A Real-Life FBA/OBA/FWA experiment

[Sven W]

I was curious about the condition for our beloved inkjet papers in a real-life daylight situation. What happens with the paper itself, it's whiteness. We sure know a lot about the inks fade resistance, but I have seen dramatic changes in the paperwhite with some paper types. So I just put four different types of our most common papers in my studio window, and measured them the first day in every month for seven months.
Using an EyeOne Spectro + EyeOne Share, I averaged (in Lab) ten spots on each, in every reading. And the result have surprises.

The four types are:
Cotton Rag, non-FBA
ResinCoated (RC)
Fiberbased, alpha-cellulose (purified wood)
Matte, cellulose (wood)
All, but one, has different amounts and kinds of FBA
I don't mention any brands, but they are all very wellknown on LL.

After using one of Bruce Lindblooms calculators I got the differences as deltaE-values (CIE76), for the first value and the one seven months later:

Cotton Rag, 1.8
RC, 2.5
Matte, 4.9
Fiber, 8.9

Conclusion.
The Lab-values tell us more what is happening; a very modest shift in Lightness, around 1 or lower, but some huge shifts in the b-axis, resulting in a strong yellowing. Because of rapid fading of the whitening agent (which gives us the blueish reflected light).
And the big surprise, for me anyway, is the low (good)values for the RC paper and the very high (not so good) shift for the Fiber-based paper.

I had some email conversations with Mark McCormick-Goodhart,
Director at the Aardenburg Imaging & Archives, about the results, and he could just confirm the same conclusion; There are papers out there, which you really should avoid, if the concern is archival reliability. The low value for the RC depends on that the recincoating works as a blocking surface for the UV. And that the FBA is placed in a lower layer, not on the top of the paper.
(Mark can give more precise explanations, if he pop in here)

A deltaE value of 8.9 is in fact a very strong yellow cast. In seven months.

/Sven Westerlund

[JeffKohn]

So are you going to name names and tells which OBA-enriched fiber-based paper fared so poorly? Without that I'm not sure what this really tells us except that you can't make broad generalizations and really need to consider specific paper types.

Other than than that, this just confirms my preference for OBA-free papers (all other things being equal).

[Sven W]

Well, it tells you something about what happens outside the test-labs.
And that not all FBA/OBA is harmful. Preferences or not.
And that we can demand the manufacturers to give us the right information.
So my idea was to point out that these factors are widespread and show up where you least thought or imagined.

/Sven

[Rusty]

Name Names, don't be coy. To not do so is a disservice to this community.
thanks.

[JeffKohn]

And that not all FBA/OBA is harmful. Preferences or not.

I find that an odd conclusion, considering that the OBA-free cotton rag paper did the best in your test. If you're trying to say that not all papers with OBA's will behave exactly the same, there's no big surprise there. The amount and type of OBA's, as well as whether they're in the paper base, inkjet coating, or both, will all be contributing factors. And it stands to reason that OBA's in the paper base might last longer when sandwiched between layers of plastic.

And that not all FBA/OBA is harmful. Preferences or not.

At best, you've shown that some OBA's are less harmful than others. Not exactly the same thing...

[narikin]

I for one would be interested to know what the RC papers are - standard Epson ones I imagine?
yes there are those that yellow quickly/badly and those that do so slowly/ acceptably. RC or Fiber, OBA or none.

there are OBA free papers that have very poor stability with some inksets on the Aardenburg archive, so that is not a bullet proof metric.  Anyone who cares should join and subscribe to Aardenburg.

Yes - please do name and shame!

[Sven W]

OK folks, once for all;
I decided in an early stage to not care about brands.
Otherwise, to be fair, I would have to include 50 more papers. That was not possible and not the point.
If you are interested in depth, it's very simple to take look at Aardenburg, because you can actually read all there.
I was just curious how a "hardcore" test on different paper types resulted.

And Jeff;
I can't understand why you have such a negative attitude. Why a sour reply if you already knew all the facts?

/Sven

[AFairley]

Interesting, and thanks for investigating, but not so applicable to me because everything I do is framed behind UV filtering glass.  I wonder what the result would look like repeated with a square of UV framing glass or plexi over part of each sample in the test.

[Ernst Dinkla]

Well, it tells you something about what happens outside the test-labs.
And that not all FBA/OBA is harmful. Preferences or not.
And that we can demand the manufacturers to give us the right information.
So my idea was to point out that these factors are widespread and show up where you least thought or imagined.

/Sven

Things happen outside test-labs. We have seen that with unexpected dye ink color shifts due to gas fading. Quite related to what can happen with FBAs, dye based as well.

The conditions in practice are often not that well described nor uniform and most are based on few samples to come to a general conclusion on what causes the image degradation. Which limits my desire to know the actual papers tested in this case.

It wouldn't surprise if more RC papers showed nice results on similar tests. I didn't join the Fiber/Baryta bandwagon as I liked the better "abrasion resistance" of RC papers and had some reason to expect the good effects of the polyethylene barriers encapsulation, both for colorants and FBAs. Not blocking UV light but gas (like ozon) and stabilising humidity in the paper base. I think Mark must have written that. Discussions often mention the risks of desintegration of RC barriers in time, layers cracking and losing their bond to the paper, yet we have not seen any real tests of today's RC papers on that aspect. Most of the stories go back to the 1960-1970's problems with analogue RC photo papers. Then there is the association with "plastic", a word that doesn't have a positive sound in arts and crafts, photography here included. Rag, Baryta, are seen as positive qualities, alpha-cellulose not considered equal to rag and a good whitening agent as TiO2 not even mentioned.

RC papers that contain FBA are not Nirvana either. The FBAs are brightening agents acting on UV light and by that the color constancy with different illuminations will be less than what whitening agents like TiO2 show. The last will also keep its white reflectance better in frames with normal glass. RC papers when not dried enough before framing produced the glycol/glycerine condensation on the inside of the glass. Not all RC papers have the FBA enclosed between the barriers in the paper base. One of the reasons I made three UV+ spectrometer measurements for SpectrumViz so there is a bit of information where the FBA is located.

I'm confused about the terms you used to describe the papers. The "Fiber" category is not very distinct. To me it isn't the collection of Alpha-Cellulose papers but the range of papers that started to appear just before the "Baryta" wave of inkjet papers. I think that both cover a wider paper base variety than AC. I expected that most "Fiber-Baryta" papers show less paper white shift than wood based inkjet papers, matte or not matte, as the last is a wide category with many cheap qualities. So I guess a question of terminology and the number of different samples tested.

I don't think real life means daylight for the print types discussed here in this forum. Humidity can also change considerably in window testing, from an absolute dry print surface when the sun heated up the paper to high paper humidity on rainy days and UV light still active. Makes a difference. Even with Xenon lamp tests in labs the humidity in the print surface can be much lower than expected.


met vriendelijke groeten, Ernst Dinkla

spectral plots of +100 inkjet papers:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[MHMG]

I was curious about the condition for our beloved inkjet papers in a real-life daylight situation. What happens with the paper itself, it's whiteness....
After using one of Bruce Lindblooms calculators I got the differences as deltaE-values (CIE76), for the first value and the one seven months later:

Cotton Rag, 1.8
RC, 2.5
Matte, 4.9
Fiber, 8.9

Conclusion.
The Lab-values tell us more what is happening; a very modest shift in Lightness, around 1 or lower, but some huge shifts in the b-axis, resulting in a strong yellowing. Because of rapid fading of the whitening agent (which gives us the blueish reflected light).
And the big surprise, for me anyway, is the low (good)values for the RC paper and the very high (not so good) shift for the Fiber-based paper....

A deltaE value of 8.9 is in fact a very strong yellow cast. In seven months.

/Sven Westerlund

Yikes, I feel like we've covered a lot of this ground before, and in my case, I did name names, and I was left with a sense that most of the participants in this forum and elsewhere simply don't care. They like a paper, they buy it, they assume it's going to last long enough to meet their needs because they are using a pigmented ink printer. If not, they will let the cards fall where they may.

That said, Sven's test is indeed a very real world result with logical caveats. It doesn't statistically represent the whole population of OBA-free and OBA-filled papers, nor does it speak directly to more benign interior conditions. But it is real world, and this test exercised commonly encountered conditions of humidity, gas exposure, and thermal conditions for prints that are placed front facing in windows...e.g., think about a photography studio that displays examples of its craft in the front window, usually framed with glass since the prints won't be on display year after year. Sven managed to yellow two premium inkjet papers noticeably, the worst being a premium "traditional fiber print" type, in just seven months. If the print were a high-key bridal portrait and the b* value changed by 7 or 8 units in just seven months, that print would no longer represent optimum quality. If I were the owner of the studio, I'd swap it out for a new print.

Now, based on my own real world print monitoring experience that adds dataloggers to gather actual temperature, humidity, and light levels during deployment of a print at any given location, let me share some educated guesses about Sven's environmental conditions. Sven's window test produced a temperature and humidity cycle that very likely went on a daily basis (when the sun shined) from typical room temperature and RH levels like 21 degrees centigrade (70F) and 50%RH to print surface temperatures in excess of 40C (i.e.> 100F) with concurrent print moisture content levels similar to those reached at equilibrium with environments less than 15% RH.  That's a big swing, but it's also real-world. Think about your car interior parked at night and then parked in sunshine. Prints facing forward in a commercial window display experience this behavior.  Sven's light exposure dose was somewhere in the range of about 1 to 2 megalux hours of exposure per month. His test ran seven months so we're discussing a total exposure somewhere on the order of 7 to 14 megalux hours. His paper media discoloration is consistent with AaI&A laboratory tests for some papers in the AaI&A database. The worst performers in AaI&A tests are triggering AaI&A conservation display limits due to highlight color changes and OBA burnout in the 8-15 megalux hour exposure range and many OBA containing papers reach the AaI&A conservation display rating (CDR) lower limit in the 20-40 megalux hour range. Sven's media change for the matte and fiber samples would have caused the lower CDR limit to have been exceeded but the cotton rag and RC samples aren't there yet. So, the cotton rag and RC paper samples have measurably changed, but the typical fine art collector of these print samples would not be getting concerned yet. In other words, he or she wouldn't really notice yet. Not so for the matt and Fiber samples. They exhibit definitely noticeable changes and ones to be concerned about if the piece has collector's value.

Based on reciprocity law treatment, and using the industry-standard extrapolation of 450 lux average illuminance for 12 hours per day, 7-14 megalux hours of exposure and Svens' measured paper yellowing would be projected to occur in more typical interior locations in about 3.5 to 7 years from light exposure alone. Add gas fading for uncovered prints and you can find papers beginning to go yellow much much sooner, but even under glass this yellowing would happen in less than 10 years. So, how come industry ratings don't flag these "bad apples"? Well, a delta E = 8.9, even if it is all due to the b* shift upon yellowing, is apparently not enough to trigger the industry criterion for easily noticeable print fading. This level of paper yellowing is deemed acceptable for consumer photos, and it's this consumer toleranced test that is still being applied to fine art print media ratings (except at AaI&A). In the industry's defense, one way to look at it is that we routinely buy new papers that range from bright white to warm white, and as such, they have initial b* values of -10 to about +5. That's a 15 delta E spread, so if we don't mind this media color range when new, why would we mind any specific paper if it changed anywhere within that range from -10 to +5? Ok, Industry defense over. Fine art printmaker's hat on!... because if I desired a bright cool-white paper and carefully printed my image content to suit, I wouldn't want the highlight areas of the print to turn warm white any time soon!  That's the funny thing about ratings criteria. It's a matter of personal standards for what is acceptable, and at least for me, that also depends on the application. I don't care if the 4x6 photo I tape on my refrigerator dies quickly. I do care if I just bought a fine art print worth big bucks and it noticeably loses its pristine image quality during my ownership period!

kind regards,
Mark
http://www.aardenburg-imaging.com

[JeffKohn]

and Jeff;
I can't understand why you have such a negative attitude. Why a sour reply if you already knew all the facts?

And I'm not sure why you're getting so defensive. All I did was ask for some clarification about the testing, and question some of the conclusions you seem to be drawing.

I still don't understand keeping the papers anonymous. Nobody's expecting you to test 50 different papers; but not naming the papers only makes sense if you think the ones you tested are representative of their categories. It would be far more useful to know which fiber-based paper did so poorly, or for this who are interested in an RC paper, which one did comparatively well.

[deanwork]

Mark,

Would you like to share what you have learned the last 6 months or about some of the new high-end rag media being sold by Canson? I'm seeing a white base here just as bright as Hahnemuhle but with no OBA content at all. Is this a game changer, suggesting that a year from now we may not be talking about OBAs at all in the better papers?

john

[MHMG]

Mark,

Would you like to share what you have learned the last 6 months or about some of the new high-end rag media being sold by Canson? I'm seeing a white base here just as bright as Hahnemuhle but with no OBA content at all. Is this a game changer, suggesting that a year from now we may not be talking about OBAs at all in the better papers?

john

Both Canson Platine Fiber and Baryta Photographique recently entered AaI&A light fade testing. They do indeed possess nice high L* values for media white point, and the Platine does get very close to "neutral" without any OBA.  The first test results at the 10 megalux hour interval will be posted in the AaI&A database around November 10, so there is no info in the database about them just yet. And at such an early exposure stage, we won't yet see much change unless the relatively high level of OBA in the Canson Baryta is really problematic. However, it may be incorporated in a Baryta subbing layer and/or paper core rather than in the microporous ink receptor layer. That would in all likelihood provide much better resistance to OBA burnout. I suspect the paper like Sven tested has those bountiful OBAs located right in the microporous top coat. It's chemistry, concentration, and location in the media that leads to the OBA performance variations we are documenting in our tests.  The Platine as advertised (and as measured here at AaI&A) is indeed OBA-free. But there are reasons for paper whitepoint discoloration/bleaching other than OBA burnout, so even OBA-free papers really need to be evaluated for paper white stability.

I'm optimistic that the Canson papers will do well in my tests. I sure hope so. I really like the surface texture and appearance of both of these papers.

kind regards,
Mark

[narikin]

Mark and Ernst:

thank you for your considered replies.
they are much appreciated for all of us out here trying to make some basic 'right' choices and then get back to photography.

[neil snape]

Why CIE76?

The numbers posted don't say much of anything to me. Delta E without an actual Lab value of both the measured original and time related shifts would be the only way it would mean much to me.

Also are these measurements overall on ink tank colours, reference flats with composite , or blends. Or is this only white point tests?

I think what you are posting is valuable, you need only expand a little the material tested and details.

[Ernst Dinkla]

And at such an early exposure stage, we won't yet see much change unless the relatively high level of OBA in the Canson Baryta is really problematic. However, it may be incorporated in a Baryta subbing layer and/or paper core rather than in the microporous ink receptor layer. That would in all likelihood provide much better resistance to OBA burnout. I suspect the paper like Sven tested has those bountiful OBAs located right in the microporous top coat. It's chemistry, concentration, and location in the media that leads to the OBA performance variations we are documenting in our tests.  

kind regards,
Mark

The Canson Baryta has the FBAs in the coating according to the spectral plots I made. The plots are identical to the Ilford Galerie Gold Fibre Silk plots. The FBA effect is comparable to Photorag but that paper has the FBAs throughout its mass. Harman Matt FB Mp Warmtone has a similar FBA effect but the FBAs are mainly  in the paper base. Several Baryta/Fiber papers have a much higher load of FBAs, some less than Canson Baryta.

I prefer my spectral plots to estimate (relative) FBA content. When overall white reflectance plots of the papers run parallel it is quite easy to compare the FBA brightening effects. How good that relates to FBA content/FBA quality is a guess but with Lab numbers that will not be better. I'm not sure whether Lab gives the same insight.


met vriendelijke groeten, Ernst Dinkla

spectral plots of +100 inkjet papers:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[MHMG]

I prefer my spectral plots to estimate (relative) FBA content. When overall white reflectance plots of the papers run parallel it is quite easy to compare the FBA brightening effects. How good that relates to FBA content/FBA quality is a guess but with Lab numbers that will not be better. I'm not sure whether Lab gives the same insight.


met vriendelijke groeten, Ernst Dinkla

spectral plots of +100 inkjet papers:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

To Ernst, your spectral plot library for modern media is a tremendous resource for printmakers, and I totally agree that the full spectral analysis gives some additional insight where Lab values are somewhat more of the "executive summary".  I also like your white/black backing comparisons as a useful estimate of paper opacity.  I generally only make measurements with white backing because I believe the vast majority of framed prints are backed with white matt board or white foam core not a black substrate. Anyway, both LAB and spectral plots impose a learning curve for the average photographer or printmaker and in that learning curve many people's eyes will glaze over with tears!  But for those that take the time to learn how to interpret CIELAB data and spectral plots, it's well worth it.

To All, I chose to estimate OBA content by a tried-and-true method of comparing the LAB values (especially the b* value) measured using both UV and UV filter light source. To do this one needs a spectrophotometer, typically employing a tungsten gas-filled light source that emits some UV energy and that accepts a mountable UV blocking filter in the light path so that we can illuminate and measure the paper under both UV-included and UV-excluded illumination conditions (the Xrite Isis spectrophotometer uses a dual LED variation on this theme). By doing the dual measurement and comparing the delta b* value in particular, we can gage the amount of paper fluorescence and that correlates directly in these modern times to the amount of OBA (or if you prefer, FWAs) in the media. I recently added two columns to the AaI&A database to make some of the initial media whitepoint data contained in the AaI&A reports more readily accessible and interpretable. One column is labeled "Optical Brightener" and the other "UV ∆b* influence". The Optical Brightener column is the full-blown photographer/printmaker's crib sheet! It categorizes the paper into four categories of OBA content, "No", "yes(low)", "yes(medium)", and "yes(high)". The ratings in the Optical Brightener column are fully accessible to the public and probably serve as a nice complement to Ernst's more detailed spectral analyses. You don't need to be an AaI&A member to see the OBA content information in this column.  Categories are expedient, but because they use a bin ranking method to sort materials into categories, they can be potentially misleading when products fall right on one of the boundaries for a category.  For example, one paper might be just pegged into the high category while another is right near high but gets pegged as medium. So both papers may be very close in total fluorescence but get sorted into two different categories by the go-no-go nature of category boundaries. For more thorough differentiation, AaI&A also provides a second column of data that reports the ∆b* value which is the difference in the measured b* values (the blue-yellow visual component of the paper) between the UV-included and UV-excluded measurements. This information allows one to determine if two similarly categorized products are closer or farther apart in their actual flourescence behavior. It thus delineates the categories more finely. Also, it tells you and/or your framing person whether the paper is going to visually shift in media color a lot or just a little when the choice of glazing is made (UV blocking versus standard acrylic versus glass, for example). The information in this column is partly available to the public but mostly reserved for AaI&A members.

Collectively, Ernst, myself, and others are starting to get a much better handle on the OBA content issue. Another point I'd like to make is that none of the spectral plot/LAB evaluations can completely determine the location or light and gas fade resistance of the OBAs in the media. Location of the OBAs in the media is one of the critical factors in fade resistance. I also carefully examine papers under blacklight, and this technique is easy for anyone to do. It can give additional insight into quantity and location of the OBAs.  With a magnifying loop, you can even look at the media cross-section to see OBAs glowing, for example, in the paper core versus the coatings.  Also, be aware that companies with a history in photographic emulsion making like Ilford, Harmon, Fuji, Kodak, etc. have coating equipment that can simultaneously extrude multiple layers of coatings, but to the end-user it usually seems like just one overall coating layer on the front and/or on the back of the substrate. What this means, is that these manufacturers have sophisticated options as to where they incorporate OBAs. If may look like its in the top coat, for example, but in a cross-sectional microphotograph using UV-rich illumination one could determine that the OBA content has been strategically placed in a subbing layer rather than dispersed through the contiguous multiple coats.

Lastly, I've commented on this subject before and base my conclusion on the light fade test results located in the AaI&A database along with other real-world monitoring studies now in progress. There is considerable performance overlap in media whitepoint stability between papers that use OBAS sparingly and OBA-free papers. While it's wise to avoid papers with high OBA content for a host of reasons, the media that use OBAs more sparingly and in discrete locations within the media can sometimes outperform other OBA-free papers, and some OBA-free papers can manifest other whitepoint stability issues such as light bleaching where the paper gets whiter and less warm (i.e., b* value decreasing) in visual appearance over time.  Light, heat/humidity, and gas fade testing is required to determine the relative strengths and weakness among the wide variety of media choices we have today.

I apologize for such a long post.
cheers,

Mark
http://www.aardenburg-imaging.com

[deanwork]

Mark,

I'm glad you have some of the Canson media in test now. I only started to use them after I had sent in the many samples of Hahnemuhle, etc.

But there are some tests well along with non-oba fiber gloss media like Crane Silver Rag and Hahnemuhle Photorag Baryta, both of which I like a little better than the Canson fiber gloss textures.

My main focus with the Canson papers is the group of very fine matte rag papers they are offering that are as bright as Photorag but without the oba issues. Those papers like Canson Rag Photographie, Canson Edition Etching, etc, are more what I was referring to.

I don't know what "whitening agents" they are concocting here but they sure seem to work well to me. I was a little disturbed to see how many of your tests of Hahnemuhle Photorag are showing burnout of that white point earlier than we have been lead to believe in the past, without spraying with uv coat.

For all of you out there who have NOT joined Aardenburg please help out the imaging community by doing so. It is the biggest bang for the buck in real image permanence info we could hope for.

John

[MHMG]

My main focus with the Canson papers is the group of very fine matte rag papers they are offering that are as bright as Photorag but without the oba issues. Those papers like Canson Rag Photographie, Canson Edition Etching, etc, are more what I was referring to.

I don't know what "whitening agents" they are concocting here but they sure seem to work well to me. I was a little disturbed to see how many of your tests of Hahnemuhle Photorag are showing burnout of that white point earlier than we have been lead to believe in the past, without spraying with uv coat.

John

Filter the AaI&A database for [Paper/media] [contains] "Canson" and you will find six samples already in test that fit your "very fine matte rag" paper category, including Rag photographique. More on the way in the next batch of samples that will be posted in November.

As for HN photo rag paper media stability, I actually think it's doing remarkably well because it only adds the OBA sparingly in the paper core, and the OBA influence on the whitepoint color is only about 1 point in the b* value, so when it finally burns out the paper does not become noticeably yellow. In fact, one of the very first samples I started over two years ago in test  is HN photo rag printed with a Canon 5000 and Lucia pigment ink. It achieved a Conservation display rating (i.e. little or no noticeable fade) with lower limit in excess of 100 megalux hours in test... the test is still ongoing to see how the fading rate proceeds in later stages of "easily noticeable" fade. This sample is a top contender for the "gold medal"  so far in the database although I expect some newer samples in test (perhaps some HpZ3100 samples) will eventually meet or exceed this benchmark. Nonetheless, that 100+ megalux hour CDR score means HN photo rag is an excellent choice with respect to system light fastness for Canon ipF prograf printer users. Other samples of photorag are performing similarly in terms of media whitepoint stability, but the overall conservation display ratings are poorer mainly  due to choice of inks used in those samples.

Mark

[deanwork]

You are totally right. I haven't looked at those figures for awhile. It seems all of the samples of Photorag are showing it to be doing well with all the OEM inks in regard to the white point stability. And that includes the Smooth Fine Art version made for HP.

What I did just notice, that I wasn't aware of, is that it appears that the Canson Rag Photographie is showing a quicker dmax loss compared to H. Photorag with Epson K3 when you get up into the longer exposures. Not sure about the other oem dmax yet with that media.

john





As for HN photo rag paper media stability, I actually think it's doing remarkably well because it only adds the OBA sparingly in the paper core, and the OBA influence on the whitepoint color is only about 1 point in the b* value, so when it finally burns out the paper does not become noticeably yellow.