Integrating spheres are ideal optical diffusers; they are used for radiometric measurements where uniform illumination or angular collection is essential, for reflectance and transmittance measurements, or even to mix two light sources. We offer five families of integrating spheres in sizes from 2 to 8 inches (51 to 203 mm), with three coating types to cover the following spectral ranges:

• Barium Sulfate for 350 – 1300 nm
• PTFE for 250 – 2500 nm
• Diffuse Gold for 7.5 – 10 µm

How Do Integrating Spheres Work?

Fig. 1 shows the principle of operation of integrating spheres. Light is collected by an integrating sphere and uniformly reflected and scattered around the sphere’s interior so the output is a uniform, spatially integrated field (Lambertian) of radiation which is insensitive to the spatial, angular and polarization changes of the input. Our integrating spheres come with 1.5 Inch Series input and output flanges so they can be mounted to any of our Oriel instruments and components.

LS-280NbFig. 1 Principle of operation of an integrating sphere. The incoming beam is diffusely reflected inside an integrating sphere. The output port cannot “see” the region of first impact.

The Ideal vs. The Real

These integrating spheres are equipped with internal baffles to prevent direct ray path from input, to output ports. These baffles and small wall imperfections make these “real” diffusers less perfect. That is, they are the closest to Lambertian diffusers of any simple optical devices, but exhibit some departure from the ideal.

Why an Integrating Sphere?

For critical radiometric measurements, always use an integrating sphere; they are close to ideal diffusers. Integrating spheres are considerably more expensive, and suffer similar or higher throughput losses than disk diffusers, but they provide a true cosine response and very repeatable uniform illumination of a spectrometer grating or detector. Disk diffusers are suitable for applications requiring relatively uniform illumination, such as detectors which are not uniformly sensitive across the photosensitive area.

Problems with Radiometric Measurements

The variation in the spatial and angular response of many detectors leads to incorrect results when the flux to be measured is non-uniform or the beam shifts slightly. Non-uniformity can originate from the source or optical path. Beam movement can come from moving components or refractive index variation in the air path. Measurements with fiber optics can be influenced by launch or fiber output pattern changes, and light moving across or off a detector.

Integrating Spheres as the Solution

Optical disk diffusers, properly used, significantly reduce sensitivity to these effects but for critical measurements they are not sufficient - you need an integrating sphere. Integrating spheres are also recommended for diffuse reflectance and transmittance measurements. Integrating spheres are standard for UV/VIS/NIR or IR analytical surface spectroscopy. Most backscatter and turbidity measurements are improved by use of an integrating sphere due to higher angle collection (a full 180° hemisphere). Flux measurements made with integrating spheres are more reliable. The sphere reduces or removes sensitivity to beam shape and angle, and detector spatial response variations.

High Reflectance

Most of our integrating spheres are available with a choice of three interior materials. For VIS-NIR applications choose our barium sulfate-based white coated spheres. The coating is highly reflective, >97% in the visible.

For UV, VIS or NIR measurements, we offer spheres with interiors machined from a unique PTFE thermoplastic material that is very rugged and highly reflective down to 250 nm. For applications in the NIR and IR regions our diffuse gold spheres provide near-lambertian characteristics and reflectance values up to 95%. All of our spheres are designed to provide high reflectance, stability and low throughput loss over their usable wavelength range.

High Damage Threshold

The excellent reflectance of our spheres gives them a high damage threshold. Measurements with a pulsed laser at 1.6 µm showed the damage threshold of the coated spheres to be 1.5 W cm-2. The PTFE spheres have an even higher damage threshold of 8 W cm-2, and the gold spheres exhibit a damage threshold of 19 W cm-2.

5496Fig. 2 Reflectance and throughput of 4 inch PTFE and BaSO4 Spheres.5496Fig. 3 Reflectance of Diffuse Gold

Which Sphere Do I Choose?

We offer five integrating sphere designs with a choice of interior coatings covering the UV-VIS through the IR. Your choice is often a tradeoff between throughput and uniformity needs. For example, our 8 inch spheres give the most uniform output, but suffer the greatest throughput losses due to the large diameter of the sphere.

Monochromator Spheres



These spheres have excellent diffusing properties, proper baffling to prevent first strike radiation from entering the instrument. 1.5 Inch Series flanges and a 1/4-20 tapped hole facilitate mounting.

Fiber Optic Spheres



These spheres are a smaller version of the General Purpose Sphere. They are 4 inches in diameter, suitable for small light beams. There are three ports on the equator; two have 11 mm internal diameters to fit any Oriel Fiber Bundle directly. For SMA terminated fibers you will also need the 70463 Adapter. The third port is a 1.5 Inch Series male flange.

General Purpose Spheres



These General Purpose Integrating Spheres are 4 and 6 inches (102 and 152 mm) in diameter. They include standard Oriel 1.5 Inch Series flanges and 1/4-20 tapped holes for rod mounting. If you need to couple these spheres to instruments with male flanges (such as a monochromator), order the 77829 Coupling Ring. A diagram of the light path for the 70451 is shown in Fig. 1.

Uniform Light Source Spheres



Place one of these integrating spheres at the output of a light source and you have an almost perfectly uniform Lambertian, broadband source. Typical radiance uniformity is within 1-2% of the average over the 2 inch (51 mm) diameter exit port as long as the FOV of the imaging system stays on the baffle within the sphere. Irradiance uniformity at the port is also better than 1-2%.

For a tunable, uniform light source, couple the sphere to the input of a monochromator (or motorized filter wheel). You\'ll need the 77829 Coupling Ring to attach one of these to an Oriel Monochromator. To make a uniform calibration source with several orders of luminance range, attach a filter wheel with neutral density filters to the sphere\'s input. The input port of the sphere has a 1.5 Inch Series male flange so it can be mated to an Oriel Light Source. The output port is a 2 inch (51 mm) diameter port frame for simple mounting of samples.

Reflectance and Transmittance Spheres



The 70679 and 70682 are for measurement of specular and diffuse hemispherical reflectance and total transmittance using a 8/D geometry (8 degree beam incidence/Diffuse collection). The 70679 is good for the VIS-NIR regions and the 70682 is good for UV-VIR-NIR (Note the ID of the 70682 is ~7.0”). The spheres have 5 ports:



A reflectance sample holder with 1 inch (25 mm) diameter, 0.5 inch (13 mm) deep cell holds the sample at 8° to the beam. The clear aperture is 0.7 inch (18 mm). The holder is spring loaded so that square, rectangular or irregular shapes with dimensions up to 2 inches (51 mm) can be held against the sphere wall. (The sample must fill the aperture.) The spring loading allows you to quickly insert and remove the sample and the 70496 White Spectral Calibration Disk. Light, which is reflected from the sample strikes the port and is turned 8 degrees to the specular exclusion port. If a plug is at the exclusion port, the light is re-captured in the sphere (specular inclusion – total reflectance 8/D). If a light trap is used in the exclusion port, the specular portion of the sample’s reflectance is subtracted from the measurement (Specular excluded – diffuse only reflectance 8/D– allows characterization of “gloss”). One exclusion port plug is included with the sphere and one extra port plug is included for the north pole port.

Conversely, if the 0° Sample Holder (also included with the 70679 and 70682 R/T Spheres) is used at the sample port and the input beam is placed behind the sample port, then you can qualify diffuse, normal or total transmittance of samples. If a plug is used at the 1.0 inch port opposite the sample (former input port), then all energy is included in the measurement (total transmittance). If a light trap is used at this port opposite the sample, then the normal transmitted beam (specular transmission) is excluded by the trap (diffuse or normal excluded transmittance – allows “haze” characterizations).

White Spectral Calibration Disk

The 70496 is a 1.25 inch (30 mm) diameter, 11 mm thick white Zenith® disk, for spectral calibration purposes. It has >95% reflectance from 250 to 2500 nm (>98% from 400-1800 nm), is durable, and is an almost perfect reflecting Lambertian diffuser. It may be lightly abraded to restore performance if it becomes dirty. The measured Hemispherical Reflectance Factor from 250 to 2500 nm is supplied for each disk. This calibration data is traceable to NIST.

Calculating Sphere Throughput

Sphere throughput is the ratio of the total flux out to the total flux in. It depends on the sphere\'s reflectance properties, diameter, and number of ports.

mod

Where
φe = total flux exiting port
φi = total incident flux
Ae = area of exit port
As = surface area of sphere
Ap = sum of all port areas
ρ = sphere wall reflectance (0 ≤ρ ≤1)

For practical purposes, it should be noted that the equation is specific to a single port and that for a fixed reflectance ρ at a specific wavelength, it is solely dependent on the ratios of port areas to total surface area of the sphere. It can be easily shown that for our series of spheres in which the output/input ports are nominally the same dimension (35 - 51 mm CA), Asdominates. Therefore, when choosing a sphere, select the smallest diameter available that provides the output port characteristics (dimension and uniformity) to maximize throughput. The basic equation also neglects the Lambertian quality of the output and is only shown to reflect the major dependencies for selection of basic spheres rather than to provide specific throughput values for the spheres.

Guidelines for Best Performance

1. Plug unused ports with diffusing plugs.
2. Avoid large changes in reflectance when making comparative reflectance measurements. For example, compare a sample with a standard of similar reflective properties.
3. Keep samples and detectors “in the sphere wall,” particularly if high scatter angle measurements are involved.
4. Use spectral filters before light enters the sphere rather than at the detector, especially if changes are involved during a measurement run. Remember that the output is a Lambertian radiator so any filter “sees” light incident at a wide range of angles.