42 |
References for 1.1 |
99Ber |
Bergmann/Sch¨afer: Lehrbuch der Experimentalphysik, Vol. 3, Optics, Niedrig, H. (ed.), |
New York: W. de Gruyter, 1999. |
|
99Bor |
Born, M., Wolf, E.: Principles of optics, Cambridge: Cambridge University Press, 1999. |
99Jac |
Jackson, J.D.: Classical electrodynamics, New York: John Wiley & Sons, 1999, p. 410. |
00Dav |
Davies, C.C.: Laser and electro-optics, Cambridge: Cambridge Univ. Press, 2000. |
01I |
I ¨ander, R.: Solid State Lasers for Material Processing, Berlin: Springer-Verlag, 2001, |
p. 257–318. |
|
01Men |
Menzel, R.: Photonics, Berlin: Springer-Verlag, 2001. |
01Vog |
Vogel, W., Welsch, D.G., Wallentowitz, S.: Quantum optics – an introduction, Berlin: |
Wiley-VCH, 2001. |
Landolt-B¨ornstein
New Series VIII/1A1
Ref. p. 51] |
2.1 Definition and measurement of radiometric quantities |
45 |
2.1Definition and measurement of radiometric quantities
B. Wende, J. Fischer
Radiometry is the science and technology of the measurement of electromagnetic energy. Here we confine ourselves on the subfield of optical radiometry which covers the measurement of electromagnetic radiation in the wavelength range from about 0.01 µm to 1000 µm. Radiometric quantities are derived from the quantity energy. The corresponding photometric quantities on the other hand involve the additional evaluation of the radiant energy in terms of a defined weighting function, usually the standard photometric observer. In the following only the definitions of the radiometric quantities are explained in detail. Starting from the radiant energy the other fundamental radiometric quantities radiant power, radiant excitance, irradiance, radiant intensity, and radiance are derived by considering the additional physical quantities time, area, and solid angle.
The radiometric quantities defined in abstract terms are practically embodied by radiometric standards. Radiometry is based on primary detector standards and primary source standards. Primary detector standards are mostly electrical-substitution thermal detectors whereas for primary source standards the emitted radiant power is accurately calculable. For the radiometric measurement of cw laser emission radiation detectors or radiometers calibrated against primary detector standards are the preferred secondary standards. The detection principle of the radiometers could be thermal (thermopiles, bolometers, and pyroelectric detectors) or photoelectric (semiconductors). As secondary standards for pulsed laser radiation mostly thermally absorbing glass-disk calorimeters are used. These standards are derived from the cw standards using accurately measured shuttering of the laser radiation to produce pulses of known radiant energy.
2.1.2 Definition of radiometric quantities
Radiometric and photometric quantities are represented by the same principal symbol and may be distinguished by their subscripts. While radiometric quantities either have the subscript “e” or no subscript (as in the whole Chap. 2.1), photometric quantities have the subscript “v”, where “e” stands for “energetic” and “v” for “visible”. The most frequently used radiometric quantities are listed in Table 2.1.1 together with their symbols, defining equations, and units. The additional physical quantities applied in Table 2.1.1 are the time t, the element of solid angle d ω, and the angle θ between the line of sight and the normal of the radiating or receiving surface with the area element dA, see Fig.2.1.1.
In the case that the quantities are functions of wavelength their designations must be preceded by the adjective “spectral”. For example, the symbol for spectral radiance is L(λ). This has to be well distinguished from the convention for the spectral concentration of a quantity, which is also preceded by the adjective “spectral”. In that case, however, the symbol has the subscript λ, i.e. dL/dλ = Lλ.
Landolt-B¨ornstein
New Series VIII/1A1
46 |
2.1.2 Definition of radiometric quantities |
[Ref. p. 51 |
Table 2.1.1. Radiometric quantities, their defining equations and units.
Quantity |
Symbol |
Defining equation |
Unit |
Radiant energy |
Q |
J |
|
Radiant power |
Φ |
Φ = d Q/dt |
W |
Radiant excitance |
M |
M = d Φ/dA |
W m−2 |
Irradiance |
E |
E = d Φ/dA |
W m−2 |
Radiant intensity |
I |
I = d Φ/d ω |
W sr−1 |
Radiance |
L |
L = d2 Φ/(cos θ dA d ω) |
W m−2 sr−1 |
n |
|||
dAn |
|||
θ |
|||
d ω |
|||
dA |
|||
ϕ |
|||
Fig. 2.1.1. Geometry for definition of the radiance. |
|||
To explain the defining equations given in Table 2.1.1 a radiation source of finite extent is considered. If we surround the radiation source with a closed surface and calculate the radiant energy Q penetrating the surface per unit time we get the total radiant power Φ emitted by the source. For clarity, the above mentioned symbols for the spectral properties of the radiation are omitted in this chapter. The radiant power per unit area of the radiation source associated with the emission into the hemispheric space above dA is defined as the radiant excitance M . At this point it is appropriate to introduce the radiation incident from all directions in the hemispheric space above the surface of a detector. The irradiance E is defined as the radiant power incident on a surface per unit area of the surface. The irradiance represents also the energy which propagates per unit time through the unit area perpendicular to the direction of energy transport. This is known as the density of energy flow identical to the magnitude of the Poynting vector averaged over time.
Coming back to the source-based radiometric quantities we consider now the radiant power proceeding from a point source per unit solid angle d ω in a specified direction. The corresponding quantity appropriate especially for nearly point-shaped sources is denoted as radiant intensity I. If we generalize and consider again a source of finite extent the directional nature of radiation has to be taken into account accurately. From Fig. 2.1.1 we formally define as radiance L the radiant power emitted in the (θ, ϕ) direction, per unit area of the surface normal to this direction and per unit solid angle. Note that the area dAn used to define the radiance is the component of dA perpendicular to the direction of the radiation. This projected area is equal to cos θ dA and in e ect, this is how dA would appear to an observer situated on the surface in the (θ, ϕ) direction.
Although the directional distribution of surface emission varies according to the nature of the surface, there is a special case which provides a reasonable approximation for many surfaces. For
Landolt-B¨ornstein
New Series VIII/1A1
Ref. p. 51] |
2.1 Definition and measurement of radiometric quantities |
47 |
an isotropically di use emitter the radiance is independent of direction:
L(θ, ϕ) = L . |
(2.1.1) |
Such an emitter is denoted as a lambertian radiator which emits in accordance with Lambert’s cosine law:
I(θ) = I(0) cos θ . |
(2.1.2) |
The radiant intensity of a perfectly di use surface element in any direction varies as the cosine of the angle between that direction and the normal to the surface element. It is noted that this law is consistent with the definitions of radiance and radiant intensity given in Table 2.1.1. It may be helpful to derive the relationship between radiance and radiant excitance for a lambertian radiator. The radiant excitance into the hemispheric space above dA is calculated from the radiance by integration over the solid angle d ω = sin θ dθ dϕ:
M = dA |
2π π/2 |
(2.1.3) |
|||
= |
L(θ, ϕ) cos θ sin θ dθ dϕ . |
||||
d Φ |
|||||
0 |
0 |
||||
By removing L(θ, ϕ) from the integrand according to (2.1.1) and performing the integration we get
M = π L . |
(2.1.4) |
Note that the constant appearing in the above expression is π, not 2π, and has the unit steradian (sr).
Depending on the application primary source and primary detector standards are used to establish radiometric scales. Black-body radiators of known temperature with calculable spectral radiance are operated as primary source standards at temperatures up to about 3200 K [96Sap]. Due to the steep decrease of their Planckian radiation spectrum in the UV spectral range radiometry with black-body radiators is limited to wavelengths above 200 nm. In comparison with a black-body radiator, the maximum of the synchrotron radiation spectrum emitted by an electron storage ring is shifted to shorter wavelengths by several orders of magnitude [96Wen]. In a storage ring electrons move with nearly the velocity of light along a circular trajectory and emit a calculable radiant power through an aperture stop situated near the orbital plane. Radiometry can thus be extended into the X-ray region up to photon energies of 100 k eV.
Electrical-substitution thermal detectors operated at ambient temperature have been the most frequently used primary detector standards. However, their performance is limited by the thermal properties of materials at room temperature resulting in complicated corrections that have to be applied. Hence, their uncertainties remain near 0.1 % to 0.3 % [89Fro, 79Wil]. Cryogenic radiometers have been developed to satisfy the increasing demands for more accurate detector standards from users especially in new and expanding fields of optical fibers, laser technology, and space science. Today, these instruments with absorption cavities at nearly the temperature of liquid helium
Landolt-B¨ornstein
New Series VIII/1A1
48 |
2.1.3 Radiometric standards |
[Ref. p. 51 |
Helium reservoir
Cavity
Quadrant diodes
Brewster window
Photodiode
Laser |
Stabilizer |
Spatial filter |
Monitor |
||||||
Fig. 2.1.2. Cryogenic radiometer for the calibration of photodiodes. The stabilized laser beam enters the cryostat via a Brewster window and is aligned by quadrant photodiodes. In the cavity the laser radia-
Mirror tion is absorbed and electrically substituted.
are the most accurate among all primary standards, with relative uncertainties of less than 0.01 % [85Qui, 96Fox]. The principle of operation of both the cryogenic radiometers and the instruments at ambient temperature is that a thermometer measures the temperature rise of an absorption cavity, relative to a constant-temperature heat sink, during radiant and electrical heating cycles. By adjusting the electrical power so that the absorption cavity temperature rise is the same for both types of heating, the radiant power can be equated to the easily measured quantity of electrical power. For cryogenic radiometers the corrections due to the limited absorptance of the cavity, the lead heating of electrical connections, the radiative heat loss, and the background radiation can be made su ciently small to reach very accurate equivalence of optical and electrical heating [96Fox]. Today, high-precision calibrations of laser radiometry secondary standards are mostly traceable to cryogenic radiometers. In Fig. 2.1.2 a typical experimental arrangement for the calibration of transfer photodiodes is shown [93Fu].
Secondary standards serve to disseminate a metrologic scale or quantity to the user in science and industry. In this section, first, the common detectors used in the secondary standards for laser radiometry are shortly described, and second, some examples for laser radiometers and calorimeters are given. The detection principle of the secondary standards is usually thermal or photoelectric. The thermal detectors have the remarkable advantage of a flat spectral responsivity function which makes the calibration for di erent laser wavelengths not necessary or at least easier compared to that of photoelectric detectors. Among the thermal detectors we distinguish between thermopile detectors, bolometric and pyroelectric detectors.
A thermopile consists of a number of thermocouples in series to provide a thermoelectric voltage proportional to the temperature di erence between the receiver and its thermal environment. Its optimization in detector applications has received considerable attention [68Smi, 58Sch, 70Ste]. At this point the term responsivity s is introduced which is the ratio of the detector output to the detector input. Whereas the detector input is a radiometric quantity, the detector output is usually an electrical quantity, for example current, voltage, or change in resistance. In order to optimize the responsivity of a thermopile one has to maximize the Seebeck coe cient of the two materials used for each thermocouple, the thermal resistance between the receiver and the environment, and the absorptance of the surface. The materials used for thermocouples are either metals, alloys, or semiconductors, for examples see [89Hen].
A bolometer is a temperature transducer based on the change of electrical resistance with temperature. The important quantity is the temperature di erence between the receiver and its
Landolt-B¨ornstein
New Series VIII/1A1
Ref. p. 51] |
2.1 Definition and measurement of radiometric quantities |
49 |
1 |
5 |
2
6 |
|
3 |
7 |
4
9 |
8 |
10 |
Fig. 2.1.3. Cross section of a cone-shaped laser radiometer. 3 : blackened cone, 6 : aperture, 7 : heat protection tube, 8 : electrical heater, 9 : electrical connections, 10 : thermopile; 1, 2, 4, 5 : parts of the heat sink.
thermal environment. Therefore one resistance element is needed to measure the temperature of the receiver and one to measure that of the thermal environment. AC and DC bridge techniques are applied for the comparison, the most common employing Wheatstone bridge configurations. The second resistance element should be physically close to the radiation-measuring element to compensate for convective disturbances, pressure fluctuations, changes in temperature of the housing, and instabilities in the bridge supply. The resistors are preferably made of metal wires or films of nickel, platinum, or gold [65Ble]. Thermistors are also used which have a larger temperature coe cient of the resistance. At lower operation temperatures the signal-to-noise ratio of bolometers can be increased considerably [82Mat, 87McD].
Pyroelectric detectors produce a current proportional to the rate of temperature change. The detection mechanism is based on the temperature dependence of the electrical polarization in ferroelectric crystals. Since pyroelectric detectors respond to modulated radiant power only, their use in laser radiometers for measuring cw radiation requires chopping of the incident beam. This can provide considerable drift immunity and allows for the use of drift-free AC amplification techniques [70Put, 75Tif].
Beside the thermal detectors also photoelectric devices or quantum detectors are used in laser radiometry. Photoelectric detectors for laser radiometric applications are either photoconductors or photodiodes. In a photoconductor made of a thin film of a semiconductor material the incident radiation generates additional carriers. These intrinsic band-to-band transitions or extrinsic transitions involving forbidden-gap energy levels result in an increase of conductivity [81Sze]. For sensitive infrared detection, the photoconductor must be cooled in order to reduce thermal ionization of the energy levels. In photodiodes the carriers are mainly generated in the depletion layer of the diode junction. The electron-hole pairs separated by an internal or external electric field recombine by driving an external current. Photodiodes are operated in two di erent modes: In the photovoltaic mode no bias voltage is applied and the photodiode can be considered as current source. In contrast, in the reverse-bias mode the photocurrent generates a voltage drop at an external load resistance which is used as measuring quantity. The reverse-bias mode is preferred for the detection of pulsed laser radiation.
A practical example of a radiometer for cw laser radiation is shown in Fig. 2.1.3. It measures radiant power in the range from 1 mW to 10 W, whereas the lower limit is set by detector and amplifier noise and the upper by the load limit of the electrical heater [89Moe]. The radiation absorber is a polished hollow cone electro-plated with a nearly specular reflecting black nickel layer. The temperature di erence between the absorber cone and the heat sink is measured by a thermopile. The electric heater for moderate-accuracy in-situ calibrations of the instrument is wound around the cone. Another design of a thermopile-type radiometer with an integral alignment module can be found in [88Ino]. Further similar systems are described in [77Gun, 91Rad]. A commercial version of a laser radiometer based on a pyroelectric lithium tantalate crystal is described in [89Hen]. For higher radiant power levels of up to 1 kW cavity absorbers cooled by a surrounding jacket of flowing
Landolt-B¨ornstein
New Series VIII/1A1