An alternative look at ink wetness

[InesF]

You have already experienced that some inks are wet while others are dry and, furthermore, you know about the existence of wet and dry nibs. If you write or draw a line with your fountain pen, you see either a wet trail left behind – or not. So far the facts.

At this point I asked myself: why is it so?

My two friends, Google and DuckDuckGo, both helped me to find hypothesis, theories and explanations in blogs, videos and any form of review – some more, some less convincing. But, to my surprise, no hard facts and no evidence given. The outcome of my recherché didn‘t satisfy me.

During the last three month I tried to figure out how I can measure ink wetness. After some initial fails, I found a way! Heureka! And now I like to present an alternative view at ink wetness.

 

The setup

I measured surface tension, dynamic viscosity and ink consumption of a couple of inks from my collection. Some of them are widely considered to be wet inks, others are classified as dry inks and for some no general opinion can be found.

When submitting this text to the forum I have measured surface tension of 18 inks, viscosity of 13 inks and had made the ink consumption measurement with 2 inks with 3 different fountain pens and one ink with 2 pens, each on three different papers (which is 24 measurements total). This data set will grow over time. As some of the measurements are extremely time consuming (see below in the methods section), the data set will grow slowly. Sorry for that.

The inks were mainly from Pelikan, Waterman, deAtramentis, Rohrer & Klingner, plus one Iroshizuko and one from Diamine. Some of my inks were used as they were delivered, others were already modified long before the start of my test series by addition of Gum Arabic (in the range of 0.3 to 1.5%) to reduce their feathering, spreading and bleeding when used in ’wet‘ fountain pens. I did this empirically, based on my experience with calligraphy inks and with watercolour in dip pens, I didn‘t have any evidence about which ink properties will change with Gum Arabic.

 

The data

This is the collective data table of all measurements so far (more to come):

 

This is how a typical ink consumption measurement looks:

 

Currently, I can‘t measure line width with sufficient precision. Only a more general, nevertheless, consistent classification can be done over all currently tested inks and all pens: ink friendly paper resulted in the narrowest lines that appeared to be wet during drawing, the low quality paper resulted in wider lines that appeared dry but fuzzy and the glossy photo paper resulted in the sharpest and broadest lines with intense colour but did not provide the feeling of being wet.

 

The Interpretation

All of the tested inks belong to one of two groups of typical surface tension: either 60-70 mN/m (water is 73 mN/m) or 40-50 mN/m with the deAtramentis series the Iroshizuko and the Diamine being the inks with the lowest and Waterman, Pelikan and Rohrer & Klingner being among those with the highest surface tension. The iron gall ink Scabiosa is the only exception with low surface tension (and with alcohol smell).

The dynamic viscosities of all inks were highly similar, all in the range of 1.02 to 1.08 mPa*s (water is 1.00 mPa*s) with no clear pattern per brand or per colour type (not all colours measured yet - black, green and brown inks are missing!). Inks with added Gum Arabic showed viscosities of 1.1 or 1.2 mPa*s with one exception: R&K Magenta with 0.5% Gum Arabic and a viscosity in the ’normal‘ range of 1.045 mPa*s. Gum Arabic increases viscosity and, when writing on ink friendly paper, it increases the subjective impression of wetness. When writing on low quality absorbent paper, G.A. does reduce or inhibit feathering, spreading and bleeding but does not change the wet (or dry) appearance when writing.

Remark to Gum Arabic: while a small addition can significantly improve the behaviour of an ink on low quality paper and can improve wetness and shading on ink friendly paper a bit, adding too much can result in ink starvation! Be careful, start slow, test a lot!

The deepest insights and the biggest surprises came from the ink consumption results. The pens were able to deliver around 14 g ink per km line on photo paper while only 4-5 g of the same ink was delivered on high quality ink friendly paper. As a consequence – and most important fact – no pen was in ink starvation (limited ink delivery) when writing on high quality or low quality paper. The lines on the good paper appeared generally wet (with 4 g/km), the lines on the photo paper (14 g/km) and on the absorbent paper (5-7 g/km) appeared more dry!

The setup of the ink spreading test was not optimal. There was next to no difference in the diameter of the spreading circles (or ellipses) – the difference was more in the time needed until the spreading stopped. Some of the inks reached their final diameter within an hour, others moved over night. The time was not measured and I don‘t know how to measure it without watching the spots for hours. However, the so called ’friendly‘ inks showed more smooth while the ’runny‘ inks showed more rough edges at the diffusion fronts. Only the Rohrer & Klingner SketchInk (water prove) and the iron gall ink showed significantly smaller spreading circles and the most highly Gum Arabic modified R&K Alt-Goldgrün showed a bit reduced spreading.

 

The ad hoc conclusions are:

• Different nibs of similar point size lay down similar amounts of the same ink on the same paper.
• The width of an ink line is inversely proportional to its surface tension (drawn with the same nib).
• Feathering and ink spreading seems to be mildly dependent on surface tension for those inks within the ’normal‘ range of viscosity. Higher viscosity reduces this influence on low quality paper and results in ’sharper‘ lines – maybe because the spreading speed is slower
• The surface tension of an ink has bigger influence on wetness than the viscosity – although, the correlation is not strict.
• Wetness is not a consequence of limited nor of un-limited ink flow!
• The paper type has bigger influence on ink flow than any of the ink properties!

So far the hard facts.

 

Outlook & open points

This first set of results is a start. In about 6-8 month I will have collected enough data to start a principal component analysis (multivariate statistics) to find the exact dependencies of ink delivery and ink properties. Please be patient and stay with me while I do more measurements.

Furthermore, I‘m looking for your help:

• I have drawn some model sketches of ink surface and flow patterns – maybe someone can help me with refined drawings based on computer simulations? I had to correct my drawings, sorry, I will post them later!
• I have currently no possibility to measure the line width with a precision better than 0.1 mm. A resolution of ca. 0.01 mm (or better) would be needed to calculate ink consumption per surface area! Do you have any idea?
• Viscosity plays next to no role for the wet appearance of an ink, but counters feathering and spreading of low surface tension inks. Looking back to historic ink making, Gum Arabic, Glycerol and Glycol, sometimes Sugar or Ethyl Alcohol (wine) were the traditional ingredients of choice to customize an ink. As far as I know, none of these shows direct chemical interaction that can explain non-linear behaviour of ink on paper that is not reflected in surface tension and viscosity. What is your idea, what is missing here?

[InesF]

Material and Methods (mad scientist section)

 

The three papers were: a no-name low-quality notepad (as you get for free at meetings and conferences) which behaves absorbent and ink-unfriendly, a Kyome letter pad of 70 g/m² which is very ink friendly (and behaves similar to 68 g/m²Tomoe River paper) and a hp inkjet glossy photo paper (which makes crazy ink lines and shows the true colour of any fountain pen ink – you should try it!).

 

The three fountain pens were: a Waterman Perspective M (considered to be a dry pen), a Pelikan M600 F (considered to be a wet pen) and a Cross Botanica M (which is a firehose). The pens were selected to represent at least one widely classified ’dry‘ and one ’wet‘ writer, they were previously tested to show no signs of ink starvation nor nib drying under normal use and, somehow, have similar line widths with the same ink. The consumption measurement is the most time consuming. As the both pens Pelikan M600 and Cross Botanica showed very similar behaviour, I will skip the Pelikan and continue with only the Waterman and the Cross for all upcoming tests.

 

Surface tension was measured by the drop size method. 10 drops of ink were released from a graduated glass pipette and the volume of those 10 drops was measured. The accuracy of the volume measurement is 2 µL absolute (equal to 0.2 µL per drop), the deviation between repetitions is ca. +/- 3%. Air temperature was not controlled but was measured to be in the range between 22 and 23°C. Each measurement was done in either 3 or 4 replicates, outliers were eliminated. The system was calibrated with water (42.0 µL per drop; 72,8 mN/m) and ethyl alcohol (16.5 µL per drop; 22.6 mN/m). The calibration is valid, as all inks had a surface tension between the two calibration points. The glass pipette was de-greased after each measurement as some inks left a hydrophobic film behind (imagine what happens in a pen), rinsed with distilled water and dried at ambient conditions (no air blowing, as compressed air contains oil residues and blowing with the mouth will introduce aerosols; no heating as it is not much faster but increases the risk of glass breakage). The same pipette is used for all measurements, because the individual tip diameter is a critical factor for the drop size and for the calibration validity and because glass pipettes differ in their dimensions. About fife to seven measurements can be done per week.

 

Dynamic viscosity was measured by the capillary flow method. A conventional injection needle of 0.8 mm outer and ca. 0.25 mm inner diameter and 40.0 mm length was equipped to the outlet of a small plastic funnel and set on top of a 100 mL Erlenmayer flask. 1 mL of ink was put in the funnel and the time was measured until all ink had run through. The accuracy of the measurement is ca. +/- 1 second (the stopwatch is more precise, but it had to be operated by hand) resulting in a deviation between repetitions of not better than +/- 5%. The system was calibrated with water (31.5 sec; 1.002 mPa*s) and ethyl alcohol (48.0 sec; 1.20 mPa*s) and was tested with a 1.5% Gum Arabic solution in water (60.5 sec; 1.35 mPa*s). The calibration is valid, because all non-modified inks were in the range between the two calibration liquids and, by far, closer to the water side. All components in the setup were de-greased, rinsed with distilled water and air-dried between measurements. The same capillary was used for all measurements to be on the safe side as production deviations of the injection needles are unknown. About three to fife measurements can be done per week.

 

Ink consumptions were measured by inking up a certain fountain pen at least 8 (mostly 24) hours before the measurement and with a warm-up of ca. 1/3 of an A4 page text in normal handwriting speed immediately before the test to remove artefacts from nib drying or other imbalances. A sheet of paper was fixed on a plane, hard, water resistant pad (Faber-Castell pad 1094 F) to have minimal surface deformation during the drawing and no suction effect of the underground. The weight of the pen was determined at the start of the test with an analytical balance (0.1 mg accuracy), then lines were drawn along a ruler with ca. 4-5 cm per second drawing speed until ca. 5 to 6.5 meter total length was reached. Immediately, the pen was weighted a second time. The weight difference (mostly between 25 and 75 mg) was considered as ink consumption for the total line length (the sum of all drawn lines) and calculated to express gram ink per kilometre. The drawing speed had some influence on the amount of ink laid down on absorbent paper – the reproducibility was with +/- 6% in an acceptable range but much lower than expected. The influence of the drawing speed was much lower for the ink friendly paper, resulting in a reproducibility of +/-2%. The inkjet photo paper was not tested for reproducibility.

 

The pens were carefully washed, de-greased, rinsed with distilled water and dried at ambient conditions for at least 48 hours (in two exceptional cases only 40 hours) before refilling and before made ready for the next test. Not more than one ink can be tested in one or two fountain pens per week, because the necessary cleaning routine is time consuming and access to the analytical balance is not provided every day – mostly suffering from asynchronous availability of either the balance or a pen. For all that reasons, the measurements were done only once per ink per pen after the initial precision determination.

 

Ink spreading test was done by putting 10 (+/- 0.1) µL ink on a ca. 1 mm diameter spot on the paper and letting it dry completely in horizontal position for 24 hours. Initially all three paper types were used, but the hp inkjet and the Kyome letter pad paper, both showed next to no ink spread – all spots looked exactly the same so that those papers were not used after some initial tries.

[TheDutchGuy]

Wow. Consider me impressed. The (former) chemistry teacher in me admires and respects the rigour of your approach. Awesome! As a fountain pen hobbyist, some of your preliminary conclusions go against the grain of my experience, but I cannot argue with your methods. Will let this sink in for a while. PS in your second conclusion I assume you mean inversely proportional instead if indirectly proportional…?

 

Thanks for sharing this!!

[7is]

Thank you InesF! I've waited for some research on ink surface tension for ages. I'm so happy and I really appreciate all your time and effort!

[LizEF]

I'm still reading, but while I'm reading, I'd be curious to know what you, @InesF, think of An Ink Guy's method of testing viscosity.

[bunnspecial]

Thanks for the thorough report. Your post needs another read or two from me to fully digest it, but I am intrigued by your conclusion.

 

In particular, I'm not surprised that surface tension impacts perceived wetness than does viscosity.

[txomsy]

Yeah, thanks, that was an impressive work. Really professional.

 

Now, for the measuring. I think it is pretty straightforward. Just write on 1, 4 or 5 mm grid paper and take a close up picture. Then, on a computer, simply enlarge the picture and take the measures using as reference the grid whose fixed size you know.

 

If you like these kinds of experiments you may even consider getting a cheap USB microscope. That would simplify image acquisition and analysis. The main problem then might become dealing with the definition of "line", once you start distinguishing feathering in the image...

[LizEF]

And I'm done reading.  Wow is right!  That's some serious dedication.   And your conclusions strike me as what my instincts would have said.  I'm looking forward to any additional testing you do.

 

And thank you for showing the methods you use in a way that others can repeat them (if so inclined, and if they can get the tools / supplies).

 

Out of curiosity, how are you degreasing the pens?

[LizEF]

2 minutes ago, txomsy said:

The main problem then might become dealing with the definition of "line", once you start distinguishing feathering in the image...

Yes, once you're zoomed in enough, it gets hard to decide which pixel is part of the line and which isn't.

[InesF]

On 5/23/2021 at 10:02 AM, TheDutchGuy said:

Wow. Consider me impressed. The (former) chemistry teacher in me admires and respects the rigour of your approach. Awesome! As a fountain pen hobbyist, some of your preliminary conclusions go against the grain of my experience, but I cannot argue with your methods. Will let this sink in for a while. PS in your second conclusion I assume you mean inversely proportional instead if indirectly proportional…?

 

Thanks for sharing this!!

Thank you! Indeed, you are right about the inverse proportionality.

[InesF]

On 5/23/2021 at 10:21 AM, 7is said:

Thank you InesF! I've waited for some research on ink surface tension for ages. I'm so happy and I really appreciate all your time and effort!

Thank you, you're welcome!

[InesF]

On 5/23/2021 at 7:27 PM, LizEF said:

I'm still reading, but while I'm reading, I'd be curious to know what you, @InesF, think of An Ink Guy's method of testing viscosity.

I started my own tests as I can't figure out what the Ink Guy really measured. His method with the flowing drops on thew slanted glass plate is a mix of surface tension and viscosity. So it can measure the combined fluidity of an ink to a certain degree. The falling drop and the capillary flow methods are as simple but can measure both properties separately. Furthermore, all methods can and should be calibrated. The required tools are cheap and commonly available.

The Ink Guy has measured so many inks, he has filled a giant database meanwhile. However, my data so far show that the inks belong to two loose groups and except the massive change in viscosity from Gum Arabic addition, viscosity is of lower importance.

PS: I like your ink reviews on YT!

[InesF]

On 5/23/2021 at 7:31 PM, bunnspecial said:

Thanks for the thorough report. Your post needs another read or two from me to fully digest it, but I am intrigued by your conclusion.

 

In particular, I'm not surprised that surface tension impacts perceived wetness than does viscosity.

Thank you! It was a surprise for me too.

[InesF]

On 5/23/2021 at 8:06 PM, txomsy said:

Yeah, thanks, that was an impressive work. Really professional.

 

Now, for the measuring. I think it is pretty straightforward. Just write on 1, 4 or 5 mm grid paper and take a close up picture. Then, on a computer, simply enlarge the picture and take the measures using as reference the grid whose fixed size you know.

 

If you like these kinds of experiments you may even consider getting a cheap USB microscope. That would simplify image acquisition and analysis. The main problem then might become dealing with the definition of "line", once you start distinguishing feathering in the image...

Thank you! Indeed, an USB microscope will be the next investment.

The idea is to measure the delivered ink on area base. Lines from low surface tension inks are wider and more ink is delivered. I'm curious if the amount per area ist maybe constant.

[LizEF]

1 minute ago, InesF said:

I started my own tests as I can't figure out what the Ink Guy really measured. His method with the flowing drops on thew slanted glass plate is a mix of surface tension and viscosity. So it can measure the combined fluidity of an ink to a certain degree. The falling drop and the capillary flow methods are as simple but can measure both properties separately. Furthermore, all methods can and should be calibrated. The required tools are cheap and commonly available.

The Ink Guy has measured so many inks, he has filled a giant database meanwhile. However, my data so far show that the inks belong to two loose groups and except the massive change in viscosity from Gum Arabic addition, viscosity is of lower importance.

Thank you!  It's been a long time since I took a science class (and I never took any physics classes), so I don't always trust my memory or instincts about these things, and I really appreciate you explaining the differences in what your technique vs his measure / demonstrate.  (We have a thread (that I need to get back to work on) where we're trying to give some better instructions to potential ink reviewers so they're not overwhelmed by being asked to evaluate various ink attributes.  This thread of yours is very timely and will prove useful for our goals there, I think.)

 

5 minutes ago, InesF said:

PS: I like your ink reviews on YT!

Thanks!

[InesF]

On 5/23/2021 at 8:08 PM, LizEF said:

And I'm done reading.  Wow is right!  That's some serious dedication.   And your conclusions strike me as what my instincts would have said.  I'm looking forward to any additional testing you do.

 

And thank you for showing the methods you use in a way that others can repeat them (if so inclined, and if they can get the tools / supplies).

 

Out of curiosity, how are you degreasing the pens?

I use a laboratory quality detergent called 'Extran'. It is highly concentrated and of very high pH-value. It needs to be used with care, as it defats your skin and harms your eyes. It is aggressive enough to remove the almost solid vacuum silicone grease from surfaces and it attacks glass!

So yeah, not a simple thing. But I did the de-greasing after I have seen the remaining film some of the inks leave on the inside of the glass pipette.

Not necessary for daily use, but maybe critical if I like to measure some fine details.

[LizEF]

14 minutes ago, InesF said:

I use a laboratory quality detergent called 'Extran'. It is highly concentrated and of very high pH-value. It needs to be used with care, as it defats your skin and harms your eyes. It is aggressive enough to remove the almost solid vacuum silicone grease from surfaces and it attacks glass!

Um.  Wow!  Won't be trying that any time soon.

 

15 minutes ago, InesF said:

So yeah, not a simple thing. But I did the de-greasing after I have seen the remaining film some of the inks leave on the inside of the glass pipette.

Not necessary for daily use, but maybe critical if I like to measure some fine details.

You have officially elected yourself the resident inky scientist.  We're gonna let you continue with this task while the rest of us maintain a safe distance - with some PPE, just in case!

[DrDebG]

Fascinating and time consuming study!  Thank you for taking the time to do this.  It sounds like it will be a multi-year project.  Are you going to use the results to write a doctoral thesis?  😁

 

After you have nailed down your technique and satisfied yourself with consistent, replicative results, you may want to expand the study to include different "fountain pen friendly" papers (i.e. Tomoe River, Clairfontaine, Cosmo Air Light, etc.)

 

Thank you again for doing this.  I look forward to see more of your results. 

[sansenri]

1 hour ago, InesF said:

I use a laboratory quality detergent called 'Extran'. It is highly concentrated and of very high pH-value. It needs to be used with care, as it defats your skin and harms your eyes. It is aggressive enough to remove the almost solid vacuum silicone grease from surfaces and it attacks glass!

So yeah, not a simple thing. But I did the de-greasing after I have seen the remaining film some of the inks leave on the inside of the glass pipette.

Not necessary for daily use, but maybe critical if I like to measure some fine details.

Impressive work.

The measurements on surface tension really do shed a light on some ink's similarities.

 

The Extran seems to be the next best cure for reluctant converters...

[inkstainedruth]

What an amazing amount of work you've done, InesF.

The only concern I have is in the use of gum arabic, because a lot of people on here warn against its use (and inks containing it such as India ink) for fountain pens, because it can gum up a feed something awful.

And of course pens can be inconsistent, even between brands/models/nibs.

Ruth Morrisson aka inkstainedruth