Quantitative analysis of granular pha... Log Out | Topics | Search
Moderators | Register | Edit Profile

NIR Discussion Forum » Bruce Campbell's List » I need help » Quantitative analysis of granular pharmaceutical products « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

Sean Cresswell (Sc19478)
Posted on Wednesday, September 10, 2003 - 10:24 am:   

I will soon be performing a feasibility study to determine if NIR can be used as an alternative to the KF and HPLC methods currently used for moisture and active content repectively.

I have an FT-NIR system with reflectance and transmission capability.

My problem is that the product is a spray granule formulation with some heterogeneity with regard to active content. I need to transfer and present a representative sample of the product to the NIR instrument. I then need to ensure that micro-heterogeneity does not influence the calibration models.

The product contains a wax componant to protect the hygroscopic active from moisture pick-up so grinding is an issue, although freeze grinding may be an option.

Does anybody have experience, or know of papers that have dealt with similar issues.
Top of pagePrevious messageNext messageBottom of page Link to this message

MPDC
Posted on Saturday, September 13, 2003 - 10:37 am:   

Heterogeneity can be handled by averaging out a large enough area. Think sample rotation and long scans. Your biggest problem will be segregation. Anything that vibrates your sample will bring smaller particles on top.
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Saturday, September 13, 2003 - 12:02 pm:   

Sean - much of your question is not an NIR question as such, but simply a question about good chemical practice: how to obtain a representative sample for analysis. This is a very old question, and there are many devices commercially available to help solve that problem: devices for sampling a large quantity of material in the first place, devices for reducing the amount without changing the relative compositions, devices for obtaining representative aliquots, etc. Which devices are appropriate will depend on the initial quantity of material and the particle size of the samples, but a good place to start looking is in the catalogs of some of the suppliers of chemical equipment. If the particle size is already so small that you're worrying about micro-particles, then I suspect that grinding isn't needed, just proper sampling and subsampling equipment.

You didn't say whether your application is off-line although commentary indicates that it is. In this case, you should be able to get answers to most of that part of the question related to the NIR measurement from the manufacturer of your instrument.

Howard

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Kathryn Lee
Posted on Monday, September 15, 2003 - 6:14 am:   

It is counterintuitive, but larger particles can go to the top if you vibrate a heterogeneous sample. The segregation is based on mass and size.
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Monday, September 15, 2003 - 6:59 am:   

It may be counterintuitive, but it's not uncommon: it's the same reason rocks pop up on your lawn in the spring. If a small rock (or clump of soil) is on top of a large one it can fall off, but if it gets under the large rock it's trapped, and the large rock can't fall back into place. That's an oversimplification, but describes the basic maechanism.

Howard

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Bruce H. Campbell (Campclan)
Posted on Monday, September 15, 2003 - 8:25 am:   

The particle size question and the distribution reminded me of a question, which is, "Would you rather have a barrel of dimes or a barrel of half dollars?" The answer is the barrel of dimes is more valuable, even though the monetary ratio is 1:5. The reason is the smaller particles pack more densely. This means the bulk density is higher for the dimes, i.e., the smaller particles. The heavier the bulk density, the lower the tier is, with respect to other tiers.
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Monday, September 15, 2003 - 8:36 am:   

Interesting. But I wonder how does it come out on a strict weight basis? I've got some dimes I could weigh, but I haven't even SEEN a half-dollar in years.

Howard

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Monday, September 15, 2003 - 8:46 am:   

My own message got me to wondering, so I weighed some dimes and quarters I happened to have in my pocket. I weighed 4 together of each; here's the data:

Four dimes = 9.2 grams
Four quarters = 22.6 grams

The ratio is 2.3, so I guess quarters are intrinsically slightly more valuable than dimes. So going back to the original comparison, it makes you wonder if the difference in the packing of dimes vs quarters would be enough to offset the lower value and give the same result as for the half-dollars.

Howard

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

djdahm
Posted on Monday, September 15, 2003 - 10:37 am:   

As I recall, back when U.S. coins contained silver, there was almost a perfect straight-line relationship between the weight and value of the dime, quarter, and half dollar.
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Monday, September 15, 2003 - 11:56 am:   

Well, Don, that makes sense, since originally the value of coins were set by the ACTUAL value of the precious metal they contained. Go back far enough, and you find that coins were invented so that you could know the value of metal they contained without having to weigh it out each time - you would rely on the emperor's specifications for the coin.

But in more recent times, the intrinsic value of the coinage was set to a fraction of the face value (or actually, vice versa, I suppose); this was to keep people from melting it down for the metal, to conserve the limited amount of precious metal available, and also to prevent runaway inflation as the metal became more valuable. But it was still pegged to the intrisic value. Those of us old enough to remember "Silver Certificates" know that it was also true for paper money; in theory you could go to a Federal Reserve Bank and demand a quantity of silver for a Silver Certificate, although I don't know what would happen if you actually did that. Probably you'd get a quantity of silver equal in value to the intrinsic value of the certificate, which, of course, was very much less than the face value (since it was only a piece of paper, after all). So there was considerable incentive to NOT do that!

Now, with both coins and paper money representing the "good faith" of the government, there is no tie at all to the intrinsic value of the money object, so there needn't be a relationship to the actual weight. The only thing I can think of that would cause the relationship to be retained is the existence of automatic machines to collect and exchange money, and that have to recognize both the old silver coins and the new sandwich coins as being the same. But we see that there is already an easily measurable difference in the relationship between dimes and their content (assuming they were silver) and quarters and their content.

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

MPDC
Posted on Tuesday, September 16, 2003 - 2:50 am:   

Hmmm, that should teach me not to post replies when suffering from a hangover. But to bring the debate back to NIR, can Sean also play with wavelength ranges in order to solve his problem? If certain ranges penetrate deeper, can they be considered to have covered more sample and therefore be more "averaged out"?
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark
Posted on Sunday, October 05, 2003 - 10:08 am:   

FYI - Speaking of the weights (or masses) of coins, as we did during our recent discussion on this thread, during my recent move I came across some half-dollars and even some "silver" dollars. So, thinking of the recent discussion involving them, I found my balance first thing and weighed them, along with some coins as I happened to have, of other denominations. Here are the results (the weights are the totals for the specified number of coins):

10 US pennies: 25.8 g
1 Canadian penny: 3.2 g
4 Nickels: 20.0 g
6 Dimes: 13.6 g
3 Quarters: 16.9 g
3 Half-dollars: 33.9 g
3 Silver dollars: 68.2 g

Howard

\o/
/_\

Add Your Message Here
Posting is currently disabled in this topic. Contact your discussion moderator for more information.