the present invention has a shell
with a radiation port at one end and a base at the other end. A wall
extends between the radiation port and the base to define inner and outer
wall surfaces. Radiation introduced through the radiation port bounces
(reflects) between the walls as it is internally reflected by those
surfaces, when the radiation is introduced at an angle greater than that
of the critical angle (relative to a reference line normal to the surface)
and a surrounding medium has a lower index of refraction than that of the
material forming the shell. The shell has a large surface area for its
length, in comparison to fiber optic rods, and therefore is compact.
In one embodiment, the inner wall surface converges to substantially a
point centered beneath the radiation port. This arrangement presents an
inclinded surface to virtually all incoming radiation signals to assist
propagation by total internal reflection. The radiation port, which
receives incident radiation and passes emission radiation, may be planar,
convex, or concave. The shell may have a pyramid shape, such as to form a
cone. The interior of the shell is hollow, and the base defines an opening
into the interior.
A reactant coating may be disposed on one of the inner and outer wall
surfaces when the sensor is for use in an assay of an analyte in a medium,
and the reactant coating includes a binder of the analyte such as an
immobilized antibody, an immobilized antigen, an enzyme, or other binding
molecule. The shell material is transmissive to radiation which can excite
fluorescence of a fluorescent tag and is transmissive to fluorescent
radiation from the fluorescent tag when fluorescence is monitored.
Alternatively, another parameter such as absorption may be monitored.
Preferably, the sensor further includes a handle attached to the base, such
as a tab, an annular flange, or other projection which extends away from
the base in a direction transverse to that of the inner and outer wall
surfaces. The handle enables manipulation of the shell during manufacture,
coating with the reagent, and installation in an apparatus for analyzing a
medium.
It is also preferable for the shell material to be suitable for manufacture
by injection molding. Thus, the shell can be made precisely and cost
effectively. Acceptable materials include plastics including
polymethylmethacrylate, polystyrene, polycarbonate, or even a moldable
glass.
This invention also features apparatus for analyzing a test medium
including a shell such as described above and receptacle means for
defining a chamber about a portion of the shell to contact the test medium
to the shell for analysis. Preferably, the receptacle defines an access
port which enables radiation signals to enter into and emerge from the
radiation port of the shell. The apparatus may further include means for
guiding radiation through the access port from a radiation source, and
means for guiding fluorescent radiation from the shell to means for
detecting fluorescent radiation. The receptacle may surround the shell and
the access port communicates with the chamber. The access port preferably
is defined by an inclined surface in the receptacle which converges toward
a narrow region disposed below the radiation port of the shell. The
inclined surface establishes a well which minimizes the effect of
overfilling the chamber. The apparatus may further include means for
introducing the test medium to be analyzed into the chamber, and means for
establishing a fluid-tight seal between the base of the shell and the
receptacle.
OBJECTS OF THE INVENTION
It is among the objects of the invention to provide an improved sensor,
apparatus, and method for analyzing test media and, in particular, for
conducting fluorescent immunoassays.
Another object of the invention is to provide a sensor having a large
surface area for its length.
Yet another object of the invention is to provide a sensor which generates
a large number of internal reflections per unit length.
A still further object of the invention is to provide a sensor which is
easily alignable with incoming radiation signals so that the radiation
enters the sensor at close to the critical angle of the sensor.
A further object of the invention is to provide a sensor which is easier to
manipulate and less costly to manufacture.
Yet another object of the invention is to provide such a sensor which may
be injection molded.
DESCRIPTION OF THE DRAWINGS
The foregoing and other objects and advantages of the invention will be
appreciated more fully from the following further description thereof with
reference to the accompanying drawings wherein:
FIG. 1 is a schematic cross-sectional view of a sensor shell according to
the invention;
FIG. 2A is a schematic side view of a portion of the sensor of FIG. 1
inserted within a test medium to be analyzed;
FIG. 2B is an enlarged schematic view of total internal reflection of
radiation having an angle greater than that of the critical angle,
relative to a reference line normal to the surface;
FIG. 2C is a schematic representation of the refraction of radiation
entering the sensor;
FIG. 3 is a schematic representation of a sensing apparatus according to
the invention;
FIG. 4 is a more detailed illustration of the optical components for one
embodiment of the apparatus;
FIG. 5 is a chart of counts (in hundreds) per second (CPS) over time for
binding of different concentrations of labelled digoxin using the
apparatus of FIG. 4;
FIGS. 6A and 6B are schematic representations of convex and concave
surfaces, respectively, of the radiation port of alternative embodiments
of the sensor; and
FIGS. 7A and 7B are schematic partial cross-sectional views of sensor
shells having converging and diverging wall surfaces, respectively.
DESCRIPTION OF THE ILLUSTRATIVE EMBODIMENT
A sensor according to this invention and apparatus and method for using the
sensor can be accomplished by a shell having a radiation port at a first
end and a base at a second end. The base has a dimension greater than that
of the radiation port, and inner and outer wall surfaces of the shell
extend between the radiation port and the base. Preferably, the wall
surfaces are substantially parallel to each other in the direction
extending between the radiation port and the base to reflect light between
those surfaces by total internal reflection when the radiation signal(s)
is/are introduced at an angle greater than that of the critical angle and
a surrounding medium has a lower index of refraction than that of the
material comprising the shell. The base has a dimension, such as a
diameter, greater than that of the radiation port so that the walls of the
shell diverge as they approach the base.
In a preferred construction the sensor shell is frustoconical in shape as
shown for sensor 10, FIG. 1, which illustrates a cross-sectional view. The
sensor 10 includes a shell or cone 11, having a planar radiation port 12
and base 14. A wall 16 extends between the radiation port 12 and the base
14 and defines outer wall surface 18 and inner wall surface 20. The
interior of the shell 11 is hollow and the base 14 defines a circular
opening 22 which communicates with the interior. The inner wall 20
converges to a point 24 beneath the radiation port 12 so that virtually
all radiation signals entering through radiation port 12 impinges at a
desired angle on inner wall surface 20.
In this preferred construction, the sensor 10 further includes handle 26
which is an annular flange connected to the base 14. In other
constructions, a tab or other projection may serve as a handle. The handle
26 is a non-active surface, and therefore can be grasped without damaging
an optically active surface. In contrast, the radiation port 12 and wall
surfaces 18 and 20 are optically active surfaces which are optically
polished. Both incoming excitation radiation signals and outgoing
fluorescent radiation signals (or other monitored radiation) reflect off
the surfaces 18 and 20 without encountering the handle 26. Radiation which
exits through the base 14 or the handle 26 will not affect the assay.
Although it is preferred that the handle is integral, it need not be
integral but can be a separate member attached to the sensor.
Alternatively, the lower portion of the shell 11 can be designated as a
handle and not used as an optically active area. One technique for holding
a sensor without interfering with the optically active area is disclosed
in U.S. Pat. No. 4,671,938; which is incorporated herein by reference.
The dimensions for the sensor 10 in the preferred embodiment are as
follows. The radiation port 10 has an outer diameter of 0.92 mm, and the
base 14 has an outer diameter of 10.26 mm. The handle 26 has an outer
diameter of 16 mm which provides approximately 5 mm of graspable surface
along all sides of the shell 11. The sensor 10 is approximately 11.5 mm in
length, including a thickness of 0.5 mm for the handle 26. The wall 16
also has a uniform wall thickness of 0.5 mm to provide a large number of
internal reflections along its length. The material is
polymethylmethacrylate (PMMA) having an index of refraction of 1.4917 at a
wavelength of 589.3 nm.
The sensor 10, in comparison to fiber optic rods used in the art, has a
large surface area for its length. For the above-described sensor having a
length 11.0 mm exclusive of the thickness of the handle 26, the outer wall
surface 18 has an area of approximately 170 mm.sup.2. This area is
slightly greater than the 157 mm.sup.2 of a 50 mm fiber optic rod having a
diameter of 1 mm as is presently used in fluorescent immunoassays.
There are several factors to be considered in selecting material for the
sensor 10. It is most preferred that the material be injection moldable so
that the sensor may by rapidly and inexpensively formed, and easily mass
produced. Further, optically polishing the mold will establish optically
polished surfaces which do not require further polishing following
manufacture. Another factor is that the material must have an index of
refraction greater than that of the intended medium to be analyzed, as
described below. Additionally, it is desirable for the material to be
optically pure and provide low attenuation of the radiation of interest.
Silica glass is suitable for ultraviolet or visible radiation, plastics
such as polymethyl methacrylate (PMMA), polystyrene, and polycarbonate are
suitable for visible radiation, and fluoride glass or chalcogenide are
suitable for near infrared radiation. Other organic polymeric materials
such as silicones, acrylates, fluoroacrylates, and the like can also be
used as the sensor material. It is also desirable for the material to be
nonfluorescent to the radiation of interest for assays involving
fluorescence.
Additionally, it is desirable for the material to have suitable surface
properties or characteristics for binding of a reactant coating to it, or
to be amendable to modification to assist bonding. PMMA is preferred not
only for its optical purity and its injection molding characteristics, but
also because it is hydrophobic which enables antibodies and proteins to be
attached to the surface simply by bringing them in contact with the PMMA.
For glass, it is desirable to add a silane coating to provide either a
hydrophobic surface or one amendable to covalent coupling chemistries.
The shell 10 has an angle a as shown in FIG. 2A. Angle a represents the
inclination of the axis of propagation 32 relative to the cone axis 30
which passes through the center of the radiation port 12 and the base 14.
The axis of propagation 32 passes through the center of wall 16, halfway
between outer wall surface 18 and inner wall surfaces 18, 20 and the cone
axis 30 is also angle a.
Light ray 34 is shown propagating through the wall 16 after passing through
the port 12, and bouncing between inner wall surface 20 and outer wall
surface 18 due to total internal reflection. Total internal reflection
occurs when the angle of the ray 34 is greater than the critical angle,
which in turn depends upon the index of refraction n.sub.1 of the wall 16
relative to the index of refraction n.sub.2 of a first medium, typically
air, through which radiation passes to enter and exit the radiation port
12, and the index of refraction n.sub.3 of a sample medium, typically a
fluid L, i.e. a liquid, which surrounds a lower portion of the sensor 10.
In the construction shown in FIG. 2A, air surrounds the remainder of the
shell 11 including radiation port 12 and all of the inner wall surface 20.
The relative indices of refraction, the calculation of the critical angle,
and the desired angle of the radiation entering and propagating through
the sensor are described in more detail below.
In this construction, the fluid L is contacted to a portion of the outer
wall 18, which is coated with a reactant coating 36 which may be an
immobilized antibody, an antigen, a receptor, a nucleic acid, an enzyme,
or other binding substance as is known in the art. Coating as used herein
shall be understood to include specific and nonspecific reactions
including noncovalent binding and covalent binding. It is desirable for
the reactant coating to bind an analyte in the sample medium or test
sample. To prepare a sensor formed of PMMA for use in a sandwich
immunoassay, a first antibody is attached to the sensor by dip coating
after cleaning the outer wall surface 18 by sonicating the sensor for
several seconds while it is immersed in a Freon TF bath. A typical
antibody has a height of approximately 100 angstroms (A) and binds and
antigen having a typical thickness of 100-200 A in the case of a large
molecular weight antigen. A second antibody having an attached fluorophore
label is then contacted against the antigen to form a tagged complex
having a fluorophore spaced approximately 300-400 A from the surface of
the outer wall 18. When a light ray 34 bounces against the surface of the
wall 18, as shown for point 38, an evanescent wave excites the fluorophore
which induces emission at a longer wavelength. The fluorescent emission is
indicated by rays 40, shown in phantom. The portion of the rays 40 which
are internally reflected are propagated back through the radiation port 12
in the described embodiment, and detected as described below. This portion
depends strongly on the distance of the fluorophore from the PMMA/sample
medium interface, and decreases rapidly with increasing distance as
expected for a tunneling-like effect.
The parameters of a particular cone construction according to the invention
are as follows. The critical angle c relative to reference line RL of FIG.
2B, is calculated according to the formula:
c=sin.sup.-1 (n.sub.2 /n.sub.1)
where n.sub.1 is the index of refraction of the shell and n.sub.2 is the
index of refraction of the first medium (air) contacting the interior of
the shell. When the index of refraction n.sub.3 of the sample medium (the
test sample to be analyzed) is greater than n.sub.2, then n.sub.3 is used
as described below.
Radiation having an angle greater than that of angle c, such as angle b of
ray 42, will be totally internally reflected as shown by ray 42a. A
plastic material such as PMMA has an index of refraction n.sub.D of
approximately 1.49 and fused silica has index of refraction of
approximately 1.46. For the media surrounding the sensor, air has an index
of refraction of approximately 1.00, whereas many biological liquids have
an index of refraction of approximately 1.33. For analysis of such
liquids, it is therefore desirable for the sensor to have an index of
refraction of greater than 1.33. Likewise for the analysis of solid
coatings, the sensor material index (n.sub.1) must be greater than index
n.sub.3 or that associated with a solid coating of interest. By way of
example, a polystyrene sensor having n.sub.D =1.59 may be utilized in
evanescent wave interrogation of methyl cellulose (n.sub.d =1.49) or
natural rubber (n.sub.D =1.52) polymeric coatings.
Alternatively, the test sample to be analyzed may be contained in a gas or
a liquid phase which is exposed to a solid composite construction. For
example, a solid silicone containing a fluorescent material such as a
ruthenium-based dye is coated onto the outer surface of a PMMA sensor as a
layer having a thickness of 1-10 microns. The sensor is then exposed to
another test medium such as blood or gas, and the oxygen contained therein
diffusing into the silicone layer quenches (reduces) the fluorescence of
the dye. The amount of quenching affects the level of detected
fluorescence. Silicone has an index of refraction of approximately 1.43,
and therefore the sensor substrate in this application requires an index
of refraction greater than 1.43 for waveguiding and evanescent pumping of
the silicone layer. In this case the evanescent wave does not propagate
into the gas or blood sample, rather the interaction is confined to a
fraction of a wavelength depth into the silicone coating layer. In other
words, the reactant coating has a low refractive index and a sufficient
thickness so as to preclude direct interaction of light between the sensor
and the medium to be analyzed. The critical angle c would thus be
calculated to be sin.sup.-1 (1.43/1.49)=73.7.degree. for dye-doped
silicone and PMMA. It is to be noted that a reactant coating typically is
sufficiently thin so that it does not noticeably refract radiation passing
through it, especially when the reactant coating is in a fully hydrated
state. Otherwise, the effect of the reactant coating must be accounted
for.
Because the critical angle is greater at the interface of the sensor and
the fluid L than that of the sensor-air interface, in the case of a liquid
based sensor, the critical angle defined by the sensor substrate and the
measured fluid L is used to establish the minimum acceptable angle of a
ray 42. For example, where the fluid to be analyzed is a liquid having an
index of refraction n.sub.D of 1.33 and the sensor is formed of PMMA
having index of refraction n.sub.D of 1.49 at a wavelength of 589.3 nm,
the critical angle is 63.2.degree.. To allow a deviation in angle of up to
3.8.degree., angle b is selected to be 67.degree.. Subtracting this angle
from 90.degree. establishes angle a as 23.degree..
Next, an acceptable launch angle e is calculated as shown in FIG. 2C. A
light ray 44 has an angle of refraction d according to Snell's Law:
n.sub.2 sin e=n.sub.1 sin d
If angle d is allowed to be as large as 3.8.degree., angle e is
5.6.degree..
In summary, the cone is constructed by selecting the index of refraction of
the cone material and the test medium to be analyzed, and determining the
critical angle at the interface of the cone and the test medium for
waveguiding of radiation signals at the desired wavelength. The cone angle
is calculated with respect to the axis of propagation by subtracting the
critical angle from 90.degree.. For a collimated radiation source, this
angle may, in principle, be used. However, in practice, the cone apex
angle is made somewhat less to account for misalignments and mechanical
tolerances. If there is an angle of incidence at the radiation port 12 of
greater than 0.degree., the cone apex angle is reduced to make the walls
steeper. The length of the cone is selected to provide the required
surface area. It is desirable to form the walls of the shell as thin as
possible to increase the number of reflections between the outer and inner
wall surfaces, and it is desirable for the radiation signals introduced
into the cone to be as close to the critical angle as possible, but still
within the waveguiding angle(s), to maximize the evanescent wave component
delivered to the interface of the shell and the medium.
A receptacle 50 for defining a chamber about a sensor 10 and an optical
system 51 for delivering radiation signals to and collecting radiation
signals from the sensor are shown in FIG. 3. The receptacle 50 defines a
chamber 52 into which fluid to be analyzed is delivered. A fluid-tight
seal is established between the base 14 and the receptacle 50 by a weld 54
which is formed by sonically welding the handle 26 to the receptacle 50.
The receptacle 50 and the sensor 10 in the above construction form a
disposable cartridge suitable for one time use. When the sensor is
intended for reuse, a removable seal may be established using an o-ring
and a clamp, for example. In either case, the attachment 54 is made to a
non-active surface which does not affect sensor performance.
Fluid to be analyzed is introduced into the chamber 52 through an
introduction port 56 and a passage 58. The test sample may be injected
through introduction port 56 using a syringe, or a well 60 may be provided
having a wall 62, shown in phantom. In the later construction, liquid to
be analyzed is poured into the well 60 and is drawn by capillary action
through the passage 58 and upwardly through the chamber 52. The walls of
the chamber match the shape of the sensor 10 to minimize the volume of the
chamber. In one construction, the gap between the outer wall surface 18
and the receptacle is 0.25 mm with a fluid level height of 10 mm, and the
chamber has a working volume of approximately 0.05 ml.
The receptacle 50 further defines an access port 64 which enable the
radiation port 12 to communicate with radiation source 66 and detector 68.
In this construction, the access port 64 also serves another function. The
inclined surface 67 diverges away from narrowed region 70 which serves to
limit capillary action beyond the narrowed region 70. Further, the widened
area accommodates excess liquid which may be introduced into the chamber
52. Providing an overflow area avoids contaminating the radiation port 12
and minimizes the surface area of the core which is exposed to the excess
liquid. Even if overfilling raises the level of the liquid by a
millimeter, the surface area of the additionally exposed region of the
sensor 10 is far less than the surface area that would be exposed without
the widened region of the access port 64, and is also far less than that
which would be exposed if a corresponding length of the sensor toward the
base of the cone were exposed. These are advantages which result from
introducing liquid near the base of the cone and providi | | |
