Very clearly, your observation must relate to an artifact of the measurement, otherwise, please solve the world's energy problems with those extra 50% of energy you're getting... :-).

Unfortunately, I am not aware of what your spectrometer looks like (what is the optical design). Can you provide a link or some description, maybe a figure of the schematics of the setup?

How is the light for reflectance collected, and how exactly does the baseline measurement work?

I have encountered intensity calibration difficulties in the past in setups relying on intensity measurements using integrating spheres. Special caution is required here in setting up the measurement. There were also publications which gave me good hints at the time, like

ROOS, A; RIBBING, CG; BERGKVIST, M: "Anomalies in integrating sphere measurements on structured samples",  Applied Optics 27 (18), 3828 (1988)

This is pretty common and is an artifact of your calibration.

If you did no calibration whatsoever (on a Cary system it’s called a “baseline”), then you really don’t know how the displayed reflectance correlates to real reflectance of a sample - you’d have to apply some calibration yourself in the data analysis afterwards.

If you first acquire a calibration spectrum, where you tell the software that you’re about to measure a “100% baseline” for example, the software will assume that the power reflected at all wavelengths from the baseline sample is 100% reflectance.  Then, if the baseline sample actually only reflected 80% of the power, then your real sample could reflect more strongly and the software will show that your sample reflectance is >100% at those wavelengths.  It’s all relative to the calibration/baseline you’re currently using.

You can order known high-reflectivity mirrors from Newport, Edmund Optics or somewhere similar, these work well as “100% reflectance” baseline samples:

http://search.newport.com/i/1/q1/products/q2/Optical+Mirrors/x1/pageType/x2/section/q3/Broadband+Dielectric+Mirrors/x2/section/x3/chapter/nav/1

http://www.edmundoptics.com/optics/optical-mirrors/laser-mirrors/precision-broadband-laser-mirrors/1931/

Dear all,

This link might give you an idea of the setup .. 

let me tell you step by step of what I do .. 

1) "Connect" the software with the equipment

2) Auto zero and the do baseline with barium sulphate

3) Then remove the barium sulphate on the slot below and keep my sample with the deposited layer facing towards me .. 

4) Auto zero and start again .. 

http://www.shimadzu.com/an/spectro/uv/accessory/solidsample/solid.html

Just to get it right:

• you're using a dual beam instrument with integrating sphere (for diffuse reflectance measurements)
• the two beams enter the same sphere on different paths and illuminate different ports, one of them being used for the sample and one for the reference object
• your reference is a Barium Sulphate diffuse refletance standard

Things to consider in my opinion:

• Is your specimen a diffuse (Lambertian) reflector or is there a significant amount of specularly reflected light to be expected (if the answer is the latter option, which I hold probable, then do consider the article(s) by A. Roos et al. or similar/related material)
• It is not simple to do accurate reflectance measurements  with this kind of set-up. You cannot expect the sphere to behave ideally, especially small BaSO4 spheres.
• Do test measurements with interchanging positions for the refletance standard and sample and see what you get.
• Use good specularly reflecting mirrors as the samples to see what you obtain for these and make yourself an opinion of what you get. (I'd use glass slides and metallic (aluminum, silver gold) mirrors where it is clear what I should be getting.
• Use a second BSO4 standard at the sample position. What do you get? Is the response uniform?
• For the measurements to be good it is of utmost importance that the detector is well protected from direct exposure to light reflected from the sample. Are there baffles inside the sphere which guarantee this to be the case?

All this are tests to make sure you understand the optical behaviour of your spectrometer. Only then, probably, you will be able to design your experiments such that you can trust the results. 

I do believe (not 'know') that dual beam instruments with a single integrating sphere are unlikely to produce very accurate results, because of the different light paths. You can always proceed like in a single beam instrument, if you make two consecutive measurements with an appropriate standard and the sample, both positioned at the sample slot, and dividing those.

Please, keep us updated on the progress you make. 

Thanks Kai Fauth for your elaborative answer.. it enlightened me to learn a lot about this UV spectroscopy itself .. Thanks again .. ! ! 

Thanks Kandhadi Jaipal, Actually I did one without autozero after baseline and one with autozero after baseline .. 

I was thinking that the reflectance I measured without autozero was wrong .. after reading your comment, I feel that my measurement is correct .. 

Thanks .. 

I am agree with above written comments. Nevertheless I have one supplemental comment. Contrary to absorbance measurement in measurement of reflectance it is necessary to remember that working white standards such as Barium Suplhate, opal glass, Spectralon, etc. has own absolute reflectance which is not 100%. That means for example in NIR region at 1450 nm can be reflectance around 90% and at 1900 80% Based on that it is necessary to have during baseline routine inside of spefo software  absolute reflectance data of used working white standard. 

Totally agree with Michaels comment. Standards should come with a calibration/specs sheet specifying the actual value of diffuse reflectance.

My experience was that Spectralon was quite superior to BaSO4 (but that was more than 20 yrs ago...)

hello

I need to know what i am doing wrong as I am not getting a 100% reflectance for the spectrolon standard labspher. My uv-vis spectrophotometer is cary 100 varian.