by Ken Goldstein

Regarding the recent editorial questioning the lack of interest in the ISO cleanroom standards [“Dirty little secret,” CleanRooms, Jan. 2000, p.6] and the accompanying “Myth” column [“You're in trouble!,” p.46], here's why the publication of ISO 14644-1 might not be the biggest news story of the year 2000.

First, let me state my bias: I rather like 14644-1. I believe it's an improvement over Federal Standard 209E. And I believe that it should be adopted with all deliberate speed.

Herewith the improvements of 14644-1 over FS 209E:

1. It adopts the metric system. This means that everything is based on orders of magnitude (powers of 10). This makes arithmetic conversions easy. No problems in going to or from cubic centimeters, milliliters, liters, cubic meters, etc. Compare that with navigating between inches and feet (12 inches = 1 foot, 144 square inches = 1 square foot, 1728 cubic inches = 1 cubic foot, etc.) I know, Americans are not supposed to be keen on the metric system but for anyone having to crunch the numbers, it’s a good thing.
2. It recognizes a “standard” particle size of 0.1 micrometer compared to the 0.5 micrometer “standard” used in FS 209E. Because of shrinking geometries and increasing densities, the microelectronics people long ago shifted their attention to particles well below the 0.5 micron level.
3. At the same time, 14644-1 retains the conversion formula (slightly modified from the FS 209E version) between different particle sizes. This makes it possible to readily convert the number of 0.1-micron particles per unit volume of cleanroom air to an equivalent number of 0.2- or 0.3- or 0.5- or 1.0- or 5.0-micron particles per unit volume. And although these particle sizes are the most commonly used, the formula can be used for any particle size of interest within this range.
4. 14644-1 adds three new cleanliness levels: two cleaner than the cleanest FS 209E level and one dirtier than the dirtiest FS 209E level. The relatively dirty new standard (ISO Class 9) will be useful for those industries just beginning to exert control over the number of airborne particles in their manufacturing facilities: food and beverage, paint spray applications, consumer optics, plastic finishes, etc. And the two cleaner standards (ISO Classes 1 and 2) will be useful for the microelectronics people. For over a decade, these people have been operating “off scale” (meaning cleaner than FS 209E Class1) but having problems when communicating and defining precisely what they had achieved. 14644-1 will eliminate this ambiguity.
5. Because both FS 209E and 14644-1 are statistical procedures (and quite similar ones at that), they are susceptible to the effects of outliers. [Think of an “outlier” as any data point that is far from the average.] But 14644-1 explicity recognizes this and provides a method of dealing with these under limited circumstances. FS 209E makes no mention of the subject.
6. The new 14644-1 is the first in a series of documents (14644-2, 3, 4, 5, and the 14698 series). While it is too early to be certain, we can hope that these new standards will be useful to practitioners in the field. By contrast, FS 209E was essentially a stand-alone piece of work.

Now that I've covered the improvements of ISO 14644-1 over FS 209E, here is a brief summary of the actual differences between them:

1. There is an exponent in the formula described previously in item 3. The 14644-1 document changes the value of this exponent from 2.2 to slightly less than 2.1. While this may turn out to be useful and more accurate, it is not earth shattering in its effects.
2. The 14644 standard changes the minimum number of sample locations required, For the cleaner classes (ISO Classes 5 and cleaner or equivalently, FS 209E Classes 100 and cleaner), the number of sample points required by 14644-1 is reduced over that required by FS 209E. (I’m not sure that I am completely happy with this change. But nothing is perfect. And in the overall scheme of things, this too is no big deal. Besides, these are only minimum values. We are always free to use a larger number of sample locations if it suits our purposes.) Please note that although 14644-1 changes the number of sample locations, it does not change the sample volume(s) required at each location.
3. The two cleaner classes and one dirtier class described in item 4.

And that's just about the whole story. The new 14644-1 offers some real benefits along with only a few significant changes. Here's a table that translates between ISO 14644-1 and FS 209E:

Click here to enlarge image

null

In other words, by looking at this table and learning to automatically think of the equivalent cleanliness classes, it should be obvious that we were already very close to using the new ISO standard, even if we were not familiar with it.

Now, the controversial part: the new 14644-1 is not a big deal precisely because it is not very different from FS 209E. The similarities between the two far outweigh the differences. And to be sure, this is probably the biggest selling point of 14644-1. Anyone who understands FS 209E will recognize and understand 14644-1.

Let me go a bit further and address the changeover from FS 209E to ISO 14644-1. As pointed out elsewhere, this change has already gone into effect throughout the world with the notable exception of the US. In the US, things will take a bit longer. To begin with, private companies doing business with other private companies in the microelectronics sector are generally free to adopt 14644-1 whenever they please. It is strictly a matter of negotiation between buyer and seller. If the government gets involved on either side of the transaction, things are a bit more complicated. Note that this includes the pharmaceutical, medical device and other industries that deal with the FDA. Currently, the US government officially recognizes FS 209E in terms of defining and testing cleanroom airborne cleanliness. Before it can switch to 14644-1, it must ask for input from the affected government departments and agencies (Department of Defense, Department of Commerce, NASA, Food and Drug Administration, etc.). This means that the General Services Administration will have to go through the lengthy process of publishing the new proposed standard, asking for feedback, evaluating these comments and then (ultimately) approving this change. I am certain this will occur. But predicting when is somewhat more difficult. Government departments and agencies are not noted for their quick and nimble action. I would be surprised if this took less than a year. And I would not be surprised if it took up to two years.

Ken Goldstein, Ph.D., is a consultant on cleanrooms and high-purity systems based in Scottsdale, AZ.