Integrating sphere reflectance accessories offer a simple method to extend the measurement capabilities of a UV-VIS or UV-VIS-NIR spectrophotometer. By adding a sphere reflectance accessory to a spectrophotometer, one is able to expand from traditional transmittance measurements of liquid samples and non-scattering solids to incorporate both reflectance measurements of opaque solids, powders, or pastes, as well as total transmittance scans of translucent films and scattering creams. Minimal preparation is required for most samples when a reflectance accessory is used, as samples can often be measured in an unprocessed form. Factors such as sphere efficiency, signal throughput, spectrum noise level, measurement accuracy, sphere port fraction, detector baffles, and sample beam size are key components of sphere design that are considered here.
Integrating Sphere Efficiency and Throughput
Smaller, 60 mm, integrating spheres are efficient collectors of light. The throughput of the system, that is, the amount of incident light which reaches the detector, is governed by the laws of probability. In a 60 mm integrating sphere, a photon will have to take a certain number of bounces before it reaches the detector. In a 150 mm sphere system, however, statistics dictate that a photon will have to take more bounces than were required in the 60 mm integrating sphere to reach the same detector. In the 150 mm sphere, since more “bounces” are required, the photon must undergo many more interactions with the sphere wall and, therefore, is more likely to be absorbed before it actually reaches the detector. Thus, a large 150 mm integrating sphere acts as an attenuator of signal and is inherently less efficient than a 60 mm diameter sphere. The larger the sphere, the greater the attenuation which results. As a rule of thumb, all other factors being equal, the relative attenuation is roughly equal to the square of the ratio between the two sphere diameters. This rule can only be used as a rough guide.
Spectrum Noise Level
A sphere’s efficiency may play an important role in the measurement of highly absorbing samples. When the high absorbance sample being tested limits the amount of light reaching the sphere’s detector, a high throughput, 60 mm sphere will generate better spectra with less noise than a 150 mm sphere mounted to the same spectrophotometer and using the same detector. Thus, spectral results of 60 mm spheres tend to have greater signal-to-noise levels than those of 150 mm spheres. In addition, the 60 mm sphere does not attenuate the instrument’s sample beam as significantly as a 150 mm sphere, thereby, only slightly affecting the overall linearity range of the instrument.
Port Fraction
Customers of spectrophotometer sphere reflectance accessories generally have access only to generic sphere designs which cannot be modified to fit individual needs. In this case, it is important to understand the effects which the sphere’s diameter and port fraction have on the performance of the integrating sphere accessory.
The port fraction is defined as the ratio of the total port area relative to the total internal surface area of the sphere. All beam entrance ports, sample ports, and detector ports which are filled with material of lower reflectance than the Spectralon sphere wall contribute to the calculated port fraction. The port fraction is significantly lower for 150 mm diameter spheres than it is for 60 mm spheres. For example, the port fraction of a representative 150 mm double beam integrating sphere accessory is 2.5 %, while a 60 mm sphere for the same instrument has a port fraction of 11.3%. The design of both accessories includes sample and reference beam transmittance and reflectance ports, as well as PMT and PbS (or InGaAs) detector ports. In order to adhere to many ASTM and CIE methods for measurements using integrating spheres, the port fraction of the sphere must be minimized. For instance, CIE recommends that the sphere’s port fraction be lower than 10% for color reflectance measurements, whereas ASTM D1003-95 requires the sphere to have a total port fraction less than 4% for haze measurements on transparent plastics. Thus, 150 mm diameter integrating spheres can be used for these methods. A 60 mm diameter integrating sphere with the standard transmittance, reflectance, and detector ports is often unable to meet these strict port fraction requirements. A low port fraction ensures good integration of the sample signal before it reaches the sphere’s detector. The influence of port fraction on sphere radiance is discussed further in the next section.
Measurement Accuracy
Obviously, the integrating sphere’s design will affect its measurement accuracy. The size and location of ports, detectors, and baffles will influence how the light bounces around the sphere. As will be discussed in this section, large 150 mm diameter spheres have better light integration and their measurements are less likely to be affected by hot spots. The signal integration is not as good in smaller spheres, and the large port fraction typically found in 60 mm spheres can introduce significant errors in measurement due to flux loss. All of these factors must be considered when choosing an integrating sphere accessory which is appropriate to the user’s application.
When designing integrating spheres, it is important that the detector’s field of view does not include any portion of the sphere surface directly irradiated by the sample beam or the first reflection from the sample. This would introduce a false response into the measurement. Baffles are typically made from thick pieces of Spectralon or from metal which has been coated with the same material as the integrating sphere wall. Baffles are used to block the detector’s view of light which has not undergone at least two reflections from the sphere surface. Thus, the baffle is positioned to prevent the so-called ‘first-strike’ reflections from entering the field of view of the detector.
The size and position of baffles within the integrating sphere are very important factors which influence the system’s measurement accuracy. As described in ASTM E903, “large errors can arise if the angular distribution of the light reflected from the specimen is different from that reflected by the standard.”An example is found in transmittance measurements of translucent samples. The translucent sample, which scatters light, is measured with respect to the non-scattering open port (air), which is used for the background correction. Careful baffle design can substantially reduce errors due to the different light scattering distribution of samples and standards. However, baffle design must always be performed with respect to the overall radiance characteristics of the sphere. The balance between baffle design and sphere flux is an important consideration when choosing an integrating sphere design.
The distribution of light within an integrating sphere will drastically affect its measurement accuracy. While small spheres do have higher energy efficiency than their 150 mm diameter counterparts, large integrating spheres will yield measurements with greater accuracy since the light in large systems can be ‘integrated’ or distributed evenly about the sphere’s surface. The large internal surface area and the small overall port fraction of 150 mm spheres allows the light to reflect properly around the sphere, creating a homogeneous flux. However, in the design of small integrating sphere accessories, sphere flux homogeneity must often be compromised in an attempt to reduce the effects of hot spots.
Hot spots are areas within the sphere which appear brighter to the sphere’s detectors than other portions of the sphere. Measurement errors, sometimes termed regular reflectance screening errors, can result from hot spots, especially when measuring glossy or specular, mirror-like samples. The reflectance of mirrors may appear higher than the true value if the sphere’s detectors are not baffled from the spot on the sphere wall where the first-strike radiation hits. As described above, spheres are designed with baffles between the detectors and each sample port to minimize such occurrences. A sphere’s baffles are intended to eliminate hot spots or to shield the sphere’s detectors from viewing them directly. Hot spots are more prevalent in small 60 mm diameter spheres. In these spheres, it is often impossible to design baffles to the proper dimensions needed to adequately shield the sphere’s detector from view of a sample port or another bright spot.
Sample Beam Size
Due to their size, 150 mm integrating sphere accessories have proportionally larger sample beam spot sizes, typically 50% larger than those of small 60 mm spheres. A large spot size is an advantage for inhomogeneous samples, where large beam coverage ensures representative reflectance measurements over the entire surface of the test sample. However, the typical large beam size of the 150 mm integrating sphere accessory is not optimized for the measurement of samples smaller than one inch in diameter. For such small samples, the sample beam must be reduced so that it does not overfill the sample. Two methods can be used to reduce the size of the sample beam so it better matches small samples; 1) either a lens can be used to focus the beam down, or 2) the sample can be masked so that the correct portion of the beam strikes the sample. Both beam reduction methods result in sometimes considerable loss in beam energy, which can increase the noise in the scan and will require slower scan times to compensate for the energy loss.
Options for 150 mm Spheres Only
While all sizes of integrating sphere accessories are able to measure diverse samples such as powders, liquids in cuvettes, and translucent or opaque solids such as fabrics or syringes, the 150 mm diameter integrating sphere has extended sampling options that are unavailable on its smaller counterpart. Integrating spheres of 150 mm diameter or greater are able to accept center mount sample holders. These center mounts enable variable angle reflectance measurements of opaque samples, absorbance scans of thin films or translucent samples,or fixed angle liquid measurements to be performed. In addition, the standard 150 mm sphere accessories have removable reflectance port covers for the measurement of large or bulky samples.
