I've got the blues - Cyanotype pain

[Alan Townsend]

Looks really nice and I am glad your making progress. I had severe trouble getting Ware's Simple Cyanotype working last year and never had a desire for NEW,

 

[Alan Townsend]

I just noticed you have a Pixle phone too. I use mine for negative and print scanning all the time.

 

[Raghu Kuvempunagar]

The result is a sparkly, crystal like finish when dry. The is in contrast to using the classic formula or when using the new formula on Arches Aquarelle when dry. Those tend to a matte finish.

This "sparkly" finish has a direct correlation with the properties of the final product.
-The more "sparkly" the finish, the lower the dmax and the more run-off in the wash water.
-The dmax improves with pre-humidification and the sparkly appearance diminishes.
-"Sparkly" surface papers loose much, if not most, of their denisity in plain or a citric acid first wash. Using a mineral acid such as sulfamic acid is a must. This retains the image.
-"Sparkly" surface papers appear to have much of the prussian blue sit atop the paper surface! You can feel the difference between areas of shadow and uncoated areas. The dense areas have a rough texture and tend to leave blue residue on your fingers.

Could the "crystal like finish" be due to residual Potassium Ferric Oxalate in the sensitiser which crystallises on the surface of the paper after coating? During the making of New Cyanotype sensitiser, Potassium Ferric Oxalate crystals form and are removed by filtration. However, as Mike Ware notes in his writings that this separation is not always 100%:

"Of course the separation of potassium from ammonium cations is not 100%: some K+ remains in the solution and a little NH4+ will be in the crystalline product, so the conditions of time and temperature adopted for the crystallization in the following recipe must represent a compromise."

 

[nmp]

What appears to be happening is the water in the solution is readily absorbed by the paper, but the contents tend to get deposited on the surface. The result is a sparkly, crystal like finish when dry. The is in contrast to using the classic formula or when using the new formula on Arches Aquarelle when dry. Those tend to a matte finish.

In a true solution, water can't be absorbed independently from the solute, in this case the sensitizer. All of it is absorbed or none of it. (unless there is reverse-osmosis going on which I don't think it is.)

It is fairly well known that the paper should be matte after it has been dried to get optimum result which is true even for Classic cyanotype. If it is not, it can be due to one of many reasons:

1) the viscosity of the sensitizer is too high so the as you brush it, most of it stays above the paper structure: remedy - dilute the sensitizer, moisturize the paper, add surfactant

2) there is not enough time given to the sensitizer to be absorbed into the paper before it is heat dried. Rule of thumb is to apply heat only after the paper has become matte: remedy - leave the paper in a enclosed box for a time after coating while still wet until it turns matte, only then use blow drying with or without heat

3) if there is too much sensitizer which again leaves out a film above the paper structure: remedy - use less sensitizer, brush excess off with a dry brush, don't leave any pooling of sensitizer on the paper

This "sparkly" finish has a direct correlation with the properties of the final product.
-The more "sparkly" the finish, the lower the dmax and the more run-off in the wash water.
-The dmax improves with pre-humidification and the sparkly appearance diminishes.
-"Sparkly" surface papers loose much, if not most, of their denisity in plain or a citric acid first wash. Using a mineral acid such as sulfamic acid is a must. This retains the image.
-"Sparkly" surface papers appear to have much of the prussian blue sit atop the paper surface! You can feel the difference between areas of shadow and uncoated areas. The dense areas have a rough texture and tend to leave blue residue on your fingers.

Indeed. When the sensitizer is fully dried, it will crystallize into its components. When those crystals are embedded within the paper, they won't sparkle due to scattering. When they are high up above the paper surface, they do. Essentially when sensitizer is sitting on top of the paper, the resulting Prussian blue collodial structure does not have any anchor to hold on to as it is formed so it can get washed away by physical forces during developing. Little UV is available to the sensitizer that might be within the paper fibers underneath so what remains is a much muted blue density.

Best Dmax can be obtained if the sensitizer is sitting right below or at the peaks and valleys of the paper but not too far down. The latter will also result in reduced density.

:Niranjan.

 

[koraks]

Or (4) there are compounds in the paper reacting with the sensitizer, producing insoluble compounds that collect on the surface. I'm leaning towards this as an explanation.

Best Dmax can be obtained if the sensitizer is sitting right below or at the peaks and valleys of the paper but not too far down.

Overall printing as well as coating properties are IME best if the sensitizer doesn't soak in too deeply. I never use surfactants for this reason. They are also a liability w.r.t. clearing the highlights in heavier papers as the sensitizer embeds itself very deeply inside the paper texture and then any insoluble compounds formed there are difficult to remove through washing.

 

[NedL]

Could the "crystal like finish" be due to residual Potassium Ferric Oxalate in the sensitiser

That was my first thought also. But I'm not sure what's going on here

 

[cirwin2010]

Could the "crystal like finish" be due to residual Potassium Ferric Oxalate in the sensitiser which crystallises on the surface of the paper after coating? During the making of New Cyanotype sensitiser, Potassium Ferric Oxalate crystals form and are removed by filtration. However, as Mike Ware notes in his writings that this separation is not always 100%:

"Of course the separation of potassium from ammonium cations is not 100%: some K+ remains in the solution and a little NH4+ will be in the crystalline product, so the conditions of time and temperature adopted for the crystallization in the following recipe must represent a compromise."

Re-reading through Christina Anderson's book, she has brief section about the crystals under the troubleshooting chapter. Her possible explanation is the crystals are from ferric ammonium citrate that hasn't been fully absorbed by the paper.

In a true solution, water can't be absorbed independently from the solute, in this case the sensitizer. All of it is absorbed or none of it. (unless there is reverse-osmosis going on which I don't think it is.)

It is fairly well known that the paper should be matte after it has been dried to get optimum result which is true even for Classic cyanotype. If it is not, it can be due to one of many reasons:

1) the viscosity of the sensitizer is too high so the as you brush it, most of it stays above the paper structure: remedy - dilute the sensitizer, moisturize the paper, add surfactant

2) there is not enough time given to the sensitizer to be absorbed into the paper before it is heat dried. Rule of thumb is to apply heat only after the paper has become matte: remedy - leave the paper in a enclosed box for a time after coating while still wet until it turns matte, only then use blow drying with or without heat

3) if there is too much sensitizer which again leaves out a film above the paper structure: remedy - use less sensitizer, brush excess off with a dry brush, don't leave any pooling of sensitizer on the paper




Indeed. When the sensitizer is fully dried, it will crystallize into its components. When those crystals are embedded within the paper, they won't sparkle due to scattering. When they are high up above the paper surface, they do. Essentially when sensitizer is sitting on top of the paper, the resulting Prussian blue collodial structure does not have any anchor to hold on to as it is formed so it can get washed away by physical forces during developing. Little UV is available to the sensitizer that might be within the paper fibers underneath so what remains is a much muted blue density.

Best Dmax can be obtained if the sensitizer is sitting right below or at the peaks and valleys of the paper but not too far down. The latter will also result in reduced density.

:Niranjan.

I didn't take a picture of the paper prior to exposure, but that edge bleeding really looked like the water was pulling away from the ring of crystal deposits around the edge. Very well could be another explanation such as @koraks.

I have tried diluting the sensitizer, moisturizing the paper, and adding tween20. Admittedly I only used moisturizing/pre-humidification and tween20 in combination. I did not combine diluting the sensitizer with these other variables. As far as per-humidification goes, I'm leaving the paper on some screen above a tray of water. Another try is placed atop and left for several hours before coating. I use this same contraption for post-coating humidification for my kallitypes, but doing it after coating appears to invite paper fog for new cyanotype.

I ran into the same issues with splotches using more or less sensitizer on HPR. I think HPR is just difficult to coat using the new formula.

As long as I can get HPR to continue to coat well using the classic cyanotype formula I still think I'm going down that path. Exposure time vs the new formulation is not that much longer and the dmax is almost as good. Though the color is quiet different using a 20% FAC solution. Very turquoise by comparison.

 

[Raghu Kuvempunagar]

Re-reading through Christina Anderson's book, she has brief section about the crystals under the troubleshooting chapter. Her possible explanation is the crystals are from ferric ammonium citrate that hasn't been fully absorbed by the paper.

But New Cyanotype doesn't have Ferric Ammonium Citrate, her explanation could be for Classic Cyanotype and not for New Cyanotype (Mike Ware does mention that traditional FAC based sensitiser is absorbed poorly: "The traditional sensitizer is poorly absorbed by cellulose paper fibres and some tends to remain on the surface or in the pores." He also says FAO based sensitiser is better in this respect: "Its solution penetrates the fibres of cellulose paper much more readily than the citrate, and it mordants better onto fabrics. It is nondeliquescent. Image substance is better retained.")

I ran into the same issues with splotches using more or less sensitizer on HPR. I think HPR is just difficult to coat using the new formula.

Have you tried giving HPR a rinse in plain water before using it as @Andrew O'Neill does? Or a rinse in mildly acidic water to remove any problematic additive?

Quoting Mike Ware:

"It is possible that this ‘decalcification’ treatment also brings an added hidden benefit: the small anions such as chloride or sulphamate from the acid will tend to neutralise the cationic polyelectrolytes that are present as retention aids"

 

[cirwin2010]

But New Cyanotype doesn't have Ferric Ammonium Citrate, her explanation could be for Classic Cyanotype and not for New Cyanotype (Mike Ware does mention that traditional FAC based sensitiser is absorbed poorly: "The traditional sensitizer is poorly absorbed by cellulose paper fibres and some tends to remain on the surface or in the pores." He also says FAO based sensitiser is better in this respect: "Its solution penetrates the fibres of cellulose paper much more readily than the citrate, and it mordants better onto fabrics. It is nondeliquescent. Image substance is better retained.")




Have you tried giving HPR a rinse in plain water before using it as @Andrew O'Neill does? Or a rinse in mildly acidic water to remove any problematic additive?

Quoting Mike Ware:

"It is possible that this ‘decalcification’ treatment also brings an added hidden benefit: the small anions such as chloride or sulphamate from the acid will tend to neutralise the cationic polyelectrolytes that are present as retention aids"

Ooops, meant to type ferric ammonium oxalate. I edited the original comment to correct that. Sorry for the confusion.

I have not given HPR a rinse. If I was to rinse it in plain water, do I fully dry it before coating? Should the paper be pre-humidified again before coating?

 

[Raghu Kuvempunagar]

Ooops, meant to type ferric ammonium oxalate. I edited the original comment to correct that. Sorry for the confusion.

"Dr. Ware in an email conversation said that if the coating is allowed to stand on the paper without being promptly absorbed, it evaporates and can leave these crystals (possibly of potassium ferric oxalate or potassium ferricyanide)."

This is what @NedL and I suspected. The chances of Potassium Ferric Oxalate forming crystals in the sensitised paper would be higher if during the preparation of the sensitiser, Potassium Ferric Oxalate didn't precipitate fully (green crystals) and some of it remained in the solution.

 

[nmp]

"Dr. Ware in an email conversation said that if the coating is allowed to stand on the paper without being promptly absorbed, it evaporates and can leave these crystals (possibly of potassium ferric oxalate or potassium ferricyanide)."

This is what @NedL and I suspected. The chances of Potassium Ferric Oxalate forming crystals in the sensitised paper would be higher if during the preparation of the sensitiser, Potassium Ferric Oxalate didn't precipitate fully (green crystals) and some of it remained in the solution.

Given that PFO has low but finite solubility in water, there will be some left over in the sensitizer - more or less depending on the exact concentration in the liquid and the temperature of recrystallization (this might be a source of batch to batch variation if not controlled precisely.) Lower temperature will remove more of it but so will be some of the other ingredients, throwing off the ratios. So ultimately, the solution to the problem is not to let the coating evaporate on the top of the paper without being absorbed by the pores of the paper first, as suggested by Dr. Ware.

:Niranjan.

 

[cirwin2010]

Success! I got something I think I'm mostly happy with.

Classic cyanotype on HPR
20% FAC
10% PF
no tween20
paper pre-humidified before coating

The result of the above gives me something that is easy to coat and coats repeatably well. Dmax is good and the exposure time is short. The attached image is of an un-calibrated test chart exposed for 6.5 minutes. I think somewhere between 5-6 minutes is better. Oddly this is shorter than the exposure time I was getting for the new cyanotype formula which is supposed to be more light sensitive?
The resulting color is certainly different than the new cyanotype formula. This is much more green blue where the new formula gives a more purple/grey blue result. Subjective, but I wish this was slightly less turquoise with HPR. New cyanotype also results in better dmax, but as previously noted, much harder to coat well.

My only real complaint with this combo is some mid-tone grain/unevenness. It might be hard to make out in the attached image, but it really is only visible in the midtones. This is present on HPR regardless of the inclusion of tween20 or if I use 10 or 20% FAC. I'm assuming this is some unevenness in the absorption of the solution in the paper fibers. New cyanotype does not have this grain.



I ran into some issues using tween20 with this combination. When using 1 drop of 10% tween20 per ml, overexposed shadows would start looking permanently solarized and gross. This problem gets worse with more overexposure. The second image depicts this issue. Also note that the white grid lines are starting to disappear. Also note that the edges of the squares adjacent to the white border have a bit of an edge effect. This issue is reduced with less exposure and eliminated with the exclusion of tween20.

My guess is that the tween20 is causing the solution to absorb too far into the paper. Light is then penetrating into this deeper layer, exposing it and changing how light interacts with the surface causing an apparent loss of density and change in color. Additionally I think there might be some sort of light piping effect happening here where light is penetrating and diffusing into the paper beyond its surface. This then exposes some of the solution that resides under the paper surface, even if it is being masked by the negative. This would explain some of the edge effects and why the grid between the squares appears to be getting encroached upon.




I also tried combining new cyanotype with classic cyanotype (10% FAC) 1:1. The results were in between classic and new formulas. The color was still a bit green-blue, but less so than classic on its own. Dmax is maybe slightly better than classic 20% FAC. There is no grain. Coats better than new cyanotype, but still has issues. Some uneveness in the coat and random light spots with jagged edges. Did not have the above described tween20 issue. No edge bleed. May work well for a different paper, but additional complexity seems hard to justify.

 

[koraks]

some mid-tone grain/unevenness

Yeah, papers differ in this regard.

Light is then penetrating into this deeper layer, exposing it and changing how light interacts with the surface causing an apparent loss of density and change in color. Additionally I think there might be some sort of light piping effect happening here where light is penetrating and diffusing into the paper beyond its surface. This then exposes some of the solution that resides under the paper surface, even if it is being masked by the negative. This would explain some of the edge effects and why the grid between the squares appears to be getting encroached upon.

Yeah, that's a decent explanation, although the term 'edge effects' is potentially confusing due to their different meaning in other fields of photography (cf. Mackie lines).
Collimation of the light sources plays a role here as well, as does (over)exposure. Your chart seems to hit close to dmax way before the last step suggesting you could reduce exposure significantly and still get virtually the same density, but better tonality. This will also keep your adjustment curve more sane/less extreme.

I also tried combining new cyanotype with classic cyanotype (10% FAC) 1:1. [...]y work well for a different paper, but additional complexity seems hard to justify.

I'd call that approach "the worst of both worlds" and just not bother.

 

[Raghu Kuvempunagar]

I'd call that approach "the worst of both worlds" and just not bother.

Several printmakers seem to have had more success with this approach than with New Cyanotype. It is discussed in Christina's book and there are some very nice prints in the same book by Sam Wang made using this approach. Peter Mrhar has high praise for Dr. Wall's Cyanotype recipe which can also be seen as a combination of New Cyanotype and Classic Cyanotype.

 

[nmp]

Several printmakers seem to have had more success with this approach than with New Cyanotype. It is discussed in Christina's book and there are some very nice prints in the same book by Sam Wang made using this approach. Peter Mrhar has high praise for Dr. Wall's Cyanotype recipe which can also be seen as a combination of New Cyanotype and Classic Cyanotype.

Mike Ware has some reservations (to put it mildly) about this.

:Niranjan.

 

[Raghu Kuvempunagar]

from https://archive.org/details/photographicfact00wall/page/258/mode/2up

The use of sodium or ammonium ferric oxalate gives greater speed than the above, but the paper does not keep so well. The sensitiser should then be:

Potassium ferricyanide 133 g
Ferric ammonium citrate (red) 166 g
Sodium or ammonium ferric oxalate 33 g
Water 1000 ccm

All these solutions are sensitive to light. Other iron salts may also be used, for instance:

A. Sodium or ammonium ferric oxalate 250 g
Water 1000 ccm
B. Potassium ferricyanide 250 g
Water - 1000 ccm

Filter both solutions, mix, and again filter.

First of these is the one Peter Mthar called a combination of New and Classic Cyanotypes. According to him "this formula stands out for its remarkable tonal range and intense deep blue hues. It also handles whites beautifully, making it a versatile choice for creating striking contrasts and intricate details."

Second is similar to New Cyanotype.

 

[cirwin2010]

Success! Calibrated my profile and made a test print. Not meant to be an artistic masterpieces, just wanted something that was tonally challenging to ensure what I am reading on my test chart matches what I expect in a print. Some fine tuning may be needed as the deep shadows are a little muddy, but overall looks good.

This was classic cyanotype on HPR. 20% FAC, pre-humidified before coating, and no tween 20. Exposure was 4.5 minutes. 0.25% citric acid first bath.

My results are a bit at odds with what is expected per Christina Anderson and Mike Ware. With the above combination, my exposure time is less than new cyanotype. Granted dmax is not as high as new cyanotype, I get to dmax faster. I also find that a higher concentration of 20% FAC to expose faster than 10% FAC. This is the conventional wisdom, but C. Anderson claims that she go faster results with 10% FAC. I wonder what variables are contributing to this.


In any case, the results I'm getting with the listed combo are good and ready for use.

I am going to go back to trying to get the optimal combination with Canson XL Watercolor paper next. It is cheap, easy to coat, and I think I might be able to tweak it to get something I like. 20% FAC is already an improvement over what I was getting with 10% FAC in terms of color, dmax, and exposure time. It is very grainy when coating a dry piece of paper. Pre-humidifying seems to help, but adding tween20 reduces dmax and creates that sort of edge bloom effect I described easier. Solution penetrating too deep into the paper. Going to try acidifying it first too and see what that does.

I also will probably return to arches aquarelle cold press using the classic solution. That didn't seem to have as many problems, but exposure time was too long. Might be better with 20% FAC.

Also got new Richson 9010 brushes which are a vast improvement over what I was using. I can coat using less solution and get much more even strokes. Maybe this will help with coating new cyanotype?

Oh, and can't forget about trying double coating.

I have all this paper and chemistry, I'm going to try to figure out what I like best. It's cold outside here in the northern hemisphere, what else am I going to do?

 

[koraks]

Very nice print indeed! Doesn't matter in the end how you get there, as long as it works in your hands.

 

[Raghu Kuvempunagar]

My results are a bit at odds with what is expected per Christina Anderson and Mike Ware. With the above combination, my exposure time is less than new cyanotype. Granted dmax is not as high as new cyanotype, I get to dmax faster. I also find that a higher concentration of 20% FAC to expose faster than 10% FAC. This is the conventional wisdom, but C. Anderson claims that she go faster results with 10% FAC. I wonder what variables are contributing to this.

This is indeed quite intriguing and if you find out the reasons for the unexpected results in the future, please do share with us. In my limited experience with New Cyanotype, I have found it to be 2-3 times faster than Classic Cyanotype. Having said that, I make the sensitiser fresh everytime and I figured out a way to avoid crystallisation. In general, Cyanotype recipes with some Oxalate/Oxalic Acid in them are much faster than Classic Cyanotype. So your observations are quite interesting.

Very nice print indeed! Doesn't matter in the end how you get there, as long as it works in your hands.

+1

 

[cirwin2010]

This is indeed quite intriguing and if you find out the reasons for the unexpected results in the future, please do share with us. In my limited experience with New Cyanotype, I have found it to be 2-3 times faster than Classic Cyanotype. Having said that, I make the sensitiser fresh everytime and I figured out a way to avoid crystallisation. In general, Cyanotype recipes with some Oxalate/Oxalic Acid in them are much faster than Classic Cyanotype. So your observations are quite interesting.




+1

I will definitely share if I ever find out! Literature doesn't seem to suggest that dichromate impacts sensitivity, but I'm wondering if my omission of the ingredient has anything to do with it. It is used as a sensitizer in carbon printing after all. Perhaps it is absorbing some of that energy and transferring it to neighboring molecules? Someone with a better understanding of chemistry will probably correct me if I'm way off base. In any case, I don't plan to test this anytime soon as I don't really want to mess with dichromate.

Do you use dichromate in the new cyanotype formula?

I don't have a blog or anything like that so I've been using this thread as a way to document things and maybe aid anyone having similar troubles. I'll probably make some sort of blog or online publication at some point for kallitypes as I've found a number of issues with various resources and I want to put better information out there. But given my struggles with cyanotypes, I'll probably leave that to someone else.

 

[Raghu Kuvempunagar]

I will definitely share if I ever find out! Literature doesn't seem to suggest that dichromate impacts sensitivity, but I'm wondering if my omission of the ingredient has anything to do with it. It is used as a sensitizer in carbon printing after all. Perhaps it is absorbing some of that energy and transferring it to neighboring molecules?

I don't think adding Dichromate increases the speed of Cyanotype. Its primary purpose in New Cyanotype is to prevent dark reaction in the single-bottle solution during storage.

 

[nmp]

I don't think adding Dichromate increases the speed of Cyanotype. Its primary purpose in New Cyanotype is to prevent dark reaction in the single-bottle solution during storage.

If anything, it should make it slower as it acts as an oxidizer here and it will quench a certain amount of photoinduced Fe(2) in the beginning of the exposure until it itself is all depleted, pushing the exposure curve to the right.

It is photosensitive primarily in presence of colloids such as gelatin etc which is not the case in the New cyanotype.

:Niranjan.

 

[cirwin2010]

I don't think adding Dichromate increases the speed of Cyanotype. Its primary purpose in New Cyanotype is to prevent dark reaction in the single-bottle solution during storage.

If anything, it should make it slower as it acts as an oxidizer here and it will quench a certain amount of photoinduced Fe(2) in the beginning of the exposure until it itself is all depleted, pushing the exposure curve to the right.

It is photosensitive primarily in presence of colloids such as gelatin etc which is not the case in the New cyanotype.

:Niranjan.

Makes sense. Thanks for the clarification!

 

[cirwin2010]

Got another winner! I'll post some images of my results later once I'm done with more calibration.

Canson XL Watercolor paper is awesome. It's cheaper and easy to coat right off the pad with classic cyanotype. It yields a somewhat muted blue-gray color with similar dmax to classic cyanotype on HPR. It has a distinct "blotchy" grain which appears to be from variations in the papers structure. But, very acceptable.

This paper completely transforms when you pre-acidify it in sulfamic acid. Classic cyanotype has greatly improved dmax and the color changes from that muted gray-blue to a much more striking cyan. Double coating it improves the dmax even more and reduces the grain somewhat. It looks very stunning, especially compared to classic cyanotype on HPR. There is not contest when it comes to dmax.

The paper also becomes compatable with the new cyanotype formula once acidified. It coats well (so far), free of much of the issues I experiences with other papers, but, it must be pre-humidified before coating. The dmax is impressive, somewhere in between the single and double coated classic formula. There is very little grain compared to classic giving a much smoother appearance. The color is also a nice royal blue color. Double coating does not work and worsened the image with lots of washout and grain.

I'm not sure which of the two combinations I'm going to utilize going forward, but I'll know once I complete my ink profiles and make some images to compare.

 

[cirwin2010]

I've made three prints of the same image with the following recipes:

Left:
-Hahnemuhle Platinum Rag
-Classic Cyanotype (CC) 20% FAC
-Paper pre-humidified before coating
-single layer coating
-exposed 4.5m
-0.25% citric acid first wash

Middle:
-Canson XL Watercolor
-Classic Cyanotype (CC) 20% FAC
-Paper coated dry
-Paper acidified in 10% sulfamic acid
-double layer coating
-exposed 6m
-0.20% sulfamic acid first wash

Right:
-Canson XL Watercolor
-Mike Ware's New Cyanotype (NC)
-Paper acidified in 10% sulfamic acid
-Paper pre-humidified before coating
-single layer coating
-exposed 4m
-1.00% sulfamic acid first wash

I feel like this is the culmination of all my efforts. Canson XL Watercolor after acidification has given me results I am very happy with. As you can see from the attached images, I get better dmax and smoothness with both combos using the Canson paper.

The double coated CC combo definitely has more dmax giving more visual punch to the image. Though it does have some "grain" that is visible in the midtones. This appears to be some slight unevenness to how the solution absorbs into the paper.

The NC combo has improved dmax over the HPR CC combo, but not as much as the double coated CC Canson combo. It does have much better smoothness than using CC and the royal blue color is beautiful. Not sure which I like better.

Oh and for fun I attached an image of CC HPR toned in instant coffee for a few hours.