Infrared sensor and method of calibrating the same

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates generally to semiconductor processing, and more particularly to a system to sense infrared radiation from a semiconductor chip and to methods of calibrating the same.

2. Description of the Related Art

Infrared thermal imaging is a common analysis technique used on semiconductor devices for failure analysis and design. In the past, typical thermal imaging of a functional device was done in an open air setup, that is, without any structures in the optical path of the detector. In such designs, air is used to cool the device undergoing testing. An open air setup is acceptable for parts that operate below certain power densities.

Some more recent designs of semiconductor devices exhibit much higher power densities. In some cases, more exotic cooling is required to keep the semiconductor device from failing due to thermal run away. Standard copper heat sinks used to cool the semiconductor devices in testing environments do not allow for optical access to the device itself. Yet optical access is required for thermal imaging.

One solution found in the industry for cooling a device with optical access is known as a diamond heat spreader. Since diamond is mostly transparent to the infrared spectrum, it is a good window material for thermal imaging. At the same time, the diamond can physically contact a device under test to spread and remove the heat during thermal imaging. In another conventional variant, a sealed fluid chamber is positioned on top of a semiconductor device. The fluid is infrared transparent and facilitates heat removal. The top of the chamber has a window made from an IR transparent material.

A difficulty with the conventional diamond spreader is the propensity for Newton's rings to degrade the infrared image of the semiconductor device. The Newton's rings appear due to inherent non-planarities in the upper surface of the semiconductor device and the lower surface of the diamond window. A difficulty with the conventional liquid setup is that the liquid and the upper window mask the actual count of photons emitted by the semiconductor chip. The liquid and the upper window both absorb and reflect percentages of any incident radiation, whether from the semiconductor chip, or in the case of the upper window, from both the semiconductor chip and the liquid. Without an accurate actual photon count from the semiconductor chip, a correct emissivity for the chip remains elusive.

The present invention is directed to overcoming or reducing the effects of one or more of the foregoing disadvantages.

SUMMARY OF THE INVENTION

In accordance with one aspect of the present invention, a method is provided that includes determining a transmission of a transmissive window and a transmission of a transmissive fluid. In addition, an infrared emission of the transmissive window is determined along with an infrared emission of the transmissive fluid for at least one temperature. In a system that has an infrared sensor and an optical pathway to the infrared sensor, the transmissive window and the transmissive fluid are placed in the optical pathway. A semiconductor chip is placed in the optical pathway proximate the transmissive fluid. Radiation from the optical pathway is measured with the infrared sensor. An emissivity of the semiconductor chip is determined using the measured radiation and the determined transmissions and emissions of the transmissive window and the transmissive fluid.

In accordance with another aspect of the present invention, a method is provided that includes determining a transmission t_w of a transmissive window and a transmission t_f of a transmissive fluid. In addition, an infrared emission b_w(T) of the transmissive window is determined along with an infrared emission b_f(T) of the transmissive fluid for at least one temperature. In a system that has an infrared sensor and an optical pathway to the infrared sensor, the transmissive window and the transmissive fluid are placed in the optical pathway. A semiconductor chip is placed in the optical pathway proximate the transmissive fluid. A photon count MPC from the optical pathway is measured with the infrared sensor. An actual photon count APC from the semiconductor chip is determined according to:

MPC=t_wt_fAPC+b_w(T)+b_f(T).

In accordance with another aspect of the present invention, an apparatus is provided that includes an infrared sensor that has an optical pathway, a first member for holding a semiconductor chip in the optical pathway, and a second member for holding an infrared transmissive window in the optical pathway between the infrared sensor and the semiconductor chip. The transmissive window has a known transmission and a known emission at least one temperature. Either the first or the second member is operable to separate the transmissive window from the semiconductor by a preselected gap. A film of infrared transmissive fluid is in the gap for establishing fluid communication with the semiconductor chip and the transmissive window. The infrared transmissive fluid has a known transmission and a known emission at at least one temperature. A count of photons measured by the infrared sensor may be converted to a count of photons emitted by the semiconductor chip using the known transmissions and emissions of the transmissive window and the transmissive fluid.

BRIEF DESCRIPTION OF THE DRAWINGS

The foregoing and other advantages of the invention will become apparent upon reading the following detailed description and upon reference to the drawings in which:

FIG. 1 is a pictorial view of an exemplary embodiment of a device under test diagnostic system;

FIG. 2 is a sectional view of FIG. 1 taken at section 2-2;

FIG. 3 is a portion of FIG. 2 shown at greater magnification;

FIG. 4 is another portion of FIG. 2 shown at greater magnification;

FIG. 5 is a sectional view of an exemplary embodiment of an emissivity target calibration setup;

FIG. 6 is an overhead view of an exemplary emissivity target;

FIG. 7 is a sectional view of an exemplary embodiment of a setup for calibrating the transmission of a transmissive window;

FIG. 8 is a sectional view of an exemplary embodiment of a setup for calibrating the transmission of dual transmissive windows; and

FIG. 9 is a sectional view of an exemplary embodiment of a setup for calibrating the transmission of dual transmissive windows and a transmissive fluid.

DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

In the drawings described below, reference numerals are generally repeated where identical elements appear in more than one figure. Turning now to the drawings, and in particular to FIG. 1, therein is shown a pictorial view of an exemplary embodiment of a semiconductor chip diagnostic system 10 that includes an infrared sensor 15 that is operable to sense infrared radiation projecting upwardly from a device under test (DUT) that is not visible in FIG. 1 but will be shown in subsequent figures. The infrared sensor 15 may be an infrared microscope or other type of infrared sensor. The system 10 includes a platform 20 that is suitable to have seated thereon a member or test circuit board 25 that may be connected to a computing device 30 that is operable to both cause the device under test (not shown) to implement certain electronic functions and to take readings therefrom and also possibly to control the operation of the microscope 15 as desired. The computing device 30 may be a general purpose computer, a dedicated computer, or other type of computing device. A data link 35 is used to connect the computing device 30 to the test board 25. The data link 35 may be a hard wired or wireless connection as desired. A temperature controlled member 40 may be seated on the diagnostic board and provided with a coolant supply and return lines 45 and 50 respectively. The temperature controlled member 40 may be a thermal plate that is provided with a window 55 through which infrared radiation may transmit up through an objective lens 60 of the microscope 15. The microscope 15 contains one or more radiation sensors (not visible). In an exemplary embodiment, the microscope sensor(s) may be a charge couple device (CCD) operable to sense infrared radiation in the 1.0 to 5.0 μm wavelength range. The CCD may include an array of pixels of virtually any number. One exemplary microscope may be the Infrascope3 model supplied by Quantum Focus Instruments Corp. of Vista, Calif.

Attention is now turned to FIG. 2, which is a sectional view of FIG. 1 taken at section 2-2. Note that section 2-2 passes through the thermal plate 40, the test board 25 and the platform 20. The member or test board 25 is designed to hold a semiconductor chip or DUT 70. The test board 25 may be provided with a socket 65 that is operable to receive a semiconductor chip package substrate 75 upon which the DUT 70 is mounted. The DUT 70 may be a semiconductor chip, multiple such semiconductor chips, a circuit board or virtually any other device. A compression ring 80 is mounted to the semiconductor package substrate 75. The compression ring 80 serves two functions: first to provide an upper seating surface 85 upon which the thermal plate 40 may be seated; and second to provide a bath in which a liquid 90 may be filled to provide an infrared transmissive but thermally conductive liquid medium to transfer heat away from the DUT 70. The compression ring 80 may be fabricated from a variety of materials such as, for example, copper, brass, aluminum, nickel, combinations or laminates of these or the like. The transmissive fluid 90 may be a Galden® liquid or other infrared transmissive fluid. Galden® liquids are low molecular weight perfluoropolyether (PFPE) fluids having the general chemical structure of:

The thermal plate 40 is provided with one or more internal chambers, one of which is shown and labeled 95 that are operable to provide a circulation of cooling or heating fluid 100 in the thermal plate 40. Note that in this view, the supply/return line 45 is visible. The window 55 extends downwardly to a central bore 105 that is slightly smaller in diameter than the window 55 itself. The thermal plate 40 has a lower projection 110 that extends downwardly and encompasses the bore 105. The thermal plate 40 may be fabricated from a variety of materials, such as copper, brass, aluminum, nickel, combinations or laminates of these or the like. A transmissive window 115 is coupled to the projection 110. The transmissive window 115 is advantageously fabricated from a material that is highly transmissive of infrared radiation 120 that will be picked up by the objective lens 60 and sensed and analyzed by the microscope 15 and computing device 30 depicted in FIG. 1. Exemplary materials for the transmissive window 115 include diamond, sapphire, silicon or the like. Note that the compression ring 80 is provided with a height sufficient to elevate the transmissive window 115 above the device under test 75 so as to leave a small gap 125 between the two. The gap 125 is provided in order to eliminate or reduce the unwanted effects of Newton's rings that would otherwise be presented to the objective lens 60 due to non-planarity of the device under test 75 and/or the transmissive window 115. The transmissive fluid 90 serves as heat conductive and radiation transmissive film in the gap 125. In the setup depicted in FIG. 2, an optical pathway 123 to the camera 15 includes the transmissive fluid 90 and the transmissive window 115. The semiconductor chip 75 is also in the optical pathway 123. As described in more detail below, the infrared radiation 120 that actually traverses the optical pathway 123 and actually reaches the objective lens 60 and camera 15 will be an amalgam of infrared radiation emitted from the device under test 75, the liquid 90, and the transmissive window 115.

Note the locations of the dashed ovals 130 and 135. The portion of FIG. 2 circumscribed by the dashed oval 135 will be shown at greater magnification in FIG. 4 and used to describe in more detail the emission and absorption of infrared radiation from the various components depicted in FIG. 2. The dashed oval 130 will be shown at greater magnification in FIG. 3 and used to describe in more detail the coupling between the transmissive window 115 and the projection 110 of the thermal plate 40.

Attention is now turned to FIG. 3, which as just noted, is the portion of FIG. 2 circumscribed by the dashed oval 130 shown at greater magnification. The location of the dashed oval 130 is such that a small portion of the thermal plate 40 including a right hand side of the projection 110, as well as small portions of the window 55, the bore 105 and the transmissive window 115 are visible. The transmissive window 115 may be supplied with one or more metal rings, one of which is shown and labeled 145 and joined to the projection 110 of the thermal plate 40 by way of an adhesion layer 150, composed of solder or an adhesive or other well known fastening materials. The metal ring 145 may be fabricated from a variety of materials that are suitable to both adhere to the transmissive window 115 as well as whatever material is used to secure the ring 145 to the projection 110. Examples include gold, silver, copper, aluminum, combinations of these or the like. In an exemplary embodiment in which the transmissive window 115 is composed of diamond, the ring 145 may be composed of gold. Well known flash plating or other gold application techniques may be used. The liquid 90 may be filled to at least to a right edge 155 of the transmissive window or all the way up to the projection 110 as desired.

The behavior of the various infrared emissions and absorptions associated with the components in FIG. 2 will now be described in conjunction with FIG. 4, which is the portion of FIG. 2 circumscribed by the dashed oval 135 shown at greater magnification. In order to distinguish between various photons, different symbols are used for photons emitted from the transmissive window 115, photons emitted from the device under test 75 and photons emitted from the liquid 90. These various discrete symbols are shown in the key in FIG. 4. The device under test 75 will emit infrared photons as a function of temperature. Some of these photons will be absorbed or reflected by the liquid 90 and others will be absorbed or reflected by the transmissive window 115. Thus, the total number of infrared photons that actually pass through the transmissive window 115 and up through the bore 105 and the window 55 to the objective lens 60 shown in FIG. 2 is actually some fraction of the total infrared emission of the device under test 75. However, both the liquid 90 and the transmissive window 15 also emit photons that pass through the bore 105 and the window 55 and reach the objective lens 60. Thus, the total infrared radiation that reaches the objective lens 60 is an amalgam of: (1) the photons that are emitted by the device under test 75 and that are not absorbed or reflected by either the liquid 90 or the transmissive window 115; (2) the photons that are emitted by the transmissive window 115; and (3) a fraction of those photons that are emitted by the transmissive fluid 90 since some of the photons emitted by the transmissive fluid 90 are absorbed or reflected by the transmissive window 115. The techniques disclosed herein provide for a calibration so that the mixed population of infrared photons that actually reach the objective lens 60 can be parsed appropriately so that the actual photon count from the device under test 75 may be accurately read and thus provide an accurate diagnostic of the operation of the device under test 75.

An objective of the techniques disclosed herein is to measure a photon count with the microscope 15 (see FIG. 1) and map that photon count to a particular temperature in a DUT undergoing testing. For any photon radiator, the following expression applies:

R=eσT⁴   (1)

where R is the radiance of the radiator, e is the emissivity of the radiator, σ is the Stefan-Boltzmann constant, and T is the temperature of the radiator in Celsius or Kelvins. The radiance R is normally expressed in units of W/cm². However, any arbitrary unit may be used, such as total photon count, average photon count per sensor pixel or something else. The value of e varies with the composition and temperature of the radiator. Thus, it will be useful to obtain a data set to calibrate the lens 60 and the microscope 15 (see FIG. 1) based on a black body radiator and thereafter use that data set to determine temperatures of the DUT 70 (see FIG. 2) from photon counts of the DUT 70 during an electrical test thereof. The calibration procedure will account for the emission and absorption effects associated with the liquid 90 and the transmissive window 115.

The goal is to calibrate for the emission/absorbance characteristics of the components positioned in the pathway between the DUT 70 and the lens 60. As noted above, the presence of the components in the pathway between the DUT 70 and the lens 60 masks the actual photon count from the DUT 70 since the transmissive window 115 and the transmissive fluid both absorb and reflect some of the photons emitted by the DUT 70, both emit some photons themselves, and the transmissive window absorbs and reflects some of the photons emitted by the transmissive fluid 90. The relationship between the photon counts measured by the camera 15 and the actual photons emitted by the DUT 70 is given by:

MPC=t_wt_fAPC+b_w(T)+b_f(T)   (2)

where MPC is measured photon counts, t_w is the transmission of the transmissive window 115, t_f is the transmission of the transmissive fluid 90, APC is the actual photon counts, b_d(T) is the emission of the transmissive window 115, b_f(T) is the emission of the transmissive fluid 90, and T is the temperature. The transmission of a given film, either the transmissive window 115 or the transmissive fluid 90, is a measure of radiation reflected and absorbed by the film. Using the transmissive window 115 as an example, the transmission t_w is given by:

t_w=1−a_w−r_w   (3)

where a_w is the absorption by the transmissive window 115 and r_w is the reflectance by the transmissive window 115. The parameters t_w, a_w and r_w may be determined experimentally.

The quantities t_w, t_f, b_d(T) and b_f(T) may be determined experimentally as described below. Note from Equation 1 that the emissions b_f(T) and b_d(T) of the transmissive fluid 90 and the transmissive window 115 are functions of temperature T while the transmissions t_w and t_f of the transmissive window 115 and the transmissive fluid 90 are not dependent on temperature. Applicants have determined experimentally that the transmissions t_w and t_f for a transmissive window 115 composed of diamond and a transmissive fluid 90 composed of a Galden fluid are independent of temperature. The experiment to examine the impact of temperature on transmission involved sandwiching the transmissive window 115 and the fluid 90 between a radiation sensor, such as the camera 15 shown in FIG. 2, and a light source (not shown) and measuring the radiation reaching the sensor at various temperatures. The sensor was capable of Fourier transform infrared analysis. The results of the experiment established the temperature independence.

Calibration of Camera, Transmissive Window and Transmissive Fluid

In an exemplary embodiment, photon counts are taken from an experimental setup that initially includes just a black body emissivity target. Thereafter, additional components that affect the actual photon count, e.g., the transmissive window 115 and the transmissive fluid 90, are added to basic setup and photon counts are measured after each component is added. The result is a data set for a given temperature.

The basic initial experimental setup is illustrated in FIG. 5, which is a sectional view like FIG. 2, but of an exemplary emissivity target calibration setup which includes a platform, which may be the same platform 20 depicted elsewhere, a heater stage 160 positioned on the platform 20, the aforementioned compression ring 80 seated on the heater stage 160, and the thermal plate 40 seated on the compression ring 80 but without need for the transmissive fluid 90 (see FIG. 2) at this point. In this calibration setup, in lieu of a device under test, an emissivity or black body target plate 165 is seated on the heater stage 160. The emissivity target 165 is advantageously composed of a material(s) that is relatively thermally conductive, such as copper, gold, platinum, silver, nickel, combinations of these or the like. A black coating may be applied to the target 165 to enhance the black body effect. The black body target plate 165 may be provided with plural openings, two of which are visible in the sectional view in FIG. 5 and labeled 170 and 175 respectively. The opening 170 may be provided with a diameter D₁ that may be selected to correspond roughly in size to the field of view of the objective lens 60. The additional opening 175 may be provided with an opening diameter, D₂, that may correspond in size to a field of view of an additional objective lens on the microscope system that is not shown in FIG. 1. In this regard, the skilled artisan will appreciate that the microscope system 15 depicted in FIG. 1 may actually include several objective lenses that may be selectively used to focus on particular targets. A thermal grease (not shown) may be applied between the plate 165 and the heater stage 160 in order to facilitate the flow of heat from the stage 160 to the plate 165.

To obtain photon emission data for the target 165 alone, the heater stage 160 may be brought up to a first selected temperature to in-turn bring the plate 165 up to a first selected temperature. The temperature in the target 165 may be sensed via a thermocouple or other sensor (not shown) associated with the target 165. When the selected temperature is reached, the infrared radiation 180 emanating from the opening 170 may be picked up by the objective lens (shown broken in this and subsequent figures) 60 and the camera 15. The microscope 15 will determine a photon count for some selected period of time t. In this illustrative embodiment, the time t may be about 2.0 seconds. The foregoing steps may then be repeated at two or three or four additional temperatures to obtain a range of data of photon counts from the opening 170 as a function of four different temperatures.

As noted in conjunction with FIG. 5, the emissivity target plate 165 may be provided with a plurality of openings. In this regard, attention is now turned to FIG. 6, which is an overhead view of the emissivity plate 165. The aforementioned openings 170 and 175 are shown with their respective diameters D₁ and D₂. Additional openings 190 and 195 may be provided in the target plate 165 to provide the capability of calibrating additional objective lenses as necessary. The number of openings 170, 175, 190 and 195 is largely a matter of design discretion.

Determination of Transmission t_w and Emission b_w(T)

The transmission of the transmissive window 115 t_w, is given by:

t_w=MPC_wcold/MPC_blackbody   (4)

where MPC_blackbody is the measured counts with just the black body target 165 in place and MPC_wcold is the measured counts with the black body target 165 heated to some temperature and the transmissive window 115 cooled via the thermal plate 40 to below an emission threshold temperature for the window 115. An exemplary temperature may be about 15° C. To obtain values of MPC_blackbody, experimental runs were performed with the basic setup shown in FIG. 5 with the black body target 165 heated to four temperatures. Three measurement runs were performed for each temperature. An objective lens 60 with a 1/2× magnification was used. The data is summarized in the following table where the values for MPC_blackbody are an average for three runs.

TABLE 1

Black Body Target

Temperature ° C.

MPC_blackbody

44.7

1573

60.3

2735

75.2

4682

90.2

7575

To obtain values for MPC_wcold, two measurement runs were performed with the basic setup shown in FIG. 5 modified as shown in FIG. 7 where the transmissive window 115 and the thermal plate 40 are included between the black body target 165 and the lens 60 of the camera 15. Thermal plate 40 is seated on the compression ring 80, which is seated on the heater stage 160 and platform 20. The values for MPC_wcold were obtained with the transmissive window 115 held at about 15° C. The transmissive window 115 will not emit at this temperature. The data is summarized in the following table where the values for MPC_wcold are an average for the two runs:

TABLE 2

Black Body Target

Transmissive Window

Temperature ° C.

Temperature ° C.

MPC_wcold

44.7

15

1234

60.3

15

2087

75.2

15

3514

90.2

15

5596

The data from TABLES 1 and 2 may be combined in another table as follows:

TABLE 3

Black Body Target

t_w (according

Temperature ° C.

MPC_blackbody

MPC_wcold

to Eq. 4)

44.7

1573

1234

0.78

60.3

2735

2087

0.76

75.2

4682

3514

0.75

90.2

7575

5596

0.73

Determination of Emission b_w(T)

The emission b_w(T) due to the transmissive window 115 is given by:

b_w(T)=MPC_whot−MPC_wcold   (5)

where MPC_whot is the measured photon count when the transmissive window 115 is heated to a given temperature above an emission threshold temperature. In this illustrative embodiment, a temperature exceeding an emission threshold temperature for the transmissive window 115 of about 80° C. was used. The transmissive window 115 is advantageously heated to a temperature appropriate for calibrating an emissivity. The data is summarized in the following table where the values for MC_whot are an average for three runs:

TABLE 4

MPC_wcold

Black Body

Transmissive

(from

b_w(T)

Target

Window

TABLE

(according

Temperature ° C.

Temperature ° C.

MPC_whot

3)

to Eq. 5)

44.7

80

1985

1234

751

60.3

2848

2087

761

75.2

4269

3514

755

90.2

6358

5596

762

Determination of the Transmission t_f of the Transmissive Fluid

The determination of the transmission t_f and the emission b_f(T) due to the transmissive fluid 90 requires more complicated experimental setups than the setup depicted in FIG. 6. Two exemplary setups are depicted in FIGS. 8 and 9, respectively. In each of the setups, two thermal plates 40 and 200 are stacked over the black body target 165 such that the transmissive window 115 of one thermal plate 40 is facing towards but separated from a transmissive window 205 of the other thermal plate 200 by a gap 210. The thermal plates 40 and 200 may be substantially identical in construction with one thermal plate 200 flipped over relative to the other thermal plate 40. The thermal plates 40 and 200 may be supported by a frame 215 that may be seated on the platform 20 and include a support 220 for the thermal plate 200 and a support 225 for the thermal plate 40. An adjustment member 230 may be interposed between the thermal plate 40 and the support 225 and fitted with one or more set screws 235 and 240. The adjustment member 230 may be a ring coupled to both set screws 235 and 240, or discrete pieces, one for each set screw 235 and 240. The set screws 235 and 240 may be turned to adjust the vertical position of the thermal plate 40, and thus the vertical dimension of the gap 210. Of course, a myriad of designs could be used to support the thermal plates 40 and 200. For ease of illustration, the gap 210 is shown greatly exaggerated in size. The gap 210 should have about the same vertical dimension as the gap 125 in FIG. 2. In an exemplary embodiment the gap 210 may be about 120.0 microns, though other sizes are possible.

Note that the setups in FIGS. 8 and 9 each include the compression ring 80. The setup shown in FIG. 9 includes the transmissive fluid 90 in the gap 210 and contained by the compression ring 80. The second transmissive window 205 is necessary at this phase so that a transmissive pathway exists for photons from the black body target 165 to the transmissive fluid 90 and the transmissive window 115. Although data will eventually be taken using the setup in FIG. 9, the transmission characteristics of just the two windows 115 and 205 must first be determined using the setup of FIG. 8.

It will be necessary to first establish baseline photon counts for the dual transmissive windows 115 and 205 at cold and hot temperatures and without the transmissive fluid 90 in place. Using the setup depicted in FIG. 8, the black body target 165 is again heated to four temperatures using the heater stage 160. This time, both windows 115 and 205 are held at a constant low temperature of about 15° C. and dual transmissive window photon counts, MPC_wwcold, are measured by the camera 15, where “wwcold” denotes a window-window cold arrangement. Although FIG. 8 depicts emission of photons from both the transmissive windows 115 and 205 for purposes of illustrating the next step, there will not be such emissions at 15° C. Next, the black body target 165 is heated to each of the four temperatures while the dual transmissive windows 115 and 205 are heated to a temperature appropriate for an emissivity calibration and dual transmissive window photon counts, MPC_wwhot, are recorded, where “wwhot” denotes a window-window hot arrangement. In an exemplary embodiment, the dual transmissive windows 115 and 205 are heated to about 80° C. With both transmissive windows 115 and 205 heated to at least 45° C., there will be photons emitted from each as depicted in FIG. 8. The data is summarized in the following two tables where the values for MC_wwcold and MC_wwhot are each an average for three experimental runs:

TABLE 5

Temperature of Black

MC_wwcold (both transmissive

Body Target ° C.

windows @ 15° C.)

45

1129

60.2

1813

75.2

2910

90.1

4500

TABLE 6

Temperature of Black

MC_wwhot (both

Body Target° C.

transmissive windows @ 80° C.)

45

2530

60.2

3206

75.2

4286

90.1

5882

With data in hand for the measure photon count with the dual transmissive windows 115 and 205 but without the transmissive fluid 90, the calibration procedure is switched to the setup depicted in FIG. 9 with the transmissive fluid 90 in place. Again, the black body target 165 is heated to four temperatures using the heater stage 160, while the combination of the dual transmissive windows 115 and 205 and the transmissive fluid 90 is held to about 15° C. and photon counts, MC_wfwcold, are measured by the camera 15, where “wfwcold” denotes a window-fluid-window cold setup. Next, the black body target 165 heated to the four temperatures while the transmissive windows 115 and 205, and the transmissive fluid 90 are heated to a temperature appropriate for an emissivity calibration and photon counts, MC_wfwhot, are measured, where “wfwhot” denotes a window-fluid-window hot setup. The data is summarized in the following two tables where the values MC_wfwcold and MC_wfwhot are each an average for three experimental runs:

TABLE 7

MC_wfwcold

Temperature of

(dual transmissive windows and

Black Body Target ° C.

transmissive fluid @ 15° C.)

45

1054

60.2

1696

75.2

2725

90.1

4256

TABLE 8

MC_wfwhot

Temperature of

(dual transmissive windows and

Black Body Target ° C.

transmissive fluid @ 80° C.)

44.9

3362

60.2

3991

75.1

5014

90.2

6533

It will be useful at this point to combine the data from TABLES 5, 6, 7 and 8 into TABLE 9 as follows:

TABLE 9

MC_wwcold

MC_wwhot

MC_wfwcold

MC_wfwhot

1129

2530

1054

3362

1813

3206

1696

3991

2910

4286

2725

5014

4500

5882

4256

6533

A few qualitative observations may be made about the data in TABLE 9. First, the addition of the transmissive fluid 90 caused the photon counts to go down slightly. For example, at a temperature of 45° C., the photon counts decreased from 1129 without the fluid to 1054 with the fluid, a drop of 75 photons. At a temperature of 60.2° C., the photon counts decreased from 1813 to 1696, a difference of 117 photons. Qualitatively, the decrease in photon counts with the addition of the fluid 90 makes sense since the fluid 90 is absorbing some photons. However, the applicants have also discovered that the thickness of the transmissive fluid 90 can impact the measured counts in a counterintuitive way. If the thickness of the fluid 90 is dropped from about 120.0 microns to about 30.0 microns, the measured counts MC_wfwcold with dual windows 115 and 205 and fluid 90 becomes larger than the measured counts MC_wwcold with just two windows 115 and 205. Applicants believe the increase is due to the fluid 90 reducing the reflectance of the interface between the top transmissive window 115 and the fluid 90. Second, heating the transmissive fluid 90 produces more fluid emission as evidence by the larger counts with fluid MC_wfwhot versus counts without fluid MC_wwhot.

With the data from TABLE 9 in hand, the transmission t_f and the emission b_f(T) due to the transmissive fluid 90 may be calculated. The fluid transmission t_f is given by:

t_f=MC_wfwcold/MC_wwcold   (6)

and the fluid emission b_f(T) is given by:

b_f(T)=MC_wfwhot−(MC_wwhot)(t_f)   (7)

Plugging the data from TABLE 9 into Equations 6 and 7 yields:

TABLE 10

Temperature of Black Body

Target and dual transmissive

windows and transmissive

Transmission

Emission of

fluid ° C.

of fluid t_f

fluid b_f(T)

44.9

0.9336

1000

60.2

0.9355

991

75.1

0.9364

1000

90.2

0.9458

969

The quantities t_w, t_f, b_w(T) and b_f(T) set forth in TABLES 3, 4 and 10 satisfy Equation 2 and characterize the general transmission and emission characteristics of the transmissive window 115 and the transmissive fluid 90. The data and Equation 1 may be used to calibrate the photon measurement for an actual sample or DUT.

Calibration of a DUT

To calibrate an actual sample or DUT 70, the basic setup depicted in FIG. 2 may be used where the DUT 70 is positioned in the optical pathway 123. The DUT 70, the transmissive fluid 90, and the transmissive window 115 are heated to some temperature, for example 80° C., and measured photon counts MPC_pixel are taken on a per pixel basis. The heat may be supplied by the thermal plate 40. During the measurement, the DUT 70 chip is substantially isothermal. If desired, the measurement may be repeated at other temperatures of interest. The photon counts MPC_pixel measured during the test are run through Equation 2 using the data from TABLES 3, 4 and 10 to yield an actual photon count per pixel APC_pixel at a set temperature T, in this case 80° C. The basic radiance equation, Equation 1, may be modified and used to solve for emissivity on a per pixel basis as follows:

R_pixel=APC_pixel=e_pixelσT⁴   (8)

Rearranging Yields:

e_pixel=APC_pixel/σT⁴   (9)

Actual Temperature Measurement on a DUT

Still referring to FIG. 2, to make actual temperature measurements on the DUT 70, the transmissive window 115 and the transmissive fluid 90 are kept cool, at perhaps 15° C., by appropriate coolant circulation in the thermal plate 40. The DUT 70 is caused to perform one or more test patterns or scripts via the computing device 30 shown in FIG. 1 and photon counts MPC_pixeldata are measured on a per pixel basis. The transmissive window 115 and the transmissive fluid 90 act as transmissive heat sinks for the DUT 70, which is generating heat non-uniformly across its surface. Since both the transmissive window 115 and the transmissive fluid 90 are cooled, there should be no emission by either. Accordingly, the terms b_f(T) and b_w(T) from Equation 2 are set to zero. The terms t_w and t_f from Tables 3 and 10 may be used to convert measured photon counts MPC_pixeldata to actual photon counts APC_pixeldata using Equation 2. The actual photon counts APC_pixeldata and the emissivity per pixel values e_pixel from Equation 9 may be used to solve for a temperature at a given pixel to yield a temperature map of the DUT 70.

Referring again to FIG. 1, the computing device 30 may be provided with instructions to enable the automated gathering of data and calculations necessary to solve for the variables in Equations 2-9 and, if desired, create and store the calculated variables for subsequent temperature mapping of a given semiconductor chip. The data may be stored in the form of look-up tables or the like. The instructions and data may be stored in a computer readable medium associated with the computing device 30.

While the invention may be susceptible to various modifications and alternative forms, specific embodiments have been shown by way of example in the drawings and have been described in detail herein. However, it should be understood that the invention is not intended to be limited to the particular forms disclosed. Rather, the invention is to cover all modifications, equivalents and alternatives falling within the spirit and scope of the invention as defined by the following appended claims.